The Cambridge History of Africa, Volume 6 PDF

THE CAMBRIDGE HISTORY
OF AFRICA
General Editors: J. D. FAGE and ROLAND OLIVER
Volume 6
from 1870 to 1905
Cambridge Histories Online © Cambridge University Press, 2008

THE CAMBRIDGE HISTORY
OF AFRICA
i From the earliest Times to c. 500 B.C.

edited by J. Desmond Clark
2 From c. 500 B.C. to A.D. 1050
edited by J. D. Fage
3 From c. 1050 to c. 1600
edited by Roland Oliver
4 From c. 1600 to c. 1790
edited by Richard Gray
5 From c. 1790 to c. 1870
edited by John E. Flint
6 From 1870 to 1905
edited by Roland Oliver and G. N. Sanderson
7 From 1905 to 1940
edited by A. D. Roberts
8 From c. 1940 to c. 1975
edited by Michael Crowder

THE CAMBRIDGE
HISTORY OF
AFRICA
Volume 6
from 1870 to 1905
edited by
ROLAND OLIVER
and
G. N. SANDERSON
I CAMBRIDGE
UNIVERSITY PRESS

P U B L I S H E D BY T i l l ; PRESS S Y N D I C A T E OF TIIH U N I V E R S I T Y OF C A M B R I D G E
The Pitt Building, Trumpington Street, Cambridge, United Kingdom
CAMBRIDGE UNIVERSITY PRESS
The Edinburgh Building, Cambridge C112 2RU, UK
40 West 20th Street, New York, NY 10011-4211, USA
477 Williamstown Road, Port Melbourne, vie 5207, Australia
Ruiz de Alarcon 13, 28014 Madrid, Spain
Dock House, The Waterfront, Cape Town 8001, South Africa
http://www.cambridge.org
© Cambridge University Press 1985
This book is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.
First published 1985

Reprinted 1990, 1994, 1997, 2001, 2004
Printed in the United Kingdom at the University Press, Cambridge
Library of Congress catalogue card number: 76-2261
ISHN o 521 22803 4
British IJbrary Cataloguing in Publication data

The Cambridge history of Africa
Vol. 6: From 1870 to 1905
1. Africa-History
I. Oliver, Roland and Sanderson, C>. N.
960 DT20
L'P

CONTENTS
Maps page x
Preface xiii
Introduction i
by R O L A N D O L I V E R , Professor of African History,
University of London
Africa on the eve of partition 10

by A. E. A T M O R E
A. North of the equator 10
The shores of the desert 11
The Maghrib and Egypt 25
Ethiopia and the Horn 36
West Africa 40
B. South of the equator 61
Western Equatorial Africa 63
Eastern Africa: the trading sphere of the
Swahili-Arabs 70
From Angola to Mozambique:
Portuguese economic activities and
their outreach 77
Southern Africa 83
Conclusion 92

CONTENTS
The European partition of Africa: Origins and
dynamics 96
b y G . N . S A N D E R S O N , Professor of History, Rqya/
Holloway College, University of London
Seeing Africa whole 96
The motives for territorial acquisition 100
The breakdown of informal empire 117
The scramble for the interior 13 5
South Africa and the Upper Nile 145
The partition and the ideology of imperialism 154
N o r t h Africa 15 9
byjEAN G A N I A G E , Professor of Contemporary History,
University of Paris IV, translated by Y V O N N E
BRETT
Algeria 15 9
Tunisia 171
Morocco 187
The Saharan regions 204
Western Africa, 1870—1886 208

by the late Y V E S P E R S O N , Professor of African
History, University of Paris I, translated by Y V O N N E
BRETT
From Senegambia to the Rivers of
Guinea 209
From Sierra Leone to the Bandama 222
From the Bandama to the Volta 227
From the Volta to the Niger 231
From the upper Niger to the upper
Volta 238
From the upper Senegal to the middle
Niger 245
The central Sudan 248
VI

CONTENTS
Western Africa, 1886-1905 257

b y j . D. H A R G R E A V E S , Professor Emeritus of
History, University of Aberdeen
2
The lines of siege, 1886-89 57
The European offensives 266
Initial patterns of colonial collaboration 280
Early impacts 289
Western Equatorial Africa 298

A. French Congo and Gabon, 1886—1905
by C A T H E R I N E COQUERY-VIDROVITCH,
Professor of History, University of Paris Vll
translated by Y V O N N E B R E T T
Early colonial contacts, 1886-97 300
Economic experiments, 1898-1905 305
B. King Leopold's Congo, 1886—1908 315
by J E A N S T E N G E R S , Professor of History, Free
University of Brussels, and J A N V A N S I N A , Professor of
African History, University of Wisconsin, Madison
In Europe (by J. Stengers) 315
Occupation and administration of the state (by J.
Vansina) 327
The economy 337
Christian missions and African religious
movements 345
Social life in the Congo State 350
Conclusion 357
Southern Africa, 1867-1886 359

by S H U L A M A R K S , Professor of Commonwealth
History, University of London
The discovery of diamonds and its impact 362
Confederation and the annexation of the South
African Republic 376
Wars of conquest south of the Limpopo 382
Christianity and class formation 399
South West Africa and Southern Rhodesia 405
Across the Zambezi 412
Vll

CONTENTS
8 Southern and Central Africa, 1886-1910 422

by S H U L A MARKS
The discovery of the Witwatersrand and
the end of African independence south
of the Limpopo 431
The scramble for Zambezia 439
The establishment of colonial states:
South West Africa and Rhodesia 442
The colonial presence north of the
Zambezi 448
Land, labour and taxation 45 5
Labour on the Rand 465
The conquest of the South African
Republic 472
The construction of the modern South
African state 481
9 Portuguese colonies and Madagascar 493

A. Angola and Mozambique, 1870-1905 493
by A L A N K. S M I T H , Professor of African History,
Syracuse University, and G E R V A S E C L A R E N C E
S M I T H , Lecturer in African History, School of Oriental
and African Studies, University of London
Angola and Mozambique before partition,
1870-85 493
The establishment of the frontiers, 1885—91 502
The occupation of the interior regions,
1895—1905 506
Social and economic structures, 1895-1905 513
B. Madagascar and France, 1870—1905 521
by the late H U B E R T D E S C H A M P S , sometime
Professor of African History, University of Paris I,
translated by Y V O N N E B R E T T
Rainilaiarivony, 1870—85 521
The phantom protectorate, 1886-94 526
The French conquest and the
insurrection, 1894—96 529
Gallieni's first tour of duty, 1896—99 531
Gallieni's second tour of duty, 1900-05 535
viii

CONTENTS
10 East Africa, 1870-1905 539

by M A R C I A W R I G H T , Professor of African History,
Columbia University
The commercial sphere and regional patterns 54°
The new sultanate and the Swahili commercial
system 5 5°
Towards a climax of Western penetration 561
The initial colonial experience 567
Military conquest in a context of ecological crisis 572
Sub-imperialism and colonial hierarchies 579
Towards a colonial economy 585
Conclusion 5 89
11 T h e Nile basin and the eastern Horn, 1870-1908 592
by G. N. SANDERSON,
Egypt: the end of Isma'Il's reign, 1876—79 592
Tawflq, 'UrabI and the British occupation,
1879-8} 599
The Mahdist revolution in the Sudan, 1874-85 609
Cromer's Egypt, 1883—1907 619
The Mahdist state under Khalifa 'Abdullahi 631
The birth of the Anglo-Egyptian Sudan,
1898—1907 640
Ethiopia: the reign of Yohannes IV, from 1875 645
Ethiopia: Menelik as emperor, 1889—1907 656
Somalia, c. 1870-1910 665
12 The European scramble and conquest in
African history 680
by J O H N L O N S D A L E , Fellow of Trinity College,
Cambridge
Frontiers of change 682
The causes of conquest 692
The African race for power 700
Confrontation and disorder 722
Reconstructions and evasions 750
Bibliographical essays 767
Bibliography g24
Index g9J
ix

MAPS
1 Northern Africa on the eve of partition page 12

2 Southern Africa on the eve of partition 63
3 T h e climax of British unofficial empire: Africa on
the eve of partition, c. 1878 118
4 T h e first phase of partition, to c. 1887 140
5 T h e scramble for the interior, to 1895 146
6 Africa partitioned, 1902 152
7 Algeria at the beginning of the twentieth century 164
8 The Sharifian empire at the beginning of the
twentieth century 194
9 West Africa, c. 1870 212-3
2 2
10 West Africa, c. 1905 9 ~3
11 French Congo and Gabon at the beginning of the
12 T h e Concessionary system in the French Congo 306
13 T h e Congo Independent State 328-9
14 Annexations of African territory south of the
Limpopo, 1866-97 384-5
15 Peoples of South West Africa before the 1904-5
uprising 407
16 Mashona and Matabeleland in the late nineteenth
century 410
17 Labour migration routes in the early twentieth
century 428
18 British Central Africa in the late nineteenth century 450
19 Southern Africa, c. 1910 482
20 Angola and Mozambique at the beginning of the
21 Madagascar 528
22 East Africa, late nineteenth century 542
23 East Africa, 1905 573
24 The Nile Valley in the Mahdist era, c. 1880-r. 1900 612

MAPS
25 N o r t h e r n Ethiopia under Yohannes I V and Menelik
II 646
26 Ethiopia, the Nile Valley and the Horn in the era of
Menelik II 658-9
27 Somalia, 1884—c. 1910 666
XI

PREFACE
In the English-speaking world, the Cambridge histories have

since the beginning of the century set the pattern for multi- volume
works of history, with chapters written by experts on a particular
topic, and unified by the guiding hand of volume editors of senior
standing. The Cambridge Modern History, planned by Lord Acton,
appeared in sixteen volumes between 1902 and 1912. It was
followed by The Cambridge Ancient History, The Cambridge Medieval
History, The Cambridge History of English Literature, and Cambridge
Histories of India, of Poland, and of the British Empire. The
original Modern History has now been replaced by The New
Cambridge Modern History in fourteen volumes, and The Cambridge
Economic History of Europe is now complete. Other Cambridge
Histories recently undertaken include a history of Islam, of Arabic
literature, of the Bible treated as a central document of and
influence on Western civilisation, and of Iran, China and Latin
America.
It was during the later 1950s that the Syndics of the Cambridge
University Pressfirstbegan to explore the possibility of embarking
on a Cambridge History of Africa. But they were then advised
that the time was not yet ripe. The serious appraisal of the past
of Africa by historians and archaeologists had hardly been
undertaken before 1948, the year when universities first began to
appear in increasing numbers in the vast reach of the African
continent south of the Sahara and north of the Limpopo, and the
time too when universities outside Africa first began to take some
notice of its history. It was impressed upon the Syndics that the
most urgent need of such a young, but also very rapidly advancing
branch of historical studies, was a journal of international
standing through which the results of ongoing research might be
disseminated. In i960, therefore, the Cambridge University Press
launched The journal of African History, which gradually demon-
strated the amount of work being undertaken to establish the past
xiii

PREFACE
of Africa as an integrated whole rather than — as it had usually

been viewed before - as the story of a series of incursions into
the continent by peoples coming from outside, from the Medi-
terranean basin, the Near East or western Europe. This movement
will of course continue and develop further, but the increasing
facilities available for its publication soon began to demonstrate
a need to assess both what had been done, and what still needed
to be done, in the light of some general historical perspective for
the continent.
The Syndics therefore returned to their original charge, and in
1966 the founding editors of The Journal of African History
accepted a commission to become the general editors of a
Cambridge History of Africa. They found it a daunting task to draw
up a plan for a co-operative work covering a history which was
in active process of exploration by scholars of many nations,
scattered over a fair part of the globe, and of many disciplines -
linguists, anthropologists, geographers and botanists, for example,
as well as historians and archaeologists.
It was thought that the greatest problems were likely to arise
with the earliest and latest periods: the earliest, because so much
would depend on the results of long-term archaeological investi-
gation, and the latest, because of the rapid changes in historical
perspective that were occurring as a consequence of the ending
of colonial rule in Africa. Therefore when, in 1967, the general
editors presented their scheme to the Press and notes were
prepared for contributors, only four volumes - covering the
periods 500 B.C. to A.D. 1050, A.D. 1050 to 1600, 1600-1790, and
1790—1870 — had been planned in any detail, and these were
published as volumes 2-5 of the History between 1975 and 1978.
So far as the prehistoric period was concerned, the general
editors were clear from the outset that the proper course was to
entrust the planning as well as the actual editing of what was
necessary entirely to a scholar who was fully experienced in the
archaeology of the African continent. In due course, in 1982,
Volume 1,' From the earliest times to c. 500 B.C. ', appeared under
the distinguished editorship of Professor J. Desmond Clark. As
for the colonial period, it was evident by the early 1970s that this
was being rapidly brought to its close, so that it became possible
to plan to complete the History in three further volumes. The first,
Volume 6, is designed to cover the European partition of the
xiv

PREFACE
continent, and the setting up of the colonial structures between

c. 1870 and c. 1905; the second, Volume 7, is devoted to the
'classical' colonial period running from c. 1905 to c. 1940; while
the focus of the third, Volume 8, is on the period of rapid change
which led from about the time of the Second World War to the
ending of formal control from Europe with the dramatic final
collapse of the Portuguese empire in 1975.
When they started their work, the general editors quickly came
to the conclusion that the most practical plan for completing the
History within a reasonable period of time was likely to be the
simplest and most straightforward. Each volume was therefore
entrusted to a volume editor who, in addition to having made a
substantial contribution to the understanding of the period in
question, was someone with whom the general editors were in
close touch. Within a volume, the aim was to keep the number
of contributors to a minimum. Each of them was asked to essay
a broad survey of a particular area or theme with which he was
familiar for the whole of the period covered by the volume. In
this survey, his purpose should be to take account not only of all
relevant research done, or still in progress, but also of the gaps
in knowledge. These he should try to fill by new thinking of his
own, whether based on new work on the available sources or on
interpolations from congruent research.
It should be remembered that this basic plan was devised
nearly twenty years ago, when little or no research had been done
on many important topics, and before many of today's younger
scholars — not least those who now fill posts in the departments
of history and archaeology in the universities and research
institutes in Africa itself - had made their own deep penetrations
into such areas of ignorance. Two things follow from this. If the
general editors had drawn up their plan in the 1970s rather than
the 1960s, the shape might well have been very different, perhaps
with a larger number of more specialised, shorter chapters, each
centred on a smaller area, period or theme, to the understanding
of which the contributor would have made his own individual
contribution. To some extent, indeed, it has been possible to
adjust the shape of the last three volumes in this direction.
Secondly, the sheer volume of new research that has been
published since many contributors accepted their commissions
has often led them to undertake very substantial revisions in their
xv

PREFACE
work as it progressed from draft to draft, thus protracting the

length of time originally envisaged for the preparation of these
volumes.
However, histories are meant to be read, and not to be
commented on and analysed by their general editors, and we
therefore present to the reader this further volume of our
enterprise.
September 1984 j. D. FAGE

ROLAND OLIVER
XVI

INTRODUCTION
This volume deals with a period of about thirty-five years, from

about 1870 till about 1905. This was, of course, the period which
saw, first, the scramble by European powers and interests to stake
out territorial claims in Africa, next, the paper partition of the
continent by those powers, and finally the colonial conquest and
occupation. In 1870, the only large areas to have suffered such
inroads lay either to the north of the Sahara or else to the south
of the Limpopo. By 1905, Ethiopia and Morocco were the only
truly independent African states, and the innumerable petty
polities of pre-colonial Africa, as well as a few larger ones, had
been consolidated into the forty-odd colonies and protectorates
which were destined to become, with only a few subsequent
changes, the sovereign states of modern, post-colonial Africa.
Yet, to characterise the period wholly in this way, is to see it
too much through European eyes. As the succeeding chapters
show, African history throughout this period pursued paths still
largely separate from those of the European colonisers. The
treaty-making expeditions of a Binger or a Brazza, of a Johnston
or a Lugard, of a Cardoso or a Serpa Pinto, were scarcely to be
distinguished by any African observer from the trading caravans
of the Dyula or the Hausa, of the Sudanese jallaba or the
Swahili-Arabs, of the Mozambican Chikunda or the Angolan
pombeiros. The diplomatic partition of the continent passed
almost unnoticed by the Africans whose territory was at issue. The
colonial conquest and occupation was experienced by them as a
piecemeal phenomenon, which affected some of those living near
the coasts as early as the 1870s and 1880s, but which reached most
of the interior peoples only during the 1890s and the 1900s. Even
by the end of the period, only a small minority of Africans had
seen a white face or had any idea that their countries were subject
to foreign rule.
Seen from the African end, the European infiltration and

INTRODUCTION
conquest was, as Dr Lonsdale makes clear in the concluding

chapter, a process and not an event. Within the confines of any
particular colonial territory, it lasted a full generation, in the
course of which African power progressively diminished while
European power correspondingly increased. At the start of the
process, European agents travelled singly or in pairs, accompanied
by two or three score of lightly armed caravan porters, very much
in the style of the explorers and missionaries who had preceded
them. The porters carried the usual trade goods, which were used
to buy food and protection, and also to aid the negotiation of what
were probably understood by the Africans as vague alliances,
somewhat akin to blood brotherhood, which involved the gift of
flags and the signing of crosses on pieces of paper. It was normal
for several years to elapse between the execution of such treaties
and the appearance of any colonial official appointed to carry
things a stage further. Meanwhile the political and economic life
of the African peoples continued on its accustomed way.
But this did not mean that nothing happened. The opening
chapter by Mr Atmore, while in a sense only recapitulating the
lessons of earlier volumes, does provide a massive summary of
the extent to which the commercial opening up of Africa had
already been carried out by Africans before the beginning of the
colonial period. He concludes that by 1870 there was hardly a
corner of the continent which was unfamiliar with some of the
manufactured goods of Europe and Asia. Manchester cottons
crossed the Sahara on camelback. The small arms of Birmingham
and Liege were paddled up the Niger and the Congo in dug-out
canoes. Portuguese brandy and Brazilian tobacco were carried on
human heads from Luanda and Benguela to the Luba and Lunda
states situated at the very centre of the sub-continent. The calicos
of Gujerat and New England travelled up the East African
caravan routes from Zanzibar. The cowries of the Maldives,
which in medieval times had reached West Africa via the Red Sea
and Cairo, were now carried round the Cape in European
shipping, and moved inland on a scale which caused serious
inflation all the way from Kano to Timbuktu. The main limitation
on this kind of commerce was the cost of transport. The imports
were expensive, and the exports which paid for them had likewise
to be luxury items, valuable enough to justify human or animal
portage — gold and copper, ivory, palm-oil, coffee, copra, ostrich

INTRODUCTION
feathers and rare skins, and of course the continuing traffic in

slaves, no longer significant in the intercontinental trade, but still
a vital lubricant of most of the internal long-distance commerce
of Africa. In the long run, a breakthrough into bulk trading of
primary produce against cheap manufactured goods could only
come through the development of mechanical transport. In the
short term, the indigenous caravan trade continued to grow
through most of the period dealt with here. The early years of
colonial rule merely added somewhat to the volume of goods that
had to be carried long distances by men and beasts.
Besides the economic opening up of Africa by the extension
of the traditional caravan trade, there was also detectable a parallel
growth of political paramountcies, which continued through most
of the period under review, and which constituted an important
part of the context in which European colonisation took place.
The most obvious examples are the conquests of Menelik,
undertaken first as king of Shoa and later as emperor, which more
than doubled the territory of Christian Ethiopia. In this case, the
systematic rearmament of the Ethiopian forces with western
weapons was effective enough to bring about the defeat of the
Italians in 1896 and to secure the freedom of the country from
further European aggression until 1935. Other instances of
African imperialism were shorter lived, but at least as significant
while they lasted. There was Khedive Isma'H's attempt to found
a North-East African empire extending to the Great Lakes and
the Indian Ocean, and the simultaneous attempt of Sultan Bargash
to extend the dominions of Zanzibar from the Indian Ocean coast
into the East African interior. More typical, however, were the
loose, military overlordships created by the magnates of the
caravan trade to safeguard the sources of their wealth. Especially
where traditional polities were small and weak, traders defended
themselves by travelling with armed escorts and establishing
fortified depots for their goods. From there, it was a short step
to taking hostages or seizing cattle in order to assure the supply
of ivory, slaves or rubber at attractive prices, and this in turn often
led on to the trader settling down as ruler, taking tribute from
the indigenous population by military means. Such were the
origins, to name but a few, of Msiri's empire in Katanga, of Tippu
Tip's between the Lualaba and the Lomami, of al-Zubayr's on the
Bahr al-Ghazal, of Rabah's in Kanem, of Samory's on the upper

INTRODUCTION
Niger. All these hegemonies reached the peak of their power

within our period, and together they probably affected the lives
of more people than the nascent colonial states which were
eventually to supersede them. Some of these hegemonies were,
like the Mahdist state in the Sudan and the movement of
Muhammad 'Abdallah Hasan in Somalia, the direct result of
militant Islamic protest against alien and infidel encroachment.
Such hegemonies could ultimately be destroyed by European
armed force; but not the ideas and attitudes that had generated
them. Mahdism and other militant manifestations of Islam often
survived colonial conquest as a major influence upon the political
behaviour of Muslim populations. The fear that this survival
might once more generate major resistance, perhaps on a scale
uncontainable by local power-resources, was often a real constraint
upon the policy of early colonial administrations, whose ' internal
security' was almost always precarious.
Militantly Islamic hegemonies were by no means the only forces
emanating from the world of Islam which took part in the
'alternative', internal, scramble for Africa. Moreover, there can
be no doubt that in the period treated in this volume Islamic
religious propaganda, and Islamic knowledge of the wider world,
were vastly more significant than the incipient Christian rivalry.
In the northern third of Africa the cultural predominance of Islam
of course survived without difficulty. Here Islamic preponderance
had long been established, sometimes for many centuries; and
political prudence often constrained colonial governments to
discourage or even prohibit the deployment of the full panoply
of missionary education even in the later colonial period. In the
middle third of Africa, however, the situation was very different,
and had the forces of the alternative scramble been allowed even
a little more time to establish themselves, the cultural predominance
of Islam might today extend to the Zambezi and the Cunene rather
than to the Bahr al-Ghazal and the Niger. It was not only where
Muslims seized political power that Islamic influences spread.
Where indigenous political systems were strong and centralised
enough to afford effective protection to strangers, as in Yorubaland,
or Buganda, or among the Yao principalities of the upper
Rovuma, Muslim traders and clerics could re-enact the processes
of infiltration from above familiar in the Sudanic belt of Africa
since medieval times. Where societies were small and weak, like

INTRODUCTION
those of eastern Zaire, the porters and camp followers of the

long-distance traders, recruited often from the dregs of society,
became the nucleus of a Muslim community which would long
outlast the temporal overthrow of the Swahili-Arab masters by
a Dhanis or a Chaltin.
Christianity, though still far out-distanced by Islam, was
nevertheless another factor in the history of the period, which
needs to be distinguished from the secular outreach of the
competing European nations. In several key areas of tropical
Africa, to the west, the east and the south, Christian missionary
propaganda had anything from twenty to fifty years' start of the
forces of imperial expansion. Pioneering outposts apart, however,
most of the early penetration of the interior regions occurred
during the 1870s and the 1880s, and at this stage missionaries from
overseas were almost invariably accompanied, if not indeed
preceded, by African converts, auxiliaries or independent evan-
gelists from the neighbourhood of the earliest outposts. The
spread of the formal cult, inevitably, was slow, depending as it
did on long years of language study, on the translation of scripture
and liturgy, and on the confrontation of radically different moral
and social traditions. Within our period baptised Christians
remained few, but what Livingstone called 'the wide sowing of
the good seed' was in itself a revolutionary event, particularly in
areas little affected by Islam. Even the first, superficial talk of
Christianity turned minds and imaginations outward, beyond the
immediate circle of small communities, and created an attitude of
expectancy and questioning which prepared the way for later
change. When colonial governments began to appear on the scene,
societies with even the faintest degree of Christian contact were
better able to pursue the politics of survival than those which had
none. A later generation of colonial Africans would be brought
up on the French revolution or the constitutional history of the
Tudors and Stuarts. For the last generation of politically
independent African tribal societies, it was perhaps more apposite
to have heard the story of Jesus, with its background of imperial
relationships between the Romans and the Jews.
At the time of its colonisation by Europe, therefore, Africa was,
to a greater extent than ever before in its history, opening itself
to the outside world as well as being forced open by that world.
Its myriad societies were also taking some steps, however

INTRODUCTION
inadequate, to improve their own defences. Firearms, long

familiar on the Guinea coast and along the sub-Saharan savanna
fringe, were now to be found almost everywhere. In most of the
larger polities at least the royal bodyguards were so armed. Of
course, the weapons traded to Africa were usually out-of-date
ones that had been replaced in European armies - usually, but by
no means always. From about 1890 Menelik was able to refuse
all but the latest models; and from about 1900 Ethiopia became
a centre for the distribution of up-to-date firearms through much
of north-eastern Africa. Again, there were no machine-guns in any
African armoury. But until about 1890, when the Maxim gun came
on to the market, there was no effective automatic weapon in any
European armoury either. Of course, in any conflict, forces with
modern equipment had the advantage of African armies many
times their number. But the fact is that the military resources of
colonial governments during the first twenty years of their
existence were almost ridiculously inadequate also. Johnston was
sent to ' rule' Nyasaland and North-Eastern Rhodesia in 1891 with
a force of seventy Sikh soldiers seconded from a Punjab regiment
and commanded by a single British subaltern. Lugard, as the first
British high commissioner for Northern Nigeria in 1898, had two
thousand African troops of the West Africa Frontier Force, led
by less than a score of British officers and NCOs, with which to
establish his authority over the Fulani empires of Sokoto and
Gwandu, as well as over a host of independent ethnic societies.
Clearly, in circumstances like these, no conceivable degree of
superiority in weapons and training could enable a so-called
'government' to lay down the law to its subjects. With skilful
management, it might just be able to protect its headquarters and
its main line of communications with the outside world. But any
wider operations to meet the inevitable challenges to its authority
could only be undertaken on the basis of alliances formed with
some of its 'subjects' against some others. Levies drawn from
' friendly tribes' were a regular feature of early colonial warfare,
and they were rewarded with booty seized under the eyes of the
colonial authority. In any balanced view of the period, therefore,
an embryo colonial government was but one of many forces
jostling for power in a given region. During their early years
several of them came to the very brink of disaster and had to be
rescued by special expeditionary forces, like those sent to
Madagascar in 1894, to Asante in 1896 and 1900, to Uganda in
6

INTRODUCTION
1897, to German East Africa in 1888 and 1905, to German South

West Africa in 1904. In the end, after twenty years of training and
experience, and following the development of rapid communica-
tions by road and rail, colonial forces became more adequate to
their task; but that situation was not reached in many parts of
tropical Africa during the period treated here. And meanwhile the
military potential of the African subjects, whether friendly or
hostile, remained of great importance.
Essentially, then, early colonial governments in Africa were
weak governments, and they were so because the European
powers which controlled them were not sufficiently interested in
their new possessions to subsidise stronger governments. During
the period under review, the colonial powers were mainly
concerned in reserving tropical African territories for possible
development in the future. Meanwhile, all that mattered was that
they should be as nearly as possible financially self-supporting,
while maintaining enough internal security to command the
respect of neighbours and rivals. The functions of colonial
governments were thus seen as very limited. Ideally, the process
involved the creation of a viable nucleus, either on the coast or,
in the case of an inland territory, in some strategically central area.
The populations of the nuclear area were seen as allies. They paid
tax at an earlier stage than others, but they enjoyed the privileges
of the trusted intermediary. All innovations, like roads and
schools, reached them first, and the colonial government by taking
its tithe of their prosperity, was gradually able to extend the
administered area outwards until it reached the frontiers. In
practice, things seldom worked out quite so conveniently. Those
living beyond the pale did not always sit down quietly to await
their incorporation. They raided caravans, or they cut down
telegraph wires to turn them into bangles and necklaces, or they
crossed the borders and threatened neighbouring colonial
governments, and so they had to be 'dealt with' forcibly and out
of due turn, sometimes by the despatch of a special, and very
expensive, expeditionary force which put the colonial revenues
quite out of kilter. Nevertheless, large areas in most African
colonies did continue to administer themselves throughout our
period with little more than an occasional glimpse of the new
masters. Early colonial governments divided, but it can hardly be
said that they ruled.
It was only, in fact, at the northern and southern extremities
7

INTRODUCTION
of the continent that a different kind of colonialism, initiated much

earlier, and centring around the perceived interests of substantial
bodies of European migrants, had, at least by the end of our
period, established full governmental control over the lives of the
indigenous Africans. In the case of Algeria, the decisive transition
from military rule of most of the native population to civil
administration conducted in the interests of the French settlers,
occurred with the transfer of soldiers to the metropolis during the
Franco-Prussian war. It continued with the great land seizures
following the suppression of the al-Muqranl rebellion. In southern
Africa, although the history of white settlement was so much
longer, the vital transition from a situation of wary coexistence
between various European and African polities to one of com-
prehensive conquest, incorporation and the authoritarian rule of
black societies by whites, was almost contemporary with that in
Algeria. The precipitating factor in this case was mineral develop-
ment, which led to a rapid reinforcement of the white population,
to a surging demand for black labour in the industrial areas, to
a rapid extension of the railway system and to an agricultural
revolution stimulated by the food requirements of the new towns.
All this made for the formation of a southern African colonial
system which, as we now know, was to prove uniquely resilient
by comparison with the rest of the continent. But that uniqueness
was by no means so apparent to contemporaries, for whom
developments in southern Africa could just as well have been a
portent of what was to happen elsewhere. The mineral resources
of southern Africa were, after all, not unique: a comparable
situation existed round the Copperbelt. And, minerals apart, the
southern African pattern of colonialism was widely assumed to
be reproducible wherever Europeans were prepared to commit
themselves to migration and settlement. The great nineteenth-
century migrations to North America and Australasia were still
fresh in the European memory. If, as Cecil Rhodes and so many
others imagined, that trend were to continue into the twentieth
century, then all Africa, and particularly all Africa south of the
equator, might prove to be the next major destination. That it did
not in the event prove to be so, was in large measure due to the
response of the tropical African peoples to the alternative, much
weaker style of colonialism established in their midst.
Enough has been said to show that much of the significance

INTRODUCTION
of the period 1870—1905 in African history arises from local

situations. Hence, most of the volume is composed of regionally
rather than thematically oriented chapters. It should perhaps be
explained that the volume was planned, and the contributors
chosen, by Professor Sanderson, who was the original volume
editor. It is well known that collaborative works are subject to
the hazard that the speed of production cannot be faster than that
of the slowest contributor. When contributors are chosen from
the leading figures in the field, it is almost inevitable that death,
promotion or competing literary obligations will remove or delay
some of them at an embarrassingly late stage in the programme.
Often, it is the volume editor who has to step into the breach.
On this occasion, however, it was the volume editor himself who
found himself too burdened by other duties both to maintain the
momentum of production among his colleagues and to write the
two important chapters which stood against his name. It was in
these circumstances that one of the general editors, Professor
Oliver, was asked in 1982 to take over the remaining editorial
responsibilities. He did so in the knowledge that progress on the
volume had already been overtaken by that on volumes 7 and 8
which had been intended to follow it, and that, in the interests
of completing the series within a reasonable time, some unevenness
of presentation and coverage would have to be accepted. Professor
Deschamps, for example, had shortly before his death written a
chapter on Madagascar which was rather more concerned with
the history of colonial administration than most other chapters.
Professor Person likewise became ill and died before his contri-
bution was quite complete. It will be noticed that the French
contributors have proved much readier than their English
colleagues to observe the editorial request to keep footnote
references to a minimum. Ideally, of course, a volume editor
should have all the draft contributions on his desk together, and
in time for co-ordinating changes to be made: in practice, it is
doubtful whether any editor of a collaborative work has ever
enjoyed such a privilege. The production line of a great university
press may be more accommodating than some others, but in the
end there comes a moment when it, too, must be served.

CHAPTER 1
AFRICA ON THE EVE OF PARTITION
A. NORTH OF THE EQUATOR

It has become a truism of historical writing to conceive of Africa
in the course of the nineteenth century as becoming increasingly
a part of, and a product of, the expansion of Europe, which,
beginning in the fifteenth and sixteenth centuries, had integrated
ever larger areas of the world into a single economic system. In
attempting an overview of the state of the continent on the eve
of partition — roughly over the decade of the 1870s —a large
number of questions arise from a consideration of this truism. To
what extent was Africa already an adjunct of an economic system
dominated by Europe ? What was the relationship between Africa
and this European system — was it one of an equal or an unequal
exchange of commodities ? To what extent was Africa dependent
economically, if not yet politically? What social and ideological
changes were beginning to follow from this dependency? Was
Africa a fruit ripe for plucking in the 1870s, was there a certain
inevitability about the forthcoming imperialist carve-up, or was
partition an extraneous historical occurrence forced upon a
continent which had within it other options for coping with the
future?
There are no answers to these questions that are at the same
time simple and sensible. Certainly the answers to all such queries
will differ, according to the region of Africa which is under
scrutiny. Even within particular regions, the situation of indi-
vidual states, societies or groups of people, their relations with
each other and with the outside world (especially with the
European capitalist economies) varied greatly. To avoid empty
generalisations, an overview of the African 1870s is bound not
only to break down the continent regionally, but also to explore
themes such as independence or dependence of action among the
diverse and changing societies who lived within these regions.
10

THE SHORES OF THE DESERT
The metaphor of the Sahara as a sea, separating the lands to the

north and to the south, and yet providing routes by which
commerce in men, goods and ideas passed from one side to
another, is very ancient. In this analogue, oases are islands, the
camel is the ship of the desert and caravans are convoys. Arab
geographers spoke of the country on the edge of the Sahara as
the sahel, the shore. The French, the most imaginative of the
European scramblers for Africa, revived the idea. After the
invasion of Algeria it became a regularly stated theme of French
political comment to regard North Africa and the western and
central Sudan as the border regions of a ' second Mediterranean'
which was to be a French mare nostrum.1 In the actuality of
historical events, this analogy is too fanciful and it became too
much of a cliche to bring much enlightenment to an obdurate and
complex reality. Yet for heuristic purposes, in attempting a
bird's-eye-view of the northern part of Africa at the end of the
1870s, it is tempting to pursue it a littler further.
There were many kinds of trade carried on in the desert and
the countries surrounding it. There was short-distance trade, of
which a good example was the age-old exchange of grain and
dates. Surplus wheat and barley was produced in many of the
settled lands bordering the Sahara on the north, and surplus millet
in many of the countries to the south. Dates were produced in
the oases and to some extent in the Sahelian lands. It has been
suggested that the merchant class of Ghadames, who were the
operators of the routes crossing the desert from Tripoli to Bornu
and Hausaland, became involved in this long-distance trade as a
result of their participation in an exchange of grain and dates with
the southern Tunisian and Algerian borderlands.2 A nineteenth-
century example of this local trade, which again provided an
entrance to the more lucrative fields of international trade for
some of the people concerned, was that of the bedouin Arabs of
southern Cyrenaica. The Mejabra at Jalu oasis, and the Zuwaya
at al-Kikharra and Kufra, had date plantations which produced a
' huge surplus', but were unable to grow sufficient cereals for their
1
Marion Johnson, 'Calico caravans: Tripoli-Kano trade after 1880', / . Afr. Hist.,
1976, 17, 1, 99; A. S. Kanya-Forstner, The conquest ofthe western Sudan: a study in French
imperialism (Cambridge, 1969), 26, 60-71.
2
Stephen Baicr, An economic history of central Niger (Oxford, 1980), ch. 3.
I I

• Towns (European and Egyptian)
Ceuta(Sp) u KAN EM African states and peoples
Z/72 Egyptian territory
Tetuan V *!? i e . r e Turns • •••*• Trans-Saharan trade routes
TUKOLOR
EMPIRE
SOKOTO
CALIPHATE >
oKano Zeila (Eg
Berbera
H)OPIA#Harar(Eg)
I YORUB
°Abeokuta
Elmina (B
1872Bf)
i Northern Africa on the eve of partition

inhabitants. They therefore traded their surplus dates for wheat
and barley grown by settled Arabs living on the Jabal al-Akhdar
and along the wadis of northern Cyrenaica.
It was the Mejabra who pioneered the international trading
route across the desert from Cyrenaica to Wadai in the early 18oos,
and the Zuwaya became involved after the foundation of %awiya
(fortified settlements and caravanserais) by the SanusI order
brought a greater degree of peace and security to the whole region.
Further west, the Awlad Sulayman Arabs traded in dates and grain
in the region between the Gulf of Sirte and the Fezzan. During
the 18 30s the Awlad Sulayman came into conflict with the
Ottoman authorities, who were taking over the Fezzan from the
old Karamanli rulers, and in 1842 they migrated across the desert
to Borku on the northern edge of the Chad basin, to escape the
alien intruders. In Borku they became involved with a similar
network, trading dates and grain with the settled people of
Kanem.3
A somewhat more complex variety of the same basic pattern
of regional trade was to be found in the Sahel west of Lake Chad.
In the country of Damergu, to the north of Zinder, the Tuareg
nomad aristocracy engaged in an annual transhumance which
took them from Air to the Hausa cities of Sokoto, Kano and
Katsina. While pasturing their camel herds in Air, the Tuareg sent
caravans to collect salt from Bilma. Moving south, the Tuareg
visited their estates in Damergu and distributed salt and dates to
the village heads of the local, semi-independent Hausa- and
Kanuri-speaking farmers. On their return northwards from
Hausaland, the Tuareg collected a prescribed quantity of millet
from the villagers. The farmers of Damergu also engaged in
longer-distance, but still regional, trade. By no means all the
population of Air practised transhumance. Their basic diet of milk
was insufficient for their needs, especially in the dry season. Millet
was transported the 300-400 km to Air by Damergu farmers in
caravans of pack animals, usually oxen. The millet was traded for
cattle, sheep, donkeys and camels. Many of these found their way
to the large markets of the Hausaland savanna. The inhabitants
1
Dennis D. Cordell, 'Eastern Libya, Wadai and the Sanusiya', J. Afr. Hist., 1977,
18, 1, 26-7 and 27, n.29; Cordeil, 'The Awlad Sulayman of Libya and Chad: a study
of raiding and power in the Chad Basin in the nineteenth century', unpublished M.A.
thesis, University of Wisconsin-Madison (1972), 104-7; A. Adu Boahen, Britain, the
Sahara, and the western Sudan, 17S8-1861 (Oxford, 1964), 91- 2.
1}

of Damergu had moved into the district towards the end of the
eighteenth century, when the climate of much of the Sahel was
improving greatly, following the disastrous drought of the middle
of that century. Their complex regional trading networks — which
appear to be wholly unrelated to the European economic system -
were at a peak during the last three decades of the nineteenth
century, before and after the establishment of European rule. By
this time Damergu was also an important component in the
international trade which was carried north across the Sahara.4
A somewhat simpler version of short-distance trade between
desert herders and savanna farmers grew up in the Maraka
country in and north of the upper Niger valley. The Maraka were
' Malinke-ised Soninke', not dissimilar in their trading activities
to the Dyula and Diakhanke of the western Sudan. The basis of
their internal economy was their trading relationship with nomadic
Moors who lived in the desert to the north of the Bambara of Segu
and Kaarta. In the dry season the Moors migrated south; and their
herds grazed the harvested fields of the Maraka, eating the husks
and manuring the soil. The Moors traded hides and some of their
livestock for millet and cotton cloth. Like the Tuareg of Air, the
Moors needed cereals to supplement their milk diet in the dry
season, and used cloth for clothes and tents. In the nineteenth
century the scale of this trade increased considerably, and the
Maraka turned to a kind of plantation system to produce surplus
grain. These plantations were operated by slave labour. Larger
numbers of Moors were attracted by the larger amounts of grain,
cloth and numbers of slaves available for trade in Maraka country.
On their part, the Moors brought in horses and salt. The enlarged
scale called for more sophisticated methods of conducting the
trade. Plantations worked by slave labour were a type of surplus
production which was common, in a number of varieties,
throughout much of the savanna of West Africa in the nineteenth
century; for example, in the Sokoto caliphate. And what has been
termed the 'desert-side' trade was also widespread.5
A consideration of these pivotal local or regional trading
4
Stephen Baier, 'Trans-Saharan trade and the Sahel: Damergu, 1870-1930', J. Afr.
Hist., 1977, 18, 1, 51-3.
5
Richard Roberts, 'Long distance trade and production: Sinsani in the nineteenth
century',/. Afr. Hist., 1980, 21, 2, 170-4; M. Klein and P. Lovejoy, 'Slavery in West
Africa' in H. Gemery and J. S. Hogendorn (eds.), The uncommon market: essays in the
economic history of the Atlantic slave trade (New York, 1979); P. E. Lovejoy and S. Baier,
'The desert-side economy of the Central Sudan', Int. J. Afr. Hist. Studies, 1975, 7, 4.
'4

networks of the Sahel and northern savanna leads naturally on to

the trading systems linking the Sudanic countries to the Mediter-
ranean lands, which appeared in the 1870s to be both generally
prosperous and fairly secure. The western routes ran from
Morocco to Senegambia and the Niger. Sijilmasa, in the Tafilelt
country south of the High Atlas, which had for centuries been
the chief northern entrepot, was no more. Its place had been taken
by Taroudant in the Wadi Sous and Tindouf in the al-Haba oasis,
south of the Wadi Dar'a. Northwards, trade from Taroudant and
Tindouf reached Agadir, Mogador and Marrakesh, and thence the
rest of Morocco. Southwards, one route ran more or less parallel
to the Atlantic coast, via Idjil and the oasis of Chingetti in Adrar
to the Sahel markets of Tichit, Walata, Kasambara, Nioro and
Sokolo, to cross the upper Niger at Sinsani, Segu, the Maraka
towns of Nyamina and Banamba, or Bamako. From the upper
Niger, trans-Saharan trade interconnected with the complex
trading systems of Senegambia and the Niger flood plain north
of Mopti. By the 1870s this route from Tindouf to the upper Niger
was being used only spasmodically, at least in its entirety.
Arabic-speaking Berber groups such as the Trarza, Brakna and
Kunta - the people known as the Moors - traded rock salt from
Idjil both north to Morocco and south to Senegambia and the
upper Niger. It was Idjil salt that formed one of the exchange
commodities in the relationship between Moors and Maraka.
Some cloth from Senegambia and Upper Niger found its way to
Morocco, but most was used by the desert people themselves. A
little gold from the ancient mines of Bambuk and Bure trickled
north, and a few slaves continued to be transported into Morocco.
But of all the trans-Saharan routes, the most westerly suffered
most from competition of the networks spreading inland from
the European trading posts on the Atlantic coast. In Upper
Guinea these networks had penetrated well over 1,000 km inland.
The volume, or the potential, of the Senegambian and Upper
Niger trade was great enough to tempt Moroccan merchants to
establish themselves in Dakar and St Louis at the end of the 1870s.
In doing so, they did not seem to incur open French hostility,
although the Moroccans and the French had the same aim in
mind - to capture and divert the trade of the interior. The
Moroccans traded up the Senegal and as far as Timbuktu for gold,
gum and other commodities which they shipped up the coast to
15

Mogador. This activity on the Senegal river by subjects of the
Sultan can be seen as one aspect of a wider project to reaffirm
Moroccan influence (and even control) over much of the western
Sudan.6
If the western Saharan trade route was less used in the second
half of the nineteenth century, that which led to Timbuktu still
carried a considerable quantity of goods. This route started at the
same north entrepots - Tindouf or the oases in Tafilelt — and ran
via the salt-mining oasis of Taodeni to the Sahel town of Arawan,
and so to Timbuktu. Another route to Timbuktu began at the
great market oasis of Ghadames and ran via Tuat to In Salah and
Arawan. Further east, the route to Hausaland went south from
Ghadames to Ghat, and so through Agades to Zinder and Kano.
East of the Hausaland-Ghadames route was the ancient ' Gara-
mantian road' from Tripoli to Bornu. This also had its main
northern entrepot at Ghadames. Tracks went either to Ghat or
to Murzuk in the Fezzan, and then through Kawar, with its great
salt-producing oasis of Bilma, before terminating at the market
of Kukawa, just west of Lake Chad in Bornu. In the middle of
the nineteenth century this route had been in decline, possibly
because of the weak political structures at either end.
The most recent of the trans-Saharan routes was that between
Benghazi and Wadai. This had been pioneered early in the
nineteenth century by Mejabra Arabs in southern Cyrenaica. The
desert journey on this route started at the oases of Awjila and Jalu,
proceeded to a refreshment station at Kufra and then crossed the
Tibesti massif to the Wadai capital at Abeche. Still further east
was the famous Forty Days' Route, the Darb al-Arba'In, from the
Nile at Asyut to Kutum and Mellit, to the north of El Fasher. This
was the main means of communication between Egypt and the
sultanate of Darfur. The major importance of the Darb al-Arba'In
was as a route to the Negro lands beyond the frontier of Darfur,
in Dar Fertit and the Bahr al-Ghazal.
There were also many lateral routes, running roughly east and
west across the Sahara, the Sahel and the Sudan. One went from
Cairo to Tripoli, the Fezzan and Tunis via Jaghbub, Jalu and
Awjila. Another went from southern Morocco to Ghadames,
6
See C. W. Newbury, 'North African and western Sudan trade in the nineteenth
century: a re-evaluation', J. Afr. Hist., 1966, 7, 2, 244, especially notes 34-6; cited in
Philip D . Curtin, Economic change in precolonial Africa: Senegambia in the ear of the slave
trade (Madison, 1975). 280-1.
16

Tripoli and Murzuq, via Tuat. A third ran from the Red Sea port
of Suakin across the Atbara to Berber, Shendi and Khartoum, and
thence westwards to Darfur, Wadai, Bagirmi and Kanem-Bornu
into Hausaland, reaching as far south and west as market towns
such as Salaga on the edge of Akan country. These lateral routes
catered for pilgrims and short-distance. commerce rather than
long-distance international trade. Trading networks emanating
from the western Sudan and Hausaland were concerned with a
wide range of commodities, almost always carried out according
to well-regulated conditions. On the other hand trading activity
from the Muslim states of the central Sudan - Bornu, Bagirmi,
Wadai and Darfur - was almost entirely limited to slaves and
ivory, obtained by raiding the non-Muslim peoples living in the
savanna north of the Cameroun and Congo forests - in Adamawa,
Dar Runga, Dar al-Kuti and Dar Fertit. The jallaba merchants
operating from the Nilotic Sudan, and foreigners based at
Khartoum, commenced their activities in Equatoria and Bahr
al-Ghazal as ivory traders, but this became inextricably linked
with raiding for slaves and livestock.
Only far to the east, beyond the Nilotic Sudan, was there trade
on a scale comparable with that of the western and central Sahara.
The products in the greatest demand by the outside world came
from the Oromo (Galla) and Sidama country deep in the hinterland
of the Horn. This was partly because these southern highland
areas were the richest in the whole of the Ethiopian region.
Surplus cereals and dairy foods were produced, and coffee and
tobacco were grown. Elephants roamed the lower country of the
Gibe and Omo valleys, and the Rift Valley lakes, as also in Kaffa
and Walamo. There were civet cats, valuable for their musk-like
perfume. Gold was produced in the BanI Shanqul region which
later became part of Wallaga, but was until the Mahdiyya still
under Turco-Egyptian sovereignty.
The other reason why the economic heartlands of the Horn
were situated so far in the interior was cultural or ideological. The
enslavement of Christians and Muslims by their co-religionists
was unlawful (in theory at least), and the greater part of the
slave-raiding in the Ethiopian region was carried out over the
pagan land to the south and west, mainly, as will be seen, by
Muslim slave-dealers. Hence most of the 'Abyssinian' slaves,
especially young girls, who continued to be in high demand in
17

Arabia and other parts of the Muslim world, were Oromo (Galla)
and Sidama in origin, rather than Amhara. Highland Ethiopia,
and the Oromo-Sidama lands, lacked salt, the main deposits of
which were in the desert country inhabited by Afar (Danakil), near
the Red Sea coast. The exchange of salt, and externally
manufactured goods, for the products and slaves of the interior
led to the development of regular caravan routes. Camels were
used as beasts of burden in the lowlands, donkeys on the plateau.
There were three main lines of access to these economic
heartlands, all passing over very difficult terrain. From the Sudan,
parallel routes ascended the Atbara valley to Tigre and Metemma
and the Blue Nile to Wallaga and Shoa. This last passed through
the gold-producing borderland of BanI Shanqul which was largely
Islamised and was frequented hyjalldba traders from the riverain
Sudan.7 Another line of access was from Massawa, up the
escarpment to Tigre, Gondar and Gojjam, and across the Abbai
(Blue Nile) into the country of the Oromo and the Sidama. The
third was from ports on the Gulf of Aden either up the Awash
valley, through Afar country, or across Somali lands to Harar and
thence over the Chercher highlands. There were two revivals of
trade in the nineteenth century. The first followed the general
pacification of the Red Sea area by Muhammad ' All Pasha, and
was accompanied in the south by the rise to power of the Oromo
(Galla) kingdom of Innarya. The second followed from the
opening of the Suez Canal in 1869, and the increasing use of
steamships, first in the long-haul and then in short-distance
transportation. The second boost coincided with the revival of
Shoa under Menelik, who became its ruler in 1865. Shoan
expansion into Somali, Gurage, Sidama and Oromo lands, during
the whole of the rest of the century, was a direct response to the
economic resources and potential of the south-western regions.
By the last third of the nineteenth century, the trans-Saharan
routes carried generally similar staple and other goods - the
distribution varied according to time and circumstance. From the
Sahelian and Sudanic countries came ivory, tanned skins (mainly
of goat), ostrich plumes, gold dust, gum, resin, slaves, some cloth
and leather work, and kola nuts. The Ottoman government had
abolished the slave trade in 1857, except to the holy cities of Hijaz,
7
Alessandro Triulzi, 'Trade, Islam and the Mahdia in north-western Wallaga,
Ethiopia',/. Afr. Hist., 1975, 16, 1, 55-71.
18

but the Moroccan authorities resisted European pressure for

abolition and emancipation until the French occupation in 1912.
There is no doubt that some slaves continued to be transported
across the desert, the greatest number, towards the end of the
century, on the Wadai—Benghazi route. Black slaves were also
used extensively in the desert, working in extremely harsh
conditions in the rock-salt mines, and on the date plantations and
vegetable gardens of the oases. When the supply of slaves finally
dried up, towards the end of the century, the desert economy was
gravely disrupted.
From the north came cotton cloth, some woollens and satins,
silks, waste yarn to be made up in Sudanic Africa, guns, horse
trappings, scents and perfumes, pepper and spices, needles, tea and
sugar, cups and saucers, and writing paper. The greater part of
these imports were produced in western or central European
countries. Manchester cotton goods amounted to three quarters
of the total. Woollens came from Austria, satins from Bohemia,
raw silk from the waste of the Lyons mills, prayer carpets from
Dundee, sugar from Austria, tea from China (via Malta), and guns
from every arms manufacturer in Europe and the United States.
Taking all of the trade into account, British manufacturers
dominated the desert-side imports to the Sudan. It appears that
this was increasingly so as the century drew to a close, particularly
in the case of the crucial Tripoli entrepot. In the 1850s Barth
considered that British cottons were less than an eighth of the
total; in 1891 the British consul in Tripoli estimated that goods
for the Sudan were worth £100,000, of which over £70,000
consisted of Manchester cottons. At Mogador, the southernmost
Moroccan port at which Europeans and their agents were allowed
to operate, and the gateway for commerce to the western Sudan,
the British trading peak was in the 1860s and 1870s, but even after
this the annual value of British trade was often worth twice that
of French trade.8
To a far greater extent than in the Horn, the staple items in
the long-distance trans-Saharan trade were dependent upon the
fluctuations of foreign markets. The European and American
fashion boom in ostrich feathers lasted barely two decades, the
1870s and 1880s, the main European market being, predictably,
8
Jean-Louis Miege, Documents ctbistoin iconomiqut et social marocaine au XIXe siicle
(Paris, 1969), 134-7.
»9

Paris. The demand for ivory from the woodland savanna between
Cameroun, the upper Ubangi and the Bahr al-Ghazal (as from the
other ivory regions of tropical Africa) was sustained at a high level
throughout the second half of the century, and here it seems that
London was the chief, but by no means only, staple market. The
trade in tanned and dyed goatskins reached its peak in the later
1880s, the skin being used for women's footwear. Most of the
market was in the United States. The trade began to decline with
the adoption in America of the chrome-tanning process, and a
growing preference for undyed skins. The widespread and deep
recession in most of the European and North American capitalist
economies, which reached its nadir between 1878 and 1884, had
a damaging effect on much trans-Saharan economic activity. In
particular it profoundly affected the Moroccan political economy.
In a set of highly complex causal relationships, the great recession
was a key element in determining the European partition of
Africa. But recent research has shown how resourceful African
producers were in switching to different trade items when demand
fell off, and equally how resourceful merchants were in rapidly
adapting to new means of transportation, such as the railway and
the parcel post.9
European commercial firms involved in the trans-Saharan trade
tended to be represented in North Africa by agents rather then
by company branches. Many of these agents were Jews. Most
British commercial activity was based on its Mediterranean
possessions - Gibraltar in the case of Morocco, and Malta as
regards Tunis and Tripoli - and Gibraltarians and Maltese (some
of whom were Jews) worked as agents for British firms. The large
Jewish communities in Tripoli and Tunisa lived reasonably
peaceably with their Muslim neighbours while others had done
so in Algeria until Napoleon Ill's senatus consulte of 1865 and the
Cremieux decree of 1870 embittered communal relations.10 But
in Morocco there was much harassment and some persecution of
the local Jews. Nevertheless, most of the trans-Saharan commerce
at Mogador was in the hands of Jews, who made up well over
one-third of the population of about 18,000 in the 1870s. Hostility
' Johnson, 'Calico caravans', 108; Marion Johnson, 'By ship or by camel: the
struggle for the Cameroons ivory trade in the nineteenth century', J. Afr. Hist., 1978,
19, 4, JJ9-49; Jean-Louis Miege, Le Maroc tt /"Europe, 1880-1894 (Paris, 1962), m,
375-459-
10
See below, pp. 22 and 28.
2O

by Muslim Moroccans would seem to have been in an inverse

proportion to Jewish prosperity.
The caravans from Morocco to the western Sudan were
operated by Moroccans from the oases of the Wadi Sous, and
Wadi Dar'a and Tafilelt, and Berbers from the Tuat oases, with
some assistance from Chaamba Arabs. Further south, Tuareg
took over. Political control over Timbuktu was contested between
Tuareg chiefs and Arab shaykhs of the Kunta family, which had
also migrated from the desert. The Kunta collaborated with Fulani
from Masina in a revolt against the domination of al-Hajj 'Umar
in 1863, but Tuareg took advantage of their success, and
re-established control over Timbuktu, which they held until the
French occupation in 1894. A council of notables conducted the
day-to-day business of the city, and its commercial life returned
to normal after the disruptions of the jihad of al-Hajj 'Umar.
Members of the Moroccan Jewish community managed to set
themselves up in Timbuktu soon after the revolt of 1863, for the
first time in the city's history. Moroccan policy in the Sahara was
expressed mainly in religious and strategic terms, in relation to
Morocco's position in the world of Islam, and to its aggressive
and hectoring European neighbour in Algeria. But behind these
declared aims lay the requirement to secure and control the
lucrative trade.
If Moroccan traders and their clients in the desert played a
major role in the Timbuktu and other western Sudanese com-
merce, that between Tripoli and Hausaland was dominated by
merchants from the oases of Ghadames. By the second half of the
nineteenth century, Ghadamasi firms were to be found in Tripoli
and Tunis, where they were involved not only with the Sudanese
trade of those cities, but also in trade between them. They had
also spread across the Sahara into the central Sudan and as far
south as Nupe and Adamawa. These Ghadamasi merchants,
established all over the trading network, facilitated the exchange
of credit and provided brokerage; arranged the collection, storage
and distribution of goods, and arranged for their transport. At
each relay point Ghadamasi merchants contracted the local
Tuareg groups who provided transport animals and armed escorts
for the caravans. The leading Ghadamasi had marriage ties with
the Tuareg nobility and also with the notable Hausa families,
which greatly eased commercial transactions. In northern Hausa-
21

land, Ghadamasi only became commercially dominant towards

the end of the nineteenth century. Previously Moroccan and
Fezzani merchants had played a major role in the international
trade. The Ghadamasi were able to take full advantage of the
boom in the export of ostrich feathers and tanned goat skins,
which was at a peak in the 1870s and 1880s, and they ousted
longer-established Magribi merchants on the central Saharan
routes.
As in Morocco, most of the merchant houses in Tripoli were
owned by Jews, but with a significant difference; whereas the
Moroccan Jews were indigenous, many of the Tripoli Jews were
outsiders — the majority from Italy. Indeed, these were probably
the first important group of Italians domiciled in Tripoli. The only
serious competitors of the Jewish merchants were Maltese, who
were British subjects. These Jewish and Maltese commercial
houses established in Tripoli (and later in Benghazi) were in direct
contact with European financiers and manufacturers, and these
were predominantly British. Thus, the sources of finance for the
trans-Saharan trade, the production of goods imported into Africa
and the chief markets for African produce, were all firmly situated
in capitalist Europe and, to a lesser extent, North America.
The operation of the trans-Saharan trade on the most recently
opened route, between Wadai and Benghazi, was in the hands
of Mejabra and Zuwaya Arabs. But in this instance a religious
institution provided much of the support system which made the
hazardous trading network possible: the SanusI tariqa or
brotherhood. By the 1870s and 1880s the Sanusiyya was at the
height of its power and influence. Muhammad al-SanusI had died
in 1858, and was succeeded as head of the order by his son Sldl
al-Mahdi. There is some irony in the constant concern of the tariqa
to avoid European and Ottoman influence and interference by
withdrawing its headquarters even further away from the Medi-
terranean and North Africa, and its distinctive role in promoting
international trade which was so heavily dependent on European
capitalism. The SanusI headquarters were moved from al-Baida,
near the Cyrenaican coast, to the oases of Jaghbub in 1856, to
Kufra in 1895 and to Quru in 1899. In fact these moves
represented a shift in the economic interests of the Sanusiyya, as
it became more heavily involved in the trans-Saharan trading
system. The chief sphere of its political influence remained in
22

Cyrenaica, where the tariqa formed the de facto government
among the bedouin groups, and where a number of the Ottoman
officials became members of the order. New %aribas were, however,
set up ever further south, all over the trading network, not only
in the desert oases, but in the Sahelian and Sudanic states, from
Darfur in the east to Timbuktu in the west. Many of the leading
Ghadamasi merchants became members of the tariqa, as did the
rulers of Wadai. In northern Hausaland, on the other hand, the
SanusI brotherhood was distinctly for foreigners, merchants and
artisans. The Sanusiyya was responsible for maintaining a con-
siderable degree of stability among the factious Saharan groups
whose quarrels had previously disrupted the trade. The order
intervened in disputes between the Zuwaya and their neighbours,
the Awlad Sulayman, and the non-Muslim Bideyat and Tubu. The
Sanusiyya, while remaining ideologically and politically hostile to
European influences, actively promoted the European-orientated
commerce of much of central northern Africa.
The equivalent of the Ghadamasi in the huge eastern section
of the central Sudan - Darfur, Kordofan, the Bahr al-Ghazal -
were northern Sudanese, mainly Danaqla (arabised Nubians from
the Dongola region) and Ja'aliyyun, known collectively zsjalldba.
These had followed close on the heels of the European traders,
who took advantage of the forcing of the White Nile by
Turco-Egyptian expeditions in the 1840s. This provided the
opportunity for the frontier of economic exploitation to move
south from southern Darfur, the Nuba mountains, the Blue Nile
and the Ethiopian marches, deep into Equatoria and the Bahr
al-Ghazal. Thereafter, the immense plain stretching from the Nile
westwards and southwards, towards the Congo watershed and
beyond Darfur to Wadai, became a huge ivory frontier, open to
the northern traders. These established their own settlements,
daym, or fortified enclosures, âriba, in the midst of Nilotic peoples
such as the Dinka, Nuer and Bari, the Eastern Nigritic Azande and
Banda, and the Central Sudanic Kreish, Bongo and Sara. Only the
Dinka and the Azande had political structures strong or flexible
enough to resist for long the exploitative impact of the Muslim
traders. Some of these traders achieved the status of merchant
princes, such as al-Zubayr Rahma Mansiir, but most operated on
a smaller scale. Sudanesejalldba moved steadily westwards, setting
themselves up in Darfur, Wadai, Kanem and Bornu. This was yet
23

another example, in the Saharan and Sahelian context, of a trading
diaspora. By the 1870s jalldba had reached as far south as the
Mbomu and Uele rivers, and had penetrated Mangbetu country
in the Zaire forest.
The main article of trade was ivory. This poured in large
quantities into Egypt, where it was purchased by European and
American dealers operating from the safety of Cairo and
Alexandria. The working capital for the traders seems to have
been provided (at usurious rates) by capitalists - mainly Egyptian
Copts — in Khartoum. Throughout tropical Africa the trade in
ivory reached a peak in the 1870s and 1880s. It was not the most
valuable item of trade to come out of Africa: Egyptian cotton,
West African palm oil, South African diamonds and latterly gold,
to name but four others, brought larger overall profits. The ivory
trade, however, was the most widespread. Elephants were hunted
for their tusks from eastern Nigeria and Cameroun eastwards to
the Nile valley and the Horn, and southwards through Gabon to
the lower Congo; they were hunted from the Horn southwards
throughout East Africa, and across the whole belt of middle and
south Africa from the Congo and Rovuma rivers, southwards
over the Zambezi to the Orange river and the Drakensberg
mountains. The pursuit of ivory was generally destructive and
often violent. It involved the organisation of hunting bands,
which even when developed locally, were disruptive of older
social groupings. Too frequently, it involved raiding across the
countryside by alien gangs. By the last third of the century,
elephants were nearly always hunted with firearms. It was the use
of European guns which transformed a series of minor local
activities into an exploitative international trading system. The
European and American demand was for ivory to make knife
handles, piano keys, combs and billiard balls. It has been estimated
that 65,000 elephants were killed annually in various parts of
tropical Africa in response to this demand."
As far as the ivory frontiers of the Nilotic Sudan were
concerned, attempts were made by merchants to build up a
balanced trading system, by importing cloth and other European
goods to be used in exchange for ivory. But conditions in the Bahr
al-Ghazal, along the Congo watershed and in Equatoria were too
unstable and societies were too inflexible for the more usual and
11
Marion Johnson, 'By Ship or Camel', 548, n.41.
24

THE MAGHRIB AND EGYPT
profitable trading patterns to develop. Violence became the easy

resort. Cattle were necessary not only to feed the traders but also
to provide inducements to reluctant tribesmen. The traders
therefore raided cattle and stole food crops. It was easier to use
slave labour rather than free. Labour was required to take part
in the elephant hunts, to run the fortified %ariba> and to work on
the riverain farms of thejalldba in the northern Sudan. A vigorous
trade in slaves developed.
Regular government under the Turco-Egyptians was even
more parasitic than were the traders. By 1880 soldiers, officials,
traders, private retainers, their wives, concubines, children,
servants, servants' families, etc., were so numerous (not less than
50,000, perhaps nearer 100,000) that there was something like a
colony of settlement in the southern Sudan.
The Bari in southern Equatoria first co-operated and then,
embittered by the tactics of the traders, resisted. During the 1860s
the Zande chiefdoms entered into a series of uneasy relationships
with the Khartoumers, but by the 1870s they were succumbing
to the onslaughts of al-Zubayr Rahma Mansiir and other Ja/Idba.
The Turco-Egyptians attempted to conquer central Zandeland in
1878. Later, Mahdist forces also attacked, but by then the Zande
were armed with guns, and could defend themselves. These
weapons came into Zandeland from the Atlantic coast, via the
Congo river system. Towards the end of the 1880s the Zande were
being raided by slave traders from the east coast, ' so that Zande
country became the place where all three trade networks finally
met'.12
The situation of Egypt and the Maghrib countries and their

response to European influence and interference, and t o ' modern-
isation ' in general, varied considerably at the beginning of our
period. They all shared a Muslim background, but the kinds of
Islam they had experienced were diverse. Politically, Morocco was
an independent sultanate, and indeed had been so since the eighth
century. Algeria was a French colony, of which the settled areas
were administered as overseas departments of metropolitan
France, while the pastoral areas further inland were subject to
11
Philip Curtin, Steven Feierman, Leonard Thompson and Jan Vansina, African
History (London, 1978), 441.
25

military rule. Tunisia was a beylicate, over which the Ottoman
sultan exercised a nominal suzerainty, but which was in practice
largely independent. Tripolitania and Cyrenaica were provinces
of the Ottoman empire. Egypt was a kingdom (khedivate),
politically independent, over which Ottoman suzerainty was even
more nominal than over Tunisia. But, overshadowing the political
processes of nineteenth-century North African states were a series
of economic revolutions which, in various degrees, reduced the
local economies to dependence upon the European capitalist
system.
The Spanish invasion of Morocco in 1859 and the capture of
the northern city of Tetuan the following year, had important
consequences for the sultanate. The price of withdrawal by Spain
was an indemnity of 100 million pesetas. Such a sum in specie was
quite beyond the means of the Moroccan economy, and the
sultan's government, the makhîn, was forced to borrow the
money from British financial houses. The repayment of this loan
gravely distorted the Moroccan economy - a large proportion of
the country's customs receipts for the next twenty-five years was
used for it. This aggravated an already difficult financial position.
European economic penetration had slowly mounted since the
1840s, and already by i860 the balance of trade had shifted from
Moroccan to European interests. This tendency continued, as did
inflation of the complex Moroccan currency, and the consequent
rise in price of basic foodstuffs and goods. Morocco's principal
exports to Europe — wool, hides and wheat — were undermined
by international competition (a consequence of the increased use
of steamships and railways). By the last quarter of the century
imports from Europe greatly exceeded exports. Cheap Manchester
cottons flooded the Moroccan market, and many artisans in the
indigenous textile industry, in towns such as Larache, Sale and
Rabat, were ruined. Natural disasters, such as periodic famines
and epidemics, exacerbated the effects of the economic crisis. One
immediate consequence was the drift of population from the rural
areas to the towns, in particular to the coastal cities, which grew
rapidly in size. The loss of population disturbed the political
machinery which regulated the relations between government and
the interior tribal groups, particularly in the Atlas mountains — the
relationship of the makhîn with what was called the bilad al-sibd
('the country of dissidence').
26

As well as the insidious penetration of European commerce,

in the 1870s still predominantly British, there was the more direct
threat of foreign invasion. On the wider diplomatic front, Mawlay
al-Hasan was able to forestall intervention by balancing the
contending concerns of the European powers. His policy was to
keep to a minimum the concessions necessary to pacify the
Europeans and to take such advantage as was possible of their
mutual rivalry. Britain wanted a reformed, independent Morocco
to safeguard the southern shore of the Strait of Gibraltar. Spain
generally wanted the same, but solely to prevent French inter-
ference. France was in favour of a weak, crumbling, Morocco, so
that when the time came, she could step in and pick up the pieces.
The immediate points at issue between Morocco and the European
governments were the position of European protected persons
and the immunities these enjoyed from the enforcement of
Moroccan law; and the European demands for favoured nation
treatment, which would enable each country the more easily to
penetrate the Moroccan economy. In regulations similar to the
capitulations of the Ottoman empire, the Moroccan government
had restricted access of Europeans to Tangier and Mogador, but
agents - Gibraltarians and Jews, as well as Muslim Moroccans —
could operate more widely and more freely and were seen as a
threat to the political authority of the state, as well as to its
economy.13
In September 1870 the French Second Empire collapsed with
the fall of Napoleon III, and early in 1871 a series of violent
uprisings broke out in eastern Algeria, particularly Kabylia, the
most vehement of which was led by the wealthy landowner,
Muhammad al-Muqranl. The causes of this revolt were complex
and are examined in an earlier volume.14 For present purposes,
the vital point of the revolt was that it failed, disastrously. Only
about one-third of Algeria was affected. Al-Muqranl's rebellion
was not the signal for a general uprising against the French. The
revolt was late in getting off the ground - when it broke out,
fighting had stopped in France, and the new French republican
government was able to divert troops to deal with it. After it was
suppressed (with the usual French colonial ferocity), much of the
13
F. V. Parsons, Tie origins of the Morocco Question, 18S0-1900 (London, 1976); also
Frank E. Trout, Morocco's Saharan frontiers (Geneva, 1969).
14
Douglas Johnson, "The Maghrib' in John E. Flint (ed.), T*» Cambridge history of
Africa, v: c. 1790 to c. 1870 (Cambridge, 1976), 118-19.
27

land of the rebels was sequestered. Algerians could regain their

land only at prices which few could afford, and the French settlers
were ready and able to take advantage of the situation. Further
land expropriation followed in the 1870s and 1880s.
Governments of the Third Republic, like those of the Second,
were more sympathetic to the demands and aspirations of the
settlers than was the regime of Napoleon III. The administrative
apparatus first instituted by the Second Republic remained in
force, as did the highly ambivalent senatus consulte of Napoleon III.
This had extended French citizenship to Muslim Algerians if they
renounced their Muslim legal status. Few Algerians were either
able or willing to take advantage of this assimilationist device. In
1870 Jews were given no choice; by the Cremieux decree they
became French citizens. Administratively, coastal Algeria - that
part most densely settled by Europeans — was divided into three
departments, equal in most respects to French mainland depart-
ments. Algerian affairs were dealt with by an appropriate ministry
in Paris. The Third Republic introduced important changes in
local government. The first actions of the new regime towards
Algeria was to lessen the power of the army in the management
of its affairs and to reduce the authority of the military bureaux
arabes, which controlled the predominantly Muslim regions. The
suppression of the 1871 revolt led, paradoxically, to an increase
in civil administration and control.15
The quelling of the al-Muqranl revolt, the subsequent land
consolidation, the gradual increase of the authority and legality
of the Third Republic, and the emergence from the recession of
the late 1870s and early 1880s, ushered in the hey-day of settler
colonialism in Algeria. Republican politicians with their liberal
democratic and their nationalistic ideals, were faced with some-
thing of a dilemma, similar to the apparent dilemma that
confronted their counterparts in Britain in regard to South Africa.
They could not deny the full exercise of citizen rights to
Frenchmen living across the Mediterranean, even when this
exercise meant the withholding of rights to Algerians. Besides this
ideological sympathy, the settlers built up powerful lobbies which
often were able to operate most successfully in the flux of
metropolitan politics. The settlers then, were assured of the
15
Wilfrid Knapp, North West Africa: a political and economic survey (Oxford, 1977),
67.
28

support of French governments. The rigours of the first decades

of colonisation, when Algeria was very much a frontier situation,
subsided. Life for many settlers became comfortable and well
organised. Some of them prospered and became wealthy. They
behaved with the apparent self-confidence of petits blancs on the
make. Both the senatus consulte of 1865 and the land law of 1873
favoured individual ownership of property on French lines as
against the various Muslim land systems, including communal
land-holding. There was prolonged drought from 1866 to 1870,
which hit all agriculturalists in Algeria; it was one of the
immediate causes of the al-Muqranl outbreak. The settlers
recovered more easily than the Algerians, especially as more land
became available to them as a result of the revolt. Between 1871
and 1880 the land in settler hands almost doubled, reaching
800,000 hectares as against 480,000 in the period 1830—70.
By the 1870s Algeria had entered an era which Jacques Berque
described half a century later, and which had many echoes of
settlerdom in South Africa:
The most provocative symbol of the colonial period in the Maghrib is that
of the titled farmhouse, a cheerful dwelling standing amid vineyards. It aroused
the" most violent, and violently opposed, reactions from Frenchmen and the
people of the Maghrib. The fact that it was surrounded by more significant
forces matters little; it implied all the rest. Banks, military camps, factories and
schools may have played at least as important a part, but none made so deep
an impression on everyone's feelings as this French farmstead, this heraldic
emblem on African soil."
Muslim Algerians benefited very little from the amenities of the

colonial state. Apart from obvious religious, cultural and linguistic
differences, they had a different and separate legal status, and
enjoyed only minimal civil rights. But they contributed in various
forms of taxation - both traditional Muslim and the new French
colonial taxes - a wholly disproportionate amount of the revenue
of the colony. Algerians had to bear most of the burdens of alien
rule but gained from few of its improvements. Apart from a small
handful, Algerians were educated in the 1870s and 1880s in
traditional Muslim schools, not in the more modern French
schools: this handful did, however, form the core of the new elite.
Algerians were conscripted into the French army in 1870, and
were again to be so treated in 1914. The old landed aristocracy
16
Jacques Berque, French North Africa ( L o n d o n , 1967), 35.

were reduced to little more than peasant farmers; peasants were
pauperised on the land, or moved into the rapidly growing towns
as casual labourers, beggars and criminals. Their numbers began,
by the 1880s, to increase fairly rapidly, as public health measures
improved. The Algerian petty bourgeoisie, in the towns and the
countryside, found niches in the structures of the colonial state,
and some did quite well for themselves. Characteristically, it was
members of this group, who also took advantage of the few
modern educational facilities which were available for Algerians,
who slowly emerged as political leaders, both within the colonial
state and in opposition to it. These were the early stages of a classic
situation of deprivation examined by Franz Fanon; the
confrontation of what a perceptive scholar of the Maghrib has
termed ' anarchic society' with the French imperialist state.
On the one side was an aggressive, dynamic, insatiable Europe... On the other,
North Africa was neither barbarous nor uncultured... careful study reveals a
perfect coherence between the spiritual, economic and social aspects of its
culture. That culture was however, different from Europe's, particularly in that
nothing arose within it to stimulate the society as a whole. Its population
growth was too late and too abrupt; there was not enough vitality in urban
crafts or foreign commerce; and there was no intellectual renewal. The
Maghrib lived at a slower pace, and this no doubt was one of the reasons for
its colonization... 17
Algeria, however, was not just a matter of three overseas French
departments with a French settler population alongside an
indigenous one. It was the forcing ground for much subsequent
French imperialism in Africa. By the end of the 1870s, especially
after the Congress of Berlin in 1878, which indirectly and
unofficially recognised Tunisia as a French sphere of influence,
France had regained international status and confidence, and plans
for expansion from the Algerian base toward the western Sudan
began to be seriously considered. These centred around designs
for the construction of a trans-Saharan railway, which was mooted
by geographers and engineers in France and Algeria from about
1875. By 1879 Freycinet, minister of Public Works, and
Jaureguiberry, minister of Marine (responsible for colonies) were
to become enthusiastic supporters of the trans-Saharieti, and
provided the scheme with the stamp of official support.' 8
17
Lucctte Valensi, On the eve of colonialism: North Africa before the French conquest
(London, 1977), 80.
18
Kanya-Forstner, Conquest of the Western Sudan, 65.
3°

In 1873 the Circassian mamluk, Khayr al-Dln, was appointed

Chief Minister of the government of Muhammad al-Sadiq, the bey
of Tunis. This move was partly in response to European,
especially French, pressure. Khayr al-Dln initiated a series of
thorough-going reforms. He belonged to the establishment elite
of the Ottoman empire, and he attempted on the smaller scale of
Tunisia what the Tanzimat reforms had set out to achieve for the
empire as a whole. Central and local administration, and the
complex taxation system, were overhauled, and every effort was
made to make these operate more effectively. Army reforms,
which had first been put into motion in the 1840s, were pressed
forward. Efforts were made to modernise the economy, but these
largely resulted in the continued growth of foreign-owned
concerns, such as railways, telegraph lines, textile factories and
large agricultural estates. Nevertheless, the Tunisian economy in
the 1870s was more prosperous than it had been, at least since the
abolution of slavery in 1846.
Perhaps the most crucial of Khayr al-Dln's reforms were in
education. At the peak of his secular school system was the Sadiqi
College, a school to train civil servants and members of the liberal
professions, with a curriculum which included French and Italian,
mathematics and science. This college and its pupils were to have
a lasting effect on the course of Tunisian history during the
protectorate period. But in all his efforts at reform Khayr al-Dln
was hamstrung by the restraints laid upon Tunisia by the
International Financial Commission, which had been imposed in
1869. During the years following the suppression of the revolt
of 1864, the bey's government had piled up an enormous debt to
European financial interests and had become bankrupt. The
commission, made up of representatives of French, Italian and
British bankers, took over control of the beylicate's finances,
and to all intents and purposes, Tunisia lost its economic
independence.
The intrigues of members of the commission, and internal
opposition to Khayr al-Dln, led to the minister's fall from office
in 1877 (he retired to Istanbul and became for a time grand vizir
to the sultan). By this time, the conflicting interests of Britain,
France and Italy brought a series of crises to Tunisia: if the
scramble for the rest of the continent had yet to unfold, that for
the old Roman province of Africa was well under way.
31

Between Tunisia and Egypt lay Tripolitania and Cyrenaica,

which had been ruled directly by the Ottoman sultan since the
fall of the Karamanli dynasty in 1835. In Cyrenaica the SanusI
brotherhood exercised almost undisturbed religious and political
authority. Cattle were exported from Tripoli to Benghazi, as were
sponges and the trade goods of the western and central Sudan.
In the late 1860s, esparto grass was developed as a new export
crop, which went mainly to Britain for paper-making. But
economic activity was on a low scale. Italian steamship lines
handled most of the export trade. Apart from the katatib, the
Koranic schools, and the SanusI âwtya, the only educational
establishments were a handful of Italian and Jewish schools. The
Ottomans were present in Tripolitania and Cyrenaica as colonial
rulers. They used techniques of 'indirect rule' not unlike those
of some later European imperialists.
Egypt was in the 1870s still the wealthiest of all African
countries. Since the agricultural revolution of the neolithic age,
Egypt had maintained this position, and if by this time it had any
rival, it was the British colony of the Cape of Good Hope. In 1875
Egyptian exports and imports were worth £13.3 million and £5.6
million respectively; figures for the Cape Colony were both £5.7
million. Government revenue in that year was £10 million in
Egypt and £1.6 million in the Cape. In 1875 exports to Britain
from what was termed the West Coast of Africa were valued at
£1.7 million, and it has been estimated that the total trade from
the western Sudan and the Maghrib - mainly Morocco and
Tripoli - in the same year was £1.5 million.19 The total Egyptian
indebtedness to European interests in 1876 was nearly £100
million — possibly an amount greater than the sum total of the
fiscal value of the wealth of the whole of the rest of the continent.
Roughly i o million people lived densely packed in the immensely
fertile countryside and towns of the Nile delta and valley. The
scale and magnitude of the political economy of Egypt put it on
a par with European and Asian countries rather than those in
Africa.
The first attempt to set up a modern state structure to utilise
and develop the potential of this political economy occurred
during the reign of Muhammad 'All; the second, and in many
respects more radical restructuring was the work of the govern-
" P. M. Holt, 'Egypt and the Nile valley', in Cambridge history of Africa, v, 59-50.
32

ments of Khedive Isma'Il. This is not the place to detail the

reforms and other changes introduced by the khedive in the 186os
and 1870s, in commerce, industry and the agricultural economy;
in local government and education; in the institutions and
policies of the state; and in the army.20 The results of those
reforms were manifold. Legally, and in theory, the fallahin,
peasant farmers, benefited from legislation brought in by Sa'id
and Isma'Il. They obtained individual rather than collective rights
to their land holdings. Isma'Il's muqabala law of 1871 granted full
rights of ownership. But rural society was becoming much more
differentiated. Big landowners other than the ruler himself had
emerged, and there was also increasing differentiation among the
fallahin themselves.
There was massive migration of poorer fallahin to the towns,
if for no other reason than to escape the payment of increasingly
burdensome taxation. The importance of cash crops, especially
cotton, was another factor in this differentiation. Peasant land was
heavily taxed, while the great estates escaped lightly. Military
conscription and forced labour added to rural pauperisation. One
group of peasants, however, achieved a kind of kulak status.
These were the families of village headmen {shaykh al-bildd), who
became substantial land-owners. Sa'id and Isma'Il opened the
ranks of the officer corps to sons of these headmen, a measure not
entirely popular, but with far-reaching consequences. But the
village headmen were small fry compared to the rural aristocracy.
These were members of the viceregal family (not least Isma'Il
himself) and their proteges from amongst the Turco-Egyptian
establishment (Turks and Circassians) who amassed huge estates.
' By the time of Isma'Il, there had developed in Egypt something
like a Turkish-speaking landed aristocracy, the creation of the
Turkish-speaking dynasts.'21 There was also the emergence of a
' service class' and a (largely bureaucratic) petty bougeoisie, and
simultaneous relative eclipse of the traditional merchants and
'ulama" as important social groups. What was lacking was an
affluent and self-confident upper bourgeoisie, its functions being
largely usurped by foreigners.
Social differentiation was as apparent, if not more so, in the
towns and cities, in which a cosmopolitan class of nouveaux riches
10
P. M. Holt, Egypt and the b'trtile Cresctnt, IJ 16-1922 (London, 1966), 203.
Zl
Ibid., z i i .
33

inhabited the recently constructed, fashionable quarters of Cairo,

Alexandria and Port Said, while the urban poor lived as slum
dwellers in the ancient parts of the cities. The new quarters had
the appearance of European cities, and in particular had a
pronounced French stamp, as did many aspects of Turco-Egyptian
culture. Railways, postal services and telegraphs linked these
cities, as steamships linked Egypt with Europe and with near and
further Asia; after 1869 ships steamed direct from Europe via the
Suez canal to the east. Egypt had become one of the two or three
key communications areas in the world in the new era of
technological innovation in transport and production. Many
aspects of Egypt in the 1870s were as modern and as up-to-date
as anything elsewhere on the continent, or beyond. Yet there were
very serious contradictions and eccentricities in these develop-
ments. Little of the modernity was home-spun. Practically all the
infrastructure of the Turco-Egyptian state was owned and operated
by foreign concerns. The price in money and in power of Isma'Il's
policies was immense.
When Isma'Il became khedive in 1863, he inherited not only
a considerable debt from his uncle Muhammad Sa'Id (some £7
million, £3 million of it to one firm of European bankers), but
also a thriving economy. The exports of long-staple cotton
increased over six-fold between 1850 and 1865, by which time the
American civil war had put a premium on Egyptian cotton; its
value increased four times. But with the end of the civil war prices
suddenly collapsed, and the cotton boom ended. This was an
economic disaster, from which Egypt did not recover until well
into the period of the British occupation. Large numbers of big
land-owners and innumerable falldhtn were ruined. Faced with
high taxation and the threat of conscription into the army, there
was much political discontent among the falldhtn, which Isma'Il's
hesitant social and political reforms only partially alleviated. The
khedive introduced an assembly of delegates {majlis shura al-
nuwwab) in 1866, which was part of a system of checks and balances
which aimed to control the falldhm through the village headmen
and at the same time to keep an eye on the headmen themselves.
Very significant advances were made in the establishment of a
nationwide modern state educational system (schools were set up
for girls in 1873), so that by the end of the 1870s Egypt had
perhaps the most advanced schools on the continent.
34

The collapse of the cotton boom did not deter Isma'U's declared
aim of aggrandisement, both within Egypt and beyond. He was
forced to turn to extra-Egyptian sources of finance. In so doing,
he ran up huge foreign debts, the very high rates of interest of
which crippled whatever recuperative powers the Egyptian
economy possessed. It is perhaps not very remarkable that Isma'Il
utilised foreign capital in this manner or that European bankers
were so willing to aid and abet the aggrandisement schemes: this
was the normal modus operandi of risk capital. What has to be
questioned is the motive behind Isma'U's schemes, especially the
Turco-Egyptian expansion into tropical Africa. There is a
measure of sense in suggesting that the earlier Turco-Egyptian
expansion into the Nilotic Sudan in Muhammad 'All's time
represented the economic interests of the governing elite. But by
1850 Muhammad 'All's system of monopolies and compulsory
cultivation had collapsed. In the mid-1850s Muhammad Sa'Id
seriously contemplated abandoning the Sudan as unlikely ever to
pay for its own garrison and administration. After extensive
experience as a khedival governor in the 1870s and 1880s,
Gordon was to write it off as financially hopeless. The further
one moved away from the Khartoum—Gezira heartland, the more
economically hopeless did the whole empire become. Neither the
khedive's ostentatious ' modernisation' policies in Egypt, nor the
expansion of the empire, can be explained in socio-economic
terms. Isma'Il was motivated solely by self-aggrandisement. What
has to be explained, perhaps, is the nature of Turco-Egyptian
society that allowed him to get away with so much for so long.
Isma'Il had some of the means of imperial expansion at hand:
a fairly well equipped army, technological innovations — railways,
the telegraph - and expertise, either native Egyptian or foreign
hired. These he largely squandered. The Egyptian empire was
(metaphorically as well as literally) built on sand. Isma'Il lacked
the financial and human resources necessary to provide a long-term
administration which might have induced modest economic
development. The timing of the collapse of Isma'Il's empire was
set by the major financial and political crisis in Egypt, by the
Mahdiyya in the northern Sudan and by insurrection in the
Equatoria province.
35

ETHIOPIA AND THE HORN
In 1872 Dejazmach Kassa, governor of the northern province of

Tigre, was crowned king of kings of Ethiopia, Yohannes IV, after
a power struggle in the interregnum following the suicide of
Tewodros in 1868. The empire he ruled over was as politically
divided as it was ethnically, culturally and religiously diverse. It
was also relatively isolated, both geographically and ethno-
graphically, from the outside world. Ethiopia enjoyed the strength
and suffered the weakness of this isolation: the unifying influence
of a long (and literate) history, and the dead hand of the past. Most
of the emperor's subjects were highlanders, surrounded by alien
lowlanders inhabiting deserts, arid plains, swamps and sea coasts.
Only its border with the Turco-Egyptian province of the Sudan
in the north and north-west was in any sense a frontier. Here, a
Christian state and a Muslim state confronted one another with
no intermediary peoples. Between Massawa and the Takazze
(Tekeze) river, south of Kassala, the frontier ran across northern
Eritrea, the Ethiopian side keeping more or less to the mountains.
It then continued south to Gallabat, an important town on the
Atbara river close to Begemder province, and the gateway to
Gondar, and from there to the Blue Nile, dividing Gojjam from
Sennar.
Following the border region southwards, the Oromo (Galla)
of Wallaga were practically independent of the centre of the
empire; they were more closely involved with Shoa. This, the
southernmost province of the old Christian kingdom, was ringed
by a group of states or chiefdoms inhabited by western Cushitic-
speaking Sidama peoples, who practised intensive settled agri-
culture with ensete, the false banana, as the staple crop. This was
the wealthiest region of the Horn. From the seventeenth century
onward much of it had been overrun by Oromo pastoralists. Most
of the Sidama had passed under Oromo rule, while retaining their
Sidama languages and cultures, but some such as Kaffa and
Walamo, situated in the most mountainous and densely forested
areas, had retained their independence. Kaffa was an isolated,
nominally Christian and wealthy state, which resisted Shoan
expansion until nearly the end of the century. The Oromo were
eastern Cushitic-speaking peoples, like the Somali; it was the
pastoralists among them who had migrated over the surrounding
36

region, though many of these had later abandoned cattle-keeping

for settled agriculture. In the early nineteenth century Oromo-
ruled kingdoms emerged from the old Sidama states on the upper
Gibe river, the most important of which were Innarya and Jimma.
Under the influence of Afar and Harari merchants the two states
became islamised. Innarya was eclipsed in the 1850s and 1860s by
its rival Jimma, East of Innarya and Jimma were the Gurage
chiefdoms, some Muslim and others Christian, but all speaking
a Semitic language, and all heavily influenced by the Amhara.
Highland Ethiopia was in fact surrounded on all sides by
pastoral, nomadic peoples: Nilotes to the west, Cushites to the
south and east, Beja and Arabs to the north. The mainly Christian
peoples of the highlands were Semitic-speakers, of two main
groups, Amhara and Tigrinya, with smaller groups of other,
Cushitic-speaking peoples, such as the Agau, Damot and the
Judaic Falasha. Isolated from the main concentration of Semitic-
speakers were the Muslim Harari of the mountainous area east of
the Awash valley, on the borders of Afar and Somali country. The
city of Harar, originally the capital of a small Muslim emirate, had
grown to be the foremost trading centre of the Horn. But by the
nineteenth century it had become hemmed in by the settlement
of Oromo, who interfered with the commerce of the townsfolk.
There were about ten long-established provinces of the
Christian empire: in the far north, Tigre and Simien, then
Begemder, Amhara, Wag, Lasta, Yejju, Wollo, and in the south,
Gojjam and Shoa. The last, though once the central province of
the medieval Christian empire, had been practically severed from
the main body of Christian territory by Oromo infiltration up the
eastern side of the northern highlands. It was thus practically
independent of the central authority, and its young King Menelik
maintained his own links with the Europeans at Aden and the Red
Sea ports, and obtained from them the firearms necessary to
defend himself from his theoretical overlord as well as to practise
an imperial expansion of his own in the Oromo/Sidama states.
By the later nineteenth century the Oromo living under Christian
rule had mostly become Christians or Muslims, the others
remaining pagan. Many had become Amharic speakers. The
Oromo had penetrated mainly the eastern side of the highlands.
In the north, the Azebo Oromo lived in the eastern parts of Wag
and Lasta. Then there were the Yejju and Wollo Oromo, whose
37

Amharic-speaking aristocracy had played leading roles in the

troubled times of the eighteenth and early nineteenth centuries.
In the south very large groups of Oromo such as the Barentu and
the Mecha, had settled within and around the borders of Shoa,
and many of them therefore were Muslim. By the 1870s Ethiopia
was a multi-cultural, multi-religious, multi-ethnic/linguistic
empire; but in its ethos or ideology, in the thrust of its politics
and (very largely) in the personnel of its governmental and
religious establishment, it was Amharic and Christian.
Ethiopia was also an empire based upon Amhara and Amhara-
ised peasant farmers. Land ownership and land holding were
complex, and differed from one province to another. Some
peasants owned the land they tilled, and paid taxes to the
sovereign. Others were tenants on the lands of the nobility or of
the church, and paid taxes and other dues to these landlords.
Church ownership of land was permanent and no taxes were paid
to the state. Part of the programme of centralising reform
promulgated by Tewodros was an attempt to tax church lands;
the attempt failed, and brought down the wrath of the religious
establishment upon him — Yohannes and Menelik did not repeat
the attempt. 'Fiefs granted to other landholders were either
permanent and inheritable' - called rest land - ' or temporary' -
called gult land - ' burdened by a variety of obligations, such as
the collection of taxes and tribute and the raising of armies. In
the final resort the burdens were shifted to the tenants, who paid
taxes usually in kind or in the form of corvee labour'.22 The
political domain of the empire was made up of a smaller number
of great landlords, secular and ecclesiastical. During the eighteenth
century these had been the real rulers of Ethiopia, and their power
did not greatly diminish during the reigns of the later nineteenth-
century emperors. Tewodros tried to force them into submission;
Yohannes and Menelik, more wisely, used diplomacy to achieve
a measure of co-operation. Both Tewodros and Yohannes, in their
different ways, were keen ideologues, proclaiming the glories of
the Solomonic empire and the crusade of Christian Ethiopia
against the Muslim 'Turks and Ishmaelites'.
Of all the provinces, Shoa was the most independent, a vassal
state of the empire. It was distant from the northerly centres of
power, and because it was close to the wealthiest regions of the
22
Sven Rubenson, 'Ethiopia and the Horn', in Cambridge history of Africa, v, 70.
38

Horn of Africa, it was in a position to benefit economically from

the rich trade from the south-western interior to the coast.
Menelik astutely used the economic circumstances of Shoa to
strengthen and expand what he considered to be his kingdom.
Neither Yohannes nor Menelik conceived of any political or
economic threat from European states in the 1870s and early
18 80s, and all the evidence suggests that their appreciation was
correct. The threat, as far as the emperor was concerned, came
from the Turco-Egyptians, both materially and ideologically.
Islam was feared not only as an external and inimical religion, but
also as a source of severe internal strain. A number of the great
territorial magnates were secret (or even open) Muslims, and some
elements of the army could not be relied upon. Although an
examination of the political economy of the Turco-Egyptian
expansion from Egypt in the nineteenth century leads one to
question the reality of this threat, there can be no doubt that, to
Isma'Il and the members of the Turco-Egyptian establishment,
encirclement of, encroachment upon and subversion of Ethiopia
was a serious policy. Hence Yohannes positively courted
European intervention, as Tewodros had done before him. These
repeated blandishments towards the European powers were as
repeatedly rebuffed. Politically, no European states were con-
cerned to intervene in the Horn of Africa in the 1870s.
Commercially there was some interest, particularly in supplying
arms and ammunition. Menelik, being a southerner, had not the
responsibilities of Yohannes in defending the empire from the
Turco-Egyptians, but had his own imperatives — the expansion of
Shoa and control of the trade of the south-west — which in this
period made him compete with Yohannes for European support.
This pro-European (in Ethiopian eyes, pro-Christian) alignment
was to change dramatically at the end of the 1880s, when
European imperialist moves in the Red Sea and Gulf of Aden
forced Yohannes and Menelik to switch policy, and to attempt
to gain the co-operation of the Mahdists in the Sudan against the
Europeans. In the earlier decade, Ethiopia was concerned to break
out of its dangerous (as well as glorious) isolation, imposed in
part by geography and in part by Turco-Egyptian, Muslim
encirclement.
39

WEST AFRICA
In the remarkably uniform ecological zones of West Africa the

patterns of economic production and trade on the one hand, and
political development on the other, had by the 1870s undergone
a century or more of rapid change. The 'desert-side' trading
concerns of Sahelian countries merged with a great variety of local
trades across the Sudanese savanna and into the more thickly
wooded lands to the south. Long-distance continental trade
flowing across the Sahara penetrated deep into the central and
western Sudan and even into the forest belt. Again, the economic
impact of the long-established European trading enclaves was
being felt several hundred kilometres inland from the coast, all
the way from the Senegal to the Bight of Biafra. The political
configuration of West Africa was, by the third quarter of the
nineteenth century, immensely varied. If there was a tendency or
pattern, it was rather in the direction of agglomeration (economic,
military, demographic, and even administrative) but many com-
munities remained very small.
The coasts of West Africa had, of course, attracted the attention
of Europeans since the fifteenth century, yet the predominant
feature for most of the region, in the 1870s, was the small and
restricted degree of European influence. Unlike some other parts
of tropical Africa, this was not a case of European neglect of
economically unprofitable lands. On the contrary, it was the
richness and diversity of the economies of West Africa which gave
its societies the strength to withstand European influence. During
the succeeding two decades, it was information about this wealth
that was one factor at least which impelled France, Britain and
Germany to partition. The fact that these alien colonisers did not
reap a great deal of economic benefit or profit from their hastily
acquired possessions is profoundly indicative of the nature of
colonial rule and systems, not of the economies of West Africa.
The richness and diversity of these economies were not an
illusion: the Europeans failed to understand in a general manner
how they worked, and in particular they failed to grasp that they
were dealing with indigenous varieties of economic operations
which, while often being themselves exploitative, could be more
intensively exploited only at the peril of their being extinguished.
On the eve of partition, even from a wholly Eurocentric point
40

WEST AFRICA
of view — that of the impatient colonels in St Louis, the harassed
British consul on Fernando Po, or the combative traders in the
Niger delta - what was apparent, and a matter of grave concern,
were West African strengths, not weaknesses, assertiveness, not
passivity.
In the context of long- and middle-distance trade, West Africa
can be divided conveniently into two great regions, east and west,
to which the old European geographical terms can usefully be
applied: Central Sudan/Lower Guinea and Western Sudan/Upper
Guinea. Neither of these two regions was in any way homogenous
economically, but each did comprise production and trading
networks or systems, some of which were of considerable
antiquity, which were closely interconnected and interrelated.
This approach to the political economies of West Africa avoids
the dichotomy of' European oriented' and ' Sudanic' economies,
which increasingly is seen to be false. West African systems of
production and trade, and the wider political economies in which
these operated, were not specifically differentiated according to
whether or not they were associated with the international trade
of capitalist Europe and North America. From the West African
viewpoint, European traders were one species of a familiar genus
of trading diasporas. Their operations were limited to coastal
lands precisely because they were foreigners in a much more
immediate and obvious sense than any of the other trading
diasporas, such as those of the Hausa and the Dyula. Because the
Portuguese, French, Dutch, British and other European traders
were so comparatively disadvantaged, they increasingly tended to
resort to force to achieve their ends. The broad pattern of
partition - certainly in West Africa - reflects the extent to which
Europeans were able and willing to use force in competing
against one another and against Africans.
Viewed from the perspective of partition, the European traders
cannot of course be dismissed so easily. They were the West
African representatives of European international capitalism,
which had pronounced inclusive tendencies. All trading diasporas
in West Africa were instruments of some change - religious and
social as well as economic and political - and the European trading
diasporas came increasingly to be so. They sought to bend the
native West African systems to their demands, implicitly or
explicitly. Throughout the nineteenth century - and indeed

earlier — expanding zones of European economic influence (in
some cases, predominance) spread inland from coastal enclaves.
By the 1870s these zones of European economic activities had
possibly reached a critical stage. Here were technical economic
problems of profitability and terms of trade which required drastic
means for resolution. These networks of European economic
activities operated, however, within the larger West African
economic regions; they were not detached or differentiated from
them. If the late eighteenth century and the nineteenth century
witnessed an economic revolution in West Africa, then the focus
of this was internal rather than external: '... the economic
revolution identified with the nineteenth century and the shift
from slave exports to so-called legitimate trade in vegetable oils
and other goods... began at least half a century earlier and was
more closely related to internal West African economic develop-
ments than European-induced policies and demands.'23
The Central Sudan/ Lover Guinea economic region

The Central Sudan had recovered from the long-lasting drought
of the mid-eighteenth century, and thereafter enjoyed what must
have been unprecedented economic growth. This was maintained
until well into the nineteenth century. The upheavals of the jihads
of 1804-20 caused some economic disruption, but the
establishment of the Sokoto caliphate brought political stability
and enlargement to much of the region. Economic prosperity
continued until at least the end of the century, by which time the
central Sudan had been partitioned between the French, British
and German colonisers. Trade flowed into and out of the Central
Sudan from many directions. Maghribi and Ghadamasi merchants
exchanged Manchester cloth, brought from Tripoli, at Kano for
ostrich feathers and hides, and Tuareg and Hausa engaged in more
regional trade from the old Air sultanate and the Hausa Sahel.
Relations between Sokoto and Bornu were generally peaceful, and
Kanuri participated in much of the regional trade of the north-
eastern Hausaland (Zinder) and Bornu borderlands, while in
Bornu itself, Hausa had become the commercial language.
Indeed, not only was the greater part of the trade of the Central
" Paul E. Lovejoy, 'Interregional monetary flows in the precolonial trade of
Nigeria',/. Afr. Hist., 1974, 15, 4, 563-85.
42

WEST AFRICA
Sudan in Muslim hands (nearly all the traders in Bornu and the
Sahel of northern Hausaland were members of the SanusI order),
but most of these Muslims were Hausa speakers. The Hausa
trading diaspora had begun before the nineteenth century, but that
century witnessed a huge expansion of Hausa production at home
and trade abroad. By the early nineteenth century the strength of
the Central Sudan economy, combined with the collapse of the
Yoruba empire of Oyo, had allowed Hausa traders to assume a
large share of the long-distance trade all the way from the middle
Volta basin to Adamawa and south to the Atlantic coast. From
the main economic centres of the Sokoto caliphate, Hausa
exported livestock, potash, salt, textiles, leather goods and slaves.
In exchange they obtained kola from Asante, west of the Volta
basin and from the Yoruba towns textiles and re-exported
European firearms and cloth. Lancashire cottons reached Kano
both from the Mediterranean and from the Bight of Benin. A
crucial re-export from the Yoruba towns was cowries, which
formed the common currency and the main medium of exchange
for almost the whole of the Western Sudan/Lower Guinea
economic region.24
It is also apparent that trade did not consist entirely, or perhaps
even primarily, of 'luxury' products. Bulk goods, especially
foodstuffs and raw materials, were carried to and from the
caliphate, the Yoruba towns and the Ibo country. Indeed, the
distinction between luxury goods and bulk goods is misleading,
since goods such as kola, potash and salt, for example, were
transported in considerable quantities. Production of manu-
factured goods and agricultural crops involved large numbers of
people, many of whom were slaves or semi-servile labourers. By
the middle of the nineteenth century there were big concentrations
of slaves on plantations in the textile belt of Hausaland - Kano,
northern Zaria, southern Katsina and Zamfara. The forced
relocation of agricultural populations from the peripheral areas
to the prosperous central regions was the most significant
economic effect of the Fulani jihads.2S Much of the collection of
14
Marion Johnson, "The cowrie currencies of West Africa', pans i and 2, / . Afr.
Hist., 1970, ix, 1, 17-49 ant* lx> 3- 331—5 3-
" Paul E. Lovejoy, 'Plantations in the economy of the Sokoto caliphate', / . Afr.
Hist., 1978,19,341-68; J. S. Hogendorn,'The economics of slave use on "Plantations"
in the Zaria emirate of the Sokoto caliphate',/»/. J. Afr. Hist. Studies, 1977, 10, 369-83,
and 'Slave acquisition and delivery in pre-colonial Hausaland' in R. Dumett and
B. K. Swartz (eds.), West African culture dynamics: archaeological and historical perspectives
(Chicago, 1979).

oil in the palm-growing lands around the southern Yoruba city
states was undertaken by slave labour. Among the trading
communities of the south-eastern area, commerce was greatly
dependent upon domestic slaves. They were used as canoe
boatmen, in the porterage of the oil, in marketing and in raising
food crops to feed those people employed in production and
trade - this last being a common feature throughout West Africa.
Some of these domestic slaves could reach the highest positions
in these socially mobile communities.
It is a curious fact [wrote Consul H. H. Johnston] and an evidence of the mild
character of the slavery in the Niger Delta, that nearly all the leading men in
the oil rivers at the present time are ex-slaves, — such as Yellow Duke of Old
Calabar, Ja Ja of Opobo, Waribu and Oko Jumbo of Bonny, and William Kia
of Brass. 26
In northern Iboland large numbers of slaves worked on yam
plantations, the produce of which was sold to the people to the
south, where most of the fertile land was given over to the
growing of palm trees.
The use of slave and other forms of coerced labour in the
Central Sudan and Lower Guinea is characteristic of particular
forms of political economies, but it is not an indicator of economic
decline or stagnation. Even when the prices and volume of palm
products in the south-eastern area fell after the 1860s, the variety
and scale of internal trade remained high. The huge dug-out
canoes used in the coastal lagoons were manufactured scores of
kilometres up-country. Dried fish and some sea salt were traded
inland for yams and other vegetables and for local tobacco. The
transmission to the interior of European goods took up more and
more of the internal trade — an index of dependence on such
foreign goods but also a sign of growing prosperity. Manchester
cloths and clothing, European firearms and ammunition, iron and
other metal goods, salt from British mines, cowries, from the
Indian Ocean, American tobacco, Jamaican rum — these and other
goods worked their way through the old established local markets
and made use of the routes and techniques of trade developed
when slaves were the major item of export. The goods manu-
factured in capitalist Europe had not by the 1870s seriously
undermined local industries and production. The south-eastern
26
H. H. Johnston, 'Report on the British protectorate of the Oil rivers (Niger
delta)', 1888, cited in D . Northrup, Trade without ru/ers: prc-colonial economic development
in south-eastern Nigeria (Oxford, 1978), 222.
44

WEST AFRICA
area was wealthy enough, and palm-oil production and trade

greatly increased the number of people involved in these economic
processes, so that expanded demand and preferences could absorb
without much difficulty both local goods and services and the
flood of inexpensive European manufactures and other imports.
The palm oil and palm-kernel oil production belt coincided
more or less with the area of tropical rain forests on either side
of the southernmost stretch of the lower Niger river, from
Cameroun to the hinterland of the lagoons from Lagos to
Whydah, and along the eastern Gold Coast west of the Volta
mouth. The relationship of African producers to the European
oil firms differed from one part of the coast to another. In the
eastern section — the Niger delta and Cross river areas — African
merchant princes intervened between producers and European
firms. These merchants had become so powerful that distinctive
polities, such as Bonny, Brass, Opobo and the other delta or Oil
river states, played a dominant role in the most lucrative of the
palm regions. In the Yoruba and Aja-speaking section, from
Lagos to Whydah, the role of the middlemen was much less
marked, and many merchants from Ibadan, Ijebu Ode and
Abeokuta were also producers.
On the Gold Coast west of the Volta mouth, trade in palm
products was much less developed, but such as it was, it had been
in the hands of Fante merchants, who dealt with the producers
inland, in the Fante states and in Asante. Along the Gold Coast,
however, a major transition had already taken place by the 1870s.
Much of this medium-distance trade, and also much of the local
trade, especially the exchange of European manufactured goods
for cash or produce, had been taken over by Europeans. Many
Fante firms had failed; some had gone bankrupt, and this enabled
Europeans, in particular the firm of Swanzy, to take the kind of
initiative that in most other parts of West Africa followed rather
than preceded the imposition of colonial rule. The formation of
the southern Gold Coast as a crown colony in 1874, following
the Asante war, was a recognition of the British orientation of
much of the commercial life of the territory. This failure of the
Fante to continue to operate commercially in their own country
lay behind both the birth of the Fante confederation in 1868 and
the breakdown of its politics and constitution in 1873. It must be
emphasised, however, that these European-owned businesses on
45

the Gold Coast were largely concerned with local exchange; the
Gold Coast had still not succeeded in finding a viable legitimate
export trade as an alternative to the trade in slaves.27
These differences in the economic relationship between African
producers and merchants and the European traders reflected the
political diversity of the southern zone. Asante had moved much
more into the production of kola nuts for the northern markets
than into palm products, and its concern with maintaining routes
to the coast was to secure an adequate supply of firearms. In the
1870s kola nuts were carried south to the Gold Coast, and
transported by sea to Lagos, for sale in Yoruba country (as also
were kola nuts from Sierra Leone). The cession of the Dutch forts,
including Elmina, to the British in 1871-2, resulted in a severe
crisis in the Asante political economy, and led to the Asante—British
war of 1873. On the British side, this was an unusual incident in
its relations with tropical African states. It had been foreshadowed
in the Ethiopian campaign of 1865, and these two military
adventures seem to be symptomatic of an increased confidence by
Europeans in their own military technology and organisation, so
that expeditions deep into the interior of Africa no longer seemed
a complete impossibility, or at least a quite unacceptable military
and political risk. The Asante campaign was a massive use of force
on a scale out of all proportion to the merits or necessities of the
situation on the ground.28
Dahomey presented both contrasts and parallels to Asante. It
was still militarily one of the most powerful states in West Africa.
It had certainly retained its internal cohesion. Unlike Asante,
Dahomey had obtained its independent access to the sea, with the
ports of Whydah and, less directly, Porto Novo. This was
immensely important to the kingdom. It was assured a supply of
firearms. Although it had been notorious as a slave-exporting state
during the first half of the century, by the 1870s Dahomey had
managed effectively the changeover to palm produce. Palms were
tapped in the forests behind Whydah and Porto Novo, in both
royal and private plantations. Oil was traded to merchants, many
of them Brazilian Creoles (who also owned some of the plantations)
and the representatives of French firms. These plantations were
17
Edward Reynolds, Trade and economic change on the Gold Coast, 1807-1874 (London,
•974)-
28
Freda Harcourt, 'Disraeli's imperialism, 1866-1868: a question of timing', Hist.
J., 1980, 23, 1, 104.
46

WEST AFRICA
operated with slave labour. In 1876-77 the British navy was to

blockade the Dahomian coast, including the port of Cotonu, over
which the French has possessed vague rights since 1868. This
blockade was ostensibly to protect British traders and to stamp
out the vestiges of the export of slaves. The real reason was the
wish of the British traders, and intermittently of the British
government, to close the gap, to extend British control over the
coast from the Volta mouth to Lagos, and thereby to increase
greatly the profitable customs returns. It achieved little. The
Lagos merchants prospered; the French were slighted, Dahomey
suffered hardly at all. After less than a year, the British called off
the navy, after collecting a nominal fine from the foreign traders.
Gelele's kingdom emerged unscathed. But the change-over from
being primarily a slave-exporting to a palm-oil producing state
led to considerable social and political tensions, which found
expression in one group supporting the old-established militaristic
nature of the kingdom, and another in favour of more peaceful
production and commerce. The conflict between Dahomey and
the Egba and western Yoruba country continued, but here the
basic reasons were not purely military, slave-raiding campaigns,
but control over routes to the sea - particularly to Badagry - and
over populations that could supply labour in the palm plantations.
The prolonged internal conflict between the Yoruba states, and
between the Egba and Dahomey, should not lead one to under-
estimate their strength as political units and their considerable
prosperity. During the middle years of the nineteenth century,
Ibadan, Abeokuta, Ijebu Ode and other towns were growing in
population and in economic diversification. A thriving trade
existed between these states and the emirates of the Sokoto
caliphate. One of the chief causes for dissension was competition
over economic (which also meant human) resources and over
trade routes. The Yoruba civil wars were all marked by a concern
to control the vital links between these inland cities and their
surrounding plantations of oil-palm trees, and the coastal ports.
There was also the need to have access to the north. Most of the
routes leading northwards went through the powerful emirate of
Ilorin to Hausaland. Of the Yoruba states, Ibadan, over 200 km
from the coast, was the most dynamic. It attempted to spread its
political and economic influence, especially eastwards into Ekiti
country, and to procure an independent outlet to the sea. These
47

moves were resisted by the Egba state of Abeokuta and by other

Yoruba states such as Ijebu Ode and Ilesha. British activities at
Lagos, proclaimed a crown colony in 1861, and at Badagry, taken
over in 1862, were aimed largely at controlling the seaward (or
lagoon) termini of the Yoruba trade routes, and imposing tariffs
on imports and exports. British activity incurred the ire of French
firms established further west along the coast at Cotonu and Porto
Novo, and of the German firms which were to try to build up
a trade based on Lagos in the 1870s and 1880s.
Although much stronger economically and much more mili-
taristically oriented, the Yoruba and Egba city states in the 1870s
resembled the Fante states in one important characteristic. Since
the 1840s Christian missions had been established in all the main
urban centres by repatriates from Sierra Leone. In 1859
Yorubaland's first newspaper was launched in Abeokuta, to be
followed by another one in Lagos in 1863. While the northern
Yoruba were strongly under the influence of Islam, those in the
south were open to Christianity, and Western education, conveyed
in the English language. Abeokuta reacted to the British presence
in Lagos by establishing the Egba United Board of Management
in 1865, with a president-general and a high sheriff. A secular
school was set up, some measures of town planning instituted,
there was a court to settle commercial disputes, and a postal
service to Lagos; this was thirty years before British colonial
expansion into Yorubaland. It was from Yorubaland that the
African clergyman Crowther set out in 1857 to proselytise the
people of the lower Niger and the delta, and indeed this mission
introduced Christianity to the merchant princes of that region.
The African-run Niger mission was an active and successful
Christian endeavour; only in the late 1870s did the white
personnel of the Church Missionary Society, the parent body of
the mission, begin their racist attack on the African clergy.
The Yoruba city states remained politically independent. Only
the ports of Lagos and Badagry were in foreign hands. The official
British presence in Lagos had little more impact than the unofficial
Brazilian Creole or French presence at Whydah and Cotonu. The
British colonial government maintained itself on import and
export dues, as many contemporary African rulers did. British
traders prospered, but no more so than up the Niger or on the
delta. So did German traders at Lagos, in spite of tariffs weighted
48

WEST AFRICA
against their goods, especially spirits. The colonial government

became involved in the Yoruba civil wars, but only as yet another
competing state, concerned to keep open the trade routes, which
were the source of the colony's wealth and revenue.
If the Yoruba and Egba states were dynamic and expansive
during the middle part of the nineteenth century, the empire of
Benin was very much on the defensive and lost so many of its
provincial areas, that by the 1870s it had shrunk to its Edo
heartland. From the north, armies from the Muslim emirate of
Nupe pushed down the Niger river into Benin territory; Ibadan
moved in from the west. To the south, the kingdom of Warri
became economically independent, especially after Itsekiri oil-palm
producers and traders set up new centres of exchange with
European firms at the mouths of the Benin and other rivers. The
Itsekiri effectively excluded Benin from the profitable oil trade.
For many years Benin was largely ignored, both by other Africans
and by Europeans: it was not until 1897 that Britain sawfitto
conquer the heartland of the ancient empire and to incorporate
it into colonial Nigeria.
The Niger delta country, including the Cross river area to the
east, was the oil-producing region par excellence. In the earlier part
of the nineteenth century the European demand was for palm oil,
used for the production of soap and lubricants. By the 1870s the
demand was more for the oil from palm kernels, which was used
in the manufacture of margarine. Earlier, much of this oil trade
had been handled by British merchants. From 1841 French
merchants, in particular the Marseilles firm of Regis, competed
on the Dahomian coast and, to a lesser extent, at Lagos and on
the Niger delta. The 1870s saw the rapid growth of German trade
in the palm zone; by 1880 Hamburg had come to appropriate
nearly one-third of all West Africa's overseas trade. This increased
European economic activity on the coast of Lower Guinea was
accompanied by, and was in part a response to, a depression in
value of palm oils to the African producers, and to the middlemen
engaged in the trade. In the twenty-five years from i860 to 1885,
the price of palm oil fell by nearly 50 per cent, and palm-kernel
prices fell by about one-third. The terms of trade, for the first time
in the century, turned distinctly against African merchants and
producers. These reciprocated by taking protectionist and
monopolistic measures, some of which entailed the use of force.
49

Force had always been latent in the highly competitive and

socially mobile milieu of the Niger delta oil barons. By the middle
of the century civil wars, with underlying economic causes, were
beginning to break out all over the delta/Cross river area: Calabar
in 1850—i, Bonny in 1869, Kalahari in the 1870s. In 1869 Jaja broke
away from Bonny to found Opobo. Around 1879 Olumu set
himself up at Ebrohimi at the mouth of the Benin river, in
competition with Warri, and in 1883 he was succeeded by the
famous Nana. The newer delta city states appeared to be stronger
economically than their older rivals. Not all the European firms
attempted to by-pass the oil barons by making contact with
African producers inland. Some continued to trade satisfactorily
with the delta states. It was rival firms to these that had used the
Niger river as a means of penetrating the interior. A British firm
traded in ivory and shea-butter with the emirate of Nupe from
1871, and a little later was able to tap the oil trade at markets such
as Onitsha. There were thus rival trading systems on the Niger
river and in the delta, although the former was small compared
to the latter. The European river traders were protected by
warships of the British navy, but the four main British firms were
in competition with one another as well as with the delta oil
barons. Only in 1879 did George Goldie succeed in merging the
British Niger river interests in the United African Company. By
this time French competitors had established themselves up the
Niger. Although prices for palm products had fallen, the
increased — and sometimes violent - competition among African
and European traders did not adversely affect African producers,
and there is little evidence to support the notion of an economic
crisis enveloping the delta states and Iboland in the 1870s.
In the 1870s Ibo country (and also the land of the Ibibio and
Efik) and its coastal outlets were still heavily committed to the
trade in palm oil and palm kernels, even though the boom
conditions of the previous two decades had given way to the
beginnings of a recession in this trade. The trade in slaves had
come to an end, as had consequently the transitional period when
both slaves and oil were exported. The slaves were now employed
internally on palm plantations and on food-producing farms.
Opobo, which was founded by Jaja in 1869, developed as a direct
result of the oil-dominated export trade. The site of Opobo
effectively cut off Bonny from participation in the trade, and
50

WEST AFRICA
Opobo became the premier oil port. Production was in the palm
belt of the southern half of Ibo and Ibibio country, and involved
large numbers of people, unlike the specialised slave trade, which
had been in the hands of groups of people who specialised in this
economic activity.
Apart from the fairly widespread use of slave labour, the
complex and laborious operations of production were essentially
egalitarian. Most free producers were members of both village and
age-set communities; and even where so-called secret societies
were involved, the membership of these was widespread and easily
acquired. With the cessation of the slave export trade, the
formerly vital role of the diaspora of Aro traders gradually
diminished in importance, though some Aro groups took part in
oil production and trade. Most oil was transported to the coastal
ports by canoe, using the numerous rivers and creeks of the
region. The construction of the large craft, their manning and the
services required, all led to intense economic activity, The
northern part of Iboland was not left out of this widespread area
of prosperity, because the palm belt increasingly became
dependent upon it for food, especially yams. What the palm tree
was to the production, commerce and transportation of the south,
yams were to the north. The development of a mass-production
force in Iboland generated a demand for cheap goods, which
coincided neatly with the importation of mass-produced European
manufactured goods for the first time on a large scale. At this
period such imports stimulated rather than depressed the local
economy: 'Iron bars actually stimulated hinterland industry by
freeing smiths from dependence on inefficient local iron-smelting,
thus greatly increasing the availability of locally made tools and
weapons.'29 All this economic activity - on a scale perhaps
unparalleled in tropical Africa - took place in societies notable for
the absence of states or any form of political units larger than
lineage groups, age sets and secret societies. 'The political
organisation of south-eastern Nigeria', concludes Northrup, 'was
able to cope with the conditions of the pre-colonial period
without major structural changes.'30 In this respect, Iboland could
not have been more dissimilar from the Sokoto caliphate.
The caliphate of Sokoto was the largest political unit in West
Africa: it was not a state, in either the general or the theoretically
20
Northrup, Trade without rulers, 226. *° Ibid., 250.
51

technical sense of the term, and neither is ' empire' an adequate

description. It was a Muslim theocratic federation, comprising
nearly thirty emirates, most of which were ruled by the descendants
of the Fulani jihadists of the first three or four decades of the
century. These emirates were linked, in a varity of ideological,
constitutional and administrative ways, to the two rulers of the
divided caliphate, the emir of Gwandu (in the west) and the caliph
of Sokoto (in the east). The caliph was the ultimate source of
religious and political authority. The emir of Gwandu related to
the caliph in much the same way as individual emirs related to
the two higher authorities. The caliph was responsible for
appointing emirs, and his wa^tr (or the watrtr of the Emir of
Gwandu) toured the federation to ensure that the emirs kept to
the complex rules.
Not all the peoples within the confines of the federated
caliphate accepted the position of the emirs or the caliph. The old
Hausa state of Kebbi, for instance, remained independent (except
for an eighteen-year period), as did many non-Muslim peoples
scattered throughout the caliphate. Borgu to the west, Dama-
garam, with its capital at the important trading city of Zinder, to
the north, and Bornu and Mandara to the east were outside the
control of the caliph. There was considerable internal friction,
over succession disputes within the emirates, and with revolts of
emirs against the caliph's authority. The emirates on all except the
northern borders tended to expand militarily. Ideologically the
caliphate can be seen as being in a state of continuing jihad; or
rather, by the last third of the century it was a vast geopolitical
area living with the consequences of the earlier jihads.
In spite of conflict and revolt, there was a large measure of
stability within the caliphate as a whole, and within each emirate.
Administrative, judicial and fiscal systems had been established,
and generally functioned satisfactorily. The caliphate was multi-
ethnic, with other groups besides the Fulani and Hausa, and it
was also stratified socially. With increased commercial and general
economic activity, this stratification was becoming more marked.
Society was divided between freemen and slaves. At the top of
the non-servile social ladder was the aristocracy, newly established
Fulani lords or more ancient Habe (Hausa) lineages. This was a
feudal aristocracy with many lords living on their fiefs and others,
employed in the local or central bureaucracies, living on the

WEST AFRICA
proceeds of their estates. Below the aristocrats, socially, were their
clients. The religious authorities, 'ulamd and others, were a distinct
group. At the bottom of the scale were commoners, talakawa,
peasant farmers and urban craftsmen. Although stratified, it was
not a closed society; social mobility was possible, and the rate of
slave manumission was high.
By the 1870s the caliphate was gaining in cohesion and strength
from the effective operation of its religious and administrative
linkages, and from its active and varied economy: 'Economic
growth which had begun after 1750 continued through the
nineteenth century, and the steady enlargement of the caliphate
(throughout the middle years of the century) facilitated the
movement of merchants and led to the economic supremacy of
Hausa traders throughout the savanna from the Volta to
Adamawa, the northern Yoruba country, Borgu, and the middle
Volta basin.'31 Kano was the great commercial and industrial
centre, but by no means the only one. In each emirate the capital
and certain other towns were markets, and were the focal points
of agricultural and manufacturing production. Only in the
outlying emirates, such as Kontagora in the west, Adamawa
(Yola) in the east, and Nupe and Ilorin in the south, was the accent
on predatory rather than on peaceful commerce. Trade between
Zinder (Damagaram) and Kano flourished, in spite of hostility
between the rulers of the two emirates, the one without, the other
within the caliphate. Barth's mid-century comments on Kano are
worth repeating, because the economic activity of the city was a
barometer for most of the caliphate as a whole.
The great advantage of Kano is that commerce and manufactures go hand in
hand, and almost every family has its share in them. There is really something
grand in this kind of industry, which spreads to the north as far as Murzuk,
Ghat and even Tripoli; to the west, not only to Timbuktu, but in some degree
even as far as the shores of the Atlantic, the very inhabitants of Arguin dressing
in the cloth woven and dyed in Kano; to the east, all over Bornu, although
there it comes in contact with the native industry in the country; and to the
south it maintains a rivalry with the native industry of the Igbira and Ibo, while
to the south east it invades the whole of Adamawa, and is only limited by the
nakedness of the pagan sans-culottes, who do not wear clothing... 32
31
Lovejoy, 'Interregional monetary flows', 571-2.
12
Heinrich Barth, Travels and discoveries in northern and central Africa, j vols. (London,
1857), 1, 510.
53

The Western Sudan/ Upper Guinea economic f(one

The second economic zone of West Africa comprised the immense
area from the Atlantic coast between Cape Blanco and Arguin
island, north of the Senegal mouth, around to the mouth of the
Bandama river on the Ivory Coast, on the seaward side; on the
landward side, from the Sahel north of the Senegal and upper
Niger rivers, as far east as Timbuktu, across the upper Volta basin
and the woodland savanna of Bonduku, Kong and Baule. The
interior was dominated by two great empires, one declining, the
other emerging: the northern Tukolor (Fulani) conquest state
ruled by Ahmad, the son of 'Umar Tal (al-Hajj 'Umar); and the
southern Dyula conquest state of Samory Ture.
To the east, in the borderlands between this region and the
Central Sudan/Lower Guinea economic zone, were a series of
much smaller states, such as the kingdoms of the Mole-Dagbane-
speaking peoples. Inland from Sierra Leone were the small pagan
chiefdoms of the Mende, who had reached the sea in the
eighteenth century, and north of the Rokel river the Muslim
Temne chiefdoms were consolidating their position. The Muslim
Fulani theocratic state of Futa Jalon, founded in the eighteenth
century, continued to exercise great cultural and religious prestige
even after its military prowess had waned. In the middle of the
century the aristocracy of Futa Jalon had to contend with a
complex religious movement, that of the Hubbus, composed of
followers of the Tijaniyya teachings introduced by al-Hajj 'Umar,
as opposed to the Qadiriyya brotherhood held to by the majority
of Fulani. This religious civil war was not brought to an end until
the alliance of the Fulani of Futa Jalon with Samory between 1879
and 1893. Futa Jalon exerted considerable influence over the small
coastal populations of what the French called the Rivieres du Sud,
between the Rio Grande (Portuguese Guinea) and the Melakori
and Scarcies rivers, north of the Sierra Leone peninsular.
Senegambia experienced a series of politico-religious revo-
lutions in the 1860s and 1870s, which led to the establishment of
a number of dynamic states, and which spread Islam among the
populations of most of the region. The revolutions were sparked
off (as were many other political events in the whole zone) by the
presence in the region of 'Umar Tal, a Muslim cleric from Futa
Toro, who returned from pilgrimage to Mecca in 1840 and
54

WEST AFRICA
proclaimed a jihad in 1852. Large numbers of Fulani from Futa

Toro migrated east to settle in 'Umar's domains. The Islamic
revolution in Senegambia began in Babidu, a Mandingo kingdom
on the north bank of the lower Gambia river, led by Maba
Diakhou Ba, who conquered the Serer country of Salum. Maba
converted two Wolof leaders to the rule of the Tijaniyya — Lat
Dior, damelof Czyor, and Alburi N'Diaye, who held the traditional
Wolof kingly title of burba, and who exercised considerable
influence over the Wolof. The last of the great Senegambian
jihadists was Muhammad al-Amln, a Soninke who in the mid-1880s
established an anti-French and anti-Fulani state centred on Bundu.
The physical impact of what Curtin terms the ' trade diasporas
from overseas' on the western half of West Africa was still very
small. A few scattered forts and trading posts on the Senegal river,
such as Podor, Bakel and Madina, inherited from the burst of
activity early in the nineteenth century or during Faidherbe's
governorship, barely secured French trading and political interests
along the river. Such power as the French enjoyed was exercised
by a few gun-boats which could only move when the river was
in flood. During the dry season their neighbours resumed
superiority. Futa Toro controlled much of the southern bank of
the river. One hundred and fifty kilometres to the south of St
Louis were the French island of Goree and the coastal settlements
of Rufisque, Dakar and Joal.
The British position on the Gambia was similar but even more
minuscule, with Bathurst on St Mary's island in the river mouth,
a few miles of river bank, the ex-French factory of Albreda, and
MacCarthy Island, a trading depot 200 km up-river. French
traders competed with British, Portuguese and Afro-Portuguese
at the mouth of the Casamance. The Portuguese retained their
ancient tiny settlements on the Bissagos islands and the neigh-
bouring mainland, including the small port of Bissau. The British
colony of Sierra Leone consisted of the stony peninsula on the
estuary of the Rokel river, the Isles de Los to the north and Sherbro
island to the south. Kennedy, who was governor of the British
West African settlements from 1868 to 1872, built up a delicately
balanced informal sphere of British political and commercial
influence among the Temne and Mende chiefdoms in the
hinterland. But the system of British informal influence broke
down after 1873, a n ^ t m s contributed to the upsurge of European
55

aggressiveness in the area. Some 10,000 repatriated African slaves

from the United States occupied coastal settlements at Cape
Mesurado (Monrovia) and a few other colonies scattered along
the Liberian coast, between the Gallinas river and Cape Palmas,
but the coast of the Kru country attracted very little long-distance
or overseas trade. The tiny trading settlements at Grand Bassam
and Assinie on the Ivory Coast, which the French had occupied
in 1842, formed a westward extension of the Akan/Gold Coast
economic region.
The traditional economic heartland of the Western Sudan/
Upper Guinea zone was the great basin formed by the tributaries
of the upper Niger and the upper Senegal. The axes of trade spread
out from this core area. The great majority of the long- and
middle-distance traders were Soninke or Malinke in origin, or a
mixture of the two, their collective name being Dyula. The Dyula
dominated the trade and commerce of the western zone to an
even greater extent than did the Hausa of the eastern. The
majority of the Dyula were members of lineage groups whose
members included not only traders but also Muslim clerics. They
often set up distinctive clerical towns along the line of their
trading routes, or they were congregated in separate quarters in
the towns of the host community. In an increasingly differentiated
society, they occupied distinctive petit bourgeois positions. The
Dyula not only had a remarkable economic effect on most regions
of the western zone, but also spread Islamic influences far and
wide, and they were instrumental in the growth of two large
polities — in the eighteenth century, the Wattara empire of Kong,
between north-western Asante and the Bani tributary of the
Niger, and, in the nineteenth, the state built up by Samory Ture.
The main axes of Dyula trade radiated in almost all directions
from the Upper Niger/Upper Senegal core area. The upper and
middle reaches of the Niger were traversed by canoes, some of
them very large, which were crewed by specialised rivermen — the
Bambara Somono above Timbuktu, the Songhay from Timbuktu
to Gao and Say - and this crucial waterway carried goods, largely
in the hands of Dyula merchants, from the woodland savanna
areas of Lower Guinea to the desert boundary in the north. The
trans-Saharan trade to Morocco and Tripoli from Timbuktu
experienced a revival in the 1860s and 1870s. Some of the desert
trade was regional rather than international. Moorish merchants
56

WEST AFRICA
transported salt from Idjil to Galam on the middle Senegal, to be

exchanged for gold from Bambuk, which was then carried north
to Morocco 'in exchange for manufactured products in a form
of triangular trade'. 33 A major and long-established trade route
ran from the river ports of the middle Niger and Bani through
Bobo Dioulasso, Kong and Bonduku to the Akan forests and the
Gold Coast. By the nineteenth century the main item of trade was
kola nuts. These were obtained in great quantities from Asante,
and from the Senufo lands south of Kenedugu (one of the
successor states of the Kong empire).
The chief Dyula group on the section of the Niger around Segu
was the Maraka, who were of Muslim Soninke origin. Their main
commercial settlement was Sinsani (Sansanding), some 25 km
downstream from the Bambara capital of Segu. In the early
nineteenth century Sinsani had taken over from the Jenne the role
of the major trading entrepot in the middle Niger valley. But the
conquest of the Bambara states by al-Hajj 'Umar in the i860 led
to a sharp decline in the fortunes of Sinsani, and thereafter the
main trading focus shifted to the Maraka city of Banamba, in the
savanna west of the river. 'Umar's conquest of Segu and Kaarta,
the two eighteenth-century Bambara states, and of the Fulani state
of Masina, seriously disrupted, but did not suppress, trade.
'Umar's empire did not succeed in establishing political and
administrative stability, and therefore in benefiting from its great
increase in geographical scale, before its long-drawn-out conflict
with the French.
The Maraka engaged in a thriving 'desert-side' economic
interchange with the Moors of the Sahel and Sahara. Millet and
other grains, and cotton cloth were exchanged for salt, horses and
animal products. This desert-side trade depended on the ability
of the Maraka to produce a surplus of millet and cotton cloth.
This in turn depended on the availability of slaves, to grow the
millet and cotton on plantations and to operate the textile
industry. The desert-side economy was the basis for more
widespread trading activities. Additional surplus grain and cloth
was transported by donkey to the south, to Kong and Tangrela
in Kenedugu, and up the Niger by canoe to Kankan, as were
re-exported animal products earlier traded with the Moors. In the
woodland savanna areas, these goods were exchanged for kola
33
Curtin, Economic change, 283.
57

nuts and slaves. A further branch of Maraka trade was by canoe

down the Niger to Timbuktu, with slaves from the woodland
savanna regions and salt from the Sahara. The profit on selling
slaves at Timbuktu was sufficient to cover the expenses of the
trading voyage and to provide money — in the cowrie currency
common to the western Sudan, but not to Senegambia - to
purchase further slaves from the south. The Maraka economy was
thus both slave using and slave exporting.34
The greatest kola nut emporium in the woodland savanna
region of Lower Guinea was Kankan, on the Milo river, a
tributary of the upper Niger. The city had grown from a village
to a Dyula merchant republic in the eighteenth century, eclipsing
Bamako. It worked in close alliance with the Fulani theocracy in
Futa Jalon, which indeed was an economic necessity, as Futa
Jalon controlled the routes between Kankan and the coastal
outlets on the Rivieres du Sud and Sierra Leone. In the aftermath
of the Umarian disruption, Kankan adopted a more aggressive
policy in the 1870s. This resulted in an alliance of most neigh-
bouring peoples against 'the imperialism of Kankan', and the
fortunes of the city were saved only by the actions of Samory,
whose involvement on behalf of the city was his first move in
taking up a distinctly Muslim position in his great military and
political campaigns.
Kankan was the junction of two of the major trading axes of
the western zone - that which ran north-east, down the valley of
the Niger to Timbuktu, and the north-south routes which started
in the Kaarta Sahel around Nioro and ran via Kita and Siguiri
to Kankan, Bissandugu and Bayla. This was one of the main kola
nut trade routes, one item given in exchange being Saharan rock
salt. To the west of the Kaarta-Kankan axis were other kola nut
routes - one set starting in the Freetown/Port Loko region of
Sierra Leone and crossing the Futa Jalon to Bakel and other towns
on the Senegal. Some kola nuts and other forest produce had been
shipped by European merchants from Sierra Leone to the Gambia
and the Senegal since the sixteenth century.
While there was some through traffic from the Rivieres du Sud
and Sierra Leone to the lands of the upper Niger and its
tributaries, especially in firearms, Senegambia was the area where
the European-dominated maritime trading systems most signifi-
34
Roberts, 'Long distance trade and production', 169-88.
58

WEST AFRICA
cantly penetrated the western Sudanic heartlands of the western

economic zone. Senegambia was the nearest sub-Saharan African
region to Europe. The two waterways, the Senegal and the
Gambia, gave a means of access to the economies of the interior
unparalleled in West Africa, and indeed, with only a few similar
examples in the continent as a whole (the Kwanza river in Angola
was one, and the Zambezi in Mozambique another). French and
British traders used the Senegal and Gambia rivers respectively;
in the opposite direction, various Dyula groups similarly exploited
these routes. Foremost among the French and British were two
particular groups, both of great importance in the history of West
Africa. One of these were the people of mixed descent - Afro-
Portuguese, Afro-French (cre'oks) and Afro-British — who as small-
scale traders dominated many of the coastal economies from the
Senegal to the Rivieres du Sud until the intervention of big
European, mainly French, firms in the groundnut trade in the
second half of the nineteenth century. The other group was that
of the recaptives or' liberated Africans', the Creoles in the English
sense of the term, who were people of mixed culture in Sierra
Leone and the Gambia, though they were mainly of African
descent. These recaptives played a major role in the economic and
political history of Sierra Leone and, to a lesser extent, of the
Gambia; and also, as migrants, in the economic and political
history of the Gold Coast, Lagos, and Oil rivers and, especially,
in Yorubaland.
There were a number of axes of Dyula trade leading from the
interior to the Senegambian ports. The Gambia river routes ran
from Bamako and other towns on the upper Niger to Barakunda
Falls or MacCarthy Island, the head of navigation on the river.
These routes were controlled by Diakhanke and Malinke Dyula.
In the wet season goods were transported down the Senegal, in
boats owned by the French or Afro-French of St Louis. There
was also a desert camel route from Senegambia to the interior,
from the country of the Trarza Moors north of the Senegal via
Tichit, Walata and Arawan to Timbuktu. By the second half of
the nineteenth century, after the export of slaves had ended, these
interior routes to Senegambia were involved in regional rather
than international trade. The only overseas import of significance
was firearms, and the majority of these destined for the interior
entered via the Rivieres du Sud and Sierra Leone rather than via
59

Senegambia. European goods on the upper Niger, mainly textiles,
were still imported by way of Morocco and Tripoli and the
trans-Saharan routes at the end of the 1870s. Curtin concludes,
Even though it was fifty to seventy percent further from [the Middle Niger
valley] to North Africa, the desert routes were dominant until the beginning
of the colonial period; which suggests that camel transportation was at least
fifty to seventy per cent cheaper than the combination of boats on the Senegal
and donkey caravans from [Galam] eastward. The east-west route of Sahel or
savanna, however, was vastly superior to the route through the forest from
the Gulf of Guinea — not because the forest itself was a barrier, but only because
on that route donkeys gave way to human porterage.35
Senegambia, however, was important in the nineteenth century

not so much because of the trade generated between the coast and
the interior, but because of the cultivation in parts of the region
of groundnuts for export. The cultivation of groundnuts by local
African peoples began along the Gambia in the early 1830s, spread
south towards Sierra Leone, and north into Senegal in the 1840s.
The groundnut (aracbis hypogaea) is a plant of the dry savanna.
Along the Rivieres du Sud and in northern Sierra Leone
groundnuts were produced largely by slave labour, which was
supplied from Futa Jalon and elsewhere in the interior by Dyula
and other merchants. Along the Gambia, in Sine-Salum, the
Rufisque/Petite-Cote region, in Cayor and along the lower
Senegal, groundnut production was mainly by free labour — that
of the owners of the land being supplemented by seasonal migrant
labour from the interior. These ' strange farmers' of the Gambia
and 'navetarns' in Senegal were the people, Dyula and others, who
previously had come down to the coast to trade in wax, ivory and
slaves.36 As in the palm-oil regions of the eastern economic zone,
this African production of a major export item stimulated the
development of a host of service and supplementary activities,
such as transportation, marketing and feeding. Groundnut
production also led to considerable social change and diversifi-
cation, in regions which were either already predominantly
Muslim by the 1870s, or were becoming so.
The pioneer merchants of the groundnut export trade had been
African and Eurafrican traders, but from the 1860s these were
35
Curtin, Economic change, 181—3.
36
C o l i n W . N e w b u r y , 'Prices and profitability in early nineteenth century West
African trade' in Claude Meiilassoux (ed.). The development of indigenous trade and markets
in West Africa (London, 1971), 96.
60

WEST AFRICA
being put out of independent business by the representatives of

the big European firms. By this time, along the whole of the t(one
arachide from Senegal to Sierra Leone, French commercial interests
had gained the upper hand over their American and British rivals.
British industry preferred palm oil for soap-making and other
uses, leaving the way open for French industrialists to promote
the groundnut as a source of oil for similar processes, and for
Marseilles (and to a lesser extent, Bordeaux and Nantes) merchants
to dominate the trade. French tariffs discriminated against
' colonial' produce imported in foreign vessels, and this meant in
effect that West African groundnuts were transported in French
ships. At the outset of peanut commercialisation, French mer-
cantilist policy also imposed metropolitan processing on Sene-
gambia, a pattern of colonial dependence which became
increasingly restrictive after partition. By the 1860s and 1870s the
French had established a commercial hegemony over the Upper
Guinea coast. This corresponds to some extent with the British
position over the Lower Guinea coast, although this was by no
means hegemonic; the British had to face German, and to a lesser
extent, French competition. The commercial framework in the
coastal regions, so clearly apparent by the 1870s, indicated fairly
exactly the pattern of later colonial penetration. The persistence
of British Sierra Leonean and Gambian interests, and of French
activity on the Dahomian coast, and the failure of the Anglo-
French exchange proposals of 1875 and 1876, emphasise the
ability of small pressure groups to exert negative influence upon
apathetic governments rather than either a departure from the
predominant overseas trading patterns of the two economic zones
or the opening moves in a territorial partition of West Africa.
B. SOUTH OF THE EQUATOR

Africa south of the equator is nearly synonymous with Bantu
Africa. Since the Late Stone Age, this part of Africa has been less
densely populated than that to the north. South of the equator
farming populations only started to build up their numbers within
the Iron Age. Forest and bush remained largely uncleared. Tsetse
infestation was widespread, and the beast of burden was practically
unknown. Human societies were smaller and more isolated. The
61

combinations of production and trade were markedly less

developed than those which had taken root in most parts of West
Africa. The growth of more intensive economic activity,' market-
oriented ', in which trade freed itself from the shackles of sub-
sistence and kinship, depended very largely on the intrusion into
areas of Bantu Africa of overseas trade and commerce (and also
settlement), whereas in West Africa this overseas factor was only
one among other entirely indigenous elements of economic
growth. Economic change in Bantu Africa, and the social and
political restructuring which accompanied it (often violently),
was mainly dependent upon Indian Ocean and European/
Atlantic economic systems. In contrast to West Africa, there were
few instances in Bantu Africa of the growth or exploitation of
agricultural resources for overseas markets. Also in contrast to
West Africa, many of the outposts of those overseas economic
systems were colonies of settlement rather than mere trading
enclaves.
Three distinctive and complex trading and colonial systems had
by the 1870s spread their economic tentacles over much of Bantu
Africa. The Arab Indian-Ocean network operated in eastern
Africa from its African base on Zanzibar island. Portuguese
colonisation and especially trade had penetrated far inland from
the coasts of Mozambique and Angola, while the southern lands
of Bantu Africa were already well within the outreach of South
African British colonial and Afrikaner republican economies. The
north-western segment of Bantu Africa was the exception to this
type of economic penetration. Here, the situation in the coastal
region from the Cameroun estuary to the Congo estuary resembled
that in West Africa: the difference was in the hinterland, which
contained no equivalent of the western Sudan and Sahel, and no
long-standing connections with the world of Islam.
This north-western equatorial hinterland also differed from
other isolated hinterlands in Bantu Africa, notably the inter-
lacustrine region and the savanna lands of southern Zaire. These
other comparatively isolated countries were agriculturally rich
and fertile, and could thus support large populations. This was
especially the case with the interlacustrine lands, where large
political units were able to take root. But the hinterland of
north-western equatorial Africa was poor in economic and human
resources. Farming was difficult, and political communities small
62

WESTERN EQUATORIAL AFRICA
MO?
TP^?IO^ ^
• Towns (European)
K\\NI European colonial possessions
i::::i Afrikaner republics
LOZI African slates and peoples
^ Zanzibar! commercial
influence
Cape Town 1000 km

500m iles
2 Southern Africa on the eve of partition
and fragile. The differences between the equatorial country to the

north-west of the Congo basin, and that to the north-east (the
interlacustrine lands), had a profound ecological effect on the ways
in which these regions reacted to the introduction of European
colonisation.
From Mount Cameroun south to the Congo estuary economic
patterns similar to many of those of the Central Sudan/Lower
Guinea zone had been established by the 1870s, but on a smaller
scale. Leading groups among the coastal peoples, notably the
Duala, the Mpongwe and the Vili acted as entrepeneurs for the
European commercial firms which dealt in palm and kernel oil
and ivory. For a century or more before the 1870s, however, much
of this region experienced very large movements of population,
63

similar in size, and in some respects in political consequences, to

the Nguni/Ngoni migrations of southern Africa. These popula-
tion movements modified agricultural production and trade, in
many instances adversely.
Fang people, who had previously lived north of the Cameroun
rain forest, began to move south in the late eighteenth century.
By the beginning of the nineteenth century they were already
entering the forest zone, adapting their grassfields culture to the
more difficult forest environment. These forests were sparsely
inhabited by Pygmy people, who were eradicated or absorbed by
the Fang migrants. The Fang were politically acephalous, but laid
great stress on the acqusition of wealth, which resulted in
considerable social differentiation. Their migration was very
rapid. By the 1840s Fang groups were exchanging ivory for guns
with European traders and their African agents on the Gabon
river estuary; thirty years later they had reached the Ogowe river,
after having expanded southwards some 800 km from their
northern homelands. Elephant hunting by the Fang accounted for
much of the ivory collected in the forests of Cameroun and
Gabon, and their infiltration towards the coast was aimed at
eliminating the Duala and Mpongwe middlemen.
In the early nineteenth century slaves were the chief export of
western equatorial Africa. Some slaves were exported northwards,
across the Sahara trade routes; others from the coastal hinterlands
of Cameroun and Gabon were sold for export to the Americas
till as late as the 1880s. But by the middle of the century the export
trade in slaves was clearly in decline. The trade goods that
replaced slaves were ivory, palm oil, kola, hardwoods and, later,
wild rubber. Most of the ivory came from the Cameroun savannas
and woodlands. The history of the trading systems by which this
ivory reached the outside world is complex, and provides a
striking example of the capacity of African merchants to wrest
trade from European factors, who, at first sight, would appear to
have been better advantaged.
The middlemen of the northern Cameroun coast were the
Duala, who exercised the same kind of economic role as did the
Ibibio and Ijo people of the Niger delta. Most of the Duala
merchant princes had successfully manipulated the change from
slave exporting to palm oil and ivory production and exchange.
British firms dominated the Duala coast, where British missionary
64

societies had been operating since the 1840s. The Hamburg firm
of Woermann had begun to compete with this British presence
from 1863, and it was representatives of this firm who signed
treaties with the Duala merchant princes in 1884.
South of Cameroun, from the Spanish station at Bata to the
Congo estuary, was a long and rather inhospitable coast. The
French had founded their freed-slave settlement of Libreville on
the Gabon estuary in 1848, but external trade on the coast was
dominated by the British. The illegal slave trade persisted until
the end of the 1860s; 'legitimate' trade commodities were ivory
and hardwoods. The role of middlemen, similar to that of the
Duala in Cameroun, was played by Mpongwe merchant princes,
but the business of trade was complex and protracted. Inland,
Mpongwe trading activities were challenged by the Fang, who
had reached the Gabon coast south of the Ogowe river by the
1870s. The Fang traded for guns, and were formidable
adversaries.
By the 1870s British and other traders were becoming more
assertive. The old complicated system of credit had broken down.
The traders advanced up the rivers, setting up stations and using
Mpongwe as their agents. Wild rubber supplanted ivory as the
main export. When slaves were no longer exported, they were
used as labourers by Fang and Mpongwe to gather the wild
rubber, and to transport it by canoe down the west-flowing rivers.
There was little or no French participation in this trade.
South of the Ogowe river a different pattern of trading had
grown up. Here, the economy of the region had been dominated
by the powerful Vili state of Loango and the lesser kingdoms of
Kakongo and Ngoyi on the coast, and by the Teke (Tyo) kingdom
inland to the north of Malebo Pool (Stanley Pool). These states
had been active in the slave trade, and indeed the power of the
kings of the Loango coast had been largely usurped by an official
known as mafuk, the governor of harbours and ' a minister of trade
and Europeans'.37 Power was passing from agnatic noble groups
to powerful businessmen and commoner village headmen. Fang
and other people of the upper Ogowe and the coastlands to the
south of Cape Lopez traded with the Vili and Teke kingdoms
rather than with the Mpongwe trading stations around the Gabon
estuary. By the second half of the nineteenth century these
37
Jan Vansina, Kingdoms of the savanna (Madison, 1966), 19).
65

kingdoms were in economic contact with an even wider trading

system in the interior of western equatorial Afria.
When in 1877 Stanley was travelling down the Congo river,
he left the sphere of Swahili-Arab ivory and slave-trading
activities at Stanley Falls (Kisangani). At Lisala, 450 km further
downstream, he came upon a trade in firearms being undertaken
by the Ngala people. Indeed, by the 1870s the outreach of Ngala
trading was as far east as Basoko, only 200 km from Stanley Falls.
South of the area where the Ikelemba tributary joined the Congo,
the Ngala gave way as the river-side traders to the Bobangi. The
Bobangi controlled the trade on a 5 00 km stretch of the river, as
far south as the Kwa river confluence, at no great distance from
the borders of the Teke kingdom.
Before the impact of the European Atlantic slave-trading system
on the central Congo basin in the eighteenth century, the Bobangi
were fisherfolk living in the triangle between the Congo and
Ubangi rivers. There they had developed a typical river-bank
trading network within the forest zone, similar in its main aspects
to the desert-side economies of the Saharan and Sudanic lands.
Fishermen exchanged fish for agricultural products, mainly
cassava. They made pots from river-bank clay and extracted salt
from river-side grasses, which they exchanged with the 'inland'
people for vines which were used for making fish nets, cloth and
mats. Such local trading networks, seldom spreading over more
than a hundred square kilometres, were common throughout the
whole of the Congo river basin region.
During the eighteenth century the central part of the Congo
basin was drawn into the expanding Atlantic slave-trading system.
Slaves, bought from all the major ethnic groups of the basin, were
taken by canoe to markets, like those around Malebo Pool in the
Teke kingdom, and then marched overland to the European
shippers on the Loango and Kongo coasts. The Bobangi fishermen
responded with alacrity to these enlarged economic circumstances.
They established settlements on the Congo river south of their
homeland, and by the middle of the nineteenth century they
dominated the river trade. The key section for Bobangi control
of trade was the narrow 'channel' between Tshumbiri and
Missongo, where the river narrows from being several kilometres
wide upstream to a single channel only a few hundred metres
across. As the slave trade came to an end, so, opportunely, the
66

price of ivory rose rapidly in response to the increasing demand

in Europe and North America. Slaves, however, continued to be
used in large numbers by the Bobangi traders. A number of these
traders rose to prominence, creating riverain firms based on slave
labour, some slaves becoming trusted agents and rich men on their
own account. The transition from local fishermen traders to
powerful merchant houses, fiercely competitive amongst them-
selves, caused severe social tensions, which encouraged the
merchant chiefs, known as makonî, largely to abandon their
matrilineal kinship structure and to turn to slaves even as a means
of increasing their population. The economic success of the
Bobangi was won at a heavy social price: by the end of the century
they had a very low birth rate, which continued into the colonial
period.
Ivory followed the same main routes to the coast as had slaves
in earlier decades, coming from far and wide over the Congo basin
and from as far north as the upper Ubangi river. Here the Bonjo
were famous ivory hunters; they traded the tusks to the Loi who
in turn traded them to the Bobangi. The upper Ubangi was close
to the territory of the somewhat amorphous Bandia sultanate of
Bangassou, which presided over the trade in ivory and slaves that
was conducted either through the Nilotic Sudan or via Wadai and
the trans-Saharan route to Benghazi. The hunters of the Sanga
river valley were Pygmies, and as the headwaters of the river
reached towards Adamawa, there was competition for ivory in
this region between traders from the Congo basin, the Cameroun
coast and Bornu. Ivory transported out of the central basin in the
great canoes of the Bobangi merchants converged on the markets
around Malebo Pool: it was then taken overland to the European
factors on the coast, mainly to Ambriz via Sao Salvador. Earlier
in the century the Bobangi had attempted to take over the Teke
markets by force. After failing to do so, they agreed to continue
their profitable trade, using the Teke as intermediaries, but not
to set up their own market towns on the Pool.
The Bobangi and other Congolese peoples, however, did not
limit their response to the economic opportunities created by the
international system merely to providing its basic requirements:
slaves and ivory. A regional economy was built up, covering the
entire river basin, which made intensive use of the facilities of the
slave and ivory trades: canoe transportation, brass-rod currency,
67

brokerage agencies, commercial relationships. The Bobangi con-
trolled the middle Congo, the artery of the regional economy; but
other riverain people controlled the commerce on all its main
tributaries. The chief item of this regional trade was cassava: the
more both the international and the regional networks grew, the
more the demand for cassava increased. Cassava, which was a food
crop that could be stored in the ground for up to two years, was
grown by the' inland' people away from the rivers, and was traded
to the riversiders, who took it by canoe down the tributaries to
the Bobangi towns on the main river. The volume of the cassava
trade was immense - in 1885 it was reported that 40 tons a day
went down the Alima river alone. Other trade crops included
tobacco, palm oil and palm wine, and vegetable salt. Craft
manufacturing produced cloth, pottery, iron knives and spears,
and canoes. The construction of the great trading canoes, which
were up to twenty metres long, involved the concerted efforts of
large teams of men. Much of the iron and iron work came from
the rich iron-ore district between the headwaters of the Alima and
Kuku rivers (tributaries of the Congo) and the Ogowe. This iron
work was transported by canoe south to Mbe (the capital) and
to other markets in the Teke kingdom, and north and east up the
Ikelemba and Tshuapa rivers to the lands of the Mongo. European
textiles, firearms and ammunition, and other goods were
exchanged for the cassava, iron and other products of the Congo
basin, as well as for ivory, which was the main item of export to
the international system. Up to the 1870s at least, the incursion
of European goods seemed to stimulate the regional economy
rather than stifle it.
The situation in the central Congo basin by the early 1870s was
paradoxical. Despite the disruption and destruction of the slave
and ivory trades, the regional economy flourished. The increased
production was
real growth in the sense that it was based on the application of previously
underemployed productive capacity... The income was widely distributed
because in most cases it went directly to producers... and did not go to enrich
political authorities... It was the traders, above all, who benefited from the
trade, for the output of many producers passed through the hands of each
trader...the trading centres had emerged as the points at which the wealth of
the central Zaire basin was being concentrated.38
38
Robert W. Harms, River of wealth, river of sorrow: the central Zaire basin in the era
of the slave and ivory trade, ijoo-ifyi (New Haven and London, 1981), 70.
68

It was these centres, and the regional trade as a whole, that proved
so attractive to the colonial officials of King Leopold, and yet were
so vulnerable to their intervention.
The Teke kingdom, based on the sandy plateau north of Malebo
Pool, had maintained its integrity, although it had become greatly
dependent upon overseas trade. From the markets on the banks
of the Pool, the Teke supplied the Bobangi with European trade
goods, and took over the produce transported from the interior
forests. This was carried, not generally down the main Congo
river, which passed over a series of hazardous cataracts between
the Pool and the estuary, but via the small rivers and streams that
led through the hilly region north and south of the lower Congo,
to the coast. In the north, the trade was handed over to the Vili
and Lumbo peoples of the Loango coast. To the south, among
the Bakongo, most of the traders were also Vili, who transacted
their business with the European firms on the coast, from the
harbours of the estuary to as far south as Ambriz. British, Dutch,
French, Portuguese, American and other merchants operated on
the Loango coast, on the Congo estuary, and to the south up to
the ill-defined frontier of Portuguese Angola, but the British
predominated. Slaves were still being exported in the early 1870s,
but this was the final flicker of the old trade. The main exports
were ivory, palm oil and palm kernels, gum and beeswax, cotton
and coffee, mostly traded illegally out of Angola.
Ivory was traded from the Kwango river valley by Zombo
traders, but the outreach of the coastal trading economy had not,
by 1870, reached the basin of the northern Kasai and its tributary
rivers. Here, Kuba, Lulua, Pende and other peoples, were still
largely untouched by the outside world. Their rich agricultural
countryside supported only small-scale, regional trade. Much
nearer the Congo mouth, however, vine rubber was being
collected and exported by the middle of the 1870s. Soon wild
rubber was to outrun all other exports. Imports included textiles,
alcohol and firearms. By the 1870s the political cohesion of the
states of the Loango coast had broken down almost completely,
and trade suffered accordingly. Increasingly, more and more of
the trade of the whole Congo area was being concentrated on the
ports of the estuary itself. The agents of European commerce were
making concerted efforts to cut out the African middlemen, and
to reach the economically productive areas. The estuary region
69

had become one of the most important trading areas in Africa for
the British, comparable in value to the oil trade of southern
Nigeria. As in southern Nigeria, the international economy
accelerated political, social and economic diversification. The
economic base of much of western equatorial Africa was pre-
cariously balanced in the pre-colonial nineteenth century. The
region lacked natural resources, such as the oil-palm of West
Africa, and the skilled labour which produced the manufactures
of the towns in the Sudanic zone. This balance was seriously
disturbed by the colonial conquests and by the harsh exploitation
of the concessionary companies. A region with sparse human and
natural resources, it underwent progressive impoverishment
until well into the twentieth century.
EASTERN AFRICA: THE TRADING SPHERE OF THE

SWAHILI-ARABS
By the 1870s the Swahili-Arab penetration of East Africa was

largely responsible for the major restructuring of much of the
region. The focus of this penetration was Zanzibar. From his
island capital, Sultan Bargash ibn Sa'Id exercised some sort of
political authority over much of the 1,200 km coastline from
Brava in the north to Tongui on Cape Delgado in the south.39
He was also attempting to strengthen a tenuous authority over
a few centres in the interior. Almost all the external trade of East
Africa was funnelled through the port of Zanzibar. American,
British, French and German commercial firms dealt in ivory, gum
copal, cloves and hides; most of the ivory went to America and
India, while the sesame went to France. Foreign goods imported
into Zanzibar included cotton cloths, beads, firearms and specie,
the last mainly in the form of Maria Theresa dollars and Spanish
piastres. The trade in cowries was a form of export of specie.
Cowries from the Indian Ocean were transported to West Africa
by German firms in a triangular trade between northern Europe
and the Indian- and Atlantic-Ocean coasts of Africa.
In the earlier years of European economic activity at Zanzibar,
during the rule of Sultan Sa'Id, British trade had been insig-
nificant compared with that of America, France and Germany,
39
C. S. Nicholls, Tbt Swabili coast: politics, diplomacy and trade on the East African
littoral, 1798—il)6 ( L o n d o n , 1971), 321.
7O

EASTERN AFRICA
although the activity of Indian traders, who were already British

subjects, was sufficient to warrant the appointment of a British
consul from 1844 onwards. After the accession of Sultan Bargash
in 1871, and the appointment of Sir John Kirk as British consul
the same year, this situation was reversed, and more British
shipping used Zanzibar harbour than that of any other state. All
shipping companies were able to take advantage of the opening
of the Suez Canal in 1869. In the 1870s British and British—Indian
trade was twice that of Germany and almost three times that of
America. The sultans of Oman and Zanzibar had signed a number
of diplomatic and commercial treaties with America, France and
Britain - the last being primarily anti-slave trade treaties. The
culmination of these was the treaty of 1873, which formally
abolished the export of slaves from Bargash's domains.
Strategically, and to a considerable extent economically,
Zanzibar was an enclave of British Indian-Ocean power; the
consuls were directly responsible to the Bombay government. But
if the island was 'a satellite of Europe's growing power in the
Indian Ocean ',40 it was a client state upon which converged a host
of economic endeavours which were only partly dependent upon
the European capitalist economic system. One of these quasi-
independent economic factors was the role of Indian traders and
bankers. The finances of Zanzibar were controlled by Indians,
who came originally from Kutch, Surat and Bombay. The
controller of Zanzibar's customs — a vital post in the economic
life of the island and of much of East Africa - was an Indian.
Indian bankers financed the caravan trade to the interior, offering
credit at very high rates of interest. Many of the island's clove
plantations were mortgaged to Indian money-lenders. The flow
of trade to and from India — carried in East African or Arabian
ships — amounted to at least two-thirds of that to Europe and
America. Another largely independent axis of trade was that
between Zanzibar on the one hand, and Oman and the Persian
Gulf, the southern coast of Arabia, and the Red Sea on the other.
Arab traders brought silk and cotton goods, coffee, dates and salt
in exchange for slaves, cloves and timber. These Arab traders
benefited from the general economic expansion of the East
African coast.
After ivory, the most important of East African exports were
40
John Iliffe, A modern history of Tanganyika (Cambridge, 1979), 40.
71

cloves, gum copal, hides and cowries. The last three were
gathered or collected, and traded, by the coastal Swahili-speaking
people and their slaves. They were transported by local ships to
Zanzibar, which monopolised the trade of the coast. All these
productive and trading activities profited from the overall
development of the economy of the region. The increase in the
production of cloves on Zanzibar and Pemba islands was
particularly marked. These clove plantations were worked by
slaves in conditions not dissimilar to those of the American south.
Sugar plantations, also worked by slaves, were established by
Arab landowners along the Pengani estuary.
The history of the Swahili-Arab penetration of the East African
mainland can be summed up in one word: ivory. The collection
of wild rubber ousted ivory from its predominant position only
in the late 1870s. Until then, the hunting of ivory and its
transportation to the coast had brought about an intense trans-
formation of all the major, and most of the smaller societies of
East Africa: few people were able to escape its exactions.41 In the
course of the nineteenth century there were a series of ivory
frontiers, which advanced ever deeper into the interior of East
Africa. By the 1860s most of modern Tanzania and eastern Kenya
had been denuded of elephants. The frontier had moved south
into northern Mozambique and the Lake Malawi (Lake Nyasa)
region, west across Lake Tanganyika into the Manyema region
(Zaire), north-east into Bunyoro and Buganda, and north towards
lake Turkana, and the Ethiopian borderlands. Here the elephant
hunters who had trading connections with Zanzibar came into
contact with Oromo and Sidama hunters who sold their ivory into
the Ethiopian trading system. In northern Bunyoro the outreach
of Zanzibar came into contact with the jalldba and other traders
who sent their ivory down the Nile to Khartoum and Cairo. In
the lands between Lake Malawi and Katanga, the Zanzibar system
connected with ivory traders who fed into the Portuguese trading
diasporas from Angola and the lower Zambezi. By the 1870s much
of Bantu Africa was, or had been in the recent past, the scene of
unprecedented slaughter of elephants in the interests of the ivory
trade.
The consequences of this predatory economic activity were
varied. In what is now Tanzania, the passing of the ivory frontier
41
Ibid., 40.
72

EASTERN AFRICA
left something of an economic void, although the manning,

feeding and servicing of the trade caravans from the regions
beyond was still a considerable enterprise. The withdrawal of the
dominant economic factor led to greater political insecurity and
uncertainty, especially as the importation of firearms tended to
increase. Survival became the dominant theme for the local
political leaders and, in the last resort, for their peoples. Some
states became conservative and emphasised the unifying force of
ritual. Other states opted for the barrel of the gun, and recruited
mercenary troops to defend or extend their interests. In practice,
the more successful states combined both the old and the new
techniques, in desperate attempts at political restructuring in the
face of external and internal dangers. Ngoni or Ngoni-like states
still dominated Songea and the southern highlands of Tanzania.
The new Hehe state reached the peak of its power in the 1870s,
and was remarkable for its reliance on spears rather than guns.
Among the Nyamwezi there was two outstandingly successful
war-lords, Mirambo and Nyungu ya Mawe, who incorporated
many formerly independent small kingdoms. Both employed
mercenary troops, called rugaruga; both were politically conserva-
tive and administratively attenuated. In the north-eastern
Tanzania and in southern Kenya, the power of the strife-torn
Masai had greatly declined, to the advantage of the Chaga
chiefdoms and of the stateless Kikuyu. The Kikuyu were
expanding territorially from the mountain pastures of Mount
Kenya and the Nyandarua range; they kept Arab traders at bay,
having transactions only through Kamba intermediaries. In the
western highlands of Kenya, the Nandi had risen in power as the
Masai declined. A new kind of leader emerged, borrowed from
the Masai and called the orkoiyot, who predicted the future, read
the omens, and enabled the Nandi to reach military and economic
decisions. Under the inspiration of their orkoiyot, the Nandi
became the dominant people between the Rift Valley and Lake
Victoria.
Whereas the old-established interlacustrine kingdoms of
Rwanda and Burundi remained largely unaffected by the ivory-
based Arab trading system, Buganda and Bunyoro reacted
vigorously to it. As in Tanzania, there was an acceleration in the
incidence of violence. Bunyoro, under Mukama Kamurasi
(d. 1869) and his successor Kabarega, greatly revived in power,
73

and entered upon a further period of conflict with Buganda. Both

sides equipped themselves with firearms, obtained from the
Swahili-Arabs of Tabora in exchange for ivory. Kabarega
employed Khartoumers to train and lead his newly armed
regiments. Sir Samuel Baker, the agent of Khedive Isma'U,
reached Bunyoro at the head of an armed expedition in 1872.
Mutesa's response was to turn to Zanzibar for support. Stanley's
arrival in Buganda in 1874 via the Swahili-Arab caravan route led
the way to the advent of British Anglican missionaries at the
capital in 1877. Buganda was potentially the stronger of the two
kingdoms. The processes of bureaucratisation and centralisation
were as marked in Buganda as in any East African polity. The
kabaka was at the head of the twin hierarchies into which the
society was divided, the territorial chiefs and the clan heads.
Furthermore, the ruler was able to appoint his own men to most
of the major military and administrative offices of state. The
kabaka personally dominated the kingdom of Buganda.
The partially arabised and islamised region of Manyema to the
west of Lake Tanganyika was likewise personally dominated by
the Swahili-Arab trader, Tippu Tip. Like other Arabs and Swahili
operating in the interior of East Africa, Tippu's aims were
primarily commercial, but when he commenced operating in the
forests of Manyema, among mainly stateless and politically
unstable peoples, he was forced to exercise a kind of suzerainty,
with a Swahili-Arab bureaucracy and with slave armies, so as to
procure ivory. The centre of Manyema, around Tippu Tip's
headquarters at Kasongo, was well-ordered and productive, but
the ivory frontier, which reached well to the west of the Lomami
river and north down the Lualaba towards the future Stanley Falls,
was lawless and violent. It was with Tippu Tip's approval and
support that Stanley started his journey down the Lualaba in 1876,
a journey which was 'decisive in opening the whole Congo basin
to European influence, and indeed in transforming the central belt
of Africa into a paper playground for European politicians.'42
In the wake of the ivory trade came the internal trade in slaves,
and the spread of domestic slavery as an institution among the
agricultural peoples of the East African interior. By the 1870s, in
spite of Zanzibari prohibitions, some slaves continued to be
exported from Africa to the Arabian peninsula and surrounding
42
Ruth Slade, King Leopold's Congo (London, 1962), 23—4.
74

EASTERN AFRICA
lands. But most slaves were used inside East Africa. Many
laboured on the clove plantations of Zanzibar and Pemba, or on
the new sugar and cereal estates on the Swahili coast. The majority
were involved in agricultural production, in a region where
traditional rural economies were coming increasingly under
strain. Slaves were used for porterage and as soldiers. There was
some local slavery - people losing their liberty for a variety of
reasons, but staying put. There was also long-distance slave
trading, mainly from the Lake Malawi region, Katanga, and what
is now north-eastern Zambia. Here Yao (to the east and south of
Lake Malawi), Nyamwezi and Swahili-Arabs had penetrated the
lands of the Maravi, Chewa, Bemba and other peoples west of
Lake Malawi and north of the Zambezi river.
In the mid-century there were three main Yao chiefs who
competed against each other for control of the lucrative slave and
ivory trade routes to Kilwa.' Their towns became bustling centres
of population far larger than any previously known in that part
of East Africa.' By about 1880 'their successors realised that their
long term interests dictated a greater need for co-operation than
for competition... [they] were attempting to deal more efficiently
with their coastal trading partners and with the Sultan of Zanzibar
by encouraging the development of literacy in Arabic script'.43
The Yao were reacting to changes in the East-African/Indian-
Ocean trading system that were only very indirectly, if at all,
influenced by European activities.
The area south of Manyema and Lake Tanganyika, and west of
Lake Malawi, had been penetrated by Swahili-Arab and Nyamwezi
traders since the beginning of the century. By 1870 the Nyamwezi,
known locally as Yeke, under the leadership of a trading chief
known as Msiri, 'the mosquito', had built up a large area of
political patronage in Katanga. This region, called Garenganze,
threatened the economic livelihood and political stability of the
western Lunda kingdom of the Mwata Kazembe and, to a lesser
extent, of the eastern Lunda kingdom also. Msiri based his power
on ivory hunting, elephants still being plentiful in the borderlands
between the two old kingdoms. He also controlled the rich trade
of the Copperbelt. Msiri not only used the caravan route to Ujiji
and Zanzibar, but also by the 1870s had linked up with Ovimbundu
43
Edward A. Alpers, 'The nineteenth century: prelude to colonialism' in
B. A. Ogot (ed.), Zamani: a survey of East African history (Nairobi, new cdn 1974).
75

traders travelling from Lovale, and thus had an outlet to Angola.
North of Garenganze the once powerful Luba state, centred on
the Lomami river, which had previously benefited from Bisa trade
routes to the east and Bihe (Ovimbundu) traders from the west,
was also succumbing to east coast pressure. Tippu Tip conquered
the northern parts of the kingdom, while Msiri encroached from
the south. The rapidly expanding Bemba polity, on the plateau
between the Luapula and Luangwa valleys, owed much of its
political and military success to links with Swahili-Arab and
Nyamwezi traders, notably Tippu Tip. Bembaland, which was
poor country agriculturally, became another great reservoir for
ivory and slaves.44
It must be stressed that alongside, and supporting, the non-
productive ivory and slave trades went a great amount of local
and regional trade, in copper, iron and ironware and, in particular,
food. In 1872 Stanley described the vigorous mingling of regional
and long-distance trade at Ujiji market:
There were the agricultural and pastoral Wajiji, with their flocks and herds;
there were the fishermen from Ukaranga and Kaole, from beyond Bangwe, and
even from Urundi, with their whitebait, which they call dogara... there were
the palm-oil merchants, principally from Ujiji and Urundi, with great five-gallon
pots full of reddish oil, of the consistency of butter; there were the salt
merchants from the salt plains of Uvinza and Uhha; there were the ivory
merchants from Uvira and Usowa; there were the canoe-makers from Ugoma
and Urundi; there were the cheap-Jack pedlars from Zanzibar, selling flimsy
prints, and brokers exchanging blue mutunda beads for sami-sami, and
sungomazzi, and soft.45
By the 1870s maize, which in 1800 had been confined to the coast
where it had been introduced by the Portuguese, was grown
extensively throughout East Africa. Yet even if people had the
means to feed themselves better, there was undoubtedly a
widespread economic and political crisis in the area. A few rulers,
notably Mutesa, Mirambo and Tippu Tip, responded to the crisis
by attempting to increase the scale of their various relationships
and dealings. Most, however, merely struggled for survival.
Sultan Bargash himself, egged on by Consul Kirk, attempted for
the first time to assert his authority in the interior. The crisis was
basically associated with the international trading system, a system
44
A . D . R o b e r t s , A history of the Bemba: political growth and change in north-eastern
Zambia before ifoo ( L o n d o n , 1973), 6, 'Politics and trade'.
45
H . M . Stanley, How I Found Livingstone ( L o n d o n , n e w e d n 1895), 3 8 7 - 8 .
76

FROM ANGOLA TO MOZAMBIQUE
which was experiencing' rising prices, falling profits, and growing

competition for trade. Zanzibar had preferred the cheapness of
informal empire but now found it no longer worked.'46
The island of Madagascar had economic ties with south-eastern
Africa of long duration, but more significant were its growing
links with French commercial expansion. Those went hand in
hand with a remarkable religious and cultural revolution
experienced by the Hova of the dominant state of Merina. Many
of the Hova accepted Christianity in a more radical manner than
did the majority of societies exposed to the alien religion on the
continental mainland at that time. The dilemma facing the Hova
was that this was a Protestant version of Christianity, propagated
by British missionaries.
FROM ANGOLA TO MOZAMBIQUE: PORTUGUESE ECONOMIC

ACTIVITIES AND THEIR OUTREACH
The whole of a wide belt across Bantu Africa, from between the
Dande and Cunene rivers in the west to Cape Delgado and
Lourenco Marques in the east, was influenced, to a greater or
lesser extent, by economic activities generated by the presence of
Portuguese mercantile endeavours on the Atlantic and Indian
Ocean coasts. The influence was largely indirect; the Portuguese
offered the opportunity, the gateway to the world overseas, but
initiatives remained largely in African hands. Even within the
colony of Angola, African chiefdoms were often powerful enough
to be able to control trade and production.
In the 186os the boundaries of the Portuguese colony of Angola
were receding and its economy was passing through a deep
recession. The trade in and status of slaves were abolished
officially between 1858 and 1878 (formal abolition took place in
all Portuguese colonies in 1875), and these measures caused
considerable disruption. The colony experienced a minor cotton
boom during the American Civil War, but this soon collapsed,
and was followed by a severe financial crisis. Furthermore, the
Portuguese conducted a series of disastrous wars, against the
outlying Kongo people in the north and against the ruler of
Kasanje in the east. In 1863 Kasanje was abandoned for the second
time, and Malange, some 50 km from the Kwanza valley, became
46
Iliffc, A modern history of Tanganyika, 48.
77

the staple market, or feira, for northern Angola; Bihe, in
Ovimbundu country, was its southern counterpart. By this time,
however, Kasanje had lost much of its economic importance. By
1866 the Portuguese had withdrawn from most of the centres
north of the Dande river, in the old territory of the Kongo
kingdom, which had been painfully reoccupied in the previous
decade. The ancient kingdom of Kongo had long disintegrated
into three or more separate chiefdoms, although the royal dynasty,
which had religious significance for the Kongo people, still
reigned in Sao Salvador, which had been reduced by the latter part
of the eighteenth century to no more than a village.
There was also a withdrawal in the south from Humbe on the
Cunene river, to Huila, the only Portuguese settlement on the
plateau inland from the fishing port of Mocamedes. Only about
2,000 of the colony's 400,000 inhabitants were white Portuguese,
and most of these lived in Luanda, the capital. The descendants
of mixed marriages settled and traded beyond the frontiers of the
colony, some as pombeiros (agents of Portuguese merchants) or
sertanejos (backwoodsmen). Coffee was grown in the Kwanza
valley, the heartland of the colony, by African farmers and by
Portuguese or Brazilian plantation owners, both groups using
slave (or servile) labour. It was not until the 1890s that the
plantation owners achieved a dominant position, to become the
coffee barons of Angola.47 Sugar was also grown in small
plantations near the coast. Vine rubber, gathered by the Chokwe,
was first exported in 1869, and in time came to rival ivory as the
commercial alternative to the trade in slaves in providing a
foundation to the precarious economy of the colony.
Portuguese control over Mozambique was even more tenuous
than that over Angola. Their rule comprised footholds - or
toeholds — on some of the coastal islets or harbours, from
Mozambique Island in the north to Lourenco Marques (the
modern Maputo) in the south. Lourenco Marques was rapidly
growing in importance as the chief outlet from the Afrikaner
South African Republic (Transvaal) in its hinterland, but the
development of its port and commercial facilities was largely in
British, Dutch and German hands. In 1875 the Portuguese
undertook to construct a railway from Lourenco Marques to
47
David Birmingham, 'The coffee barons of Cazengo', J. Afr. Hist., 1978, 19, 4,
5*3-38.
78

Pretoria. The older pattern of trade was breaking down, with the
Tsonga people acting as middlemen in an ivory-dominated
economic zone covering northern Zulu country, the eastern
Transvaal and Gaza country in southern Mozambique. The most
significant feature of this breakdown was the collapse of the
Afrikaner republic of Zoutpansberg in 1867, after two decades of
intensive elephant hunting.48 The Portuguese trader and free
booter, Albasini, played a dominant role in the affairs of this
hunting state and continued to operate in the north-eastern
Transvaal into the 1880s. By this time, southern Mozambique was
already intertwined with the growth of the mining economy of
South Africa.
The ports north of the mouth of the Zambesi were an extension
of the East African coastal system. Even where the Portuguese
maintained their toeholds — on Mozambique island, Angoche
island and Quelimane — the commerce of the coast was dominated
by Muslim Africans. These traded with Yao and Makua middle-
men for ivory and slaves from the interior. Slaves continued to
be transported in considerable numbers to the Hova and Sakalava
kingdoms on Madagascar until the beginning of the twentieth
century. The only really significant Portuguese presence in the
interior of Mozambique consisted of the immense estates called
praqos da coroa (grants of crown land). The coastal pra^ps, which
produced coconuts and sugar, were amenable to some govern-
mental control. The inland pra^ps, up the Zambezi valley, were
to all intents independent African chiefdoms. By the 1870s these
had coalesced into four conquest states, each ruled by a dynastic
clan which had become to a large extent Africanised. Three were
in the Zambezi valley, between the Shire confluence and Zumbo;
the fourth was in Manicaland, south of the river. This was ruled
by the da Sousa family, and acted as a buffer between the Gaza
Ngoni and the Zambezi praôs. The pra^p chiefs entered into close
relations with neighbouring states and peoples, north and south
of the river. They traded in slaves, gold and ivory, often
employing aggressive and violent methods. By the end of the
1870s Manuel Antonio da Sousa was extending his power
amongst the eastern Shona peoples, who continued to trade with
Portuguese agents.
48
Roger Wagner, ' Zoutpansberg: the dynamics of a hunting frontier, 1848-67', in
Shula Marks and Anthony Atmorc (eds.), Economy and society in pre-industrial South Africa
(London, 1980), 313-49.
79

The Shona chiefdoms occupying the central and western areas

of the Zimbabwean plateau were subject to varying degrees of
tribute relationship to, and at times cattle raiding from, Ndebele
kingdom. This had settled down from being a predominantly
militaristic and raiding state into one based on pastoralism and
agriculture. Ndebele and Shona lands were being increasingly
penetrated by white ivory-hunters and their agents from the
colonies and republics of South Africa. By the end of the 1870s
the demand for labour in the developing diamond-mining industry
of Kimberley was making its impact on these societies. The
Ndebele continued to raid periodically in all directions; one such
raid north of the Zambezi was defeated decisively in the late 1860s
or early 1870s by a combined force of Tonga, Shona and
Chikunda, and the Ndebele did not venture in that direction
again for twenty years. The slave armies of the praêros, known
as Chikunda, became a distinctive socio-economic group, who
traded for ivory far up the Zambezi, beyond the feira of Zumbo,
which was reopened by the Portuguese in 1862, into what is now
southern Zambia and, in the late eighteenth and early nineteenth
centuries, as far north as the Luapula homelands of the Lunda
king, Kazembe. Chikunda founded marauding settlements in the
middle Zambezi valley, up the Luangwa valley and in southern
Malawi; these represented scattered and far-flung enclaves of an
outreach that was in its remote origins Portuguese.
The main trading group north of the Zambezi were the Bisa,
who sometimes competed with the Chikunda, and at other times
co-operated with them. In the eighteenth century the Bisa had
established trading links between Kazembe's kingdom and the
Portuguese on the Zambezi, involving copper and other goods
for African consumption, as well as the usual slaves and ivory.
At Kazembe's capital the Bisa came into contact with caravans
which had started their journeys in Angola. But this long-distance
trade was seriously disrupted by the Ngoni invasions and
migrations of the second quarter of the nineteenth century; it had
never been a route of major importance. In the 1860s and 1870s
the more northernly Bisa traded from Kazembe to Lake Malawi,
where they handed over their goods to Yao, who transported
them to Kilwa; this was despite interruptions by the Ngoni and
the Bemba. But most of the trade of Kazembe, the rich Katanga
(Shaba) region, and the expanding Bemba polity, had been
80

absorbed into the Arab-Swahili East African trading zone. Other
Bisa were trading, mainly for ivory, well to the south of the
Luapula basin, amongst the peoples in what is now western
Malawi and central Zambia. They exchanged ivory and slaves
with Portuguese merchants at Zumbo and Tete, and also with the
praêros.
One of the largest and also one of the most stable states in Bantu
Africa was the eastern Lunda kingdom ruled by the mwatayamvo.
This kingdom was at the centre of a series of conquest states which
had been spread by Lunda lineages across much of the central
savanna south of the Congo basin. The savanna, which was
traversed from north to south by the wooded valleys of the
southern tributaries of the Congo, was rich agricultural and
hunting country. The kingdom of the mwatayamvo reached the
peak of its power and prosperity in the middle of the nineteenth
century, and remained fairly stable for the next fifteen or so years.
The eastern Lunda had trading links with the Portuguese in
Angola, exchanging slaves and later ivory for various luxury
goods, and textiles, but not firearms. The main traders were
Imbangala people, based on Kasanje.
In the earlier nineteenth century the Ovimbundu kingdoms of
southern Angola expanded politically and economically, and by
the second half of the century had eclipsed the Imbangala.
Ovimbundu expansion coincided with shifts in the economic
exploitation of the hinterland of Angola. European demand was
for ivory rather than for slaves. In their search for sources of ivory,
Ovimbundu caravans ranged far and wide. Some were royal
caravans, organised by the rulers of the Ovimbundu kingdoms,
of which Bihe was the largest. Others were organised by rich
commoners, sometimes with Portuguese or mulatto participation.
The ivory trade was largely diverted from the former east-west
route of the Imbangala, running via Kasanje to Luanda, to a more
southerly route which passed through the Ovimbundu lands to
reach the coast at Benguela. Some trade still went through Dondo,
on the Kwanza river, into the old regions of the Angola colony;
but even Dondo, only some 100 km from Luanda, was controlled
by a local chiefdom, only nominally under Portuguese rule.
Initiative in the ivory trade was firmly in African hands.
Ovimbundu caravans penetrated to the Luba Lomami empire
and the Kuba kingdom to the north-east on the fringes of the
81

tropical rain forest, to Lovale and Bulozi to the west, and to the
Ovambo and Okavango regions to the south. In Ovamboland the
Ovimbundu traders competed in the early 1870s with Portuguese
from Mocamedes and African and white South African traders
from Namibia. Similarly, the riverain Lozi kingdom felt the
pressure of competition for ivory of traders linked both to the
Portuguese and to the South African economic systems. The death
of Sekeletu, the second and last king of the Kololo (Sotho)
conquerers of the Lozi in 1864, ushered in a long period of strife.
Kololo rule was overthrown, but this was followed by factionalism
among the Lozi princely lineages, in which traders based in
Angola and South Africa played their parts. It was not until 1878
that Lewanika had managed to consolidate his authority.
Like the Nyamwezi of East Africa, the Ovimbundu were
migrant traders rather than the settler hunters. Outside their
homelands, the Ovimbundu presence was commercial and
peaceful. But the warlike Chokwe, like the Fang to the north of
the Congo basin, became the great elephant-hunting specialists of
the Angolan savanna. In the 18 5 os the Chokwe occupied a small
sandy, infertile country, called Quiboco, on the headwaters of the
Kwilu and Kasai rivers. They had for several decades participated
in the trade in wax, slaves and ivory with Imbangala and
Ovimbundu. They then began to expand dramatically.
Chokwe values and their social system supplied the key factors in giving
Chokwe expansion its unique migratory character.. .With them, it acquired
slave-raiders and thousands of migrants. The pawnship system allowed the
Chokwe to integrate enough alien women to create over-population in
Quiboco, and lineage headmen easily converted their normal village-moving
techniques into methods of migration.49
Chokwe expansion brought a train of violence and disorder into
a previously relatively stable region. Chokwe raiding and hunting
parties were armed with guns; most of their neighbours were not.
By the late 1870s, the collection of wild rubber rivalled the
hunting of elephants in the Chokwe political economy, but
methods remained much the same. Not until the late 1880s and
early 1890s was this expansion at its height, but already in 1874-5,
the Chokwe were intervening in the succession dispute after the
49
Joseph C. Miller, 'Chokwe trade and conquest in the nineteenth century', in
Richard Gray and David Birmingham, Pre-colonial Africa trade: essays on trade in central
and eastern Africa before ifoo (London, 1970), 201.
82

SOUTHERN AFRICA
death of the Lunda mrvatayamvo. The stage was set by the Chokwe
in the west and by Msiri in the east for the collapse of old
Lunda-based order in the southern savanna.
Economically, the ivory trade of Bantu Africa was not totally
disruptive. Nor indeed was the concomitant trade in and use of
slaves. Both trades were exploitative, but both could support
economically productive systems, albeit on a limited scale. But the
political and long-term economic effects of enslavement and
elephant hunting tended towards instability and insecurity, and
this tendency was undoubtedly increasing by the 1870s. There was
much more competition, which resulted in an increase in the
violent methods by which trade was accompanied. Many African
societies in Bantu Africa were undergoing a period of rapid and
unpleasant change; they needed relief from the violence which
surrounded them and which threatened to submerge them. When
direct European involvement came, it largely followed the
well-worn lines of approach of the African, Swahili-Arab and
African-Portuguese traders. In East Africa this was from the coast
westwards; but in southern central Africa it was on the north-
south axes pioneered by the Kololo and the Chokwe rather than
the older east—west routes of the Bisa and Ovimbundu.
SOUTHERN AFRICA
By 1870 the processes of historical change and elaboration in

southern Africa had produced societies which were markedly
different from those in tropical Bantu Africa, and indeed, in the
rest of Africa north of the Limpopo. The only areas that were
comparable were Algeria and especially Egypt. Algeria had
become a French colony of settlement, and in Egypt the infusion
of European capital had resulted in considerable economic and
social change. South Africa contained large areas of European
settlement and was beginning to receive inflows of capital, mainly
at this stage from Britain. South Africa - the region south of the
Limpopo river and the Kalahari desert - comprised four Euro-
pean areas (and a number of more ephemeral smaller ones) and
several African kingdoms and chiefdoms which were still
politically independent. The European areas were the British
colonies of the Cape and Natal and the Afrikaner (Boer) republics
of the Orange Free State and the Transvaal (the South African
83

Republic). The larger independent African states included the

Zulu and Swazi kingdoms, the Tswana kingdoms, and most of
the chiefdoms of the Africans living in the Transkei. Although
the 1860s were a fairly muted decade in South African history,
European territorial expansion continued. The Ciskei, which as
British Kaffraria had become a British territory in 1847, was taken
over by the Cape in 1866. In 1868 Lesotho was annexed by the
British, and three years later was handed over to the Cape to
administer. In 1871 the newly discovered diamond fields of
Griqualand West were annexed as a British crown colony. The
Portuguese foothold of Lourengo Marques had become meshed
into the evolving South African political economy. There were
living in South Africa and neighbouring European colonies and
states, some 2 5 0,000 whites, well over one million Africans, as well
as several tens of thousands of people of mixed descent, who lived
in the Cape, and people from Asia, mainly in Natal. There were
perhaps another million Africans living within the region but
outside the European-ruled lands. White settlers were thus
outnumbered by some eight to one, but this large numerical
disadvantage did not prevent them from dominating much of the
region. The natural environment, however, still imposed
constraints upon effective total domination by either white or
black people.
Apart from the short line built from the harbour to the tiny
town of Durban in Natal in 1859-60, the first railways constructed
in South Africa were from Cape Town to Wynberg and Wellington
in 1863. This 70 km line served the wine and grain farms of the
European settlers of the Western Cape; its track was laid down
on the expensive standard gauge of European countries. When
construction was resumed in the 1870s the lines were built on a
cheaper, narrow gauge of 3 ft 6 in. There were short telegraph
lines in the Cape and Natal, but generally all communications were
poor. Roads, when they had been constructed at all, were bad.
Most routes were mere tracks. There was some horse-drawn
transport, but most of it was by ox-drawn wagon. Economically,
ox wagon transport was much more effective than the head
porterage of Bantu Africa, and somewhat more efficient than the
ox, donkey and camel transport of the lands of the Sudan and West
Africa. But it still meant that the passage of goods and people was
slow, expensive and difficult. Societies and peoples of southern
84

SOUTHERN AFRICA
Africa were isolated one from another. Social and economic

integration, where it occurred, was small and localised. If this
region later became much more integrated as the political
economy evolved and expanded, during the 1860s it was still a
region of societies greatly differentiated by particular circum-
stances, including isolation. This differentiation did not necessarily
follow the racial categories of 'African' and 'European'. While
poor communications emphasised isolation, the business of
transport riding provided opportunities to break out of this.
Africans were much employed in the laborious transport system.
Some Africans owned both wagons and oxen; some African
groups, such as the Sotho, possessed large numbers of horses.
Nearly all peoples in southern Africa in the late 1860s gained
their livelihood from farming. Industries, such as tanneries,
wool-washing plants, and sugar refineries in the European areas,
and iron forging in African areas, were small in size and
organisation, although the distribution of their products could be
quite extensive. The scale of commercial operations had always
outstripped local production, and many of the goods sold in South
Africa were imported from Europe. Most Africans and many
Europeans were subsistence farmers, in the sense that they grew
their own food and could provide many of the other necessities.
There was usually some surplus in these rural societies, which was
appropriated politically or socially (in the form of taxes or
marriage dowries) or was used to purchase consumer goods such
as textiles and coffee. The units of production were families or
segments of lineages, often with the help of servants. Where
markets influenced production, their range and impact was
circumscribed. This type of agricultural production was pre-
capitalist, while the modes of production were varied. Many
European farmers, and some Africans, were however wholly or
partly market oriented. These markets might be confined to South
Africa, with the cycle of production and consumption being local
or regional, or else they might be the outposts of overseas, mainly
British, trading networks. The two agricultural systems existed
side by side, uneasily, their divergencies apparent but not yet
overwhelming.
Superimposed on these rather different agricultural systems,
and closely connected with them, were political and commercial
factors. The political systems of the British colonies, Afrikaner

republics and African kingdoms, in varying degrees, exploited the

agricultural systems. Ulundi (the Zulu capital) and Cape Town
had to be fed, and government officials had to be remunerated.
Capitalist and pre-capitalist states lived in uncomfortable
proximity, the actions of the one affecting the other; the colony
of Natal and the Zulu kingdom are a pair of several possible
examples. Commerce was the means by which agriculture became
capitalist, but it had also, by 1870, deeply penetrated the areas of
pre-capitalist productivity, and this had resulted in further
dimensions of change. Many Africans were already responding to
European political and, more frequently, commercial pressure, by
making available the one item that was in continuous demand by
Europeans, namely their labour. Labour migrancy had been a
feature of some parts of southern Africa for a century, if not
longer.
At the end of the 1860s the Cape Colony had the most
developed economy in South Africa. It had three to five times the
white population of all the other states and colonies in South
Africa put together, and a similar proportion of the foreign trade.
Indeed, this imbalance rather than the precise extent of the
capitalist mode of production was, in the 1870s and 1880s, perhaps
the crucial factor in the historical development of white South
Africa. In 1870 the Cape's exports were valued at £2,569,000, and
the imports were £2,3 5 2,000. The amounts for the colony of Natal
were ,£382,000 and £429,000 respectively. Wool had outstripped
wine as the Cape's principal export, in an overwhelming manner,
as these figures reveal:
Wine GO Wool (£)

1826 98,000 545
1842 45,000 72,000
1861 34,000 1,460,000
Merino sheep, which were the main source of wool, had been
introduced to the eastern districts of the colony in 1827, and wool
farmers, many of them of British origin, had taken the place of
the cattle farmers who trekked north across the Orange river onto
the high veld. Wool production spread from the northern Cape
districts into the Orange Free State. The eastern districts were
86

SOUTHERN AFRICA
more prosperous than the western areas of old European settle-

ment around Cape Town, which nevertheless remained the
capital of the colony. Port Elizabeth, Grahamstown and other
towns were growing quite rapidly in size.
This economic imbalance was reflected in the political friction
between the two parts of the colony. The Cape was governed
under a constitution which had been in operation since 1853. This
provided for a legislative assembly, the members of which were
elected on a limited franchise, based on property qualifications.
Political power was firmly in the hands of the European settlers.
A mainly white electorate voted for white members of parlia-
ment. However, under this representative government system,
ministers were responsible to the British governor, not to
parliament. The settlers in the eastern districts agitated sporadi-
cally for a separate political identity, but these internal differences
were overshadowed by the conflict which dominated most of
the 1860s, between settler politicians and the British governor.
In a period of recession, Britain was concerned to cut down
its financial commitments in the Cape by making the colony
economically self-sufficient. The settlers (or their representatives)
consistently refused to acept the financial implications, notably in
the field of defence, of the British plans for responsible govern-
ment. There was a complete impasse. Elections held in 1868 and
1869 failed to return parliaments which were acceptable to the
British government. Only the economic recovery of the 1870s,
and a change of governors at the Cape, produced a way out of
the doldrums of Cape politics. Responsible government was
achieved in 1872. Combined with the Cape's great economic and
demographic preponderance, this enabled the colony to assert
itself very effectively both against the other South African states
and against pressure exerted by the British imperial government.
The settlers in the eastern Cape were mainly of recent British
descent, whereas those in the western districts traced their origins
to the Dutch, French and German settlers of the seventeenth and
eighteenth centuries. The majority of the non-European people
of the western Cape were landless labourers working on the
European-owned farms; some, especially those living in Cape
Town, were skilled artisans. The African inhabitants of British
Kaffraria/Ciskei retained some of their land. They, and their
neighbours beyond the river Kei, had recovered from the
87

disasters of the 18 5 os, in particular the cattle killing of 18 5 7. Some

Africans living in the colony were already responding to economic
pressures and opportunities by changing their farming practices
and producing food and materials for local markets, such as King
Williams Town (the main town in British Kaffraria), or wool for
the export trade. African agricultural production greatly increased
in the last three decades of the century, spreading from the more
fragmented African groups living in the Ciskei (the Fengu) to the
larger Xhosa, Thembu and Mpondo chiefdoms of the Transkei.
North of the Orange river frontier of the Cape, many Sotho
farmers had also started to grow maize and other crops, and to
produce wool, for the European markets, inside South Africa and
overseas. Economic ties between Basutoland and the eastern Cape
were close. Concern by Port Elizabeth merchants was one of the
factors which led the British government to intervene in the war
between Basutoland and the Orange Free State, which had broken
out in 1865, by annexing the kingdom in 1868.
The republics of Afrikaner farmers on the high veld had been
formally recognised by conventions with the British, those in the
Transvaal in 1852, and in the Orange Free State in 1854. As the
result of its war with the Sotho, and its occupation of the old
Griqua state on the Orange river in the 1860s, the Orange Free
State extended over most of the high veld between that river and
the Vaal. The Afrikaner lands in the Transvaal were not united
(on paper) until i860, and even then the ivory hunters' state in
the Zoutpansberg remained a practically independent polity until
its destruction in 1867. In the northern parts of South Africa there
was still an ivory frontier. In the eighteenth century, European
elephant hunters had found an outlet for their ivory at Cape Town,
while their African counterparts traded at Lourenco Marques. By
the middle of the nineteenth century this ivory frontier had moved
to the northern Transvaal and the fringes of the Kalahari desert.
European hunters and traders from South Africa, and their
African agents operated deep into Matabeleland and Ovamboland
in northern Namibia. This was a hunting system similar to those
of Bantu Africa. Zoutpansberg, the small Afrikaner republic in
the mountains south of the Limpopo river, was an elephant-
hunting state with many similarities to that of Tippu Tip in
the eastern part of present-day Zaire. Hand in hand with ivory
hunting and trading went the spread of firearms. Indeed firearms,
88

SOUTHER N.AFRICA
of models which had become obsolete in Europe, were one of the

largest imports to southern Africa, and by the 1870s nearly all
African peoples were heavily armed with guns.
The Transvaal was ringed by still effective African states - the
Venda and Gaza kingdoms to the north and north-east; the Pedi
and Swazi states to the east, and the Tswana kingdoms, of which
the Ngwato kingdom was the largest, to the west. The Free State
exported some wool to the Cape ports, but the majority of the
Afrikaner farmers were subsistence pastoralists. They shared the
occupation of this land with African farmers who, in return for
the right to' squat', looked after the herds of the white men. North
of the Vaal river the miscellany of Afrikaner and African states
exported some hides and much ivory. In 1864, 85 per cent of the
ivory came from Zoutpansberg, much of it exported through
Lourenco Marques. But most of the people in the Transvaal were
cattle farmers.
The British colony of Natal was sparsely populated by
Europeans and heavily populated by Africans. Some of these
Africans lived on reserves, but many were farmers on European-
owned land. The 18 5 os were a decade of intense speculation in
land by European companies and individuals in Natal and on the
high veld. In consequence, much of the land was in the hands of
absentee companies, which were paid rents by the African
occupiers (a practice known as Kaffir farming), and a shortage of
available land was felt by Africans and Europeans alike. After
Natal became a British colony in 1845, many of the Afrikaner
cattle farmers withdrew to the high veld; the sprinkling of British
settlers then experimented with various export crops. Sugar,
which had been introduced from Mauritius in 1847, survived with
difficulty until in i860 Indian labourers were imported to work
on the estates. The Natal authorities could not persuade Africans,
from within the colony or beyond, to engage in plantation labour.
Apart from sugar, customs dues on the export of ivory from the
interior and taxes collected from the colony's African population,
were the chief sources of revenue. A large proportion of Natal's
trade was with African people living throughout south-east
Africa, so much so that it was, in the scathing words of the Cape
politician John X. Merriman, 'a white forwarding agency in [an
African] territory'.
To the north-east of Natal was the Zulu kingdom, militarily
89

perhaps the most powerful African state in southern and central
Africa. In 1870 the kingdom was still ruled by Mpande, brother
of Shaka. The main external pressure on the kingdom was from
Afrikaner farmers in the Transvaal. There was some internal
friction in the 1860s between factions led by Mpande's eldest sons,
one of whom, Cetshwayo, succeeded him in 1872. The Zulu
economy was both socially complex and materially self-sufficient;
it was based on cattle, access to which was according to age and
rank. The activities of traders from neighbouring territories seem
to have had only a superficial effect on the way of life of the people
of Zululand. The mainstay of this trade was the importation of
firearms, and the most successful European trader, John Dunn,
significantly adopted Zulu culture, and became a quasi-Zulu chief.
After forty or more years of contact with Europeans, the Zulu
political economy remained largely intact.
The majority of the Zulu were still held firmly in the different production
communities of the kingdom, moving from one type to another as they grew
older and their status altered... throughout the reigns of the kings Zulu labour
expanded within the commoners' homesteads, continued to support the bulk
of the population, and the surplus which was drawn from them by the king
through the military system created the basis for his material power and
authority, together with that of the officials with and through whom he ruled.so
Although there were some obvious similarities between

agricultural systems operating in the same environment, the
contrast between the political economy of the Zulu kingdom and
that of the colony of Natal was potentially a source of severe
conflict.
The 18 5 os had been a decade of moderate prosperity for many
of the communities in South Africa, one of the major exceptions
being Africans living in the Transkei. As far as the Europeans
were concerned, more immigration took place during this period
than in any previous one. In contrast, the 1860s were a period of
acute depression. In 1862 much of South Africa was afflicted with
drought, which persisted season after season without much
interruption until 1870. Some parts of the high veld and
Basutoland had had poor rains for much of the previous decade,
but the climatic situation in 1862-3 was exceptionally bad. As well
as food supplies, a wide range of economic activities were
s0
J e f f G u y , The destruction of the Zulu kingdom ( L o n d o n , 1979), 15, 18.
90

SOUTHERN AFRICA
adversely affected. A French missionary in Basutoland wrote in
1863:
There is not a trace of vegetation left; cattle, horses, and sheep are dying by
the thousand. The sand has accumulated in drifts.. .Hunger is beginning to
make itself felt everywhere. Basutoland [Lesotho] which... is the granary of
the Free State and part of the [Cape] Colony, has been completely drained. The
Dutch farmers continue to go there with wagons full of sheep on the hoof
which they transport in this way because the latter are no longer in fit condition
to attempt to walk. They also have ploughs and merchandise of every
description with which they hope to induce the Basuto to part with their
remaining corn. Provisions can no longer reach them from the seaports for
lack of transport. Laden wagons are stranded on the highway because the teams
have died.51
Naturally, severe economic recession followed in the wake of this

drought, a recession exacerbated by a number of external factors.
As a result of technological changes, the Yorkshire cloth manu-
facturers switched from short-staple (the wool produced in South
Africa) to long-staple wools. During the American civil war
shippers found it difficult to penetrate the American market, and
after the war the wool market collapsed; sheep farmers in South
Africa faced lean times. The nascent banking system found itself
with too many creditors and too little money, and imported goods
could not be sold. The British Indian government placed a ban
on the flow of Indian workers to Natal. Europeans migrated from
South Africa to the United States and Australia. Within South
Africa, whites left the coastal colonies for the high veld. The
amorphous South African Republic in the Transvaal was in
particularly bad shape. Its government, which was divided by
internal dissensions, could find no ready cash and paid its few
employees with credit notes. The recession was also one of the
main factors in the political instability of the Cape.
There seemed to be no end to the drought and depression. As
the Suez Canal approached completion in 1869, some people in
South Africa considered that there would be even graver effects
on the economy of the region, with the diversion of the valuable
shipping trade. But the commercial infrastructure of South Africa
in general, and of the Cape in particular, did not entirely lack
resources and enterprise. South Africa was able to participate in
51
F. Daumas, February 1863, published in F. C. Germond, Chronicles of Basutoland
(Morija, 1967), 459.
91

the improvement of the international economy in the early 1870s.

The price of wool rose again. Many Cape farmers prospered
during the ostrich-feather boom of the 1870s, as did their
counterparts in the western and central Sudan. But above all, the
South African political economy was able to react positively (if
not efficiently or humanely) to the discovery of diamonds in 1868.
Diamonds - and later gold - had the most profound effect on the
course of South African history, and were to accentuate the
differences between it and the other regions of the continent.
CONCLUSION
What can be said to draw some distinctive strands out of this

tangle of human activities and aspirations that make up an
overview of Africa on the eve of partition, an overview that is
of necessity discursive and fragmentary? The dominant factor in
the history of the continent at that time was the growing
relationship between Africa and the wider world, in particular
Europe. This relationship of course varied immensely from one
part of Africa to another. There is no doubt that some European
commodities had penetrated to almost every part of the continent
by the 1870s: historians have only to look hard enough, and they
will find. This commodity penetration, however, did not
necessarily imply that an alien capitalist economy had gained the
upper hand. Even on the West African coast, where features of
the European economy were an integral part of the economic
system, the balance between African and European — between
indigenous and alien - was still finely poised, down to the
early 18 80s. Nor was European economic penetration yet a pre-
ponderant political factor among the societies of Lower and
Upper Guinea. The forces making for state building and for state
disintegration were basically still those that had been operating
for the previous two or three centuries, and perhaps even for a
millennium or so. Certain elements of European (or Middle
Eastern) political technology had been introduced, notably
firearms, but these had been easily incorporated into existing
political structures.
In certain areas of the continent there was already European
political control. In South Africa and Algeria this was quite open:
in Egypt it was more veiled. In these countries the European
92

CONCLUSION
capitalist system had made major and disruptive inroads into the
indigenous political economies, and was already a powerful
influence on the remote peripheries of these centres - over much
of south-central Africa, the hinterland of Algeria and, as far as
Egypt was concerned, the upper Nile. In the Swahili-Arab zone
of East Africa some elements of the European capitalist economy
had intruded quite drastically, but with little in the way of serious
political control. It can be argued that Zanzibar was already an
enclave of European capitalism, and that East Africa was an
enormous economic periphery of this enclave. But in East Africa
the special demands of the European economy went hand in hand
with the special demands of Middle Eastern slave-using
economies. These dual, often conflicting demands, were one of
the main reasons for the scale and the violence of the ensuing
disturbances. The situation was not dissimilar on the upper Nile,
where European methods and ideologies of economic exploitation
clashed with those of the Turco-Egyptian ruling elite and with
the practices of local Muslim entrepreneurs and merchant princes.
Another obvious point that emerges from this overview of
Africa on the eve of partition is the kaleidoscopic variety of
African political societies. The 'political world' seen by different
African polities (or, to be more precise, by their leaders) differed
vastly in size and shape. The ' political world' of the Sultan of
Sokoto was enormous. It extended from the Senegal to Istanbul
and Mecca, from the Mediterranean to the Guinean forest zone,
possibly even to the coast of West Africa. Europeans were
certainly a part of this world, but in the 1870s only a very remote
and peripheral part. Ideologically, the Weltanschauung of the sultan
of Sokoto comprised the faith of Islam, from which Christians,
while not quite infidels, were excluded as inferiors. At the other
end of the scale, and of the continent, it is doubtful whether the
world of Cetshwayo extended beyong his Zulu kingdom and
those black and white South African groups who were his
immediate neighbours. The cosmology of the Zulu was self-
contained; there was no place in it for aliens such as Europeans.
But for Cetshwayo these very Europeans in South Africa were
a major - perhaps the major - political and economic problem.
This exercise of considering different variations of what con-
stituted the 'political world' can be applied almost endlessly: a
short list of examples might include Yohannes IV of Ethiopia,
93

Tippu Tip, Sultan Bargash of Zanzibar, the king of Dahomey
and the asantehene, the merchant princes of the Niger delta trading
states, Msiri and Mirambo, Kabaka Mutesa of Buganda, Lewanika
of Bulozi, the reth of the Shilluk, a Zande chief.
Of course this is only an historiographical game, which should
not be pushed too far. Serious studies of 'political worlds',
Weltanschauungen and cosmologies require very precise analysis of
complex and often obtuse data. A study of the early colonial
experience of the Lozi provides an illuminating insight of how
a considerable proportion of the processes and boundaries of Lozi
history and society were deliberately hidden from alien
investigation:
The most important result of Lewanika sanctioning an official version of Lozi

history was to strengthen the shutters behind which lay much of the propulsive
force of Lozi society, for the litaba z" sichaba (history of the nation) set limits
upon what the Lozi elite were prepared to allow whites to be interested in
through its very subject matter.52
But to return to the more straightforward concept of size and

shape, it does seem that, although some of these ' worlds' were
very large and extended well beyond Africa - to Europe, the
Middle East, even to India - none of them embraced the whole
of Africa, or saw 'Africa as a whole'. Perhaps some members of
the Turco-Egyptian ruling elite (almost certainly Isma'Il) could
see political geography in this way. This group, however, were
not really 'African' any more than South African whites were
' Africans' in this context. And very few whites in South Africa
thought in this way.
This kind of approach leads to at least one tentative conclusion,
namely that the partition occurred, when it did, so quickly and
completely, because Europeans (rather than Africans) were the
first to develop the idea of 'Africa as a whole'. Jules Ferry,
Bismarck, Lord Salisbury, King Leopold, Rhodes - all these -
were able to conceive of the whole of Africa in the context of the
wider strategies of imperialism. Menelik II of Ethiopia, the one
African ruler to resist the Europeans with complete success was
also the only one, it appears, to see the local conflict in Ethiopia
52
Gwyn Prins, The hidden hippopotamus: reappraisal in African history: the early colonial
experience in western Zambia (Cambridge, 1980), 29.
94

CONCLUSION
as part of a universal struggle of ' the Europeans' against ' the

Africans'.
It is noticeable that the size and shape of African-perceived
worlds seems to narrow quite sharply below latitude 10 °N and
again, still further, below about latitude 10 °S. Southern Africa
in the 1860s had the same 'end of the world' quality, though not
in such an extreme form, as places like Tasmania or Patagonia.
It was certainly so considered by most British governors of the
Cape Colony. Even the whites tended to narrow their horizons.
The trekboers had more or less deliberately chosen to cut their
links with the wider world, and the whites of Natal had an out-
look hardly less parochial than that of the Zulu. The limited
nature of African perceptions of their political world in the
southern part of the continent when compared to the north might
explain some of the different consequences of the colonial
experience in the major areas of Africa.
Long-term effects of the relationships between Africa and the
outside world were discernible by the beginning of the last third
of the nineteenth century. Widespread ideological and cultural
changes had already taken place as a result of African experiences
of the Muslim Near East and Christian Europe. The introduction
of new food crops, such as maize and cassava from the New
World, was producing beneficial changes in the performances of
African population groups. Nonetheless, there is epidemiological
and some cultural evidence which suggests that the population
of many of the peoples of tropical Africa was declining by the
middle of the century, a pattern that the early decades of colonial
rule does not seem to have greatly altered.
Finally, why did the partition take place? The overall evidence
suggests that the reasons were results of shifts in the international
balance of power and of trade amongst the European capitalist
states rather than changes within Africa itself. Trying to look at
the world as it must have appeared from within Africa in the early
1870s, one can see few signs to suggest that alien partition was
inevitable a decade later.
95

CHAPTER 2
THE EUROPEAN PARTITION OF AFRICA:

ORIGINS AND DYNAMICS
SEEING AFRICA WHOLE
In 1873 Jules Verne's Around the World in Eighty Days celebrated

the spectacular conquest of distance, especially since about 1860,
by the adaptation of the steam-engine to trans-continental and
trans-oceanic transport. To the suggestion that the world was still
' assez vaste', Verne's hero retorted laconically:' II l'etait autrefois'
('It used to be'). The terrestrial globe had indeed suddenly
contracted; and itsfinitedimensions had now become of practical,
and not merely of 'philosophical', interest to the industrialised
societies of the West. The planetary stock of markets and
resources was evidently none too large; it seemed already possible
in principle to foresee the day when its saturation and exhaustion
might impose an absolute limit upon the growth of industrial
economies. This ultimate catastrophe seemed indeed to be fore-
shadowed by the onset in the early 1870s of the so-called 'Great
Depression', essentially the effect of the failure of domestic and
foreign demand to keep pace with the productive capacity of
increasingly mechanised industries. Existing markets seemed to
be satiated. Prices, profits and interest-rates fell, apparently
inexorably, until the mid-1890s. The overall effect was a sharp
retardation in the growth-rate of the industrial economies after
some twenty-five years of unprecedented acceleration, and, for
some of these economies, an especially severe recession between
1878 and 1884. Moreover, the existing overseas outlets for
European capital now no longer seemed sufficiently safe or
remunerative. British capital export between 1875 and 1879 was
negligible by comparison with the previous five-year period; and
from 1880 to 1904 the average annual export of British capital was
little more than half that of the period 1870—4.'
1
H. Feis, Europe: the world's banker, rSyo-iyif (New Haven, 1930; reprinted New
York, 1965), 10-23.
96

SEEING AFRICA WHOLE
In France, where retardation of the growth-rate had already set

in during the 186os, industrial growth remained slow and hesitant
until the later 1890s. The more dynamic German economy
resumed a high rate of growth during the 1880s. But even in
Germany the phase of comparative stagnation initiated by the
'great crash' of 1873 did not fully bottom out until 1879-80; in
the early 1880s recovery was still hardly perceptible, and the crisis
of confidence persisted. Moreover the depression had overtaken
both the French and the German economies before either had
developed sufficiently to offer an effective challenge to the
world-wide economic preponderance of Britain. British manu-
facturers and merchants, thanks to their half-century lead in
industrialisation and their expert knowledge of the requirements
and functioning of overseas markets, were in the 1870s still able
in most commodities to out-trade and undersell their continental
competitors; and this not only in Britain's vast overseas posses-
sions, but in the markets of Latin America, East Asia and coastal
Africa - in fact, wherever free trade prevailed, or could be caused
to prevail by the guns of British warships. Only in the com-
paratively limited colonial holdings of other powers could the
keen edge of British competition be blunted by discriminatory
tariffs and other restrictions. The obvious solution for these
powers was to add to their colonial holdings; but after 1815
British naval hegemony was very effectively used (and nowhere
more effectively than on the African coasts) to discourage
annexations which would create new quasi-monopolistic trading
enclaves for other powers. This system of 'informal empire',
whereby British sea-power safeguarded and even created the
conditions of free trade which guaranteed Britain's economic
preponderance, was usually in British eyes clearly preferable to
formal possession, with its attendant expense and responsibilities.
Until the 1880s, therefore, British policy-makers normally
thought, not in terms of'staking out claims', but of preventing
others from doing so. This was particularly true of tropical Africa,
where experience throughout the nineteenth century seemed
clearly to demonstrate the greater profitability of informal empire.
For other powers, however, formal possession meant the
elimination of British competition and the acquisition of a
permanent title to a share in a limited market. Moreover, by the
later 1870s formal possessions on the coast were acquiring a new
97

THE EUROPEAN PARTITION OF AFRICA
importance for some makers of opinion and even of policy. In
the 'consumer famine' of the 1870s the prospect of opening up
previously inaccessible and supposedly lucrative interior markets
became increasingly attractive, and a coastal establishment might
become the base for the penetration of these markets by that
potent, almost magical, symbol of the ultimate omnipotence of
steam technology — the railway. The long-distance railway already
appeared to have wrought economic miracles in North America,
and as early as 1867 Ibsen's Peer Gynt was in fantasy applying
its magic to the African interior, building railways across the
Sahara to develop Timbuktu and Bornu and travelling through
Ethiopia 'safe in a train to the Upper Nile'. Reality (of a sort)
did not lag far behind fantasy. In 1873, the very year of Verne's
Tour du monde with its ' lyrisme ferroviaire', Khedive Isma'Il was
inaugurating at Wadi Haifa a Sudan railway intended to open up
his African possessions. In 1879 m e French launched in the
Senegal hinterland the first deliberate European attempt to create
a large territorial empire in tropical Africa. This move was in
origin a plan to open up the supposedly rich interior markets of
the western Sudan by railway construction on a grandiose scale:
the Senegal-Niger line and the celebrated (but still non-existent)
Trans-Saharien.
Another technological revolution dating from the 1860s, that
in the design of small-arms, may also have encouraged ideas of
staking out claims by discounting the factor of local resistance.
By 18 70 the development of an effective breech-loading mechanism
had enabled the rifle, with its long-range accuracy and rapid rate
of fire, to supersede the muzzle-loading musket, unchanged in
essentials since the seventeenth century. By 1889 Hiram Maxim,
abandoning the multi-barrelled monstrosities of Gatling and
Nordenfeldt, had perfected, in a conveniently portable form, the
first truly automatic firearm. Less than a decade later, Hilaire
Belloc's 'Modern traveller' could blithely sing:
Whatever happens, we have got
The Maxim gun; and they have not.
These improvements were not of course a necessary condition for

the European partition of Africa. Frenchmen had already
subjugated Algeria, Turco-Egyptians the Sudan, and Afrikaners
and Englishmen South Africa, without benefit of armes perfection-
98

SEEING AFRICA WHOLE
nets. But the small-arms revolution may well have not only
encouraged, but accelerated, the scramble - especially by some-
times enabling quite small European expeditions to plant flags
and extort 'treaties' deep in the African interior.
Before Africa could be partitioned as a whole, it had to be seen
as a whole. Until the 1870s, 'Africa as a whole' had been a purely
geographical concept, of no practical relevance to the European
politicians and merchants concerned with the continent. Much of
Africa still remained what it had been to the first Europeans who
circumnavigated it: a series o f coasts' - Barbary Coast, Windward
Coast, Grain Coast... Swahili Coast, Somali Coast — surrounding
a vast enigmatic blank. Nor, except in South Africa and to a
limited extent in the West African bulge with its trans-Saharan
trade, had the precise political and economic contents of this blank
usually been relevant to the mainly water's-edge activities of
Europeans. As late as the 1840s, Anglo-French confrontations on
African coasts were seen by London and Paris, not as disputes
about Africa, but as disputes on the periphery (as it happened,
an African periphery) of fields of conflict whose centres lay
elsewhere. Disputes on the northern and north-western coasts
were a subordinate part of Mediterranean naval strategy. East-coast
disputes, and disputes about Madagascar, were similarly related
to naval hegemony in the Indian ocean; while from Zanzibar
northwards, these disputes were also linked to power-conflicts in
the Persian Gulf, on the western approaches to India. Strategically,
'Africa' had no existence except on its coasts. Except in South
Africa and marginally in West Africa, the interior of Africa
was, from the European viewpoint, almost as non-existent
economically.
In the early 1840s, with so much of the interior still a blank,
'Africa as a whole' was still a concept almost without meaning
except in its application to a mere land mass. A generation later,
however, increased knowledge of the interior had inspired, "in
Winwood Reade's Martyrdom of man (1872), a serious attempt —
perhaps the first - to consider the past, present and even future
of all Africa and its peoples. By 1876 European policy-makers
were also beginning to think in terms of' Africa as a whole'; and
some of them already believed that 'Africa' was a prize worth
competing for. In February 1876 the direction des Colonies at Paris,
pressing for more active competition with the British, evidently
99

assumed that before long the whole continent, coasts and interior
alike, must fall under either British or French domination. By
December 1876 the British colonial secretary, Lord Carnarvon,
not content with plans for the consolidation of informal British
hegemony on both the East and West coasts, was anxious to
extend British influence far into the interior, northward along the
spine of southern and eastern Africa - to extend it indeed until
it reached the African possessions of an Egypt which was already
a British client-state. 'We cannot', wrote Carnarvon, 'admit rivals
in the East or even the central parts of Africa... To a consider-
able extent, if not entirely, we must be prepared to apply a sort
of Munro [sic] doctrine to much of Africa.'2
THE MOTIVES FOR TERRITORIAL ACQUISITION
Even to the British, the extension of informal control to the

interior now seemed a possible way of opening up new markets.
Although the water's-edge trade was still modestly flourishing,
especially in the Niger delta, at the Congo mouth and at Zanzibar,
its volume depended on the activity of African producers and of
African or Arab middlemen. It seemed to ' go without saying' that
Europeans would be more efficient both in organising production
and in marketing the product. The water's-edge trade, especially
in West Africa, was moreover notoriously vulnerable to political
conflict in the interior. At Lagos, such conflict had hindered the
development of trade for decades; in Sierra Leone, it led to
unpredictable fluctuations in the volume of trade and therefore
of revenue. European attempts to pacify these conflicts were rarely
successful unless, as in the Niger delta, the African belligerents
were accessible to gunboats. Informal control in the interior,
which would not of course exclude the judicious application of
armed force, would in principle solve this problem. In spite of
the disappointing results, both political and economic, of earlier
small-scale experiments in European control of the tropical
interior on the Senegal and the lower Niger, this solution seemed
increasingly attractive as economic depression persisted and
technological confidence developed. It seemed especially attractive
2
Public Record Office (P.R.O.) 50/6/54 (Carnarvon Papers), Carnarvon to Sir Bartle
Frere, 12 Dec. 1876: cited N. Etherington,' Frederic Elton and the South African factor
in the making of Britain's East African empire', J. Imp. Catallh Hist., 1981. 9, 5,
z5 5-74. at 267. The potential rival was King Leopold (see below, p. 27).
IOO

where, as on the Guinea coast, the failure of trade to expand in

the later 1870s was often attributed to 'unproductive' African
middlemen, or to the tolls levied by African states in the interior.
By the later 1870s economic depression and European wishful
thinking, reinforced by the reports of explorers who rarely failed
to emphasise the supposed economic potential of the lands they
had explored, had generated totally mythical El Dorados in
tropical Africa: El Dorados of vast, fertile, empty lands, African
'Sleeping Beauties' awaiting the magic kiss of European energy,
skill and capital; or, in bewildering contradiction, manufacturers'
El Dorados of millions of eager potential customers. Some
leading policy-makers, especially in France, saw in the interior of
Africa markets and outlets for capital which would not merely
palliate but cure the malaise of the industrial economy. This myth
neatly complements (and may indeed have helped to generate) the
'Leninist' 3 myth that by the late nineteenth century political
control of new African markets and outlets for capital had become
an ineluctable necessity for the further development - indeed, for
the survival — of European capitalism. The statistical evidence
gives no support to either myth, except conceivably in South
Africa. In the late 1860s, when Britain was by far the most active
of tropical Africa's trading partners, this trade probably represented
less than 1 per cent of Britain's total extra-European trade. But
after partition, and even as late as 1909-13, it still represented only
about 2 per cent. Capital export to tropical Africa, whether by
Britain, France or Germany, was statistically insignificant before
1914. British capital export even to Egypt represented by 1913
only about 1.3 per cent of her total capital export to non-European
regions; as a trading partner for Britain, Egypt under Lord
Cromer and his successor was for many years considerably less
active than she had been under Khedive Isma'Il.4
Nevertheless, by the 1880s the belief in tropical African
markets, if not as El Dorados at least as economic palliatives,
seems to have become 'conventional wisdom' in Europe. In
1
'Leninist' is convenient rather than strictly accurate. For Lenin himself, the age
of imperialism as the 'highest stage of capitalism' did not begin until c. 1900; and in
his treatise he was concerned to analyse, not the panition of Africa, but the First World
War as an imperialist conflict for the re-partition of the world: V. I. Lenin, Imperialism,
the highest stage ojcapitalism (Petrograd, 1917; English-language reprint, Moscow, 1970),
9-10, 79, 85-6, 118-19.
4
R. Robinson and J. Gallagher, Africa and the Victorians (London, 1961), 6—7. Feis,
Europe: the world's banker, 17-23, 51-7, 75—8.
IOI

September 1884 the French diplomatist Alphonse de Courcel,

who was certainly no believer in El Dorados, had to decide
whether Bismarck's offer of collaboration against Britain in Africa
was anything more than a stratagem for embroiling Britain and
France. Courcel concluded that the prolonged crisis of 'over-
production' had at last constrained Bismarck to acquire African
colonies and that his offer was therefore not a mere trap. Nor,
considered simply as palliatives, were all the supposed African
markets necessarily fantasies. Britain, apparently with the most to
lose by the transition from informal to formal empire, did not in
fact suffer economically. Her £2.3 million of annual trade with
tropical Africa in the late 1860s had by the early 1890s nearly
doubled to £4.4 million, in spite of the general fall in prices by
about one-third and the loss to other powers of trading outlets
previously under informal British control; by 1909-13 it was
worth £14 million.5 This was of course still only about 2 per cent
of Britain's total extra-European trade, and these figures evidently
do nothing to rehabilitate Leninist models of capitalist imperialism.
But they do suggest that, for Britain at any rate and probably for
other economically developed powers, formal empire in tropical
Africa brought with it appreciable though modest economic
advantages.
The economic advantages actually achieved by formal possession
are not however strictly relevant to the problem of motivation.
What matters is not the results, but the expectations; nor does
it matter whether these expectations are rational or 'mythical',
so long as they inspire specific decisions to acquire territory. Two
at least of the policy-makers who promoted the French advance
into the Niger Sudan in 1879 — Charles de Freycinet, a polytech-
nicien and a railway engineer, and Maurice Rouvier, a Marseilles
banker and businessman - were certainly inspired by economic
expectations: expectations which turned out to be almost entirely
'mythical' but which as motives to action were thoroughly
effective. They were soon followed, both in France and else-
where, by other economically motivated imperialists: Jules
Ferry, Leopold II, George Taubman Goldie, Cecil Rhodes,
Eugene Etienne, Joseph Chamberlain, to name only the most
prominent. Some of these men, notably Leopold and Rhodes,
combined the roles of political policy-maker and great financial
5
Robinson and Gallagher, lot. cit.
IO2

magnate. Even when the financial magnates were not themselves

political policy-makers, they were usually well placed to exercise
influence over policy. Yet this influence was by no means always
crucial; and the actual operations of the capitalists in tropical
Africa, though often of course vitally important to the particular
interests concerned, were before 1914 never a statistically sig-
nificant proportion of the capitalist economy as a whole.
The Leninist model does not however exhaust the category of
economic theories possibly relevant to the partition of Africa. By
the mid-1870s consciousness of a shrinking and all too finite globe,
the apparent satiation of existing markets, the temporary absence
of new opportunities for safe and remunerative overseas invest-
ment, and the evolution of a new map of Africa embodying some
determinate interior content, had conspired to present the African
interior as the world's last great untapped reservoir of markets,
resources and possible investment opportunities. This image
developed precisely at the moment when the growth of techno-
logical confidence was encouraging the belief that political control
and economic exploitation of the interior had at last become
feasible operations; within five years or so of the development of
this image, the scramble for Africa had begun. This model can
also be used to explain, with apparent plausibility, why European
annexations of tropical African territory were so trivial and
infrequent between 1815 and c. 1880. For Britain, down to about
1850 the sole effectively industrialised Power, informal empire
sufficed; while France and Germany, even after their rapid
industrialisation from about 1850, had until the mid-1870s
encountered no problems of economic growth acute enough to
act as a spur to African empire. Nor had either the new technology
or the new African geography yet inspired the belief that the
political control of Africa might become an important factor in
the development of European economies. Hence the long stand-
still, followed by very rapid action after about 1880.
This model is however certainly misleading as an explanation
of the standstill. British relations in Africa with other European
powers, and especially with France and Portugal, demonstrate
very clearly that the crucial factor between 1815 and 1880 was not
the mere absence of a strong economic incentive, but the very
positive presence of a British deterrent to annexation. Only in
1830, when France seized Algiers, was this deterrent successfully
103

challenged; thereafter London was determined that there should

be no further French North-African acquisitions or hegemonies,
with their implied threat to British naval preponderance in the
Mediterranean. Palmerston's reaction to the French sponsorship
of Muhammad 'All of Egypt is well known; the magnitude of the
crisis reflects the fact that the policy of the French Prime Minister,
Adolphe Thiers, threatened British preponderance not only in the
Mediterranean but throughout the Middle East and the western
approaches to India. Less publicly, however, in the 1830s France
had been successfully deterred from seizing Tunis, and when in
1844 she became involved in war with Morocco, the naval
operations of the French were monitored by the Britishfleetand
London successfully forbade to a victorious France any annexation
of Moroccan territory.
Thiers' major challenge to British hegemony in the Mediter-
ranean and the Middle East had been accompanied by subsidiary
challenges, prompted above all by the ministere de la Marine, to
British informal empire on the west and east coasts of Africa. On
the Guinea coast, it was hoped to foster French trade, and possibly
to develop a naval station, by the annexation of comptoirs protected
from British competition. On the east coast there was to be an
outwork of the planned citadel of French influence in the Middle
East, but the Marine also hoped to restore the French position in
the Indian Ocean by establishing, in Madagascar or on the
mainland, a naval base as a replacement for He de France
(Mauritius), lost to the British in 1815; some senior French naval
officers hankered after the whole of Madagascar as a ' substitute
India'. Francois Guizot, who replaced Thiers in October 1840,
was anxious on grounds of general European diplomacy to
resume cordial relations with Britain. But his withdrawal from
these adventures had, for domestic reasons, to be gradual and
unobtrusive. It required some British pressure to bring him to
'neutralise' the three new West African comptoirs by imposing a
regime of complete free trade and to announce in March 1843 a
politique de points d'appui which abandoned all serious territorial
expansion. Had there been in the early 1840s another major power
interested in Africa, and willing to support France in these
challenges to British informal empire, the coastal partition of
Africa would probably not have been delayed until the 1880s. But
until 1884 France remained, to her disadvantage, isolated against
104

Britain in Africa. Although the Marine continued intermittently

to look for strategic footholds in East Africa, it was invariably
checked when its activities threatened a serious confrontation
with London. In 1860 Napoleon III rejected proposals for a major
French establishment on the Red Sea and Somali coasts rather than
risk such a conflict; and the Anglo-French Agreement of March
1862, guaranteeing the independence of Zanzibar, was in effect
an undertaking by France not to challenge Britain's hegemony on
the East coast between Cape Guardafui and Mozambique. After
1870, the dangerously weak and isolated position of France in
Europe for some years constrained the French Foreign Ministry
at the Quai d'Orsay to even greater caution in handling local
disputes with England on both the east and west coasts of Africa.
British sea-power, merely displayed against the French, was
used with brutal directness against a minor power like Portugal.
In 1846 Palmerston, determined to keep the Congo open to
British trade, repudiated earlier British acknowledgements of
Portuguese sovereignty on both banks of the Congo mouth. In
1856 he warned Lisbon that any attempt to extend Portuguese
occupation to this region would be 'opposed by Her Majesty's
naval forces';6 and in July 1857 the navy expelled a Portuguese
military detachment from the north bank of the maritime Congo.
In 1857 the industrial economy was non-existent in Portugal; and
in the early 1840s it had been rudimentary even in France. Yet
the absence of industrialisation did not prevent both these Powers
from attempting to stake out claims on the coasts of Africa. It was
the British naval deterrent which restricted even France to merely
derisory acquisitions between 1830 and 1880.
By the later 1870s, nevertheless, Africa had become far more
economically interesting to the industrial Powers than it had been
a generation earlier. This increased interest was due less to any
great increase in the actual volume of trade than to the supposed
economic potential, in a period of depression, of an Africa which
now consisted not merely of' coasts' but of a possibly controllable
and exploitable interior. Without this development of economically
motivated interest, the full-blooded scramble of the 1880s and
1890s is indeed hardly conceivable. But no economic model can
6
[ P . R . O . ] , Foreign Office ( F . O . ) 6 5 / 1 1 1 3 , H o w a r d ( L i s b o n ) t o F . O . , S / T n o . 4 3 ,
5 J une 1856: cited R. Anstey, Britain and the Congo in the nineteenth century (Oxford, 1962),
47-
IO5

explain, even in the broadest and most general terms, the pattern
of partition that actually took place. In 1875, the most likely
outcome of increased European economic interest in Africa
seemed to be the consolidation by the British of their informal
coastal empire and its selective extension into the interior; and
between 1875 and 1879 this process was actually taking place, only
to be overtaken by the onset of partition. But the mere fact of
partition implies the failure of the British deterrent against
encroachment upon their informal empire; no economic model
seems likely to account for this failure. Partition began as an
exclusively Anglo-French contest, which was indeed the only
pattern that had seemed even remotely possible in the late 1870s.
Apart from the apparently trivial intervention of King Leopold,
no other power was involved until in 1884 partition suddenly
developed into an entirely unforeseen multi-power scramble - in
which, by a crowning paradox, the non-power Leopold II carried
off one of the major prizes. Even if each individual act of
annexation had been economically motivated, explanations of a
different kind would still be needed to account for this remarkable
pattern of partition.
Moreover, the demonstration that economic incentives existed,
and that they stimulated the kind of interest that led to annexation,
does not necessarily imply that economic motives were crucial,
or even that they existed at all, in any particular act of territorial
acquisition. And for some of the partitioning powers, the economic
model does not work even as a generator of interest. In the 1870s
Portugal, still virtually a pre-industrial power, was presiding
listlessly over the last shrinking relics of an African empire ruined
by the final suppression of the transatlantic slave trade. Yet in the
18 80s she reasserted her control over very extensive territories
which for long remained a crippling burden upon her under-
developed economy — although they may of course have enriched
particular interests within it. Italy, a very sketchily industrialised
country of whose many economic problems industrial over-
production was surely the least, nevertheless had genuine economic
interests and prospects in Tunisia. But she failed to make good
her position there; instead, she acquired in the Horn of Africa an
empire which offered prospects not of markets or profits but of
a military commitment so heavy as to be a political danger as well
as an economic burden.
106

The revival of effective Portuguese imperialism in the 18 80s was

mainly a response to the contemptuous arrogance with which
other powers, especially Britain, brushed aside Portugal's' historic
claims' in Africa. Politically conscious Portuguese reacted to this
affront by treating the maintenance of the Lusitanian imperial
heritage as crucial to national self-esteem and indeed to the
Portuguese national identity. Italy had no 'historic claims' in
Africa — or at least, none more recent than those of imperial
Rome. But when in the 1880s African imperialism became
fashionable among the great powers, the Italian foreign minister,
Francesco Crispi, in particular seems to have felt that Italy would
look more like a great power if she followed the fashion. As
Crispi's political rival, di Rudini, remarked, Italy's African empire
was acquired 'in a spirit of imitation... for pure snobisme'' P
'Prestige' is an uncomfortably nebulous concept; but it is
impossible to discuss Portuguese and Italian imperial motivations
without using it as a convenient shorthand.
'Prestige' also seems the most appropriate category in which
to discuss Khedive Isma'lPs African empire, which was no
fortuitous agglomeration of territories, but the first of the
consciously-planned African empires of the later nineteenth
century. Isma'Tl intended it to embrace the entire north-east
quadrant of the continent, including Ethiopia, eastwards to the
Indian Ocean and southwards to the equator — he optimistically
called his southernmost Sudanese territories the ' Equatorial
province'. This grandiose plan suffered a mortal blow when in
1875 and 1876 the khedive's attempts to subjugate Ethiopia were
ignominiously routed. Soon after Isma'lPs deposition in 1879, m s
empire in most of the Sudan was destroyed by the Mahdist
insurrection and by local risings of the southern Sudanese
peoples; and by 1888 the surviving remnants had been dismantled
as a measure of retrenchment by the British in Egypt. Isma'il, like
Leopold II and Rhodes, expatiated on his 'civilising mission': he
claimed for Egypt' the position of the head of civilisation in North
Africa'.8 Although the khedive sometimes attempted to promote
economic development, even so blindly optimistic afinancieras
Isma'Il can hardly have hoped to recover much of his enormous
7
Cited W. L. Langer, The diplomacy of imperialism (2nd edn, New York, 1951), 281.
8
F.O. 84/1571, Granvillc to Stanton (Alexandria), no. 6, 2 Apr. 1873: cited
M. F. Shukry, Equatoria under Egyptian rule (Cairo, 1955), 25.
107

outlay. He created his empire, not in serious hope of gain, but

to the greater glory of himself, his dynasty and his adopted
country, Egypt. Like the Cairo opera house, African empire was
a symbol of Egypt's claim, as a modernised, civilised and
'civilising' power, to equality of status with the great historic
nations of Europe.
The imperialism of prestige would have been comparatively
unimportant had it been confined to decayed powers like Portugal
or over-ambitious parvenus like Egypt. But it was not. During the
later nineteenth century the great powers (and a fortiori a
doubtfully great power like Italy) lived in a ruthlessly competitive
world of constantly fluctuating antagonisms and alignments. The
system was so complex and so delicately balanced that diplomatic
setbacks even on quite minor issues could sometimes jeopardise
a power's perceived effectiveness and general international
standing. Hence the constant and anxious concern for 'prestige'
— a concern which became almost obsessive when a previously
preponderant power like France was struggling to avoid rele-
gation to second-class status, or when a powerful and ambitious
newcomer like Wilhelmine Germany was striving to assert itself
far outside Prussia's traditional political arena. Until the mid-
18 80s the British were usually confident enough of their metro-
politan security and oceanic hegemony to accept minor setbacks
without great anxiety, and to treat with ironic condescension the
excessive 'touchiness' of lesser breeds. But this supercilious
detachment did not survive the steady erosion after c. 1884, by
rivals both old and new, of Britain's former easy preponderance.
By 1898 Britain too had become intensely concerned for her
'prestige' - for her credibility as a great power; and when that
credibility appeared to be at stake, even in African conflicts over
territory of little or no intrinsic value, she was prepared to risk
or even to threaten European war in order to reassert it.
In the early 1840s the French challenges to Britain on African
coasts had been prompted far less by real or supposed economic
necessity than by the determination of French statesmen and
French admirals to restore France to her rightful status as lagrande
nation, especially by reducing British preponderance in the Indian
Ocean. Forty years later, the necessity for France to vindicate her
claim to great-power status had become even more acute; and the
crucial factor in the French decision to occupy Tunisia in 1881
108

was not pressure from financial interests, influential though these

were, but the intolerable prospect of being elbowed out of an
established unofficial hegemony by Italian collusion with the
Bey. It was the diplomatists, not the businessmen, who nerved
Ferry and his strong ally, Leon Gambetta, to act in spite of their
fears of unpopularity at home and complications abroad. The
diplomatists insisted that failure to meet the challenge from a
parvenu power like Italy would be seen by a watchful and critical
Europe as evidence that France was ' sinking to the level of Spain':
a disaster all the more deplorable in that 'one single act of
firmness, of will and of determination' could restore France to
her ' rightful place in the eyes of other nations '.9 In the later 188os
and the 1890s, French bitterness at the continued British occupation
of Egypt was prompted less by considerations of the Mediterranean
power-balance than by the British usurpation of France's traditional
role, which she had reasserted with considerable success in the
later 1870s, as the preponderant European influence in Egypt
itself. It was the determination to wipe out this affront, not the
pursuit of strategic advantage in the Mediterranean, that in July
1898 brought Marchand to Fashoda in order to challenge the
British occupation of Egypt by an implicit threat to the Nile
waters.
German expansion, and attempted expansion, in Africa
was also strongly driven by motives of prestige. Bismarck's bid
for colonies was complex in origin and motivation. Economic
motives were certainly present; but by May 1884 his strongest
single motive seems to have been to force London to abandon
a doctrine of' paramountcy' which purported to exclude German
political influence from African territory which Britain neither
occupied nor claimed by legal right. This 'Monroe doctrine for
Africa' (as Bismarck called it, unconsciously echoing Carnarvon's
words of December 1876) was less an economic injury to
Germany than, in Bismarck's own words, 'an affront to our
national self-esteem'.10 The later German pressure for repartition
in Africa, maintained fairly steadily from about 1895 to 1914, was
9
J. Ganiagc, Let origints du protectorat fraitfais en Tunisie (Paris, 1959) 632, 634—5,
citing: St-Vallier to Barthclemy St-Hilaire, 26 Jan. 1881; Roustan to Noailles, 23 Feb.;
St-Vallier to Noailles, 21 Mar. 1881.
10
Die Grosse Politik dtr Europdischtn Kabinette, 1871-1914, iv, no. 742, Bismarck,
minute on Hatzfeldt to Bismarck, 24 May 1884; no. 743, Bismarck to Munster (London),
1 June 1884.
109

driven not by economic necessity but by the conviction that

German prestige was being impaired by Bismarck's failure in the
18 80s to acquire an overseas empire of dimensions more in
keeping with Germany's new status as a world power.
Although during the 1890s the Egyptian question became for
France a matter of prestige rather than of strategy, Egypt was
nevertheless a crucial region strategically. In 1882 the real or
supposed need to defend the Suez Canal against 'UrabI and the
Egyptian nationalists was one of several considerations which
gradually overcame Gladstone's shrinking from armed inter-
vention. Another quasi-strategic motive was the fear that by
mid-1882 French policy under Freycinet was moving towards
active diplomatic support for 'UrabI - a development which
might well have inhibited British intervention and have established
France as the patron of Egypt's new rulers. For some six years
after 1882, however, British policy-makers did not see Egypt as
an imperial acquisition. They still looked forward to military
withdrawal as soon as they had, by financial and other reforms,
restored Egyptian solvency and political stability, thereby re-
moving pretexts for intervention by other powers. Lest main-
tenance of the debris of Isma'Il's African empire should hinder
reform in Egypt, the British hastened to liquidate these remnants,
ignoring the possible long-term threat to the Nile waters implicit
in the vacuum of power they were creating.
In the course of 1888 the British decided to remain in Egypt
indefinitely - a development fundamentally more important than
their spectacular armed intervention of 1882. Maintenance of the
British position in Egypt was henceforth seen as an imperative
strategic necessity. Turkey seemed to have become, in Lord
Salisbury's words, merely 'the janitor of Russia' at the Straits;
defence of the Mediterranean balance by supporting Turkey
against Russia at Constantinople was therefore, at best, an
obsolescent strategy. Russian influence in Turkey was expected to
increase steadily, and with it Russian naval strength in the eastern
Mediterranean; sooner or later a strong naval base in Egypt would
become indispensable to Britain. Once the British had decided to
remain, they could no longer ignore the potential threat to Egypt's
water-supply entailed by the vacuum of power on the upper Nile;
and by 1889 the strategic defence of the upper-Nile waters had
become 'a separate and dominating factor' in Lord Salisbury's
110

diplomacy." After 1889 this strategy usually dominated British

policy not only towards the upper Nile itself but also towards all
those very extensive regions of Africa which could be regarded
as 'approaches' to the Nile basin. After 1893, when the French
were attempting, in pursuit of a goal dictated by prestige, to
establish a strategic prise de possession on the upper Nile, strategy
similarly dominated French policy towards this region.
Strategy therefore appears, in the Nile valley at least, as yet
another category of motive driving on the partition. This is not
to say, however, that' from start to finish the partition of tropical
Africa was driven by the persistent crisis in Egypt'. 12 By 1882 the
partition of tropical Africa was already in full swing in the Niger
Sudan, on the Guinea coast, on the Congo, and in Madagascar.
Moreover, for nearly two years after September 1882 the' Egyptian
crisis', as an Anglo-French confrontation, was hardly a crisis at
all. Paris still believed that London would ultimately conclude that
French friendship was worth the continued sharing of financial
control in Egypt. To assist the British to see reason, until 1884
the French refrained from local obstruction in Egypt; the French
consul in Cairo was even instructed to co-operate actively with
his British colleague.'3 Elsewhere in Africa however, and especially
in Guinea and Madagascar, France had already become more than
usually combative towards England at least eighteen months before
the British occupation of Egypt. Immediately after September
1882 there was little if any perceptible increase in this com-
bativeness; and until thefinalcollapse ofAnglo-French negotiations
on Egyptian finance in summer 1884, there is no convincing
evidence that French forward policies in Africa were prompted
by the desire to 'retaliate' for Egypt.
Only in August 1884 did Jules Ferry begin cautiously to
respond to Bismarck's proposal, first put forward in May, for a
Franco-German alignment against Britain on African questions.
This alignment, while it lasted, certainly produced some spectacular
results in Africa. But it did not long survive the fall and disgrace
of Ferry in March 1885; and its demise was followed, in the later
" Cecil, (Lady) G., Life of Robert, Marquis of Salisbury, iv (London, 1952), 48, citing
Salisbury to A. Austin, 3 July 1887; to Lord Lyons, 20 July 1887. Ibid., 139-40.
12
Robinson and Gallagher, Africa and the Victorians, 465.
13
For French policy towards Britain in Egypt from Sept. 1882 to Aug. 1884, see:
Documents Diplomatique! Francois, Ie serie, iv (Paris, 1932), nos. 525-97, passim; v
(Paris, 1933), nos. 26-402 passim.
I I I

18 80s, by a marked improvement in Anglo-German relations. By
1890 the Quai d'Orsay saw little prospect of progress on the
Egyptian question until the return to office of the Liberals. But
when the Liberals came back in 1892, Lord Rosebery at the
Foreign Office proved no less intransigent than his predecessor
Salisbury. French overtures were rebuffed, with no excess of
courtesy; and thereupon in February 1893 Theophile Delcasse,
then at the Colonies, launched on his own responsibility the first
French attempt (the abortive Monteil mission) to enforce
evacuation by a threat to the upper Nile waters. As a propellant
of partition in the Nile basin and its approaches, the ' Egyptian
crisis' evidently had a very long fuse. The British had seen no
threat to the Nile waters from anyone until 1889, and none from
the French until mid-1892, when the possibility of an advance to
the upper Nile was first discussed in the French Chamber. To the
partition of West Africa, at any rate down to 1894, 'the persistent
crisis in Egypt' had no relevance at all. Between 1879 and 1883,
when there was as yet no serious quarrel over Egypt, the French
had advanced very rapidly in West Africa. Precisely when the
Egyptian dispute became embittered, in 1884—5, there was a pause
in this advance; and when the advance was resumed in 1889—90
it was directed exclusively to the creation of a large territorial
empire centred on Lake Chad — a project which tended, until
1894, to distract the attention of French colonialists from the Nile
valley.
The 'Egyptian crisis' - not that of 1882 but the open Anglo-
French quarrel of 1884 — did indeed isolate Britain diplomatically;
and the short-lived Franco-German alignment in Africa enabled
Bismarck to help himself, with a minimum of risk, to African
territories which had previously formed part of Britain's informal
empire. These African acquisitions have been seen as purely a
move in Bismarck's European policy, a means of picking a quarrel
with Britain so as to make his desire for a Franco-German
rapprochement more credible to Paris. But it was never very clear
why, on purely European grounds, Bismarck should have preferred
a rapprochement with France to the renewed isolation into which
France had been driven by her Egyptian quarrel with Britain; and
it is now well established that by 1884 Bismarck had concluded,
albeit rather reluctantly, that Germany needed to acquire, if not
formal colonies, at least 'spheres' of her own in Africa. He
112

therefore needed, as he himself told the French, a Franco-German

alignment in African disputes in order to create an 'equilibrium
on the seas'14 against Britain's massive naval preponderance over
Germany alone.
Advances or acquisitions in Africa undertaken primarily to
secure a diplomatic advantage in Europe are not however quite
unknown. The most striking is the Anglo-Egyptian advance into
the Sudan in March 1896. This move was not originally intended
to destroy the Mahdist state, still less to solve the strategic
problem on the upper Nile. It began as little more than a token
military demonstration undertaken, at Italy's request, in order to
distract the Mahdists from attacking Italian outposts in Eritrea
and the eastern Sudan after Italy's catastrophic defeat at Adowa
by Menelik of Ethiopia. The British solicitude for Italian interests
was prompted by the more important consideration of Britain's
relations with Germany, which had for some two years been
severely strained, both in Africa and elsewhere. But these
conflicts had brought Germany no significant advantage; on the
contrary, they had recently led to a marked and disquieting
improvement in Anglo-French relations, while the Italian catas-
trophe had shaken the Triple Alliance. The Germans were
therefore now willing to improve their relations with Britain; but
only if London would reciprocate by supporting the Triple
Alliance and breaking off the flirtation with Paris. A gesture of
military assistance to Italy in Africa would be a public demon-
stration of support for the Triplice; and Berlin correctly calculated
that an advance into the Sudan would destroy the tentative British
rapprochement with France. Berlin therefore saw in London's
response to Italy a crucial test of the entire orientation of British
foreign policy, which was in effect being constrained to 'choose'
between France and Germany. Salisbury decided to satisfy the
Germans. He feared that a negative response would leave Britain
dangerously isolated against an unfriendly Germany; the conver-
sion of the Anglo-French rapprochement into a full entente
seemed impossible so long as the two countries were still divided
by the Egyptian question. On 12 March 1896 Salisbury therefore
ordered the token Anglo-Egyptian advance that was during the
next two years gradually to develop into the total reconquest of
the Sudan (below, pp. 149-50).
14
Ibid., v, no. 407, Courccl (Berlin) to Ferry, 23 Sept. 1884.

Examples of'negative' diplomatic motives - the renunciation
of African territory in order to gain a diplomatic advantage in
Europe - are of course not uncommon. Salisbury's comparative
generosity towards Italy in Africa between 1889 and 1891 reflects
his recognition that his relations with Germany would suffer if
Anglo-Italian quarrels in Africa became so acute as to jeopardise
Britain's support for Italy, and therefore for the Triple Alliance,
in the Mediterranean. When in 1893 Lord Rosebery, foreign
secretary in the Liberal government, failed to convince Berlin that
British support for Italy would be forthcoming in a crisis,
Anglo-German relations did promptly deteriorate: notably in
Africa, where Germany became stiffer towards Britain and more
generous towards France — especially where such generosity
would tend to embroil Anglo-French relations. Although ' diplo-
matic ' motives of this kind are of course motives for abstention
rather than acquisition, they are nevertheless important for the
detailed pattern of partition.
So too, and not merely for the detailed pattern, are motives
generated by the immediate pressures of domestic politics. Such
motives in fact usually boil down either to economics or prestige;
but the type of domestic motivation known as ' manipulated social
imperialism' perhaps deserves a category to itself. It has been
succinctly defined by its leading practitioner, Bernhard von
Biilow: ' T o direct the gaze from petty party disputes and
subordinate internal affairs...on to world-shaking and decisive
problems'. 15 This motive, as well as the desire visibly to assert
Germany's greatness as a world power, certainly underlies the
German pressure for repartition between 1895 and 1914-a
period when the Social Democrats were capturing a rapidly
increasing proportion of the Reich electorate. Crispi's imperialism,
too, had a distinct Biilovian streak: he hoped that the grandeurs
and burdens of empire would distract Italians from their all too
intense preoccupation with their multiple internal divisions and
conflicts. On the whole, however,' manipulated social imperialism'
does not seem to have been a very effective propellant of the
scramble. It seems to have played only a minor part in initiating
partition: it is in particular doubtful how far theorising of this
kind influenced Bismarck.' Manipulated social imperialism' seems
15
Biilow, cited P. M. Kennedy, 'German colonial expansion: has the " Manipulated
Social Imperialism" been ante-dated ?', Past and Present, 1972, 54, 134 41.
114

indeed to be a phenomenon of the 1890s and after rather than of

the 1880s. But even in this later period its practical effects were
meagre: the German pressure for repartition was, after all, almost
entirely unsuccessful.
Two further factors — hardly ' motives' - are sometimes seen as
influential, or even crucial, in promoting territorial acquisition.
'Missionary influence' is sometimes regarded as having instigated
the partition; but although the flag did very often follow the
Gospel into Africa, only exceptionally was this flag the national
flag of its missionary precursors. In Madagascar and Cameroun,
British Congregationalists and Baptists ultimately found them-
selves not under the Union Jack, but under the French Tricolour
and the flag of Bismarck's Reich. British Baptists on the Congo
found themselves in King Leopold's empire. French White
Fathers and Holy Ghost Fathers ended up respectively in British
Uganda and in German East Africa. Perhaps only in Nyasaland
is there a clear and direct causal link between missionary influence
and colonial annexation. Salisbury was anxious to avoid assuming
political responsibility for this awkwardly stituated Scottish
mission field, which was then virtually inaccessible except through
Portuguese Mozambique; but his hand was forced by the fear of
losing Presbyterian votes in Scotland if he permitted a Catholic
Portuguese takeover. In general, missionary influence was probably
a less important factor in partition than the career ambitions of
French military officers who, especially but not exclusively in the
Niger Sudan, often initiated advances and offensives in order to
accelerate by/aits de guerre the snail's pace of routine promotion.
More interesting is the situation summed up in the phrase ' the
local crisis'. Here annexation is seen as virtually forced upon the
imperial power by the breakdown of a previous informal control
operated through a collaborating local elite; and this breakdown
is often the result of a successful challenge to this elite by a popular
or traditionalist opposition. This model works excellently in
Egypt (not surprisingly, as it was originally generated by an
analysis of the Egyptian situation); and well enough in Tunisia
and Zanzibar. In South Africa, although applicable to some
aspects of Anglo-Afrikaner relations between c. 1870 and 1899, it
is clearly inadequate. An attempt has been made to apply it to
French expansion in the Niger Sudan by interpreting the French
conquest as an enforced response to Muslim militancy provoked
"5

by resentment of informal French influence. In fact, the Muslim

states usually went to considerable lengths to avoid war; and the
French campaigns were not' involuntary imbroglios' but deliberate
aggression, 'a determined European bid for territory'.16 As a
covering theory for the partition as a whole, the local crisis is not
even superficially plausible. What local crisis, of the type specified
in the theory, brought Leopold or Brazza to the Congo? — or Italy
to Eritrea or Somalia? — or Bismarck to his curiously distributed
set of colonies, in most of which the previous informal influence
had been not German but British ? — or France to Madagascar,
where the previous collaborative relationship had again been with
the British, and was deliberately destroyed by the French? Nor
does the local crisis, even where it works, represent a separate
category of motivation. In handling any local crisis, and in finally
resolving it by annexation, European policy-makers were
defending interests - economic, strategic, or other - falling
within the five categories already established.
These five categories - expected economic gain, prestige,
strategy, diplomatic advantage and ' manipulated social imperial-
ism ' — do however seem to be exhaustive, or very nearly so.
Missionary influence and military careerism can perhaps be treated
as merely local and exceptional phenomena; or else subsumed as
varieties of 'prestige'. Indeed, who can draw the line between
military careerism and ' the honour of the flag', especially in the
minds of the military officers themselves? An analysis with five
variables is necessarily complex; but in this labyrinth of motivations
there is one sure clue. None of these motives, and no combination
of them, could have generated a general multi-power partition of
Africa if the British had continued successfully to deter other
powers from encroaching upon their informal African empire.
The collapse of British hegemony is the great negative factor
which alone enabled the ' positive' motivations of other powers
to achieve practical results. If this factor is neglected, the partition
seems on detailed analysis to arise from mere coincidence: from
the fortuitous 'convergence of many forces', embarrassing in
their multiplicity and diversity, which just happened to' converge',
suddenly and inexplicably, in the early 1880s. Seen in this way,
16
Cf. R. Robinson and J. Gallagher, 'The partition of Africa', New Cambridge modern
history, xi (Cambridge, 1962), at 609, 619—20; and A. S. Kanya-Forstner, 'Myths and
realities of African resistance', Canadian Historical Association, Historical papers, 1969,
185-98.

THE BREAKDOWN OF INFORMAL EMPIRE
the scramble is simply incomprehensible as a historical process.

Rather, 'c'est une histoire idiote... Si tu la regardes de loin, elle
se tient a peu pres; mais si tu te rapproches, tout fuit le camp'. 17
The concept of partition as a consequence of the collapse of an
almost exclusive British hegemony, at a time when both Britain
and other powers were beginning to see significant economic
advantage in the extension and consolidation of their formal or
informal control over African territory, permits the ordering of
an otherwise unmanageable variety of phenomena. However,
emphasis on the 'background' of growing economic interest
does not imply that all annexationist decisions were somehow,' in
the last analysis', economically motivated. On the contrary. The
collapse of British hegemony created a very complex and fluc-
tuating pattern of conflicts and alignments in both Africa and
Europe. Amid this complexity, policy-makers often simply could
not afford to handle African questions by giving economic
interests absolute priority over the demands of diplomacy and
strategy, or even of prestige.
British hegemony had never seemed more powerful and complete

than in the mid-1870s, on the very eve of partition. It extended,
virtually unbroken apart from Algeria and Senegal, around the
entire African coastline. Other than Britain and France, the only
African coastal 'powers' were Egypt, Zanzibar and Portugal.
Egypt and Zanzibar were by now British client-states; and in
1875-6 their subordinate position was emphasised when a clash
on the Somali coast between an Egyptian naval expedition
(commanded by a Scotsman) and the Zanzibari forces (soon to
be commanded by an Englishman) was settled by the fiat of the
Foreign Office. Portugal was not quite a client-state; but she held
her African territories virtually on British sufferance, as her
expulsion from the maritime Congo in 1857 had demonstrated - a
demonstration repeated two years later on the Guinea coast, when
the forcible' removal' of Portuguese residents from Bulama island
was followed by its annexation as British territory. By 1876,
17
I... H. Gann and P. Duignan, 'Reflections on imperialism and the scramble for
Africa', in Gann and Duignan (ed.). Colonialism in Africa, I (Cambridge, 1969), 127-8.
J.-P. Sartre, lês mains sales, premier tableau, scene iv (Hugo).
"7

Za
f\
L TanganytkaKX zz
-
Sultanate of Zanzibar
Portuguese
( • h * j 5 p e 9 l O " s o * •paramouni'Bntishinfluence.
British other than Egyptian and Zanzibar! dominions
^ X N F r e n c h * ^ " R e 9 ' o n s o f small-scale Anglo-French

^COLONY'
^SSl competition on the West African coast
"~ — "~ Tunis, Tripoli and the Egyptian dominions

were under Ottoman suzerainty
3 The climax of British unofficial empire: Africa on the eve of partition,

c. 1878
118

moreover, certain British policy-makers were consciously working
to extend and consolidate informal hegemony, notably by the
application of the doctrine of'paramountcy'. In the many regions
where Britain exercised exclusive influence without having
acquired formal possession, she was increasingly ready to claim,
as 'paramount power', a quasi-legal right of objection to any
encroachment by other powers.
In West Africa, 'paramountcy' seems to have originated simply
as an administrative expedient, a solution to the difficult problem
of exercising control in the interior in order to protect the internal
trade of small coastal possessions. Paramountcy permitted total
flexibility in the extension and application of control while
avoiding the unwelcome legal and financial consequences of
formal sovereignty - consequences so unacceptable that aban-
donment of the coastal possession itself had sometimes been
seriously considered as the preferable alternative. The concept was
also used to cover situations like that in the Niger delta, where
Britain enjoyed ' paramount influence' backed by a preponderance
of armed force, but had no formal possessions at all. Paramountcy
of this kind could hardly be deployed unless no other European
power exercised significant influence in the region concerned; but
from 'paramountcy' as an expedient applicable only where other
powers were absent, it was an easy if illogical step to the doctrine
that where paramountcy existed no other power was entitled to
intrude its presence. By 1875 this was the Foreign Office position
on the Niger delta. In the same year, the Colonial Office was
deliberately refraining from piecemeal annexations in the Pacific
lest these, by alerting other powers, should ' defeat the object and
prevent us from quietly acquiring paramount influence among the
Islands';18 and in 1875—6 Lord Carnarvon was hoping to obtain
French recognition of British paramountcy along the whole West
African coastline, excepting only Senegal, from Morocco to
Gabon.
By 1876-7 some British policy-makers were extending the
concept of paramountcy to cover not merely African 'coasts' and
their immediate hinterlands, but enormous areas of the accessible
or potentially accessible interior. In southern Africa British
paramount influence, alike over British colonies, Afrikaner
18
R. Herbert (P.U.S., Colonial Office), 5 May 1875, cited W. D. Mclntyrc, The
imperialfrontier in the tropics (London, 1967), 356.
119

republics and African polities, had been regarded as axiomatic at

least since i860. This paramountcy had never had any fixed
northern limit; but in December 1876 Carnarvon wanted to
extend it until it met, at the Great Lakes, another sphere of
paramount British influence — that over ' Egypt and the country
that belongs to Egypt'.' 9 In April 1877 the Transvaal was
annexed. In February 1877 British influence had been exerted to
secure the appointment of Gordon as governor-general of the
Egyptian Sudan; and in September Khedive Isma'Tl bound
himself not to dispose of Egypt's African coastal possessions
without British permission. These possessions extended to Cape
Guardafui, where they adjoined the northernmost coastal sphere
of Britain's other client, the sultan of Zanzibar. Further south on
the Zanzibari mainland, the consul, Sir John Kirk, was in 1877
encouraging the sultan to extend and formalise his influence; at
the same time Kirk was pressing him to grant to the Scottish
shipowner, Sir William Mackinnon, mainland concessions which
would have transferred to Mackinnon the reality of power and
profit. In May 1877, Sir Robert Morier, the British ambassador
at Lisbon, proposed the enrolment of Portugal as a third British
client-state with functions similar to those of Egypt and Zanzibar.
In this way, he believed, it would be easy for Britain to acquire
paramount influence throughout the whole of Africa. The Foreign
Office did not object, but gave Morier little active encouragement.
Perhaps an attempt openly to subordinate the proud and sensitive
Portuguese hardly seemed worth the trouble: by mid-1877 Britain
already seemed well on the way to imposing the ' Munro doctrine
[for] much of Africa' which Carnarvon had demanded in
December 1876.
The aggressive and enterprising behaviour of Britain in Africa
during the mid-1870s did not pass unnoticed at Paris. For the
French, the one redeeming feature of British informal empire was
its purely de facto existence, devoid of legal warrant and therefore
instantly vulnerable should the power-balance ever tilt in favour
of France. Any British attempt to formalise informal empire -
whether by outright annexation or by expedients such as para-
mountcy or a ' Monroe doctrine' - was therefore a serious threat
to France in that it might confine her permanently to her Algerian
ig
Carnarvon t o Frere, i z D e c . 1876, cited E t h e r i n g t o n , J. Imp. C'wealtb Hist., 1981,
I2O

and Senegalese enclaves. As early as February 1876 this danger
had been emphasised by Benoist d'Azy, the directeur des Colonies.
Benoist, seeing in a recent British blockade of Dahomey an
extension to West Africa of the British forward policies in South
and East Africa, pressed urgently for more active French
competition with the British. But Benoist's pleas fell on deaf ears
at the Quai d'Orsay, which saw Africa in a very different per-
spective from that of the Colonies. In February 1876 barely nine
months had passed since France had, to all appearances, been
saved from a German preventive war only by the diplomatic
intervention of Russia and Britain. To the diplomatists, possible
colonial opportunities in Africa were not only worthless compared
with the security of France in Europe: they were a positive
menace, for by generating Anglo-French disputes they might
cause France to forfeit British goodwill. Rather than run this risk,
the Quai d'Orsay preferred to sacrifice the colonial opportunities
en bloc. Early in 1876 the French foreign ministry was prepared,
in return for the trivial compensation of the Gambia, in effect
to recognise British paramountcy over the whole west coast apart
from Senegal. The Marine and the Colonies protested, and so did
the firm of Regis Freres at Marseilles; but it was British ministerial
incompetence, rather than the strength of the French opposition,
which frustrated this remarkable bargain. The Quai d'Orsay,
having thus failed to impose its radical solution, was forced to
handle with the utmost caution local West Coast disputes, which
were now becoming sharper and more frequent as governors of
Senegal and of Sierra Leone attempted to increase their depression-
starved revenues by planting customs posts in the coastal no-
man's-land between the two colonies.
This extreme caution was however no longer necessary after
the spectacular improvement in 1878 of the international situation
of France. At the Congress of Berlin, France ceased to be an
isolated power and almost a diplomatic outcast, and suddenly
became the key to the power-balance in the Near East. After the
Russian victory over Turkey in 1877-8, it was the refusal of
France to align herself with Russia which enabled Austria and
Britain to check Russia in the Balkans and at the straits without
calling upon overt and active assistance from Germany. Such a
call would have forced Bismarck to choose, to his intense
embarrassment, between the denial of support to Austria and a
121

direct confrontation with Russia. At Berlin, therefore, France was

courted by Bismarck hardly less ardently than by Salisbury: both
suggested that she should take her reward in Tunisia, which
neither Bismarck nor Salisbury wished to see fall into Italian
hands. But the real reward for France was that, so long as Russia
maintained her ambitions in the Balkans and the Levant, the
attitude of France remained of crucial importance to no less than
three great powers: not only to Britain and Germany, but also
to Russia herself. Salisbury, with his acute diplomatic perception,
responded instantly to the changed power-balance by striving to
restrain all British activity in Africa which might alarm or irritate
the French. He sabotaged, by devious means, the proposed
Mackinnon concession on the Zanzibari mainland; he frustrated
Morier's Portuguese machinations by slow-motion correspond-
ence and ambiguous instructions. With Carnarvon no longer in
office, Salisbury was able effectively to preach restraint to his
colonial colleague: but the Colonial Office was not always
able to restrain its own agents, whether in Zululand, where
Bartle Frere's forward policy precipitated military disaster at
Isandhlwana, or on the Guinea Coast, where 'insupportable
proconsuls' infuriated Salisbury by continuing to quarrel with
the French. It was now Britain's turn to be cautious and con-
ciliatory in these local disputes; and Salisbury sometimes carried
his 'graceful concessions' to lengths which disconcerted the
Colonial Office.
Had France remained as isolated and vulnerable in Europe as
she had been between 1871 and 1877, t n e Marine and the Colonies
would hardly have been permitted by the French Foreign Ministry
to launch any significant challenge to Britain in Africa. Even after
1878, the foreign ministry remained for a year or two less than
combative in African disputes with Britain. But it was now no
longer necessary deliberately to sacrifice all colonial opportunity
in Africa for the sake of European security, nor to be so strict
in restraining the colonial activists; and in November 1880 the
Quai d'Orsay issued a directive to this effect. The year 1879, which
in French domestic politics marked the definitive take-over of
state power by bourgeois, anti-clerical republicans after a seven-
year struggle with aristocratic, ultramontane royalists, had
already seen the launching of a plan for large-scale territorial
122

empire in the Niger Sudan by the minister of Public Works,
Freycinet, who originally thought in terms of peaceful penetration
by means of railways; and by Admiral Jean Jaureguiberry at the
Marine, a former Governor of Senegal who did not shrink from
outright military conquest.
The Niger Sudan was chosen as a 'substitute India' largely
because of its supposed value as a market. Earlier attempts at
purely commercial penetration had not been successful; and
formal empire was obviously by far the best safeguard against the
suspected intention of the British to forestall France by annexa-
tions in the upper Niger basin. Moreover, as the upper Niger was
accessible from Senegal, the French advance could take place
without violating British coastal hegemony; it therefore avoided
a direct confrontation with London which would in 1879 still have
been unwelcome to the Quai d'Orsay. The Senegal—Niger advance
was not indeed an overtly anti-British move. It injured no material
British interest, for London had by now abandoned any serious
hope of profiting by the trans-Saharan trade. But the advance was
nevertheless a resounding announcement that France would not
stand idly by while the British extended to West Africa the
expansive hegemony which they had already established, and
appeared to be consolidating, in southern and eastern Africa.
On the West African coast, Anglo-French competition was by
now obvious and explicit: in 1879 Briere de l'lsle, the governor
of Senegal, introduced the discriminatory Senegal tariffs and
harbour dues into the territories in dispute with Sierra Leone. But
Briere believed that the scramble had already begun not only on
the coast but in the interior. He was convinced that the British were
planning to forestall France in the upper Niger basin. No such
project existed; but, in the context of recent British policy
elsewhere in Africa, Briere's fears were not unreasonable. Less
reasonable was his readiness to credit, as proof of London's
machinations, every unconfirmed and implausible rumour of
covert British activity in the interior. Against this background the
Gouldsbury mission to Futa Jalon in January 1881 was seen, not
only in Senegal but also in France, as the first overt and official
move in the British plan. In February 1881 Jaureguiberry told the
senate: 'We have rivals, implacable rivals, who constantly oppose
the influence we exercise in Senegal. They strive to frustrate us
123

20
in every possible way'. Like many French naval officers,
Jaureguiberry was intensely combative towards Britain. In April
1882, after his return to office in January, he established a formal
French protectorate at Porto Novo, in an area of particularly acute
coastal competition to the west of Lagos. By J anuary 18 8 3, he was
planning not only to extend this protectorate westwards to the
Gold Coast frontier, but to establish further French protectorates
in the Niger delta, on the Benue river and in Cameroun. In
these projects he had the strong support of Maurice Rouvier, by
then Minister of Commerce, and the full concurrence of the Quai
d'Orsay, which had since early 1881 been increasingly combative
in the ever-recurring minor coastal disputes.
Although the British attempt to formalise their informal
empire, and their adoption of forward policies in East and South
Africa had, in principle at least, been halted by Salisbury in 1878-9,
Paris continued to fear the supposed British threat; and France
had by the end of 1882 responded by a counter-challenge far more
effective than that of the early 1840s. This counter-challenge was
not confined to West Africa. In 1881 what appears to have been
a routine exercise of British informal influence in Madagascar
provoked a French response which was at least as combative as
that in the Niger Sudan and in Guinea. These French moves were
in no way ripostes to the British occupation of Egypt. Both in
West Africa and in Madagascar, they had been launched while the
two countries were still co-operating quite closely on the Nile;
and in Madagascar at least, the 'point of no return', whereafter
France could not retreat without visible loss of prestige, had been
reached as early as May 1882.
Since the 1860s France had not seriously challenged the British
monopoly of informal influence over the dominant Hova dynasty
in Madagascar, in spite of pressure from the sugar-planters of
Reunion, who saw in the island a potentially valuable source of
labour and foodstuffs. Meanwhile, through the conversion of the
dynasty by British missionaries, Congregationalism had become
virtually the Hova state religion. A British naval visit to
Madagascar in July 1881, intended to encourage the Hova to
extend and consolidate their rule, was seen in Paris as an indirect
20
Jaureguiberry, 17 Feb. 1881: cited C. W. Newbury and A. S. Kanya-Forstner,
'French policy and the origins of the scramble for West Africa', J. Afr. His/., 1969,
10, z, 25 3—76.
124

attack upon certain non-Hova (Sakalava) groups with whom
France had made treaties in 1859—60 and over whom she still
retained some influence, and therefore as a deliberate attempt
finally to destroy the 'historic links' between France and Mada-
gascar. Jules Ferry promptly recalled his consul Theodore Meyer,
whose conciliatory disposition had qualified him for the Mada-
gascar post so long as the Quai d'Orsay had been concerned not
to quarrel seriously with London, and replaced him in November
1881 by the notoriously combative Auguste Baudais. In January
1882 Maurice Rouvier offered to Gambetta and Freycinet his full
support for positive action in Madagascar, and for energetic
opposition to 'that pretension to rule the waves which the British
navy is all too ready to assume'. In March 1882 Freycinet gave
Baudais carte blanche to engineer a crisis by instructing him that
' our one and only preoccupation is to secure the defence of our
rights and interests'.21 Baudais complied by raking up old
disputes over French property-rights; by May he had broken off
relations with the Hova. In December 1882 a Hova delegation
visited Paris and London. In Paris they were offered, and rejected,
terms which amounted to a demand for complete French control
of Madagascar. In London, they found no effective support; the
Foreign Office was unwilling to exacerbate the Anglo-French
difference in Egypt by a serious dispute over Madagascar. French
naval action against Madagascar was initiated in February
18 8 3 - by a Deputy for Reunion who happened to be acting
minister of Marine in a stop-gap ministry; and in December
1885 the Hova queen was constrained to accept a French
protectorate.
These French challenges to British informal empire were
undoubtedly encouraged by the belief that Britain had overreached
herself by her forward policies of the later 1870s. In October 1879
the governor of Reunion had suggested, in a despatch marked by
Jaureguiberry for 'most particular attention' at the Quai d'Orsay,
that' the time has come to reassert our rights in Madagascar, now
that Britain is engaged in so many disastrous wars all over the
world'.22 The Isandhlwana disaster of January 1879 had indeed
21
Ministere des Affaires Etrangeres (M.A.E.), Madagascar 13, Rouvier to Gambetta,
22 Jan. 1882; Freycinet to Baudais, 2 Mar. 1882. Cited J. S. Swinburne, 'The influence
of Madagascar on international relations, 1878-1904', M.Phil, thesis. University of
London 1969, 118-19.
" M.A.E. Madagascar 11, Jaureguiberry to Waddington, 15 Nov. 1879, enclosing
despatch from governor, Reunion: cited Swinburne, 'Influence of Madagascar', IOJ-6.

been widely regarded in Europe as proof of Britain's military

decadence; and in October the British position in Afghanistan
collapsed with the massacre at Kabul. British weakness seemed
further demonstrated when in 1881 Gladstone failed to restore
British control in the Transvaal after the humiliating defeat at
Majuba. Moreover, as Admiral Jaureguiberry must have been
well aware, the expansion and technical modernisation of the
French navy since 1878 had by the early 1880s considerably eroded
the British naval preponderance that had been the sure shield of
informal empire in all its Protean forms. By 1882 France already
had, in commission or approaching completion, almost as many
up-to-date capital ships as Britain; at the same time most of the
numerically enormous British fleet had become technologically
obsolete - ' mere ullage', in the sardonic words of a British
admiral.
In 1882 the British found their informal empire under French
challenge not only in Guinea and Madagascar, but also at the
Congo mouth, where a trade comparable with that of the Niger
delta was at stake. The maritime Congo, in spite of the falls and
rapids that obstruct the river above it, was nevertheless the main
area of convergence for the trade-routes between the Atlantic and
the vast interior basin. In the hands of a European power, this
bottle-neck might have become a formidable barrier to British
trade; but London had kept it open by discouraging, when
necessary by force, any Portuguese occupation either at the Congo
mouth or on the adjacent ocean coasts. The sudden emergence
of French official interest towards the end of 1882 was quite
unexpected, even by the French themselves; for this interest was
neither the result of a planned search for empire, as in West Africa,
nor even in its initial stages a response to a supposed British
challenge, as in Madagascar. Its origins were indeed almost
fortuitous. Savorgnan de Brazza was hardly ' sent' by the French
government to the Congo: he chose his own objective for his
almost unofficial, largely self-financed, expeditions. Nor would
Brazza's activities necessarily have achieved any political result
but for the presence in the Congo basin of the agents of King
Leopold II.
Leopold II, king of the Belgians, had for decades been
convinced that the future of the Belgian economy depended on
its ability to exploit overseas markets and resources. Failing to
126

convince his subjects that ' Belgium needed a colony', Leopold
resolved to acquire, in his personal capacity and by the use of his
enormous personal wealth, a privileged sphere overseas for his
own economic activities and ultimately for those of Belgium. He
had sought, not political sovereignty, but vast monopolistic
economic concessions in underdeveloped South American states
or in European possessions in South-East Asia. It was only after
he had repeatedly failed to obtain a 'colony' of this kind that he
at last turned, in 1876, to central Africa. In July 1879, after earlier
attempts to reach the Congo basin from the East African coast,
he commissioned H. M. Stanley to work upstream from the Congo
mouth, founding stations and making treaties with local African
rulers which transferred to the king extensive economic conces-
sions and monopolies, but not formal political sovereignty.
Leopold of course intended from the outset to exercise ' informal'
political control; but the acquisition of formal sovereignty might
involve him in politico-legal difficulties in Europe, and possibly
in unnecessary administrative expense in Africa.
Leopold's project of creating a personal 'informal empire' in
the Congo basin seemed likely to be stillborn once Brazza had (as
the king quaintly put it) 'introduced politics into Africa' by
concluding in September 1880 treaties with the Teke ruler,
Makoko, whereby this chief ceded 'son territoire a la France'.23
Makoko's territory lay just upstream of the Congo rapids, at the
interior terminus of the trade-routes to the Congo mouth and the
Atlantic; and French control here might well isolate Leopold's
interior stations from the sea. It seemed impossible for Leopold's
organisation, the misleadingly-named Association Internationale du
Congo, even to compete with France. Whatever Leopold's 'non-
descript Association' was, it was certainly not a state; and at this
time it was not even attempting to acquire political sovereignty.
Leopold nevertheless exerted his considerable influence in Paris
in an attempt to prevent the ratification of Brazza's treaties; and
at first his prospects seemed by no means hopeless. Brazza had
been given no authority to conclude treaties of any kind.
Jaureguiberry, who as minister of Marine was both Brazza's
service chief and the minister responsible for colonies, disavowed
13
M.A.E. Memoires et Documents Afrique 59, Gabon—Congo 11, Leopold to
de Lesseps, 18 Sept. 1882: cited H. Brunschwig, ISavenement de t'Afrique noire (Paris,
1963), 159-60. Brazza's treaties of 10 Sept. and 3 Oct. 1880, printed ibid., 147-8.
127

and disapproved the explorer's activities on the Congo. These he

regarded as a mere distraction from his own priority areas on the
Guinea coast; and he had no intention of ratifying Brazza's
treaties, still less of actually occupying any part of the Congo
basin. France had neither 'historic links' with the Congo, nor as
yet any serious material interests there. Moreover, it soon became
clear that London, disturbed by the increasingly restrictive
trading policies adopted in French-controlled Guinea, would
strongly oppose the extension of similar restrictions to the Congo.
Brazza's treaties were nevertheless ratified by Charles Duclerc's
ministry in November 1882; and not by the usual decret but with
all the formality and publicity of a lot debated in the Chambers.
Ratification was not however an anti-British gesture, nor even a
sign that after the occupation of Egypt' Paris did not need to pay
the old deference to British susceptibilities'.24 In November 1882
London had still not formally communicated its decision to
exclude France from the control of Egyptian finances; and in Paris
hopes were high that London would have second thoughts.
Duclerc was therefore not looking for yet another quarrel with
Britain: he already had on his hands Madagascar, bequeathed to
him by Ferry and Freycinet. But the ratification of Brazza's treaties
did seem to offer, because of the diplomatically negligible status
of Leopold's Association, an opportunity of scoring a popular and
perfectly safe success overseas after the depressing fiasco in Egypt.
The focus of public excitement had swung abruptly from Egypt
to the Congo; Brazza had become the social lion and popular hero
of Paris. Ratification would be hailed as a' victory' for France both
over Leopold, whose attempts to prevent ratification were deeply
resented, and over Stanley, who as an ' Anglo-Saxon' was serving
as a convenient lightning-conductor for French anglophobia.
Duclerc therefore offered to London assurances both public and
private of complete free trade in the future French Congo. This
was a painless concession, in view of the absence of French
material interests and of Jaureguiberry's flat refusal, which
Duclerc did not at that time contest, seriously to involve France
in the Congo basin. On 1 December 1882 a British note, evidently
in confirmation of earlier verbal exchanges, acknowledged
(though without explicitly accepting) these assurances. The note
contained no hint of further British objection; indeed, twelve days
24
128

earlier Duclerc had felt able to assure the Chamber, very
emphatically, that ratification of Brazza's treaties entailed not the
slightest danger of 'international complications'.
However, Lord Granville at the Foreign Office had, even
before the ratification, decided to seek a more reliable safeguard
for British trade on the Congo than the mere assurances of a rather
lack-lustre French ministry. Prompted by Morier's earlier schemes
for using Portugal as a cat's-paw in Africa, in December 1882
Granville offered to Lisbon recognition of Portugal's 'historic
claims' at the Congo mouth, subject to conditions which would
guarantee complete freedom of trade. The opening of the Anglo-
Portuguese negotiation came as an unpleasant surprise to Duclerc;
but it was a challenge that he could not ignore. In January 1883
he countermanded Jaureguiberry's instructions forbidding further
French activity on the Congo; and insisted that the Marine should
at once organise a prise de possession effective of coastal areas giving
access to the interior.
Granville's expedient might have succeeded if it had led
without delay to the conclusion of an Anglo-Portuguese agree-
ment. But the Foreign Office guaranteed slow progress by sitting
down ' to think of all the things we can ever want from Portugal
in Africa'.25 When the treaty was at last signed on 26 February
1884, it was born into a diplomatic world very different from that
in which it had been conceived. Britain was now confronted not
only by France but by a Bismarck already affronted by Britain's
attempted application to Germany of her ' Monroe doctrine for
Africa', and soon to intervene decisively in the general politics
of the scramble. However, even before Bismarck's intervention,
Gladstone's Liberal government was finding it difficult to adjust
to the abrupt decline of an African hegemony which had
previously seemed almost as immutable as the laws of nature. The
Liberals had been elected in 1880 on a platform of impassioned
rejection of Conservative forward policies and annexations, but
the continued existence of informal empire had been the un-
expressed premise underlying this commitment to imperial self-
denial. Even when Granville had begun to understand what was
happening (it is doubtful whether Gladstone ever did), he was for
some time at a loss for a policy, especially in regions like the
" J. D. Hargreaves, Prelude to the partition ofWest Africa (London, 1963), 502, citing
Cfolonial] O[ffice) 147/52, F.O. to CO., 25 Nov. 1882 and minutes.
129

Niger delta where it was impossible to interpose a client-state as

the caretaker of informal empire. However, as early as June 1883,
Percy Anderson, the head of the Foreign Office's Africa
department, had drawn the painful, inevitable, conclusion that
informal empire was dead and gone, at all events in West Africa:
'Protectorates are unwelcome burdens, but in this case [the
Lower Niger] it is...a question between British protectorates,
which would be unwelcome, and French protectorates, which
would be disastrous.'26 The cabinet seemed to have grasped this
point when in November 1883 they authorised the British consul,
Edward 'Too-Late' Hewett, to establish protectorates in West
Africa. But then, characteristically, Treasury reluctance to finance
Hewett delayed his departure until May 1884.
Nevertheless, until the British began to gather the bitter fruit
of their mishandling of Bismarck in connection with Africa, their
defensive responses were by no means completely ineffective. At
the Niger delta, Hewett was not too late. By early 1884 the French
Niger companies had been completely discouraged by cut-throat
British competition and by the vigorous use of gunboats against
Africans who seemed inclined to trade with them. But Hewett and
the Foreign Office were alike taken aback to find Germany already
installed in Togo and Cameroun. London believed, in spite of
mounting recent evidence to the contrary, that Bismarck was still
unalterably opposed to colonial acquisition; but by mid-1884
Bismarck had decided that he could not prudently exclude
Germany from the African scramble. Already by 1883 Britain and
France appeared to be shutting the coastal doors to West Africa
at a pace which would soon exclude all other powers; and exclude
them commercially as well as politically, for in West Africa British
'free trade' was in practice almost as discriminatory as French
protection. British tariffs did not indeed formally discriminate by
country of origin. But they fell very lightly on textiles and
hardware, in which Britain could out-trade all her competitors;
and very heavily on commodities imported from other European
countries, such as tobacco and above all spirits, by far the most
important German export to West Africa. Once all the doors had
been closed, it would be too late for second thoughts on the value
of colonies: the world contained only one Africa. Bismarck was
26
F . O . Confidential print, no. 4 8 1 9 , m e m o r a n d u m by H. P. A n d e r s o n , 11 June 1885:
printed C. W. Newbury, British policy towards West Africa, select documents 1871-1914
(Oxford, 1971), 179-81.
130

also under considerable domestic pressure to safeguard German

commercial interests in Africa: the Hanse towns (important not
merely economically, but politically as constituent states of
Bismarck's federal Reich) were complaining bitterly against
British and French customs policies; and in July 1883 Hamburg
had asked for the acquisition of a ' trade colony'. In 1881 Bismarck
had deprived Hamburg of her greatly prized ' customs independ-
ence ' by coercing her into the Zollverein. If a West African colony
would now conciliate Hamburg, it would be sensible tofindher
one, provided that the diplomatic price was not too high. By
mid-1884, with Britain isolated by the consummation of her
Egyptian dispute with France, this price was at rock-bottom.
Yet Bismarck remained sceptical of the material value of
colonies; moreover, colonies under Reich sovereignty would
imply an annual colonial budget and therefore an additional
opportunity for the Reichstag to interfere in haute politique. Had
he not been affronted by Britain's 'Monroe doctrine for Africa',
it is possible that he would have found other ways of safeguarding
Germany's commercial interests. It was British behaviour in the
South-West African dispute between January and June 1884 that
made decisive and even showy acquisitive action an imperative
political necessity for Bismarck. The chancellor had originally
asked, in February 1883, for British protection for the coastal
establishment of the Bremen merchant, Liideritz, at Angra
Pequena. London had replied non-committally: but it was in fact
very unlikely that the Cape government, then in acute financial
difficulties, would be willing to extend its responsibilities to the
north of its Orange river frontier. Receiving no clear answer to
his original request, Bismarck then enquired in September 1883
whether Britain claimed sovereignty at Angra Pequena and, if so,
on what grounds. Neither the Colonial Office nor the Foreign
Office originally saw any political reason to claim such sovereignty,
the legal basis for which was in any case non-existent. A reply
disclaiming British rights had already been drafted when in
November the Cape premier, Sir Thomas Scanlen, then in London
in search of a loan, urged the Colonial Office to assert 'a South
African Monroe Doctrine, [with] all European powers given to
understand that it must be "hands off"'.27 The Colonial Office
27
Merriman Papers, T . Scanlen t o J. X. Merriman, 29 N o v . 1883: cited D . M.
S c h r e u d e r , The scramble for Southern Africa 1877-1#9> ( C a m b r i d g e , 1980), 121.

thereupon withdrew its approval of the proposed reply to Bis-

marck; and the Foreign Office began to experience misgivings
about the establishment of Germany so close to the nodal point
of British imperial communications at the Cape. Bismarck was
therefore informed on 21 November that the establishment of a
foreign Power in South-West Africa 'would infringe [Britain's]
legitimate rights'. 28 But he was not told why these rights would
be infringed; the British reply had carefully refrained from raising
the issue of formal sovereignty. On 31 December 188} Bismarck
riposted by asking for further and better particulars of Britain's
alleged rights.
The British had played their' paramountcy' card: Bismarck had
promptly trumped it by implicitly refusing to recognise anything
but formal sovereignty. Over six months passed without a reply
to his enquiry of 31 December. The Foreign Office had of course
no convincing reply to make; but Granville, virtually abdicating
responsibility, referred the question to Lord Derby at the Colonial
Office, who interpreted it as an invitation to annex — a course for
which the Cape Government was by now strongly pressing. The
question was, of course, who should pay — London or the Cape?
Six months were consumed in leisurely discussion of this question:
London, still convinced that Bismarck was basically opposed to
overseas empire, was evidently far more concerned for its
relations with Cape Town than with Berlin. Meanwhile, in April
1884 Bismarck informed the Foreign Office that he had extended
German protection to Liideritz. The precise nature of this
'protection' was obscure, as Bismarck almost certainly intended
it to be; but Granville sought no clarification. He merely passed
on the information to Derby, who appears to have ignored it. On
16 May Derby publicly stated that while Angra Pequena itself was
not ' claimed... as British territory, we had claimed a sort of
general right to exclude foreign powers from that coast'.29 On
3 June Bismarck learned that the Cape government was proposing,
with Colonial Office approval, to annex the entire coast as far as
the existing British enclave of Walvis Bay, to the north of Angra
Pequena.
28
F.O. 64/1101, Granville to Miinster (London), 21 Nov. 188}: cited W. O.
Aydelotte, Bismarck and British colonial policy: the problem of South West Africa
(Philadelphia, 1957; reprinted Westport, Conn., 1970), 36.
20
Derby to 'a deputation of South African merchants' 16 May 1884: cited (from
The Times, 17 May 1884) Aydelotte, Bismarck, 6;. Cf. Derby's speech in the Lords on
19 May, cited ibid., 66.
H2

As early as 14 May Bismarck had approached Paris for

assistance in resisting the British' Monroe doctrine'; but the insult
early in June, whereby London had permitted the Cape to take
action which had been forbidden to Berlin, demanded instant
retaliation. On 7 June Bismarck refused to recognise in any form
the Anglo-Portuguese Congo treaty of February, to which France,
Germany and other powers had already raised objections of detail.
He then drove Granville from his hiding-places behind the
Colonial Office and the Cape government; and on 17 June
extorted from the foreign secretary a veto on the proposed Cape
annexation at Angra Pequena and a verbal admission that London
had no right to object to Germany's establishment there. The
Colonial Office nevertheless continued, with Granville's approval,
to egg on the Cape to make annexations in South West Africa
which would have reduced the German holding to a mere enclave.
But by May 1884 Bismarck had already become convinced that
London was attempting by evasion and deceit to forestall him not
only in southern Africa but also in West Africa, for which Hewett
at last departed at the end of May; and he had decided to forestall
Hewett by sending his own expedition under Nachtigal.
In August 1884, after the final Anglo-French rupture on
Egyptian finance, Ferry not only responded to Bismarck's
proposal for a Franco-German ' equilibrium on the seas' against
Britain, but also accepted Bismarck's suggestion that the whole
complex of Congo questions, including the Anglo-Portuguese
treaty, should be referred to an international conference. The
Berlin West Africa Conference (November 1884 to February
1885) did not of course 'partition Africa'. It did however,
strongly prompted by Bismarck, establish the rule that henceforth
'effective occupation' was to be the sole valid title to African
coastal territory. 'Effective occupation' sounded the death-knell
of paramountcy and informal empire - concepts which had
always been indefensible in international law and which the
British did not attempt to defend at Berlin. Gone for ever were
the halcyon days when (as Lord Salisbury wistfully recalled in
1890) Britain had dominated almost the whole of accessible Africa
'without being put to the inconvenience of protectorates or
anything of that sort'. 30
J0
Salisbury in the Lords, 10 July 1890: cited Cecil, Life of Salisbury in (London,
•95*). « J - 6 .
133

Only in southern Africa, where Bismarck had originally

attacked British paramountcy, did a remnant of it survive his
onslaught. Bismarck did not wish to incur London's permanent
and unappeasable hostility by completely destroying British
hegemony in a region where not only ocean strategy but the
interests of British settlers were at stake. Once it had become clear
that Britain would recognise his West African acquisitions (and
his claims in New Guinea), he withdrew the claims he had made
on the Zululand coast and in March 1885 undertook in effect to
respect local British paramountcy outside his new South-West
African colony. He also refrained from extending this colony so
far eastwards as to obstruct the Cape's ' road to the north'. No
such delicacy marked Bismarck's proceedings in the Anglo-
Zanzibari East African sphere. By November 1884 this region was
under obvious German threat from the activities of Carl Peters;
and Granville now wished to establish a British protectorate over
the Zanzibari mainland, including the economically promising
Kilimanjaro region. However, Granville emphasised particularly
the strategic importance of the mainland coast for the routes to
India, especially since the French had become established in
Madagascar. But Gladstone, who had never really budged from
his Liberal fundamentalism on the wickedness and impolicy of
'annexations designed... to forestall other nations', scornfully
rejected Granville's proposal to protect 'the mountain country
behind Zanzibar with an unrememberable name',31 and simply
ignored the strategic argument. Granville then attempted to melt
Bismarck's heart by communicating to Berlin a historical account
of Britain's special relationship with Zanzibar. This was precisely
the kind of pseudo-title that Bismarck was now rejecting on
principle. With London distracted by the catastrophicfinaleof the
Sudan crisis and then by a dangerous confrontation with Russia
in Afghanistan, he was able almost effortlessly to destroy the
Anglo-Zanzibari informal empire, and force London to share the
Zanzibari mainland, by ruthless pressure upon Britain's bewil-
dered client, the sultan.
31
The letters of Queen Victoria, 2, iii (London, 1928), 593-4: Gladstone to the Queen,
23 Jan. 1885. S. Gwynn and G. M. Tuckwell, Life of Sir Charles Dilkt (London, 1917),
11, 83-4: Gladstone to Dilke, 14 Dec. 1884.
134

THE SCRAMBLE FOR THE INTERIOR
Until about 1880 British naval preponderance, by preventing

foreign annexations, and by making British annexation unneces-
sary and therefore apparently unprofitable, had preserved on the
coasts of Africa an almost completely stable situation in terms of
European political control. This stability was disturbed when,
after 1880, France successfully challenged British hegemony in
West Africa and Madagascar. Madagascar was written off by
Britain, probably in the hope that it would console France for her
disappointment in Egypt. On the African mainland, however, the
British attempted, by the Anglo-Portuguese treaty over the
Congo mouth and by the proposed establishment of protectorates
from the Gold Coast to Cameroun, to confine the area of
'competitive instability' to western Guinea and the upper Niger
basin. These responses might well have succeeded but for the
intervention of Bismarck. By frustrating the British attempt to
formalise by protectorates most of their informal West African
sphere, by destroying the Anglo-Portuguese treaty, and above all
by insisting on the criterion of effective occupation, Bismarck
perpetuated the existing area of instability. He then promptly
extended it from West to East Africa by destroying the Anglo-
Zanzibari hegemony. His East African acquisitions, which
included not only German East Africa proper, but Witu on the
Somali coast, had hinterlands which could be extended in the
south to Rhodes's ' sphere of aspiration' between Lakes Tangan-
yika and Malawi, and in the north to the newly-created vacuum
of power on the upper Nile. Indeed, almost all the coastal
annexations since 1882 had hinterlands which clashed if extended
far enough, and were therefore an invitation to competition in
the interior. Once interior competition had begun, and the
criterion of effective occupation had prevailed in the hinterlands
as well as on the coast, the scramble became virtually self-
propelling almost throughout the continent. No new stable
situation could now be reached until the whole continent had been
carved up into internationally-recognised European 'spheres'.
Even in Morocco, where European annexation encountered very
severe constraints arising from the country's key position in
Mediterranean and Atlantic strategy, partition was merely post-
poned, not prevented.
135

The extension of the scramble to the interior was also stimulated

by the emergence of King Leopold's Congo Independent State.
Although Leopold had responded to the ratification of Brazza's
treaties by claiming sovereignty in his own treaties with Congo
chiefs, 'sovereignty' acquired by a non-state recognised by
nobody was evidently well-nigh worthless. Leopold owed his
success to his brilliantly skilful manoeuvring between Britain and
France; and to Bismarck's sponsorship of a Leopoldian Congo
State as a neutral solution to the Anglo-French Congo dispute.
Leopold's willingness to give (though not to fulfil) any and every
assurance of complete free trade in his new state spoke strongly
in his favour; for Bismarck wished if possible to avoid admitting
French protectionism into the Congo basin as the price of
destroying British informal empire at the Congo mouth. But
Bismarck could not have imposed the Leopoldian neutral solution
against French opposition. Leopold had, however, in April 1884
disarmed this opposition by granting to France a right of first
refusal {droit de preference) should he ever wish to dispose of his
territories. Jules Ferry, with the reversionary title-deeds in his
pocket, now saw no reason to oppose the creation of a ' state'
which nobody then expected to survive for long. In November
1884, with French concurrence, Bismarck recognised the flag of
Leopold's Association as that of 'a friendly state'. London at first
refused to follow suit. The Foreign Office was scandalised by a
request for recognition of a ' state' which as yet had neither real
existence nor known frontiers; and there was intense resentment
at Leopold's 'shabby and mischievous trick' 32 in granting the
droit de preference to France. But Bismarck soon extorted recog-
nition by threatening to withdraw support at the Berlin Conference
for the British claim to a commercial monopoly on the lower
Niger.
On the lower Congo, the difficulties of adjusting the Portuguese,
Leopoldian and French claims are reflected in the curious
territorial jigsaw that still survives. In the interior, however,
Leopold established with astonishing ease the frontiers of an
enormous empire. It was indeed typical of the king that he
hastened to stake out these grandiose interior claims while other
32
P.R.O. 50/29/198 (Granville Papers), T. V. Lister, memorandum, 20 May 1884:
cited J. Stengers, 'King Leopold and Anglo-French rivalry, 1882-1884' in P- Gifford
and W. R. Louis, France and Britain in Africa (New Haven and London, 1971), 1 21—66,
at 162.
136

powers were still contemplating, often rather ruefully, the
problems of establishing 'effective occupation' in their new
coastal possessions. In August 1884 Leopold put forward a claim
to a great quadrilateral of territory comprising ' all central Africa,
the very core of the continent'.33 When this claim was opposed
by neither Paris nor Berlin, he steadily added to it, especially in
the south-east; in February 1885 the frontiers recognised by
Belgium were not unlike those which exist today. These frontiers
were then submitted to the powers. Britain had every reason to
contest frontiers which in the south-east now extended almost to
the Zambezi and which were, thanks to the droit de preference,
potentially the frontiers of French territory. Yet the Foreign
Office accepted them without even a query. It was August; Percy
Anderson was on holiday, and Anderson's deputy made the
'stupid blunder' of confusing Leopold's extended frontiers with
the boundaries of the Congo Free Trade Area. The stimulus to
interior competition created by this vast internal sovereignty,
whose frontiers now approached the hinterlands of all the
partitioning Powers except Italy, was enhanced by Leopold's
apparently insatiable appetite for even more African territory. By
1886-7 he was already reaching out towards Uganda and the
upper Nile, and planning to persuade Emin Pasha to transfer his
allegiance from Egypt to the Congo State. When this project failed
he began in 1890 to despatch expeditions of his own to establish
' effective occupation' in the western basin of the upper Nile, parts
of which were in fact occupied, though rather precariously,
between 1892 and 1894.
Until the mid-1890s the Congo State brought Leopold no
profits, but losses on a scale which threatened to exhaust even the
king's immense resources. The motivation of King Leopold's
imperialism was nevertheless overwhelmingly economic. As a
non-Power, the king could conduct his acquisitive operations
undistracted by considerations of prestige; and he played strategic
games only in return for tangible territorial rewards. His ultimate
objective of ensuring the long-term viability of the Belgian
economy was ' Leninist' avant la lettre, although he had to play
in person the Leninist role which Belgian capitalism for long
refused to assume. Leopold was of course not alone as an
economic imperialist: in the first stage of partition down to 1885,
" Stengers, "King Leopold', 165, citing Courcel (Berlin) to Ferry, 50 Aug. 1884.

before experience had rubbed some of the gilt off African El

Dorados, economic motivation was often preponderant. But
the precise nature and strength of this motivation varied greatly
from power to power and from region to region; and in the 1880s
no other power acted from economic motives so far-sighted and
so rationally compelling as those of King Leopold. Some
Frenchmen saw in the Niger Sudan an exporters' El Dorado; and
there is evidence that in the early 18 80s some French policy-makers
valued West African coastal acquisitions mainly as gateways to
the 'mythically' rich interior. But in this period the British in
West Africa were merely attempting to protect their established
trading areas, doubtless vital to individual firms but trivial in
relation to the overall metropolitan economy. Even more trivial
in this context were the customs revenues of French and British
colonies; but the urgent need to increase them during the acute
depression of 1878-84 was a very sharp spur to local Anglo-
French competition on the Guinea Coast. As for Bismarck's
economic imperialism, it was in his own eyes little more than the
reluctant but unavoidable taking-up of rather dubious options.
Even in the initial stage of the scramble, however, considera-
tions of prestige were often crucial, especially for France. French
policy towards Madagascar in 1881—2 was driven almost entirely
by prestige: the British naval visit of July 1881 had in no way
injured the purely economic interests of France in Madagascar.
What stuck in the throat of even so capitalist an imperialist as
Maurice Rouvier was the off-hand arrogance with which Britain
appeared to be using her naval hegemony to extinguish the last
remnants of 'historic' French influence in the island, thereby
treating France as a negligible quantity in the Indian Ocean.
Again, for the French occupation of Tunisia prestige was even
more important than the safeguarding of investments; eviction
at the hands of the despised Italians was utterly intolerable to a
France still struggling to recover her international standing after
the catastrophe of 1870—1. Prestige, too, played its part in
Bismarck's colonialism. Bismarck moved from rather hazy
projects of informal trading spheres to outright territorial an-
nexation largely in reaction to the affront to German self-esteem
(and to his own political reputation) implicit in the apparent
arrogance and duplicity with which London treated Germany's
claim at the economically negligible site of Angra Pequena.
138

Strategy was also present, mainly as a British motive. It was

certainly relevant to the occupation of Egypt in 1882, and
overwhelmingly dominant in the 1888 decision to remain in
Egypt. However, the African consequences of this decision - a
further wave of strategic imperialism directed to threatening or
defending the Nile waters - do not emerge until the second active
phase of partition from 1889 onwards. In South and East Africa,
British opposition to Bismarck's acquisitions was in part prompted
by strategic defence of the routes to India. But the most powerful
motive for London's evasive resistance to Bismarck in South
Africa was evidently an over-riding anxiety to defer to the wishes
and susceptibilities of the Cape government, which was all too
often touchy and demanding in its dealings with 'the imperial
factor'. With this semi-domestic, semi-diplomatic motivation may
perhaps be compared Bismarck's relations with Hamburg.
Another type of domestic motivation - perhaps a rather puny
specimen of ' manipulated social imperialism' - appears in
Duclerc's ratification of Brazza's Congo treaties. Until the British
riposte to this ratification injected an element of great-power
rivalry which Duclerc felt compelled to take seriously, French
' imperialism' on the Congo had been little more than an exercise
in public relations: Duclerc had not originally contemplated, and
Jaureguiberry had been determined to prevent, its development
into a genuine prise de possession.
The political map of Africa in 1886 reflects the extent to which,
since 1879, Britain had failed to defend her informal empire. On
the western coasts of Africa, from Senegal to the Orange river,
complete British informal hegemony had been replaced by new
French, German, Leopoldian and (at Cabinda and on the south
bank of the maritime Congo) even Portuguese holdings. Only
between Lagos and Calabar had Britain succeeded in retaining a
remnant of her previous informal empire. In eastern Africa,
Madagascar had been irrevocably lost to France; on the mainland
Germany had taken a very large bite to the south, and a smaller
bite (Witu) to the north, of what she had permitted Britain to retain
of the former Anglo-Zanzibari sphere. In the former Egyptian
coastal possessions France had in 1884-85 sought to facilitate her
communications with Madagascar and Indo-China by reviving
and extending a claim in the Obok region which had been
dormant since 1862. To forestall France on the northern (Gulf of
• 59

of ZANZIBAR
Boundary of Congo State

notified to the powers. Aug 1885 SWAZILAND
ZUIULANO
• — • — Boundary of Free Trade Zone
established by the Berlin Act NATAL
BASUTOLANO
I It I Ml Ottoman suzerainty
I Portuguese
^ _ _ _ ^ French 15OOkm
j - ' - ' ; ' ' / j Congo Independent State
I l l l l l l l German 1000 miles
4 Thefirstphase of partition, to c. 1887
140

Aden) Somali coast, London would have accepted Italy as a

substitute for Egypt; but when the Italians hesitated, the British
established protectorates of their own. But soon even the cost-free
Italians were no longer welcome; by March 1887 Salisbury was
regretting that in 1885 they had been permitted to replace Egypt
at Massawa on the Red Sea. Nevertheless, in 1889 they acquired
the former Egyptian and Zanzibari possessions on the Indian
Ocean coast from Cape Guardafui to Kismayu.
However, during the second active phase of partition, from
1889 to 1899, enormous British gains in the interior were to
compensate for the coastal losses of the 1880s. By 1900 the
political map of Africa had again been transformed. The British
positions in Egypt and South Africa had acquired interior
extensions so vast that they were now separated only by German
East Africa. In West Africa, British diplomacy had indeed proved
no match for French military action in the upper Niger basin, and
Sierra Leone had lost its potential hinterland; but the Nigerian
coastal remnant of informal empire had expanded northwards to
the latitude of Sokoto, Bornu and Lake Chad. During the later
1880s the British adjusted to the loss of informal empire and
largely abandoned their inhibitions about formal annexation. By
the 1890s they had become anything but 'reluctant imperialists'.
In 1885-6, however, British policy-makers did not yet hanker
after interior expansion. To Lord Salisbury, the only useful test
of the value of African territory was whether British traders were
willing to 'go in': where, as in East Africa, they seemed reluctant
to 'go in', Salisbury preferred to barter African territory for
diplomatic advantages in Europe and the Near East - for
Bismarck's 'help in Russia and Turkey and Egypt'. 34 In 1886
Nile-waters strategy still of course lay in the future; and in spite
of open hinterlands, the collapse of Isma'Il's empire and the rise
of King Leopold's, the British, and soon the Germans too, paused
for breath in East Africa. There was a similar ' loaded pause' from
1885 to 1889 in West Africa, where the French were temporarily
dismayed by the financial and military costs of empire-building.
Moreover, the fall and disgrace of Jules Ferry in March 1885 had
revealed the intense disquiet of French opinion at the dispersal
of military resources overseas - especially when Bismarck
>4 Salisbury to Iddesleigh, 24 Aug. 1885: cited Cecil, Life of Salisbury 111, 230.

appeared to be encouraging the expansionist policies which

caused this dispersal.
By 1888, Bismarck too was 'sick and tired of colonies'. They
had brought with them the financial losses and domestic political
difficulties which he had foreseen; worse still, the clamour from
the colonialists for their further extension, largely at Britain's
expense, was beginning to compromise Germany's European
security. By 1888, Bismarck needed British support to balance the
rapprochement between a Russia once more active in the Balkans
and a France now apparently dominated by a 'saviour with a
sword' and obsessed by bellicose fantasies of revanche. He alienated
the German colonialists, who were normally among his most
reliable political supporters, by refusing to compete with Britain
either in Uganda and the upper Nile, or between Lakes Tanganyika
and Malawi. So successful was his veto on expansion that by 1889
he found himself quite without territory with which to bargain
with London for an African settlement which had now become
an urgent European necessity. October 1889 therefore saw the last
and oddest of Bismarck's African acquisitions, on the Somali coast
from Vitu northwards to Kismayu. The territory itself was almost
worthless; but London would be reluctant to leave it in German
hands, because its hinterland could be extended to the upper Nile.
After Bismarck's fall in March 1890 his successors, having 'cut
the line to Russia' by refusing to renew the Reinsurance treaty,
were all the more anxious for an African settlement which would
permit close Anglo-German collaboration in Europe; but they
were far less able than Bismarck to resist the colonialist demand,
now strongly backed by Kaiser Wilhelm II, for massive
acquisitions in Africa. Salisbury too was alarmed by the Franco-
Russian rapprochement and anxious for an African settlement
with Germany; but he was quite unable to make territorial
concessions large enough to satisfy the German colonialists. In the
Tanganyika-Malawi region, he dared not sacrifice either Rhodes's
recent treaties or the interests of the electorally influential Scottish
missionaries. In the German hinterland west of Lake Victoria he
was under very strong pressure, even within the cabinet, to
support extravagant claims which H.M. Stanley had made on
behalf of Mackinnon's East Africa Company. Much British
opinion, having at last recognised that informal empire was gone
beyond recall, was by 1890 in rampantly acquisitive mood; and
142

for the moment saw Germany as Britain's most dangerous

colonial rival. Salisbury, for all his mockery of African El
Dorados whose very lakes turned out to be mere ' beds of rushes',
could not ignore these pressures; and he was himself determined
to retain Zanzibar in the interests of Indian Ocean strategy and
to exclude Germany from the upper Nile in the interests of
Egyptian security. He risked an open breach in his cabinet by
refusing to press the very dubious Stanley—Mackinnon claims, on
which indeed Berlin was quite determined not to yield. But he
had nothing more to offer to the Germans in Africa; and he
achieved an otherwise quite unattainable settlement by throwing
Heligoland into the bargain as a bait particularly attractive to the
German emperor. Berlin gave way in Zanzibar and the Nyasa
hinterland; relinquished Bismarck's new Somali coast protec-
torate ; and recognised a ' British sphere' on the upper Nile - a
curiously belated triumph for 'paramountcy', in that Britain,
without the slightest claim of historic right or effective occupation,
had nevertheless since the mid-1870s regarded herself as para-
mount over 'the country that belongs to Egypt'.
In the Anglo-German Agreement of 1 July 1890 the more
subtle imperatives of diplomacy and strategy had triumphed over
unrestricted competition for African territory; and for the next
two years or so Anglo-German relations, both in Africa and in
Europe, were so cordial that French and Russian diplomatists
sometimes suspected that Britain had secretly joined the Triple
Alliance. Meanwhile, British naval strength was being reinforced
by the building programmes of 1885 and 1889. In 1890 and 1891
Salisbury concluded major African territorial agreements not
only with Germany but with France, Italy and Portugal. These
agreements secured for Britain much of her lion's share of the
interior: Britain's successes in Africa in the early 1890s were, no
less than her failures between 1880 and 1885, a direct reflection
of her European diplomatic situation and her relative naval
strength. By mid-1891 the broad pattern of interior partition had
been settled throughout the continent, except for the British and
French frontiers with German Kamerun and of course the bound-
aries between the British Upper Nile sphere and the French and
Congolese hinterlands to the west and south-west of it. So long
as the Egyptian dispute remained unsettled, France would never
willingly recognise the upper Nile sphere. Nevertheless, in
143

August 1890 Paris was willing to conclude an approximate

delimitation of spheres in the Nigerian hinterland — if only as a
demonstration that the new Anglo-German friendship was not
systematically directed against France. Paris accepted a line from
Say on the Niger to Barrawa on Lake Chad as the southern limit
of the French 'Mediterranean possessions' in Africa; London
recognised the French protectorate over Madagascar. This settle-
ment enabled the French ' Chad plan' to go forward north of the
line without diplomatic complications; it would at the time have
seemed not ungenerous but for Salisbury's notorious quip about
the ' light land' ceded to France. South of the line, the agreement
gave George Goldie's Niger Company a free hand from Sokoto
to Bornu: a free hand of which he had made embarrassingly little
use when French expeditions from the west, relying on a very
literal interpretation of the words ' Mediterranean possessions' in
the 1890 Agreement, began to appear in these regions a few years
later.
Agreements with Italy in March and April 1891 recognised the
Italian sphere in the eastern Horn and excluded Italy from the
British Nile valley sphere, after a period when Crispi had
embarrassed Anglo-Italian, and indirectly Anglo-German rela-
tions, by attempting to insist on British recognition of an Italian
title in the eastern Sudan. Southern Africa was partitioned by the
Anglo-German Agreement of July 1890, which settled the
internal frontiers of German South West Africa; and by the
Anglo-Portuguese Agreement of June 1891. The friendly relations
between London and Berlin enabled Salisbury to impose upon an
isolated Portugal a settlement which the Portuguese bitterly
resented; for Salisbury used Rhodes's 'effective occupation' to
brush aside as 'archaeological claims' Portugal's 'historic rights'
in south-central Africa. Effective occupation had now triumphed
not only over informal empire but also over decayed formal
empire. After 1891 it soon became the only title worth having in
the interior; and the remaining years of partition are the classic
period of' steeple-chases' by rival expeditions racing to establish
effective occupation in those West African and Nilotic regions still
open to competition.
144

SOUTH AFRICA AND THE UPPER NILE
Until the mid-1890s the European scramble had surprisingly little

effect upon Anglo-Afrikaner rivalries in South Africa. The British
occupation of Bechuanaland in 1885 was certainly hastened by the
fear that Germany might obstruct the Cape's northward expan-
sion. But imperial action against the Transvaal Afrikaners who
had recently set up the ' freebooter republics' of Stellaland and
Goshen athwart the Cape's road to the north was already being
contemplated; and could in any case hardly have been much
longer delayed if it was to be effective. In fact, Bismarck interfered
neither with the Cape's access to the north nor with Britain's local
paramountcy. Until 1894 the crucial new factor was not the
German presence but the spectacular gold-nourished rise of Paul
Kruger's South African Republic, which challenged the previously
overwhelming economic and (white) demographic preponderance
of the Cape over all the other South African states and colonies,
and could therefore be seen as a potential threat to British
paramountcy throughout the sub-continent. From 1891, in the
interests alike of British imperialism and Cape sub-imperialism,
Cecil Rhodes as Cape premier attempted to harass Kruger into
a closer economic (and ultimately political) South African union.
But in 1894 his attempt to strike the decisive blow, by acquiring
from Portugal Lourenco Marques (Kruger's only rail outlet to the
sea not under British control), was frustrated by German diplomatic
opposition.
By 1894 Berlin had become deeply suspicious of British policy
both in Europe and in Africa. Germany retaliated by attempting
to undermine the British position in South Africa through
diplomatic support for the Transvaal. Fear of preponderant
German influence at Pretoria brought British policy-makers,
especially Chamberlain, to connive at the campaign of internal
subversion which Rhodes had launched in the Transvaal, and
which in December 1895 culminated in the fiasco of the Jameson
raid. But the German attempt to exploit this situation by the
'Kruger telegram' was hardly less a fiasco. It provoked a very
combative British response and a disquieting improvement in
Anglo-French relations. By March 1896, especially after the shock
which the Triple Alliance had sustained through the Italian
catastrophe at Adowa, Germany was ready to abandon her
145

PORT
GUINEA^
III 11IJ Ottoman suzerainty

8 Portuguese
Br,,,sh
BRITISH
French BECHUANALAN
K-- •:! Congo Independent State
[ M i l l German
u — - J Spanish
= (Italian
1500 km
1000 miles
5 The scramble for the interior, to 1895
146

unfriendly intervention in South Africa and to resume more

normal relations with London. But she demanded in return a
guarantee of British support for the Triplice; and this Salisbury
provided by his token advance in the Sudan (above, p. 113).
The Kruger telegram had been the climax to a deterioration in
Anglo-German relations dating from 1893, when Berlin had
become convinced that the Liberal government in London was
seeking to evade the commitment, undertaken by Salisbury in
1887, to support the Triple Alliance in the Mediterranean, where
Italy's situation was now growing more precarious as Franco-
Russian naval strength increased. The Germans also believed, not
entirely without reason, that in Africa Rosebery tended to treat
them as a power of secondary importance. Minor African disputes
multiplied and were allowed to fester; by 1894 both powers had
adopted a policy of silent retaliation against real or supposed
unfriendly acts in Africa. In December 1893 Berlin had construed
as deliberate duplicity a minor African move by Rosebery which
was in fact no more than clumsy and discourteous. The Germans,
then negotiating their Cameroun frontier with France, retaliated
by conceding to France access to the Benue and so to the lower
Niger, regions where the political activities of the French agent,
Lieutenant Mizon, in apparent contravention of the Anglo-French
1890 Agreement, had recently aroused intense resentment in
London. Outraged by this German ' treachery', Rosebery in turn
retaliated in April 1894, by writing into an agreement with King
Leopold a lease to Britain of a 'corridor' of Congolese territory
extending from Uganda to Lake Tanganyika, and thus inserted
between German East Africa and the Congo State. London was
well aware of the strong German objection to such a corridor;
what Rosebery had not foreseen was the lengths to which Berlin
would go in opposition to it. The Germans threatened Leopold
with a refusal henceforth to treat the Congo State as neutral, the
British with an international conference on Egypt, and both
parties with concerted Franco-German opposition to the entire
agreement.
Under this pressure Leopold, lacking effective support from
Rosebery, was forced to abandon the 'corridor clause'. Mean-
while, the Germans had been encouraging the efforts of Gabriel
Hanotaux at the Quai d'Orsay to frustrate Britain's major
objective in concluding the agreement: the obstruction of French
M7

access to the upper Nile by interposing Leopold as the lessee of
the entire south-western Sudan, much of which was believed to
be already under the king's 'effective occupation'. Leopold's
Belgian ministers forced him to abandon almost all the leased
territory, thereby restoring French access to the Nile, when
Hanotaux threatened to denounce, as a breach of Belgian
neutrality, the secondment of Belgian officers to the Congolese
force publique; against this threat the king and Rosebery were alike
powerless. The dispute over the Anglo-Congolese Agreement,
and the complete collapse of Leopold's position under German
and French pressure, brought England and France for the first
time into open and direct diplomatic conflict on the upper Nile.
The crisis also alerted the French public to the close connection,
both political and hydrological, between the upper Nile and
Egypt: Eugene Etienne insisted to the Chamber that 'c'est la
question egyptienne qui s'ouvre ainsi devant vous'. 35 Finally, it
dethroned the ' Chad Plan' from its previous official priority over
the upper Nile: in future, French moves on the upper Nile would
no longer be the semi-conspiratorial initiatives of individual
ministers but formal acts of state carrying cabinet approval.
After the collapse of the Anglo-Congolese Agreement, Rose-
bery attempted to negotiate an upper Nile settlement directly with
Paris. Hanotaux refused to recognise the British sphere; but he
was prepared to bind France not to enter the upper Nile basin
if Britain would give a similar undertaking and would compensate
France generously enough in West Africa.' Mutual self-denial' on
the upper Nile would of course have solved Britain's purely
strategic problem; but Rosebery, unlike Salisbury, believed that
African territory was worth having for its own sake; and he
refused to make, either on the upper Nile or in West Africa, the
territorial sacrifices demanded. This failure to give Nile strategy
priority over territorial acquisition doomed the negotiation; and
Paris thereupon went ahead with its own Nile strategy, despatch-
ing a mission under Victor Liotard in November 1894. Liotard's
mission provoked the famous 'unfriendly act' declaration by Sir
Edward Grey on 28 March 1895, although by that time local
difficulties had brought Liotard to a standstill. Later in 1895 a
new start was made by the indomitably energetic Captain
J.-B. Marchand; but even Marchand was not firmly established in
" Journal Officiel, Debats, Chambre, 7 June 1894.
148

the Nile basin until November 1897, and did not reach Fashoda
until July 1898.
The major objective of these expeditions was to enforce by a
potential threat to the Nile waters an early British evacuation of
Egypt. To those who promoted and planned these missions —
officiers soudanais and permanent officials of the ministere des
Colonies - the importance of Egypt for the Mediterranean strategic
balance was a very remote concern. For them, the object of
enforcing evacuation was to achieve a spectacular enhancement
of French prestige; and between 1896 and 1898 even Hanotaux,
as foreign minister, consistently discussed the Egyptian question
in terms of prestige rather than of Mediterranean strategy. A
subsidiary objective was territorial acquisition in the southern
Sudan; but although French diplomatists sometimes talked in
terms of access to the navigable Nile for (non-existent) French
trade on the upper Ubangi, these 'economic' arguments were
mere fig-leaves to cover motives less avowable in diplomatic
discourse. When in October 1898 the foreign minister, Theophile
Delcasse, attempted, at the risk of diplomatic rupture and even
war, to extort from Salisbury territorial concessions in the Bahr
al-Ghazal, he was well aware that this triste pays was depopulated
wilderness where it was not a swamp. The mere territory was
worthless; not so the demonstration that even if France could not
compel the British to evacuate Egypt, she could at least bring them
to abandon their pretensions to a total monopoly of the Nile basin.
In thefinalphase of the upper Nile contest, territorial acquisition
became for the British, hardly less than for the French, a matter
of pure prestige: as a move in Nile-waters strategy, the acquisition
of territory by Britain was a quite unnecessary elaboration. Yet the
spectacular Anglo-Egyptian Sudan campaign of 1896—8 was not
originally launched for the sake of British territorial acquisition,
nor even as a move in upper Nile strategy. Indeed, in March 1896
Salisbury feared that his token advance in the northern Sudan
might provoke a genuine French advance in the south, where
effective British action was still impossible. The minimal advance
undertaken as a move in European policy was soon extended by
Kitchener's pressure for a more ambitious campaign and by
warnings from Lord Cromer, the British consul-general in Egypt,
that the khedive might publicly denounce a mere military gesture
which brought no territorial gain to Egypt. By December 1896
149

the objective of the campaign was to recover for Egypt as much
of her Sudan territory as could be reconquered, without incurring
any military risks, by Egyptian troops without British reinforce-
ment. At the first sign of danger the Egyptians were to halt
indefinitely. Nevertheless, by the end of 1897 Kitchener had
manoeuvred himself into a position apparently so precarious that
British reinforcement now seemed the only alternative, at best to
a humiliating retreat, at worst to a military catastrophe. But once
British troops had been committed, it was quite out of the
question to maintain these very scarce resources indefinitely on
the defensive. A rapid and of course victorious conclusion to the
campaign was now imperative; and any less decisive conclusion
than the complete destruction of the Mahdist state would now
have seemed a very damaging defeat for London.
After Kitchener's victory on the Atbara in April 1898, it began
to seem possible that his advance might after all be in time to
forestall, or at least to challenge, the French on the upper Nile.
By June 1898 the campaign had, at last, developed into a practical
move in upper Nile strategy. Its immediate political objective had
also been radically altered: no longer the restoration of the Sudan
to Egypt, but the creation of a British title to Sudan territory. On
3 June 1898 Salisbury asserted a joint British and Egyptian right
of conquest over ' the whole of the Mahdi State from Haifa to
Wadelai',36 thereby extinguishing the exclusive Egyptian right
derived from her former possession of the Sudan. The recon-
ciliation of Egypt's share in this joint right of conquest with
complete de facto control of the Sudan by Britain was treated by
Salisbury as a question of mere administrative detail, which was
ultimately delegated to Cromer for solution. As a prominent Tory
politician put it in October 1898,' We don't care whether the Nile
is called English or Egyptian, or what it is called - but we mean
to have it.' 37 Yet the protection of the Nile waters did not dictate
a British acquisition of the whole Sudan: and even in the southern
Sudan, the reinstallation of Egypt would have been a complete
strategic solution.
At Fashoda Salisbury was defending not merely the Nile waters
and Britain's security in Egypt but, as a prescient French
16
F.O. 78/5050, Salisbury to Cromer, tel. no. 47, 3 June 1898.
" George Wyndham to Wilfrid Blunt in October 1898: W. S. Blunt, My diaries
(London, 1919; reprinted 1932), 299-300.
150

diplomatist had predicted in December 1894, ' la grandeur meme

de l'Angleterre'.38 From the Cabinet down to the popular press,
British opinion saw in the Anglo-French confrontation a crucial
test of Britain's strength and effectiveness as a power; and so did
continental onlookers both official and unofficial. The intensely
provocative manner in which the French appeared to have
presented their challenge meant that anything less than total
success in meeting it would be construed as a defeat. In October
1898 Salisbury was not joking about' beds of rushes', monstrously
prolific though these were in much of the disputed territory. Yet
the imperative need to eject the French from the southern Sudan
did not entail, any more than the defence of the Nile waters, a
virtual British annexation of the whole of the Sudan.
Nor did any conceivable economic motive. British opinion no
longer saw El Dorados on the upper Nile; and certainly none in
the nothern Sudan, which ' picturesque' reporting of the Sudan
campaign had presented as a mere inferno of sun, sand, stones and
thorns. For Cromer, the southern Sudan was simply 'large tracts
of useless territory which it would be difficult and costly to
administer properly'; and in October 1898 Salisbury minuted
'wretched stuff' on a military intelligence report describing the
alleged wealth of the Bahr al-Ghazal.39 As for the northern Sudan,
Cromer predicted, quite correctly, that its economic development
would for the foreseeable future be a charge upon government:
private capital would not be attracted by the very long-term
investment, especially in communications, which would be
needed to make it profitable. Yet the clamour for annexation was
almost universal in Salisbury's own Tory party, and very
widespread even beyond it - not least in many normally non-
imperialist circles, which opposed the restoration of the Sudan to
Egypt on religious or humanitarian grounds. Especially when
Salisbury refused to humiliate France further by declaring a
protectorate over Egypt itself, a ' British Sudan' became for the
imperialists an indispensable symbol of Britain's imperial
hegemony over the Nile — and perhaps indeed of Britain's general
imperial greatness.
During the 1890s the theorists of imperialism had emphasised
38
Documents Diplomatique: Franfais, Ie serie, x i (Paris, 1948), n o . 303, d'Estournelles
de Constant t o H a n o t a u x , 3 D e c . 1894.
39
Salisbury Papers, Cabinet m e m o r a n d a , Cromer t o Salisbury, 5 N o v . 1897. F . O .
7 8 / J 0 5 1 , Salisbury, minute o n a report by A . E . W . Gleichen, 20 O c t . 1898.

PORT
GUINE
SIERRA'
LEONE
LIBERIA
15OOkm
1000 miles
IIIIMill Ottoman suzerainty
B8S88& Portuguese
S I—
K>^3 French
H i l l Belg.an
J-TTTTT1 In 1912. the European partition was completed by
iiiUUI German 1 The Italian annexation of Tripoli (Libya)
Spanish 2 The definitive partition of Morocco between France and Spam
Italian
6 Africa partitioned, 1902

the economic imperative even more strongly than in the 1880s:
the future greatness of any great power was, it seemed, crucially
dependent upon its acquisition of exploitable colonial territories.
But except in southern Africa economic considerations seem in
the 1890s to have played a comparatively minor role in the actual
acquisition of territory. In the great upper Nile contest, economic
motives were of negligible importance for Britain and France
alike; nor had they played any major part in the Anglo-German
and Anglo-Italian diplomatic settlements of the early 1890s.
Chamberlain seems to have accepted the apocalyptic 'colonial
empire or downfall' doctrine; but in West Africa in 1897-8, where
some economic advantages were indeed at stake, his bellicose
intransigence was apparently aroused less by the possible loss of
these advantages than by his sense of affront at French expeditions
staking claims all over 'British' hinterlands. 'Cheek' he once
minuted on a statement of French claims based upon these
exploits:40 one word could hardly say more. Chamberlain seemed
to regard the loss of obscure West African villages as an intoler-
able national humiliation; and in mid-1898 he was ready, even
eager, to break off negotiations and risk war with France over
trivial territorial differences of no conceivable economic import-
ance. It was the rather similar exalted chauvinism (and the career
needs) of the officiers soundanais which often drove on, by methods
which destroyed any rational prospects of economic gain, the
French advances to which Chamberlain reacted so violently.
Only in southern Africa were economic motives arguably
preponderant during the later phase of the scramble. Between
1870 and 1900 British capital export to South Africa had risen
from virtually nil to some £200 million, about as much as to
Canada;41 and the great bulk of this increase had taken place since
1886, after the discovery of the Rand goldfield. Nevertheless,
more was at stake in South Africa than the profits of the Rand
capitalists and their metropolitan associates. A rich enough and
strong enough Transvaal, especially if supported by Germany or
some other major power, could have challenged British hegemony
throughout the sub-continent, including the British position at
the Cape; and the crucial strategic importance of the Cape had
40
F . O . 2 7 / 5 5 7 3 , Chamberlain, m i n u t e o f n N o v . 1897 o n the Niger C o m m i s s i o n e r s '
report of 10 Nov.
41
153

not disappeared merely because powerful financial interests were

now involved in South Africa. Lord Salisbury, who detested the
prospect of fighting for the Rand capitalists - 'people whom we
despise' — was still prepared to face war in order to ' make the
Boers feel that we are 8c must be the paramount power in S.
Africa'. 'The real point to be made good', he told the under-
secretary for the colonies, Lord Selborne, in July 1899, 'is that
we not the Dutch are Boss.'42
THE PARTITION AND THE IDEOLOGY OF IMPERIALISM
Acquisitive imperialism as an ideology seems not to have been

very widespread before the 1890s. Except in Germany, where a
vociferous colonialist movement with very strong ' Establish-
ment' support was already by the early 1880s influencing public
opinion, its dissemination appears to have been a result rather than
a cause of the scramble. Imperialism was not however (as has
sometimes been suggested) the specific ideology of the so-called
' nationalist masses'. The ' masses' are presumably urban and rural
manual workers; and the attitude of these groups to colonial
expansion, so far as it is known, was normally a profound
indifference - except of course for their socialist minority, which
was hostile on principle. Imperialism was essentially the creed of
ruling elites and of their professional and white-collar auxiliaries;
and not always even of a majority of these. In Britain and
Germany, fervent nationalism and colonial imperialism very often
coincided; but in France many militant nationalists saw in colonial
expansion at best an undesirable distraction from the recovery of
the Lost Provinces, and at worst a Bismarckian plot to weaken
France in Europe. For this and other reasons, convinced
colonialists were never more than a small minority of the French
bourgeoisie. They were however a very active and influential
minority. They included politicians (often leading politicians) of
all parties except those of the extreme Right and Left; and in
Parliament these politicians operated as a permanently organised
42
Salisbury to Lansdowne, 30 Aug. 1899: cited Lord Newton, Lord Lansdowne, A
biography (London, 1929), 157. B. M. Add. Ms 48675, Sir E. Hamilton's Diary, 6 Oct.
1899; Milner MSS, 16, 1, Selborne to Milner, 27 July 1899: cited A. N. Porter, 'Lord
Salisbury, Mr Chamberlain and South Africa, 1895-9', J. Imp. Oaealth Hist., 1972, 1,
1, 3—26.
154

PARTITION AND THE IDEOLOGY OF IMPERIALISM
groupe colonial. Colonialism was strong among naval and military

officers; and among senior public officials. It was backed by
important financial and industrial interests which, especially from
about 1890, organised 'lobbies' and disseminated propaganda
though such organisations as the Comite de I''Afrique fran$aise. In
spite of their comparatively small numbers, French colonialists
were therefore often very successful in influencing the course of
policy.
In England, the home of Gladstonian principles of international
comity and peaceful economic competition, the collapse of
informal empire was soon followed by something like the mass
conversion of the dominant elites from these principles to an
aggressively rapacious imperialism with strong social-Darwinist
overtones. By the late 1890s the Liberal Party had rejected the
leadership of both Sir William Harcourt and John Morley, who
continued to oppose colonial imperialism; and its official policy
was hardly less imperialist than that of the Tories. This ideological
revolution was not however confined to party politicians. It swept
the court, the aristocracy and 'smart society', the 'city', the
universities and the great learned societies hardly less powerfully
than the armed services, and the manufacturers and merchants
represented in the chambers of commerce of the great provincial
towns. Its triumphal progress can be followed in the mounting
enthusiasm, culminating in extravagant adulation, with which all
branches of the Establishment lionised Stanley in 1890, Lugard
in 1892—3, Rhodes in 1894-5 and Kitchener in 1898-9.
By the later 1890s emotional commitment to imperial expansion
sometimes approached the hysterical in Britain and France alike.
After the battle of Omdurman, the British behaved as if they had
won a major war, and rewarded Kitchener accordingly. During
the Fashoda crisis, many Frenchmen behaved almost as if the
national existence of France itself was at stake. Partition had begun
as a diplomatic 'game', in which the skilful player won without
even a hinted threat of force against Huropean adversaries. By the
later 1890s some of the issues had become too important to be
contained within this convention; and at Fashoda in 1898 French
policy was thrown into disarray when the British bluntly refused
to observe the conventional rules of the game. As a French
publicist ruefully confessed, ' Jamais on ne previt qu'on put etre
'55

amene a quitter le terrain diplomatique.'43 This radical change of

attitude was doubtless partly a feed-back from the melodramatic
' steeplechases' and ' incidents' which marked the final years of the
scramble, and of which the encounter of Marchand and Kitchener
at Fashoda is the perfect type. But the influence of the then very
fashionable 'scientific' Social-Darwinist approach to politics
should not be overlooked. If, as Social Darwinists asserted, the
struggle itself is the key mechanism of evolution, the sole source
of selection and renewal, 'the fiery crucible out of which comes
the finer metal ',44 then the material value of the actual bone of
contention, even though it be as meatless as Chamberlain's West
African villages, becomes an almost irrelevant consideration. To
shrink from any struggle generated by international competition
is publicly to advertise irredeemable national decadence.
For a Social Darwinist, the most convincing possible demon-
stration of his own racial superiority is his ability to crush ' inferior
races' by armed force. Although force against Africans had
certainly not been spared, especially in southern Africa and by the
officiers soudanais in West Africa, in the early years of partition, its
large-scale use in operations planned by metropolitan policy-
makers is mainly a phenomenon of the 1890s. At the outset of
partition many Africans, quite unaware of the implications of
European control, had greeted with indifference the beginnings
of formal empire; or had accepted it hoping 'to deflect the costs
of colonialism on to their fellows and appropriate its benefits as
their own'. 45 Some African rulers and groups had indeed
positively welcomed European ' protection' as a safeguard against
foreign enemies, internal dissidents or the disruptive activities of
unofficial Europeans. By the 1890s many of these hopes and
calculations had been bitterly disappointed; and the negative
effects of colonial rule, upon elites and masses alike, had often
become obvious. There was a marked hardening of African
opposition to the extension and consolidation of European
control. This resistance now seemed intolerable, not merely to
local European interest-groups such as white settlers or officiers
43
J. D a r c y , France et Angleterre: cent armies de rivalite colonial/ (Paris, 1903), 389.
44
K. Pearson, National life from the standpoint of Science ( L o n d o n , 1900), 2 6 - 7 : cited
Langcr, Diplomacy of imperialism, 88.
45
J. M . L o n s d a l e , ' T h e politics o f c o n q u e s t : T h e British in western Kenya,
1894-1908', Hist. J., 1977, 20, 4, 841-70, at 870.
I56

PARTITION AND THE IDEOLOGY OF IMPERIALISM
soudanais, but to 'the chancelleries of Europe'. Especially against

the background of social-Darwinist ideology, success in imperial
expansion was now seen as an important, even a crucial test of
a power's general effectiveness and 'virility'. African resistance
was no longer a merely local problem: it had become an issue in
which 'prestige' was at stake.
Metropolitan policy-makers were now often prepared to
override the financial constraints which had previously restricted,
sometimes very severely, the scale of European military action in
Africa; and to crush African resistance by a ruthlessly systematic
exploitation of the technological gap between European and
African weaponry and military organisation. The French success-
fully sought this solution in Dahomey in 1892, and again in
Madagascar in 1895-6. So did the British in the Sudan, especially
after January 1898 when the continued existence of the Mahdist
state became politically intolerable to London. The Italians failed
disastrously in its attempted application to Ethiopia in 1895-6.
The scale of these operations deserves emphasis. Kitchener led
over 20,000 men to victory at Omdurman, Baratieri about 17,000
to defeat at Adowa. By the later 1890s expeditions of two to three
thousand men were commonplace. Even these smaller expeditions
represent a far greater deployment of force against Africans than
was usual in the early years of partition: in 1882-3 Borgnis-
Desbordes had been trying to conquer the upper Niger basin with
about 500 men.46
At the outset of the scramble the powers had staked out their
claims in the hope that African territories could somehow be made
to yield a profit with the barest minimum of administrative and
military effort. The colonial projects of both Bismarck and
Leopold II were in their initial stages designed to avoid direct
administrative burdens; they were in effect adaptations of certain
features of British informal empire. Even in the Niger Sudan,
some at least of the French policy-makers who sought formal
empire did not originally contemplate large-scale campaigns of
military conquest. But European racial arrogance and competitive
national self-assertion soon responded to African resistance by
policies of subjugation through the massive deployment of armed
force. The motives for these major campaigns of subjugation
46
E. Requin, Arcbinard et le Soudan (Paris, 1946), 13-3*-
'57

differed from region to region and were often multiple. Outside

southern Africa, economic motives were rarely preponderant: in
the upper Nile basin, they were negligible. But one motive was
always present, and was always very important: the desire to
demonstrate, in the most crushing possible way, the ' superiority'
of the European.
158

CHAPTER 3
NORTH AFRICA
When the Second Empire succumbed to the Prussian onslaught

the French had already been established in North Africa for forty
years. Even before the conquest of Algeria had been completed,
they had advanced into the Sahara, and occupied the great
northern oases. Morocco had managed to preserve its independ-
ence by a policy of systematic non-involvement in the outside
world, but the two regencies of Tunis and Tripoli, which had
hitherto been vassals of the Porte, became the objects of French
expansionism. Reconquered by Turkey in 1835, Tripoli had
become an Ottoman province. The bey of Tunis had escaped the
same fate only by placing himself under French protection. As for
the heart of the great desert, a no man's land inhabited by Arab
and Tuareg nomads, it was in practice still free from the rival
influences of French, Turks and Moroccans.
Ten years later, in 1881, France established a protectorate over
the Regency of Tunis. The momentum of European penetration
increased as the twentieth century approached. After the conquest
of the Algerian Sahara, the French decided to establish a
protectorate over Morocco, while the Italians cast their mantle
over Tripolitania. The 1904 agreements between France, England
and Spain appeared to seal the fate of the Sharifian empire. The
intervention of Germany and the start of another international
crisis merely delayed by a few years a French conquest that had
become inevitable.
ALGERIA
The fall of the Second Empire was greeted with joy by the French
in Algeria. Ever since Napoleon III had uttered the words: 'Arab
kingdom', they had gone into opposition. They had voted ' N o '
in the 1870 plebiscite, and had not waited for the September 4th
revolution to proclaim themselves republicans. The event in Paris
provided the occasion for several weeks of unrest. Officials
M9

NORTH AFRICA
appointed under the empire had lost all authority. Committees
that were no more than talking shops, such as the one run by the
lawyer, Robert Vuillermoz, aimed to take over the administra-
tion. But the French Prime Minister, Adolphe Thiers, called a halt
by appointing an energetic governor, Admiral de Gueydon, who
restored order by putting an end to such committees of chatterers.
The time for speeches was indeed over, for the government was
confronted by an insurrection affecting a third of Algeria. This
should have been foreseen, especially in the most recently pacified
areas. The establishment of a civilian regime had angered the
Muslims. The departure of the French garrison, followed by their
defeat in Europe, provided an unexpected chance of revenge. For
the natives, the establishment of a civilian form of government
meant domination by the French settlers. The Arab chiefs feared
the loss of their authority and privileges. They also feared land
confiscations and tutelage by European municipal councils. The
qadls {cadis) knew that sooner or later they would be replaced by
French judges. The rumour spread that the French were going
to force Muslims to convert.
The first measures taken by the Government of National
Defence did nothing to dispel these apprehensions. The natural-
isation of Algerian-born Jews, the setting up of assize juries and
the extension of the area under civilian rule were proclaimed one
after another. The news of the French defeats had spread rapidly.
The emperor was taken prisoner, and France no longer had a
leader. As a final humiliation, France was now governed by a
Jew.1 A letter from Mahyl al-Dln (Mahieddine), a son of 'Abd
al-Qadir (el-Kader), called on the faithful to rebel.
On 14 March 1871 the signal for the rising was given by
Muhammad al-Muqranl (el Mokrani), the bachaga of the Medjana,
followed shortly by the shaykh of the Rahmaniyya brotherhood,
who proclaimed a holy war. Kabylia obeyed his call, as did most
of the tribes of the Constantinois. But the insurrection remained
localised, as the tribes of the Oranie and the Algerois took no
part in the movement. The rebels attacked European farms and
settlements, which were devastated, but they proved incapable of
organising concerted operations. The arrival of reinforcements
enabled the French troops to regain the initiative. The Rahmaniyya
1
This was Cremieux, a member of the Government of National Defence, who had
carried through the naturalisation of his fellow Jews.
I-6o

ALGERIA
chiefs were quickly subdued. Al-MuqranI was killed in battle.

There was sporadic fighting for a further seven months, but the
struggle was finally brought to an end by raids upon the Saharan
frontier regions to which the diehards had fled.
This was followed by severe repressive measures. Fines totalling
more than 10 million francs were imposed on the rebel tribes and
446,000 hectares of land confiscated, some of which was used to
set up fresh areas for French colonisation. Some of the tribes were
ruined; it took others twenty years to recover. At least, the aim
of the settlers had been achieved: apart from a few sporadic
outbursts around 1880, there were no more armed rebellions of
any importance, and Algeria could be considered to be finally
pacified.
The defeat of the rebels ensured the political victory of the
settlers. The French army, held responsible for the insurrection,
was gradually removed from positions of power. Now that they
were all-powerful, the French people of Algeria soon tried to
make themselves independent of the Governor-General. General
Chanzy (1873—9), who succeeded Admiral de Gueydon, had been
an officer in the bureaux arabes. He favoured a policy of immigration
and the settlement of the countryside, sponsoring the creation of
new villages. He extended the territory under civilian rule at the
expense of the area under military control, and welcomed the
increased number of communes de plein exercice (responsible local
authorities), administered as in France by elected mayors and
councillors. However he resisted some of the demands of the
settlers: in the areas where the majority of the population was
native-born, he extended the institution of communes mixtes,
districts governed by civilian administrators appointed by the
government.
The first civilian governor of Algeria was Albert Grevy
(1879-81), the brother of the president of the French Republic.
Unfamiliar with the affairs of the country, he allowed himself to
be dominated by the Algerian members of the French National
Assembly. He satisfied their demands, by doubling the area under
civilian rule to almost 105,000 square kilometres. In 1881, these
lands were divided into 196 communes de plein exercice1- and 77
communes mixtes. The area under military control, now restricted
almost entirely to the near-desert regions of the high plateaux and
the fringes of the Sahara, included a mere half-million inhabitants.
2
261 in 1900.
161

NORTH AFRICA
These Algerian communes were strange administrative entities.
Although some were comparable to a small European town,
others extended over tens of thousands of hectares. Indeed, whole
sections of tribes had often been attached to communes de plein
exercice in order to increase their resources. The French mayors,
who, totally unsupervised, levied taxes on the native population,
used their budget entirely to profit the European community. In
Jules Ferry's phrase, this amounted to the 'blatant exploitation
of the natives'. 3 The proportion of Muslim municipal councillors
was reduced by a decree of April 1884 from a third to a quarter.
Moreover these native delegates could no longer participate in the
election of the mayor, who had to be a French citizen. But the
Algerian members of the National Assembly wanted still more.
They demanded administrative assimilation into metropolitan
France, the same rights as Frenchmen in France and, as a corollary
to this, the subordination of the Governor-General to themselves.
Assimilation was achieved with the agreement of Grevy by the
decree of 26 August 1881, known as the 'decret des rattachements',
which placed the various Algerian services under the direct
authority of the relevant ministers.
The role of the Governor-General was henceforth reduced to
that of messenger; he became, in the strong words of Jules Ferry,
'an inspector of colonisation in the palace of a rot faineant
[do-nothing] king'. 4 Henceforth, only Algerian members of the
National Assembly, whose numbers were doubled, had the right
to deal directly with the civil servants in Paris, who knew nothing
of local conditions. Influential deputies such as Gaston Thomson
or Eugene Etienne became the real masters of the country under
such a mediocre governor as Louis Tirman (1881—91). Despite
the criticisms which were soon levelled at the system, it remained
in force for more than fifteen years.
After the fall of the Second Empire, the French Government once
again adopted a policy of officially supporting European
settlement and the state made free grants of land as it had done
during the governorships of General Bugeaud and General
Randon. One hundred thousand hectares had been promised to
Alsatians who settled in Algeria. But those who emigrated were
3
'C'etait l'exploitation dc l'indigene a ciel ouvert.'
4
'Un inspecteur de la colonisation dans le palais d'un roi faineant.'
162

ALGERIA
largely manual workers, and out of the 1,18} families settled by

the government, two-thirds abandoned their grants. They were
more successful with the sons of Algerian settlers, and with the
peasants of the Massif Central, Dauphine and the Provencal Alps.
In total, by 1882, four thousand families had arrived from France,
free grants of land totally 347,000 hectares had been made, and
197 new settlements founded by the government. Unofficial
colonisation also made progress, thanks to the Warnier law,
passed in 1873, the provisions of which encouraged the splitting
up of lands held collectively by the tribes. Thus the Europeans
could acquire common lands that had little value in the eyes of
the natives. However this law gave rise to such abuses that in 1890
the government was forced to suspend its application.
The official settlement policy was not without its miscalcula-
tions. The National Assembly attacked it on the grounds of
expense. It was also the more reluctant to vote the necessary funds,
in that Algeria was thriving because of the ruin of the French
vineyards, devastated by phylloxera. In December 1881 the
Chamber of Deputies refused a request for funds totalling 50
million francs required to found more settlements. Thenceforth
official support for settlement gave way to private initiative;
Spanish immigration replaced French settlers. The grain harvests
had long been a disappointment. Methods used in metropolitan
France were not readily transferred to a country where the rainfall
was inadequate and irregular. The periodic recurrence of droughts
and plagues of locusts discouraged the settlers. It was only after
1890, with increased mechanisation and the introduction of dry
farming, that wheat growing became profitable. From then on
progress was rapid thanks to the growth in yields. Vineyards
benefited from the phylloxera epidemic in France, and from the
breaking off of diplomatic relations with Italy, which closed the
French market to Italian wines. Therefore Algeria was ener-
getically launched into the establishment of wine-producing
industry to replace the loss of French production. The area
planted with vines rose from 15,000 hectares in 1878 to 110,000
in 1890 and 167,000 in 1903, with wine production at that time
averaging five million hectolitres.
At the beginning of the twentieth century, the prosperity of
settler agriculture in Algeria showed that the Republic's gamble
seemed to have paid off; a class of French peasants, comprising
163

r"" sWw
1
r / T 10 °f
_^-^Bizerta
yS— r * ?
Algier , Oeilvs
J ^- j Tunis IT >feO
CABYLIA »' • . I . • • • -y -\.
/ ConstantineijJ
Tangier^ Mostaganem/THI1 -^vOrlean?«iK •"+ D
•^Set.f /
\ Oraaj)P»H?lCXr (
\\Sousse
!f \
_— — -. '
,/JrSnii bcl AbDi;s
\
J Tebessa* /
)
Kairouan»-t- J< ^N.
) 35°N
) ; Jf \( Biskra 1
Tlemcen I" Diella }
\
/ . . - •
"'"lie r
V / Laghouat
1
I /
Tuggurt«
s"\
' L''
\ Algiers • fTunis
• Ghar daia \
J
Tangier*
.Fez
/TUN'IS
L °ALG E R 1 A 1 ^
Wa,9,a# V, j > - ^
Wargla
\ s
( j Colomb Becha
V N
V
—^^
o .— (•Ghadames
>• T R 1 POL 1
/Ghadames 30°N
{ Cape Juby In Salah • \
y r
'•/ma''
y 1. ~Z Cereals MM Vines i H Olives
KDE ,' N t H ^ Market Gardening
E 3 Cork oak I Iron
ORO / AHAGGAR N
i' " P Phosphates Railways
i S A H A R A 0 E S £ R T
Department boundaries
Taudeni T BEST1
i 0 t 300 km
BORKU 6 200 miles
7 Algeria at the beginning of the twentieth century

ALGERIA
more than a third of the total population, was becoming established
in Algeria. At the same time, the towns were increasing rapidly
in size, including Oran and, in particular, Algiers, which was
taking on the appearance of a large provincial capital.
Owing to immigration and high birth rate, the European
population doubled in twenty years, from 245,000 in 1872 to
almost half a million in 1891. But foreigners - Spaniards, Italians
and Maltese - were more numerous than people of French
extraction.5 The ' foreign peril' was neutralised by the law of 1889,
which automatically naturalised the children of foreigners who
were born in Algeria and who did not formally reject French
citizenship. Military service and mixed marriages ensured the
gradual assimilation of those who had been naturalised in with
those of French extraction. In 1901, the European community
numbered 5 80,000 (638,000 including the Jews), making up more
than 14 per cent of the population of the three departments.
At the end of the Second Empire the Muslim population of

Algeria had decreased as a result of famine and epidemics; but
owning to a high birth rate it increased rapidly during the closing
years of the century. From 2,800,000 in 1881, the numbers rose
to 3,5 00,000 ten years later, and more than 4,000,000 in 1901, with
3,841,000 of these in the three northern departments. This
5
The population of Algeria according to the census of 1901:
Europeans Jews Muslim Total
Algeria (1902 frontiers)

3 departments 580,590' 57.538* 3,841,29s 3 4,479,426
(.2.9%) (1.28%) (85.7%)
Territoires du Sud 3.254 '.423 255.228 2 59.9O5
Major towns: Algiers 94,064 "'.356 30,082 '35.5O2
(69.4%) (8.4%) (22.2%)
Oran 64,839 11,187 >3."7 89. 2 5 3
(72.6%) (12.5%) (14.8%)
Constantine '7,9°9 7.252 23.750 48,911
(36.6%) (14.8%) (48.6%)
Bone 26,683 1,461 . 9.74O 37,884
( 7 o. 4 %) (3.8%) (2 5 7 % )
Tlemcen 5.595 5.201 24,672 5 5.468
(.5.8%) (14.6%) (69.5%)
Notes: 1 Including 1)4,693 Spaniards and 38,531 Italians. 2 Divided almost equally between
Frenchmen, foreigners and naturalised citizens. 3 Including Tunisians and Moroccans.
165

NORTH AFRICA
population was far from homoegeneous. There were, on the one

hand, the Kabyle who lived in the mountians and had remained
faithful to their Berber language and traditions, and on the other
hand the arabised tribes of the plains and the plateaux. The French
stressed the good qualities of the Kabyle, the 'Auvergnat' of
Africa, which provided a sharp contrast with the laziness of the
Arab. But though the hill people had preserved their individuality,
they were not willing to be assimilated by the French; attempts
at conversion failed, and a ' moral conquest' through the schools
was not seriously undertaken. By the end of the century, the
Kabyle mirage had begun to fade. In February j 895, the governor,
Jules Cambon, declared in the Chamber of Deputies that he did
not believe that the Kabyle was more easily assimilated than the
Arab.
This establishment of the new regime did nothing to improve
the status of the native population, for it merely transferred to
the civilian authorities the discretionary powers hitherto enjoyed
by the military officers. The Senatus Consultus of 14 July 1865
had guaranteed for the Muslims the continuation of their personal
status. It had also furnished them with the possibility of achieving
French citizenship if they were willing to renounce that personal
status. But few of them had been tempted to take advantage of
this provision; to abandon their religious status seemed apostasy,
and only 736 individuals requested naturalisation between 1865
and 1890.
A native code {code de /'' indigenai), established in 1881, gave the
authorities powers to impose fines or terms of imprisonment,
individually or collectively, without trial for a certain number of
offences. A permit was needed to travel in the interior of the
country. In the communes mixtes, the native population found itself
subject to the civil authorities, and in the communes de plein exercice
to the mayor, assisted in both cases by appointed officials, the qa'ids
[caids) or bachagas. Islamic justice retained its own organisation:
the qddis judged according to custom in all questions involving
personal status. But in cases concerning property (the decree of
September 1886) or criminal proceedings, Muslims were called
before French judges and tribunals.
For taxation purposes, the native population was still subject
to a regime dating from the time of the Ottomans, that of the
four 'Arab taxes'. The 'ushr {achour), the tithe on the harvest, had
166

ALGERIA
been converted to a fixed tax on the area sown to crops. The

hokor, only levied in the Constantinois, applied only to collective
tribal lands. The tâkdt was a tax on flocks, and the Id^tma, a
capitation tax in Grande Kabylie, took the form of a tax on palm
trees in the southern Constantinois. Though the system had a
long tradition behind it, it nevertheless weighed heavily upon the
poor. On average, an individual European paid five or six times
more tax than a Muslim, but when their means are taken into
account, Europeans were slightly less heavily taxed. The re-
organisation of the fiscal system, first proposed in 1912, was
carried out only in 1918.
The education legislation of the Third Republic remained for
all practical purposes a dead letter as far as the Muslims were
concerned. The 1883 and 1887 decrees, which extended the Ferry
laws to Algeria, did not in fact include any clause making them
obligatory for Muslims. In the towns, the Muslim population had
demonstrated so effectively against the plans of the government
that, in order to avoid further trouble, the authorities had not
insisted upon them. Since funds were in short supply, and the
Europeans on their side could scarcely be said to favour the
education of the natives, little progress was made.
In the whole of the country, the number of Muslim pupils in
primary schools in 1882 was 3,000, in 1890 11,000, and in 1900
24,000, as against 97,000 Europeans and 14,800 Jews. The
government's reply to its critics, for example Tirman in 1883, was
that one should not be surprised that there were so few Muslims
in French schools, but rather' that there should be as many as there
are, given the reservations of the parents'. 6 In fact, although
amongst boys the proportion was one Muslim to two Europeans,
in girls' schools the proportion was one to twenty-six. Only in
Kabylia was an experiment attempted to enforce attendance, with
eight schools established entirely at state expense. It was not
followed up, and Jules Ferry himself gathered from his travels in
1892 that the Muslims were opposed to compulsory education.
Three madrasas (medersas) provided an education based on the
study of French and Arabic. French secondary education attracted
only a few dozen Muslims each year. The Koranic schools
6
' Mais bien de ce qu'il y en eut autant, en raison des reticences manifestoes par les
parents': K. Vignes, Le gouverneur general Tirman et le sytteme des rattacbements (Paris,
'958), 149-
167

NORTH AFRICA
continued to survive on the margin of the official education

system. Algerian students who wished to acquire a traditional
Islamic training went to the great mosques of Tunis and Cairo.
The situation was scarcely better in the economic field. Native
agriculture made a poor showing alongside the modernised
methods of the European farmers. In the greater part of the
country, the Algerian fallah, with his fatalistic attachment to
tradition, continued to scratch the soil with primitive wooden
ploughshares. Too poor to build up reserves, he was at the mercy
of every crisis. Bad harvests and animal epidemics regularly
brought poverty to the countryside. To buy seed, the fallah often
had to borrow, thus subjecting himself to the Jewish money-
lenders, who pitilessly exploited his ignorance and credulity.
More than half the falldhin owned their own land, but most had
only a few small plots. In the department of Constantine just
before the war, three-quarters of native landowners possessed less
than 20 hectares. The large estates were subdivided into farms of
about 10 hectares, rented out to tenant farmers. The landless
peasants had to hire out their labour, either to European settlers
or to other falldhin. As the years passed, their numbers tended to
increase: from 12 per cent in 1900 to 18 per cent in 1930. The
increase in the Muslim population intensified the shortage of land
and the prevalence of rural poverty.
The Muslims cultivated barley and hard wheat, but their yields
were much lower than those of the European farmers-in an
average year five to six quintals per hectare for wheat in
comparison with eight to ten quintals in an average year for the
Europeans. When allowance is made for annual variations in the
harvests, cereal production remained stationary, as did that of
olive oil. Stock-rearing decreased slightly as a result of the
reduction in common land. In broad terms, there were in Algeria
about seven or eight million sheep, three to four million goats,
and about a million head of cattle. But these animals, under-
nourished and badly cared for by the natives, were not very
productive. Sometimes the herds would be halved by two
successive years of poor rains.
It did not take the French in Algeria long to discover the faults
of the system inaugurated in 1881. Algerian affairs, divided
amongst eight ministries, were settled often after interminable
168

ALGERIA
delays by officials totally ignorant of the country. Having no

budget and no revenues of its own, Algeria was obliged to beg
for funds that were grudgingly doled out. Having begun in
Algeria, the offensive against the system of rattachements (direct
rule by individual ministries in Paris) was soon being mounted
in metropolitan France itself.
In the Chamber of Deputies, in 1891 and 1892, Auguste
Burdeau and Charles Jonnart denounced the abuses of centralisa-
tion and the errors which had been made in administration and
settlement. In 1892 a Senate commission, led by Jules Ferry,
undertook a fact-finding mission to Algeria. Ferry's report, a
veritable indictment of the system of rattachements, concluded that
the governor's powers must be reinstated by centralising the
higher administration in Algiers.
In spite of the efforts of Tirman's successor, Jules Cambon
(1891—7), it took three more years for the reforms to be completed.
Finally, the French government, under Felix-Jules Meline, pro-
mulgated the decree of 31 December 1896, which brought the
policy of rattachements to an end. But the reform was not yet
complete. It did not satisfy the demands of the French in Algeria,
who hoped for the granting of new freedoms, the establishment
of their own budget controlled by a local assembly — this being
in their eyes the only way to bring about the modernisation and
development of the country. Now, neither Jonnart nor Ferry had
considered granting an elected assembly to Algeria, on the
grounds that the inhabitants were lacking in political maturity.
It was in these circumstances that there occurred the anti-Semitic
crisis of 1898.
The Jewish question had become politically important since the
Cremieux decree of 24 October 1870 had naturalised en bloc the
entire Jewish population. This measure was doubtless premature,
since the majority of Algerian Jews, crowded into the ghettos of
the large towns, were still very oriental. Despised by Europeans
and Muslims alike, the Jews were beginning to be envied for their
commercial success. Foreign immigrants, in particular the
Spaniards, were among the most hostile. From time to time, local
assemblies requested that the Cremieux decree be abrogated. But
there were ulterior political motives behind these public protests.
The fact was that the Jewish electors, voting en bloc on the
instructions of their consistories, held the balance between two
169

NORTH AFRICA
French groups, the opportunists on the one hand and radicals

on the other. The latter accounted for the success of their
opponents as being the result of their scandalous collusion with
important members of the Jewish community.
From 1895 onwards, anti-Jewish leagues were formed, and
their propaganda struck an answering chord amongst the rank and
file members of the European community. The radicals used this
movement as an electoral springboard. The crisis was triggered
off by student demonstrations in March 1897, and soon there were
riots in Algiers. A law student, Max Regis, took the lead,
producing a violent broadsheet, /' Antijuif. Between 20 and 25
January 1898, the mob took over the streets, looting Jewish shops
and forcing the governor-general to barricade himself in his
palace.
The French National Assembly elections of May 1898 were
taking on the appearance of a plebiscite on the Jewish question.
The anti-Semites led by Edouard Drumont took four of the six
Algerian seats. Only two opportunists, Etienne and Thomson,
managed to hold on to their fiefs at Oran and Bone. In order to
force the government to abrogate the Cremieux decree, the
Algerian deputies invoked the spectre of separatism. Meline
compromised by offering administrative concessions.
The decree of 23 August 1898, drawn up by the new governor,

the lawyer Edouard Laferriere, in effect set up in Algeria a new
assembly, the delegationsfinancieres.Subject to the ultimate control
of the National Assembly in Paris, this had the right to approve
a local budget. In this way, Algeria won financial autonomy,
almost a sort of local self-government, as a strange, not to say
unforeseen, result of the anti-Jewish riots. Thenceforth the shape
of the regime was fixed. The law of 19 December 1900 granting
a special budget to Algeria, and the separation of the Territoires
dii Sud in 1902, were merely finishing touches. The governor-
general had completely regained the whole of his former powers.
He had however to take into account the new assembly, whose
authority was superimposed on that of the conseils gene'raux. To
fulfil their role of representing the various economic interests, the
delegationsfinanciereswere divided into four sections. The French
colons (settlers) were entitled to twenty-four delegates, as were the
170

TUNISIA
'non-co/otis'. The Arabs and the Kabyles had respectively fifteen

and six representatives.
The system appeared to be a compromise, but it continued to
function satisfactorily until the outbreak of the First World War.
The administration, luckily, was led by capable individuals. After
Laferriere, there were Paul Revoil (1901-3), Charles Lutaud and
especially Jonnart, who governed the country from 1900 to 1901
and from 1903 to 1911, and who returned in 1918. Thanks to
these men, the authority of the office of governor-general was
respected, and wise management of the finances gave Algeria a
balanced budget. Stagnation had been replaced by prosperity.
Gradually society was pacified. The failure of the anti-Semitic
candidates in the 1902 elections demonstrated that the anti-Jewish
crisis was over. ' It was the triumph of the practical good sense
of country-dwellers over the agitation of the turbulent populace
of the large towns'. 7
TUNISIA
In the middle of the nineteenth century, the regency of Tunis was

a small principality, whose international status would be hard to
define. Since it was not completely outside the authority of the
Ottoman empire, it was halfway between autonomy and complete
independence; a very vague status which allowed the European
powers to treat the beys as their interests dictated, either as
independent princes or as vassals of the Porte. Since the sixteenth
century, the bonds which linked Tunisia to the Ottoman empire
had become looser and looser. After the expulsion of the
Spaniards in 15 74, the country had become a Turkish province
ruled by a pasha, supported by a corps of janissaries recruited in
the east. But the pasha had been rapidly stripped of his powers
by the dey, the commander of the militia, who in his turn was
replaced by a civilian official, the bey. The sultan had recognised
these changes by granting the bey the title of pasha and the right
to hand on his office to his heirs. But the first dynasty of beys,
the Muradids, failed to take root in the country. In 1705 the
commander of the militia, Husayn b. 'All, a soldier of Greek
extraction, seized power during the disturbances which at that
time were troubling the regency. He was recognised by the Porte,
7
C. Martin, Histoire de CAlgerie frattfaise, 18)0-1962 (Paris, 1963), 243.

NORTH AFRICA
and founded a dynasty that lasted for two and a half centuries.
Despite princely quarrels and assassinations, a rule of succession
by seniority of birth had emerged by the mid-eighteenth century.
Thus Muhammad Bey succeeded his cousin Ahmad in 1855, and
left the throne to his younger brother Muhammad al-Sadiq
(es-Sadok) (1859-82).
On the accession of Muhammad al-Sadiq in 1859, tne Husaynist
dynasty possessed both the strength built up over 150 years'
hereditary transmission of power and the traditions of independ-
ence which gave the beys the authority of sovereign princes. They
had complete autonomy; they legislated as they wished; they
governed with the help of councillors chosen by themselves. But
the beys had possessed even more extensive prerogatives that
provided them with the apparatus of sovereignty. They had their
army and their own navy, they minted coins, maintained diplomatic
relations, declared war and signed treaties. Although they had
neither legations nor consulates abroad, they could, in Tunis,
discuss political matters with the consuls of the major European
powers. These were indeed the attributes of sovereignty, empha-
sised even more by the existence of a Tunisian flag, recognised
by the Porte itself.
However, these hereditary princes continued to bear the titles
of pasha and bey, which ranked them with the governors of the
larger provinces of the Ottoman empire. At Tunis, the Friday
prayer was still said in the name of the sultan; at their accession,
the beys sent an embassy to Constantinople to inform the sultan
and obtain the firman, his confirmation of their investiture and the
mark of their dependence. Thus for many years no foreign power
had questioned the, at least nominal, suzerainty the sultan claimed
to exercise over the regency, and the beys were treated by the
European courts as vassal princes of the Porte.
Now, by a strange paradox, the conquest of Algeria caused a
rapprochement between France and Tunisia, and increased the
latter's separation from the Porte. The French expedition had been
welcomed by the regency, which rejoiced in the fall of a detested
rival. No doubt the extension of the conquest to the Constantinois
posed a delicate problem in regard to the western frontier of
Tunisia, but the danger from France was negligible compared
with the threat of an Ottoman reconquest. Indeed, in 1835 the
Porte had taken advantage of a civil war in Tripoli between rival

TUNISIA
princes of the Karamanli family to intervene in that Regency and

replace it by direct rule. Tunis feared the same fate as Tripoli, and
the bey realised that he would scarcely be in a better position than
his neighbour to resist the intervention of an Ottoman force. The
consciousness of his own weakness forced him to seek support
from the neighbour who seemed least dangerous in the immediate
future. He appealed to France, whose protection might discourage
the Turks from any attempt at offensive tactics. In the long run,
Tunisia became a principality in a semi-vassal relationship to
France, a buffer state protecting the eastern frontier of Algeria.
At this time, the sovereign was still Muhammad al-Sadiq, who
had acceded to the throne in 1859, a weak ruler depicted by a
contemporary in these terms:
The Bey has no intelligence, and he is addicted to the most shameful vices.
His hareem contains only small boys... of which he is not in the least ashamed.
He never concerns himself with affairs of state except when obliged to by the
consuls, and then merely parrots the words taught him in advance by his
minister. Twice a week he holds a court session. There, smoking his pipe, he
listens to the plaintiffs, and pronounces very brief judgements whispered to
him by one of his officials. Apart from these occasions, he remains shut up
in his hareem, where his life is nothing but one long orgy.
Politically, the only person of any account in the Bardo palace

was the prime minister, the all-powerful Mustafa Khaznadar, a
mamliik of Greek extraction, who had managed to remain in
power, under three beys, since 1837. The khaznadar, intelligent
and cunning, maintained at court a careful balance between France
and England, but his own sympathies were on the side of Great
Britain on account of his connections with Wood, the British
consul. At the palace, he alone exercised influence over the feeble
spirit of the bey.
The bey's powers were delegated to about sixty qaids, assisted
by khalifas and shaykhs, whose task was to keep order and collect
taxes. The palace dignitaries competed for the administration of
districts with sedentary populations, lands considered peaceful,
where the office provided a good living. On the other hand, they
were eager to hand over the government of the tribes to native
notables. Amongst the nomads, indeed, the qffid in particular had
become the representative of authority, a task that had grown
very difficult in the troubled circumstances of the regency at the
time.
173

NORTH AFRICA
The sedentary cultivators numbered about half the population,

half a million or so out of approximately 1,100,000 in the country
as a whole. Apart from the mountain dwellers, they were grouped
mainly on the plains and in the hills of eastern Tunisia, forming
an almost continuously populated strip from the northern coast
to the Gulf of Gabes. In the Tell, the falldhtn mainly grew wheat
and barley; those of Cape Bon cultivated fruit and olives as well
as cereals. The olive tree was the main source of wealth in the
Sahel, considered to be the regency's richest and most densely
populated area. In the south, from Gabes to the Djerid, the
peoples of the oases all depended on their date-palms.
The hard-working, resigned inhabitants bore the full weight of
beylical taxation. The northern jalldhin were mostly reduced to the
status of sharecroppers on the great estates of the dignitaries of
the bey's court circle. The sedentary inhabitants of the oases were
preyed upon by the nomads. On the other hand, the villagers of
the Sahel had for many years escaped relatively lightly. In 1864,
however, they took part in a large-scale insurrection against
increased taxation. After the revolt had been crushed, the Sahel,
pillaged by the military, was burdened with taxes and requisitions.
The falldhtn, at the end of their resources, had become the prey
of the money-lenders of Sousse.
The town dwellers were less downtrodden than the rural
population, but the towns suffered from the decline of the
traditional industries, which at both Tunis and Kairouan provided
a living for more than a third of the Muslim population. The craft
guilds could not compete with English and French manufactured
products. The checbia (headgear) industry, which had for many
years been the pride of the capital, was a mere shadow of its former
self. In the face of competition from fezzes manufactured in
Europe, exports from Tunis fell by 75 per cent in ten years.
The economic and financial crisis that adversely affected the
native artisans and shopkeepers, however, favoured the rise of a
Jewish colony oriented for the most part towards financial
dealing. Having only recently been emancipated, the Tunisian
Jews were striving to free themselves from the authority and the
taxation of the bey by obtaining some protection from abroad.
It is true that there were no more than 30,000 Jews in the regency,
but half of them lived in Tunis and its suburbs. The Europeans,
who were nearly all Maltese or Sicilian, numbered more than
174

TUNISIA
11,000 in 1870, and it was not without anxiety that the bey viewed
the development in his capital of foreign colonies over which he
had no jurisdiction.8
The interior was a zone of transhumance for the nomadic tribes,
in which the descendants of the Arabs who had invaded the
country from the eleventh century onwards mingled with those
of Berber origin who had been arabised to a greater or lesser
extent. All these tribes moved fairly regularly according to the
season. On the high plateau which stretched to the north of the
Tunisian Dorsal, the migrations were on a minor scale: small
tribes, fragmented by ancient rivalries, had only small areas at their
disposal. On the other hand, on the dry, bare steppe, migrations
were on a much larger scale.
The tribes of the High Tell were amongst the most mobile.
They were constantly quarrelling with each other, and their
reconciliations were only in order to mount a raid into Algerian
territory or to resist paying taxes. Their misdeeds kept the frontier
zone very insecure and periodically gave rise to difficulties with
the French in Algeria. The southern slopes of the Dorsal were
the domain of powerful tribes, the Majeur and the Frechich, who
were habitually at enmity with their neighbours the Zlass, who
camped in the region of Kairouan, and the Hammama, who
nomadised in the territory from the centre of Tunisia to the
borders of the Djerid.
Opportunities for strife among all these tribes were not
infrequent. Raids on herds and arguments about pasture kept alive
ancient enmities, which became keener in times of drought. From
alliances to coalitions, they were grouped into two soffs, Husaynists
against Pashists, two factions whose names perpetuated the
memory of the civil wars of the eighteenth century.9 The hostility
of the soffs dictated the political geography of the steppe, but this
hostility maintained a relative stability in the area, since the two
forces were of approximately equal strength.
The bey and his agents played on these rivalries to keep the
most turbulent tribes under control, normally supporting the
Husaynist faction. Although there was permanent insecurity on
8
Tunis at that time had less than 90,000 inhabitants, Kairouan about 15,000, Sousse
just over 8,000. In 1870, there were 15,000 Europeans in the Regency, of whom 7,000
were Maltese, nearly 7,000 Italian, 800 French and 300 Greeks.
0
The struggle between the bey Husayn, the founder of the dynasty, and his nephew,
'All Pasha.
175

NORTH AFRICA
the steppe, at least revolts were infrequent, and usually localised.

In Tunisia, there was no dissident area {bildd al-siba) as there was
in the mountains of Morocco. The only ungovernable cantons
were in the mountain massifs of the Tell, where several Berber
confederations, such as the Moghod and the Khrumirs, lived
completely independently, ruled by their elected councils.
In the spring of 1864, the bey had been faced with a general
insurrection, in which nomads and sedentaries alike arose against
the doubling of the majbd, poll-tax. The fate of the dynasty had
for the moment hung in the balance, but the rebels had not dared
to march on the capital. Making use of the traditional enmity
between Husaynists and Pashists, the kha^nadar had managed to
divide the movement, which had never been very closely knit. The
submission of the Zlass and Hammama gave him his chance to
crush the insurrection in the early autumn.
With the constitution suspended, Mustafa had returned to
traditional methods of government, but the authority of the bey
had been greatly undermined by the crisis. Impoverished by the
revolt and the subsequent repressive measures, the country
groaned beneath the expedients of the prime minister. Bankruptcy
was followed in 1867 by the results of a catastrophic harvest:
famine and epidemics, with their train of disorder and destitution.
After the ravages of cholera, the year 1868 opened with a typhus
epidemic which carried off thousands of people.
The kha^nadar, at bay, kept his office only by repeated con-
cessions, finding means of dividing enemies whom he was
unable to confront. In the interior, the Zlass and the Hammama
took advantage of the toleration of the Bardo, the beylical
government, to bring to a triumphant conclusion their old
struggles with their enemies. In the capital, Mustafa negotiated
with the creditors of the state, trying to play off the Jews against
the Europeans, the Italians and the English against the French.
He managed to gain two years' grace by his skilful delaying
tactics. But confronted by the trenchant attitude of the French
government, supported by England, he had to resign himself to
the imposition of international supervision of the finances,
underlining the bankruptcy of the policies which he had been
pursuing for a decade.
Although the kba^nadar was mainly responsible for the crisis,
the financial difficulties of the regency could not be laid entirely
176

TUNISIA
at his door. Since the middle of the century, the country had
suffered from the effects of European competition, with which it
had difficulty in coping. The steady increase in imports from
France and England, not paralleled by a similar growth in exports,
had necessarily been the cause of a progressive worsening in the
balance of payments and a fall in the purchasing power of the
currency. The first difficulties had made themselves felt when,
following Egypt's example, Ahmad Bey wanted to equip his army
in the European fashion. The upkeep of 10,000 regular soldiers
was all the more costly for the bey's treasury, in that all the orders
were the object of shameless acts of embezzlement.
The resources of the state were indeed strictly limited: on
average it had an annual revenue of 17 to 18 million piastres, the
equivalent of 11 million francs. Moreover, this revenue varied
from year to year according to the state of the harvest and the
degree of success in gathering taxes from the tribes. The principal
taxes were the 'ushr {acbour), a tithe in kind on cereal production,
the qdniin, on olive trees and date palms, and the majbd, a qualified
poll-tax; there were also various other indirect duties and taxes.
It is true that the state had few needs. It left the few services,
such as looking after the poor and education, in the hands of the
pious foundations, the habus {habous). As for roads and bridges,
there were virtually none. The taxes were used for the upkeep of
the court and the beylical army. The financial administration was
periodically thrown into confusion by the flight of some high
official, fleeing to Europe with his ill-gotten gains. In 1864, it was
the Jew Nessim Samama, the director of finance, who left for
France with nearly 20 million piastres. Nine years later, it was the
turn of his nephew and successor, who found refuge in Corfu.
The prime minister set the tone; he was tolerant of any malversa-
tion as long as he was the first to profit from it. However, on the
death of Muhammad Bey in 1859, the national debt did not
amount to more than 19 million piastres.
All-powerful during the reign of Muhammad al-Sadiq, Mustafa
had used this power to play the role of a reforming minister. With
the overt aim of modernising the country, he encouraged a policy
of public works. Telegraphs were installed, the aqueduct of
Carthage was restored. Many orders were placed for equipment
to modernise the ill-furnished army. The national debt had risen
to about 28 million francs at the beginning of 1862, and the
177

NORTH AFRICA
kha^nadar had the idea of raising a loan in Europe. Sharing half

the commission with Erlanger, a swindler from Frankfurt recently
settled in Paris, in 1863 he floated a loan of 35 million francs on
the Paris Bourse, and then another in 1865. On the pretext of
paying for government purchases, the money of the Parisian
subscribers was soon dissipated in brokerage fees in greasing the
palms of intermediaries. Ships, guns, uniforms - all reject stock
invoiced at the highest price - uselessly cluttered up the bey's
arsenals. But no order was so famous as that for guns 'rifled on
the outside', which caused so much merriment among
contemporaries.
When the amount produced by the two loans had been
squandered, they naturally thought of raising a third loan. In an
effort to repay the local debt, immeasurably swollen by innumer-
able teskeres (treasury bonds) that the kba^nadar issued daily to pay
for current expenditure, they decided on a large-scale operation,
a loan of 100 million francs which was put in hand by Erlanger
in May 1867. But the French investors held aloof from the new
operation. The result was bankruptcy.
The intervention of the French government, followed soon
after by Great Britain and Italy, resulted in a compromise between
the three powers, which imposed a tripartite control of the bey's
finances. A nine-member international commission was set up by
the beylical decree of 5 July 1869. Its chairman was General Khayr
al-Dln (Kheredine), a Circassian mamluk, the son-in-law of the
khavyiadar, but on bad terms with him. The driving spirit was the
French Treasury inspector, Victor Villet, who held the position
of vice-chairman. Thanks to the good understanding between
Villet and Khayr al-Dln, the Financial Commission was able to
draw up the balance sheet of the Tunisian bankruptcy. The debt
had been estimated at 160 million francs, but after all the evidence
had been assembled, this was reduced to 125 million francs, and
the interest was cut back to 5 per cent per annum. In return, the
bey had to yield half his revenue to his creditors until the debt
had been completely paid off.
The financial control in effect established a three-fold protec-
torate over the regency, but the presence of a French official in
a key position in the commission underlined the privileged
position of France. The personality of the man who came to be
called Villet Bey, and his friendly relations with Khayr al-Dln, did
178

TUNISIA
still more. Thanks to his exceptional authority, France after 1870

kept her credit at the court of Tunis, which, following her defeat
in Europe, would otherwise have been impossible.
The war of 1870 dealt a very severe blow to French influence

throughout North Africa. Italy, a new power with numerous
settlers in the Regency, attempted to take advantage of the war
to substitute her own influence for that of her neighbour, but the
clumsiness of the Italian representatives played into the hands of
the English, represented by Richard Wood, who had been in
Tunis for nearly fifteen years.
Helped by his good relations with the kha^nadar, Wood tried
to ensure that the policy that he had always advocated prevailed
at the Bardo. In order to put a stop to French ambitions, the bey
was to re-establish his former relationship with the Ottoman
empire and to recognise the sultan as his overlord, provided that
the traditional liberties of the regency were recognised. A mission
led by Khayr al-Dln was sent to Constantinople, and in November
1871 it brought back afirmanwhich expressed in legal terms the
connection between Tunisia and Turkey. Of its own volition, the
Regency of Tunis reverted to the status of a vassal province of
the Ottoman empire, but a province whose autonomy was
formally guaranteed by the firman.
In the interior, Wood tried to increase his country's influence
by attracting British companies to the Regency. In 1872, a British
company commenced a railway line to Goletta, soon known as
the TGM (Tunis-Goulette-Marsa). It also obtained an option for
the construction of railway lines in the interior. At the same time,
a British bank established itself in Tunis. Unfortunately for Wood,
none of these undertakings was successful. The bank closed its
doors after three years; a gas company went bankrupt. As for the
TGM, it contrived with difficulty to run a railway which was
always in deficit. The fall of the kha^nadar from power in 1873
struck a further blow at the influence that Wood had for so long
exercised at the court of the Bardo.
For several years, Mustafa ibn Isma'Il, whom the kha^nadar had
not succeeded in ousting, had exercised an increasing influence
over the mind of the senile ruler. In order to get rid of the prime
minister, whose corruption was hindering financial recovery,
Villet and Khayr al-Dln set the favourite against him. Having lost

NORTH AFRICA
power, the kha^nadar was imprisoned and his goods confiscated.

He died five years later in 1878.
His successor was none other than General Khayr al-Dln, who
governed the Regency for four years from October 1873 to July
1877. Owing to his tactful and restorative administration, helped
by the return of good years for agriculture, Tunisia experienced
a relatively prosperous period. Khayr al-Dln was anxious to
pursue a policy of the 'middle way'. He considered that Tunisian
autonomy was guaranteed by the firman of 1871: 'the vassalage
of Tunisia', he said, 'is a guarantee of its independence. As long
as the Turkish empire endures, the tributary or vassal states will
have nothing to fear.' In the interior his policy was one of balance:
'The English already have concessions and privileges; let them
now be given to the French and the Italians.' Thus, in 1877, he
granted a French company the concession for the Medjerda line,
a railway line between Tunis and the Algerian frontier. This was
a result of the efforts of Theodore Roustan, who had been consul
in Tunis since the end of 1874.
Khayr al-Dln was not popular with the bey, and his policies
gave rise to many criticisms. Overthrown in his turn by an
intrigue of the favourite, he resigned in July 1877, and retired to
Constantinople. The new prime minister, Mustafa ibn Isma'II,
revived the wasteful and corrupt policies which were the hallmark
of the Bardo court.
This situation could have lasted even longer, had not the fate
of the regency been decided behind the scenes at the Congress of
Berlin in 1877, which brought to a close the three-year-old Balkan
crisis, which had begun in 1885. The British, who had secretly
arranged for Cyprus to be ceded to them by the sultan, proposed
some compensation to the French, so that the balance of power
in the Mediterranean could be preserved. ' Prenez Tunis si vous
voulez', Salisbury suggested to William Waddington, the French
foreign minister, 'Do what you like there. You will be obliged
to take it, you cannot leave Carthage in the hands of the
barbarians.' These words were soon confirmed by Disraeli, and
Bismarck shortly afterwards gave Germany's support.
The British were certainly hoping to kill two birds with one
stone, for they hoped that this initiative would ensure France's
goodwill in an Egyptian settlement that they had in mind. As for
Bismarck, his main aim was to make the French forget the loss
180

TUNISIA
of Alsace-Lorraine. On the other hand, a few months earlier Italy
had failed to take advantage of her opportunity by refusing to
make common cause with Great Britain in opposing Russian
ambitions in the Balkans. In this way, Italy had deeply disappointed
Disraeli and helped to bring about a rapprochement between
Britain and France.
France had thus been given carte blanche in Tunisia. But it took
her nearly three years to make up her mind to act, three years of
hesitation and half-measures before launching herself almost
reluctantly into the expedition which obliged the bey to accept
the Protectorate treaty. Public opinion in France at that time
tended to be hostile to colonial enterprises. People could see no
reason for once more resuming the overseas expansionist policies
of the Second Empire. There were neither commercial nor
demographic reasons for such a course. Moreover, the republican
regime was not yet firmly established, so that there was little
encouragement to embark on bold enterprises. Even the most
ardent hoped to avoid the expense of a military expedition by
establishing a protectorate over Tunisia by peaceful means. On
two separate occasions Waddington presented proposals to the
bey, but they were rejected. In December 1879, Waddington was
replaced at the Quai d'Orsay by Charles de Freycinet. These
prevarications took up months and even years, while rivalry with
Italy was becoming more and more acute.
At Tunis, the French and Italian consuls, Roustan and Maccio,
were at daggers drawn over the pursuit of economic concessions.
The Italians were helped by the sheer numbers of their settlers;
the French had on their side money and the support of Mustafa
ibn Isma'Tl, now the power behind the throne at the Bardo. In
July 1880, the Italians seemed to win a point by buying out the
British TGM company. But a month later, thanks to Mustafa,
Roustan obtained from the bey the right for a French company
to construct a harbour at Tunis, and also the concession for an
entire railway network in the regency. A French bank, the Societe
Marseillaise de Credit, bought up the property of General Khayr
al-Dln, and thus became the owner of the domain of Enfida, a
province of 96,000 hectares between Sousse and Tunis.
The French policy of economic penetration seemed to be
succeeding, but Mustafa, who had hoped to get his hands on
Enfida without loosening his purse-strings, quarrelled with
181

NORTH AFRICA
Roustan. Having reconciled himself with Maccid, he made it his

business to obstruct all the designs of the French: he prevented
the Societe Marseillaise from taking possession of Enfida, and
delayed the construction of the railway lines for various specious
reasons. At Tunis, as in Europe, the situation was getting
gradually worse. In Britain, the defeat of the Conservatives in
the 1880 elections had once more put the Liberals in office. The
Quai d'Orsay was soon made bitterly aware that the Gladstone
cabinet was much less well disposed towards France than Disraeli
and Salisbury had been.
Nevertheless, in the spring of 1881 the French government
decided to act. Brought on by the Enfida crisis and by Italian
provocation, the decision taken by the Ferry cabinet was the result,
of a long campaign waged by Baron de Courcel, the director of
political affairs at the Quai d'Orsay. Courcel had long attempted
to influence the minister for Foreign Affairs, but Freycinet had
resigned without doing anything. Courcel succeeded in convincing
his successor, Barthelemy-Saint-Hilaire, a minister in the Ferry
cabinet. But the ministry had no intention of engaging upon an
expedition far from home, and Jules Ferry was entirely taken up
with educational matters. The date of the elections was approach-
ing, while in the National Assembly the opportunist majority was
indifferent or even hostile. But Courcel approached Leon
Gambetta and convinced the president of the Chamber of his case.
The interview was decisive. With Gambetta's support, Ferry was
covered on his left; he therefore made up his mind and acted
energetically.
All that remained was to find some pretext for intervention,
certainly not a difficult task. For months past, Roustan had been
striving to demonstrate this fact. ' Do not doubt that every week
we have a casus belli on the frontier,' he wrote to Courcel in
September 1880; 'it is up to us to make use of it.' At this juncture,
a group of Khrumirs appeared on the Algerian frontier. An
engagement between French troops and Tunisian tribesmen on
30 and 31 March 1881 gave Roustan the pretext he was waiting
for. A last survey of diplomatic opinion showed that Germany
was still favourable, Britain hesitant, and Italy isolated and
powerless. France was thus free to act as she wished.
On 4 April, Ferry informed the National Assembly of the
Tunisian incident. After demonstrating the necessity for a punitive
182

TUNISIA
expedition, he obtained from the Chamber of Deputies a

unanimous vote for the military credits of 5 million francs that
he had requested. At the end of April 30,000 men invaded the
Regency from the west, while a naval squadron landed 8,000 men
at Bizerta. As the bey made no attempt at resistance, the Tunisian
campaign was merely a military progress. Three weeks' march
took the French troops to the gates of Tunis without a fight. In
consultation with General Breard, Roustan presented the bey
with a draft protectorate treaty, which was signed at the Bardo
on 12 May. Twelve days later, the treaty was unanimously ratified
by the Chamber of Deputies.
But even while the expeditionary force was being withdrawn,
the French military command was surprised by an insurrection
in central and southern Tunisia, which occurred at the end of
June. Sfax and Gabes rebelled, and the tribes of the interior took
up arms. The Hammama, the Zlass, the Methellith, the Swassi and
the Beni Zid gathered under the leadership of the qa'id of the
Neffat, 'All ibn Khalifa, who was relying on Turkish intervention.
To the west, the Majeur, the Frechich and the Oulad Ayar were
also on a war footing. The French were forced to campaign during
the summer, sending reinforcements of 50,000 men taken from
garrisons in France and Algeria. Sfax was bombarded from the
sea before being taken by storm by the marines on 26 July.
Finally, Kairouan was occupied without a struggle on 26
October by three military columns converging from Tebessa,
Tunis and Sousse. Gabes and Gafsa were taken in November. In
their flight from the French, more than 100,000 nomads sought
refuge in the Tripolitanian borderlands, placing themselves under
the protection of the sultan. Most of them returned very soon,
making an act of submission, but there were some groups which
held out for nearly three years, and their attack kept that disputed
frontier in a state of permanent insecurity.
The Bardo treaty was a copy of previous similar ones. It dealt

with the military occupation of the Regency, and the establishment
of a resident who would act as intermediary between the French
government and the Tunisian authorities, at the same time acting
as the bey's foreign minister. In addition to the reorganisation of
the country's finances, the supervision by French officials of local
administration was also provided for. In the event nothing was
183

NORTH AFRICA
spelt out in detail; the word 'protectorate' did not even appear
in the treaty.
After the bey died in 1882, the fate of the Regency immediately
became a major concern of French opinion. Was Tunisia to remain
a protectorate, or was it to be joined to Algeria as a fourth
department? Jules Ferry's point of view prevailed. He preferred
to keep the protectorate formula, and France signed the conven-
tion of La Marsa on 8 June 1883, which regulated the status with
the new bey, 'All.
The form the protectorate was to take was determined by Paul
Cambon, the resident from 1882 to 1886. The regime rested on
the fiction of the absolute power of the bey, a convenient shield
concealing the real power of the resident-general. In practice, the
bey reigned but did not govern. The resident was not merely the
foreign minister, he was by right of office the chairman of the
council of ministers. All the administrative services in the
Protectorate were under his control, as were the commanders of
the land and sea forces. No decree could be promulgated by the
bey without his seal of approval. The central government was
reduced to its simplest form: a prime minister assisted by a sort
of secretary-general adorned with the romantic title of minister
of the Pen. Alongside this traditional administration, Cambon
created autonomous departments of state, which were entrusted
to French officials. At the local level the hierarchy of beylical
officials was not altered, but the number of qffids was reduced to
about forty, and their administration was supervised by thirteen
controleurs civils {district commissioners). The treaty had not specified
representative institutions, but under pressure from the French
colony the successors of Cambon from 1891 onwards instituted
a consultative assembly based on the example of the Algerian
delegations.
The important public services in the Regency were organised
according to the same principles of economy and respect for
tradition. The reform of Muslim justice was an undertaking that
would take some time, but as early as 1883 French courts of law
were set up for the Europeans. The reorganisation of finance was
one of the most urgent tasks, but, before that could be done, the
International Commission had to be wound up. This was achieved
in 1884, when the old debts were consolidated at 5 per cent and
184

TUNISIA
a new loan at 4 per cent was guaranteed by the French Government.

The former taxes were maintained, but the reduction in the rate
of interest on the debt, and the establishment of strict control of
revenue collection, made it possible to balance the budget.
Surplus revenues, which became a regular occurrence, were to be
used for a large-scale programme of public works.
State education was organised on the lines then in force in
metropolitan France, but private education still attracted many
adherents. The Italians provided their own schools, officially
recognised by the conventions of 1896. On the other hand, it
proved impossible to reform traditional education to any great
extent: in the Koranic schools, children still intoned the
innumerable suras of the Koran under the rod of the mu'addab.
Despite its reputation in the Muslim world, the education
dispensed by the Great Mosque of the Zituna consisted mainly
of learning by rote.
The provision of health care throughout the country was also
a long-term project. Doctors and nurses began the struggle
against malaria, trachoma and syphilis, and against epidemics
such as cholera, typhoid and smallpox, which periodically
attacked the native population. With their open drains, and with
heaps of rubbish piled up at the gates in their walls, all the towns
were permanent breeding-grounds for infection, the capital most
of all. In 1910, Tunis had five hospitals and one institut Pasteur,
but in the interior, health services were still rudimentary. The
administration tried to remedy this by providing mobile dispen-
saries, which made their rounds in the countryside at regular
intervals.
There was much to be done in Tunisia in the agricultural
domain, but the authorities had no intention of taking this upon
themselves. Deliberately eschewing the Algerian policy of official
colonisation, they opened the way for individual initiative. But
in order to encourage the acquisition of land by Europeans, the
extremely complex system of land tenure had to be clarified.
Cambon instituted a programme of land registration on the lines
of the Torrens Act in Australia. This was the object of the Land
Law of 1885, which rendered a general survey unnecessary. Land
was cheap. Companies and individuals bought vast estates from
the impoverished Tunisian notables. In 1892, in the Sfax area, the
185

NORTH AFRICA
government put up for sale 150,000 hectares of uncultivated land

which had become state property. New groves of olive trees were
soon planted by the French and Tunisian buyers.
At the end of the century, under pressure from the settlers, who
demanded the same advantages as in Algeria, the residency
embarked upon a policy of settlement, which until then it had
always opposed. Between 1895 and 1900, several settlements were
founded in the area round Tunis, a new departure which was soon
to arouse the wrath of the local bourgeoisie. As the land under
cultivation by French and Italian settlers increased, the production
of cereal crops made rapid progress. At the beginning of the
twentieth century, it was estimated that production had increased
by 50 per cent since the establishment of the Protectorate. A
vine-growing area had been planted near Tunis, while, the
number of olive trees increased from 8 million to 12 million
between 1881 and 1913, as a result of the development of
plantations in the Sfax region.
The exploitation of the mineral wealth of Tunisia depended on
the development of communications. The establishment of a rail
network and the development of the ports of Tunis, Sfax and
Sousse at the end of the century made it possible to exploit the
phosphate deposits. Beginning in 1899, by 1905 the production
of phosphates reached 521,000 tonnes. But it was not until 1907
that the iron ore deposits on the Algerian border began to be
exploited.
On the eve of the 1914-18 war, Tunisia seemed a prosperous
country, and the French could with reason congratulate them-
selves on their achievements. But this success did not eliminate
other difficulties: the problems posed by the presence of a larger
Italian community and, in addition, after 1906, the political
demands of the Tunisian nationalist movement.
From 1896 on, the Italian community suffered under a French

slogan which depicted them as the ' Italian peril', a peril, to speak
the truth, as old as the protectorate itself, but of which the French
in Tunisia had only become properly aware after the signing of
the agreements defining the status of Italians under the regency.
For reasons connected with its European policies, the French
government, desiring closer relations with Italy, was prepared to
extend the privileges granted to Italian subjects settled in the
186

MOROCCO
Regency by a treaty dating from 1868 which was about to expire.
This resulted in the conventions of September 1896, signed by
the French foreign minister, Gabriel Hanotaux, and the Italian
ambassador in Paris, Count Guiseppe Tornielli. The French
refrained from automatically naturalising settlers of Italian
extraction as they had done in Algeria since 1889. The Italians
could engage in any occupation in Tunisia, including the liberal
professions, and Italian diplomas were recognised as the equivalent
of French qualifications. Italians had the right to acquire property,
to form associations, to provide schools and a grammar school.
Hanotaux had paid a high price for the official recognition of the
French protectorate by Italy.
These privileges were all the more dangerous in that the Italian
community in the regency was a large one. From 15,000 in 1881,
its numbers had risen to 71,000 in 1901, three times the size of
the French community. The danger would not have been great,
had it merely been a question of a mass of Sicilian immigrants
driven out of their overpopulated island by poverty and unem-
ployment. But a Jewish bourgeoisie, which had originated in
settlement from Leghorn, and which had become wealthy as a
result of the financial crisis just before the protectorate was estab-
lished, offered considerable resistance to gallicisation.
The 1896 agreements gave the Italians in Tunisia the means to
create a state within a state. With their primary schools, their
grammar school, their hospital, their clubs and their newspapers,
they could lead an entirely independent existence without ever
departing from their Italian way of life. Some of them had not
given up hope of seeing the kingdom of Italy occupy both shores
of the straits. For two more generations the Italian question, at
times dormant, at times fiercely active, was to remain an insoluble
and irritating problem for the French.
MOROCCO
In the closing years of the nineteenth century, Morocco was

the only country in Africa, apart from Ethiopia, to preserve its
independence. Protected by its mountains, it had remained
resolutely closed to foreign influence, despite the proximity of
Algeria. The battle of Isly (when, in 1844, in pursuit of 'Abd
al-Qadir, a force of 11,000 Frenchmen had routed a Moroccan
187

NORTH AFRICA
army of 30,000) had destroyed the myth of the sultan's military

strength, but the country, protected by its isolation and by the
rivalries of the great powers, still appeared to foreigners as a
mysterious, hostile world.
In many respects the Sharifian empire could be seen as a
medieval survival at the gates of a rapidly expanding Europe. A
theocratic monarchy, set in its archaic traditions, was trying to
maintain its authority over a country one and a half times the size
of the British Isles by means as rudimentary as those employed
by the first Norman kings. Ever since the time of the Sa'dlds, the
sultan, both temporal ruler and high priest of Islam, had necessarily
to be chosen from a Sharifian family, among the real or supposed
descendants of the Prophet. Even more than a sovereign by
'divine right', he was the intercessor between his people and the
divinity, sacrosanct in his person, with the gift of baraka, the
benediction bequeathed to him by his ancestors. As Commander
of the Faithful, it was in his name that the prayer was recited in
the mosques. For townsfolk and tribesmen alike, he was sidna, a
revered being whose name could not be invoked without respect
and awe.
Since the eighteenth century, power had continued in the hands
of the 'Alawid dynasty, whose origins were in Tafilelt. The
founder of the dynasty was Rashld, but it was Isma'Il, a con-
temporary of the later Stuarts, to whom it owed its renown.
Following him, the dynasty spent years exhausting itself in
fratricidal quarrels that occurred periodically as a result of
uncertainty in the succession. This danger had not entirely
disappeared in the nineteenth century, but by designating their
successors during their lifetime, the later sultans managed to avoid
the worst troubles. Muhammad had no difficulty in succeeding
his father 'Abd al-Rahman in 1859. In '873, his eldest son, Hasan,
succeeded to the throne without being seriously challenged. But
it was not enough merely to accede to the throne. A sultan could
not survive without periodically campaigning in the frontier
regions of an empire whose greater part was in a permanent
state of dissidence. Although the low-lying area posed fewer
problems, the sultan had only a primitive organisation to
administer it, the makhîn, the result of centuries-old traditions.
The makhsyn, the central administrative body and the personal
household of the sultan, concerned itself equally with the affairs
188

MOROCCO
of the empire and of the ruler, with religious as well as civil or
judicial matters. This small administrative body did not even have
a fixed seat, since it followed the sovereign about in all his travels.
Active sultans were by definition nomadic, and the government
was sometimes to be found in a tent, sometimes in one of the
imperial cities of Fez, Meknes, Rabat or Marrakesh. However in
the nineteenth century, Fez as the usual residence of the sultan,
seemed to have achieved the status of the capital city of the empire.
In spite of ideas of modernisation, the makh^tn's organisation
still displayed all the characteristic features of a medieval institu-
tion. The council was restricted to about half a dozen people who
had the ear of the sultan, court dignitaries whose areas of
responsibility were never clearly defined. Their authority was
wholly dependent on the pleasure of the sultan who, according
to circumstances, could entrust them at will with the most diverse
tasks.
The grand vizier acted as head of the central administration,
and was the person who worked most closely with the sultan. He
was in charge of all matters of state, but was more particularly
concerned with relations with the tribes. In this respect, he could
be thought of as a minister of the interior. Relations with foreign
powers were the province of the vizier of the sea, who thus filled
the role of a foreign minister. But the vizier, when not
accompanying the sultan on his travels throughout the land,
resided at Fez. Therefore, in order to maintain permanent contact
with the diplomatic representatives, who were quartered at
Tangier, he was represented there by an agent or nd'ib who often
combined this office with that of pasha. Beneath an archaic system
of titles, there were also the equivalents of ministers of finance,
war and justice. The commander of the sultan's guard and the
chamberlain were also highly placed officials in the makhvyn. The
chamberlain was in charge of the internal administration of the
palace and held the official seal of the sultan. Since he was in
constant touch with the sultan, controlling access to him, he came
to play the part of confidant, or even adviser, though he was often
of very humble origin, a eunuch or a freed slave.
The ministers had their offices at the palace. But these offices
merely consisted of a single room, with rudimentary furnishings,
the baniqa, which opened off the courtyard of the mashwar. There
the ministers held audience or dictated their correspondence to
189

NORTH AFRICA
a handful of secretaries. The archives did not amount to more than

one or two boxes. When travelling, the entire correspondence of
the state was wrapped up in a few handkerchiefs. Apart from one
or two traditional gratifications, the remuneration of the ministers
was purely symbolic. The grand vizier was given seven douros
a month. Since the office had to support its holder, corruption
was the norm in the Sharifian administration. Servants and
dignitaries of the palace laid hands on what they wanted, and
frequently importuned the visitor by openly begging.
The sultan chose his councillors from rich bourgeois families
of Fez, as well as from the leaders of the makhîn tribes. But slaves
and freed slaves could equally well rise to the highest positions,
as was the case with the famous Ba Ahmad, son of a chamberlain,
and chamberlain himself, before becoming grand vizier of Sultan
'Abd al-'AzIz. The favour of the prince was unreliable; the rise
of a favourite might be rapid, but his fall still more abrupt.
Dismissal or summary execution was quite normal at the palace.
However, a few important families, the 'Ashrln, the Gharnit, the
Bargash and the Jama'I, had managed to stay in power. As so often
in despotic regimes, the great offices of state tended to become
hereditary. The Gharnit and the Jama'I were viziers from father
to son, while two families of black slaves monopolised the offices
of chamberlain and qa'id al-mashwar. At the death of Mawlay
al-Hasan, three of his ministers were Jama'I; ten years later, they
had been replaced in the favour of 'Abd al-'AzIz by the Gharnit.
The revenues of the Sharifian government came from direct
taxation collected by the qa"ids: the 'usbr {achour), or tithe on
cereals, the âkdt, a tax on herds, and a war tax that became
permanent, the nd'iba, collected either in cash or in kind. The Jews
of the interior had to pay a tribute calledy/sya. These regular taxes
were periodically augmented by offerings called hadtya, which
towns and tribes had to grant to the sultan when they acted as
hosts during his travels, and also on the occasion of religious and
family festivals.
In addition to the demands of the makhîn, there were the
exactions of those who collected the taxes. The pashas and qffids
who did not receive a regular salary, lived largely at the expense
of those they administered. Almost everywhere, afloggingwas
considered the best method of making the falldhin pay. These taxes
were all the heavier in that they bore unequally on the various
190

MOROCCO
parts of the country. By definition, the dissidents of the mountain
areas did not pay taxes to the sultan, but even in makhîn territory
it was difficult to find anything to tax. Apart from the officially
exempt gish (Jaysh) tribes, the makhîn personnel, the shurafd'
{chorfa),10 the âwiyas and the majority of the notables found some
means or other to be exempt. In fact, the apportionment of
taxation followed a feudal system, in which the subjects of the
prince were taxed in inverse proportion to their ability to resist.
Whereas the weakest and most faithful tribes had to make regular
contributions, the makhîn had to fight or negotiate every year
with the great tented nomads. Although generally speaking the
makhîn managed in the end to extract a hadiya in money or in
kind, this periodic tribute was an acknowledgement of a some-
what remote overlordship rather than the payment of a proper
tax. From time to time, the sultan despatched a military expedition
{harka, baraka) to the centre or south of the country with the task
of gathering in the outstanding amount. The success of the harka
depended mainly on its military strength, but the nomads knew
how to disappear when it suited them. In general, the system was
both burdensome and of doubtful utility.
For military matters, the sultans had for long been content with
employing irregular forces, whose organisation and equipment
had altered little with time. The standing army consisted of no
more than 20,000 men supported as necessary by contingents of
cavalry sent by the makbîn tribes.
The black guards (âbid) of the Bawakhir (Bouakhar) per-
petuated traditions founded in the seventeenth century by Isma'Il
but at the end of the nineteenth century, their numbers had fallen
to less than four thousand. The most numerous elements in the
standing army were taken from the bedouin gish, military tribes
who in exchange for the services of their men had received
concessions of land and exemption from taxation. Having been
reorganised many times, the gish consisted of the three tribes of
the Udaya, the Sheraga and Sherarda. In addition, there were
contingents from the large makhfin tribes of the south, the Abda,
the Uled Ahmar, the Rehamna, the Harbil and the Menabha. The
soldiers that they supplied, the makha^nis, received their equipment
and their horses from the sultan. The more technical tasks, such
10
Plural of sharif (cbirif).
191

NORTH AFRICA
as the command of the corps of engineers, or the artillery, were

entrusted to Christian renegades, who had for the most part
escaped from Spanish penal servitude.
These forces had for a long time been adequate to protect
Morocco from foreign invasion, but their reputation collapsed at
Isly. In i860, the army was once more cut to pieces by the
Spaniards at Castillejos. Following Egypt's example, two sultans,
'Abd al-Rahman and Muhammad, undertook the reorganisation
of the army on European lines. Hasan in his turn developed the
tabors, regular infantry battalions created after the defeat at Isly.
Troops were levied in the main cities of the empire, and arms were
bought in Belgium, Germany and England. To officer these
forces, which now numbered seven thousand, the sultan could no
longer rely on renegades. He sent recruits to be trained in
Gibraltar, invited military missions from Europe, and took into
his service foreign officers, like the famous Harry Maclean, a Scot
who rapidly gained considerable influence at court. In order to
lessen his dependence upon supplies from abroad, the sultan had
decided to build modern factories supervised by European
engineers: at Marrakesh, a cartridge factory founded by Mawlay
Muhammad; near the imperial palace of Fez, the arms factory of
the Makina was founded in 1890. Meanwhile, the traditional
contingents from gish, increasingly neglected, were gradually
restricted to the role of garrison troops in Tangier, Larache, and
the four imperial cities.
But the upkeep of a modern army proved a heavy burden for
the Sharifan treasury. Rates of pay had to be cut and the level
of equipment reduced; worn-out equipment was not replaced.
Soldiers began to desert or to sell their arms. Uncared for, the
cannons which had been the sultan's pride fell into disrepair.
However, there were a few field guns, the equivalent offiveor
six batteries, which were still maintained to some extent. Every
year a detachment of regular troops took them on tour with the
harka. Despite the proverbial ineptitude of the gunners, they were
an indispensable auxiliary in reducing the mud-brick walls of
kasbahs and ksurs.
From time to time, to collect the taxes or to restore order among
rebel tribes, a harka of several thousand men gathered at the gates
of Fez. Normally, the sultan would strengthen the numbers of the
irregular troops by adding a few tabors and one or two field
192

MOROCCO
batteries. Weighed down by its baggage, the expedition moved
slowly, leaving deserters and laggards in its wake. In difficult
terrain, lack of order rendered manoeuvres impossible. If they
were taken by surprise, the troops had to abandon everything to
the enemy. Fortunately, the dissidents normally avoided pitched
battles. Military operations consisted of sudden sorties, clashes
between groups of foragers, and ambushes laid for isolated
groups. But the warfare was pitiless on both sides. The soldiers
gave no quarter, and the heads of the victims brought back as
trophies were set to rot on the gates of the imperial cities.
On most occasions, the leader of the expedition tried to
negotiate. On the plain, the army lived off the country. Requi-
sitions and raids weakened the tribes, who preferred to make a
bargain. For many years the harka had not dared to venture into
mountainous Berber territory. His artillery notwithstanding, the
sultan was in no position to undertake the pacification of the bildd
al-sibd. But since the dissidents proved incapable of uniting in
order to take the offensive, this derisory force succeeded in
keeping some sort of order among the tribes of the plains and,
helped by fortifications, it was sufficient to secure the makhîn
against surprise attack.
Bildd al-makhîn and bildd al-sibd, the pacified region and the
dissident - this was the traditional division in Morocco on which
the whole internal organisation of the empire reposed; it had
existed since the time of the Saadians, and was to last throughout
the 'Alawid dynasty. The boundaries might vary slightly from
one year to the next, according to the fortune of a campaign or
the success of a rebellion, but in its main lines the division of
the country had remained astonishingly stable since the beginning
of the nineteenth century.
Broadly speaking, the bildd al-makhîn corresponded to the
region of the plains, the cities, and the areas more or less
profoundly arabised. The mountain area was bildd al-sibd, a place
of refuge where Berber language and tradition were kept alive.
The bildd al-makhîn covered about a third of the country's area
and roughly half of the population (2.5 million out of 5 million),
since it covered the most densely populated area.
In the north, between the Rif and the Atlas mountains, were
the plains of the Sebou, a vast triangle of land facing the Atlantic
from Tangier to Rabat. In this area, the dynasty had its capital,
193

NORTH AFRICA
w'°w
i
3S°N
Bitad al-makhzin
K::::::::::::i Zone of influence
© 100 0 0 0 inhabitants
• More than 2 0 0 0 0 inhabitants
Rabat Port open to international trade
Meknes Imperial city
0 300 km
tol°w 0 200mile!
8 The Sharifian empire at the beginning of the twentieth century
Fez (with a population of 100,000), which had replaced Meknes,

founded by Mawlay Isma'Il and deserted by his successors. Fez
was not just a political and religious centre; with its businessmen
and its handicrafts, it was also the economic centre of the empire.
The Atlantic plains, which stretched for more than 400 km
between Rabat and Mogador (Chaouia, Dukkala, Abda and
Chiadma), were lands that were properly pacified. The develop-
ment of commercial relations with Europe had favoured the birth
or rebirth of Casablanca and Mazagan, encouraged and then
discouraged the development of Mogador. The Jews arrived in
large numbers, leaving Marrakesh and the mullahs of the interior.
At the close of the century, they formed almost 25 per cent of
the population of the Atlantic ports. At Mogador, they numbered
9,000 out of the 20,000 or so inhabitants at that time.
Thanks to the fidelity of large tribes like the Rehamna, the
194

MOROCCO
interior plateaux as far as the fringes of the Tadla, and the lower
slopes of the Atlas, were also bildd al-makbîn. In the south,
Marrakesh was still the second capital of the empire, even though
the dynasty had deserted it in favour of Fez. Owing to the regular
trade with the Sahara, it was an important commercial centre, with
very active artisanal industries that provided a livelihood for a
population estimated at 100,000. The sultan had a palace in
Marrakesh, and was normally represented there by one of his sons,
who administered the region with the far-reaching powers of a
viceroy.
Links between north and south were difficult. Not far from the
coast, the Zaer-Zaian massif and the Zemmour territory were bildd
al-sibd. The stronghold of Rabat, the strategic position of which
was of major importance, was a necessary staging-post for
messengers, caravans, and the sultan's armies. Everywhere else the
authority of the makhîn was uncertain or shaky. The Tafilelt was
obedient, but its comunications with the capital were irregular.
The Sousse was still not stable. The great chiefly families of the
south, the Mtugga, Gundafa and Glawa, who controlled the
passes through the High Atlas, were allies rather than vassals.
Since he had been unable to forsee or to prevent the growth of
their power, the sultan had granted each of them the title of qd'id,
which attached them legally to the makbîn. But in these areas his
authority was theoretical rather than real. As the sultan himself
admitted, a new power was emerging, a Berber feudal state which
was independent of the makhîn, but no longer part of the
anonymous dissidence of the bildd al-sibd.
The sultan had merely a rudimentary administrative network
to keep the country under control. The large towns were subject
to the authority of a governor, usually with the title of pasha. The
tribes and smaller towns were administered by qa'ids, of whom
there were about three hundred. Under the qa'ids were the
shaykhs, district or village chiefs; amongst the nomads, they ruled
one or two douars, sometimes a whole section of a tribe.
In the most obedient regions, the qa'ids were not normally
members of the tribes. The sultan chose them from the dignitaries
of the makhîn. Many of them did not take up residence there,
but delegated their powers to a khalifa, whose task was to
represent them. In other cases, the sultan granted the appointment
either to representatives of important families, or to local chiefs

NORTH AFRICA
who had gained some authority. Everywhere administration

consisted only of the most elementary tasks. The duty of the qffid
was to maintain order and to collect taxes, and if need be to take
command of a contingent of cavalry raised in the tribe. Often
these offices were sold to the highest bidder, since the appoint-
ment was always accompanied by a more or less voluntary
distribution of gifts. Moreover, the qaids saw their functions
mainly as a source of revenue, enabling them to live as befitted
their rank and to maintain their households. Their greed made
all the heavier the load of taxation borne by those whom they
administered.
Owing to the extent of their powers, some qaids became
influential figures, but this authority was tempered by the caprice
or mistrust of a ruler, who could appoint them or dismiss them
at will. Fortunes made too rapidly, or too independent an attitude,
could arouse jealousy or suspicion. This was followed by sudden
dismissal, often accompanied by prison and the confiscation of
goods. Sometimes there was a revolt of a tribe or a section of a
tribe that had been too harshly exploited. The qitid was then
confined to his kasbab, while the rebels pillaged his barns and stole
his flocks.
From time to time, the sultan had to move his qetids around,
dividing or deporting rebellious tribes. The brutality of the system
was only equalled by its instability. In the event of troubles, or
at a time of dynastic crisis, the fidelity of the tribes was as
unreliable as the devotion of the qaids.
Beyond the bildd al-makhîn, the bildd al-sibd seemed a hostile
world, difficult to penetrate, with an area twice as great as that
of the pacified territories, an outer Morocco which was totally
outside the sultan's authority. Although the boundaries were
often fluid, in general they corresponded to linguistic divisions.
The bildd al-sibd was mainly traditional Berber country: in the
north the range of the Rif, with its Mediterranean coastline, and
in the south, from the Bou Regreg to Cape Rhir, the vast crescent
of high land enclosing the Atlantic plains and plateaux. But
although, geographically, the dissident area formed a compact
bloc, politically it was completely broken up. Being incapable of
forming any kind of federation, the Berbers in their mountains
were not able to challenge seriously the central power. Despite
its faults, the makhîn was the only organised state in the country.
196

MOROCCO
From north to south, three main groups could be distinguished,
the people of the Rif, the Beraber and the Chleuhs, differing in
their origins, their dialects and their traditions. The Rif was
inhabited by about thirty tribes who lived in the eastern part of
the mountain range. Connected with them were the Trifa and the
Banu Snassen on the borders of the Algerian Oranie. A threat
to shipping in difficulty along their coasts, these mountain peoples
were even more of a threat to the caravans venturing to pass
through the Taza gap. On the other hand, the Djebala, to the west
between Ouezzane and Tetuan, heavily arabised, had no great
difficulty in accepting the authority of the sultan.
The Beraber of the Middle Atlas, settled or semi-nomadic, were
the most dangerous of all. Their hostility, which closed the Tadla
to the movements of the sultan and his agents, forced the monarch
to make lengthy detours along the coast whenever he visited
Marrakesh. The tribes were spread over a vast area where they
nomadised with their flocks. The Banu Ouarain, the Ai't Youssi
and the Banu M'guild were among the largest, as were the
scattered sections of the Ait S'rhoushen. To the west on the
plateaux overlooking the Rharb and the Atlantic plain, the Zaer,
the Zai'an and the Zemmour were a constant menace. The plain
of Meknes did not escape their incursions. Brigands, who made
the routes impassable, carried on their activities with impunity-
right up to the gates of Rabat, where the Zaer regularly came to
exact tolls.
The Chleuhs of the south were no readier to submit than the
Beraber, but they were less troublesome, as they generally
remained on the defensive. Sedentary for the most part, they rarely
ventured down to the plain. Withdrawn in their valleys, they were
grouped in fortress villages frequently dominated by the
fortifications of their agadirs or granaries. The Anti-Atlas was
totally impenetrable. The oases of the Dar'a and the Dades,
populated by Negroes and half-castes, the haratin, had fallen under
the domination of the Ait Atta, who forced them to pay tribute.
On the other hand, in the central part of the High Atlas, the sultan
had managed to gain a few advantages. Favours granted to
important Berber qaids enabled him to cross the mountains
without fighting. The Gundafa opened the way to Taroudant
through the N'fiss valley and the Tizi n'Test; the Glawa did the
same for the Dar'a and the Tafilelt via the Tizi n'Telouet. The
197

NORTH AFRICA
Glawi had been rewarded for his good will with the gift of a
Krupp cannon from Mawlay al-Hasan. But the progress made by
the great chieftains did more to encourage their aspirations
towards independence, and nothing to strengthen the authority
of the makbîn.
The organisation of the Berber tribes perpetuated ancient
traditions found from the north to the south of the country under
various names, and also in Kabylia. Starting with the douar or the
village, each social unit also constituted a political unit, organised
on more or less democratic lines. Douars and villages, cantons and
tribes, each had their assembly, the djemaa, consisting of the heads
of the most important families. Important decisions were taken
in council, and their execution was entrusted to an amghar,
normally elected for one year. But the system naturally tended
towards oligarchy, with the most influential families agreeing
amongst themselves to run the assembly. From time to time, an
ambitious character gained power through cunning and violence.
Once he was amghar for life, he had merely to fill the djemaa with
his proteges to wield undivided power. But such despotism was
under constant threat from the vengeance of enemies or the
betrayal of disappointed supporters.
The djemaa governed, or meted out justice, according to
custom. When danger threatened, the tribe would occasionally
elect a war leader, and if necessary form alliances with neigh-
bouring tribes. But these confederations lacked stability. There
was too much rivalry between the tribes for them to bring
themselves to form a united front to oppose the schemes of the
makhîn. The sultan played upon their disagreements to try to win
supporters. Tours of pacification were also diplomatic campaigns
in the bildd al-sibd. But in order to make efficacious use of the
short-lived alliances that resulted, the sultan would have required
a stronger force, and above all more continuity of action. In vain
the army periodically went on campaign in the same regions; the
authority of the makhîn dissolved as it left the plains, and faded
out completely on the lowest mountain slopes.
The Moroccan question suddenly became of vital importance in

1900, because of an internal crisis which arose from a change of
regime, and which created a state of anarchy giving the great
powers an opportunity to intervene in the country's affairs.
198

MOROCCO
Suddenly dragged out of its ancient isolation, Morocco at once
became the object of international greed; a new source of discord
for a divided Europe, and a pretext for repeated trials of strength
which endangered world peace.
In spite of the country's weakness, the prestige of the 'Alawid
dynasty had been maintained as long as Mawlay al-Hasan had
lived. Right up to the close of his reign, the old sultan had waged
war on the frontiers of his empire. Death took him in 1894, on
the way back from an expedition to the Tafilelt. To avoid
rebellion, the chamberlain, Ba Ahmad, prudently concealed the
death of the monarch. ' For five days on end, a litter conveyed
the imperial corpse, which arrived at Fez in a dreadful state of
decomposition.'11 But Ba Ahmad had already proclaimed one of
the sultan's younger sons, 'Abd al-'Aziz, whom his father had
designated as his successor. 'Abd al-'Aziz, son of a Circassian
slave, Lalla Rekia, had always been one of the sultan's favourite
sons, but was only thirteen years old, so that his entourage was
left free to act as it thought fit.
With the connivance of Lalla Rekia, Ba Ahmad in fact seized
power. As soon as the sultan was proclaimed, he had himself
appointed grand vizier, and cast into prison the Jama'I brothers,
the ministers of Mawlay al-Hasan, on the pretext that they were
plotting against the regime. To replace them, he called in two of
his own brothers; one became minister of war and the other took
over his old office of chamberlain. This made Ba Ahmad master
of Morocco. He governed as Hasan had done, with a mixture of
harshness and negotiation, and kept reasonable order in the
country. Although he made sure of his own fortune and that of
his family he left a well-filled treasury when he died in 1900.
'Abd al-'Aziz was then almost twenty years old. He was old
enough to govern, but although he showed a vague desire in that
direction, he quickly made it clear how little interest he took in
his role as ruler. He did not lack intelligence or willingness, but
he was careless and erratic. He was shy in public, and he enjoyed
being in a small company of familiars. His entourage took
advantage of this weakness to bend him to their will.
On the advice of his mother, 'Abd al-'Aziz took as his vizier
the secretary of Ba Ahmad, Al-Hajj Mukhtar, and as his minister
for war one of his mukha^nis, Mahdl al-Manabhl. The latter
" Ch.-A. Julien, Histaire de I''Afriqut du Nord (Paris, 1931), 7Z8.
199

NORTH AFRICA
quickly made his mark. In 1901, he engineered the appointment

of a vizier who could not stand in his way, Fadil Gharnit, a witty,
sceptical old man from a great Fez family. From then on, Mahdl
al-Manabhl became 'the all-powerful favourite whoflatteredand
amused his master'.12
As 'Abd al-'AzIz had developed a childish fad for all the
novelties of Europe, Mahdl became the sultan's major supplier.
He attracted to the court middlemen and commission agents, an
entire fauna of adventurers who came to the palace to proclaim
the merits of a varied array of shoddy goods. Every afternoon,
the sultan received friends and visitors. Among the ranks of the
entertainers the foremost place soon went to the qdyid Maclean,
the Scottish military instructor who had now spent twenty years
in the sultan's service. The palace became a vast bazaar in which
all sorts of trash piled up at enormous expense: mechanical toys,
billiard tables, gramophones, cameras, often thrown away before
they had been used. 'The sultan's wives put on silk dresses and
plumed hats. Some even had to wear wigs, and on Thursdays,
which were rest days for the imperial harem, the younger members
began to learn to ride bicycles.'13
The whims of the sultan and the liberties he took with the
customs of the court soon aroused public opinion against him.
At Fez, a way of life so alien to tradition caused indignation.
Before long, the sultan was accused of impiety, of despising
Muslims and living with Christians. Disquieting rumours spread
through the land. Everywhere the tribes, no longer curbed by fear,
began to stir. After two years of waste and muddle, money was
in short supply at the palace. In order to obtain more, 'Abd al-'Aziz
had the idea of replacing the traditional taxes, the %akdt and the
'usbr whose collection gave rise to innumerable abuses, with a new
tax, the tarttb (tertib), which would be levied on all agricultural
resources without exception, arable land, orchards and herds. The
introduction and collection of the tax would no longer be the
responsibility of the qaids, but would be entrusted to a special
body of umand' (fiscal agents). In exchange, pashas and qaids
would henceforth receive regular salaries.
It was difficult to change long-standing habits. Although
excellent in principle, the reform damaged too many vested
12
E. Aubin, Le Maroc d'aujourcThm(Paris, 1905), 146.
13
Ibid., 167.
2OO

MOROCCO
interests to be accepted without recrimination. Notables, religious
dignitaries, and in general all those who to some degree or other
enjoyed exemptions, protested loud and long against fiscal
innovations thought to violate the tenets of the Koran. The qd'ids
and their followers fanned the flames of resentment by attacking
the European demands, and accusing the sultan of selling himself
to the Christians. To get the reform accepted, an exceptionally
energetic sultan would have been needed. But as usual, 'Abd
al-'Azlz had already found a new interest. Unconcerned by the
news reaching him from the tribes, he allowed the agitation to
grow throughout the country.
A marabout from the Zerhoun, JilalT ibn Idrls, better known
by the soubriquet of Bu Himara, 'the man with a donkey', took
advantage of the circumstances to raise the standard of rebellion.
As secretary to one of the sultan's brothers, Jilall had a few years
previously been involved in some rather shady business and been
imprisoned for two years. Ever since then he had travelled the
country, gaining a reputation for holiness by using his talents as
a conjurer. The agitation provoked by the introduction of the
tartib was a godsend to him. He preached against the new tax,
passing himself off as an elder brother of the sultan, who had been
denied the throne. The Riata of the Innaouen accepted his claim,
and acclaimed him as sharif. Everywhere in the country, tribes
were in a state of revolt or refusing to pay the tax. The Berbers
took advantage of the inactivity of the makhîn, and descended
upon the plains. The Ait Youssi pillaged the kasbah at Sefrou,
while the Zemmour attacked the siiq (market) at Meknes. Mean-
while Bu Himara, fortified by the adherence of the Riata, settled
in Taza, making it his capital. At his call, the whole of eastern
Morocco rebelled. Oujda came over to his side in 1903, together
with all the tribes of the lower Muluya.
'Abd al-'Azlz finally made up his mind to leave Marrakesh,
which had been his residence for several years. He sent Mahdl
al-Manabhl to fight Bu Himara, but the imperial army scattered
ignominiously. This failure caused the fall of the favourite, and
he had to resign and go into exile. With bolder tactics, Bu Himara
could have taken Fez, but he was not capable of organising a
proper army. The disparate bands of soldiers that followed him
were good for nothing but pillaging and fighting among them-
selves. Neither side was strong enough to win a victory. As the
201

NORTH AFRICA
civil war dragged on, it degenerated into a monotonous series of

razzias, while the little authority that the makb^tn still enjoyed was
gradually dissipated.
The disaggregation of the Moroccan empire made it an easy
prey for the foreigner. In France, as in Algeria, it awakened
long-standing ambitions aimed at the southern Sahara. As in the
case of Tunisia, twenty years earlier, the increasing number of
incidents along an ill-defined frontier provided France with a
pretext for intervention. In 1901 and 1902, the French foreign
minister, Theophile Delcasse, forced the sultan to sign agreements
concerning the boundaries between Algeria and Morocco. Apart
from delineating the frontier, these agreements made provision
for the setting-up by France of an administration to police the
border and set up customs posts. The pacification of the area was
entrusted to Colonel Lyautey, who soon extended French influence
to the banks of the Muluya.
'Abd al-'AzIz was obliged to seek foreign loans. Faced by the
worsening financial situation, he would sooner or later have to
make political concessions. But the Quai d'Orsay could not
remain indifferent to the rival ambitions of the other Great
Powers. Great Britain was keeping a close watch on the area
around the Straits of Gibraltar. Spain, with a toehold in the
presidios of Ceuta and Melilla, was intent on having her share of
the Sharifian empire. Germany, which was carving out a place for
itself in the Moroccan market, had recently begun to cast envious
glances at the Atlantic coast.
The diplomatic campaign waged by Delcasse achieved its aim
in 1904. The agreements of 8 April, which brought to an end many
long years of colonial rivalry with England, settled the Moroccan
question by means of a quid pro quo arrangement between the two
powers. France would not stand in the way of British supremacy
in Egypt, as long as England gave France a free hand in Morocco,
with certain reservations about commercial concessions and the
neutralisation of the zone of Tangier.
Six months later, Delcasse made a treaty with Spain, based on
the partition of the country. In addition to a Saharan zone south
of the Dar'a, the agreements of 6 October 1904 recognised as
Spanish a narrow coastal strip stretching from the Atlantic to the
mouth of the Muluya, together with Tetuan and Larache. But,
202

MOROCCO
as a concession to England, Tangier was to be given the status
of an international city.
The denouement seemed imminent. In January 1905, Delcasse
sent a diplomatic mission to Fez, with the task of inviting the
sultan to 'restore order in his empire with the help of France'.
It is true that the qestion of a protectorate was not formally raised
in the French proposals. But 'Abd al-'AzIz could have no illusions
on that score. If he accepted, that was the end of his independence.
The overt intervention of Germany, added to the Russian defeat
in the Far East, brought about an international crisis. The Tangier
coup, William IPs visit to Morocco on 31 March 1905 forced the
French to negotiate. After the resignation of Delcasse, who was
blamed for the crisis, the government agreed to participate in the
international conference that Morocco and Germany had
proposed, and which met in the small Spanish town of Algeciras
on 16 January 1906. For three months, the discussions concen-
trated on the policing of the Sharifian empire. It was the aim of
France to obtain a ' mandate' from Europe, a mandate which she
would share with Spain. On the other hand, Germany preferred
the idea of an international police force run by Belgian, Swiss or
Dutch officers. But the United States and Italy accepted the French
proposal, which was also supported by England and Russia.
Germany was isolated, with Austria as her only ally, and even
Austria advised compromise. Finally the German foreign minister,
Prince von Biilow, had to yield.
The final act of the conference, on 7 April 1906, was to leave
to France and Spain the organisation of law and order in the
Moroccan ports, with, as a concession to the Germans, the
theoretical supervision of a Swiss inspector-general who was to
reside in Tangier. The officers of the police force were to be
French at Rabat, Mazagan, Safi and Mogador, Spanish in Tetuan
and Larache, and mixed in Casablanca and Tangier. In addition,
the procedure for the allocation of contracts for public works was
defined; a state bank with the power of issuing money was to be
set up, in which the powers would have equal shares. In fact,
Germany had been forced to accept the French proposals which
she had hoped to defeat. France had obtained de facto the
predominant influence in Morocco. But this dominant position
had strict limits set by the Algeciras agreements. By inter-
203

NORTH AFRICA
nationalising the Moroccan question, Germany had succeeded in

putting considerable obstacles in the path of her rival. There could
be no question of a protectorate in the immediate future. The
Moroccan affair was far from settled, and Germany was still able
to intervene at will, should France ever act imprudently.
THE SAHARAN REGIONS
In order to protect their Algerian possessions, the French were

gradually led to use the regular troops at their disposal to conquer
one by one the oases of the northern Sahara. Mistress of the high
plateaux since 1848, the arme'e a" Afrique was confronting the desert
from which from time to time devastating raids erupted. Laghouat
and Biskra had been occupied first of all, then Ghardaia, Wargla
and Tuggurt in 1853 anc^ 1854. Fortified posts were spread along
this frontier, controlling the main water holes. But the system was
both costly and disappointing. It immobilised garrisons, which
had to be supplied at great expense, and did nothing to prevent
the incursions of the nomads.
The massacre of the mission led by Colonel Flatters by the
Tuareg in 1881 had put an end to hopes of any peaceful
penetration of the Saharan heartland. Plans for a railway line from
Algiers to Timbuktu, which had fascinated French opinion for
some years, were adjourned sine die. In the west, the French were
held back by the Bou Amama uprising in the southern Oranie,
and the idea of a punitive expedition against the Tuareg, which
had already been indefinitely adjourned, was finally abandoned.
The establishment of a protectorate over the Regency of Tunis
did little to alter the main elements of the problem.
The Turks were having the same difficulties in the eastern
Sahara. Although the sultan claimed that his sovereignty extended
to the shores of Lake Chad, the authority of his governors was
restricted almost entirely to a narrow coastal belt in Tripolitania
and Cyrenaica.
For many years, Tripoli had been a semi-independent state, one
of the three Barbary regencies governed by the Karamanli
dynasty. But in 1835 a Turkish expedition had restored the direct
rule of the Sublime Porte. The country had been divided into two
provinces or vilayets ruled from Constantinople. But the duties of
204

THE SAHARAN REGIONS
the valis were virtually sinecures. The Ottoman governors confined
themselves to the administration of the coastal oases, exercising
a distant protectorate over the Fezzan. The great Saharan nomads
fought each other unhindered in the desert. Only the main
religious brotherhoods had any real authority, particularly the
Sanusiyya, which had grown up since the mid-nineteenth century.
In fact, over the greater part of the Sahara, from Borku to the
Atlantic, authority was mainly in the hands of the chiefs of tiny
tribes scattered over an immense area. Everywhere, the nomads
had reduced the population of the oases to a state of semi-slavery.
Although they were all to some extent of mixed blood, Tuareg,
Moors and Tubu thought of themselves as white, or biddn. The
Tuareg, Berbers who were superficially islamised, dominated the
central mountain massifs. They were divided into three con-
federations, the most powerful being that of the Ahaggar
(Hoggar) and the Ajjer. Warriors wearing the lithdm (veil), armed
with poor rifles, they charged the enemy with their barbed iron
lances and their shields of antelope hide.
To the west of the Tanezrouft, all the lands which stretched
from the Senegal to the Wadi Dar'a were the zone of transhumance
of the Moors. The most formidable were the Reguibat, arabised
Berbers who had long since given up the rosary of the marabout
for the warrior's rifle. Indefatigable pillagers, they were the terror
of the whole of Mauretania from Cape Juby to Timbuktu.
The settled peoples of the oases were Negroes or half-castes,
the descendants of Sudanese slaves, sometimes of freed slaves, the
haratin, who were forced to cultivate the land for their masters.
The date palm, associated with the production of vegetables, was
the only source of wealth of the oases. Reduced to the condition
of share-croppers in the over-populated oases, most of the settled
population had scarcely enough to provide their most elementary
needs. The mining of salt at Taodeni, Bilma and Borku was merely
a supplementary source of income, as were the tolls levied on the
caravans which still ventured into the great desert. Chronic
undernourishment, periodically aggravated by seasonal famines -
such was the common lot of the Saharans, whether sedentary or
nomadic.
The twentieth century history of the Sahara opened on 29 December 1899 and
2 April 1900 with two events which had been expected for decades, the taking
205

NORTH AFRICA
of In Salah by the Chaamba goum of Captain Pein, and the conquest of Chad
by the forces of Commandant Lamy, who was killed in the same battle as his
adversary Rabah.14
In fact, after long hesitation, the French in the end made up
their minds to undertake the conquest of the Sahara, which was
seen as the keystone of their African possessions. The only means
by which the French could reach the nomads, and force them to
fight a pitched battle, was to enlist native Saharan, traditionally
accustomed to exhausting marches through the desert. France was
able to recruit such auxiliaries amongst the Chaamba Arabs, the
sworn enemies of the Moors and the Tuareg. In a few years, the
expedition of the Chaamba, who had been organised by Com-
mandant Laperrine into Saharan companies, gave France a third
of the Sahara.
The victory of Tit in May 1902, won by the.goum of Lieutenant
Cottenest, quickly entailed the submission of the Ahaggar Tuareg.
The Chaamba then attacked the Ajjer, and their raids then
threatened the Turkish oases of Ghat and Ghadames. Only the
orders of the French government then prevented the conquest of
the Tripolitanian hinterland, an area that had come into the
Italian sphere as a result of the 1900 and 1902 agreements.
In April 1904, Laperrine joined up with the forces from French
West Africa, which were organised on the same lines as his
Saharan companies. There still remained the task of occupying
Borku and Tibesti in the east, and above all of subduing the
Moorish tribes of the west. Tripolitania was becoming ever
smaller as a result of the encroachments of the French and the
British, but from now on the Turkish officers put up a resistance
which was supported by the Sanusiyya, who had reconciled
themselves with the sultan in the face of the common enemy.
At the beginning of the century, the French Sahara covered
more than 4 million square kilometres, taking into account the
areas which had not yet been pacified. The greatest part had been
divided between Algeria (Territoires du Sud, 2,171,000 sq. km),
and French West Africa, but French Equatorial Africa also
possessed its Saharan annexes, as did Tunisia. It was still difficult
to evaluate the population of recently pacified areas. Nevertheless
it was estimated that this population could not amount to more
14
A. Mattel, in V'Afriqut au XXe siecle by J. Ganiage and H. Deschamps (Paris,
1966), 272.
206

THE SAHARAN REGIONS
than a million inhabitants, of whom 450,000 were in the Algerian

south.
But in the opinion of the geographers, this kingdom of sand,
sun and wind was of virtually no economic value, even as a means
of access to the Sudan. The mirage of a great caravan trade had
faded. The French had dealt this trade a mortal blow by
forbidding the traffic in slaves. Both by value and by weight, the
trade between the Maghrib and the Sudan was insignificant,
perhaps a thousand tonnes and not much more than 600,000 francs
a year. Most of it consisted of loads of sugar, coffee, tea, and a
few spices.
Of what use was it to evoke once more the dream of a
trans-Saharan railway, that would be the main artery of a unified
French Africa? Maybe the military would find a use for it; but
from an economic point of view, what was the point of a line laid
down at great expense across 1,600 km of desert? Ever since
routes into the interior had been pioneered in the direction of the
Sudan, it was still Dakar and the ports of the Gulf of Guinea which
attracted the major part of the trade of French West Africa.
207

CHAPTER 4
WESTERN AFRICA, 1870-1886
One may have doubts about the appositeness of treating the fifteen
or sixteen years between 1870 and 1886 as a phase in the evolution
of West Africa. These dates are significant only in the context of
the beginning of the colonial era, by which 90 per cent of the area
was still untouched at the end of our period. It could be said that
the decline in Saharan trade, the growth in imports of European
products and increased production of export crops, foretold the
imminent end of free Africa, by making certain areas dependent
on the world market. But, as Lord Salisbury pointed out, before
1880 no one in Europe was aware of this, let alone anyone in the
savannas of the Sudan. It is only with hindsight that we see these
things as premonitory signs.
The same applies to the date of 1870, even in the perspectives
of colonisation. It is clear to us today that the upsetting of the
balance of power in Europe as a result of the Franco-Prussian war
acted as a catalyst of economic and social evolution, impelling
Europe to occupy Africa, by taking advantage of its technological
superiority. But nobody foresaw this at the time, and in fact 1870
opened a period of colonial retrenchment. It was only after 1875,
with the Belgian and French activities on the lower Congo,
followed by those of the French on the upper Niger, that the
imperialist advance began; and the reluctance of Great Britain in
this respect is well known. On the other hand 1885/86 is a much
clearer dividing line, coming after the Berlin Conference, which
had tried to limit the damage caused by an attack of imperialist
fever that nobody had predicted. They did not yet realise that the
whole of Africa was about to be engulfed in colonialism, but a
sort of truce had been initiated — Hargreaves' 'loaded pause',
which would be followed after 1890 by the final stage of the
scramble for Africa.
The object of this chapter is not however to study the dynamics
of the partition of Africa, but rather those of African societies still
zo8

FROM SENEGAMBIA TO THE RIVERS OF GUINEA
in control of their own fate, at least in appearance. And here it

must be admitted that it is difficult to deal with so vast an area
for such a short period. Indigenous African history, which still
dominated most of the arena, moved at a slower rhythm than the
history of the colonial advance. It is only in Senegambia, on the
upper Niger, and among the Asante and the Yoruba, that major
developments can be seen as the direct or indirect result of
growing European pressure. Elsewhere, the balance brought
about by the revolutions of the preceding period was either stable
or else changing only slowly.
In an area stretching from the Atlantic to the central Sudan and
from the Gulf of Guinea to the fringes of the Sahara, it is hard
to find a single general direction before the colonial partition,
which occurred ten or fifteen years later. In this vast area, with
its huge diversity of environment and cultural traditions, a
thematic study is precluded, so each section of the area will be
dealt with in turn. Since the fundamental unity of the effects of
European economic and political action is to be treated in other
chapters, we can here only study its unequal effects, sector by
sector, on societies which were then subject to a range of different
historical processes. Some coastal areas apart, these sectors must
be determined in line with the organic realities of traditional
Africa; they must not be shaped tofitthe arbitrary frontiers that
colonialism was about the draw across the continent. We shall
therefore look first at the zone of acculturation, stretching around
the coast from Senegambia to the Bay of Biafra; next at the
immediate hinterland which, from the upper Niger to the Volta,
was open to influences from both the coast and the Sudan; and
finally at the great belt of the Sudan itself, from the Senegal to
Wadai, where events still seemed to move at the traditional pace.
Within this framework we shall endeavour to keep within the
chronological limits set for the chapter; sometimes, however, if
justice is to be done to the natural flow of events, there will have
to be excursions beyond these limits.
FROM THE SENEGAMBIA TO THE RIVERS OF GUINEA
Senegambia occupies a unique position in Africa, for it is the only

region in which the ancient acculturation zone of the Sahara cuts
across the much more recent one of the sea coast, subjected to
209

WESTERN AFRICA, 187O-1886
the corrosive influence of the Europeans. Here more than

elsewhere, this influence was limited by the strong cultural
tradition of the Sudan, which included Islam. As an easy outlet
from the Sudan, thanks to the two great rivers to which it owes
its name, which are blocked by rapids only at a considerable
distance from the sea, this area very early attracted firmly based
European settlement. The colonies of Senegal and Sierra Leone
were at this time the biggest in tropical Africa, with those of the
Gold Coast and Lagos coming a poor second. The smaller
settlements of Gambia and Portuguese Guinea lay between them.
From the time of General Faidherbe, Senegal ceased to be a mere
trading station and became a true colony. Walo, annexed in 1855,
provided more scope to St Louis, a charming little town like those
of the West Indies, clustered on its island around the governor's
palace. Along the Senegal river as far as Bakel, French posts were
set up to supervise the river traffic and its landing places: It was
the route into the Sudan, whose vast cities, described by Barth,
had haunted the colonial imagination since the time of Faid-
herbe.
Further south, opposite Goree, the port of Dakar was beginning
to develop, along with Rufisque, as the main outlet for the
groundnuts of Cayor. The annexation of Ndiander together with
the outpost of Thies created an unbroken stretch of territory
under French control. Further south still, on the Petite Cote and
the Salum, only isolated forts protected the landing places of
Portudal, Joal and Kaolack. Casamance, cut off by the Gambia,
will be dealt with later.
With at least 30,000 inhabitants, Senegal was no ordinary
colony. Three towns, St Louis, Goree and Dakar, had been
electing municipal councils since 1872. Rufisque made a fourth,
and the inhabitants, the 'originaires', had the status of French
citizens. After 1879 the citizens elected a deputy to the French
National Assembly and a conseil-general for the colony, which
made life difficult for the governor. At first the deputies were
former governors or white businessmen, but the celebrated ethnic
group of 'mulattoes' of St Louis was, despite its small numbers,
already playing a role similar to that of the Creoles in Sierra Leone.
Their strength lay in their bilingualism and their acclimatisation.
The annexed territory was divided into cantons, grouped into
' circles', under civilian or military administrators, following the
210

Algerian system introduced by Faidherbe. Already the economy

was dependent largely on the production of groundnuts; the hides
and rubber that were brought from up-river were now of little
significance. The nuts were exported in their shells, since the
French government had put a prohibitive tax on the import of
oil into France, thus delaying for half a century the birth of a local
industry. Exports in 1882 amounted to 83,000 tons, but the total
declined to 21,700 tons in 1886, owing to the slump of 1883 and
to competition from India. Camel caravans took the groundnuts
to the Petite Cote, where they were delivered to the merchants
of four large firms, mostly from Bordeaux: Maurel et Prom,
Maurel Freres, Buhan et Tesseire, and CSCO (later CFAO).
These firms maintained eighteen trading stations in the territory,
the more important being at Rufisque, Joal, Fatick and
Foundiaoune, near Kaolack. The river was ceasing to be an
economic lifeline, just as it was about to become the axis for the
conquest of the Sudan.
The large trading companies controlled the chamber of com-
merce, and were opposed to territorial expansion, fearing that the
trade routes would be upset. Steam navigation, however, had
reduced the cost of the import-export trade, thus making it
possible for small businessmen, including many mulattoes, to
participate. Some of these were also officials, who formed a radical
republican group, inspiring lively, critical newspapers. Filled with
patriotic fervour, they supported conquest. Although there might
be argument about an advance into the Sudan, there was none
at all about the conquest of nearby lands. These were the great
groundnut-producing areas, which were thus part of the world
market, even before they lost their political independence. In
order to facilitate their operations the trading companies demanded
a railway line from Dakar to St Louis, crossing the very
productive area of Cayor.
And yet at the beginning of our period, this extension of the
annexed area was not inevitable. Successive military governors
respected their instructions, which, after 1870, forbade military
adventures. Colonel Francois-Xavier Valiere ruined his reputation
by his strict obedience to these orders from 1869 to 1876. The
imperialistic atmosphere was such that Louis Briere de l'lsle was
able to take the offensive on all fronts from 1876 to 1881, followed
less enthusiastically by H. P. Canard, who had once served at St
211

‘The map which appears here in the printed edition has
been removed for ease of use and now appears as an
additional resource on the chapter overview page’.

WESTERN AFRICA, 187O—1886
Louis as a simple spahi (trooper), and more actively by the first

civilian governors, Rene Servatius (1882—3) and A. Seignac-
Lesseps (1884-6). From 1859 onwards, the local commander of
Goree controlled all the southern parts as far as the Rivers of
Guinea in the modern Guinea Republic. As the growth of the
groundnut trade attracted interest to this coast, this command was
extended southwards and became increasingly autonomous. The
semi-autonomous territory of Rivieres du Sud, established in
1882, now stretched only from Salum (Kaolack) to Benti, and was
governed by Dr Jean Bayol. In 1884 it reached no further than
Casamance in the north, but in 1886 it was extended to include
the trading posts of the Cote d'Or (i.e. the modern Ivory Coast),
which had been separated from Gabon.
I
The period of retrenchment coincided with that in which
Senegal was faced with the persistent efforts of militant Tijani
Islam to supplant the old forms of animism, which in places still
had some life left in them, but lacked the unity to confront
colonisation. In the Wolof and Serer lands from the Senegal to
the Gambia, a very productive area for groundnuts, the death of
Maabu Jaaxu, who had been killed in the Siin in 1867, had
weakened the Tijani offensive. His Serer conqueror, the bur Sine
{mad-i-Sinik) Kumba Ndofen Juf, tried to control the merchants
of Joal, only to be assassinated by one of them, Beccaria, in 1871.
He was succeeded by the brutal Sanumon Fay, who in 1887 signed
a protectorate treaty with Captain Reybaud, but was killed the
following year by Semu Mak Juf (1878-82). The death of this last
mad was followed by a long civil war, one side being supported
by the Tijani of Nioro and the other by the animists of Salum.
The animists won, under Mbake Mak and Mak deb Njay
(1887—98), in whose time the French administration was estab-
lished. The other Serer kingdom, Salum, which had been almost
destroyed by Maaba, was rebuilt by Faaxa Fal (1864-71) with the
support of the French post at Kaolack. Sajoka Mboj (1874—79)
helped Semu Mak to come to power. Subsequently Gedel Mboj
(1879-94) breathed new life into animism, and in 1882 started to
take the south away from the Tijani. The Muslims kept only the
eastern part of the area.
Pushed back from the Rip to Nioro, the Muslims remained
strong as far as the banks of the Gambia. Mamu Ndari, who
succeeded his brother Maaba, soon came under the influence of
214

his nephew Saer Mati (Said Mathi), a fierce warrior, who in 1877
incited him to make war on Biram Sisa, who had been Maaba's
lieutenant, but was now becoming too independent. Niany and
Wuli, higher up the Gambia, were at this time beyond Mamu's
reach, but Fodi Kabba came to the assistance of the Tijani from
Casamance. Having refused Bayol's advances in 1884, Saer Mati
entered into negotiations with the Gambia in 1887, but his
continual raids as far as the Senegal and a new attack on Salum
exhausted the patience of the French, and in 1887 they destroyed
Nioro. Saer Mati fled to Bathurst a few months before the
occupation of Niani and Wuli on the upper Gambia by Gallieni
in his pursuit of Mamadu Lamine.
While the heirs of Maaba were managing to hold their own in
the south, his former partisans had seriously upset the Wolof
territories, whose importance for the French lay in their position
between Dakar and St Louis. In 1869 Lat Dior, the former darnel
of Cayor, had returned to his native land as a district chief with
Valiere's agreement. At first he allied himself with Amadu Madiyu
Ba (Amadu Sheku), a religious warrior who was gaining control
of Toro (the downstream part of Futa) on the lower Senegal, and
expelled the bur of Jolof in 1870.
In 1871 Lat Dior was once more recognised as darnel. He then
turned against Amadu, who was killed with French help in 1875.
The darnel installed his nephew Alburi N'Diaye the legitimate
heir, as bur, leaving Bra, the brother of Amadu, to conduct
guerrilla warfare until his death in 1881. In 1877 the darnel, like
many of his ancestors, was recognised as tegne of Baol. Lat Dior
appeared to be collaborating with the French, but in reality he
was engaged on a profound transformation of the country, which
was being integrated into the world market through the trade in
groundnuts. To establish a united front against the Europeans he
imposed Tijani Islam, which abolished the barriers between
classes and castes. In his entourage, and then in Alburi's, was the
brother of Ahmad Bamba who later founded the Mouridiyya
brotherhood. In 1879, however, he had to sign a treaty with Briere
de l'lsle allowing the Dakar—St Louis railway line to cross Cayor,
although in 1882 he opposed the start of the construction. Driven
out by the French, he fled to Baol and the puppet darnels who
succeeded him, including his nephew Samba Laobe, allowed the
construction to proceed. With the help of Alburi, however, Lat
215

Dior continued with guerrilla warfare, and when Samba Laobe
was killed by the French in 1886, he returned in force. After his
death on 26 October 1886, Cayor was annexed and divided into
six cantons. Playing a double game, Alburi held on in Jolof until
1890, and then went to join Ahmad b. 'Umar at Nioro. He went
eastwards with him and died on the borders of Sokoto in 1899.
The Wolof had thus lost their independence and their warrior
aristocracy just when they were entering into Islam and the
colonial era. Groundnut cultivation, with its tendency towards
monoculture, was to upset completely their economic and social
structures. This process was already well under way when the
Dakar-St Louis railway was opened on 7 July 1885.
In the north of Senegal, river traffic in 1870 seemed to assure
the prosperity of St Louis, even though it may have decreased in
relative importance. Though Faidherbe had driven away the
Moors, the Tukolor living on the river banks were unaccom-
modating and impeded the river traffic from Bakel. Walo seemed
under control, despite the regrets of the district chief, the heir of
the brak, and Toro appeared secure after the death of Amadu
Madiya, but the upper Futa was now dominated by the powerful
figure of a maker of almamis, Abdul Bokar Kan. In a land which
had lost 20 per cent of its population at the call of al-Hajj 'Umar,
and whose young men were still emigrating to Nioro, this
aristocrat from Bosea had taken the lead in combating such
suicidal mysticism. After eliminating his rival, Tierno Brahin, in
1869, he then had to contend with the Wan family, in particular
Ibra Almami, but on the upper reaches of the river his power was
now unchallenged. He engaged in ceaseless intrigues with the
French, whom he supported against Lat Dior and Mamadu
Lamine, but when forced to choose in 1890, he sheltered Alburi
and took to the bush, only to be assassinated the following year
by the Moors.
Beyond Futa the Soninke (Sarakule) territory of Bakel - which
was already much under French influence, providing many laptots
(crew men on river boats) - appeared to be well under control.
But it was there that Mamadu Lamine, on his return from a long
pilgrimage to Mecca followed by imprisonment by Ahmad at
Segu, posed a threat to colonial power by appealing to the young
and the uprooted. After getting on good terms with the French,
he raised troops on the pretext of fighting the pagans of the
216

south, seized Bundu, which was allied to the French, and attacked
the river port of Bakel in February 1886. Frey then interrupted
his campaign against Samory and forced Mamadu Lamine back
towards the upper Gambia, where, pursued by Gallieni, he died
in November 1887.
To the north of the Senegal, nobody had yet thought of
occupying the desert country from which the Moors hadfled.The
latter's relations with St Louis were defined by the 1868 treaties,
and consolidated by subsidies paid to them from there. Their great
negotiator, Shaykh Siddya, had died in 1868 at Butilimit and his
descendants did not have the same influence. When the Emir of
Trarza, Ali Ould Mohammad al-Habib, died in 1886, he was
succeeded by Ahmad Fall, who was accused of having designs on
Walo. In 1887 with the support of the French, he was killed and
replaced by Amar Salum, who was to see the creation of the colony
of Mauritania twelve years later.
Thus the territorial extent of Senegal was being determined
between 1880 and 1890, at the time when the colony was used
as a base for the conquest of the French Sudan. It is within this
context, linked to the economic and social transformation of the
country, that there spread the Tijaniyya brotherhood, which had
been formed to Islamise the pagans still outside the European
world. The birth of the Mouridiyya marked the acceptance of
this new situation.
The British Gambia, an isolated dependency of Sierra Leone,
was not as insignificant as might be thought. Restricted mainly
to the islands of Banjul, the site of the capital Bathurst, and
MacCarthy Island further up the river, the oldest British colony
in Africa was inhabited by a group of Creoles from Sierra Leone,
who were motivated by intense patriotism. The opposition to the
exchange of the territory for the French trading posts of the Gulf
of Guinea, abandoned for the first time in 1876, came from them.
But it was then that the country, already an outlet from the Sudan
and for the gold from Bambuk, took on fresh importance with
the groundnut trade. These were in fact produced on French
territory, but the British customs duties favoured the trading
companies at Bathurst, some of which were French. Lacking both
men and means, however, the British officials were not able even
to control the 'ceded mile' on the north bank of the river. Except
when a warship was in port, they were unable to intervene in the
217

marabout wars, the Islamic movement that was destroying the
animist structures of Senegambia. On the north bank they were /
confronted by Maaba's Tijani and by Saer Mati, whose end in 1887
has been described.
Next door to Bathurst, the pagan Malinke kingdom of Kombo
had been in a state of revolution since 1854, as a result of the
activities of a reformer of Tukolor origin, Fodi Kabba Ture of
Gunjur. His brother Fodi Silla, who succeeded him about 1868,
became master of Kombo after taking the capital, Brikama, in
1874. It needed joint action by the British and French to relieve
Bathurst by eliminating him in 1894. But in 1870, much higher
up on the south bank, another power appeared, the Malinke chief
Fodi Kabba Dumbia, whose story can only be studied in relation
to that of Casamance. Gray confused these two Fodi Kabbas, who
were totally unconnected,1 and it is strange that all anglophone
historians have persisted in this error to this day.
We return to the rhythm of Senegalese history if we now study
Casamance, that isolated dependency of Senegal to the south of
the Gambia, with a river which, although navigable, cannot be
used as an outlet from the Sudan, since its whole course lies within
the forest zone. In 1870 the French possessed two posts there:
Carabane was established downstream among the Diola (Djola),
an animist society of sturdy rice-growers without rulers; Sedhiou,
up-river among Malinke who were in the process of islamisation,
was separated from Carabane by the old Portuguese presidio of
Ziguinchor in Banhun territory. Open to some European trade
but fiercely independent, it was only slowly and gradually that the
Diola were brought under control, particularly between 1888,
with the Seliki affair, and 1900, with some resistance continuing
as late as 1914.
From 1882, until the Berlin Conference, the Portuguese
attempts to consolidate their hold led to increasing clashes with
the French. The treaty of 12 May 1886 ceded Ziguinchor to France
in exchange for Rio Cassini, further to the south. After the
transfer in April 1888, Zinguinchor at first remained a separate
village, becoming the capital of Lower Casamance only in the
twentieth century.
Up-river all the land as far as the Gambia belonged to Kabu,
the centre of which was further south (in Guinea Bissau), and
1
J. M. Gray, A history of the Gambia (Cambridge, 1940).
218

which had almost entirely absorbed the old Banhun population.
However this great Malinke state had just collapsed as a result
of the attacks of Futa Jalon, coupled with the revolt of its Fulani
minority. Its last king had perished in the fall of the capital
Kansala in 1867. Vanquished and oppressed in the east, the
Malinke fell back towards the centre, around Sedhiou, where the
French authorities were trying to control trade and prevent
attacks on the pagan Balante. Sunkari Kamara, chief of Buje (in
the territory of Sedhiou), a brave but brutal man, then organised
opposition to the French. On two occasions, in 1872 and 1881—2,
he vainly attacked the French post at Sedhiou. He was finally
crushed by Commandant Dodds, sent from St Louis, to whom
the Malinke surrendered. Sunkari surrendered in 1887.
Sunkari had obtained assistance from Fodi Kabba, a member
of a family from Bundu settled in Jimara on the south bank of
the Gambia. As an early ally of the Tijani of the Rip, Fodi Kabba
gathered together in the name of Islam the Malinke who were
threatened with destruction by the Fulani. Therefore Alfa Mollo
captured his residence at Kerewane and massacred his family in
1871. Fodi Kabba regrouped further to the west in Pakao and in
1876, with the help of Sunkari, broke the Fulani offensive. He
then moved to Kiang, lower down from the Gambia, whence
from 1878 onwards he set out to vanquish the Diola of Fonyi
and convert them to Islam. Thus he built up a domain, half
Malinke, half Diola, astride the Senegal-Gambia frontier that
was to be drawn in 1890. By his cunning he was able to hold out
there at Madina until eliminated by the French in 1901. Other
Diola, settled more to the west, in particular the Jugur and the
Kalunay, were to be attacked by Ibrahim Njay, a merchant from
St Louis, between 1884 and 1888. Finally, he was disposed of by
the joint efforts of the French and Fodi Silla.
However, it was in upper Casamance that the most profound
changes took place. A Fulani of slave origin, Mollo Ege, soon to
be Alfa Mollo, who had been converted to the Tijaniyya by al-Hajj
'Umar, took part in the siege of Kansala in 1867. On his return
home, in Firdu, he launched a general revolt of the strong Fulani
minority, and easily overcame the Malinke who now lacked the
support of Kabu. Cut off in the west by Fodi Kabba, he set up
a powerful and dynamic new state stretching from the Gambia
to Rio Geba (in Guinea Bissau), supported in the south by the
219

Fulani of Futa Jalon to whom Alfa Mollo remained faithful all his
life. From his capital Hamdallahi, the conqueror established an
extremely centralised despotic government, striving successfully
to 'fulanise' the vanquished Malinke. After his death in 1881 his
son, Musa Mollo, broke with Futa Jalon, and in order to
overcome the opposition aroused by this policy, he allied himself
to the French, with whom he signed a protectorate treaty in 1883.
In 1887 he crossed the Gambia to help them to eliminate Mamadu
Lamine. However, enraged by the cessation of the south of his
domain to the Portuguese, he broke with France in 1903 and went
to spend the rest of his days in the Gambia. For fifteen years he
had enjoyed a semi-independent status, wielding, virtually
unchecked, a harsh and at times despotic power.
The colony of Senegal was in the final stages of being
constituted between 1885 and 1890, but the same cannot be said
of Portuguese Guinea further south. At first there were only a few
sleepy trading posts, such as Cacheu and Bissau, with no measure
of autonomy, since they were under the direct rule of the governor
of the Cape Verde Islands. Moreover the English had occupied
Bulama from 1858 to 1868, only handing it back after arbitration
by President Grant of the United States. The fact that this
territory was inhabited by a long-established population of
half-castes, who had made Portuguese Creole the lingua-franca all
the way from Petite Cote to Sierra Leone, made no difference to
its lack of economic and social significance. Worried by the
activities of the French, the Portuguese, in March 1879, made
Guinea into a colony separate from Cape Verde and moved its
capital from Bissau to Bulama. With poor, but relatively numerous
troops, the governors, Agostinho Coelho (1879-81), Pedro
Ignacio Gouveia (1881-4), and F. Paolo Gomes Barboso (18 84—7),
engaged in increasing military activity, usually with disastrous
results — in 1881—2 against the Beafades, in 1884 at Cacheu, in 1886
against the Beafades once more. Their only success was in 1886,
when Lieutenant Geraldes repulsed Musa Mollo from the Rio
Geba. The treaty of 1886, redefined in 1905, fixed the frontiers
of the territory, but it took many more difficult years before the
Portuguese made themselves masters of it, a process not completed
until 1907— 15.
From Rio Nunez to the Melakori a series of French trading
posts formed the major part of the domain of the Rivieres du Sud,
220

the first nucleus of French Guinea. The French at Boke, first

occupied in 1866, had difficulty in controlling the Landuma and
especially the Nalu (the wars of Dina Salifu taking place from 1884
onwards). Boffa, occupied at the same date, could not stop the
' war of the half-castes' from laying waste the Rio Pongo from
1865 to 1870. The Koba war was to follow from 1880 to 1888.
In Melakori, the post at Benti had been built in 1867 with the aim
of blocking the influence of Sierra Leone, which since 1819 had
been in possession of the Los islands right in the middle of the
area. This explains why the French initially gave up Kalum and
Dubreka.
In this region, where a clandestine slave trade had continued
for a long time and in which British influence was preponderant,
the rapid growth of the groundnut industry attracted Senegalese
traders, who thus found themselves in conflict with those of
Freetown. In 1878 the occupation of Melakori by Sir Samuel
Rowe, governor of Sierra Leone, greatly increased the tension,
but the convention of June 1882 provided the basis for the
Franco-British frontier. However the coastal minorities, in par-
ticular the Baga and Mmani, found it difficult to stand up to the
pressure of the Susu, who had been absorbing them for centuries,
while Islamic trade from the Upper Niger was slowly eroding the
animist traditions. It is impossible to detail here the complex
conflicts for power and trade in which, for lack of military means,
the French were helpless bystanders. In Melakori two Ture
lineages, the Maliguists and the Bukarists, were engaged in
endless civil war for the title of almami of Morea. This war broke
out again from 1878 to 1882, and it was only temporarily checked
by the Samorians between 1884 and 1887.
Dr Bayol, the lieutenant-governor of the Rivieres du Sud from
1882, was faced in 1884 with the German attempt on Dubreka
led by Gustav Nachtigal. Bayol reacted immediately by securing
the occupation of Conakry in July 1885, but the following year
the death of Bale Demba, the king of Kalum, was to unleash
hostilities which lasted until 1888. The colony of the Rivieres du
Sud was separated from Senegal in August 1889; t n ' s 'e<^ t o t n e
delineation of its frontiers in 1890, the British recognising its right
to extend its borders to the Niger, encircling Sierra Leone.
The importance attached to the Rivieres du Sud was not only
due to the known wealth of this troubled coast, but to the
221

proximity of the important Fulani state of Futa Jalon, for which

Boke and Boffa were the outlets, and which France intended to
absorb into her sphere at the expense of the over-hesitant British.
In Futa Jalon the biennial alternation between the Alfaya and
Soraya groups at Timbo was now effected with reasonable
regularity. The former were represented by the almamis Sori Dara
(1841-73) and Amadu (1873-90), the latter by Umaru (1840-72)
and Ibrahima Sori Dongol Fella (1872—89). However there were
attempted coups d'etat in 1875 and 1880. The era of conquest was
over, except in Kabu, a domain in the hands of the great lords
of the north, the alfa-mo-labe. However there was still a threat
towards the Upper Niger from the schismatic and revolutionary
Hubbu, and it was in attacking their fortress, Boketto, that
Ibrahima Sori Dara was killed in 1873. In order to dispose of them
Samory had to be called in 1884, following his arrival on the
eastern frontier of the country.
Now stabilised, the Fulani aristocracy left religious culture
largely in the hands of the Diakhanke minority. Greedy for slaves,
this aristocracy procured them by selling livestock to the
Samorians, and continued to enrich itself by its control of the trade
routes leading to the Rivers Coast and Sierra Leone. Its main aim
therefore was to eschew all risks, with the result that it thought
it prudent to welcome the Europeans who were now pushing at
the gates of its country, albeit without trusting them entirely. In
June 1880 the almami signed a treaty of friendship with Dr
V. S. Gouldsbury, the administrator of the Gambia. He thought
he was doing the same in June 1881 with Dr Bayol, but the text
of this treaty was skilfully worded to justify French claims to a
protectorate, which was recognised by the British in the treaty of
1890. The fate of the Fulani state was thus sealed without its
knowledge, a fate very easily confirmed by arms in 1896.
FROM SIERRA LEONE TO THE BANDAMA
The cautious policies imposed from London had lost for the
British the priority which was theirs as a result of the powerful
colony of Sierra Leone at the southern end of the Rivieres du Sud.
Together with neighbouring Liberia, it was the cradle of
anglophone culture in West Africa, which was to play a decisive
role not only in the establishment of British colonialism, but also
222

in the birth of African nationalism. Repatriates from America and

slaves liberated by the British fleet had already fused to form the
ethnic group known as Creoles, with only the Aku still speaking
Yoruba and at times practising Islam. By this time, after three
generations, a vernacular, Krio, had developed, together with a
culture founded on Victorian respectability as viewed by the main
Protestant denominations. The Creoles sent their children to
Britain for further education, and had already produced remark-
able intellectuals, doctors, lawyers or churchmen, of whom the
most famous were J. Africanus Horton, Sir Samuel Lewis and
Bishop Crowther. In the 'white man's grave' that Sierra Leone
then was, they soon filled the highest positions around the small
European staff surrounding the governor. They were filled with
loyalty to the Crown and wanted British domination extended to
bring civilisation to their 'barbarian' brothers. They expected to
be the means whereby this would be achieved, and it was not until
the beginning of the twentieth century that the quinine and the
racialism which accompanied the setting up of the colonial era
showed them their error. It was through their agency that the first
modern intellectuals were to arise in Nigeria and the Gold Coast,
and through them that links were established with the American
Negroes, then only just emancipated, thus preparing the way for
the great Pan-African movement.
They were less racially mixed than the ' mulattoes' of St Louis
and more numerous, and even the poorest of them had the trading
instinct. Since the northern routes running from Freetown and
Futa Jalon to the upper Niger were in the hands of the Malinke
(Mande), the Diakhanke or the Dyula, it was the forests of the
south, the lands of the Mende, as far as the Liberian frontier, that
were opened up to European products by small-scale Creole
traders. On the border between the groundnut and palm-oil areas,
the colony was also open towards the upper Niger, from which
came the cattle and the ivory and the gold from Bure. The effect
of the world depression was very severe in 1875 and 1883, when
palm-oil and groundnut prices collapsed.
Freetown and St Louis, clustered around their Christian
cathedrals, were the only modern towns in West Africa. As a base
for the British fleet, a coaling station, a telegraph post and a port
of call on the regular Elder Dempster line from Liverpool,
Freetown was after 1866 the seat of the governor in charge of
" 3

all the British West African territories from the Gambia to Lagos.
Military and civilian governors alternated: the authoritarian
Major A. Kennedy from 1868 to 1872 was unpopular with the
Creoles, who however adored the Irishman, J. Pope Hennessy
(1872—73). G. H. Kortwright (1875—7) w a s succeeded from 1887
to 1881, and again from 1885 to 1888, by an activist military
doctor, the formidable Sir Samuel Rowe, whose two terms of
office were separated by that of Captain A. E. Havelock (1881-4).
These governors never at any time took, measures for the
evacuation of West Africa, as recommended by the parliamentary
commission of 1865. Egged on by the Creoles, they dreamed of
extending the colony by further annexations along the coast, so
that its financial problems could be ended by the establishment
of a single customs regime, and then extending it towards the
Niger to open up the Sudan to its trade. But their ambitions were
not be realised, for during this period London was opposed
to any expansion. The main lines of the frontier with French
territory were drawn up in 1882. In the same year the frontier with
Liberia was also determined, when, despite the protests of
Monrovia, Havelock occupied Sulima on the river Mano. Thus,
juridically, the country consisted only of the peninsula and a few
coastal stations. The former British sphere of influence in the
Rivieres du Sud had been abandoned to the French. Beyond the
peninsula, the British acted only by diplomatic means, the
garrison having recourse to arms only as a last resort, never
occupying the territory of those whom they had defeated in
battle.
As in the case of the Guinea rivers, there is no space here to
consider the complex struggles between the Susu, Loko, Bulom,
Temne and Mende people. These derived from the expansive
tendencies of some ethnic groups, such as that of the Temne at
the expense of the Loko, or that of the Mende, whose vanguard,
the Kpa, only reached the area around the Colony after 1850, or
from attempts to control Freetown's trade routes, or from
struggles for power. From 1875 onwards Gbanya, the powerful
chief of the Kpa, intervened in the civil war between the Caulkers,
the half-caste chiefs of the Bulom of Shenge. Rowe had Gbanya
flogged and his Caulker ally hanged. Further to the north, other
Mende, the Yoni, were attempting from 1878 to advance to the
224

Rokel. Later they attacked Koya near the Colony and made an
alliance with the Caulkers. The Freetown garrison forced them
to make peace in 1887, by force of arms. From 1880 until the end
of the century, Gbanya's widow, the famous Madam Yoko, ruled
the Kpa in close collaboration with the governor.
But the Sudanese dream directed men's minds towards the
north, an area which experienced great upheavals at the end of
the period. It was in 1884 that Langaman Fali, the leader of the
Samorian army in the west, captured Falaba in Sulimana and
conquered the Limba, while Dauda's mission appeared in Free-
town in January 1885. Close commercial relations, based on the
sale of modern arms, were established between Freetown and the
Samorians, which were to last until 1887, when the great revolt
of the Dyula empire divided the two parties. Then the Samorians
led astray the Creoles with their dreams of a Sudanese empire for
the British. This explains the mission of Major Festing, who died
in 1888 on his return from Sikasso, wrongly convinced that
Samory had not pledged himself to the French.
In spite of the failure of its wider ambitions on account of
pressure from London, the colony of Sierra Leone enjoyed
considerable prestige, whereas neighbouring Liberia had a bad
reputation even though it was independent. This however, is to
reflect back into the past the sorry situation of the Liberian
Republic at the beginning of the twentieth century, for it must
be admitted that the first generations of Liberians, isolated and
with very few means, nevertheless accomplished much. The
country at that time consisted of a few coastal enclaves with no
more than 30,000 inhabitants (of whom half were Afro-
Americans), who participated in the anglophone culture and
provided a stimulus for the Pan-African movement. They
produced remarkable personalities, such as the educationalist,
Edward Wilmot Blyden, who was born in the Danish West
Indies and divided his career between Monrovia and Freetown.
The Afro-Americans were often traders, but their activity was on
a small scale, for the route into the Sudan, from Monrovia and
along the St Paul river towards Konyan, was not as easy one. It
was to this area, Samory's homeland, that Benjamin Anderson,
the secretary of the Treasury, made his way on two occasions, in
1868 and 1874, on the initiative of President James Spriggs Payne
225

(1868—70) and President Joseph Jenkins Roberts (1848-56 and
1872—8). After him the chapter of Liberian exploration was
closed until the twentieth century.
It was at this time that the half-caste element which had
brought progress to the colony and expressed itself through the
Republican party was definitively ousted from power by the
Negroes, who were poorer and supported the True Whig Party.
The first Whig President, Ed. James Roye, was however, the
richest merchant in the country. Anxious to build a railway line,
he floated a loan on the London stock market and was accused
of corruption. He was further accused of dictatorial tendencies
for having his term of office extended from two to four years, and
was deposed and assassinated. The founding father, Roberts, a
very light coloured half-caste, restored order. Nevertheless, with
Anthony W. Gardiner (1878-83), who resigned following the
arbitrary British action on the frontier, and his successor, Hilary
R. Johnson, the Whigs ensured the continuation of Negro
supremacy.
The Afro-Americans suffered considerably in the 1875 depres-
sion and withdrew into themselves in fear in the face of the native
population, whom they considered to be barbarians. In 1875
Roberts came into conflict with the Protestant churchmen, whom
he accused of giving too much education to the Africans. From
1873 onwards he clashed with the nationalistic opposition of the
acculturated native Africans of Maryland (Harper to Cape Palmas)
who proclaimed the Grebo Reunited Kingdom, invoking a
precedent from 1834. These Christians complained that they had
been deprived of their lands and their brokerage rights by the
Liberian colonists. They inflicted a crushing defeat on the
Liberian militia, and it needed the direct intervention of the USS
Alaska in 1876 to bring them under the control of the government.
Grebo nationalism, which was to produce the prophet Harris,
continued to take to arms until 1915.
The treaties of 1886 and 1892 with the British, and 1895 with
France, would confirm the Liberian boundaries with Sierra Leone
and Ivory Coast, but recognise Monrovian rights over the
hinterland as far as Konyan. These rights were ceded to French
Guinea in 1911. But the hinterland was not to be occupied until
the twentieth century. We cannot here trace the 'Brownian
Movement' of the clan groupings which can be discerned at the
226

FROM THE BANDAMA TO THE VOLTA
end of the nineteenth century. The Kran (Gere) and the Gio (Dan)
continued their advances towards the sea from what is now Ivory
Coast territory. Near Monrovia the Gola kept their hold on the
Malinke enclaves of Kondo and Boporo. However after 1883
Samory was extending his power into Loma territory (Toma) in
the north, and his traders visited Monrovia. This military threat
was soon to become stronger at the expense of the Gola who
looked to the government for support.
At this time the southern coast of Liberia and the western part
of the present Ivory Coast were inhabited by the ' Krumen' who
went on board ships as deck crew to do the heavy work on
merchant shipping during their stay in the tropics. They learned
to adapt to steam navigation and were to be found from Freetown
to the Congo. Up to the Bandama, Kru-speaking clans had little
to do with maritime trade, since the Neyo of Sassandra were
insisting on their right to the brokerage on their rivers. Some clans
removed up river as far as Soubre. However, in the interior, nearer
to the savanna, this inward-looking clan society witnessed the
birth of markets. The trade in kola for the Malinke of Seguela
was taken over by the Zebouo women from the Daloa region,
whose social status was thereby enhanced. Thus this old trade
from hand to hand was linked to the long-established Sudanese
network.
To the east of the Bandama, the Akan peoples had a tradition of

close connections with the outside world, which made them very
different from their Kru neighbours. On the Bandama estuary and
further to the east, between the lagoon and the sea, two related
peoples, the Avikam and the Alagya (Alladiahs), had set them-
selves up as brokers for the palm-oil producers of the interior.
They maintained regular links with the Liverpool ships that
travelled along the coast. A considerable proportion of the
population spoke English, and it was at this time that their great
men constructed the multi-storied houses, the ruins of which still
amaze the visitor. Their monopoly of the middleman trade had
been strengthened by the French withdrawal: since 1871 Chief
Aby of Grand Jack had been arresting on the lagoon the
steamers from Bassam which were attempting to buy their oil
directly from the Ajukru.
227

Of the two rivers, navigable on their lower reaches, which
provide a route northwards, the Bandama had been completely
blocked by the Baule since the beginning of the eighteenth
century. Tiassale, up river from the navigable reach, did not allow
middlemen higher upstream, and beyond that point trade was
carried on from hand to hand (exchanging salt for gold, ivory or
cloth), the only kind of trade which the Baule would tolerate.
Although monarchical in theory, in practice their lineages
governed themselves. Remarkably prolific and good at assimilat-
ing other peoples, they were only now fully populating the wide
savanna of the 'Baule reach' that they had first occupied in the
eighteenth century. The empty lands to the south had just been
occupied by the Aitu, and production had started at the gold mines
of Kokumbo. Trade was carried on mainly through the barter of
luxuries, without any cash changing hands, and markets were
unknown. These made their appearance in the extreme north after
1904 to barter yams for the slaves of Samory.
There was considerable traffic on the second river, the Comoe,
navigable as far as Bettye, even though it passed through difficult
forest country. This was because the Dyula of Kong were keen
to develop trade down-river. The area was occupied by organised
Akan kingdoms, the Agni of Ndenye, Bettye and Sanwi, the
Abure of Bonwa and Moosou, while the estuary of the river
belonged in principle to the French stations of the Cote d'Or (i.e.
the modern Ivory Coast), whose exchange for the Gambia was
under consideration. France withdrew from Grand Bassam and
Assinie in January 1871, and henceforth the French presence was
limited to the visits of warships and agents of the merchant
A. Verdier of La Rochelle, who in 1872 was appointed 'Guardian
of the Flag'. In 1878 he obtained the title of 'resident'. This
obnoxious character, whose only aim was to use nationalism to
further his own affairs, envisaging a monopoly, was less interested
in trade than in plantations and the exploitation of timber. After
1878 he neglected Bassam in favour of Assinie and the Aby
lagoon, planting coffee at Abima. His British rival, Swanzy, stuck
to trade and tried to establish direct links with the Asante, but
he was hindered by difficulties of navigation on the Comoe, mainly
due to the kings of Bonwa. However Dyula caravans did reach
the coast, coming from Kong or Bonduku, and thus from the
Abron kingdom, and crossing the Ndenye. In Sanwi, the old king
228

Amon Ndufu (1844-86), was ending his reign at Krinjabo on

good terms with Verdier, whose only merit was to represent a
French presence that was weak and useless. This period came to
an end in February 1885 with the arrival of the resident Charles
Bour, who received a cool welcome from Verdier and was quickly
replaced by his agent, Treich Laplene, and finally in 1886 with
the transfer of the trading stations, which hitherto had been
dependent on Gabon, to the Rivieres du Sud under Bayol. In 1889,
following Binger's journey, this tiny possession was to gain
importance as a maritime outlet from the Sudan.
Populated by the same ethnic groups, however, the British
colony of Gold Coast was very much more important, even
though it was still a dependency of Sierra Leone, whence many
Creoles such as Horton had migrated. Under their influence and
that of the Protestant missions, the Fante and the Ga from the
coastal area, who had been trading with the outside world for
centuries, began to modernise. They were soon to produce
remarkable intellectuals, such as Mensah Sarbah, make contact
with the American Negroes and join the Pan-African movement.
As early as the 1860s they were critical of the British government,
whom they accused of giving them insufficient protection against
the Asante, who in the 1860s were renewing their military efforts
to gain direct access to the coast.
One of the British problems was their inability to find revenue
from customs duties, the problem being that their coastal forts
were interspersed with Dutch ones. In 1867, therefore, they were
glad to conclude an agreement with the Dutch whereby forts were
to be exchanged so that the Dutch held forts only on the western
half of the Gold Coast and the British only on the eastern half.
This arrangement broke down, however, because of resistance to
the transfer from some of the Fante on whose lands the forts were
situated. Two consequences followed. First, the Dutch decided
to withdraw altogether from the Gold Coast, and in April 1874
the Dutch forts were taken over by the British. Secondly, Fante
intellectuals, objecting to the neglect of their interests by the
Europeans and wary of British intentions, attempted to set up a
modern state, the so-called Fante Confederation, which, without
breaking with Britain, had independence as its eventual aim. With
its king—president installed in the religious centre of Mankesim,
its three successive constitutions, and its bureaucratic administra-
229

tion, the confederation ended in failure, due to British intran-
sigence, prompted by hostility to these ' black intellectuals', and
the military threat posed by the Asante. In May 1873, with war
going on, the confederation was dissolved to make way for the
new Colony of the Gold Coast. In 1869 the Ga too, but with less
success, had sketched out a constitution for an 'Accra Native
Confederation'. In 1873 the asantehene; whose forces had been
engaged in military excursions on both sides of the Volta since
1869, launched a large army against the Fante. Disraeli's Tory
government then decided to respond by sending a strong
expedition under the command of Sir Garnet Wolseley. The
Asante were decisively beaten, and Kumasi was occupied in
February 1874. In accordance with instructions, the town was
immediately evacuated, and a peace treaty was hastily signed on
12 February at Fomena, on the way back to base.
It was then, from July 1874 onwards, that the Gold Coast was
finally separated from Sierra Leone and established as an
autonomous colony, controlling Lagos as well until 1886. Now
that security had been restored, political stability ensured the rapid
development of education and the economy (palm oil and gold
mines) under its governors, G. S. Strahan (1874-6), S. Freeling
(1876-8), Sir Samuel Rowe (1881-4), and W. Brandford-Griffith
(1885-9). The development of the Fante was accelerated. The
capital was transferred from Cape Coast to Accra. In this first
phase of imperialist advance, men like Rowe and Brandford-Griffith
were anxious to extend the territory, as Fante opinion demanded,
but were firmly prevented from doing so by London. This
explains the ambiguity of their policy as regards the Asante. Since
they were not allowed to annex the country, they feared that the
weakened Asante Confederation might regain its strength, and so
they spent their efforts in dividing it. But this created serious
disturbances harmful to trade, creating the wish that the Asante
might be strong enough to maintain order in the interior. In 1874
their confederation was in ruins. All their Brong allies to the east
and north had defected, as had their vassals in Gonja, Dagomba,
Abron (Bonduku), and Ndenye. Even one of the member states,
Juaben, had seceded. Kofi Karikari, asantehene since 1867, was
destooled and replaced in September by Mensa Bonsu. Juaben was
recaptured, followed by Adansi in 1879, and to calm the fears of
the British, the asantehene sent the golden axe to Accra in 1881.
230

FROM THE VOLTA TO THE NIGER
Captain Lonsdale's visit to Kumasi the following year did not

prevent another outbreak of trouble. Mensa Bonsu was over-
thrown in 1883, but his successor Kwaku Dua died immediately.
The Asante chiefs were then unable to agree on a successor and
an interregnum of four years followed before the election of
Prempeh in March 1888.
To the east of the Volta, the domain of the Aja was not
homogeneous, for despite a deeper unity of culture, the great
kingdoms of the east (Dahomey, Porto Novo) differed from the
petty chieftancies in the west, in Ewe territory. However in this
area bigger political units had emerged, notably on the coast, the
Anlo along the Volta and the Ge (Mina) kingdom of Glididji,
overshadowed by the commercial towns of Porto Seguro, Agwe
and especially Anecho, dominated by the very acculturated
dynasty of the Lawsons. When in 1874 the British occupied Keta
to the east of the Volta, they had established control over the Anlo,
whose alliance with the Asante was a cause for concern. In their
estimation they had natural rights over the other Ewe further to
the east, but once more their priority was to be eliminated by
prohibitions from London.
Since i860, and especially since 1868, French and German
establishments (Regis and Fabre from Marseilles, Vietor from
Bremen) had had trading stations at Bey (Lome), Porto Seguro,
and Grand Popo. Khadjovi, the chief of Anecho beach, had in
1882 signed a commercial treaty with the Germans. This treaty
was endangered the following year, when the death of the king
caused a serious crisis. The struggle was won by the ally of the
British, William Lawson, who had one of his relatives proclaimed
king, but the German ship Sophia immediately arrested him,
confirmed the treaty and carried away hostages to Germany.
These hostages asked the Kaiser to establish a protectorate over
the country. Returning on the Move in the company of Consul
Nachtigal, the hostages arrived at Anecho on 2 July 1884, just
when the British district commissioner at Keta had incited
Lawson to drive the Germans out. The protectorate was pro-
claimed that very day, then at Lome on the jth, where it was
recognised by Mlapa, the chief of the Togo community. Lawson
231

adhered to it, and the German merchant, Henri Randad, was
appointed provisional consul. He installed himself at Baguida.
However the colony was not properly organised until after the
Berlin Conference, when Ernst von Falkenthal, the imperial
commissioner, negotiated the Togo frontiers with Great Britain
and France. By this unforeseen intrusion Germany had inserted
its presence between the British Gold Coast and Dahomey, long
the object of French attention. In response to Nachtigal's action,
the latter had in July 1883 proclaimed their protectorate from
Grand Popo to Porto Seguro, of which after Berlin they kept
only Grand Popo and Agwe, arbitrarily split off from Anecho.
The major aim of these isolated positions was to keep the
Germans away from Dahomey. They only became important
after the fall of Abomey, whose territory separated them from
Porto Novo, the main area of interest for the French.
The small kingdom of Porto Novo (Hogbonu) had rejected
French protection in 1864, but Regis had considerable economic
interests there. However the influence of Lagos was preponderant,
and was strengthened by the presence of the Revd T. J. Marshall,
a Yoruba pastor. The British could not easily surrender this
position at the end of the Lagos lagoon, since it would permit
the Egba of Abeokuta to trade outside their control, as the Asante
did at Assinie. Mekpon (king of Porto Novo 1864-72), although
hostile to France, had nevertheless persecuted the Protestants, so
that he was deported and exiled to Abeokuta. His successor, Mesi
(1872-4), was on the other hand pro-British, persecuting the
Yoruba. On his death the French party took control with Tofa
(1874-1908). He immediately expelled Marshall, but was soon
confronted by British intervention. In 1879 the coastal part of the
kingdom seceded, and the British occupied Ketonou and the
channels leading to Cotonu. Threatened with a blockade, Porto
Novo recalled the French, who installed a guardian of the flag at
Cotonu in 1882, and re-established their protectorate over the
kingdom on 25 July 1883. It was to be administered under Gabon
until 1886, then under the Rivieres du Sud. Colonel Dorat, the
resident, then took energetic measures against the friends of the
British, and at the beginning of 1885, following the Berlin Act,
Lagos evacuated all its positions, in return for the renunciation
by Porto Novo of its claims in the east. The Franco-British
frontier was drawn in 1889. Porto Novo was a nuisance to Lagos,
232

but the principal aim of the French was to surround the powerful
kingdom of Abomey, and force it to surrender Cotonu, the only
means of access to Porto Novo apart from Lagos.
Dahomey (Danhome), which had reached its peak in the time
of Gezo, had difficulty in adapting to the suppression of the slave
trade, which ended about 1865, and during the reign of Glele
(1859-89) withdrew into conservative policies. After 1875 the
ageing king yielded the reality of power to his son, the vidaho
Kondo, the future Behanzin. The latter was ambitious and
energetic, and immediately provoked a crisis by arresting the
agent of Swanzy at Whydah, thus causing the British fleet to
blockade the port in 1877. After Regis's agent had agreed to pay
the fine, the grateful Kondo permitted the signing in 1878 of the
Treaty of Whydah, apparently renewing the 1868 cession of
Cotonu. This was certainly a misunderstanding, as Dahomey
merely intended to cede a commercial depot, without yielding
sovereignty as the French claimed. As the only convenient outlet
for Porto Novo, Cotonu was effectively occupied in 1882, but the
agents of Dahomey were not expelled, so that this difference of
interpretation was not immediately obvious.
Whydah (Glehove), the residence of the chacha and theyevogan,
' the minister for the white man', was the only centre of European
activity, with French and British firms, Catholic missionaries and
a Portuguese fort. The Governor of Sao Tome, J. de Souza,
anxious to establish the rights of Portugal in the follow-up to
Berlin, obtained in August 1885 - w i t h the complicity of the
chacha — the signature of a treaty setting up what he saw as a
Portuguese protectorate over Dahomey. When, as a result of
French protests, Kondo realised what was happening, he had
the chacha arrested, and presumably executed, and in March 1887
broke off relations with the Portuguese envoy. In December
Lisbon officially renounced its claims.
Once it had been deprived of the slave trade, this military state
tried to turn its energies to the export of palm oil, with the result
that the use of European goods became more widespread. The
king maintained his superiority by working vast plantations
through the labour of slaves who could no longer be exported,
but who continued to be captured each year during the dry season.
Military activity did not cease; it was directed year after year
against Middle Togo (Atakpame in 1877), the Mahi and especially

WESTERN AFRICA, 1870—1886
the Yoruba. This met with failure at Abeokuta in 1873, but the
first destruction of Ketu took place in 1883, and the final one in
1886. Kondo was politically conservative and thus was unable to
end the practice of human sacrifice at the annual festivals.
However he did try to modernise the army, buying arms and
obtaining mercenaries from Togo. In any case he was not the sort
of man to give up his sovereignty when the French sooner or later
put him to the test, since, following the Berlin Conference,
Dahomey was definitely within their sphere.
Along the coast of the future Nigeria, from Porto Novo to
Mount Cameroun, the old British preponderance was never
seriously challenged, although even here London's opposition to
action hampered its agents on the spot. Occupied during the
course of the fight against slavery, Lagos (Eko) was at first a
dependency of Sierra Leone, and then from 1873 to 1886 of the
Gold Coast. When the fight against slavery had ended, the town
increased rapidly in size, attracting all the lagoon trade in palm
oil and the trade from Hausaland, even though trans-Saharan
trade was still flourishing. The Creoles of Sierra Leone contributed
to the acculturation of the Yoruba from whom many of them had
sprung. Amongst these was Bishop Crowther, in charge of the
Niger Mission. In order that the budget might be balanced
through the revenue from customs duties, the British at Lagos
tried to prevent the export of goods by other routes. This explains
their activities at Porto Novo and their policies with regard to
the hinterland.
In this area a second centre of acculturation had appeared at an
early stage with the 'creolised' Yoruba in the Egba city of
Abeokuta. Like Ibadan, this town had sprung from the fall of the
empire of Oyo, and Protestants had been active there since 1845.
There was a Yoruba language press, and in 1867 a modern style
of government was established there, the Egba United Board of
Management (EUBM). This was an experiment analogous to that
of the Grebo or the Fante, but in this case beyond the control
of the colonial power. The product of acculturation, the Egba
government had no intention of yielding to anyone. Attacks by
Dahomey were three times brilliantly repulsed by the army. In
1867 it set up customs posts to hinder Lagos' relations with the
hinterland. This led to a violent conflict with the British and
culminated in the expulsion of all the Europeans from Abeokuta.
2
34

In fact the Egba were the natural allies of the Ijebu, who
occupied the northern shore of the lagoon to the east of Lagos,
and strove to prevent the colony from suppressing their middle-
man trade by opening up the routes into the hinterland and
trading directly with their enemy, Ibadan. The conference called
by Sir John Glover, the administrator of Lagos, in July 1871
opened up five routes, but did not put an end to the causes of
the crisis. Since the fall of Oyo, the divisions among the Yoruba
had not been healed. They had been torn by civil wars since 1810,
causing many of them to be sent as slaves to America. The alafins,
Adelu (1859-75) and Adeyemi (1876-1905), retained only local
authority, but it was in their name that the ' empire' of Ibadan
had arisen since 1845, which had limited the disaster by controlling
the entire northern part of the Yoruba lands, and thus blocking
the progress in that direction of the Muslims, now masters of
Ilorin. To the east Ekiti had been freed from the power of the
Muslims of Nupe and the empire of Benin. But to the west and
south the stubborn resistance of the Egba and the Ijebu had tried
to bar the way to Porto Novo and Lagos. The old religious centre
of Ife affected to remain neutral.
Ibadan had stopped the decline of Yoruba society, and had
rebuilt it according to a new set of values, those of the military
chiefs. But its domination was at times heavy. Momoh Latosisa,
who ruled Ibadan with the title of are-ona-kakanfo from 1871 to
1885, tried to re-establish Yoruba unity by bringing Egba and
Ijebu into line. He ran into the opposition of Oyo and Ife and
the revolt of his Ekiti subjects. Following various incidents,
during which the Egba closed the road, the are attacked them in
1877, thus opening the 'Sixteen Years' war', which was to end
in 1893 with the imposition of British rule. In the east he tried
to open the route to Ondo, his means of buying gunpowder from
Benin, but was soon threatened by the revolt of the Ekiti,
supported by the Muslims of Ilorin. The Ekiti-Parapo Confed-
eration, constituted the following year, lasted until 1893. Oyo
asked in vain for the arbitration of Lagos, and in i88z Ife joined
the coalition, only to be instantly destroyed, thus leading the Ijebu
to make peace. The Egba, howeever, remained intransigent, and
the efforts of the Revd J. B. Wood, the administrator, ended in
failure in 1884. Unable to intervene directly, Lagos was thus
powerless to end this conflict which closed the trade routes. It
235

took many years and the circumstances of the 'scramble for
Africa' before London finally agreed to act. In the interval,
despite Abeokuta and the Ekiti, the empire of Ibadan had stood
firm, virtually complete and alone against all the rest.
Further to the east, to the north-west of the Niger delta, the
old empire of Benin and the kingdom of Itsekiri, were, like the
city-states of the Ijo and Old Calabar, within the sphere of the
British consuls of the Bight of Benin. Thus it was a region which
was not juridically under British sovereignty. Activity was purely
diplomatic, carried on through the medium of the 'courts of
equity', except for the occasional intervention of warships;
everything depended on the personality of the consul, the most
famous of whom was Beecroft. The residence of these consuls was
on the Spanish island of Fernando Po until 1875, when it was
transferred to Old Calabar at the mouth of the Cross river. From
1880 to 1885 Consul Edward H. Hewett established British power
by means of numerous treaties, and the Protectorate of the Niger
Districts was proclaimed on 5 June 1885.
The old empire of Benin had, since the eighteenth century,
responded to the slave trade by withdrawing into itself, and it had
been weakened by the domination of Ibadan over the Ekiti.
During the time of Oba Adoro (1848-88), this isolation and
distress in the face of a changing world seem to have encouraged
a conservative attitude and the hardening of certain customs such
as human sacrifice. This was bound to lead to catastrophe, which
came in 1897. The mouth of the Benin river was in the territory
of the Itsekiri, with whom the consuls had close links. It was
acting-consul Easton who in 1879 appointed Olumu as 'gov-
ernor of the river', preparing the way for his son Nana's
appointment in 1884. While cleverly playing the imperial card,
the latter tried to outwit the new order in his opposition to the
traders who were ruining the middlemen. He was deported in
1894.
The classic domain of the consuls, however, was the city-states
of the Ijo to the east of the delta, a land of uncultivable swamps,
which acted as an outlet to the sea for the Ibo country, and in
which centuries of the slave trade had transformed village lineage
groups into powerful commercial structures, the canoe-houses.
Direct access by European commerce to the Niger and the
hinterland would have the effect of ruining these houses, and they
236

opposed it violently. Now it was precisely this policy that the
consuls had to defend. After 1875 the depression in world prices
resulted in. the merchants such as John Holt wishing to rid
themselves of the middlemen, and the crisis broke out. In 1875-76
the people of Brass attacked the steamers that wanted to cross the
delta. Having become the head of one of the greatest' houses' of
Bonny, Jaja, a former Ibo slave, had broken away in 1870 to found
the kingdom of Opobo. For opposition to commercial activity in
his hinterland, he was deported in 1887 by the consul, H. H.
Johnston. At Old Calabar in the fertile soil of the Cross river, a
monarchy of royal lineage dominated by secret societies cultivated
oil palm plantations with the aid of a large number of slaves, whilst
being open to acculturation as a result of the presence of
Protestant missions since 1846. Civil war between the Duke and
Henshaw factions for the two titles of obong and eyamba was
recurrent (1875, 1880, 1885). All this activity was restricted to the
coast. The hinterland was still unknown, especially the huge Ibo
population, scattered in independent lineage groups between
which links were maintained through the slave-trading network
of the Aro Chuku oracle.
The lower course of the Niger was the only easy way into the
interior, and this route had been well known since the great
explorations of the 1840s. It was thus natural for trade to turn
in this direction to break through the screen of middlemen on the
coast. Onitsha, the centre of a minor Ibo monarchy, and Bishop
Crowther's base, naturally attracted the merchants. In 1878 there
were four British companies there, to which were soon added
French and German ones. Faced with the sudden advance of the
French, the British firms from 1879 onwards were amalgamated
by an enigmatic Manxman, the famous Sir George Taubmann
Goldie, whose dreams were of empire rather than trade. His
National African Company caused the collapse of the French
traders by undercutting them, and he bought them out in 1884.
He concluded treaties with dozens of African states and possessed
thirty-seven armed ships and a private militia. The Berlin Act had
constituted the Niger basin into a free-trade zone and thus posed
a threat to his monopoly, but in 1886 he obtained the Royal
Charter which transformed his National African Company into
the Royal Niger Company, and which made it until 1899 the main
agent for British penetration into northern Nigeria.

It has seemed advisable to survey at one go the whole zone of

coastal acculturation from the Senegal to Mount Cameroun at the
outset of the first imperialist advance. This advance merely
confirmed the already high degree of acculturation. The pene-
tration of world market forces popularized European products
and ruined the local artisans. The response of the Africans was
often very positive, as is shown by the rapid development of the
groundnut and palm-oil industries in areas that were still politically
independent. Acculturation had already advanced far in Senegal,
and especially in the area of anglophone culture, where the first
generation of African intellectuals originated. Their influence
could already be seen in the rise of the modern nationalism of the
Grebo, the Fante and the Egba. In these three examples the
modern demand for political independence was fused with the
traditional aim of keeping control over one's own destiny. In the
circumstances of the time this was not enough to prevent the
establishment of the colonial era.
FROM THE UPPER NIGER TO THE UPPER VOLTA
Before tackling the Sudanic zone, which at that time was still
intact and orientated towards the Sahara, we should first examine
the intermediate zone, which from the upper Niger to the Volta
basin was already subjected to strong influences from the coastal
zone which were to bring about great upheavals. From the upper
Niger to the fiandama, the land of the southern Malinke was
plunged in 1870 into the whirlpool of the Dyula revolution, which
was to give rise to the empire of Samory. In this long-inhabited
region, the Malinke had spread as far as the rain forest, where the
trading minority group of the Dyula went in search of kola along
the borders of the 'barbarous' Misi, Toma, Kpelle (Guerze), Dan
or Guro peoples. The political unit was the small territorial
chiefdom (kafu or nyamaana). Above these, since the eighteenth
century, there had been only military hegemonies on a rather small
scale and of an ephemeral nature. However this was not a
homogeneous area. To the south it fronted the rain forest, which
was impenetrable, except to a slight degree in the centre, from
Konyan to Monrovia by way of Loma territory. In this cul-de-sac,
local trade was linked to the transport of kola towards the valley
of the Niger. On the other hand, beyond the upper Niger to the
238

west, by way of Futa Jalon or Solimana, the Rivers Coast was close
at hand, and the Dyula took part energetically in the international
trade which, since the abolition of the slave trade, was concerned
with gold from Bure and with ivory and cattle. They could be
found all along the coast from Boke to Freetown (Kempu) and
Monrovia (Dukoro).
Since about 1830 this stable system had been shaken by the
refusal of the Dyula, who were traders and Muslims, to accept
the place originally assigned to them. This was due to the
increased trade with the coast, which made it possible for their
numbers to grow and which made them indispensable as traders
in arms and cloth. Another factor was the influence of the Fulani
revolutions (Futa Jalon, Masina, al-Hajj 'Umar) which had
awakened their Muslim consciousness. Large areas were succes-
sively laid waste in the upper Toron and Konyan by Mori-Ule
Sisse from 1835 onwards, in the lower Toron by the Berete from
1840, and in Nafana (Odienne, in the Ivory Coast), where in 1845
Vakaba Ture founded Kabadugu (Kabasarana). Far to the west
these were the examples of Fodi Drami in Sankaran around 1870,
and of Mori Sulemani about 1875 among the Kisi. In the east,
also in 1875, there was Hedi Mori of Mankoro (Ivory Coast).
Animistic society was too divided to be able to resist the aggression
of a Muslim minority, open to the outside influences and
possessing a wealth of guns and military and military experience.
Therefore it accepted reformation when a man was found who
was sufficiently acculturated to be effective, but still close enough
to traditional society not to have to destroy it. This was the case
with the sons of Mori-Ule and with the Berete, to a lesser extent
with Vakaba also, but above all it was the case with Samory. He
was born about 1830 in lower Konyan (Beyla, Guinea) in a
sedentarised Dyula family which had partially abandoned Islam.
He was at first an itinerant trader, and then a soldier with the Sisse
and the Berete, before putting himself at the disposal of his
'uncles', the Kamara animists, in 1860-1. After helping the Sisse
to defeat the Berete, then fleeing from the former to the Loma,
he established himself at Sanankoro on the upper Milo from 1867
to 1870.
Samory returned to the Islam of his ancestors, but without
passion, and built up a private army owing personal loyalty to
himself. It was made up of levies from Konyan, as well as
2
39

adventurers from all parts. After crushing the animists of Toron

with the consent of the Sisse (i 871-4), he responded to the appeal
of the great city of Kankan, the ruin of whose Kaba chiefs,
adherents of the Tijaniyya, was brought about by the animists of
Sankaran. Having conquered the latter, Samory kept their land
and extended his power on the upper Niger from Kouroussa to
the gold mines of Bure (Siguiri) from 1875 to 1878. Near
Dinguiray he was stopped by the Tukolor of Aguibou Tal. The
Kaba then refused to support him, and allied instead with the
Sisse, who were worried by their isolation. In 1879 the Sisse broke
with Samory and advanced into Sankaran. From 1880 to 1881 in
a series of brilliant campaigns Samory crushed them, took
Kankan, and attracted to his cause his relatives, the Ture of
Kabadugu (Odienne).
At this time Samory was unchallenged at the head of a veritable
empire, which continued to expand first among the Malinke until,
in 1883, Bamako was reached, from which he pushed back the
Tukolor towards Segu, then to the south as far as the forest, and
finally in 1884—5 westwards, reaching Limba country on the route
to Freetown, and eastwards as far as Bagoe where he came up
against the forces of Sikasso. The new state did not entirely
overturn the ways of the older society, it provided better
conditions for the Dyula and the Muslims, and brought peace to
a huge area. For the first time the region was managed by a
superstructure imposed on the small city states or kafus, which
however kept their autonomy. The territory was divided up
between great armies, largely autonomous in respect of recruit-
ment and provisions, while Samory accumulated supplies at the
centre (Juruba) around Bissandugu, his new capital built in the
recently conquered Toron. Since the capture of Kankan, Samory
had been surrounded by Muslims, and was now turning towards
stricter religious practices. In 1884 he surrendered the title of
farma (powerful) in favour of that of almami in imitation of Futa
Jalon. Having been empirical up till then, he allowed ideology
to control him for a time. In 1886 he proclaimed his aim of
imposing Islam on the entire empire, which combined with the
siege of Sikasso (1887-8), led to the catastrophic revolt of Ban-Kele
(1888-90).
As a conqueror Samory was harsh but clear-sighted, not
particularly savage and no worse a slave trader than his contem-
240

poraries. The political edifice he constructed was dictated by the

crisis in Malinke society, and although the remote influence of
Europe had some effect on it, it had been formed according to
a properly African model. In 1882 he appeared to have triumphed
and his work seemed destined to endure. Yet it was at this moment
that his prospects were ruined by the abrupt entry onto the scene
of the French on their way to the upper Niger. Borgnis-Desbordes,
prevented by yellow fever from reaching Bamako that year,
made a raid on the Samorians whom he attacked at Kenyeran.
Although this was a decisive failure, he had given Samory a
chance to appreciate the military superiority of the white man.
The lesson was even more painful at Bamako in 1883. Samory,
whose empire continued to increase, even though confronted in
the east by the people of Sikasso, tried to avoid conflict. This
proved impossible, and in 1885 Col. A. V. A. Combes attacked
him in Bure. Samory's rapid and energetic response was the start
of one of his most brilliant campaigns, in which the French,
badly led, only escaped thanks to their superior weapons. The
Samorians besieged Nyagasola and advanced as far as the neigh-
bourhood of Bafoulabe, causing panic in St Louis.
Samory, however, felt that he was surrounded, and decided to
turn all his efforts to resolving the problem of Sikasso. To do that
he had to treat with the French. So in 1886 he fell back before
Frey, who, being anxious to return to Senegal to do battle with
Mamadu Lamine, granted him the favourable terms of the treaty
of Kenyeba-Kura in May 1885. The terms of peace were made
more onerous the following year by M. E. Peroz at Bissandugu
in April 1886. In particular, without his being aware of it, Samory
signed a clause which would be interpreted as the acceptance of
a protectorate. Thus he had mortgaged his independence when
he marched on Sikasso, to a major battle that was to have a
disastrous outcome. Samory recovered from this catastrophe and
struggled on skilfully and courageously in the upper Ivory Coast
until 1898. But his achievement had lost its meaning on the day
on which, instead of rebuilding his society, his only aim had been
to gain a breathing space before the inevitable final defeat.
To the east of the Bagoe and the Bandama, the Senufo, who
spoke the most westerly of the Voltaic languages, were solid
villagers with a truly stateless tradition. However, in the eighteenth
century, swallowed up by the Dyula empire of Kong, and
241

acculturated as a result, they had begun to form small military
states, such as Benge, Korhogo, Sinematyali or Nwele in the Ivory
Coast, and Kenedugu, later the kingdom of Sikasso, in Mali. The
latter was to arouse a truly nationalist resistance to Samory. After
the great reign of Daula (1840-60), who had almost destroyed the
Wattara of the Gwiriko, the kingdom had collapsed into a state
of complete anarchy and Tieba Traore, elected about 1868, had
to reconquer it step by step, starting from his mother's village,
Solokaa (Sikasso in the Dyula language).
Nevertheless this conquest was not total, for about 1868 some
former sofas (slave soldiers) of al-Hajj 'Umar had broken with
Segu and constructed Fafadugu, a strong military state on the
right bank of the Bani around Kinyan. In the south, on the other
hand, Tieba advanced further than Daula. He vanquished Nwele,
and with the support of Korhogo attacked Sinematyali, but
without success. He had reconquered the country as far as Bagoe
in 1883, when to his great annoyance Samory's army made
its appearance in a zone which had previously been politically
divided. In 1885, therefore, he used a Bambara revolt as the
occasion to take the offensive, but his army was crushed at Koroni,
and Samory's response was delayed only because his army was
engaged by Combes. Tieba however knew what to expect and he
used the time to fortify Sikasso as fast as he could in view of the
great war which indeed broke out in 1887. This stable state, based
on the solid mass of the Senufo peasantry from whom the army
was recruited, nevertheless had a Muslim minority, and the
Traore family, allied to Segu through their hatred of Fafadugu,
were won over to the Tijaniyya. The family moreover proudly
considered itself 'Dyula' and thus set in motion a process of
alienation which was to develop further in the colonial era.
In the extreme south of Senufo territory, the Tagwana and the
Jimini, whose political structure was weak, had within their
borders Zerma (Zaberma) and Hausa traders, the forerunners of
the warlike migrations into the land within the bend of the Niger
from the 1860s onwards. About 1883—5 a young trader from
Saxala, Mori Ture, began a systematic series of aggressive inroads,
on the pretext of a succession quarrel. He failed, and had to take
refuge among the Baule at Marabajaza about 1890, but he had
opened the way to the east for the Samorians. Having inherited
the vassals of Kong, Sikasso had in the middle of the century
242

confronted the principal successor states to this famous empire.

In the south the Kpong-Kene, in the area round Kong, had been
much weakened since the loss of Jimini and the revolt of the
Pallaka, stateless Senufo, who blocked the roads and appeared
invincible. The main routes however remained open, especially in
the south in the direction of Grumaniya and the sea, and in the
north towards Bobo-Dioulasso. This town was the capital of
Gwiriko, where the authority of the Wattara had survived longer,
even to the extent of beating back Daula Traore in 1850. In spite
of brutal repressive measures, the farma, Ali Jan (1854—78), had
lost virtually all his authority over the Sanon, who had control
of the town, and his successor, Kokoro Jan, only controlled a few
villages. The real masters of the country were henceforth
'barbarians', like Amoro, the chief of the Tyefo of Numudaxa,
who were once more raising their heads.
Near the Volta, the Abron kingdom, which had been free of
Asante domination since 1874, was spending its days relatively
peacefully under Agyimani (1855—98). Pape, the chief of the
Akiton, had in 1881 destroyed the Banda, a Senufo people who
panned for gold, and who had remained loyal to Kumasi.
Unaware of the French threat, the Abron were trying to forge
links with Accra behind the backs of their former rulers. Lonsdale
was to pay them a visit in 1887. Further north the Kulango
kingdom of Gbuna (Buna), which was becoming ever weaker,
was unable to stem the rising tide of the stateless Lobi, a prolific
peasantry advancing from the east.
The Volta basin is dominated by the huge demographic mass
of the Mossi-Dagomba states, astride the frontier between the
Upper Volta and modern Ghana. Further to the east, the Gurma,
very different both in language and culture, were linked to them
by certain political traditions. These numerous states, of which
the most important were powerful and stable, had for a long time
to come to terms with the growth of their Muslim minorities.
More recently the aggression of alien groups like the Fulani of
Masina or the Zerma had posed a threat. In the far north the
kingdom of Wahiguya under Naba Yenda (18 51-77) had difficulty
in controlling the barbarian Samo and the Fulani of Balobbo, who
were engaged in warfare against the Tukolor of Bandiagara,
before withdrawing to the land of the Bwamu (Bobo-UIe). Under
Naba Baogo (1885—95), a state of anarchy developed which wag
243

to facilitate the occupation of the land by France. In the great
empire of Wagadugu the death of Naba Kutu in 1871 had given
power to his mediocre and unpopular son, Naba Sanem. Until his
death in 1889, Sanem struggled against his exiled younger brother
Bokari, the future Naba Wobgho, who would activate resistance
to the French. In the south, the state in which the Muslims played
the major role was Dagomba, where even members of the royal
family had been converted. As early as 1874 Na Abdulay
(1867-76) broke off relations with the Asante, at the same time
beating off Zerma attacks.
These Zerma had left their country on the middle Niger after
it had liberated itself from Sokoto in the 1860s. As horse-dealers
and formidable warriors, they hired themselves out as mercenaries
from Borgu in north Dahomey to the Comoe. A large detachment,
under the orders of Alfa Hanno, from 1874 onwards thus took
service, first with the Dagomba and then with the Mamprusi,
raiding for slaves especially amongst the Gurunsi, a collection of
peasant tribes covering the southern frontiers of the Mossi, and
taking in old enclaves of Muslim traders such as the Kantoshi and
the small kingdom of Ta (Dagari-speaking). About 1870 Alfa
Hanno quarelled with the Dagomba, and attempted to conquer
the Gurunsi on his own account. His successor, Kalare (1874—82),
used islamised people such as Musa of Sati in his support. But
from 1883 onwards the new leader Babatu, a good soldier but a
bad statesman, took over Sati and then Wa, crushing the allies
of his predecessor. When he was installed in Sati, he marched
westwards against the Boboko-Kirpirsi coalition, which defeated
him at Safane in 1885. His military instincts led Babatu on to a
chequered career, not lacking in setbacks, until 1897. The Zerma,
of whom there were few, despite a constant flow of young recruits
from the homeland, had intermarried with the people they had
conquered and enrolled them firmly on their side. Around 1890
a generation born in the country, totally ignorant of their native
language, swollen by conquered people who had been absorbed,
prepared the way for the ruin of this remarkable military
hegemony.
Between the Dagomba and the Asante there stretched the vast
empty spaces of Gonja, a country that was dominated by the
Gbanya aristocracy and long visited by traders, in particular the
kola traders going to Kano by way of Sokode and Djougou. In
244

FROM UPPER SENEGAL TO THE MIDDLE NIGER
the nineteenth century the great commercial centre was Salaga,
near to Kpembe, the political capital, through which passed the
Kumasi trade. Liberated from the Asante in 1874, Salaga was
weakened by the diversion of trade routes, towards Kete Krachi
and the lower Volta on the one hand, and towards Kintampo on
the other. This impoverishment, which affected mainly the
Muslim merchants, aggravated the tensions which were already
present in the town, and brought about its destruction in 1892
on the occasion of a succession crisis in Kpembe. Further away,
on the road to Hausa territory at Sokode, the most important
chiefdom of the Tern (Kotokoli), Uro Jobo II (1860-73), was
converted to Islam contrary to the wishes of his warrior aristo-
cracy. He attempted to use a monopoly of the trade in salt and
slaves in order to provide the wherewithal for the nucleus of a
permanent force of cavalry. The revolt of Bafilo precipitated his
downfall, and Uro Jobo III publicly apostasised. In 1888 he was
to receive the Germans. At the next stage, at the Djougou (Zugu)
caravanserai, the kings of Tyilixa, Nyora III and Kpeytoni IV,
were attempting to impose their will from the mountains of the
Tangba (Taneka) to Basila, but without much success. Like the
Tern, they had problems with the Zerma bands from 1880 to 1890.
Further north at Kwande, the Bangana Seko-Wonkuru
(1857-83) waged civil war against the Bariba of Birni, and then
intervened at Teme in the direction of Bembereka, frequently
raiding the Berba and Betammaribe (Somba) of the Atakora. In
this immense underpopulated territory of Borgu, the unruly
cavalry of the Wasangari princes now recognised the authority of
the king of Nikki only in theory and not at all that of Busa. Their
considerable skills as warriors had nevertheless allowed them to
keep their independence in the face of the Caliphate of Sokoto.
FROM THE UPPER SENEGAL TO THE MIDDLE NIGER
Apart from Senegambia, the Sudanese zone around 1870 was

relatively sheltered from outside influences. Trans-Saharan trade
was still very active, and was not to decline until after 1885. The
slave trade was still carried on across the desert, albeit at a slightly
lower rate, but the goods that were arriving from Tripoli were
largely European in origin. For the rest, the influences from the
Europeans in Senegal could already be felt in the west and from
245

those on the Gulf of Guinea in the east by way of the lower Niger
and the Benue rivers.
The Tukolor empire, already much weakened after the death
of al-Hajj 'Umar, stretched from the upper Senegal to Masina. On
the death of the conqueror, many regions rebelled, and were never
to be reconquered. Thus, the empire found itself reduced to four
main regions, which had difficulty in communicating with each
other: Masina, under the command of Tijani, the nephew of the
conqueror, who was based with the Dogon of Bandiagara, but un-
able to put down the rebellion of the Fulani incited by the Kunta
Abidin and by Balobbo, who in any case often fought each other;
the Segu country, the most firmly controlled of the regions despite
the hostility of the Bambara, and in which was the residence of
Ahmad b. 'Umar Tal (Ahmadu Seku), who took the title of khalifa
in 1873; Kaarta and Khasso as far as the banks of the Niger,
together with Nioro and Koniakari, entrusted in turn to various
relatives of the master; and lastly Dinguiray, the cradle of the
empire on the boundaries of Futa Jalon.
With Ahmad, cunning management had replaced prophecy.
The constant influx of Tukolor, with their claims to influence,
tended to keep the vanquished, for the most part the Manding,
in a position of subordination, thus preventing the stabilisation
of the empire. The talibs (disciples of the Muslim shaykhs)
despised the Bambara sofas and continued to pursue in Segu the
clan conflicts of Futa Toro, between which Ahmad exercised
skilful arbitration. This skill did not ward off repeated crises at
the heart of the system itself. Thus it took four years of warfare,
from 1870 to 1874, for Ahmad to defeat his brothers Abibu
(Habibu) and Muktar, who had rebelled at Nioro and Koniakari.
In 1878 Abibu was sent to Dinguiray, the governor of which had
been killed by the Malinke, where he entered into an alliance with
the French in 1887. Lastly in 1883, Muntaga rebelled at Nioro in
his turn, and Ahmad, who came to defeat his brother, stayed for
fear of the French who had meanwhile occupied Bamako.
In this tottering empire there was no possibility of pursuing
a bold policy, and the horizon was clouded by the growing menace
of France. Following his father's example, Ahmad had sent an
embassy to St Louis in 1874, and the governor had signed a
reassuring treaty promising arms, in particular a field gun which
was never sent. But the crisis became more acute on the upper
246

FROM UPPER SENEGAL TO THE MIDDLE NIGER
Senegal after 1876 with the aggressive Briere de l'lsle. Logo, allied
to Ahmad, was destroyed by the French in 1877 (Sabugire). Then
with the project for the Niger railway, the eastward advance,
which was the forerunner of the great imperialist drive, penetrated
into the heart of the Tukolor domain, causing its subjects to rebel
along the way. Bafoulabe was occupied in 1879, Kita in 1881,
Bamako in 1883. Briere de l'lsle sent Gallieni to Ahmad in an
attempt to make him acknowledge these losses. After being
rescued just in time from the Bambara who were attacking him,
the ambassador had to spend long months at Nango waiting for
Ahmad to sign an ambiguous treaty, which was immediately
disowned by Paris. Unable to act decisively, and threatened in the
south after 1883 by the forces of Samory, Ahmad could only
observe bitterly the piecemeal destruction of his empire by the
French. On 12 May 1887 he resigned himself to signing the treaty
of Gouri, which allowed Gallieni to inform the European powers
that the Tukolor empire accepted the protectorate of France. This
did not save him. His remaining territories were occupied by
General Louis Archinard from 1890 to 1894, and Ahmad, joined
by Alburi N'Diaye, went to Sokoto, his mother's homeland,
where he died in 1902.
Beyond Masina, the withdrawal of the Tukolor and the
weakened state of the Fulani had left the banks of the Niger open
to the Tuareg. In keeping with their traditions, these nomads put
pressure on nomads and peasants alike. Thus Timbuktu was in
the hands of the Tengerege, distant vassals of the great Iwilli-
midden (Ullimiden) of Menaka. Before his death in 1885, their
chief Fandanguma played an active part in the fight against
Tijani of Bandiagara. From Menaka the amenokal al-Insar, who
reigned from i860 to 1890, controlled the Niger bend from
Bamba to Gao and Tillabery. Towards the end of his life he had
to withstand the vigorous threat of the Ahaggar, who were
raiding the Adrar. About 1875 he was forced to accept the
secession of the tribes of the Niger bend, who formed an
autonomous confederation directed by the tengeregedesh.
The eastern Iwillimidden or Kel Dinnik were also retreating
in the face of the Ahaggar of the Air and the Kel Gress. The
amenokal, Musa Ag Bedal (1840-72), was killed in Air. His
successor, Mehamma Ag Alkumati (1872-1905), lost ground to
the Air and the Kel Gress in the fight which he carried on until
247

1875 for control of the Hausa territory of the Ader. Under Sultan
Mahama al-Bakari (1860-1903), the people of Air frequently
raided Damergu, advancing as far as Bornu. But in 1888 they were
unable to reassert their authority over the Teda of Kawar who
who had just liberated themselves.
Within the Niger bend the tengeregedesh had gradually taken over
the small Songhay states (Gorwol, Kokoro, Dargol, Tera), which
they were supporting against the Fulani of Liptako. However,
from 1878 to 1885 Tera rebelled against the amenokal Elu under
the command of Gobelinga, who called for assistance from Tijani
of Bandiagara. After his death, Tera was laid waste and had to
yield. The Fulani of Liptako, like those of Say near to Niamey
under the command at that time of Emir Abdulwaidu, remained
loyal to Sokoto, or more precisely to Gwandu, but they were on
the defensive, isolated from the caliphate by the Zerma rebellion
of the Kebbi and the Dendi (north Dahomey). The Zerma had
in fact liberated themselves under the command of Daud about
1855. Gwandu had made peace with them in 1866. With support
from the kings of Tebbi (Toga, 1860-83; Somma, 1883-1915),
they could never be conquered. The sgrmakoy of Dosso, Abdula
(1870-80) and Alfa Atta (1860-86), were at peace until the
invasion of Ahmad b. 'Umar. This is doubtless the explanation
of the exodus of young volunteers, lacking adventure in their own
country, to the Niger bend.
THE CENTRAL SUDAN
The vast caliphate of Sokoto was stable and relatively peaceful,

but there were hostile enclaves in several regions. These were
groups of pagan peasants who had been pitilessly raided, like the
Gwari, who had undergone virtual genocide at the hands of
Kontagora, or the Birom, Angas and other pagans of the plateau
who were subject to the raids of the emir of Bauchi, or hostile
Muslims, such as the Hausa emir of Ningi. On the periphery, old
Habe states managed to survive after losing part of their territory,
and intermittently carried on the struggle; such is the case of
Kebbi, which withdrew to Argungu, of Gobir at Tsibiri, of
Katsina at Maradi or of Zaria at Abuja. But the entire centre of
the empire, particularly the area from Gwandu to Sokoto, Kano
and Zaria, enjoyed henceforth remarkable peace and prosperity.
248

THE CENTRAL SUDAN
Whether free or captive, the peasants peacefully cultivated the

land, and craft workers produced in quantity cloth and leather
goods, which were sold as far away as Timbuktu, Mauritania,
Tripoli and Wadai.
Contrary to the impression given by Barth, the empire was not
decadent. Central authority was still strong, having the last word
in the appointment of emirs, and deposing them unhesitatingly.
The wa^trs travelled endlessly among the emirs to keep control
of their administration. However, circumstances had changed;
the army was still powerful, but it rarely had to fight. The era of
annual campaigns had well and truly passed. The revolutionary
nature of social organisation became less clear, and a new
aristocracy with a strongly established hierarchy developed. The
diffusion of the written language, Arabic, Fulfulde or, increasingly,
Hausa, was now an established fact, but no longer were great
writers produced like those in the early years of the century. Lastly
the new caliphs {khalifa) or emirs, the heirs of the conquerors, only
acceded to the throne when already advanced in years; thus their
reigns were short and they were disinclined to embark on military
campaigns.
In the capital, Wurno (since Sokoto was merely the main town),
Ahmad Rufa'i (1867-73), A b u B a k r b - Bello (1873-7), Mu'adh
(1877—81), and 'Umar b. 'All (1881—91) ruled in turn during this
period. Apart from a few campaigns against Maradi and Argungu,
Mu'adh and 'Umar were mainly concerned with fighting Gobir,
which had launched a considerable offensive by taking Sabon
Birni. All observers agree that during this period the people with
most influence were the waîrs, who were all descendants of
Gidado dan Laima, the companion of the Shehu. The wa^trs at
that time were first Ibrahim Khalifu (1859-^.1874) and then
'Abdallah Bayero (1874-86). Relations with the emirates,
appointments or depositions, were essentially their province.
At Zaria, the most strictly supervised of the emirates, Emir
'Abdallah was deposed in 1870 for having engaged in disastrous
warfare which had started without authorisation, and Emir
Sambo in 1888 for his powerlessness when faced with Ningi and
the treason of his galadima. At Kano, Muhammad Bello b.
Ibrahim, who was to be a great disappointment, was installed
following much trouble in 1886. The only emirate outside the
control of the wa^trs was that of Gwandu, the heritage of
249

'Abdallah b. Muhammad (Abdullahi dan Fodio), which was

governed first by Mustafa (1868-75) and then by Maliki
(1876-88). But its power, which extended over the whole of the
west, collapsed as a result of the Zerma revolt of 1855 and the
revival of Kebbi. It was thus reduced to impotence and had to
appeal to Sokoto when it needed to intervene on the Niger, for
example against Bussa.
Outside the central area which was under the strict control of
Sokoto, and in which the Fulani conquerors tended to assimilate
the vanquished Hausa, whose Muslim tradition they respected,
three groups of emirates deserve closer study. Firstly, in the east,
facing Bornu, Hadeija and Gombe were responsible for a frontier
which had not changed since the conquest of the Chad empire had
been renounced, but there were constant border incidents. The
population had been transformed by the influx of Fulani who had
left Bornu, but continued to pay respect to their former country.
Hammaruwa (Muri), on the other hand, which had split off from
Gombe in 1833, looked towards the territory lower down the
Benue, where it devoted itself to fighting the pagans. Emir
Muhammad Nya (1874-95) repeatedly attacked the Jukun and the
Tiv. But the latter, sturdy peasants with a segmentary society and
a Bantoid language, brought him to a halt in no uncertain fashion.
Secondly, to the south, the characteristic of the large states of
Nupe and Ilorin was the strength of the autochthonous pagan
population, which became only partially islamised and assimilated
linguistically the people who had conquered them. The emir of
Nupe bore the title oietsu, and had to take into account the former
royal families which, though vanquished, still wielded influence.
During the reigns of the emirs Masaba (1859-81) and 'Umar
(1881—97), several revolts of the indigenous population took place
(1872, 1881). In 1881 'Umar was supported by the English
merchants and his reign was to be dominated by his relations with
Goldie, who had him overthrown in 1897. Under Emir Aliyu,
Ilorin increasingly developed into a Muslim Yoruba state. Until
1883 Aliyu put his major effort into combating the empire of
Ibadan by supporting the Ekiti.
Thefinalexample, which merits close study, is that of Adamawa.
In this area, the Fulani dispersed in the upper valley of the Benue
were dealing with stateless pagan peasants, ethnically very diverse,
whom they called by the denigratory term of Kirdi. The Adamawa
250

THE CENTRAL SUDAN
Fulani were in conflict with them before the beginning of the

jihad, and though they willingly joined the jihad and always
recognised the authority of the caliph, their movement remained
autonomous. In effect, they had constructed a separate state,
Adamawa, covering the whole of northern Cameroun with the
exception of Mandara and the Kotoko territory, the ally or vassal
of Bornu, which stretched as far as the upper Sangha in what is
now the Central African Republic. To the south of the Kirdi, they
crushed the Sudanese monarchy of Mbum and subjugated or
drove back the Gbaya, whose historic drift towards Ubangi was
thereby accelerated. But here, unlike in Hausaland or Nupe, the
victorious minority so despised the vanquished that they rejected
any assimilation and, far from adopting their language, imposed
Fulfulde as the common language. As in certain emirates in the
west, this was an unstable frontier vis-a-vis the non-Islamic world
beyond. Here slave raids played an important part, either at the
expense of Kirdi enclaves, which had been only partially subdued,
or on the periphery in Gbaya, Tikar or Bamum territory.
Adamawa did not comprise one emirate but several, subordinate
theoretically to the emirate of Yola, which was the only one to
have direct links with Sokoto. During this period the emirs of
Yola were Lawal (1848—72), who so charmed his guest, Barth, and
Sanda (1872—90), who, shut up in his harem, was thought to be
weak and soft. But there was a dominant militarism, which clearly
distinguished Adamawa from Sokoto, and which derived from the
front line emirates, which only paid lip-service to the primacy of
Yola, even when they had not formally broken with it, like Rei.
In the north, Pette and Marwa (Radawa clan) stood guard against
Mandara, Bornu and the peoples of the Logone. Under Moharam
Sali (1840-96), Marwa continued to subdue pagan enclaves and
to contain the Mandara.
The south was to a large extent the domain of the Wollarbe
clan which held the important emirates of Tohamba, Tibati,
Banyo and Ngaundere. The celebrated lamido of Tibati, Mohaman
Arnga Nyambula (1851-72), ended peacefully a troubled life
noteworthy for violent conflict with Banyo and Ngaundere,
which had forced Lawal to come in person to conduct the siege
of Tibati. His son, Mohman Buba (1872-88), devoted his energies
to extending his domain southwards as far as the Sanaga, at the
expense of the Tikar and the Klute, of whom large numbers were

vanquished and enslaved. Under the sons of Mohaman Dandi,
Diko, Saso and Usmanu, the emirate of Banyo continued to attack
the Tikar, but further south it proved powerless against the
important Bamum kingdom. Having assimilated the Mbum,
Ngaundere - under Isa (1870-80) and Mohaman Gabdo (1880-
5) — devoted its energies to raiding the Gbaya and increasing its
hold over the upper Sangha, from Kunde on the Kadie to Berbera
(in the modern Central African Republic). From there, they
pursued their advance towards Ovesso (now in the Congo
Democratic Republic). The Fulani world, so different from of
Sokoto, was still expanding, a process that was only to be arrested
by the German intervention from 1899 to 1902. It was at this time
that the inegalitarian structures were established, dividing pagan
peasants from the Fulani nobility, and which were to dominate the
society of north Cameroun in the colonial period.
The area near Lake Chad was still the domain of the old empire
of Bornu, whose administrative structures had scarcely changed,
even though the Shehu dynasty had finally eliminated the Sefawa
dynasty in 1846. The new dynasty had largely maintained the
hierarchies and the complex system of titles of the previous court,
and transferred them to the new capital of Kukawa,. Shehu 'Umar
(1835—53 and 1854—80) was a prudent, even timorous man, who
allowed effective power to be wielded by his wa^tr, Laminu Mtjiya,
and, when the latter died in 1871, by his favorite son Bukar Kura,
an active man with plenty of initiative, who took control of
military matters. The western frontier with Sokoto was then
stable, and the southern one, covered by the alliance with
Mandara, was free from danger. Beyond the Chari, Bagirmi, from
where 'Uraar's mother had come, formed closer ties with Bornu
in order to withstand Wadai. Slave-hunting raids took place from
this direction and were a cause of great concern, since the export
of slaves to Tripoli still continued. To the east of Chad, Kanem
was under some degree of control, despite the Awlad Sulayman
Arabs who pillaged the area but kept out the influence of Wadai.
The hereditary governor of Mao by now wielded only symbolic
power.
Bukar automatically succeeded his father in 1880, only to die
in 1885, thus opening a grave succession crisis. His younger
brother Ibrahim imposed his rule by force. Without power before,
he had allied himself to a group of greedy young men who wanted
252

THE CENTRAL SUDAN
to supplant the holders of great titles, and began to assassinate

them. The nobility then overthrew him, and replaced him with
the weak Hashimi, a withdrawn man of religious tendencies,
indecisive and miserly. Moreover this was the time when the
trans-Saharan trade finally collapsed, a process hastened by the
raids of the Tuareg, the Teda and the Awlad Sulayman, who were
fighting over Kawar. Having fallen into a state of extreme
weakness, Bornu was unable to withstand Rabah in 1893.
The great Hausa kingdom of Damagaram, around Zinder,
arose in the nineteenth century among the sub-feudatories of
Bornu. Developing from a small eighteenth-century vassal state
which had grown at the expense of the older Sosebaki states,
Zinder freed itself thanks to the menace of Sokoto. Without
breaking with it, Bornu left it completely free, for the existence
of this armed power, arising to the north-east of Sokoto, had the
effect of making a Fulani offensive towards Chad impossible. This
new power was then ruled by Tenimu 1851—84) who built strong
fortifications round Zinder and constructed a modern army with
guns bought from Tripoli. After 1870 he continued to extend his
frontiers, annexing Munio in 1872 and the Tuareg of Kutus in
1873, and then liberating Gumel which had been crushed by the
Fulani of Hadeija in 1872. Having subdued the last Sosebaki and
Zongo, Tenimu began to raid as far as Kano. His son Seliman
(1884—93), who left a reputation for cruelty, carried on the
struggle against Kano, the Tuareg of Air, and the Kel Gress,
whom he took in the rear in their battle against the Iwillimidden.
On the eastern bank of the Shari, Bagirmi, under Muhammad
Abu Sekin (1850-77), had proved unable to reject the rule of
Wadai, whose kolak, 'Ali, took Masenya by assault in 1871. The
people of Bagirmi continued to make their living from slave
raiding in the south among the Sara (who spoke the same
language), especially between Ouham and Gribingui and into
areas now part of the Central African Republic. Some of the Sara
collaborated with them in order to reduce the damage, and,
imitating the Masenya, some of the Mbang tended to set up petty
chieftaincies more capable of resistance. Under Aburahmani
Gaurang (1883-1918) Bagirmi continued to enjoy the double
sovereignty of Wadai and Bornu, but this was of no help to them
when Rabah approached in 1891.
In the east of the territory, Wadai, which had been conquered

by Darfur about 1835-39, had reasserted its power. Its sovereignty
over Kanem and Bagirmi had somewhat weakened, but was still
total over the Bulala of Lake Fitri and the Daza of the Bahr
al-Ghazal, as it was over Borku and Ennedi, where the local
sultanate was of Tunjur origin. There were still slave raids
between Gribingui and the Kotto, an area that had been reduced
to semi-desert by the veritable genocide of the Gala and the
Kreish. The base for these raids was Dar al-Kuti, constituted
in the south of Aouak around Cha by the Bagirmian prince,
Jugultun.
Whilst it imitated the institutions of Bornu, Wadai had been
a strongly entrenched Sudanic kingdom, and the sultans of
Abeche, the ko/ak, normally spoke the language of the Maba, the
mountain peasants who were their supporters. They also drew
support from the black Salamat Arabs in the south and the
Mahamid in the north. But in the nineteenth century their Islam
was influenced by the Sanusiyya, whose effects could already be
felt in the reign of Muhammad al-Sharif (183 5-5 8), who attempted
to make use of the brotherhood in his struggle against Darfur.
It was then that the arabisation of the court began, a process that
was to continue during the colonial period. 'All (1850-74), the
son of Muhammad and also Nachtigal's host, fought with the
Mahamid Arabs of the north and the Arabs in the east, and
succeeded in re-establishing the power of Wadai vis-a-vis Darfur.
His brother Yusuf (1878—98) pursued policies that were mainly
defensive, linked ever more closely with the Sanusiyya, who with
their establishment at Kufra were demonstrating their intention
to expand into the Sudan. However Rabah was to capture Dar
al-Kuti in 1889, thus initiating the final crisis which ended in 1894
with the occupation of the country by France.
The influence of the Sanusiyya was then growing rapidly
throughout the Saharan and Sahelian zones inhabited by the Teda,
Tubu and Daza from Kawar to Ennedi north of Wadai. This
situation was to be reinforced in 1895, when the seat of the
brotherhood was transferred from Djaghbub to Kufra, out of
reach of the Ottomans. The Tomagra sultans of Kawar, who in
1877 had exchanged the sovereignty of Bornu for that of Air,
succeeded in liberating themselves in about 1870, and the counter-
attack by the Tuareg in 1888 was a total failure.
Most of Tibesti recognised the authority of the derde of the
254

THE CENTRAL SUDAN
Tomagra, formerly a vassal of Bornu, who cleverly made use of

Islam and his reputation as a rainmaker to impose his will on the
clans. After Tahorti Walli-Mi and Togoi-Kenimi-Mi, Chai-Bogar-
Mi came to the throne about 1885. He lived until 1939. This subtle
politician, who was skilfully to use his position to play off the
Ottomans against the French, began by supporting the Sanusiyya,
thus tending to a more radical islamisation of his compatriots, who
were lukewarm Muslims and totally un-arabised.
The main agents for the extension of the Sanusiyya in the
central Sudan were the Awlad Sulayman Arabs, who came from
Libya after being driven out by the Ottomans in 1842. Settled in
the north of Kanem, and paying mere lip-service to their
acceptance of the shehu of Kukawa, these formidable pillagers had
thrown the Sahelian region into utter confusion, attacking in
particular the Daza, a cattle-raising people. After 1870 a certain
equilibrium prevailed, however, giving the Daza the opportunity
to reoccupy their positions when the colonial era put an end to
the Sanusi episode.
Thus the immense area covered by this chapter was extremely

heterogeneous during these fifteen years studied. After the
extinction of the illegal slave trade, the whole of the coastal strip,
already strongly subject to outside influences, was rapidly
integrated into the world market. At the same time, although there
was no planned design in this, it was on numerous counts the
theatre of the first phase of the imperialist advance. In 1885 this
situation was stabilised for a few years following the Berlin
Conference, the 'loaded pause', but the first outlines of the
colonial boundaries can already be discerned. This is not the case
with the immense hinterland, stretching from the upper Senegal
to Lake Chad. The commercial influence of Europe was already
making an impression, but only spasmodically, and the trans-
Saharan trade links were to be weakened only after 1885. It was
in the southern zone, in the area nearer to influences from the
coast, that long-established societies, so far untouched, came to
witness such great upheavals as those of Samory or the Zerma.
These upheavals can be interpreted as a reaction to changing
conditions, notably the development of long-distance commerce.
In the Sudanese savanna zone, the upheavals had already taken
place at the beginning of the century, so that the period we have
255

studied was one of stabilisation and relative tranquility. This state

of calm was to be abruptly broken by the colonial boundaries
which were completed some fifteen years later, about 1900, but
in 1885 nobody could foresee this, least of all those involved. This
part of Africa was still living at the centuries-old pace of its own
evolution; it could not have imagined that its days were
numbered.
256

CHAPTER 5
THE LINES OF SIEGE, 1886-89
The outlook from the savanna

Although in retrospect the flurry of European diplomacy which
attended the Berlin Conference of 1884-5 seems clearly designed
to prepare for the European invasion of West Africa, economic
and political uncertainties in Europe and indications of stiff re-
sistance in Africa inhibited the formation of aggressive policies.
In the immediate aftermath the European siege lines were
strengthened, but the scale of the impending threat to African
independence was still not generally predictable. Even if in
retrospect it appears that external and internal pressures were
producing a general crisis of authority, its local manifestations
varied greatly in nature and in intensity, and it is not easy to trace
this crisis to common causes. Obviously, the impact of the
European economy and European power was strongest in coastal
regions. Many inland kingdoms of the savanna and Sahel still knew
Europeans only as isolated and powerless travellers, and they
retained the preoccupations and priorities which their historical
experience suggested. Bornu, for example, though affected by
fluctuations in European demand for ivory and ostrich-feathers,
had more urgent problems at home; the challenge which
eventually in 1893 overthrew the al-Kanam! dynasty came from
Rabah al-Zubayr, a well-armed state-builder from the Nilotic
Sudan, who proved capable of mobilising support among over-
mighty subjects of Kukawa.'
'Umar b. 'All, Caliph of Sokoto, 1881-91, also faced difficulties
in maintaining administrative control over the empire founded by
Usuman dan Fodio and in enforcing the theocratic standards
which justified its existence; but these were largely inherent in the
attempt to hold together territories which it took four months to
1
L. Brenner, The Shcbus of Kukawa (Oxford, 1973), 111-30.
257

WESTERN AFRICA, 1886-I905
2
traverse from east to west, once the capital of charisma accumu-
lated by the founders had become attenuated. This was no archaic
monarchy crumbling at the advent of European ' civilisation'; the
basic structures of the jihadi state were still intact, even though
some of the problems of authority had never been wholly solved.
A common system of religion, culture and law was widely
accepted; taxes, tribute and military contingents were still
rendered by the emirs to the twin capitals of Sokoto and Gwandu;
settled agricultural communities continued to sustain the long-
distance trade for which Hausaland was so widely known.
Certainly, security problems were posed by contests for power
within the emirates (which, as in Kano in 1893, might involve
challenges to the ultimate authority of the caliph),3 from the
simmering challenge of Mahdism, or from frontier conflicts with
non-Muslims continuing since the early period of jihad. Though
controllable, these were serious preoccupations; by comparison
the activities of the Royal Niger Company and other European
visitors still represented not so much a menace as an opportunity
to obtain supplies of those (obsolescent) firearms which these
foreigners were usually anxious to sell. Even in riverain emirates
like Nupe and Adamawa, where armed Europeans now appeared
regularly, the Niger Company did not yet appear militarily
dangerous, and the occasional French and German reconnaissance
simply enlarged the opportunities for manoeuvre. At Sokoto, the
arrival in June 1885 of the Scot, Joseph Thomson, cannot have
seemed a particularly portentous event. Clearly 'Umar did not
assent to the cessation of sovereign rights in the Niger and Benue
valley, as appeared in the treaties which Thomson took back to
his paymaster, Goldie; but he was willing and even eager to
negotiate about the future commercial status of the company,
not perceiving how this would ultimately weaken his state.4
In Segu, Ahmad b. 'Umar Tal (Ahmadu Seku), lacking that
religious charisma which could partially compensate for defective
administrative structures, had more reason to fear a direct
European onslaught on the extensive empire which his father had
2
D. M. Last, The Sokoto caliphate (London, 1967), 63.
3
R. A. Adeleye, Power and diplomacy in Nothern Nigeria, 1840—1906 (London, 1971),
97, 103 ; Adamu Fika, The Kano civil war and British over-rule, 1S82-1040 (Ibadan, 1978),
Ch. 5 .
* Adeleye, Power and diplomacy, 132—6, J. D. Hargreaves, West Africa partitioned, 1:
The loaded pause (London 1974), 99-104.
258

T H E L I N E S OF S I E G E , 1886-89
created but not consolidated. The advancing French army had

reached Bamako in 1883; although its officers might disagree over
methods, their intention to obtain control of the upper Niger
valley had been clearly signalled. Nevertheless, Ahmad found it
difficult to establish priorities between dangers from the French
and those from his own rebellious subjects and refractory vassals,
whom they encouraged. In 1885 Ahmad, having led an army
northwards into Kaarta to subject his half-brother, Muntaga,
decided to remain in that less vulnerable province, whence he
half-heartedly harassed the French flank. But from the end of that
year his feud with the French was suspended to allow both parties
to deal with the common threat from al-Hajj Muhammad al-Amln
(Mamadu Lamine).
When Lamine returned to Khasso in 1885 his religious prestige,
greatly heightened by his hdjj, appealed most strongly to the
Soninke (Sarakule) of the upper Senegal basin, some of whom
resented Tukolor political and religious hegemony. Initially the
priority of his preaching was against Ahmad, not the French, who
provided ' powder, shot, guns and munitions of war, as well as
paper '. s Yet France's progressive political dominance, with its
arbitrary demands for labour, had produced even deeper resent-
ment among many Soninke; officials perceived latent perils in his
movement, and in March 1886 Colonel Frey took the offensive
against him. This commitment implied a reprieve for the Tukolor
empire; in May 1887 Gallieni offered a new protectorate treaty
which Ahmad, judging Lamine's threat more immediate, found
it expedient to accept.
Potentially the most formidable West African state was
Samory's empire; here, unlike Sokoto and Segu, the 'Dyula
revolution' had not exhausted its initial impetus, although it now
met increasing African opposition. After his first encounters with
the French in 1882, Samory had switched some of his forces
southwards, conquering Falaba in 1884 so as to open alternative
supply-routes for his armies in Freetown and on neighbouring
coasts. Access to the Senegal valley and the Sahel still remained
important for him, especially in relation to the supply of horses;
his military opposition to the French in 1885-6 was effective
5
Muhammad al-Amln to Governor, September 1886; D. Nyambarza, 'Le marabout
El Hadj Mamadou Lamine d'apres les archives franfaises', Cahiers d'etudes africaines, 9
(33). '41-*-

enough to secure reasonably equal terms in the Treaty of Kenieba,
28 March 1886. Having made these arrangements with his
European neighbours, Samory resumed his primary concerns, the
expansion of his empire and the enforcement of stricter Koranic
orthodoxy. In November 1886 he proclaimed his intention of
prohibiting the consumption of alcohol and the practice of
animist religions; at the same time he prepared to move his armies
against Tieba, the Senufo ruler of Sikasso, who despite his
profession of Islam was encouraging Bambara resistance to
Samory in the upper Bagoe valley.6 These moves involved
relaxing his hold on the gold-bearing province of Bure; on 25
March 1887 Samory tried to buy security in the west by ceding
to France the left bank of the Niger up to Siguiri, and by agreeing
to recognise their protectorate over his other territories. But even
these sacrifices did not secure their objective; the French secretly
helped Tieba to defend Sikasso, and in 1888 Samory's growing
demands for tribute, supplies and conscripts for this campaign
produced widespread revolts among subjects already resentful of
forced islamisation.7 In this crisis Samory made yet more
concessions to France at Nyako in February 1889, hoping vainly
to win their support or acquiescence for his policies of Islamic
expansion.
During the late 1880s the major issue in the history of the West
African interior thus appeared to be how effectively the over-
stretched structures of Sudanic monarchies could respond to
internal challenges, which might take the form either of resistance
to islamisation or of Islamic reform movements. Although the
French army on the Niger already presented a grave threat to
African independence, its magnitude was restrained by political
and financial pressures within France. So long as French govern-
ments could not afford to engage more than one African state in
battle, Muhammad al-Amln's resistance reprieved the independ-
ence of his neighbours. As for the British, their long-standing
intention of 'reaching Hausaland' was still seen in commercial
rather than military terms, and the central Sudan still felt secure.
In Mossi, and other kingdoms of the Upper Volta, British and
German reconnaissances from the south were beginning to
6
Y. Person, Samori: sou Revolution Djula, 3 vols. (Dakar, 1968, 1970, 1975) is the
indispensable work. For the islamisation of the empire, vol. 11, 808-18, 887ff.;for the
campaign against Tieba, 747-49.
7
Ibid, 11, io49ff.
260

THE LINES OF SIEGE, 1 8 8 6 - 8 9
interact with those of the French, and Binger's epic exploration
of routes from the Niger to the Gulf of Guinea foreshadowed
dangers to come; but the scale of the European military menace
was not yet evident. Only on the frontiers of the existing
European coastal settlements could clear connections be demon-
strated between internal crisis of authority and those external
forces to which an increasing number of African societies were
becoming linked.
The outlook from the coasts
European participants in the Berlin Conference of 1884-5 hoped
that, by prescribing a code of behaviour for foreigners seeking
to control the coasts and rivers which gave access to African
commerce, they had stabilised their own imperial responsibilities.8
The new procedures for partition of the littoral regulated a
process of scrambling for protectorates which was already almost
complete; by 1855 only the stretch between the Mano river and
France's outposts on the Ivory Coast remained unclaimed by
Europeans. But, apart from collecting customs duties at the coast,
most Europeans still hoped to expand their influence by moral
suasion, to secure African collaboration with a minimum of
coercion. Only the French in Senegambia were prepared to
penetrate inland by military force. Lat Dior's death in the battle
of Dyaqle (26 October 1886) marked the consolidation of French
control over Cayor and warned other Africans of the dangers of
military resistance. His cousin Alburi N'Diaye of Jolof, and Abdul
Bokar Kan in Futa Toro, tried by astute diplomacy to protect
their own lands, revenues and authority, but could secure only
temporary reprieves. By 1890 the military rulers of Senegal,
worried by the continued migration of Muslims, felt strong
enough to eliminate these remaining enclaves of Senegalese
autonomy. Abdul Bokar was killed in Mauritania in August
1891 ;9 Alburi responded to the invasion of Jolof by emigration,
to join the fighting remnant of the Tukolor empire in its long
retreat.
Other Senegalese Muslims adopted policies of emigration a
rinte'rteur. The marabout Ahmad Bamba rejected the option of
continued military resistance after Dyaqle with the words, ' I
8
For development of this point, see Hargreaves, West Africa partitioned, i, 40-9.
9
D. W. Robinson, Chiefs and clerics (Oxford, 1975), ch. 8.
261

know God sent these foreigners to pacify the country;10 his

rapidly growing band of Wolof disciples became the nucleus of
a new tariqa, the Mouridiyya, which assumed the role, filled
elsewhere by traditional religious cults, of providing a new focus
of African identity during the period of adjustment to colonial
rule. Ahmad Bamba's initial persecution by worried administrators
only increased his following, and the Mourids were eventually
able to carry on by peaceful methods the work of uniting the
Wolof peasantry under clerical leadership and expanding their
territorial base which had been begun by warriors like Lat Dior.
Yet even where no such threat of European military
intervention was apparent during the 1880s, the stability of
Afro-European collaboration was illusory. In Europe the
'stabilising factors', which had permitted Britain to maintain a
general oversight of Africa's external relationships with a
minimum of political commitment, disappeared when other
governments came to see in Africa opportunities for resolving
socio-economic crises or gratifying political ambitions.11 And
within the West African coastal belt many states experienced their
own 'crises of authority' when to the eroding influence of new
religions and cultures were added sudden dislocations in the
external trade. Those who had most successfully adapted their
economies to the production of groundnuts or palm products for
export found these new staples unreliable sources of income.
Tensions arising from the commercialisation of land and labour,
which had been growing through the century, became acute when
the terms of trade turned against African producers, and when
fluctuations in world demand threatened the incomes of producers,
traders, exporters and colonial governors.12
While some members of the commercial community sought to
maintain their profits by restrictive practices, others, both Africans
and Europeans, perceived opportunities for innovations in
production, which would lead to still deeper involvement in the
international economy. It was during the 1880s that the farmers
of the Gold Coast first turned to the cultivation of cocoa. A more
10
Tradition recorded in 1965; V. Monteil, Esquisses Se'negalaises (Dakar, 1966), 148.
See also D. B. Cruise O'Brien, The Mourides of Senegal (Oxford 1971), part 1.
11
See G. N. Sanderson, 'The European partition of Africa: coincidence or con-
juncture?' in E. F. Penrose (ed.), European imperialism and the partition of Africa
(London, 1975).
12
A. G. Hopkins, An economic history of West Africa (London, 1975), ch. 4.
262

immediately remunerative response to external demand was to

increase exports of wild rubber. The consequences were no-
where so terrible as in the Congo basin (although they included
the murder of Temne rubber-tappers recruited to work for the
Royal Niger Company)13 but they could nevertheless have
disturbing effects on established relationships. Students of Asante
history, for example, see the traders and tappers of that state as
reinforcing a small but growing bourgeois middle class with
distinct interests and aspirations transcending loyalties and
allegiances of a traditional kind.14
To liberal development theorists of the Buxton school, such
social changes should have been welcome, as indications of
advancing civilisation. But the impact of the depression on
established producers of oil-palm or groundnuts was often more
violent. One reaction by the larger producers was to try to
increase their labour supply; their ' slave-raiding and plundering
expeditions', it has been argued, were at least a contributory cause
of the Yoruba wars which had begun in 1877 and — despite the
peace negotiated between Ibadan and the Ekiti in 1886 — continued
to disrupt commerce until 1893.15 Similarly, the desire of rulers
to eliminate 'middlemen' and obtain more direct access to the
sea-coast lay behind many of the wars in the neighbourhood of
Sierra Leone, which seemed to become more frequent and intense
in the 1880s.
Outside Senegambia, European governments still hoped that
their African neighbours would resolve their crises of authority
in some economically rational way, and resume mutually advan-
tageous collaboration. Military force might occasionally be
needed to administer a short sharp shock to a ' savage horde' like
the Yoni of Sierra Leone (1887),16 but conquest and control of
the African mainland still seemed prohibitively expensive in
IJ
J. E. Flint, Sir George Goldie and the making of Nigeria (London, i960), 147—9;
Hargreaves, West Africa partitioned, 1, 104-6.
14
Ivor Wilks, Asante in the nineteenth century (Cambridge, 1975), 7 0 3 - j ; T. J. Lewin,
Asante before the British: the Prempeanyears (Kansas, 1978); E. Dumett,' The rubber trade
of the Gold Coast and Asante in the nineteenth century...', J. Afr. Hist., 1971, n ,
79-101.
15
A. G. Hopkins, 'Economic imperialism in West Africa: Lagos 1880-92', Eton.
Hist. Review, 1968, 2nd series, 21, 580-606: cf. the rejoinder by R. A. Austen and
J. F. Ade Ajiyi, ibid, aj, 305-12. The best modern account of the wars is S. A. Akintoye,
Revolution and power politics in Yorubaland, 1840-189) (London, 1971).
16
E. A. Ijagbemi, 'The Yoni expedition o f 1 8 8 7 . . . ' , ] . Hist. Soc. Nigeria, 1974, 7,
248.
263

money and lives. At Lagos, Cornelius Moloney still patiently

relied on diplomatic suasion to maintain Britain's sphere of
influence among the Yoruba; only when Egba chiefs, dissatisfied
with their dealings with Lagos, signed a treaty with the French
adventurer, Viard, in April 1888 did he suggest providing for even
nominal cession of sovereignty in his treaties.17
In the short run European hesitancy seemed to offer African
diplomatists some freedom of manoeuvre. Tofa of Porto Novo,
when he accepted France's protectorate in 1883, hoped to use her
support to eliminate dynastic rivals and to extend his authority
over outlying districts which he claimed for his state. The
presence of a French Resident supported by a minuscule garrison
emboldened Tofa to respond with provocative defiance to British
activities in the Weme valley and southern Egbado, and in 1887
to take the risks of quarrelling with Glele, his 'elder brother' of
Dahomey. But in practice the benefits derived from the French
presence proved to be limited. The frontier which French
diplomatists negotiated with the British in August 1889 left many
areas of Egbado claimed by Tofa under the influence of Lagos,
and even his capital was not secure against Dahomian power.
During the later years of Glele's reign tension had continued
within Dahomey between notables engaged in commerce and
those who favoured intransigent defence of territorial sovereignty
against all comers.18 The strength of the latter party was revealed
in March 1889, when Dahomian troops advanced down the Weme
valley, impelling the whole merchant community, and Tofa
himself, to take refuge in the British sphere. Behanzin, who
succeeded Glele at the end of 1889, wished to maintain good
commercial and political relations with Europeans but only from
a base of unconditional independence; taking on himself the
symbol of the shark (which had long impeded French connections
with Cotonu), he took an intransigent, even patronising, line to
Lieutenant-Governor Bayol's clumsy attempts to reassert French
prestige:
The King of Dahomey gives his country to no-one. God has given the white
men their share; when God gave everyone his share, it was Dahomey which
17
Hargreavcs, West Africa partitioned, I, 5 5—6j: E. A. Ayandelc, The missionary
impact on modern Nigeria (London, 1966), 49- 51.
18
D. A. Ross, 'The autonomous kingdom of Dahomey, 1818-94', Ph.D. thesis,
University of London, 1967; Catherine Coquery-Vidrovitch, 'De la traite des esclaves
a Pexportation de I'huile de palme et des palmistes au Dahomey', in C. Meillassoux,
The development of indigenous trade and markets in West Africa (1970).
264

had this place. The man who comes to take the country of Dahomey without
the King's permission will die."
As a constructive step towards better relations, Behanzin sug-

gested that France might replace the jeunes gens irre'flecbis who
seemed to be in control of the Third Republic by the wiser
governance of a restored monarchy.20
Such forthright independence on the part of an African ruler
infuriated patriotic French officials; yet France did not become
committed to the conquest of Dahomey until 1892. Military
expeditions were likely to incur political risks at home, as well
as military risks in Africa, which would be acceptable only when
an African ruler went beyond traditional boundaries of Afro-
European diplomacy (hard bargaining about duties payable by
traders, the jurisdiction of his courts, the admission of missionaries
to his sovereign territory) and questioned the structures with
which the fortunes of his state were becoming involved.
It was his claim to economic autonomy, rather than any
prophetic sense of nationalism, that brought Jaja of Opobo into
conflict with young vice-consul, Harry Johnston, during the rainy
season of 1887. Jaja's commercial strategy had originally been to
ship palm-oil directly to Liverpool and so become less dependent
upon foreign merchants; but once he discovered that this tied up
his working capital for too long he worked with the Glasgow firm
of Miller Brothers against the monopolising practices of the
African Association. Complaints from the latter body about his
wickedness as 'middleman' reflected their desire to force down
prices in order to pass on the cost of the depressed palm-oil
market, not any real intention of moving their own buying
organisation into the producing centres.21 But Johnston's
vigorous backing of their complaints introduced a new political
judgement on ' the inevitable march of events which is causing
Africa to be opened up to the white men's entreprise'.
It is no exaggeration [he wrote] to say that, from Benin to Old Calabar, all
the native chiefs are watching with interest the long struggle between the
'• Archives Nationales, S. O. M., Dahomey V/i, enclosure in Bayol to Etiennc,
4 Jan. 1890.
20
S.O.M., Dahomey III/1. Bayol to Etienne, 14, 10 Jan. 1890.
11
Cherry Gertzel, 'Commercial organisation on the Niger coast, 1851-1891', in
Leverhulme Foundation, Historians in tropical Africa (Salisbury 1962), 501-5: C. de
Cardi, 'A short description of the natives of the Niger Coast Protectorate', App. 1 to
Mary Kingsley, West African studies (London 1899).
265

traders and the Trader-King of Opobo. As either side is victorious they will
rule their conduct accordingly.22
In other words, if Europeans were to control the crises of
authority which accompanied the growth of commerce, they
would increasingly need to substitute force for moral suasion.
Jaja's deportation, deplored but not reversed by Lord Salisbury,
foreshadowed the end of that' loaded pause', which in retrospect
seems to have marked the last days of West African independence.
THE EUROPEAN OFFENSIVES

The changing military balance and the rise of colonial imperialism
Around 1890, relations between Europe and West Africa changed
their character. Hitherto there had been dialogue, though
increasingly violent and one-sided, among independent polities.
Although Europeans always controlled greatly superior military
technology, they had rarely been able to deploy it beyond the
coastline. Africans' resources included their superior capacity to
survive in a physical environment which Europeans knew to be
hostile; a terrain rendered difficult to invaders by dense vegetation,
formidable river-crossings, and sheer distance; and military
tactics well adapted to that terrain and using indigenous
weaponry like spears and bows as well as the inferior firearms
which Europeans had been happy to supply. Some nineteenth-
century rulers tried to modernise their armies. Al-Hajj 'Umar Tal
('Umar b. Sa'Id) used mountain howitzers captured from the
French in the 1850s to reduce the fortifications of his African
enemies, and both he and his son Ahmad tried hard to buy
additional ones.23 Modern breach-loading rifles reached West
African markets during the 1870s. They were used effectively by
Samory's armies and his blacksmiths learned not only to repair
but to copy them; after 1890 he imported some six thousand
modern Gras rifles through his main supply port of Freetown.24
Behanzin, likewise, after his inconclusive brush with the French,
began to supply slaves to Hamburg merchants at Whydah in
22
P. P. 1888, LXXIV, C.5365, Johnston to Salisbury, zo Aug. 1887. Cf. Hargreavcs,
West Africa partitioned, 1, 111—21.
" Y.Saint-Martin, 'L'artillerie d'El Hadj Omar et d'Ahmadou', Bulletin de
fl.F.A.N., scries B, 1965, 27.
2<
Person, Samori, n, 1104—9.
266

return for modern weapons and instruction in their use. During
1891—2 he imported four Krupp cannon, three mitrailleuses, and
nearly two thousand assorted small arms. He also claimed' to have
an armament factory where powder is made according to methods
learned from Europeans, and firearms likewise'.25 Perception of
such rearmament programmes led the Europeans attending an
anti-slavery conference at Brussels in 1890 to resolve that' all arms
for accurate firing' should become the monopoly of their own
colonial forces; although this ban was not fully enforced for
another two years, it heralded a period in which military methods
of European penetration would become the rule rather than the
exception.
Meanwhile Europeans were modernising their own colonial
armies by improving their armaments and expanding their estab-
lishments of regular African troops, preferably drawn from
peoples of 'martial' repute like the Hausa or Bambara. In the
British colonies local constabularies moved gradually from tasks
of' moral suasion' into more active military role in support of the
garrisons of the West Indian Regiment. The largest such force was
the tirailleurs se'ne'galais, created by Faidherbe in 1857 and
progressively expanded during the period of conquest; yet even
in 1900 they numbered only 6,600. African regulars provided the
shock troops; but they were invariably commanded and stiffened
by substantial cadres of European officers, NCOs and specialist
engineers or artillerymen, and were often accompanied, not only
by an extensive tail of carriers, non-combatant auxiliaries and
camp-followers, but by ill-disciplined forces of African allies,
hoping to settle old quarrels or to obtain human booty. The force
which attacked Segu in April 1890 included only 400 tirailleurs,
234 other African regular troops and four African officers; there
were 103 European officers and specialists, 545 non-combatant
Africans on the official strength, 1,466 African auxiliary troops
and 818 porters, autorises a suivre but not directly remunerated.26
The decisive advantage of all the invading armies lay in their
fire-power. Gun-boats had always been able to terrorise limited
areas accessible from coasts or river-banks, but new weapons
25
Archives de la Marine, V i n c e n n e s . BB4 1992, Return o f arms and m u n i t i o n s s o l d
by Germans at W h y d a h , j Jan. 1 8 9 ) . Report by Hocquart, 2 April 1891 (J. D .
Hargreaves, France and West Africa, 1969, 191).
26
J. Meniaud, Ltspionniers du Soudan (Paris, 1 9 3 1 ) 1 , 4 5 0 - 1 ; Marc M i c h e l , L'appela
I'Afriqut (Paris, 1982), 4 - 5 .
267

WESTERN AFRICA, 1886-I9O5
were more mobile. The British Ijebu expedition of 1892, though
hastily improvised, was equipped with three seven-pounder guns,
three rocket-troughs, two Nordenfeldt machine-guns, and one of
the first Maxims to be used in Africa; all these contributed to
inflicting fatal casualties on the Ijebu of the order of twenty to
one.27
By reserving the control of modern military technology to
Europeans, the Brussels conference established an essential pre-
condition of conquest; but the problems of logistics, transport
and organisation remained formidable, and most European
politicians still doubted whether their constituents would be
willing to meet the human and financial costs. The necessary
additional impetus might be provided by some local crisis of
authority, as at Opobo; but after 1890 it increasingly came from
the conscious initiatives of imperially-minded groups of
Europeans, who found that they could mobilise sympathy or
support from fellow-patriots.
The pace of the colonial partition of West Africa was largely
set by the French - and specifically by the ambitions of the
colonial army. Archinard's attack on Segu in April 1890 was not
a response to any local crisis but an act of aggression, inspired
by chauvinism and professional ambition, for which preparations
had been made as carefully in Paris as in Africa. His attack the
following year on Kankan finally committed the French state to
difficult and prolonged campaigns to destroy Samory's empire.
The aims of these officers often diverged radically from those of
civilians interested in doing business with Africans, but to secure
domestic support it was expedient to exhibit unity among
colonialists. The Comite de /'Afrique fran$aise, founded in 1890 by
' a certain number of persons inspired by patriotic zeal', adopted
the dramatic and grandiose design of the famous ' Chad Plan' as
a sort of covering myth, which could lend the appearance of
national purpose to the miscellaneous ambitions of speculators,
soldiers and commercial adventurers in expanding existing French
holdings in Algeria, Senegal and Equatorial Africa.28
During the 1890s other objectives were added. Binger, a
military officer turned colonial administrator and publicist,
" Robert Smith, 'Nigeria-Ijebu' in M. Crowder (ed.), West African resistance
(London, 1971), 179, 189-90.
28
T. W. Roberts, ' Railway imperialism and French advances towards Lake Chad,
1890-1900'. Ph.D. thesis, University of Cambridge, 1973.
268

stressed the importance of following up his own explorations of
1887-9 by consolidating French control of the Ivory Coast,
establishing new routes through the forest to the Niger and to
the populous kingdoms of the Upper Volta. Others worked to
link the military empire of the Sudan with the commercial ports
on the Guinea coast, thus increasing pressure on the almamis of
Futa Jalon.
The ambiguities of French relations with Dahomey too were
eventually resolved only after the French public was converted
to colonialist values. In 1890, as in 1889, France's protectorate
over Tofa had threatened to lead to a military confrontation with
Dahomey for which public opinion in France was not prepared,
and she had to safeguard her hold on Cotonu by a negotiated
compromise. Behanzin, while tolerating a French presence at
Cotonu as well as Porto Novo, refused to abate his sovereign
claims, and increased his purchase of modern firearms. To pay for
them by exporting 'indentured labourers' he resumed slave-
raiding, and so provided France with a moral charter to override
anti-colonial scruples. In April 1892 attitudes had so far changed
that an incident on the Porto Novo frontier was made the pretext
for a full-scale military expedition, which in November occupied
Abomey with a flourish of triumphal colonialism. After Behan-
zin's surrender in January 1894, French expeditions thrust north-
wards into Borgu, so reviving the design cherished by a few
merchants and imperialists for thirty years of outflanking British
trade monopolies by securing a French route to the navigable
Niger below Bussa. The parti colonial, modest though its resources
seem in retrospect to have been,29 had temporarily convinced
representatives of the French state that action to extend French
political control and build railways would produce returns far
surpassing the modest results so far achieved by French merchant
capitalism.
In Wilhelmine Germany too, relatively small groups who had
acquired material or emotional interests in Africa were increasingly
able to call on the political and military resources of the Reich.
Where Bismarck had hoped to transfer financial and military
29
C. M. Andrew, 'The French colonialist movement during the Third Republic;
the unofficial mind of imperialism', Transactions, Koyal Hist. Soc, jth series 26, 1976,
143-66; C. M. Andrew and A. S. Kanya-Forstner, France overseas (London, 1981),
Chap. 1; cf. L. Abrams and D. J. Miller, 'Who were the French colonialists: a re-
assessment of the parti colonial, 1890—1914', Hist. J., 19 (3), 1976, 685—725.
269

responsibilities to companies of merchants, his successors found

themselves launching military expeditions or arguing diplomatic
claims on behalf of their traders. Ambitious officers who explored
the interior established their own systems of political and com-
mercial collaboration; in the Bamenda grasslands Eugen Zintgraff
committed the German government to a treaty with his protege,
Galega of Bali-Nyonga, an astute state-builder.30 Both in Kamerun
and Togo, expansion tended to move towards the outlying
provinces of the Sokoto caliphate; successive expeditions had
established a stake in Adamawa by 1893, and in April 1895 a small
but well-armed expedition under Dr Griiner claimed to have
secured a treaty from the sultan of Gwandu.31 And, as in the other
empires, behind the explorers came the administrators, and newly
organised police forces, armed to enforce control on those who
rejected collaboration on German terms.
The British, too, gradually began to recognise that economic
inducements and diplomatic suasion might no longer suffice to
protect their interests against European intrusion or African
assertiveness. After 1890 they gradually introduced new mech-
anisms of control. Around Sierra Leone new treaties were made,
travelling commissioners appointed, a frontier police force
organised; though a formal protectorate was proclaimed only in
1896, its institutions had largely been invented, or improvised,
already. In the Niger delta, where foreigners had been told of a
British protectorate in 1884, the first plan for 'a paternal
government under the direction of the Foreign Office' was
prepared by Major Claude Macdonald in 1890; but it still rested
largely on moral suasion, with force held in reserve to deal with
those who had failed to draw the right conclusion from Jaja's fall.
In 1892, contrasts could still be drawn between French military
imperialism and Britain's 'policy of advance by commercial
enterprise' ;32 but the differences were more of timing and of style
than of principle.
The incentive to adopt more expansionist policies came partly
30
E. M. Chilver, 'Paramountcy and protection in the Cameroons; The Bali and the
Germans, 1889—191}', in P. Gifford and W. R. Louis, Britain and Germany in Africa
( N e w H a v e n , 1967).
" W. Markov and P. Sebald, ' T h e treaty between Germany and the sultan o f
G w a n d u ' , / . Hist. Soc. Nigeria, 1967, 4, 141—53.
31
P. P. 1892, L v m , C 6701, Salisbury t o Dufferin, 30 March 1891: cf. J. D .
Hargreaves, 'British and French imperialism in West Africa', in P. Gifford and
W. R. Louis (ed.), France and Britain in Africa ( N e w Haven, 1971).
270

from fear of European rivals, partly from the intention to impose

new demands on African collaborators. Late in 1891 merchants
engaged in West African trade united to complain that Salisbury's
diplomatic compromises with the protectionist French had
sacrificed the future prospects of Sierra Leone and the Gambia.33
At the same time they became increasingly eager to meet the
commercial depression by removing restraints imposed on trade
by Africans, and by reducing transport costs through railway
construction. With a general election pending, Salisbury agreed
to respond. He despatched an able young African, George Ekem
Ferguson, to try and raise a treaty barrier to French penetration
of the upper Volta basin; and he responded to the desire of Gilbert
Carter, Moloney's vigorous successor at Lagos, to coerce the
coastal Yoruba kingdoms. Both the Ijebu and the Egba had
hitherto strongly defended their interests as commercial inter-
mediaries by restricting through-traffic and imposing tolls; but it
was the Ijebu, whose exclusion of foreigners extended to
Protestant missionaries, who had made more enemies in the
Colony. Already the failure of an ill-planned diplomatic mission
had seemed to Sir Augustus Hemming, an expansionist in the
Colonial Office, to provide 'a fitting opportunity for breaking
down the trade monopoly which the Ijebus arrogate to themselves1
and which is extremely prejudicial to the interest of Lagos and
prevents the proper development of Yorubaland' ;34 in the spring
of 1892 a new interruption of trade, and new allegations about
slavery and human sacrifice, enabled Hemming's friend, Carter, to
profit from Salisbury's new disposition. A force of Hausa and
West Indian troops overcame determined Ijebu resistance,
occupied the capital, and compelled Awujale Tumwase to open
his roads; this resolved the crisis of authority within Ijebu and
opened the way for rapid expansion of British commerce and
western influences.35 An eminent Lagosian later recalled how
. . . ' the taking of Ijebu sent a shock of surprise and alarm
throughout the land. Men felt instinctively that a new Power had
" Colonial Office (CO.) 879/35, C.P. African 42 j , Proceedings at a Deputation...
8 December, 1891.
14
C O . 147/80, Minute by Hemming, 9 July 1891.
" A. A. B. Aderibigbe,' The Ijebu Expedition 1891', in Historians in tropical Africa:
proceedings of the Leverhulme Inter-collegiate History Conference, held at the
University College of Rhodesia and Nyasaland, September i960; E. A. Ayandele, 'The
ideological ferment in Ijebuland, 1891-1943,', African Notes (University of Ibadan) 1970,
5, 5, 17-40; R. S. Smith, 'Nigeria-Ijebu', in M. Crowder (ed.), West African resistance.

WESTERN AFRICA, l886~ 1905
entered the arena of Yoruba politics, that a new era was about
to dawn on the land.. . ' 36
Yet Salisbury had not intended to abandon his 'policy of
advance by commercial enterprise' in order to exercise direct
control. When the Egba took the message and agreed to modify
their control over the passage of trade, they were able to negotiate
a treaty formally recognising their independence. Captain R. L.
Bower, stationed as resident at Ibadan with a small constabulary
force, was not sent to exercise formal control. ' We have neither
assumed nor declared a protectorate,' wrote Hemming, ' but we
want some one on the spot to watch the movements of the king
and chiefs, and to keep us in touch with them and with the
people. ' 37 Yet within two years Bower, unable to reconcile the
moral world of the Yoruba with that of a Victorian officer, had
bombarded the town of Oyo and reduced the alafin to the status
of a British client.38
The British had in fact taken the first step which counted; the
crises of authority in the frontier zones of empire, which their
presence intensified, would only be resolved by the assumption
of more direct control. Hence Gladstone's return to prime
ministerial office in July 1892, which in Uganda led to hesitation
and controversy over British responsibility, had little effect on the
advance of the West African empire. The Colonial Office under
Lord Ripon proceeded gradually to formalise the ascendancy it
had already acquired; in July 1894 Colonel Frederic Cardew drew
up comprehensive plans for the administration of the Sierra Leone
protectorate, with a government railway as the key to its
economic development.39 Gladstone's successor, Lord Rosebery,
having failed to extend existing Anglo-French agreements into a
comprehensive diplomatic partition, prepared to peg out claims
for the future; when Goldie despatched Lugard to confront the
French thrust into Borgu, Percy Anderson indicated that the
Foreign Office 'was not prepared to take the responsibility of
stopping the expedition'.40 Nor did the government attempt to
stop, or retrospectively censure, the attack which the acting
36
Obadiah Johnson, 'Lagos past': Proceedings of the Lagos Institute, 20 Novem-
ber 1901.
" CO. 147/91, minute by Hemming 16 Jan. 1894, in Carter, 260.
18
J. A. Atanda, The new Oyo empire. Indirect rule and change in western Nigeria,
1894-19)4 (London, 1973), Ch. 2.
39
CO. 267/409, Cardew to Ripon, conf., 45, 9 June 1894.
*° F.O. 85/1315, note by Anderson, 19 Sept. 1894.
272

Consul-General, Sir Ralph Moor, launched on Nana and the
Itsekiri trading empire in August 1894. In February 1895 the
pacific Ripon seems to have been deterred only by considerations
of expediency from approving an immediate expedition against
the ' savage and barbarous power' of Asante, now advocated by
Hemming on the advice of ' Merchants, Missionaries and other
warlike classes'.41
Once the general arguments were accepted, and successfully
put into practice in particular cases, the way was clear for
' Imperialism' to become an effective credo for politicians in Great
Britain, as it was already proving to be in France, Germany and
Italy. Joseph Chamberlain's choice of the Colonial Office in
preference to the Treasury in July 1895 showed where one
bourgeois radical saw the best chance of building a reputation.
His declaration on 22 August, that he was prepared to advocate
government participation in West African railway construction
and other projects, 'in which by the judicious investment of
British money, those estates which belong to the British Crown
may be developed for the benefit of their population and for the
benefit of the greater population which is outside', provided a
manifesto no less effective because it aroused expectations beyond
anything that Chamberlain would actually achieve in West Africa.
Chamberlain's radical zeal was checked by the older, more
sceptical style of imperial statesmanship represented by Salisbury,
who saw African colonies in the perspective of a world-wide
system of power relationships; but both now operated on the
assumption that the exercise of political power in Africa could
contribute to solve Britain's long-term economic problems. The
increasingly vigorous use of military force was not justified by
unctuous appeals to nationalistic sentiment. In September 1895
Chamberlain authorised the ultimatum which led to the deposition
of Prempeh the following January. Self-righteously moralistic
flourishes accompanied the conquest and looting of Benin in 1897;
the following years saw a series of military thrusts into Iboland,
of which the Aro expedition of 1901—2 was the most important.
From the point of view of imperial strategy northern Nigeria
remained the main objective. Adamawa was partitioned with
Germany in 1893 and the threat from Togo never materialised;
41
CO. 879/42, C. P. African 486, memo by Hemming, Feb. 1895. Rosebery papers
(National Library of Scotland), box 46. Ripon to Rosebery, 20 Feb. 1895.
273

WESTERN AFRICA, I 886—1905
but the ambitions of the French colonial party proved harder to

contain. After a period of tense confrontation in Borgu this
challenge was removed by diplomacy in June 1898; and in 1900
the government took over from the Niger Company responsibility
for asserting control of Britain's allotted sphere in the Sokoto
caliphate. A short period of cold war ended with Lugard's military
occupation of Kano and Sokoto in 1903.
Meanwhile the eastward advance of the French military had
continued, checked only temporarily by the casualties which the
Tuareg inflicted near Timbuktu in January 1894. Late in 1896 they
occupied Mossi, Gurunsi and Gurma, thus setting limits to the
northern expansion of the Gold Coast and Togo. In May 1898
they occupied Sikasso; in September Samory, their most for-
midable enemy, who had retired fighting to the northern Ivory
Coast, was finally taken prisoner. Other columns, grudgingly
observing the restraints imposed by Anglo-French diplomacy,
moved towards Chad through the outlying provinces of the
Sokoto caliphate; on 22 April 1900 they killed Rabah al-Zubayr,
so defeating the last of the empires combattantes. The French thus
secured their coveted access to the northern and eastern shores
of Lake Chad, and cleared the way for the British occupation of
Bornu.
Some African responses
Scrutinised through European eyes the process of military
occupation, rapidly though it was carried through, appears a
somewhat erratic process, its detailed timing regulated by for-
tuitous combinations of circumstances in Europe and in Africa.
So it is hardly surprising that African rulers, responding to drastic
changes of behaviour in ignorance both of the underlying causes,
and of the diplomatic bargains which the European governments
had made among themselves, should show puzzlement and
uncertainty. Their individual crises might take the form (as in
Mossi) of the sudden appearance of competing white men with
well-armed escorts, in areas where they had not been seen before;
or (as in Asante) of sudden changes in demands and attitudes of
long-familiar strangers; or (as for Samory) of direct aggression.
Many of the societies most directly challenged had long been
deeply ambivalent in their responses to European civilisation,
commerce and Christianity; the European onslaught might
274

provide occasions to resolve, temporarily at least, long-standing

crises of authority or dilemmas of policy.
The French military allowed the Muslim empires of the western
Sudan little real choice; Archinard's attack on Kankan finally
convinced Samory that they would never accept co-existence on
conditions compatible with the preservation of the Dyula
revolution; in 1892 Samory decided to displace the whole military
structure of his empire eastwards, using the rifles which he was
still obtaining from Freetown to harass the French, and leaving
them only scorched earth to occupy. This policy, adopted at a time
when the region was severely afflicted by cattle disease, imposed
new sufferings upon Samory's subjects; but the French conquerors
found it difficult to identify and use collaborative authorities
among the peoples they claimed to have liberated — largely
because of their own ethnocentric attitudes, but also because the
Samorian revolution had already levelled structures on which it
might have been possible to build.
For the Tukolor rulers of Ahmad's empire there was likewise
little alternative to resistance (much less effective) and bijra in face
of aggression, although Archinard did utilise his half-brother,
Agibu, as collaborator, first at Dinguiray, then at Masina. For a
time Archinard also tried to restore Bambara rulers whom the
Tukolor had displaced; but his devious tactics and lack of basic
confidence in Africans soon led them to reject this clientage, and
here too conquest led quickly to a general levelling of historic
polities.
For the Sokoto caliphate, more options were open. Under 'Abd
al-Rahman (1891-1902) there was growing apprehension of
gathering storms. Rabeh's conquest of Bornu in 1893 established
on the eastern borders a formidable rival (with a special appeal
to Mahdist dissidents within the caliphate), just at the time when
changing European behaviour elsewhere in Africa sharpened the
fear that diplomatic skill in fending off emissaries might not be
enough. Goldie's Nupe and Ilorin campaigns of 1897 struck the
first direct blows, but as they were not followed by immediate
occupation, 'Abd al-Rahman, worried about Rabah and the
French, was still tempted by Goldie's assurances that British
suzerainty would be the lightest and best of the threatened
alternatives.42 Though the British threat grew rapidly thereafter,
42
Goldie to Lugard, 21 Dec. 1897, cited Adeleye, Power and diplomacy, 190.
275

the Caliph doubted his capacity to mobilise the resources of the
emirates in opposition; while Sokoto maintained a sort of
smouldering hostility, Lugard was able to play upon divisions
between and within the southern emirates to subdue them
separately. In the final confrontation which Lugard provoked in
1903 the new caliph, Attahiru I, found no alternative to a hopeless
battle, followed by a hijra which secured much support. Yet those
Fulani who remained to collaborate with the conqueror found it
possible to preserve the dominance, not only of the Muslim
religion but of those conservative social forces in which it had
become encapsulated. Had earlier caliphs been more intransigent
in resistance, Lugard might well have been less willing to rely on
those associated with them for the future framework of British
administration.
Coastal peoples with long experience of Europeans perhaps
found the new era even harder to understand. Their own
responses might have been equally revolutionary, as in the case
of the Ijebu. Before 1892 the dominant elements in this area had
kept alien cultures, as well as alien traders, at arm's length; but
some of the Ijebu who most fiercely resisted the British invasion
then became enthusiastic collaborators with the new order. The
undermining of traditional institutions unleashed an ' ideological
ferment'; conversions to Christianity and Islam increased spec-
tacularly, and the converts began to diffuse the new values more
rapidly than in those parts of Yorubaland where the British
implantation was more gradual.43
The Fon, too, once diplomacy had failed, saw no alternative
to resisting French aggression, and largely united behind Behanzin
in defence of their traditional state. But their military tactics, even
afforced by modern weapons, could not resist General Dodds's
column. After Abomey fell in November 1892 Behanzin tried to
secure some autonomy as a tributary of the French; when this was
rejected, he organised guerrilla resistance in the north. Only when
it became clear that the French would not maintain Behanzin on
the throne did his brother, Gucili, agree to collaborate with them,
exercising a strictly limited authority within a much circumscribed
territory until 1900. While some traditions present this action as
betrayal, others suggest that Behanzin had made a secret blood-pact
43
Ayandele,' The ideological ferment.. ', passim; cf. M. O. Abdul,' Islam in Ijebu-
Ode', paper presented to 16th Congress, Historical Society of Nigeria, Dec. 1970.
276

with his brother, accepting deposition in hope of preserving the

continuity of monarchy. Gucili's regnal name of Agoli-Agbo (' the
dynasty has not fallen') suggests that there was continuity of
purpose behind the apparent switch from resistance to
collaboration.44
For some peoples, the implications of the new imperialism were
slower to dawn, and the establishment of colonial authority might
require more than one display of force. Asante, for example, did
not seek to avoid participation in the world economy; its politics
seem to have turned on alternative methods of doing so. The
asikafo, or ' rich men', who developed the tapping and export of
rubber during the 18 80s, together with the ahiafo, or poor, seem
to have favoured a weakening of the control traditionally exercised
by Kumasi over the economic activities of individuals, while
Agyeman Prempeh, who became asantehene in 1888, favoured what
has been called 'modernisation under mercantilist direction',45
This they thought was their own internal problem. Despite the
shock they had experienced in 1873-4, the Asante were well used
to dealing with the vacillating and inconsistent representations of
British governors, and in 1891 the asantemanhyiamu (the Asante
national assembly) felt it safe to support Prempeh in rejecting Sir
William Brandford Griffith's unauthorized proposal for a British
protectorate, declaring that ' Asante must remain independent as
of old'. 46 However, when Governor Maxwell arrived in January
1896 with his Hausa constabulary and proceeded, on rejection of
his preemptory ultimatum, to arrest and depose Prempeh and
other leading chiefs, this initially had a stupefying effect. The full
implications did not dawn until tactless demands by Governor
Hodgson triggered the insurrection of 1900, with its demands for
an end of the new obligations and a return to pre-colonial
freedoms.
A different pattern of delayed reaction to colonial rule came
from the Mende of Sierra Leone. During the nineteenth century
their political and social structures appear to have been fluid.
Individuals could achieve the status of chiefs on the basis of
44
L. M. Garcia, ' A r c h i v e s et tradition o r a l e ' , Cahiers <Tetudes africaines, 1976, 16,
61-2, 189-206.
45
I . W i l k s , Asante in the nineteenth century (Cambridge, 1975), e s p . c h . 1 5 ; cf.
T. D. Lewin, Asante before the British: the Prempean jears, iS/j—ifoo (Laurence, 1978).
46
P. P. 1896, LVII, C 7917, 16, Griffith to Knutsford, ) June 1891, cncl. Prempeh to
Griffith, 7 May.
277

WESTERN AFRICA, I886-I9O5
personal qualities and skills - religious, commercial, diplomatic,
and especially military prowess; but some underlying unity and
continuity was maintained through the authority of the Poro,
which learned to act as guardian of new commercial interests as
well as of Mende traditions. Officials in Freetown knew little of
the Poro and distrusted what they knew, but when seeking
collaborators they tended to 'ossiffy] this fluid situation on the
basis of what they thought was tradition and custom'.47 The
establishment of the British protectorate was thus a gradual and
insidious process, involving not only the presence of the often
oppressive frontier police, increasing British interference in
matters like domestic slavery, and eventually the taxation of
houses, but also the imposition of ' paramount chiefs' amenable
to colonial control upon the evolving political order. In 1898 the
imposition of house-tax proved the final straw, provoking a
widespread rural insurrection in which Sierra Leone Creoles were
the chief sufferers. This traumatic event had sobering effects on
' modernising' administrators elsewhere in the British empire.
Historians of the colonial occupation have tended to concen-
trate their attention on polities which ultimately decided on armed
resistance; but many did not. Shrewd rulers like Tofa at Porto
Novo calculated that their interests, or those of their subjects,
could best be served by close collaboration with a foreign power;
but this could prove a grave miscalculation. Long before Tofa
died in 1908 his hopes had been sadly disappointed, as irascible
French officers undermined the authority he had worked to build
up in the eyes of his subjects; there was nothing left to hand on
to his heirs. Tofa's funeral effectively marked the end of the Porto
Novan monarchy;48 thereafter his dynastic enemies began to join
other displaced groups in early forms of anti-colonial resistance.40
Sometimes, however, there were genuine foundations for the
colonialists' myth that they were received as liberators. In western
Yorubaland, a historian native to the area concludes that both
French and British were initially welcomed 'as a godsend to
47
A . A b r a h a m , Topics in Sierra Leone history ( F r e e t o w n , 1976), 2; see also his Mende
government and politics under colonial rule ( F r e e t o w n , 1978).
48
A . A k i n d e l c and C. A g u e s s y , Contribution a I'etude dt I'histoire de t'ancien royaume de
Porto Novo, Memoires de 1'I.F.A.N., no 25 (Dakar 1953), 89; cf. Nathan S. Senkomago,
'The Kingdom of Porto Novo with special reference to its external relations,
1862—1908', Ph.D. thesis, University of Aberdeen, 1976.
40
J. A. Ballard, 'The Porto Novo incidents of 1923; politics in the colonial era',
Odu, July 196;, 2, 1.
278

deliver the people from their adversaries', the Egba and the Fon.50
Similarly there is no reason to doubt that many Africans were
pleased to see Samory or the Tukolor displaced by the French army,
even if they soon became disenchanted with the consequences.
Some of the coastal peoples were already so strongly penetrated
by western influences, or so firmly articulated into the commercial
economy, that no sharp discontinuity was involved in the
establishment of colonial rule. In the southern Gold Coast,
declared a British colony in 1874, the actual extension of British
authority as marked by the activities of commissioners and
constabulary had been a gradual process, closely related to
internal changes and conflicts arising from the simultaneous
spread of commerce, Christianity and literacy. Given the con-
tinuing reluctance of the British to assert their undivided
sovereignty, those who understood something of the new forces
(whether chiefs or commoners) tried to manipulate the emergent
colonial administration, for example by invoking its support in
the growing number of destoolments and land disputes. In
general, educated chiefs were the gainers; by 1910 the Gold Coast
government had conceded to them ' a significant degree of judicial
authority under weak control in order to enlist their co-operation
in administrative matters'.51 In Akim Abuakwa, chief Amoako
Atta II (1888—1911) consolidated a position in the new colonial
system which his successor and former secretary, Ofori Atta I
(1912-43), would skilfully utilise to extend the jurisdiction of his
state, to control the expanding frontiers of gold miners and
cocoa-farmers, and so to increase the revenues and patronage
available to his stool.
There was clearly great diversity in the manner in which the
new colonial order initially presented itself to African rulers and
their subjects, and in the ways by which they tried to come to
terms with it. But by 1905 -although there were still remote
districts in the rain-forest and the desert where no effective
' pacification' had yet taken place - the fact of colonial rule had
generally been accepted. Longer-term patterns of co-existence, if
not of collaboration, were now in process of establishment.
50
A. I. Asiwaju, Western Yorubaland under European rule, itfy-iyfj: a comparative
analysis of French and British colonialism, Ibadan History Series (London 1976), 53, and
chs. • and 2 passim.
sl
Jarle Simensen, 'Commoners, chiefs and colonial government; British policy and
local politics in Akim Abuakwa, Ghana under colonial rule', Ph.D. thesis. University
ofTrondheim, 1975,90.

WESTERN AFRICA, 1886-I9O5
INITIAL PATTERNS OF COLONIAL COLLABORATION
Once the French, British and Germans had acquired their new
empires, they had to find methods of governing them. Before the
conquest, they had addressed little direct attention to the specific
problems of doing this. To maintain their interests hitherto, they
had relied upon the more or less voluntary collaboration of
Africans; and, given their determination not to provide the
resources needed for direct administration, this was what they
would still have to do. Imperialist euphoria in Europe did not
mean willingness to finance large colonial bureaucracies; in 1901
the entire white staff of the German administration in Kamerun
numbered 77, in Togo 23." Yet the responsibilities which
colonial governments now assumed — not merely for governing
their enlarged territories but for encouraging a progressively
greater involvement in the commercial economy - were vastly
more complex than those of the anti-slave trade era. They would
need African co-operation in recruiting labour to build railways
and feeder roads and administrative stations, to collect taxes, and
to judge disputes and maintain order in the increasingly complex
conditions created by the spread of commerce. It was therefore
necessary to undertake 'a reconstruction of collaboration'.53
In West Africa it soon became apparent that collaboration
would continue to be based on African communities which
possessed their own land and organised agricultural production,
even though administrative as well as market constraints would
direct this towards export markets. Governors eager to expedite
the exploitation of mineral and forest resources might experiment
with the granting of concessions and monopolies, and even
blunder onto the delicate ground of land ownership, but only
rarely did they attempt to interfere with the usufructory rights
of African cultivators. In the stage which Afro-European rela-
tionships had reached in West Africa, this was a matter of
enlightened self-interest. Administrators whose revenues de-
pended largely on customs duties naturally gave priority to
expanding those modes of production on which the existing
export economy was based.
52
L. H. Gann and Peter Duignan (eds.), Colonialism in Africa, 1870-1960, vol. iv:
The economics of colonialism (Cambridge, 1975), Introduction, 18.
51
Ronald Robinson, 'Non-European foundations of imperialism: sketch for a
theory of collaboration', in Robert Owen and Bob Sutcliffe (ed.), Studies in the theory
of imperialism (London, 1972), 133.
ZoO

This might nevertheless involve imposing changes in the actual

pattern of African production. Whereas in the Gold Coast Akan
entrepreneurs were by 1900 already willingly and effectively
responding to growing European demands for cocoa, their Baule
kinsmen in the Ivory Coast still preferred to concentrate on
producing gold and cloth to exchange with their African neigh-
bours. Their French overlords were prepared to tolerate this so
long as the country's resources of wild rubber, exploited by a
cosmopolitan group of European, African and Asian entre-
preneurs, provided an alternative revenue base; but by 1908, with
the collapse of the rubber bonanza in sight, official policy
regarding taxation, labour, and land tenure was seeking to direct
Baule resources towards the cultivation of cotton or cocoa.54 This
would cause conflict. But — at least until Liberia granted the
Firestone concession of 1926-there was never any serious
question of a large-scale plantation economy in West Africa. The
self-interested conviction of established European merchant
houses, that existing commercial structures were capable of
expansion, was powerfully sustained by colonial administrators
who, quite apart from the paternalistic ideals which many of them
fervently held, knew that their existing revenues derived ultimately
from existing patterns of production, and that any threat to
African land rights might produce discontent, if not widespread
revolt on the Sierra Leone model.
Challenges to this economic policy did develop from Europeans
seeking exclusive rights over natural resources, and preferential
access to the labour needed to develop them; and they had some
successes. It was difficult for impoverished colonial governments
(or ambitious African rulers) to resist proposals for mineral
concessions; from the late 1890s there was a localised gold boom
in the Gold Coast and Asante, and once the mines were established,
the colonial authorities found themselves under pressure to assist
labour recruitment." The British Cotton Growing Association,
speaking for a key British industry, enjoyed much Colonial Office
favour, but less from the men on the spot. In 1904 the Ibadan
chiefs succeeded in resisting its demand for a 15,000-acre con-
cession; attempts to increase British cotton supplies had to rely
54
T . C . W e i s k e l , French colonial rule and the Baule peoples ( O x f o r d , 1980), 1 5 4 - 6 0 ,
172-86.
55
R. G. Thomas, 'Forced labour in British West Africa: the case of the Northern
Territories of the Gold Coast, 1906-1927', J. Afr. Hist. 1973, 14, 79-103.
281

on diverting peasant production from the local weaving industry
to Lancashire.56 However the British did permit the estab-
lishment of a few European plantations, and the French rather
more; in the Ivory Coast, where Arthur Verdier had led the way
in 1881, these would become serious competitors with African
farmers, not least in seeking access to migrant labour from Upper
Volta. In Kamerun, the colonial economy began in the 1890s to
follow the pattern of equatorial rather than west Africa; force was
increasingly used to secure land and labour for German interests,
and to restrict the trading privileges of the Duala, their original
collaborators. Governor Puttkamer, appointed in 1894, granted
substantial concessions for the extraction of rubber and ivory as
well as for agricultural plantations; in 1906, however, he was
recalled by the reforming colonial secretary, Bernhard Dernberg,
and African agriculture began to be taken somewhat more
seriously.57 In Togo, on the other hand, it was the view of the
evangelical Bremen merchant, J. K. Vietor, that the government
should collaborate with 'free peasants, happily and peaceably
cultivating their own fields with their wives and children',58
which generally prevailed against those who sought monopolistic
concessions. The main reason for the adoption of this form of
economic policy was of course that it worked. By 1905 most
colonies could feel optimistic about the prospects for increasing
the type of commercial exchanges already established between
African agriculturists and the capitalist world.
Finding appropriate methods of exercising political control was
however still a subject for experiment and improvisation. Whether
the new colonial masters arrived as military conquerors or with
the implied consent of their new subjects, their initial problem
was to identify persons of authority who could provide them
with porters, provisions and local intelligence, and assist them in
the adjudication of disputes and the maintenance of order. This
exercise of' finding the chief, though vital for future relationships
in the district, had often to be carried out under the most
56
K . D . N w o r a h , ' T h e West African operations o f the British Cotton Growing
Association, 1 9 0 4 - 1 9 1 4 ' , Afr. Hist. Studies, 1971, 4, 3 1 5 - 3 0 ; Marion Johnson, 'Cotton
imperialism in West Africa', African Affairs, 1974, 7 3 , 178-87.
57
K. Hausen, Deutsche Kolonialberrscbaft in Afrika (Zurich, 1971); M. Michel, 'Les
plantations allemandes du mont Cameroun, 1885—1914", Revue fraitfaise d'bistoire d'outre
mer, 1976, 58, 207.
58
J. K. Vietor, Geschichte und kulturelle Entwicklung unserer Schiit^gebieteti (Berlin,
•9>3). M3-4-
282

unfavourable conditions; sometimes there was no evidently
available candidate, sometimes there were several, and those who
volunteered for the role might thereby discredit themselves with
important sections of the community. Colonisers and colonised
were mutually ignorant, not only about the authority which
individuals might enjoy but about the nature of authority itself
in the societies which they were encountering.
In this early phase, less depended on the national culture or
colonial doctrine of the colonisers than on their immediate needs;
the French were no less ready than the British to work through
established hierarchies if this seemed expedient. In Futa Jalon in
1881 Jean Bayol, anxious to secure a treaty to exclude British
claims, accepted terms which left the ruling Alfaya and Soriya
houses with substantial freedom of diplomatic manoeuvre; un-
willing to risk yet another military campaign the French had to
rely on moral pressure through existing authorities to keep their
imperial option open. This meant acquiescing in attempts by the
almamis to strengthen their authority within this decentralised
state, and even in their readiness to supply firearms to Samory.
Not until 1896, when any danger of British intrusion had finally
been removed, did the development of long-standing conflicts
between Almami Bokar Biro, provincial chiefs, and other elements
of the Fulani aristocracy provide the French with opportunities
to occupy Timbo militarily and impose a Resident.59 By 1900 the
new administration was strong enough to break up Futa Jalon
into its constituent 'cantons', and to depose troublesome rulers
at will. But even these manoeuvres involved finding some
elements of the Fulani aristocracy willing to collaborate; they had
secured sufficient influence to ensure that their distinctive brand
of Islamic conservatism remained a dominant force throughout
the colonial period.
Where French officers arrived as conquerors of the Muslim
empires, they seemed to have a wider choice of collaborators;
Archinard could experiment not only with the use of Agibu
and the Bambara dynasties, but with the appointment of Mademba
Sy, a former employee of the telegraph service who had done well
in the army, zs/ama of the important town of Sansanding. None
of these expedients had any lasting success, even from the French
s9
Winston McGowan, 'Fula resistance to French expansion into Futa Jallon,
1889-1896', J. Afr. Hist., 1981, 22, 245-61.
285

point of view. In Dahomey, on the other hand, the Senegalese

General Dodds, anxious to see trade resumed, was more concerned
to seek collaborators with genuine authority among their peoples.
While accepting the restoration of Agoli-Agbo in the central core
of the Abomey kingdom, in dependent provinces like Ketu, he
made genuine attempts to restore the authority of Yoruba obas
who had been conquered by the Fon. Seeking ' responsible chiefs,
sufficiently powerful to exercise real direction over the popula-
tion', Dodds signed protectorate treaties providing that their
countries would be governed according to their own laws and
customs, and their institutions respected.60
The British too, in these early years, were groping for viable
bases of collaboration according to particular circumstances of
time and place. Sometimes they improvised as wildly as Archinard.
In 1897 the government of Sierra Leone recognized as chief of
Sembehun, Nancy Tucker, a trader who was the mistress of a
sergeant in the Frontier Police; she proved a ' loyal' collaborator
during the collection of house tax, but not an effective one when
it came to ensuring the collaboration of her subjects. In the
bewildering environment of the Ibo hinterland, it could happen
that ' the whiteman just looked at Anyigo Agwu [or some other
striking-looking personage] and called him out to be a chief'.61
When consultation did take place, Ibo might prefer to protect
their true village head by putting forward a person of small
standing. The British thus manned their new system of ' native
courts' by creating a set of appointed functionaries — the ' warrant
chiefs' and ascribing them authority of a sort unknown in Ibo
history.62
If the British were in this initial stage more likely than the
French or Germans to work through established African rulers,
this was due more to pressures upon civilian administrators to
economise in the use of force than to any general theory of
collaboration. Carter's reason for retaining the awujale, Tumwase,
in Ijebu in 1892 were severely practical; 'although essentially a
weak man, he is not wanting in sense, and it is better to deal with
a man of that kind, who will be amenable to reason and can easily
be managed, than to allow the Jebus to elect another king to be
60
Asiwaju, Western Yorubaland, 85-8.
61
Quoted in A. E. Afigbo, The a/arrant chiefs: Indirect Rule in South-eastern Nigeria,
1Sfi-1929 (London, 1972), 61.
62
Ibid., chs 2 - 3 ; E. Isichei, A history of the Igbo people (london, 1976), ch. 10.
284

a tool in their own hands'. 63 In Abeokuta, Carter, knowing his
government's reluctance to sanction further military operations,
was equally ready to work through a complicated oligarchy which
he hardly understood; once the oligarchs had learned their lesson
and agreed to open the roads, Carter did not press those clauses
in his draft treaty which would have authorised close political
control by the Lagos government over the Egba state, but agreed
that 'its independence shall be fully recognized'.64 But
independence in the age of the railway seemed to require new
institutions; British officials encouraged the conflicting authorities
to form the Egba United Government, guided from 1902 by a
Wesleyan minister who took the name of Adegboyega Edun. By
1909 Abeokuta possessed the rudiments of a modern bureaucratic
administration, employing 357 staff in fifteen departments,
including European engineers and other specialists.65 But —
although the United Government looked to Lagos for technical
advice, for consent to the raising of revenue, and sometimes for
help in maintaining order — there was no consistent policy of
encouraging this attempt at self-modernisation. When in 1914
Lugard found a pretext to abrogate Carter's treaty, the Egba
discovered that their independence had become purely nominal.
The collaborative structures which Lugard had established
after the military conquest of northern Nigeria were of course very
different; Sokoto provides the standard example of pre-colonial
authority incorporated into a colonial regime, though on the
clearly declared basis that' all these things . . . which the Fulani by
conquest took the right to do now pass to the British \ 6 6 But even
here the British did not, as is sometimes assumed, simply proceed
on the basis of preserving existing rulerships with which to
collaborate; Lugard at least considered the alternative of restoring
Hausa dynasties displaced by the jihad, before deciding to build
the new system of colonial autocracy upon the structures created
under the caliphate. The northern Nigerian system of ' indirect
rule' was a response to local circumstances before it became the
basis of colonial dogma.
63
C O . 8 7 9 / 3 6 , C P African 428, n o . 133, Carter t o Knutsford, 3 June 1892, 182.
64
P.P. 1893-94, LXII, C 7227, Carter t o R i p o n , 11 O c t . 1893, 4 - 8 , 36.
65
Agneta Pallinder-Law, ' G o v e r n m e n t in Abeokuta, 1 8 3 0 - 1 9 1 4 , w i t h special
reference t o Egba United G o v e r n m e n t , 1898—1914', P h . D . University o f G d t e b o r g ,
' 9 7 3 . '3<>. M J - J 9 -
66
Quoted in Margery Perham, Lugard: the jean of authority, i/fi-/f4} (London,
i960), 128-9.
285

WESTERN AFRICA, 1886— I9O5
Though sometimes the collaborative arrangements improvised

during the period of occupation proved remarkably durable, often
their weakness became quickly apparent. The testing time came
when colonial governments, eager to accelerate the growth of the
commercial economy, increased their demands for labour and
taxation. Until 1898 Mende and Temne leaders were, with
misgivings, prepared to accept the presence of frontier police and
district commissioners — on condition that their attempts to
exercise a 'civilising' influence on the local judicial system, and
in particular to interfere with the peculiar institution of domestic
slavery, were kept to a very bare minimum. But when Sir
Frederick Cardew, anxious to encourage commerce and finance
the railway, imposed the house tax, each people resorted to its own
style of military resistance. Similarly, the collaborators whom the
French had identified among the Baule proved reliable only so
long as no great demands were made upon them. In 1894-5 and
in 1898—1902 the French military, needing porters and labourers,
threatened to erode domestic slavery and deprive the Baule of
their labour force, and so provoked localised rebellions; in 1908
Governor Angoulvant sought to undermine the Baule economy
by greatly increased demands for labour and taxes, and reacted
to their resistance by a still more violent programme of' pacifica-
tion ', intended to inaugurate direct French control.67 It was such
crises of colonial authority which provided the real incentives to
develop more systematic approaches to colonial government.
One possibility, taken more seriously by the British than by
French or Germans, was to revive an older style of West African
policy. In 1909 the colonial secretary of the Gold Coast, Major
Bryan, argued that the jurisdiction of the chiefs should be
abridged in favour of the educated African elite, whom he
considered ' far better qualified to administer justice'.68 Such men
had indeed played distinguished roles, not merely in the
administration of the nineteenth-century coastal settlements, but
in their expansion. George Ekem Ferguson, not only a trained
surveyor but a loyal servant of Queen Victoria, could be sent at
minimal risk to health, and minimal expense, to make the treaties
which justified British control of the northern territories of the
67
Weiskel, baule peoples.
68
Memo by Bryan, 13 Sept. 1909 (enclosed in C O . 96/486, Rodger, 19 Oct.), cited
in Simensen, 'Commoners, chiefs and colonial government', 85-6.
286

Gold Coast;69 Bou el-Moghdad Seek, assessor to the Muslim
tribunal in St Louis, volunteered to promote French influence
among Senegambian Muslims; T. G. Lawson pioneered the
system of British paramountcy around Sierra Leone, on lines
which his successor, Ernest Parkes, believed could have been most
constructively developed by the appointment of Creole district
commissioners.70 Much nineteenth-century experience seemed to
encourage the continued employment of western-educated
Africans as responsible agents on the expanding frontiers of
European influence. But European attitudes and prejudices on
this matter had begun to change.
Early Victorians who hoped to train up educated Christians
committed to the social and economic values of bourgeois
capitalism had succeeded remarkably in this. Creole loyalists, with
genuine roots in both African and western cultures, saw
themselves as essential intermediaries and natural partners in the
creation of West African nationality. Senegalese whose citizen
rights had been recognised by the Third Republic had comparable
expectations. Edward Blyden, prophesying the partition in 1880,
envisaged Afro-Americans from the United States also playing
indispensable roles in that 'Providential' event.71 Europeans
anxious to spread the benefits of Christianity and commerce might
thus have counted on the continuing co-operation of this
population of several hundred Africans, literate in English or
French, possessing various relevant technical and commercial
skills, in touch (to some extent at least) with the customs and
attitudes of fellow Africans, and eager to promote the expansion
of western civilisation in return for very modest remuneration.
Yet, ironically, the zenith of Euro-American imperialism in the
1890s produced a new orthodoxy of cultural arrogance, coloured
by racial prejudice, which set serious restrictions upon the
collaboration of educated Africans in the new tasks of
empire-building.
During the 1890s (as the supply of young Europeans anxious
for appointments in West Africa increased, and medical advances
60
Roger G. Thomas, 'George Ekem Ferguson: civil servant extraordinary', Trans-
actions of the Historical Society oj Ghana, 1972, xiii; Kwame Arhin (ed.), The papers of George
Ekem"Ferguson,African Social Research Documents, 7 (Leiden and Cambridge, 1974).
70
Sierra Leone Archives: Minute paper 4602/1892, memo by Parkes 18 Nov. 1892.
" E. W. Blyden, 'Ethiopia stretching out her hands unto God; or, Africa's service
to the world'. Discourse delivered before the American Colonization Society, May 1880,
in Christianity, Islam and the Negro race (1887; reprinted Edinburgh, 1967).
287

made it more likely they would survive) these prejudices issued
in a general de-Africanisation of senior posts in the public service.
Racial discrimination was not of course new; it had long been
applied in cases where Africans might be required to administer
justice72 or medical treatment73 to white men. But now it became
exceptional for Africans to be appointed or retained in responsible
posts. In Sierra Leone, Governor Cardew, finding a number of
African officials deficient by his austere standards of public
rectitude, concluded that ' none of them are fitted at present by
character and habit to hold such responsible positions as heads
of departments', and began to reconstruct his administration
accordingly.74 In Lagos, where clergy of the Church Missionary
Society had led the way by their strictures on the ageing Bishop
Crowther, a similar process left Henry Carr the only senior
African official by 1900.75 In Senegal, the reaction centred on
attempts to abridge the rights of citizenship which during the early
years of the Third Republic had been accorded to residents of the
four communes of St Louis, Goree, Dakar and Rufisque.76 In the
lower cadres of government, missionary and commercial service
the need for literate African collaborators expanded; Europeans
could never have extended their control so rapidly without the
services - often rendered with genuine empire loyalism - of
thousands of clerks, storemen, constables, boatmen, customs
officers, medical orderlies, teachers, mechanics and catechists who
had been trained in the schools of Freetown or Cape Coast, Lagos
or Calabar, St Louis or Porto Novo. But the routes by which such
men might have advanced to administrative or political
responsibility became increasingly impassable everywhere except
Liberia — an unedifying exception which many Europeans re-
garded as justifying their own new rules.
71
E.g., CO. 267/517, minutes by Holland, Wingficld, Kimbcrley on Hennessy to
Kimberley, 16 Nov. 1872, 143.
73
E . g . , C O . 1 4 7 / 5 8 , Evans t o Holland, Conf. 10, 6 April 1887, encl. Grant,
16 March.
74
C O . 2 6 7 / 4 1 2 , Cardew to H e m m i n g , pte., 28 N o v . 1894; cf. C. H. Fyfe, A. history
of Sierra Leone (Oxford, 1962), 537—38.
75
P. D . C o l e , Modem and traditional elites in the politics of Lagos (Cambridge, 1975),
7 5 - 6 ; J. B. W e b s t e r , The African churches among the Yoruba, 1888-1922 ( O x f o r d , 1964),
part I; cf. D. Kimble, A political history of Ghana, if/o-iyit (Oxford, 1963), 93—8.
76
G . W . J o h n s o n , The emergence of black politics in Senegal: the struggle for power in the
four communes, if00-1920 (Stanford, 1971), 80-2; cf. P. Mille, 'The black vote in
Senegal', / . Royal Afr. Soc, 1900, 1, 64—79.
288

EARLY IMPACTS
This self-imposed reluctance to assign educated Africans
significant responsibilities in the expanded empire obliged the
colonial powers to consider how the working arrangements they
had improvised with various 'chiefs' might be converted into
regular systems of government. The ultimate results may be more
appropriately considered in later volumes, but already by 1905 it
was possible to discern signs of developing national styles of
'native administration'. The French, while continuing to accept
the expediency of utilising the traditional authority of some rulers,
such as the mogho naba, were already tending to level out historical
institutions in order to incorporate the chiefs into a uniform
bureaucratic order. For some of the military commanders the ideal
collaborator was a well-disciplined NCO; and that was what
some of the men recognised as chiefs actually were. The British
on the other hand were in general readier to utilise the authority
of powerful rulers like the Fulani emirs when this could prudently
be done. Initially, this might be due to calculations of financial
expediency; legitimate government was economical government.
Captain Ross, the influential district commissioner of Oyo,
prosaically estimated the collaboration of the alafin to be worth
'a saving in administrative charges of from £3,000 to £5,000 per
annum', and Governor Girouard regarded more direct methods
of ruling northern Nigeria as 'quite out of the question on
financial grounds \ 7 7 But gradually British administrators, proud
that in their own country much of the work of rural government
was still reserved to local notables, began to blend inherited
assumptions and operational experience into a working philo-
sophy of'indirect rule'.
EARLY IMPACTS
The immediate impact of the conquest upon the lives of ordinary

Africans varied enormously. Military operations meant a period
of violence; but it was rarely protracted, and in any case violence
was usually no new experience. Many scholars, reacting against
colonialist historiography, emphasise the elements of continuity
in African life under colonialism, the extent to which Africans
77
Atanda, Tbe new Oyo Empire; n o ; Colonial report, no. 594, Northern Nigeria,
1907-8, 6.
289

successfully preserved whole areas of their culture inviolate from

their new colonial masters.78 For the majority of Africans the real
world remained the ' little platoon' — an ethnic community
defined by tangible connections of family and neighbourhood, by
attachment to plot of land intimately associated with the ancestors
as well as the living kindred. But these were not the static worlds
depicted by romantics as ' unchanging Africa'; in varying degrees,
the peoples of the region were already experiencing inner tensions
or conflicts rooted in their earlier history, and responses to
colonial rule grew out of these experiences.
Abrupt changes did occur. The little platoons were now under
supreme commanders who imposed new disciplines and new
fatigues, though their orders had not by 1905 penetrated every-
where. Severe blows began to fall on the religious and cultural
foundations of many communities. Colonial governments sup-
pressed traditional rituals which involved 'barbarous practices'
like human sacrifice, and traditional procedures for punishing
wrongdoers or settling disputes; missionaries not only attacked
old deities but founded schools, where the young were taught to
question established authorities, and to reject traditional family
structures. Opportunities to earn wages in railway construction
might induce domestic slaves to abandon even the most lenient
of households.
Faced with such bewildering upheavals, some Africans
turned to traditional religion. As their political leadership was
converted into an agency of French tax-collecting, 'political
initiative among the Baule passed largely from the chiefs to the
diviners'. 79 Elsewhere secret societies operating under oath, like
the Poro in 1898, sought to reassert old values by organising
popular resistance. From that same year, Europeans living among
the western Ibo became aware of the ekumeku, a ' secret war cult'
uniting members of local town societies to express hostility to the
new order by attacks on mission stations, as well as on
establishments of the Niger Company. Later attacks were
concentrated on the newly created native courts; prolonged
fighting over more than a dozen years was necessary to control
the movement. A missionary observer described it as the secret
78
Cf. J. F. A. Ajayi, 'The continuity of African institutions under colonialism', in
T. O. Ranger (cd.), Emerging themes in African history in Proceedings of the International
Historical Conference held at Dar es Salaam, if 6} (Nairobi, 1968).
™ Weiskel, Battle Peoples, 217-30.
290

EARLY IMPACTS
police force of local chiefs weary of the European presence; 'one

cannot deny', he added ruefully, 'that the Ekumekus form the
pick of the most capable young men'.80 Other Ibo used the
resources of the dibia, traditional diviners and herbalists, as
weapons in their prolonged resistance to the foreign insurgents.81
Yet in general it seems that the resources of African primal
religions were less widely used by West African than by East
African communities as a means of cultural survival. More
commonly they responded by accepting one of the world religions,
Islam or Christianity, and making it their own. Missionary agents
of both faiths had some spectacular successes. The growth of
urban centres provided new mission fields which Muslims
generally tapped with greater success than Christians. In Senegal
al-Hajj Malik Sy, a learned Tijaniyya marabout, offered such
effective leadership to urban populations, and those of certain
rural areas, that the French soon found him a more valuable and
influential collaborator than any of their chiefs. They were slower
to overcome their fears of Islamic resistance in the case of Ahmad
Bamba, whom they twice exiled during the years 1895-1907; yet,
as they later realised, the 'organizational and ideological
salvation ' 82 which he offered to his Mourid disciples had the effect
of easing the adaptation of the Wolof to the new tensions of
colonial rule. In Nigeria too, colonial rule provided opportunities
to complete the unfinished business of the jihads; already by 1913
Muslims were thought to form 3 5 per cent of the population of
the Ibadan-Ife area.83
Christianity was both helped and handicapped by being
identified in African eyes with colonial rule. Its most spectacular
advances were in eastern Nigeria; as Catholics and Protestants
followed the flag into the Ibo interior, they offered spiritual and
practical guidance through new perplexities, and founded schools
which could provide avenues to worldly success. By 1921,
284, 835 Ibo out of four million were estimated to be Christians,
where in 1900 converts had been numbered in hundreds.84 The
80
Archives of the Societe dts missions africaines, Rome: SMA 14/80404/1579, Report
on 'Ecou-Mecou' by R. P. Straub, cf. Philip A. Igbafe, 'Western Ibo society and its
resistance to British rule: the Ekumeku movement, 1898-1911', Journal African Hist.,
1971, 12, 44'~59-
81
Isichei, History of the Igbo PeopU, 1 2 5 - 4 .
82
O ' B r i e n , Mowides of Senegal, 1 3 .
8j
C. G. Elgee, The evolution of Ibadan (Ibadan, 1914).
84
Isichei, History of the Igbo PeopU, i6jrT.
291


Ewe offered another example of Christian advance; here,

dedicated German missionaries to a certain degree succeeded in
distancing their work from that of the colonial administration.85
In almost every case the practical services rendered to Africans
by mission schools, and to a lesser degree by clinics and hospitals,
were essential elements of Christian growth.
The world religions often flourished best where they assumed
distinctively African characteristics. Among the Yoruba, where
Protestant mission churches were not only deeply rooted among
the educated repatriates of Lagos and Abeokuta but increasingly
established themselves in the inland states, the period of imperial
conquest was also one of schism. Thoughtful African Christians
were deeply troubled not only by the political and economic
aggression of Christian Europe, but by the increasing rigidity of
evangelical missionaries over polygamy and other central features
of African culture, and by the strong racial overtones which
increasingly marked clashes between European and African
churchmen over questions of church order and discipline. First
Baptists, then Anglicans and, eventually in 1917, Wesleyans,
seceded in strength from the Lagos mission churches.86 Their new
churches remained in the mainstream of Protestant tradition, but
their insistence that ' to render Christianity indigenous to Africa
it must be watered by native hands, pruned with the native
hatchet, and tended with native earth', often made them more
effective as evangelists.87 A little later, still more striking successes
would be achieved by Christian prophets offering charismatic
presentations and interpretations of the Gospel; W. W. Harris,
who began his remarkable mission to the southern Ivory Coast
in 1913, and Garrick Braide, who began to achieve mass con-
versions in the Niger delta about 1915, were both working
among populations accustomed to religious change.88
In 1905 the tide of colonial rule had not completely submerged
the West African sub-continent. Although European maps
showed the territorial partition as complete, there were still many
areas which had yet to see a white man or to feel the impact of
85
H. W. Debrunner, A church between colonial powers (London, 1965).
86
Webster, African churches, part z.
87
D . B. Vincent [Mojola Agbebi], Africa and the Gospel (Sierra Leone and Lagos,
n.d.), Sermon in Bamboo Chapel, Lagos, 7 April 1889, 9.
88
G. M. Haliburton, The prophet Harris (London, 1971); G. O. M. Tasie, Christian
missionary enterprise in the Niger delta, 1864-1918 (Leidon, 1978).
294

EARLY IMPACTS
the new regime; in desert regions of Mauritania, Mali and Niger

French military control would not be established until the 1920s.
Nearer the colonial capitals, penetration of the lands of the Baule,
the Ibo, the Tiv was still very uneven; and even in countries where
the French or British had decreed the imposition of direct
taxation, the extent to which it could be assessed or collected
depended on local conditions. Experience of resistance, like that
in Sierra Leone in 1898, made the new rulers reluctant to test their
mastery too far.
Economically, the conquest was beginning to show results.
Railways designed to channel trade towards the oceanic ports
were well advanced in every coastal colony except the Gambia
(where the river served the purpose). The French completed the
Senegal—Niger link in 1905 (though until 1924 it was connected
to the coastal lines only by unreliable river transport); Conakry
was linked with Kankan in 1914, Abidjan with Bouake in 1912,
Cotonu with Save in 1911. The Germans had begun lines from
Lome to both Palime and Atakpame (1913). Kumasi was linked
with Sekondi by 1903; the Lagos railway reached Ibadan in 1901
and was extended to Jebba between 1905 and 1909; the Sierra
Leone railway was completed to Pendembu, near the Liberian
border, in 1908. At the same time port facilities were being
improved, and the circulation of modern monetary currencies
encouraged; foreign merchant and shipping concerns increased
their capital resources by mergers or pooling agreements, and
inter-colonial banking systems established for both British (1894)
and French (1901) empires.
In partial consequence of such changes, figures for the export
of primary produce and the imports of European manufactures
had begun to show gratifying expansion. Although to some extent
this may have represented the diversion of resources from
production for the trans-Saharan trade or the regional exchanges,
there seems little doubt that these trade figures reflect increases
in Gross Domestic Product. The responsibility of colonial policy
for this remains debatable. A stimulating history of the southern
Dahomian economy suggests that production was already ex-
panding steadily before the French conquest and that the con-
tinued annual growth rate of about 2.5 per cent owed more to
' domestic and international economic factors' than to government
action; the principal effect of French taxation and labour policies
2
95

may have been to remove much of the surplus from private

Dahomian hands in order to create the new infrastructure of
transport and administration. Otherwise, Dr Manning sees no
revolutionary change: 'The methods of production, the types and
quantities of goods produced and consumed and the economic
organisation of society all remained much the same on the eve of
World War I as they had been on the eve of the French
conquest. ' 89 Moreover, although in many colonies (notably
Nigeria) African entrepreneurs could find opportunities to profit
from the expanding capitalist sector, distributive roles which in
the nineteenth century would have been filled by African
middlemen, increasingly passed into the hands of European firms
or of Lebanese and Syrian immigrants.
Although, within the period of this volume, many West
Africans had felt the impact of colonial rule only superficially,
imperial apologists were justified in claiming that the conquest
was opening a new period of African history. Africans had become
subject to new, often remote, authorities whose significance for
their lives would inevitably grow. The new colonial unities were
defined territorially (except in the case of internal boundaries
within the French West African 'federation') by European
treaties: politically by their subjection to new bodies of laws (still
limited in their scope and rudimentarily applied); administratively
by the presence of new alien bureaucracies. By 1905 most, though
not all, West Africans had experienced colonial government as a
force which taxed, demanded labour or military service, and in
certain cases judged and punished; some knew it only too well
as an armed conqueror. Yet in certain essential respects the
conquerors showed some restraint in their demands. In most areas
administrators tacitly condoned the continuation of domestic
forms of servitude for transitional periods, though their own
demands for labour might in some localities be undermining the
control which the owners could exercise in practice; and, despite
African concern over complex measures of land legislation,
relatively few West Africans were forcibly removed from their
family lands.
As colonial rule continued, these inchoate new politics would
increasingly develop distinctive identities, and with them that
*• Patrick Manning, Slavery, colonialism and economic growth in Dahomey, 1640-1960
(Cambridge, 1982), 217-25.
296

EARLY IMPACTS
social cohesion which is 'rooted in the conflicts between men's

different allegiances'.90 Increasingly, Africans recognised possi-
bilities of realising their own aims and advancing their own
interests within and through the new colonial order - not merely
the weak who had originally welcomed protection against strong
neighbours, but chiefs and aspirants to chiefly office, traders,
court-clerks, teachers, letter-writers, and interpreters. Even those
whose former hopes and aspirations seemed to have been rejected
accepted the new system; despised 'wogs' like Joyce Cary's
Mister Johnson, could still take a perverse pride in (besides
deriving certain perquisites from) the lowly status which the new
empires reserved for them.
But it was not merely collaborators and opportunists who
found their activities increasingly focused within the colonial
state; the prospect of its reform and ultimate capture provided
prophets of African nationality within the petty bourgeoisie with
concrete objectives they had previously lacked. Sir Samuel Lewis,
a conservative lawyer who had acted as loyal opposition within
the narrow compass of the Sierra Leone Colony, turned his
formidable powers of criticism upon the administration of the new
Protectorate, though with less success. Senegalese citoyens, like the
letter-writer, Modi Mbaye, and the young Lamine Gueye, could
by 1914 create political alliances which secured the election to the
French parliament of Blaise Diagne, a black deputy whose
connections, like theirs, extended beyond the four communes.91
African lawyers in the Gold Coast not only earned large fees for
litigation over development lands in the colony, but began to find
wider political roles in championing African land tenure through
the Aborigines Rights Protection Society, founded in Cape Coast
in 1897 with the explicit aim of interpreting government Bills to
those unable to read them.92 Thus, at the very time when the
vaguer aspirations of the nineteenth-century West African national
bourgeoisie were being shattered, new though distant horizons
were appearing within the boundaries of the new colonial states.
90
M. Gluckman, Custom and conflict in Africa (Oxford, 1959), 4; cf. D . A. Low, Lion
rampant (London, 1973), 2^-6.
" G. W. Johnson, Black politics, passim; Lamine Gucye, Itineraire africain (Paris,
1966), 1 5-jo.
91
Kimble, Political history of Ghana, chap. 9: 341-2 for public statement of April
1897. For litigation, see, e.g., Simensen, 133-6.
2
97

CHAPTER 6
A. FRENCH CONGO AND GABON 1886-1905

In 1885 the European opening up of Gabon and Congo had only
just begun. Following the first two missions of Savorgnan de
Brazza (1876-8 and 1879-82), and the ratification of the Makoko
treaties, which recognised the French protectorate over the right
bank of the river (the county of the Teke), France had entrusted
to the explorer the task of effectively taking control of the
territory (1883-5). It w a s n o t by chance that the attitude of the
population, elsewhere at times hostile to the white conquest, here
proved to be on the whole favourable; trade had preceded the flag,
and the occupation of the hinterland had immediate economic
repercussions. The Kande and the Duma, who had in their hands
the monopoly of traffic on the Ogowe, the Teke of the plateau
and Stanley (Malebo) Pool, and the Bobangi on the Congo river,
had long since left behind the stage of economic self-sufficiency
in favour of an economy based on long-distance trade. By 1885
the slave trade had been replaced by a varied trade in goods which
were expedited towards the Atlantic coast (ivory, dye-woods, and
then rubber).1 These populations with an outward-looking
tradition were thus favourably disposed towards the new
economic currents which seemed likely to fit in easily with
traditional networks. Some groups immediately made an effort to
take advantage of the situation, such as the Teke allies of Brazza,
or the Fang on the Ogowe, whose first migration had reached this
river in 1879.2
However, the calm was short-lived. As soon as the ' discovery'
phase was completed, the French state undertook the 'develop-
ment ' of the country. The intervention of metropolitan France
• The slave trade in war captives was still current between tribes. In 1877, Captain
Marche visited a camp of shackled slaves. But this trade was merely residual. A. Marche,
Trots voyages dans tAfriqtu occidental/ (Paris, 1879), J26.
2
Noted at Lambarene by Dr Nassau, My Ogowe (New York, 1914), 296-7.
298

FRENCH CONGO AND GABON, 1886-97
INDEPENDENT STATE
OF THE CONGO
fe
' Limit of the concession
of Haut Ogowe granted
to Oaumas in 1894
I Dense forest
V * Marsh
0 200km
iJ French Congo and Gabon at the beginning of the twentieth century
in the archaic and brutal form of the regime concessionnaire (con-

cessionary grants of land) copied from the Leopoldian model,
soon resulted in the upsetting of the fragile pre-colonial balance.
The heedlessness of the feeble administration was paralleled by
the combined ill effects of ever more demanding and remote
military operations, coupled with commercial activities of a
dubious nature, the results of which were extremely destructive.
In this new context the trading peoples proved the most difficult
to integrate into a colonial economy. Whether passive or
rebellious, they were rapidly pushed aside as the price paid for
intensive yet fruitless exploitation, the worst excesses of which
(taxes, military repression and porterage) were in fact brought to
299

an end just at the close of the period under consideration, as a

result of the Congo scandals of 1905.
EARLY COLONIAL CONTACTS 1886-97
Our study begins at an ideal moment. For the first time indigenous
peoples found themselves face to face with Europe, and modern
economic practices were making inroads into traditional society.
Only the Mpongwe, along the coast around Libreville (founded
in 1849) and a very few other trading posts, were accustomed to
acting as middlemen in a system of brokerage based on advance
payments in trade goods by the fifty-two commercial establish-
ments spread along the Gabon coast.3 The major drawback of the
system was that it placed between the producer and the purchaser
a whole series of intermediaries, whose commission sometimes
increased the price of the commodities concerned by 50 per cent.
Lately the effects of Senegalese competition had made themselves
felt. The Senegalese were former laptots (boatmen), who, having
completed their service, found employment with the largest
concerns, whether British (John Holt or Hatton and Cookson),
German (Woermann), or, most recently, French (Conquy).4 In all,
these concerns employed about 600 people.5
Upriver from the island of Lambarene, with its enviable
situation at the crossroads of routes from the Ogowe, the
Ngounie and the lakes, the peoples dwelling along the river (the
Kele, Kuta, Pinji, Kande, Duma and others) had up till that time
enjoyed virtually exclusive rights along their stretch of the river.
The opening up of the Ogowe assisted the progress of the Fang,
newcomers encouraged by the administration, of whom Brazza
wrote, ' Here as on the coast they are destined to take everything
over, for they have on their side vitality, courage, strength and
numbers. ' 6 Within a few years the Fang, in their turn, became the
indispensable intermediaries, contributing to the three- or four-
fold increase in the import trade, particularly over land.7
3
Twelve in Ogowe, two in Ngounie, one at Cape Lopez, and the rest at Libreville.
4
About eighty Senegalese were trading on the lower river in 1883; Nassau, My
Ogowe.
5
Letter from Dutreuil de Rhins to Maunoir, 28 May 1883, Archives Nationales,
S.O.M., M.O.A., iv.
6
Brazza, Commercial report, Madiville, 20 Aug. 1885, S.O.M., Mission - 38.
7
zoo tonnes (400 Fang canoes), according to Brazza, went up the river each year,
for a population estimated at '25-50,000 souls', ibid.
300

Above the Bowe rapids, the country was already being drawn
into the sphere of the Congo basin. The Teke group, isolated
on a ridge of sand whose meagre resources had long ago forced
them to turn to trade and travel, provided the link between the
upper Alima and the Congo river. But at the heart of the system
were the Bobangi canoe men, then called Likuba, who had arrived
scarcely two generations before;8 their mastery of the middle
Congo and its main tributaries - from the lower Ubangi as far
as the approaches to the Pool - enabled them to tap the wealth
of the country, and in return to despatch into the interior
the goods acquired from the Kongo middlemen on the lower
reaches of the river. They combined their international trade with
a complementary interregional one, for they furnished themselves
with cassava by trading the products of their local industries, mats,
pottery, cutlasses, nets and in particular dried fish,9 with the
landsmen of the Alima, particularly the Mbochi.10 Hence the
numerous prosperous villages of 2,000 to 4,000 inhabitants
spread out along innumerable lagoons, in an area that is now
deserted as a result of syphilis and sleeping sickness."
This Bobangi dynamism probably accounts for the haste with
which the Teke make themselves allies of the French: they saw
in this an opportunity to bring - or bring back - prosperity to the
markets of the interior. Even though there was never a ' Teke
empire' in the Western meaning of the phrase, and the power of
the Makoko, the great and famous paramount chief, was never
much more than symbolic, it nevertheless seems to be the case
that at the time of Brazza's arrival the country was already in
difficulty, excluded from long-distance trade simultaneously by
two dynamic forces:12 upstream by the Bobangi, and downstream
by the Kongo, whom the decline of the Teke allowed gradually
to settle on the borders of the plateau overlooking the river.13
Thus, around Ncuna (on the Pool), the terminus of the
immense navigable network of the Congo, the Kongo controlled
8
The grandfather of the then reigning Makoko was said to have stopped their
invasion: letter from Ponel to Dufourcq, 30 June 1885, S.O.M., M.O.A., VII.
Q
On the excellent description of this 'Civilisation of the Rivers', see G. Sautter, De
I'Atlantiquc auflcuvcCongo (Paris, 1966), 274-6.
10
About 40 tonnes a day went down-river in the dry season (between April and
September) in convoys of ten to twenty canoes loaded to the gunwales, ibid., 259.
1
' Ibid., and see C. Coquery-Vidrovitch, Bra^ga it la prise de possession du Congo (Paris,
1969), 105-10.
12
Cf. Jan Vansina, The Tio kingdom of the Middle Congo (Oxford, 1975).
13
G . Balandier, Sociologie actutlle de tAfriqut noire (Paris, 1965), 185.
301

the great transit markets from the Atlantic at the focal point of
the caravan routes on either side of the lower river: to Luanda
in the south, and Loango in the north. To be sure, the cohesion
of the old Kongo kingdom was but a memory at that time. To
the north of the estuary, the roadstead of Loango, the best on the
coast, had just been occupied by the French (1883).I4 But the
hinterland (the Mayombe massif and the Niari plain), despite the
still recent ravages of the slave trade, seemed, in the eyes of the
early administrators, rich in foodstuffs, relatively densely popu-
lated, and accustomed to commercial dealings with the merchants
settled in the villages15 — the only exception being the moun-
tainous areas peopled by hostile refugees (Yaka and Nsundi on
the right bank).
Social and political disintegration took place rapidly following
the operations of the conquest, and the rapid proliferation of
European commercial enterprises16 had disrupted the main
traditional trade routes. In the area of the townships, former
Mpongwe or Vili traders were transformed into 'boys', minor
employees and shop assistants. On the Congo the numbers of the
Likuba, deprived of their monopoly of navigation, fell from
12,000 to 4,000 within a few years. The Kongo and Vili, who had
the reputation of being tractable, became a useful source of labour
as workers in the first cocoa and coffee plantations and, in
particular, as porters along the caravan routes, but were soon
decimated by sleeping sickness.
Although as yet relatively unknown, the northern area was
nonetheless undergoing a profound change. Distance and the lack
of means of transport bore heavily on the system of communica-
tions, which had been dealt a grievous blow by the decline of the
Bobangi. The brutality used by the occasional expeditions
attempting to penetrate further inland than the Bangui rapids were
a factor in driving away the population from the river banks. The
Sabanga, Mbaya and Ndy farmers, and then, on the interfluve
between the Ubangi and the Gribingi, the Maja people, all of them
14
Coquery-Vidrovitch, Braqga..., 159-61. As far as Pointc Notre, the future
terminus of the Congo-Ocean Railway, it was chosen simply 'to prevent the others
[Belgian or Portuguese] from settling there [for in the opinion of everyone] it's a
country with no future': the Captain of the Voltigeur to the Commandant of the Div.
Navale, Libreville, 30 June 1883, Arch. Marine, BB 4, 1943.
15
Cf. Coquery-Vidrovitch, Braâ..., 167.
16
On the lower Congo: France: Daumas-Beraud, Belgium: Societe Anonyme Beige
pour le commerce et l'industrie (S.A.B.). Netherlands: Nieuwe Afrikaansche Handels
Vennootschap (N.A.H.V.).
302

FRENCH CONGO AND GABON, 1886—97
scattered in villages without any central organisation capable of

resisting penetration, were thus sacrificed in the cause of imperial
objectives on the Bahr al-Ghazal and the Nile.
Beyond all these, the animist peoples — the Ngao, Boubou,
etc. - were periodically raided by the sultans from the north-east,
who had settled there in the last quarter of the ninetenth century
as a result of the political upheavals following the Egyptian
conquest of the Upper Nile. Such were the sultans of Bangassou,
Rafai (linked to Zubayr) and Semio on the upper Ubangi or,
further north, the SanusI sultan of Ndele: most of them were of
foreign origin and supported themselves by slave raiding and
subjugating by military means the indigenous peoples, the Nzakara
and the Zande, whom they forced to collect ivory and rubber for
the Europeans in exchange for firearms.17 It was in 1895 that the
French administration took over from the early Belgian occu-
pation of the upper Ubangi. But after the junction of French forces
at Lake Chad in 1900 (the battle of Kousseri and the death of
Rabah), it was not until 1909 that the colonial influence resulted
in treaties of protectorate with these ' allied chiefs' drawn up in
due form. As for the Logone basin, it was still the scene of
an active east-west trade in slaves and livestock from the Sudan
to the lamidat (Fulani chiefdom) of Ngaoundere (Cameroun) and
northern Nigeria; at the beginning of the century in these regions
approximately ten thousand people still disappeared annually,
killed in raids, sold as slaves, or dying on their way into exile.18
At the end of the century, even though European penetration was
still thin on the ground, nevertheless by the very fact of their
presence local conditions were profoundly altered.
Apart from a few military expeditions, the element that most
changed the internal equilibrium of the country was the interplay
of rivalries between the various peoples as a result of European
influence. For 'colonisation' as such remained very limited for a
long time: the administrative presence was virtually nil, and
'development' non-existent. In 1885 the administrative establish-
ment, not of a high quality, consisted of thirty-six Europeans and
a handful of auxiliaries.19 On his return to France, Brazza had
" Cf., e.g., E. de Dampierre, Un royaume ftandia du Haut Oubangui [Bangassou],
Paris, 1967.
18
G. Bruel, 'Rapport sur l'occupation du Cercle du Moyen-Logone', *8 Sept. 1903,
Arch, de Djamena.
19
13 Algerian tirailleurs, 60 Senegalese laptots, and 45 Kru, plus 500 local workers,
of whom 310 were Vili. See Coquery-Vidrovitch, Braâ..., 182.
3°3

imprudently claimed that he could administer his conquest, with

an area bigger than that of France, with a mere million francs a
year.20 At a time when the raison d'etre of a colony was to enrich
the mother country, and public opinion had been wooed with the
idea of the supposed wealth of the Congo, parliament took him
at his word. 21 By the end of our period the situation was not much
better: in 1904 the number of administrators and equivalent ranks
had fallen to thirty - a figure which did not improve until after
the administrative reform of 1906-7. Two thirds of the budget
was taken up with administrative expenses. The lack of skill of
the personnel was all the more obvious in that, following the
wishes of Brazza, this was the first colony with a civilian
government at a time when no colonial administrative training
yet existed.22 The spectacular excesses revealed by the Gaud and
Toque scandal23 — which were the cause of the urgent recall of
Brazza for his last mission (1904-5) - laid bare the evils of a
cheese-paring system, in which the bulk of the work was left to
unsupervised auxiliaries in the shape of a local militia set up in
1897, and organised into 'veritable assault columns'. 24
In 1900 the total strength of the white community (officials,
merchants and missionaries, plus a very small number of women
and children was merely 800, of whom most were stationed at
Brazzaville (248), Pointe Noire and Libreville.25 Religious pene-
tration, restricted to the coasts and the course of the main rivers,
scarcely made up for the inadequacies of the health and education
services. The French evangelical mission, which in 1875 had taken
the place of the American Presbyterians, consisted in 1898 of a
mere two stations in Gabon. The Catholic mission of the Holy
Ghost fathers played an active political role, with the outstanding
personality of Prosper-Philippe Augouard, who had arrived at the
Pool almost at the same time as Brazza, and was the apostolic vicar
20
S p e e c h at t h e Cirque d'Hiver, 21 Jan. 1884, 24.
21
F o r t h e f o l l o w i n g years, they allocated 1.5 million francs for the C o n g o and
6 0 0 , 0 0 0 francs t o G a b o n , o f w h i c h almost all were a b s o r b e d in salaries, supplies and
military e q u i p m e n t . Loi d u 11 a o u t 1886.
" T h e E c o l e C o l o n i a l e was o n l y established in 1890.
23
T h r e e administrators o f Fort Crampel, the p i v o t o f caravan routes t o Chad, were
c o n v i c t e d o f terrible c r i m e s against Africans. O n e o f them in particular had b l o w n up
a native w i t h d y n a m i t e t o celebrate t h e Fete National, 14 July 1903. J. Saintoyant,
Uaffaire du Congo, rfoj ( e d . Ch.-A. Julien, Paris, i 9 6 0 ) .
24
D e c r e e o f t h e c o m m i s s a i r e general d u G a b o n - C o n g o , j o N o v . 1897, J. Officiel du
Congo Franfais, 1 Feb. 1898, 5.
25
2 0 0 0 in 1 9 1 1 .
304

FRENCH CONGO AND GABON, 1898-I905
from 1890 to 1919. An ardent nationalist, he was ahead of, rather
than in step with, French penetration.26
Lacking money and personnel, the colony finally opted for
repressive policies, exercised over scattered and disparate peoples
whose stability had been undermined for fifteen years by the
arrival of the Europeans. In 1897 Brazza was removed because of
his financial negligence. But what could he have done? His de-
parture marked a turning-point. Renouncing officially the dream
of 'peaceful colonisation', France wanted to make clear her
intention of developing the resources of the country, following
the norms which had just been so successful in the neighbouring
Belgian colony. But the solution that had been advocated, that
of privileged companies, was to prove a worse evil. How gravely
and rapidly the expectations of the' inventors' of the concessionary
system were disappointed, we shall see. It is not merely a question
of recalling the abuses of the system (excessive porterage, camps
of hostages virtually exterminated, even massacred). These abuses
were not exceptional or isolated, as people liked to say at the
time — the activities of evil characters or people suddenly afflicted
with madness under the combined effects of the climate, alcohol
and solitude. They were the inevitable manifestation of the
harmfulness of the system that had been installed.
ECONOMIC EXPERIMENTS 1898-1905
The idea of an economic conquest based on the opening up of

great penetration routes with the aim of' linking the mouth of
the Congo to Upper Egypt across Africa'" went back to Brazza.
After witnessing the failure in 1882-3 of a first attempt at opening
a railway line in Kouilu, to connect Brazzaville to the coast, he
turned his energies to the development of the waterways of the
Ogowe and the Niari, trying to interest private companies who
might invest in the colony, since the state would not. In 1894
Brazza granted to Daumas-Beraud - the only French concern of
any size - a concession covering an immense area in eastern
Gabon. The Societe du Haut Ogooue received 11 million hectares,
in return for a monopoly of trade, and also governmental powers
16
About twenty Fathers ran eleven schools with 750 children in Gabon.
27
Brazza, Note sur Us votes de communication..., s.d., [igiff], Arcb. Nat., S.O.M.,
Gabon-Congo, x n , 90a.
3°5

200 km *Dem-Ziber
5°/V
24 Compagnw commarciata d*
Colonisation du Congo trancais
1 Sociata da I'Afnqut francttte 25 Sociata das tactoranat da
2 Compagnw francaise du N'DdjoM
Haut-Congo 26 CompagnN da la
3 Compagnia da U Sangha Hauta-N'Gounia
4 Compagnia das produit* da la 27 Social* da t'Ongomo
Sanga Ltpa-Ouasso 28 Sociat* commafctala.
5 Sociat* da tEkata-Sangha induimalta at agiicola du
O 8 Sociat* commareiata at agricola Haui-Ogooua
CN da la Kadat- Sangha 29 Compagnia agricola at
7 Sociat* da lAfriqut aquatoriala cotoniala at mdlla da la Laf.n.
8 Compagnia das Caoutchouc! 30 Compagmtdu Kouango
ei produits da ta Lobav francai*
9 Sociat* da la Haut a-Sangha 31 Sociat* commarciala cotonista
10 Sociata da la Kadai Sangha dalaMambara-Sangtia
11 Sociata da I'OgoouaN'GouniA 32 Compagnia da la Mobay*
12 Compagnia franc*)* du Congo 33 UKono
13 Sociata igricola at commarciato 34 Compagnia du
da lAltma Kouango -Oubanghi
14 Sociata coloniaia du Bamambe 38 Sociata agricota at commarciala
15 Sociata da I'lbanga du 8aa-Ogooua
16 Compagnia franco-congol»i» 36 Compagnia du Bavili-MBanio
da la Sangha 37 SocMta das Suttanats du
17 Sociata daa atablisaamants Haut-Oubanghi
Gratry M'Poko 38 Compagnia da la N'Gofco
18 Social* da la Sangha Ouasao
aquaiofiate 39 Compagnia frtncaaa da
19 L'Alimanwnna rOubanghi-Ombalia
20 Compagnia ganarala du 40 Sociati biatonna du Congo
Famand Vai 41 Compagnia propriataw du
21 Soctttadala K'Kaniat NKama Kouilou Nian
22 Sociata da la San* Cama 42 Compagnia francaisa da
23 Compagnia frsncaiM du Congo rOuahm* at da la N«na, 3 Rua
occidanul d'Atgar
12 The Concessionary system in the French Congo

in matters of taxation and policing which were to be withdrawn

in 1897. As for the line to Brazzaville, he negotiated for its
construction with Credit Foncier (1889), then with the business-
man A. le Chatelier (189$), who finally ceded his rights in 1899
to a Belgian concern with the name of Compagnie Proprietaire du
Kouilou-Niari (bought out in 1911 by the Lever group).28 But the
railway project suffered for many years as a result of the
competing Belgian line, opened between 1890 and 1898.
The system advocated by Brazza was organised on a large scale
immediately after his removal, at a time when he was himself
beginning to appreciate the dangers of this course. The idea had
made progress in metropolitan France29 and abroad.30 Given the
dreadful state of the finances of the colony, the apparent success
of the Leopoldian system after 1896 made the decision inevitable.
The military expeditions to Chad swallowed the entire budget.
The only means of penetrating a virtually unexplored forest was
through difficult rivers, broken up by rapids. How could this
underpopulated country whose inhabitants, apart from a few
coastal peoples, knew nothing of money, have been capable of self
sufficiency? The only source of revenue lay in the tiny customs
duties, which were limited by the free trade area of the Congo
basin. The development of the colony demanded a considerable
investment in men and in capital and every kind of infrastructure.
The only perceived solution was to hand the country over to the
businessmen and let them make the necessary effort at their own
expense.
In the face of parliamentary hesitations, Andre Lebon, the
minister for the Colonies, put the process into effect by decree in
1896. This allowed his successors in 1899-1900 to grant forty
concessions, which fragmented the territory. The biggest, the
Compagnie des Sultanats du Haut-Oubangui, comprised
140,000 sq. km; the smallest, 1,200 sq. km. The only success of
the radical opposition was to obtain the rejection of the principle
of chartered companies with full governmental powers. Hence-
forth the term was entreprises de colonisation, which were granted
a thirty-year monopoly of the products of the soil (mainly ivory
28
Cf. Coquery-Vidrovitch, Le Congo au temps des grandes compagnies concession/wires
iSfg-ifjo (Paris/The Hague, 1972), ch. 1.
2
« Cf. the popular work of the economist Paul Leroy-Beaulieu, De la colonisation cbe^
les peuplcs modernes (Paris, 1882, 2nd edn), and the colonial role of politicians with
business connections, such as Eugene Etienne and Binger.
30
Great Britain and Germany had just increased the number of chartered com-
panies.
3°7
and rubber), in return for a fixed annual payment to the state plus
15 per cent of the profits.
Although only lands actually exploited were to become the
absolute property of the company at the end of the period of the
contract, a confusion rapidly arose between the ideas of exploitation
(i.e. utilisation) and ownership. The companies were not interested
in investment, and their objectives were purely commercial. They
refused to pay the producers the value of their produce, and paid
them only for their labour in collecting it, thus causing prices
to be very low, less than two-thirds of the prices current in the
open market. The colonial government tended to take no interest
in infrastructural expenses, since the companies were expected to
devote part of their profits to public works. As for the Africans,
they quickly learned that there was no real distinction between
a military commander and a merchant.
In principle, the Africans still retained the property rights of
their villages and forest lands, pastures and crops, that were
' reserved' for them, but the boundaries of these reserves, the
object of inextricable difficulties, were never properly drawn up.
The African, like his produce, rapidly came to be seen by the
concessionary companies as a possession. From the earliest days,
the specific ' rub' of the system was the lack of both competence
and capital. Unlike the Leopoldian concerns, large-scale French
investors were not anxious to launch themselves into an enterprise
that was considered risky. The concessionaries were either settlers
of limited means (such as the five Trechot brothers of the
Compagnie Francaise du Haut-Congo) or, at best, industrialists
from sectors of the economy of metropolitan France that had lost
their momentum, such as the northern textile industry, who
therefore were seeking to unload their unsold stock (such as the
Gratry brothers of the Compagnie de la Mpoko and the Al-Ke-Le
group). Given the vast extent of the territory, the capital invested
remained minimal: from 300,000 francs (Compagnie Bretonne du
Congo, 3,000 sq. km) to 3 million francs (Compagnie Francaise
du Congo, 43,000 sq. km) or, exceptionally, 9 million francs
(Compagnie des Sultanats). In all, the capital sums effectively
employed did not exceed 40 million francs, and the concessionaries
refused to invest voluntarily, out of their profits, funds that the
state had failed to supply.
308

The declaration of bankruptcy was made in 1904-5. In 1902

only one company was in a position to pay dividends; there were
two in 1903, and seven, the maximum number, in 1905. By 190}
about ten companies were already on the way to extinction, and
three of these had been granted land which, as a result of an
inadequate survey of the terrain, existed only on paper! As a
result of the scandals, the French government tried to restrict the
privileges of the concessionaries, but experience demonstrated the
complete inadequacy of local control, and the lack of an adequate
legal framework. It is true that in theory any serious default could
be punished by forfeiture, but since the occasional attempts of the
state to do this resulted in the cases being thrown out,31 it was
decided to deal with the companies by negotiation and
compensation.
The only companies that survived were either purely speculative
operations, allowing vast territories legally to lie fallow (for
example the Compagnie Ngoko-Sangha, which for years made a
profit from the Franco-German frontier dispute, until a scandal
broke out in 191i32), or concerns that owed their success to the
extreme severity of their methods (such as the CFHC, the
Compagnie des Sultanats, and a group of eleven companies which
in 1911 gave birth to the consortium called the Compagnie
Forestiere Sangha-Oubangui).
The concessionary regime, particularly deleterious in that it was
based on meanness, encouraged the most archaic practices,
showing up, in all its absurdity, the impasse to which French
policy was leading in the Congo. But it may be questioned
whether a policy of open competition would have done any better,
given the specific problems of the country (the climate, the forest,
and the sparse population), and in particular the refusal by the
government to participate in investment. Probably the only
difference would have been a less close association between the
administration and the private sector, an association which
31
A special g o v e r n m e n t c o m m i s s i o n e r for the concessionary c o m p a n i e s had b e e n
appointed in 1902, b u t for lack o f means h e vegetated until 1 9 0 ; . In 1906 o n l y six
c o m p a n i e s had been the object o f a summary enquiry. T h e First attempt at disqualifi-
cation w e n t back t o 1902 (against the M o b a y e c o m p a n y ) . See C o q u e r y - V i d r o v i t c h , Le
Congo, 2 7 6 - 7 .
32
M. Violette, La Ngoko-Sangha (Paris, 1914), and C o q u e r y - V i d r o v i t c h , Le Congo,
315-22.
309

resulted in the exploitation of the local inhabitants, who were
decimated, exhausted, and lastingly affected.
The Africans, deterred by prohibitive prices, by harvesting
techniques of which they were totally ignorant, and by their own
very limited needs, refused to take part willingly in the new
dispensation. Paralysed by the shortage of labour, the conces-
sionaries demanded the institution of forced labour. Although this
was denied in principle, it was nevertheless introduced in the form
of a poll tax, which was tried out in a restricted area in 1894,
properly instituted in 1897 and given a wider scope in 1900.33 The
idea was not new - Gallieni had already made use of it in
Madagascar. But nowhere more than in the Congo was it
developed as a colonial principle, as the only means of forcing the
native to 'produce and make use of the riches of the soil' by
imposing on them' regular methods of work, without which there
could be no proper exploitation of the concessionary lands'. 34
Financially the tax was a failure. At the very least it required a
correct estimation of the number of inhabitants;35 but the local
people soon recognised the purpose of the administrator's tax-
collecting rounds, and the villages disappeared into the forest.
Nevertheless the poll tax, which brought in only 60,000 francs in
1902, almost tripled the following year, and reached 770,000 francs
in 1905, before arriving at a ceiling of a million francs, scarcely
one-fifth of the total revenue of the colony.36
The administration was driven to make use of coercive methods
to mobilise all the country's resources. Wherever the troops had
to be regularly provisioned with foodstuffs and goods, porterage
took its toll. The nerve centres were the two interfluvial zones:
the Lower Congo (from Brazzaville to Loango) and the Upper
Chari (towards Chad). On the lower river this effort came to a head
in 1896—7, when the passage of the Marchand mission had to be
provided with porters for 3,000 loads at a few weeks' notice.
33
C o q u e r y - V i d r o v i t c h , Le Congo, ch. 5.
34
T h e Minister for the Colonies t o the Commissaire-General G r o d e t , 14 M a y 1901,
S.O.M., Cone. xiv-B(2).
35
A l t h o u g h w i l d estimates o f p o p u l a t i o n w e r e made b e t w e e n 1900 and 1912 (from
8 t o 15 million), t h e w h o l e o f t h e p o p u l a t i o n o f French Equatorial Africa hardly
e x c e e d e d 2.5 m i l l i o n o n the e v e o f the First W o r l d War ( G a b o n 4 1 6 , 0 0 0 , M o y e n - C o n g o
8 0 5 , 0 0 0 , O u b a n g u i - S h a r i 380,000, Chad 9 8 8 , 5 0 0 ) . Rolede timpotindigene, 1911-13 Arch.
Nat., S.O.M., A.E.F., ix-8.
36
Rapport de la Commission d'enquete du Congo (Paris, 1907), 30.
310

Necessity became law. The revolt of the Basundi was suppressed
with all speed.37 And the exhaustion of the Kongo and Vili
porters was a factor in the spread of sleeping sickness in Niari,
which halved the population between 1900 and 1910. In the area
near Chad, the Gentil expedition of 1899 began to affect the
country adversely. Since the strategic importance of the region
had led to its exclusion from the zone of concessionary partition,
it was the administration itself which ravaged the Manja territory
by the widespread practice of seizing female hostages and holding
them in camps in order to force their menfolk to collaborate in
supplying labour. The recruiters had to institute veritable
manhunts through empty villages and abandoned plantations:
'The exhausted Manjian can do no more and want no more of
it. At present they prefer anything, even death, to porterage. It
is more than a year now since the scattering of the tribe
began...' 3 8 1903 was the climax: the commander of the Fort
Crampel region at that time was Gaud, nicknamed Niama-
Gounda, 'the wild beast'. His activities were at the root of a
violent insurrection, and caused the recall of Brazza for a mission
of inspection. Although the losses suffered are unknown, it is
possible, according to the evidence of a local officer of the period,
that the population of the Chari region fell by half in two years.39
The construction of a road which was recognised as indispensable
was at last begun in 1904, but at first merely increased the unrest
in the labour force. In the area of the concessions, these
authoritarian methods were all the more deleterious in that they
were applied at a local level by isolated commercial agents who
lacked the colonial ideology of the government officials, and
whose only guiding light was the notion of profit. Unlike the
Leopoldian Congo, with its close association of merchants and
officials, scandals broke out in two particular sets of circum-
stances: either when an unscrupulous merchant was left to his
own devices beyond the reach of any administrative supervision
or, more rarely, when he collaborated with an agent of the
administration who had as few scruples as he himself about the
methods to be pursued. The Lobaye and Ouhame-Nana com-
37
M. Michel, La mission Marchand, /fyj-iffp (Paris, 1972).
38
Report o f 1 Jan. and 1 N o v . 1901, S.O.M., G . C . x i x ~ 4 ( b ) .
39
From 40,000 to 20,000 persons. A Britsch, Hisloire de la dcrnicrc mission
(Paris, n.d.), 27.

panies were early convicted of serious malpractices.40 But the

most serious matter was hushed up: on the Mpoko concession,
near Bangui (one of the few firms to declare a profit from 1904
onwards), the enquiry instituted by a recently appointed young
administrator concluded that there had been 1,500 murders
between 1903 and 1905, committed by forty Europeans at the
head of 400 armed guards whose task was to shoot at sight any
African caught not collecting rubber. A devastating dossier
resulted in 1907 in the citing of 236 people, of whom seventeen
were Europeans. But the whole affair ended two years later with
the case dismissed for lack of evidence.41 Meanwhile the economic
results had been conclusive: a comparison of the monthly
production of the main station on this concession with the number
of shots fired proves the truth of the lapidary formula of one of
the men in charge of the concession: 'the more armed men, the
more rubber'. 42 This was not merely a matter of individual
shortcomings, but of a systematic method of exploitation.
Contrary to accepted ideas, revolts were incessant. In the first
phase, these were mostly insurrections in relatively inaccessible
areas, where the inhabitants simply refused to allow the conquerors
to settle. Such an area was the Upper Sangha, where a precarious
balance existed between the sedentary agriculturalists of the Baya
country and their western neighbours, the slave-raiding Fulani
pastoralists of Adamawa. It was a zone under considerable stress
due to pressures from both north and south: the empire of Rabah
in the north and the incoming Europeans in the south.43 Taxation
was the cause of the most enduring resistance. Everywhere the
peoples reacted against it by passive resistance or emigration. It
was a normal consequence of any repressive operation that the
village concerned melted into the forest and never returned.
Generally speaking, the revolts were mainly sporadic expres-
sions of the exasperation of the Africans in the face of the most
blatant outrages. Following the rising of the Vili on the caravan
40
The affairs of the '119 women of Fort Sibut' and of the '60 women of Bangui'
(of whom 45 died of hunger in five weeks). Cf. Coquery-Vidrovitch, Le Congo, ch. 7.
41
Meanwhile, in order not to aggravate the 'Congo scandals', the administrator in
question had given the Minister for the Colonies his word of honour to remain silent.
S.O.M., Cone. x x x m - A ( 2 ) . Cf. Coquery-Vidrovitch, Le Congo, 177-84.
42
Lund, Rapport mtnsuel, September 1905, ibid.
43
In 1891, the insurrection of the whole of the Bayanda tribe against the first
explorer o f the region, Fourneau, which was crushed only in 1894. Clozel, Rapport
politiqut, 25 Aug. 1895, S.O.M., G.C. m-15. See Coquery-Vidrovitch, 'De Brazza a
Gentil. La politique francaise en Haute-Sangha', Revue franfaise d'bistoire d'outre-mer,
1965, 52, no. 186, 21-40.
312

total no. of shots
400
1905 1906
tonnes
PRODUCTION OF ORY RUBBER
t it tr r n w im it it xi n / u » iv v vi vn via ir xi mi i g m iv v vi VII vu ur * xi *» / H u iv v
liancs
AVERAGE PRICE PER KG OF ORY RUBBER

-
^
' * in iv v vi n k ,'r x xi V//I / h in I'V v vi ia im ir it ii ia ) h k iv \r #7 ¥ii fit IX K if V//| / II M IV V tV ni VII

1904 1905 1906 1907
Consumption of ammunition and the production of rubber at the Salanga

station.
route from 1897 to 1859, one of the next organised revolts was
on the middle Ogowe in 1901—2. This was the revolt of the Fang
chief, Emane-Tole, who tried to close the river in retaliation for
the increase in prices by the Societe du Haut Ogooue.44 The year
1902 witnessed serious troubles virtually everywhere: pillaged
factories, cases of ritual cannibalism, all coinciding with the
setting-up of the concessions and the institution of taxation.45
Very often these were acts of vengeance, some of the executed
agents ' having, sad to say, merited death not once but ten times,
each time [that they themselves had killed] for the pleasure of
causing suffering \ 4 6 The most murderous outbreak of the period
was that of the Manja, who had suffered the ravages of porterage
on the upper Chari in 1903-4.47 What was striking in this normally
fragmented country was the unanimity of the movement, since
all the villages without exception fled to the bush to engage in
guerrilla warfare. Although it is not always easy to discover the
sometimes complex causes of these revolts, some of which took
on the dimensions of large-scale resistance movements (Ngounie,
the Middle Sangha and Manja country up till 1905, Lobaye from
1902 to 1908, and Upper Ngounie from 1903 to 1909), they
incontrovertibly represented a response and a challenge to those
responsible for the overturning of traditional values.
In 1905 the country was sick, the administration enfeebled, the

companies moribund. Scandals broke out in the context of the
anti-Congolese campaign directed mainly against the ' red rubber'
of King Leopold's Congo.48 Henri Brunschwig's recent studies
tend to show that in the beginning the scandal was exploited by
the minister for the Colonies himself, in his desire to impose
reforms which he judged necessary. But he was rapidly over-
whelmed by events. Brazza's mission revealed that the system was
fundamentally pernicious, and that the only means of remedying
the situation was precisely what the French government was
44
' N o t e d'histoire: E m a n e Tole d e N d j o l e ' , Ke'alite'sgabonaises, 1965, n o . 26, 4 6 - 7 6 .
45
C o q u c r y - V i d r o v i t c h , Le Congo, 1 9 8 - 9 .
46
L. T a v e r n e , 22 Feb. 1905, S . O . M . , F o n d s Brazza, 1905-11.
47
R a p p o r t d e T o q u e et Pujol, Fort Sibut, 26 Feb. 1904, ibid, 1 9 0 5 - 1 ; Coquery-
V i d r o v i t c h , Le Congo, 200—1.
48
C o n g o R e f o r m Association o f E . D . Morel. C(. by this author, c o n c e r n i n g the
French C o n g o : ' L e s concessions au C o n g o ' , Questions diplomatique* et coloniales, 1903,
x v i , 4 2 6 ; and ' L a q u e s t i o n c o n g o l a i s e ' , ibid., 1904, x v n , 433. Cf. also the campaign
w a g e d by the Socialist team o f the Cahiers de la quinâine (Charles P e g u y and Pierre Mille).
314

KING LEOPOLD S CONGO: EUROPE
determined not to do, namely to inject technicians and capital in
large quantities, in order to develop the country in real terms
without making a sparse and over-exploited population bear the
burden of an archaic system of pillage. The French government
recoiled from the gravity of the diagnosis. In the national interest,
the ministry of Foreign Affairs refused to publish the final report,
'Conceived in a spirit of absolute sincerity, but also of implacable
severity'.49 The document was never published. The result was
limited to a few administrative reforms, but the Congolese crisis
was not resolved. For this to happen, the monopoly of the
concessions had to end: most of the companies disappeared
during the 1914—18 war, but some of them survived until 193 5.50
In 1905, in this poor country, the development of which would
have required enormous prior investment (plantations, forestry
enterprises and mineral prospection), the twin obstacles of lack
of capital and under-population had not been overcome. Not only
the adversaries of the concessionary regime, but also many of ,
those working in the Congo, knew that this was so. Their reports
tirelessly denounced this failure. If they were not granted credits,
or equipment, or personnel, they would be obliged either to resign
or to impose themselves by force. The government also was aware
of the facts. However, it persisted in its course to the absolute
limit, for reasons such as the conquest of Chad and parliament's
refusal of credits, which went far beyond the Congolese problem,
since they were matters of general French policy.
B. KING LEOPOLD'S CONGO, 1886-1908

IN EUROPE
The fact that the title of this chapter bears the name of a European
monarch is fully significant. The Congo was a political entity
brought into being on African soil completely by the will of one
man, and that man - who never visited his dominion - governed
it from his residence in Europe in a completely autocratic way.
The Congo Independent State (Etat independant du Congo),
under the personal government of King Leopold, lasted from
49
N o t e pour le Ministre d c s C o l o n i e s , confidentiel, Paris, 13 March 1906, S . O . M . ,
G.C. xix- 4 (b).
50
T h e C o m p a g n i e Forcstierc S a n g h a - O u b a n g u i kept the m o n o p o l y o f rubber until
that date.
315

determined not to do, namely to inject technicians and capital in
large quantities, in order to develop the country in real terms
without making a sparse and over-exploited population bear the
burden of an archaic system of pillage. The French government
recoiled from the gravity of the diagnosis. In the national interest,
the ministry of Foreign Affairs refused to publish the final report,
'Conceived in a spirit of absolute sincerity, but also of implacable
severity'.49 The document was never published. The result was
limited to a few administrative reforms, but the Congolese crisis
was not resolved. For this to happen, the monopoly of the
concessions had to end: most of the companies disappeared
during the 1914—18 war, but some of them survived until 193 5.50
In 1905, in this poor country, the development of which would
have required enormous prior investment (plantations, forestry
enterprises and mineral prospection), the twin obstacles of lack
of capital and under-population had not been overcome. Not only
the adversaries of the concessionary regime, but also many of ,
those working in the Congo, knew that this was so. Their reports
tirelessly denounced this failure. If they were not granted credits,
or equipment, or personnel, they would be obliged either to resign
or to impose themselves by force. The government also was aware
of the facts. However, it persisted in its course to the absolute
limit, for reasons such as the conquest of Chad and parliament's
refusal of credits, which went far beyond the Congolese problem,
since they were matters of general French policy.
B. KING LEOPOLD'S CONGO, 1886-1908

IN EUROPE
The fact that the title of this chapter bears the name of a European
monarch is fully significant. The Congo was a political entity
brought into being on African soil completely by the will of one
man, and that man - who never visited his dominion - governed
it from his residence in Europe in a completely autocratic way.
The Congo Independent State (Etat independant du Congo),
under the personal government of King Leopold, lasted from
49
N o t e pour le Ministre d c s C o l o n i e s , confidentiel, Paris, 13 March 1906, S . O . M . ,
G.C. xix- 4 (b).
50
T h e C o m p a g n i e Forcstierc S a n g h a - O u b a n g u i kept the m o n o p o l y o f rubber until
that date.
315

1885 to 1908. In 1908, it was annexed by Belgium and became
a Belgian colony, the' Belgian Congo'. Well before 1908, however,
the term ' Belgian Congo' was very often used by foreign writers.
This can readily be understood when one considers the way in
which the Independent State appeared to them. The sovereign of
the state, they thought, was the King of the Belgians, who was,
as everyone knew, ardently devoted to Belgian interests. The
central services of the state were located in Brussels, and all its
officials were Belgian. As for the Congo itself — in the administra-
tion, in the army, and in the courts - the essential role was played
by Belgians, particularly by officers of the Belgian army assigned
to African service. Even the religious missions — at any rate those
of the Catholics, which were most favoured by the state — almost
all had a markedly Belgian character. How, in such circumstances,
could contemporaries have failed to come to speak naturally of
a 'Belgian Congo'? Nevertheless, the usage was quite improper.
It was contrary both to the situation in law and - what is much
more important — to the situation in fact. In law, before 1908,
Belgium and the Congo were two absolutely distinct states
without any common organ; their sole link rested on the fact that
they had the same sovereign. But Leopold II distinguished very
clearly between his role as sovereign of the Congo and that as king
of the Belgians. In Belgium he was a constitutional monarch and
played with perfectly good grace the game of constitutional
monarchy in the English fashion. He submitted the texts of his
speeches to his ministers beforehand, in accordance with the rules.
But the same ministers could only learn by reading the newspapers
that the Congo had leased the Upper Nile or concluded a treaty
with Portugal: the government of the Congo was something in
which they had no say. Belgium had no part in it.
The Congo Independent State was not merely an absolute
monarchy in which the sovereign held all powers; it was a state
that was in a way fused with its sovereign. Sovereignty was
invested in the person of the king, who considered the state his
private property. Leopold II called himself the 'proprietor' of the
Congo. In his will, which was made public in 1890, he bequeathed
to Belgium his 'sovereign rights' over the Congo, just as if he
were bequeathing a house or a piece of real property.51 A jurist
51
Leopold's will, it should be noted, never came into effect, because the Congo was
annexed by Belgium while the king was still alive. He died in 1909, a yeai after the
annexation.
316

KING LEOPOLD'S CONGO: EUROPE
said at the time - and the remark was made in all seriousness - that
to find a precedent for such a situation one had to go back in
Europe to the time of the Merovingian kings. This does not mean
that Leopold was a relic of the past. He was on the contrary quite
modern. His spirit was that of a great captain of business. At the
price of incessant labour, and at his own expense, he had built
up his enterprise; what else could he be than the master of the
enterprise? The master he was, not only theoretically, but also in
practice. He never delegated power, he exercised it personally. All
the great decisions were his. When his advisers did not agree with
him, they had to quit. In the daily work, in his office in Brussels,
he concerned himself even with details. As a capitalist and as a
manager his main concerns were to make his enterprise profitable,
and to expand.
Curiously enough, Leopold tried to expand even before the
enterprise became self-supporting. In the first years of existence
of the Congo State, when only a tiny part of the territory of the
state was occupied, Leopold tried to extend his frontiers in all
directions. In 1888—9, l ^ e points he sought to attain were the
upper Zambezi, Lake Nyasa (Malawi), Lake Victoria and the
upper Nile. Time was pressing, he explained to his most intimate
collaborator, for 'after next year there will be nothing more to
acquire in Africa'. An elaborate policy was devised with these
expansionist ends in view. The push toward the east, for instance,
would depend on the alliance with the ' Arabs'; it was hoped to
obtain their support in order to extend the influence of the state
to Lake Victoria. This vast scheme soon encountered obstacles
from almost every direction. The expeditions projected towards
the upper Zambezi came to nothing. No agreement was reached
with the Arabs; on the contrary, it became necessary to engage
in war with them. In the direction of the upper Nile, however,
the push persisted until the very end of the Congo Independent
State; it was a great chapter in the history of Leopold's imperialism
in Africa. This 'enormous voracity', as Stanley had described it,
can be best explained by the driving impulse which was also
at the root of the creation of the Congo; Leopold was a firm, and
one might almost say a religious, believer in the economic profits
of colonial exploitation. Hence the simple idea: the more the
better. Unlike most other imperialists - with the exception of
Cecil Rhodes, who resembled him - Leopold was never guided
3*7

either in Africa or anywhere else in the world (for his imperialistic

efforts literally spanned the globe) by former entanglements which
became enticements to act. He acted from scratch, with only an
idea to guide him.
In the Congo, however, profit seemed at first very elusive.
During the first ten years of its existence, the young state needed
help for its very survival. From 1885 to 1895 its normal revenue
remained extremely limited; it could cover only a small part of
its expenditure. In 1887—8, receipts still represented only a tenth
of the expenses. The king had to cover the deficit from his privy
purse. But, although he had a huge fortune, he could not bear
the burden. In 1888 he was helped by the yield of a lottery loan,
issued in Belgium for the benefit of the Congo, which brought
in several millions. This proved still insufficient; in 1890 he had
to turn to Belgium itself. When the Belgian parliament had given
him the authorisation, which was required by the constitution, to
become the sovereign of the Congo State, Leopold had privately
promised never to ask for any financial help from Belgium. The
Congo, he announced, would be self-supporting. This was
certainly a sincere promise. It shows how deliberately and even
absurdly sanguine Leopold was about the economic prospects of
his colony. Five years later, Belgium had to save him from
bankruptcy. The Belgian government agreed to grant the Congo
a loan of 25 million francs, spread out over ten years. But given
the expansion of the Congo State's expenditure, this gave the king
only a temporary respite. In 1895, to avoid a total collapse, a
supplementary Belgian loan of nearly 7 million francs was added.
The year 1895 marked the turn of the tide. Leopold was saved
by a combination of his own deliberate policy, the regime domanial,
which he had introduced a few years earlier, and a kind of
miracle — the miracle of wild rubber.
The regime domanial dated from 1891—2. It bore the personal
stamp of the king. The logic of the system was implacable. The
state had been declared proprietor of all vacant land, which would
henceforth constitute its domain. Vacant land, it was decreed,
consisted of all the land which was neither occupied nor being
exploited by the natives. It happened that almost everywhere in
the country the two most remunerative wild products, ivory and
wild rubber, came from lands decreed vacant and so might be
regarded as products of the state's domain, to be collected by the
318

KING L E O P O L D ' S C O N G O : EUROPE
state alone. As a consequence of this system, a trader might not

buy ivory or wild rubber from the Africans without becoming
a receiver of stolen goods - stolen, in effect, from the state.
Private commerce was stopped in that way by what amounted to
a state monopoly. The state authorities protested that there was
no monopoly at all. Everywhere, they said, exchange remained
entirely unimpeded. But as the state had laid hands on virtually
the only marketable products, a contemporary critic was right
when he humorously observed that it had in effect made a law
with two articles: 'Article i, Trade is entirely free; Article 2,
There is nothing to buy or to sell.' Leopold imposed this policy
against the advice of his best and most faithful collaborators, who
were horrified by this violation of the most solemn engagements
of a state which was born under the aegis of free trade. The king
swept away all objections. He was preoccupied only with financial
results.
When Leopold established his regime, his hopes rested mainly
on ivory. The wealth of the country in rubber was not yet
suspected. But the harvests of rubber increased within a very short
period. In 1890 the Congo exported only 100 metric tons of
rubber; in 1896, exports reached 1,300 metric tons: in 1898, 2,000
metric tons; and in 1901, 6,000 metric tons. This last figure
corresponded approximately to a tenth of the world production
of rubber.
The state itself, because of the domanial regime, was the
principal beneficiary of this manna from heaven (not the only one,
for it had been compelled to make a compromise with some
trading companies and to abandon to them the exploitation of
the domanial products in some parts of the Congo). Financially
speaking, the change was a momentous one. In 1890 the state took
from its domain around 150,000 francs; in 1901 the domanial
products - with rubber rating first - brought it more than 18
million francs. From that time on, the financial difficulties of the
Congo were ended. Indeed, thanks to the domanial resources, and
thanks also to the proceeds of loans - for now that its credit was
solid, it could raise loans without difficulty — the Congo soon
began to have budgetary surpluses. In a developing country, these
were badly needed for public investments. Instead the king used
the greater part for the advantage of Belgium. Again, this was the
result of a personal creed: Leopold saw the direct enrichment of
319

the metropolitan country as the cardinal advantage to be gained

from the well-planned exploitation of an overseas possession. As
he had the soul of a builder-king, it was to a policy of public
works, urban improvement and grand monumental constructions
that the Congo funds were systematically applied. The sovereign
of the Congo paid for the construction of the Arcade du
Cinquantenaire in Brussels, for the building of the Tervuren
Museum, for enlargements to the castle of Laeken, for public
works at Ostend. After 1901 the transfer of funds from the Congo
to Belgium even took on an institutional form, with the
establishment of the Fondation de la Couronne (which was first
called Domaine de la Couronne). Its mechanism was simple: the
Foundation was granted domain lands of enormous extent —
around 250,000 sq. m, or more than a tenth of the total area of
the state. The products collected from these lands (mainly, of
course, wild rubber) brought in a very high annual income. The
king, moreover, gave the foundation part of the yield from the
state loans. It was these substantial resources that the foundation
applied to the programme assigned to it, which had as its almost
exclusive object the carrying out of great public works in the
metropolitan country. Urban developments were planned that
would completely change the face of Brussels. In that way, the
king said, the Congo would take its ' just share in the embellish-
ment of our country'. These words 'just share' show that
Leopold was profoundly convinced of the Tightness of his policy.
In his eyes, a nation that endowed a new country with its
civilisation, its labour, its capital, had the right to 'legitimate
compensation'. This is what Belgium had done for the Congo;
it was only fair that the Congo, in compensation, should participate
in the 'embellishment' of Belgium.
All the principles of the Congo policy were decided by the king.
But in the implementation of these policies, the administration of
the state often played its own game. This was particularly true
of the domanial regime. Leopold's will was that the state should
extract the maximum profit from its domain. But he was not very
much concerned with the way this aim was reached. The system
of outright exploitation of the native population which gave, as
the figures of production show, such extraordinary results, was
mainly devised by the administration, both in Brussels and in
Africa. Leopold limited himself to noting that the yield was
320

satisfactory. However when the first accusers rose up to denounce

abuses in the treatment of the Africans, the king was deeply
moved. From 1896 to 1900, as his private letters reveal, he passed
through several periods of agony. 'We are condemned by
civilized opinion', he wrote in September 1896. 'If there are
abuses in the Congo, we must make them stop.' 'It is necessary
to put down the horrible abuses,' he repeated in January 1899.
'These horrors must end or I will retire from the Congo. I will
not allow myself to be spattered with blood and mud.'
On the occasion of each of these crises of anger and disgust,
the king reiterated strict orders: cruelty to the natives should be
severely punished. The Congo administration just waited for the
storm to pass. It had elaborated a system and stuck to it. Altering
the system might weaken it. The lessening of pressure on the
Africans would naturally bring about a reduction of revenue; and
the administration was well aware that, if this occurred, it would
have more than royal anger to face. In other words, the
administration distinguished between the king's permanent and
fundamental desire — to increase the output of the domain - and
his occasional crises of conscience. It modelled its action on what
was permanent and fundamental. All those linked with the regime,
therefore, and desirous of exculpating themselves, tried to con-
vince Leopold II that the accusations against the Congo were
unjust or exaggerated and were made in great measure out of ill
will. The attitude of Leopold, who, unconsciously no doubt, was
ready to be convinced, thus came to undergo profound
modification; instead of being affected by the attacks, he began
soon to react more and more violently against them. Whereas the
king almost always dominated his entourage, it may be said that
in this case he allowed himself to be dominated by it.
Negotiations with the private sector always remained in the
hands of the king personally. He was interested by business as
much as by government. These negotiations took three main
forms. First, the king had to attract Belgian and foreign capitalists
to the Congo for enterprises which called for heavy capital
investment. His task in that respect, especially in the beginning,
was extremely difficult. It would have been made much easier if
the main Belgian financial company, the Societe Generale, had
been willing to invest. But the Societe Generale kept aloof; it had
no confidence in the Congo. Only in 1906, with the foundation
321

of the Union Miniere du Haut-Katanga, did it enter the field. From

then on, it would lose no time in coming to the top. Finding
capital for the Congo Railway Company, which was constituted
in 1889 to build the line from Matadi to Leopoldville (Kinshasa),
was in some ways as much of an epic as the building of the line
itself. The key to the success was that the Belgian state accepted
to invest in the company for a large part of the capital. Secondly,
the king had to meet the opposition of the trading companies,
which protested against the establishment of the domanial regime.
The conflict ended in 1892 with a compromise: some parts of the
Congo, and especially the Kasai, remained open to a real free trade,
i.e. to the trade in the so-called 'domanial products'. But by far
the greater part of the country was closed under a state monopoly.
The companies which were hurt by the domanial regime were
practically all of them Belgian companies. In their conflict with
the Congo State, they called for the support of the Belgian
government. This was one of the cases which led to an exchange
between two administrations both situated in Brussels - the
Congo and the Belgian administrations. Thirdly, the king had to
cope with the problem of Katanga. This was the only region of
the Congo where foreign capitalists, with Robert Williams at their
head, wanted on their own initiative to develop mining activities.
Long and intricate negotiations finally led to a British-Belgian
combination, the Union Miniere du Haut-Katanga, with the
Societe Generale as the main Belgian participant.
The fate of the Congo was for long years in the hands of one
man. But in spite of the fact that the Congo State was a totally
independent state, that fate was actually also in many ways in the
hands of Belgium. It was Belgium which helped the Congo State
to come into being and to survive. It was Belgium which decided
that its life had to end. Belgian help, without which Leopold could
not have succeeded, involved the country's name and reputation,
its men and money. The first factor was the most subtle of all,
but perhaps the most important. All the doors which opened for
King Leopold in international circles opened because he was the
sovereign of a small but respected country. Otherwise he would
have been treated for what he was - an incredible gambler,
launching into the most hazardous ventures. But being the king
of the Belgians, and having the grand manners of a king, he was
taken most seriously. Next, Belgians went to the Congo in all
322

kinds of capacities but many of them were actually lent to the

Congo by Belgium. The Belgian government allowed officers to
serve the Congo State while retaining their pay and rank in the
Belgian army. Between 1879 and 1908, six hundred Belgian
officers were recruited for the Force Publique and for the
administration of the state. In 1894—5, one Belgian regiment had
no fewer than twelve of its officers serving in the Congo. In 1901,
the total number of Belgian officers in the Congo was ninety-five.
Moreover there were 278 non-commissioned officers, who did not
draw their pay in the Belgian army. Thirty per cent of the Belgian
officers - 179 out of 600 - died either in Africa or on the boat
bringing them back to Belgium.
Finally Belgian money was vital for the king. The authorisation
to issue a lottery loan in 1888, the subscription of Belgium to the
capital of the Congo Railway Company in 1889, the two loans of
the Belgian state in 1890 and 1895, these were all shots of adrenalin
which kept the Independent State alive. But after 1895, the
financial help of Belgium would certainly not have continued: the
Congo, if it had asked for more, would have been pronounced
a desperate case. The miracle of wild rubber occurred just at the
right time.
The end of the Congo State, i.e. the annexation of the Congo
to Belgium, was the result of a decision, not of the king but of
Belgium itself. The story of the annexation begins in 1890. The
counterpart of the 1890 loan was that Belgium was given the right
to annex the Congo, if it decided to do so, in 1901. At this date
it would have a free choice. This prospect generally pleased the
parliament and the country; no undertaking was entered into, but
after a decade one would be able to judge whether the Congo was
worth while or not. Public opinion would certainly have been less
happy if it had known in what dire financial straits the Congo was.
But one of the characteristics of a state where the sovereign was
everything was that it did not bother to publish either budgets
or accounts. A few years later, however, the naked truth appeared.
At the beginning of 1895 the Belgian public learnt that the Congo
was nearly bankrupt. The remedy which both the king and the
Belgian government adopted was immediate annexation. A treaty
of annexation was signed in February 1895. To come into effect
it needed the approval of the Parliament. But that approval never
came. A violent campaign was waged against the annexation by

the Socialists, many Liberals and even some Catholics. The

opponents had a wonderful case: 'We were promised that we
would be free to choose in 1901,' they said, 'We are now asked
to take an immediate decision and the only reason for that change
is that the Congo State is doomed.' The government felt that they
would never have a majority for the treaty. They resorted to
another measure: a new loan to the Congo State. The abandonment
of the annexation treaty was approved and even encouraged by
the king, because the news of better and better rubber harvests,
which came in precisely during that period, gave him confidence
in the future.
That confidence, as we know, was fully justified. In 1901, when
the 1890 convention could normally bear its effects, annexation
was proposed only by a handful of members of parliament. That
proposal was strongly opposed by the king. Now that the Congo
had become profitable and its resources allowed him to realise his
dreams, he was not ready to resign his absolute power. Belgian
public opinion did not support a change in the Congo regime:
the fact that the king no longer asked for Belgian help, the
ordinary citizen thought, showed that he had finally succeeded;
there was no reason to deprive the Congo of such a successful
sovereign. The annexation proposal misfired.
The idea of annexation revived only five years later as a result
of the campaign against the Congo abuses. Belgian public opinion
at first had received the British denunciations of the abuses with
great hostility. The main cause of that reaction was undoubtedly
the recollection of the recent Boer war. On the South African
question the Belgians had almost deliriously espoused the cause
of the valiant little Boers. After their attack upon the Boers, to
rob Belgium of its riches, the British were turning their attention
to the Congo; was it not obvious that they were doing so in the
same spirit of greed ? ' The story of the Transvaal is beginning all
over again. These are preparations and excuses for a new
annexation,' claimed the press. The great liberal leader, Paul
Janson, was expressing a widely held opinion when, in a speech
to the Chamber in July 1903, he said: ' I cannot admit that the
Congo State should be particularly suspect. Above all I cannot
associate myself with a campaign whose last words seem to be
"Ote-toi de la que je m'y mette'" E. D. Morel, the head of the
Congo Reform Association, was particularly maligned by the
524

KING LEOPOLD'S C O N G O : EUROPE
Belgian press. He was described as the instrument of a gang of
'Liverpool merchants' with sinister aims. The Belgian public
believed in the ' Liverpool merchants' with the same simple faith
as the opponents of Dreyfus in France who believed they were
fighting a 'syndicate'. The counter-attack against the British
accusations was led by the Congo State itself. Money flowed to
many journalist and newspapers to encourage them in their
denunciations of Morel. Plenty of articles on the themes of British
greed and the ' Liverpool merchants' were furnished to the press
by a specialised office of the Congo State in Brussels. The
counter-attack extended to foreign countries. It signally failed in
Britain, where only a handful of impoverished Irishmen could be
bribed. But in America it met with remarkable success. America
was important because Morel himself had established a branch of
his association there, which received the help of some American
missionary organisations. Leopold's counter-attack, waged by a
Congo lobby, did not disregard religious arguments; a Catholic
country, the defenders of the king said, was being treacherously
attacked by Protestants.
The counter-attack was especially needed after the publication
in 1904 of the devastating report of the British consul, Roger
Casement. Leopold then sent to the Congo a commission of
enquiry of three members (a Belgian, a Swiss and an Italian)
which, he thought, would refute Casement's allegations. This
tends to show that the king did not realise the actual situation.
The publication of the commission's report at the end of 1905 was
a turning point. On the whole, the commissioners, whose moral
stature could not be questioned, confirmed Casement's findings.
The report did not make a very big impression on the public
because, contrary to Casement's report, it was not a tale of
horrors; the horrors were only described in general terms. But
in Belgian governing circles, and among all those who could
understand a report, the document led to a dramatic reversal of
opinion; the Congo system of exploitation appeared as a funda-
mentally vidious one. One doubted whether the Congo State
would be able to reform itself. Reform, it seemed, could come only
from Belgium. This meant annexation.
The year 1906 was a year of verbal struggle between the
growing number of those who opted for annexation, and Leopold,
who resisted the idea. The king however suddenly capitulated in
325

December 1906. This was the direct result of a debacle which had
taken place in the United States. One of Leopold's paid agents
had passed to the enemy for mercenary reasons and was beginning
to publish his correspondence with the king, a most damaging
series of documents indeed. This led to the collapse of the
counter-attack in America and the victory of Morel's friends; it
also seemed probable that the American government would
support the British government in its efforts to initiate inter-
national measures, such as the convening of an international
conference, to suppress the Congo abuses. The international
status of the Congo State was therefore seriously in danger; the
king resigned himself to the Belgian solution.
However, instead of the quick decisions that might have been
expected, nearly two years were to pass before the annexation was
effected. The main stumbling block proved to be the Fondation
de la Couronne. The foundation was the dearest to his heart of
all Leopold's creations. With the foundation, he was sure that his
policy of embellishing his country with the resources of the Congo
would continue even after the annexation. The programme of
great public works which had been laid down for the directors
of the foundation, who formed a quite independent body, was
sufficiently important to last for decades. To this patriotic dream,
Leopold stuck with the utmost tenacity. The foundation, however,
met with strong opposition, not only from the Socialists and a
great number of Liberals, but even from among the Catholic
majority. The use of Congo resources in favour of Belgium was
criticised. But the main objection, which made the foundation
intolerable, was that it would function under Belgian rule as an
independent institution owning a tenth of the territory of the
future colony, as a 'state within the state'. Such a restriction upon
the sovereignty of Belgium could not be admitted. Parliament, in
that respect, was so adamant that, after long and strenuous efforts,
the king had to give in; in March 1908, the Fondation de la
Couronne was suppressed.
Leopold also had to give in as regards the future institutions
of the colony. The law organising these institutions was discussed
before the annexation, and the king tried to influence it in order
to retain as much as possible of his personal power in the colony.
But if a majority in parliament had decided to annex the Congo
because the abuses of Leopold's regime could no longer be
326

KING LEOPOLD S CONGO: OCCUPATION
tolerated, it was also determined not to let the king have any
authority which could hamper the reforms. The final version of
what was to be called the Charte coloniale reduced the king to a
role in the colony which was similar to his role in Belgium: that
of a purely constitutional monarch. So the way was opened to
annexation and reforms.
The annexation took place on 15 November 1908. In June 1913,
the Congo Reform Association was dissolved; Belgian reforms,
abolishing the domanial regime and the exploitation of the
Africans, had proved so effective that Morel, a severe judge, saw
no reason for maintaining a special watch on the Congo.
OCCUPATION AND ADMINISTRATION OF THE STATE
When Stanley landed at Banana in 1879 most societies in the vast

area of central Africa were no longer isolated one from the other.
A network of long-distance trading routes stretched from coast
to coast providing an overall spatial organisation for this part of
the world, and European commodities were penetrating almost
anywhere. One system linked the coast to Stanley (Malebo) Pool,
another reached from there along the main river to the vicinity
of Basoko and branched out along many affluents, while a third
(the Luso-African sphere) tied Luanda and Benguela to points in
the Kasai and Katanga (Shaba). From the Indian Ocean, Swahili
speakers had reached as far as Katanga and the falls near
Stanleyville (Kisangani) on the upper Congo. From the Nile came
the jalldba, or Nubians, based on Khartoum, who raided as far
south as the middle reaches of the Uele. And all these networks
were expanding. Europeans had travelled along all of the major
routes.
Stanley's first task was to link the lower Congo to the Pool,
by building a road to carry a steamer and to launch it on the upper
river, and by establishing posts. The first was founded at Vivi
before 1880, and the En Avant was launched on the Pool on 3
December 1881 in the roadstead of the nascent Leopoldville
(Kinshasa) post. By April 1882 Stanley was able to enter the
trading sphere of the upper Congo, and posts were founded as
far as Nouvelle Anvers (Makanza) by November 1882, and then
at the Stanley Falls near the headquarters of Indian Ocean traders
327


by the end of 1883. Three years later these traders forced the state
to abandon this outpost.
Meanwhile, expeditions of King Leopold's front organisation,
the African International Association, left the Zanzibar coast and
settled at Karema (1879) on the eastern side, and then at Mpala
(1883) on the western side of Lake Tanganyika. This first round
of expansion was completed when Hermann von Wissmann, who
had founded the association's station at Luluabourg (Kananga)
in 1884, arriving there from Luanda and helped by Luso-Africans,
linked the area to the Pool by paddling down the Kasai. The same
year Luebo was established on the Lulua and the Luso-African
sphere was linked to that of the main river.
This early expansion did not cause major fighting, because the
Africans believed the Europeans to be traders. The new posts
were welcomed by the local populations as a means to bypass the
traditional middlemen. And in the first years this proved to be
correct, both along the main river — where the Congo State and
France were vying one with the other for the allegiance of the
people, so that prices were cheap for plentiful imported goods -
and in Kasai, where the state was competing with the overland
trade from Luanda. No prosperity came to the Lower Congo.
Here it was only by dint of showing superior power that a
connection could be made between Vivi, Manianga and the Pool.
This had been done with very few soldiers, all recruited from other
parts of Africa. The force was not sufficient to prevent the loss
of the Falls on the upper Congo river to the East Coast traders,
so that by 1887 Tippu Tip, the leading merchant in this sector,
was created vali (governor) and a resident could return to the
Stanley Falls, though on sufferance.
The army, the force publique, was organised from 1886 to 1888,
when a decree established its basic organisation. It still numbered
less than 2,000 men, of whom 875 were concentrated at Boma,
the new capital, and only 111 Congolese had been recruited for
it. This mercenary force was too expensive, and by 1891 forced
recruitment by districts was begun, while in 1900 the term of
service was raised from five to seven years. No formal provisions
were ever laid down for the use of irregular levies, and yet the
major military campaigns involved the presence of large numbers
of such troops, especially among the Azande in Uele and the
3 3°

KING LEOPOLD'S C O N G O : OCCUPATION
'Batetela' between the Sankuru and the Lomami. The regular
forces grew from 1,487 in 1889 to a peak of 19,028 in 1898, and
had fallen to about 13,000 by 1908. One military camp was to be
established in each district. More important were the base camps,
the number of which varied at different times. The most important
ones were at Eambi Luku near Boma, and Irebu and Lisala, both
in Lingala-speaking territory. In time Lingala became the language
of the army and thus acquired the growing influence it still enjoys
in Zaire. During the whole period there was always a shortage
of troops because, until at least 1906, so many were earmarked
for the proposed occupation of the Nile regions. But the distinct
features of the force publique were mostly acquired during the ' Arab
campaign' led by Baron Dhanis.
The expansion of territory began again after 1889. Two
fortified posts at Basoko and at Lusambo provided the bases, since
the Emin Pasha relief expedition of 1887—9 had not created any
new posts. In 1892 occurred the inevitable showdown with the
Indian Ocean traders, and the war ended in January 1894, when
they had been routed. The fighting had been massive, because
both sides enlisted the aid of thousands of irregulars. An estimate
of 70,000 dead on the Swahili-Arab side is probably exaggerated,
while figures for Congo State losses are not available. At all events,
tens of thousands died during this campaign. Meanwhile, progress
in the north ran into repeated difficulties, so that state forces under
Colonel L. N. Chaltin only reoccupied the Uele preparatory to a
push to the Nile. This expedition began in 1897. Chaltin defeated
the Mahdist forces and took the Lado enclave, while Dhanis, who
was to have joined him, suffered utter defeat at the hands of his
own troops.
After 1899 no similar major military operations took place.
Expansion in the south from the Kasai began with a set of
expeditions to Katanga, but for lack of personnel only a token
force remained in Katanga to show the flag until renewed
expansion could take place here in 1903. State forces penetrated
into Kivu only from about 1900 onwards. Indeed, by 1908
resistance in Kivu had prevented the occupation of large stretches
of the hinterland, while the Portuguese flag of the Luso-Africans
was still fluttering between the Loange and the Kasai. A full
history of the territorial expansion in terms of posts founded,

taxes in labour or produce levied, and populations subdued, has

still to be written.
After 1885 the occupation met with resistance of different
types, just as penetration was helped by the assistance of African
allies. Primary resistance occurred especially in Uele, south Kivu,
Kwango (Yaka), Kasai (Kaniok) and north Katanga (Kasongo
Niembo) districts. In most of these areas pre-colonial states, such
as the Zande chiefdoms, the Yaka kingdom or a portion of the
Luba kingdom, were involved. In others, such as among the
Mamvu, Lugbara or Lendu, chieness societies resisted with great
success. In 1908 some Zande chiefdoms, some Lendu, the Shi
states and Kasongo Niembo were still holding out. A second form
of resistance was put up by traders of the pre-existing networks.
Thus the Luso-Africans did not accept the authority of the new
Congo State, while the Chokwe in Kwango and Kasai found allies
among the Kasai Luba (Lulua) of Kalamba after 1895, and in
Katanga traders coalesced with forces of Kasongo Niembo and
army mutineers. By 1908 the Katanga resistance was overcome,
but the Chokwe still remained in the field.
Meanwhile other local powers collaborated with the Congo
State, in whom they saw an ally. This happened first in Kasai with
the Zappo Zap, and later with Lumpungu (Songye) and Ngongo
Leteta (Tetela), who abandoned their Swahili-Arab allies in 1892.
Even after Ngongo's execution in 1893, other Tetela leaders still
helped the penetration of the Congo State in the Upper Tshuapa.
In Katanga the son of Msiri, king of the Yeke rallied to the state,
even though his father had been killed by one of the expeditions
of 1891. His subjects, the Sanga, had risen both against him and
the state, and had been supported by the Luso-Africans. Similar
patterns of alliances were common among the Zande and
Mangbetu in the Uele district.
Yet another form of resistance became common after 1891. It
was revolt against taxation, in labour or in commodities, imposed
by the state or its concessionary companies. The rebellion of the
porters in Manianga (1893), and those against labour requisition
among the Ababua (1895, 1900—3, 1906), differed little from those
of the Budja, who had resisted penetration in 1891, and then
fought the rubber tax imposed by the Anversoise Company from
1898 to 1901, or those of various Mongo groups in the territories
of the Anglo-Belgian India Rubber Company (ABIR) from 1893
332

KING LEOPOLD'S C O N G O : OCCUPATION
onwards, or against those of other groups further south, on the
royal domain. By 1905 the Budja had submitted, but it was 1908
before the Mongo in the ABIR country surrendered, while - along
the Lokenye especially - primary resistance, and then resistance
to the collection of rubber, continued until well after 1908.
Taxation also was the main cause for insurrection in parts of the
Kasai, notably among the Kuba in 1904. And even where
populations had been subdued, rubber collectors, agents or
traders could occasionally still be murdered. The first case dates
from Tshumbiri in 1889, and as late as 1907 there were still such
incidents in the rubber collecting areas of Kasai.
The last form of outward resistance was by flight. Flight on
a massive scale took place in the equatorial forest from about 1895
to 1905 and in all directions. People hid in the forest, moved on
further when occupation forces came too near, and stopped only
when hemmed in from all sides. Substantial numbers crossed
the borders, especially to Angola and, before 1900, even to the
French Congo. This form of resistance was barely subsiding by
1908.
The most spectacular upheavals came with the mutiny of
portions ofthe forcepublique. A first mutiny at Luluabourg in 1895
was caused by ill treatment. At first the soldiers swept aside all
opposition. Later, after some inconclusive fighting, and lacking
any unified leadership, they drifted to Katanga, where they joined
the Luso-Africans, until their surrender in 1908 after a prolonged
campaign. The second mutiny broke out among the largest force
the army had fielded: the Dhanis column which was marching
towards the Nile. Hardship again was the main reason for the
revolt at Dirfi in early 1897. This time the rebels defeated the
loyalist forces and, their numbers swollen to about 3,000 men,
turned southwards. One of their columns was defeated on the
Lindi in 1898, but their fortunes revived. They occupied the key
position of Kabambare only to be defeated on New Year's Eve
in 1898. Nevertheless, remnants of their forces held the field until
part of them surrendered to the Germans at Usambura in 1900,
a portion surrendered near the Luama in 1901 and some may have
rejoined the Luluabourg mutineers in Katanga. A third, brief
mutiny occurred at the fort of Boma in 1900 and did not lead to
prolonged fighting. The conquest of the Congo had been anything
but pacific in the end.
333

Black and white held conflicting opinions about each other. To

the whites, the Africans were lazy, cruel children. To the blacks
of the Lower Congo and as far as the Kasai, the whites were
persons returned from the dead, from the land across the water
or under the sea, whence their wealth and power came. Every-
where European technical superiority and wealth was attributed
to special, supernatural power; Europeans were sorcerers or
magicians. Theirs was an ill-gotten knowledge and wealth. Their
prestige was ambivalent. Even the ' good' white man derived his
power from some unethical unknown. In practice the Congolese
never accepted European superiority. After resistance was
broken they still did not accept the new order. When, early in
1908, a river steamer, called the Ville de Bruges, overturned near
Lisala, all the whites aboard save one were killed by the fishermen
attracted to the plunder. At the subsequent trial, one of the latter
was reported to have chosen between saving the whites and
gaining 'great wealth' or killing them. He did the latter, because
that was what he saw as his primary objective. The Mangbetu met
the first state column with shouts: ' Turks, Arabs, Whites, men
of Semio (Zande), all liars, thieves and dogs!' After years of
colonial rule a headman in Mayombe, asked why his village now
'fled the catechist', replied that all whites were bad: they had
brought sleeping sickness; they asked for the children to make
them pray; and now there were only graves (1908/1909). Already
in 1883 the traders on the river had asked: 'Our customs may
seem bad to you, but leave us alone. Stay in your country as we
stay in ours,' and, after the last revolt in 1908, the Budja queried:
' The rubber is finished. You have no more to do here. Won't you
go away now?' Belgium was taking over the Congo State in 1908,
and at Boma whites were explaining that a general insurrection
was unlikely, not because the Africans did not desire a rebellion,
but because they were not unified enough to rebel. Even now the
period of the Congo Independent State is remembered in the
Equateur Province as 'the wars', and the whiteman as 'the
destroyer of the country'. The mass of the population finally
submitted, but never forgot.
The implantation of government began with the erection of a

district headquarters. Land was acquired, in the early years by
negotiation, later often by occupation. Then labour had to be
334

KING LEOPOLD S CONGO: OCCUPATION
found to build houses, warehouses and offices and to tend gardens
or fields. In most posts the yield was not sufficient to feed the
personnel of the post. After 1891 the necessary labour and food
was by compulsory work in lieu of taxes. Before that date, services
had to be paid for and obtained through the co-operation of
neighbouring local chiefs. Income up to then consisted mostly of
commercial profits or on occasion the spoils of war. Stanley soon
found out that it was not sufficient to support an administrative
post. The fortunes of such posts fluctuated according to the
personality of the European in charge, and it was not until about
1886 that some stability developed in even the most important of
them, as more personnel became available. In the early years the
prosperity of the administration depended on the relations
between the administrator and the local African authorities,
including the balance of physical power. As more soldiers became
available, and the law on taxation gave more latitude, force was
more and more applied. Still, until the end of the period, newly
founded posts simply could not withstand determined local
hostility. Furthermore, the first year or two in the life of a post
was often vital, since it set the pattern and the reputations by
which the state officials and the local populations were to live for
a long time.
The administrative structure of the Congo State took shape
after its recognition in 1885. The capital was transferred to Boma,
where the central administration resided. It consisted of the
governor-general, who could legislate in urgent matters but
almost always waited for orders from Brussels, and the directors
of the four basic services as well as the commander of the force
publique. The territory was divided into districts, numbering
between twelve and fifteen at different times. Districts in turn were
sub-divided into posts, of which there were 18 3 by 1900. The grass
roots organisation was provided by the 1891 decree on chiefs.
These were appointed by the administration as an area came under
occupation. By 1906 they numbered 440, administering about one
million people. Even at this date the bulk of the population was
still not integrated into the system, and the number of chiefs
remained manageable. Each district had its own budget. This gave
it a fair degree of decentralisation, especially in the Upper Congo.
The district commissioners also had a free hand to settle political
affairs, found posts, subdue populations, appoint chiefs and use
335

the local military detachment for these ends. Their budgetary

autonomy implied, though, that they should generate revenues.
Before 1891 they traded; later they levied taxes in kind and by
corvee.
Chiefs were auxiliary agents of the state. They did not need to
have any traditional legitimacy, and many veterans from the army
or from service in the administration were rewarded with such
a post. In many instances, however, a legitimate chief was sought
and recognised, if only because of the perceived ease of adminis-
tration which resulted. Still, it was not until 1906 that legislation
prescribed the nomination of persons with hereditary rights. This
was still very vague and still allowed larger chiefdoms to be
broken up into smaller chieftaincies, while traditional chiefs could
still, as among the Pende, have straw men appointed, when they
did not want to serve themselves. Any ambitious person still could
make a claim, if he was as much as a clan head, and be given 'the
medal'. In effect, then, local rule in the district was and remained
direct, with rare exceptions.
The Department of Justice, founded in 1886, provided for
courts in the Lower Congo, a single court of appeal at Boma and
a superior council or supreme court in Brussels. The latter
elaborated the Codes of 1886, 1888 and 1895. Upstream of
Leopoldville military law prevailed until May 1897. The com-
missioners, all officers, administered justice. Even later there was
but a slow change in the de facto system, and in 1905 these officers
were still the judges of most 'territorial tribunals', although the
Department of Justice had expanded from nine members in 1904
to fifty-three in 1906, and magistrates began to be encountered
in the Upper Congo. Only two jails were built: one at Boma for
the most serious offenders and one at Stanleyville, although
rudimentary detention centres existed in all districts.
The department was headed by a director of Justice and a
procureurd'etat, subordinate to the governor-general. Its autonomy
vis-a-vis the administration was only relative, and even such
freedom of action as existed was limited to areas where magistrates
were found, that is, until 1905, mostly to the Lower Congo.
During the first years the department did bring the arbitrary
meting out of 'justice' by the Boma settlers or by state agents in
the Lower Congo under control. It never interfered with African
law and local courts save to ban 'barbarous customs', such as the
336

KING LEOPOLD S CONGO: THE ECONOMY
poison ordeal, and later by limiting the severity of the sentences

such courts could pass. During most of the period, however, each
European handled the 'palavers' he created, or which were
brought to him, as he saw fit. In the later years of the period,
however, tensions rose rapidly between the administrators and the
increased number of magistrates who tried to impose the notion
of and the respect for law, while also attempting to force the
administrators to defer serious cases to the regular courts. This
expansion of the role of the magistrature was closely linked to the
attempts to counteract charges about lawlessness and atrocities in
the Congo. By 1908 the magistrature had succeeded in making
its presence felt throughout the State, but was still far from
controlling even the administered areas. Throughout its existence
this Department of Justice had to struggle for recognition and
expansion in the teeth of the indifference and accusations of
incompetence raised by both local administrators and critics of the
State.
THE ECONOMY
At first the growth of the Congo State was accompanied by the

development of traditional commercial companies. By 1891-2 the
government had imposed a regime actually limiting free trade to
the Lower Congo and, after 1902, to the Kasai. Elsewhere trade
was its own monopoly or that of the concessionary companies to
which it had leased certain areas. Meanwhile, the necessity of
establishing an infrastructure of communications, and later the
lure of gold, saw the foundation of other companies as of 1889
and 1891. In short, the state destroyed the previous economic
organisation of the Congo basin, primarily the trading networks
which have been mentioned but also much of the agricultural and
artisanal production. Its policy was founded on the gathering of
rubber and ivory by forced labour. The resulting situation led to
the reforms of 1906 and the annexation by Belgium in 1908, not
long before the substantial revenues the system had produced
were about to vanish with the collapse of the rubber price on the
world market, and the growing scarcity of ivory and rubber alike.
Meanwhile this policy had furnished the means to establish the
state and to create its minimal infrastructure.
The first Belgian house came to join a number of companies
already operating along the estuary of the Congo river by 1885.
337

All these companies vied with one another to buy the profitable
ivory from African traders, and also bought palm oil, palm
kernels, groundnuts and small quantities of other goods such as
timber. By 1883 the aggregate value of the trade approximated
to that of the Niger delta, and was growing as the agents of the
companies followed the penetration into the interior. At the end
of 1886 there were 132 stations. However, it was not until the end
of 1888 that a major Belgian company, the Societe Anonyme Beige
du Haut Congo (SAB) — a subsidiary of the new Compagnie du
Congo pour le Commerce et l'lndustrie (CCCI) - entered into the
arena. Along with a Dutch Company, the AHV, it soon became
dominant in the Upper Congo. Major competition from the
established African trade collapsed after 1887—90 and the com-
mercial activities of state officials hindered them to a lesser degree.
The regulations of 1891 and 1892 were a severe blow to their
expansion. Still they did survive.
The commercial station ('factory') both bought and sold. The
warehouse, where it stocked European imports and ivory or other
commodities for export, was its nerve centre. The company agent,
often working on commission, attempted to make the best profit
both on buying and selling, and his African counterparts tried to
do the same. The African traders wanted to bar agents from
dealing directly with producers, near or far, and to prevent them
from learning which commodities were prized inland, because
these items - such as beads, cowries or brass rods - were used as
a form of currency or standard of values. The basis of valuable
trade was the exchange of ivory for fabrics, guns and gunpowder.
Since the last two products could not be sold or at least not
openly, the African trader even in the Lower Congo did manage
to compete for years with the companies; because he could
provide these items from Angola, ivory was reserved for him
rather than for the companies. As the companies penetrated
further up the river and launched steamers, the advantages of the
African middlemen were eroded, so that on the main river the
whole trade in ivory had fallen into European hands by 1890. The
African traders, forced out, became sailors on the new ships, still
trading on the side in many local commodities destined to be
carried for relatively short distances. The canoe trade did not
completely die out but came to be restricted also to shorter
distances and more local goods. The river people escaped most
338

of the brunt of forced labour because of their mobility, the

closeness of the border, their qualities as seamen and, around
Bolobo, because of mission protection. The traditional trading
system survived much longer upstream of Luebo, where the
companies could not use ships. The competition with the Luso-
African trade in southern Kwango, Kasai and Katanga only began
to favour the Compagnie du Kasai after 1903.
As explained in the previous section, the Congo State claimed
ownership of the vacant lands. A strict interpretation of the
implications was not given before the secret decree of 12
December 1891 was signed. Vacant land was thereafter held to
include all natural products such as ivory or rubber. The
collection of these products was to be done by Africans in lieu
of paying tax. Implementation of this decree was violently
opposed by the commercial companies, so that the Lower Congo
was exempt from the system, as was the Kasai until 1902. To
collect the rubber and ivory, the government leased two large
areas, one in the Mongala basin and one comprising the Lopori—
Maringa basin to two new companies: the Anversoise and ABIR.
While not transferring sovereign rights to the companies, it
gave them the right to collect taxes and keep their own militia.
To the south of this area, in the regions of the Lokenye and Lake
Ntombo Njale (Lake Leopold II), the state exploited the Domaine
de la Couronne (Crown domain) itself. Later, as the territory
under Congo State control expanded, the system was introduced
into new areas. Rubber was even collected in Katanga on the
Lomami in 1902 and later. But the bulk of the exports came from
the areas cited and from the Aruwimi district.
The fiscal rules had been laid down by law in December 1892,
and during the same year commissions began to be paid to state
agents according to their 'production', a system later imitated by
the commercial companies, ABIR and Anversoise. Since the
amount of tax to be paid in labour had not been fixed, the decrees
gave the agents a means to exploit the populations at will. Only
in 1903 was the limit of forty hours a month decreed, but the
appreciation of the time it took to produce a kilogram of rubber
was left to the judgement of the agents, and only nominal payment
was to be issued for such forced labour.
Atrocities resulted. In the Equateur district a high pitch of
violence erupted in 1893, even before concessionnaire firms
339

operated there. Whereas force had been used before, and even the
burning of villages, the new regime attempted to occupy all the
villages by settling African auxiliaries, the ' sentries' or capitas, in
every single one of them. Resistance was forcibly broken, and the
populations reacted again and again by insurrection, guerrilla
warfare, submission, renewed warfare, and flight. There was not
enough time to plant crops, so food shortages became usual and,
while people were hiding in the forests, the absence of all
amenities favoured the spread of disease. Many died, many others
were killed by the sentries. If the atrocities were certainly worst
in the Equateur district, the system led to abuses elsewhere as well.
Thus in Kasai the pressure exerted by the Compagnie du Kasai
brought similar reactions. As late as 1907-8 there were 285 capitas
in its sector X (Luebo and the Kuba) or one for every two villages.
Food was scarce, because not nearly enough time was available
for agriculture.
Isolated missionary protests began as the system was put into
operation, but no reforms were made. Only after Casement's
consular report (1903) became known, and the Congo Reform
Association had been founded, did the government set up its own
commission of enquiry. Its report substantiated the essential
charges, and reforming decrees were issued in 1906, but their
enforcement was slow. Meanwhile, after 1904, both ABIR and the
Anversoise lost their rights to exploitation but continued to draw
income from rubber exports originating in their concessions. By
then, at least in the case of the ABIR, the accessible sites for rubber
collection were almost exhausted.
A railway to the Pool was the indispensable condition for opening

up the Congo river basin, and soon after its foundation the Congo
State attempted to attract capital to found a company. Finally a
subsidiary of the CCCI was formed, and work began in April 1890.
The company received land grants as an incentive. The task
proved extremely arduous, because of the topography near Matadi
and because all labour had to be imported. All the available local
men were engaged in porterage, which itself became so onerous
that it led to an insurrection in 1893—4. Working conditions were
bad and mortality was high. A railway militia soon had to be
created to maintain discipline among workers terrified by the
mortality and discouraged by the extremely slow rate of progress.
340

In four years only 26 miles were built, and at its worst the rate
of mortality and desertion ran up to 17 per cent in 75 days at the
end of 1891 and the beginning of 1892. The death-toll was given
in the end as 1,800 Africans and 132 Europeans, but the labour
force comprised about 2,000 Africans and 170 Europeans at any
given time. As the work progressed beyond the first thirty miles,
the major problems had been overcome and the mortality figures
diminished. Once one half of the line had been completed, more
local labour became available, but the core remained foreign: Kru,
Sierra Leonians and, towards the end, Senegalese rail-laying
crews. By March 1898 the job was done.
This railway allowed the launching of five 100-ton steamers on
the upper river and the transportation thereby of the equipment
needed for further lines of rail. The next company, the Compagnie
des Chemins de Fer du Congo aux Grands Lacs Africains (CFL)
was formed in 1902 and was granted 4 million hectares in
concessions, which eventually were mostly converted to mining
concessions. The CFL was originally intended to build a railroad
to the Nile. With the failure of Leopold's schemes there, it was
instructed to link the Stanley Falls with Katanga and Tanganyika.
A first section from Stanleyville to Ponthierville (Ubundu) was
built at a rather slow rate between 1903 and 1906, and by 1908
work was under way on the next section from Kindu to Kongolo.
Labour was recruited in the country, different districts being
required to provide a set amount of labour and often using
press-gang methods to round up the men. Over the first three
years, and for the first section, the mortality was 3.3 per cent
annually, much lower than the two-digit figures of the Matadi
railroad. The logistics were also much improved. By the time that
the first section was completed, a core crew had been trained.
Conditions improved so much that by 1907 labour problems eased
and forcible recruitment ceased, while the line of rail progressed
faster than before.
A third company which was to build a railroad from Katanga
to Leopoldville was also founded in 1902, then replaced in 1906
by the Societe du Chemin de Fer du Bas Congo au Katanga (BCK),
which did not start with its construction project until after 1908.
When Belgium took over, several human porterage routes still
existed. The main ones ran from Buta to Rejaf on the Congo—Nile
route, from Kasongo to Albertville (Kalemie) and Baraka, and

from Lusambo to Dilolo. Most of the traffic for Katanga already

used the Zambezi route and later the railway through Rhodesia.
Katanga was believed to be rich in gold, and a rush was feared.
So a Compagnie du Katanga, founded in 1891, was to occupy the
area and stop all British attempts to take it over. The company
received one-third of the land in ownership and held a 99-year
lease on the mineral rights in this area; on the other two-thirds
they held preferential mineral exploration rights for twenty years.
The company sent out expeditions, one of which identified rich
copper deposits. Then in 1899 Robert Williams, heading the
Tanganyika Concessions Ltd (TCL), a company founded to
exploit gold, requested to pursue his search into Katanga from
Northern Rhodesia. The Compagnie du Katanga and the Congo
State then formed the Comite Special du Katanga, the CSK, to
administer the whole of Katanga and to hire Williams's company
for prospecting. The CSK had its own police force, and its
representative dealt directly with Brussels, bypassing the adminis-
tration at Boma for many matters. In 1906 the mining company,
Union Miniere du Haut Katanga (UMHK), was created by the
Compagnie du Katanga, the CSK and the TCL. The first
exploratory trenches were opened in 1903, and by 1906 a pilot
plant near Kolwezi processed copper for a while, the main mines
then being for gold and tin. By 1908 the prospecting work had
been completed, the railhead from South Africa was close to the
border and Upper Katanga was on the verge of a profound
transformation.
Meanwhile, gold had been found in 1903 in the north-east, near
Kilo. The government decided to exploit this itself, and began
immediate production from the alluvions. The production was
still small in 1905, averaging 15-20 kg a month, but it climbed,
reaching 311 kg by 1908. During these years only a small labour
force was employed in the exploitation itself, but porterage, the
building of a compound at Kilo and the supply of food for the
labour force was already producing a drain on the local
populations, many of which, such as the Lugbara, Lese and Lendu,
were still unadministered.
In 1906 the Forminiere was created and given the mineral rights
on lease for ninety-nine years over one half of the Congo, the rest
having already been granted to the UMHK, the BCK and the CFL,
excluding only the Kilo—Moto gold-bearing deposits. The
342

Forminiere was to prosper later on, when diamonds were

discovered in its concessions.
The Congo State was anxious to promote agriculture, but its
economic policies in effect prevented this. Indeed, even the
existing exports of groundnuts and palm oil from the Lower
Congo declined in value until about 1900. Still, the state was
anxious to introduce cash crops. It recommended the planting of
trial plots on all stations, and founded the experimental botanical
gardens at Eala in 1900. It sent out several important agronomic
expeditions, while the Jesuit mission at Kisantu began planting
its botanical gardens in 1898. The state was interested in highly
priced products such as rubber, cocoa, coffee, cotton and tobacco.
Only in Mayombe did some private plantations succeed with
cacao, coffee and palm products. And even here most of the
income stemmed from palm oil, which did not come from the
plantations themselves. The tax law of 1891 envisaged that tax
should be paid as required in foodstuffs to different posts. This
was especially the case at Leopoldville and Kisantu in the Lower
Congo, where cassava in the form of chikwangue was imposed.
This, together with the development of a system of mission farms,
resulted in a growth of production by 1908. From the last years
of the Congo State onwards, the area of Madimba easily met the
food requirements of the Pool and the railroad. It was then
beginning to develop as the only real area growing food as a cash
crop for export. The same did not happen in the areas where the
Arabs had introduced thriving plantations of rice, coffee and
cotton. Rice was still available in excess of needs at Isangi
(opposite the later agricultural research station of Yangambi) and
at Kasongo and Stanleyville, but after 1900 the surplus fast
diminished, in part because of recruitment for the CFL railroad
construction, in part because of growing needs for porterage
(Kasongo), and in part because of the requisitioning of rubber.
Here as elsewhere the state's policy contributed to a lowering of
output.
As has been stressed before, the Congo State developed as a
result of its income from ivory at first, and from rubber after 1896.
Rubber then represented 43 per cent of the value of exports, its
share rising quickly to 80 per cent or more after 1900; the total
value of exports also increased dramatically. The gatherers of
rubber, the workers on the railway and even the Pygmies hunting
343

elephants thus paid for the creation of a state on the European

pattern and in addition for the dividends and other bonuses which
found their way to Europe. At the same time the sources for
African prosperity from trade, artisanal production and even
agriculture had been destroyed or were in disarray. More and
more the populations were forced into an agriculture of sub-
sistence in the literal sense of the term.
Up till 1908 the Congo State in effect still refused to introduce
money, even at Leopoldville. This would have had two un-
favourable results: it would have allowed Africans to pay their
taxes in cash and thus gradually have eliminated the forced labour
on which so much rested. It would also have allowed them to
break the direct bond between selling their produce and buying
imports, especially where they had access to competitive trade, as
in the Lower Congo or along the main river. The introduction
of money would have allowed a partial revival of regional and
long-distance trade. All of this did happen when money was
introduced. But apart from a small area around the Equateur
station in 1893, and for a short while around Boma, money was
still in 1908 regularly used only in Katanga, where the president
of the CSK had had to beg for its introduction, in part because
the TCL paid their labour in currency and in part because the
miners, like the sailors in Leopoldville, accumulated too many
goods which they could not exchange (or only at ruinous rates)
or carry home (because of their bulk) when their contracts ran
out. Since the monetary policy not only favoured the state but
the traders as well, all put up resistance to its introduction, except
in northern Katanga. An effect was the retardation in the uni-
fication of the market in the country and consequently the lack
of genuine economic development rather than growth.
On the whole, the Congo State had built up only the most
elementary infrastructure and plundered the most easily accessible
natural resources. In the final analysis the economic dynamics of
such a system were not very different from those of the Luso-
African or Zanzibar trade. Slave raiding had merely been replaced
by forced labour.
344

KING LEOPOLD S CONGO: CHRISTIAN MISSIONS
CHRISTIAN MISSIONS AND AFRICAN RELIGIOUS

MOVEMENTS
European penetration was justified by 'its civilising mission*. In

practice this ideology amounted to the justification of Christian
missions by all other Europeans. At first the activity of all
denominations was encouraged by the state, and their advance was
quickened by apostolic zeal, especially from the millenarian
groups, and the rivalry between Catholics and Protestants in
general. But after about 1891 Protestant advances were dis-
couraged by officials, while from 1906 onwards the Catholics
obtained a highly favourable status. The Christian mission
teaching constituted a frontal assault on the convictions, rituals
and ethics of the Africans. Supposedly, agents of the Congo State
were concerned only with the eradication of' barbarous customs'
such as human sacrifice, poison ordeals or the slave trade. The
missions were to transform these cultures. Islam, which had
enjoyed a modest spread in the east, was to be thwarted after
the Swahili-Arab wars by both state and missions. But, just
as the occupation of the country met with spirited opposition, so
the spread of the Gospel also met determined resistance from the
traditional African religions.
The earliest missions were Protestant. In 1878 the Baptist
Missionary Society (BMS) was at Underhill near Matadi, and
moved to Leopoldville when the station was founded. It then
founded stations at Lukolela, Upoto, Monsembe and finally, in
1896, at Yakusu. The Livingstone Inland Mission was in the
Lower Congo until 1884, when they handed their stations over
to the American Baptists and Swedish missionaries. By 1884 they
also had a station at Coquilhatville (Mbandaka). In later years
some felt that the pace of advance of the BMS was too slow, and
founded the Congo Balolo Mission, which went to the Lulonga
in 1887. Along the Kasai, a precursor arrived at Luluaburg in
1885, but died in 1888. Here durable activitity started with the
arrival of the American Presbyterian Christian Mission (APCM)
and the foundation of their station at Luebo in 1891. Meanwhile
the Plymouth Brethren (Garenganze Evangelical Mission) had
founded a post at Msiri's capital in 1886, but spread out only after
1893.
On the Catholic side, the earliest arrivals were the French Holy
345

Ghost Fathers at Banana and Boma in 1880, and the White Fathers
on the shores of Lake Tanganyika, where they took over Mpala
in 1885. King Leopold was worried by the presence of French
mission orders and, after long negotiations, persuaded the Scheut
Fathers to take over the main responsibility. The Vatican then
created the Apostolic Vicariate of the Congo in 1888 covering the
whole country, save for the eastern borders, which remained in
the hands of the White Fathers, who now recruited Belgians for
their posts. Their most significant later foundation was Old
Kasongo (Tongoni), from where, after 1903, they specifically
went to fight Islam. The Scheut Fathers founded a major mission
station at Nouvelle Anvers in 1889, one at Luluabourg in Kasai
in 1891 and one at Kangu in Mayumbe in 1898. From these nuclei
they expanded throughout these areas. The Jesuits took the area
between the Inkisi and the Kwilu, making their headquarters at
Kisantu (1893). To mention only the main other orders, the
Fathers of the Sacred Heart were established at Stanleyville in 1897
and the Premonstratensians along the Uele by 1899.
Unlike the Catholics, the Protestants had no overall plan of
occupation. In practice the denominations did not compete with
one another, so that the effect in the end was quite similar to the
Catholic expansion. By 1908 there was still no mission penetration
in Kivu. Elsewhere, both Catholics and Protestants were present,
save in Katanga, which had only Protestant stations, and in the
east beyond the Falls, which had only Catholic stations. State
intervention had affected the situation from the mid-1890s, when
it seriously curtailed Protestant expansion because of the criticisms
the missionaries voiced against official policies. This forced the
state in the end to rely on Catholics for its support, a situation
which played a role in the conclusion of a Concordat in May 1906,
by which Catholic missions were to receive favoured treatment
in the granting of land and the building of schools.
By 1908 the results on paper showed the Catholics to hold 52
stations and about 670 posts, occupied by 268 priests or brothers
and 125 nuns. They had roughly 125,000 converts. The Protes-
tants, with 40 mission stations and a personnel of 211, had perhaps
70,000 converts. The number of Christians was still small, ranging
from 2 to 3 per cent of the population, but the impact of their
religious, ethical and educational views affected the whole
population of the country with the exception of Kivu, parts of
346

Katanga and of Orientale Province. Even where no missionaries

were found, news of their faith had spread. Thus at Beni the first
missionaries found a small congregation led by a Ugandan
catechist. The portrait of the poet Francois Coppee had been
mistaken for that of the Pope.
The doctrinal differences between Protestants and Catholics,
which mattered so much in European eyes, were not perceived
so strongly by the local populations, who were more impressed
by the personal antagonism by which it was so often accompanied.
Nor did differences in training, background and mode of mis-
sionary expansion make much of an impact. Whether persons in
outposts were called evangelists or catechists did not affect the rate
of expansion nor even its manner. Only the chapel farms of the
Jesuits were really different here.
Often the first to rally around a new station were children and
liberated slaves. This was especially true for the Catholic stations,
which attempted, unlike Free Churchmen, to build up a new
community first and convert within it, rather than to gain a
personal acceptance of the Christian faith by individualsfirstand
form communities only later, although in several cases, such as
at Luanza (Plymouth Brethren), at Bolobo (BMS) or at Luebo
(APCM), people flocked of their own accord to settle around the
mission. Still, Catholic missions started often with a larger
community. Luluabourg had 300 freed slaves for a start, and the
Premonstratensians started with 1,600 'abandoned' children at
Ibembo and Amadi. After the initial years, these differences in size
lessened, as many people sought protection at the mission stations
from government soldiers or from raiders. And the tactics used
by both confessions to make their first converts were rather
similar. They aimed to raise 'liberated' children in the Christian
faith, and to convert local chiefs, hoping that the rest of the
population would then follow.
The first converts to be baptised were often personal servants
and helpers of the missionaries. Conversions at first were very
slow. Yakusu, founded in 1896, had its first three baptisms only
in June 1902. But then, late in 1905, a flood of applicants arrived.
This may bear some relation to the recruitment drive for the CFL
railroad construction, which had begun then. In this case,
however, a great demand for schools and teachers was observed,
and this may also explain the rush to the Christian faith. Still, in
347

many other cases a very slow start was followed by a temporary

boom for conversions. It seems as if the local people looked upon
Christianity as another traditional religious movement. Received
sceptically at first, the movement then snowballed in the way any
local religious movement would do. This certainly seems to have
been the case at Mbanza Manteka in the Lower Congo, where the
Kongo came to see their missionary as a ngunâ, a prophet. The
other major pattern was a steady growth among those who had
sought protection at a station, without any religious revival
aspect. In these cases conversions implied choosing sides.
Converts, by accepting baptism, identified with the European
order. This could be hard on them, as they might be ostracised,
denied marriage partners or be accused of corrupting the
traditional order and ethics. Often they left their original villages
and tried to settle near the mission. Some Christians, especially
teachers, were well received even in their own communities. They
could help the village in its dealings with Europeans. But the first
converts almost invariably became evangelists or catechists
dependent on and protected by the mission. Their help was
indispensable to the success of the mission. They had to teach the
language to the foreign missionary, interpret the Gospel in that
language and thereby transform it into a message understandable
to members of their own culture. On the other hand they
attempted to explain their culture to the foreign missionary.
Invariably they also started the outstations. Great as their impact
on the spread of Christianity has been, it remains little studied.
Three general patterns of expansion could be followed.
Christians could start an official outpost and remain in a group
in a Christian village; individuals could return to their home
villages and live there while remaining converts and acting as
catechists or evangelists; and young Christians could be grouped
in settlements nearby other villages, but subtracted from the
authority of the chiefs or elders. This latter device was systematised
by the Jesuits in 1895, who called the system chapel farms' (Jermes
chapelles) and recruited so-called 'orphans' for this purpose. By
1902 some 250 chapel farms had been established, and the number
was to grow to over 400. Protests resulted. The state-appointed
chiefs complained that the settlements did not obey them, paid
no taxes in kind or in labour and ' annoyed' practitioners of the
traditional religion. The chapel farms began to look as if they were
348

states within the state. Moreover, too many children were taken,
sometimes forcibly, from their homes to the missions. The
complaints found an echo in the report of the Commission of
Enquiry of 1904, which led to a hot debate about the propriety
of chapel farms, which was to last until 1914. Other Catholic
missions, such as the White Fathers or those of Scheut, also built
Christian villages but not as close to other settlements and they
relied much less on converting only children. The main effect of
the chapel farms was the spreading of a new style of practical life
on a large scale. The emergence of a thriving cassava plantation
area in the Madimba region may be linked to this, but the
connection has been insufficiently explored.
Too little is known about the spread of Islam during the years
after the wars with the Swahili-Arabs to allow any generalisation.
All that is known is that sizeable numbers of Africans had become
Muslims by 1894 in the Kasongo-Nyangwe area and around
Stanleyville. The further fortunes of Islam and especially the
acceptance of the Qadiriyya in these communities remains to be
studied.
Religious movements existed in central Africa before the colonial
period. Some were attempts to reorganise the central rituals of
a religion, attempts triggered by revelations made in dreams to
prophets. A movement began when a prophet was followed.
Often a new movement required the destruction of personal or
even collective charms which protected against misfortune, and
movements were collective: a whole settlement had to accept it
to be effective. In the Lower Congo kiyoka (the burning), of 1872,
was an early case. The lubuku movement of Mwamba Mputu in
Kasai started some time before 1865 and continued to gather
adherents until 1895. The mani association of the Zande may have
started before the turn of the century and bufwa, among the Shila
and Luba, antedated the colonial period as well. Still other cases
can be cited.
The turmoil accompanying the colonial occupation may have
favoured the flowering of such movements, although, so far, we
lack records about them, perhaps because to contemporary
observers many of them may have appeared as 'traditional
religion \ This was not the case with the Lugbznjakan water cult,
the importance of which in this period only appears from the
ethnographic record of the 1950s, even though Rembe brought
349

it to the area about 1890. The cult, while a general religious

movement, was closely related also to Lugbara resistance against
penetration. It already had the famous feature that ' bullets could
be turned into water'. But it was not just a collective war charm,
it implied a general reordering of Lugbara society.
A similar movement along the lower Lomami in 1892, and
one directed against the Arabs may also have been more than just
a war charm. The tonga tonga movement, appearing on the Sankuru
and in Kuba country by 1904, outlasted the rebellion, even though
its core feature was a charm to turn bullets into water, and it was
still active in 1908. It had been a general religious movement of
which the war magic was just a part. Other movements remain
to be retrieved from archives and the collective memory of the
populations involved.
A further novel manifestation of traditional religions is recorded
in missionary accounts as ill-will or plotting by ' sorcerers'. Often
traditional rituals were performed with greater ostentation than
before, acquiring symbolic value as a form of spiritual resistance.
The traditional Lilwa initiation ceremonies among the Lokele in
1900 and 1902 was an at least temporarily successful attempt to
counter the influence of the Baptist mission over the young.
Renewed activities of butwa in eastern Katanga, and initiation
rituals in parts of the Lower Congo, may have been due to similar
reactions. In areas where the population was decimated, dispersed
or under tight control as a result of the rubber system, religious
movements could not develop, nor could the traditional rituals
be held. Only where a certain measure of initiative remained with
the population could traditional religion be used as a form of
protest. And in areas where no missionary activity had penetrated,
such religions were not yet under attack. It seems probable that
further evidence will document a strong reaction of African
traditional religions against the missionary attack as well as the
early existence of religious movements inspired by prophets.
SOCIAL LIFE IN THE CONGO STATE
The society emerging from the Congo State was dominated by

a new set of vertical cleavages. The top group was formed by the
European caste. Then followed the elites, which comprised the
following subsets: non-Congolese African agents, Congolese

KING LEOPOLD'S C O N G O : SOCIAL L I F E
agents or allies of the Europeans, the legitimate chiefs and
patriarchs, and the former commercial elite. All of these groups
intermarried, but they were far from being fused into a single class.
Most of the population ranked below them. The bottom group
comprised the slaves, for slavery had not yet been abolished; it
also encompassed many of the liberes until they passed into the
elite of European auxiliaries. The new society had in effect
encapsulated all the more local social groups, promoted interaction
between them and began to give rise to horizontal regional sets
based on the novel phenomenon of ethnicity. In this whole
system, the Europeans formed the reference group for part of the
elite, and that part of the elite in turn was the reference group
for large numbers of the population. The European labels for
ethnic groups were adopted without much question, and this
identification of horizontal groups was crucial in creating the
'tribes'.
The whites in the Congo State belonged to many nationalities,
especially Belgian, British, Scandinavian, Portuguese, Italian,
Dutch and American. Many other nationalities were represented
by small numbers. At the onset the Belgians were a minority, but
by 1908 they had become a bare majority. All Catholic mission
personnel, a few Englishmen excepted, were Belgian, while all
Protestant mission personnel, with a single exception, were
non-Belgian. More and more Belgians were employed in govern-
ment service, partly because recruitment at a relatively cheap cost
from the ranks of the Belgian army was possible. By 1908 there
were 1,700 Belgians out of 3,000 whites. The number of
government agents reached 1,500 by 1906 and constituted slightly
under 50 per cent of the population. Missionaries accounted for
some 20 per cent and commercial agents for the remainder. Most
state agents in territorial service were still military men. Few
Europeans, with the exception of the magistrates, physicians,
engineers, Catholic priests and some Protestant ministers, had
university training. And although on the whole the educational
level was rather low, few whites stemmed from blue-collar
families and even fewer were employed in manual labour. Most
among them belonged to the middle classes, and were still in their
twenties when they arrived. Turnover ran apparently at a very
high rate. The shifting of careers occurred frequently, with state
agents being later employed by companies more than the reverse.

But in very many cases, especially at the beginning of the period,

Europeans did not often stay for the full length of their terms of
contract. For most years mortality rates were over 6 per cent (18
per cent for a typical term) and repatriation for reasons of health
was probably higher. An estimate of a 40 per cent loss over the
typical three-year contract period for medical reasons is not
exaggerated. And this does not include those who broke their
contracts for other reasons. With the partial exception of the
mission stations, the turnover of personnel, especially in the
interior, was high enough to threaten the continuity of local
political and economic policies.
Why did whites go to the Congo? It was widely believed that
middle-class drop-outs formed the bulk of candidates, except for
the missionaries. Debts, women, rebelliousness are most often
cited, in ditties of the period. There were other reasons. The
dullness of Belgian garrisons bred a spirit of adventure. This was
often mixed with a desire for glory, advancement and honours.
Except for company agents, the desire for financial gain was not
marked, as salaries were low. And then there were the missionary
vocations, another form of the spirit of adventure, often born in
spiritual garrisons or, especially among Free Churchmen, revealed
as a sense of personal destiny. Conrad's The heart of darkness paints
a portrait that only partially corresponds to the change in
behaviour and personality structure that affected lonely agents,
and in the interior almost all of them were lonely. His picture is
overdrawn. Still, both callousness and a deep sense of mission,
developed to levels not reached in Europe, and recurrent illness
left its mark on behaviour. A study of the psychological effects
of loneliness and disease among the colonisers is as urgently
needed as is one dealing with their detailed socio-cultural back-
grounds.
The earliest Africans brought to the Congo from outside were
Stanley's Zanzibari. In 1883 these were joined by Hausa soldiers
from Nigeria. They already found Krumen (Liberians) on the
coast, and later numbers of West Africans from Sierra Leone,
Ghana, the Lagos area and Senegal joined them. Recruitment for
the Matadi railway brought some Jamaicans, some Chinese,
labourers from Senegal and Mali; and the army recruited small
numbers of Xhosa, Ethiopians, Somali, Egyptians and Daho-
mians. Most of these left the country, once their terms were
352

KING LEOPOLD'S CONGO: SOCIAL LIFE
completed, but some, especially Senegalese, Liberians, Sierra

Leonians and Hausa remained as skilled blue-collar labour or as
traders. A few Zanzibari settled in the 'arabized' communities in
the eastern half of the country. Some of the immigrants married
in the country and settled; the best known case is the Badjoko
family, founded by the son of 'the chief of the Bangala', a
renowned sergeant in the army, who married a Senegalese girl.
In 19;) we still found a Brazilian Indian mestizo, married to a
local African and running a small plantation started before the
turn of the century. In the long run the immigrants did not play
a major role in Congo. In the short run some among them rose
to responsible posts, such as acting officer in the army during
campaigns or the equivalent of the later territorial administrators.
In the end, the opportunities open to this category of persons were
not greater than those open to local collaborators with similar
skills.
The new local elites achieved their positions as auxiliaries of
the Europeans, and even more as literate auxiliaries. Most had
acquired their skills in the country, but a few reached Europe and
returned with a solid education. By 1908 one of them, Lusambo,
was in charge of the railway station at Leopoldville, the hub for
all traffic to the upper Congo. Others had sadder experiences, as
resident European sentiment ran higher, as the years progressed,
against the 'semi-civilised' and 'the mission-boys'. Of those who
had been trained in Europe, most returned to Africa and some
landed in jail. Local education was left to the missions. In the first
years the population did not appreciate it, but later, when people
began to realise that literacy was a passport to advancement,
especially in remote districts, missions were urged to build
schools or send teachers. Thus in far away Yakusu the Lokele had
challenged the Baptist mission as late as 1902, but in 1903-4 a vast
movement swept that portion of the Congo river, begging for
teachers and crying out for alphabet cards.
The two existing state schools at Nouvelle Anvers and Boma
were run by Scheutists. Children 'adopted' by the state were sent
there and trained in military fashion for jobs in the army or in
the administration. Other schools just taught the three Rs, that
is to say, a programme equivalent to the first two elementary
grades in Europe. Persevering pupils did learn to read and write.
Besides these schools some training was given in the base army
353

camps and 'on the job' in the railroad construction camps. The
villages settled by veterans often benefited from such training and
enjoyed a higher standard of living than the 'traditional'
settlements nearby, and were envied by them. Apart from
anything else, they were exempt from taxes. More advanced
schooling was created in 1906, when vocational schools were set
up at Boma, Leopoldville and Stanleyville, and then in 1908, when
the school for nurses at Boma opened its doors. So the statistics
of 'students' do not tell much. It was only in 1907 that each
Catholic mission order was asked to establish a training school
for teachers, as well as for office workers needed by the state and
the companies; this was a result of the 1906 Concordat.
By 1908 there were probably upwards of 30,000 literate
Congolese, but very few with more than just bare literacy. They
worked for the Europeans at low-skilled operations. The most
responsible blue-collar occupation was that of river pilot. Skills
such as riveter or mason were typical. The usual white-collar jobs
were those of clerks, foremen, warehousemen or mailmen. But
the companies hired capitas able to keep accounts, and the schools
needed teachers. These people mixed with the remnants of the
older commercial elite, which in part had become European
auxiliaries. But they remained aloof and often antagonistic to the
legitimate chiefs. The 'modern' elite was spread everywhere, but
took its cues from life in the towns: Boma, Matadi, Leopoldville.
The towns remained small. By early 1914 Leopoldville numbered
only 12,000-13,000 inhabitants. Between 1895 and 1904 various
centres were proclaimed by decree to be urban areas, and in 1892
a decree providing for the registration of educated Africans was
enacted. The Congo State thus recognised the existence and some
of the aspirations of this elite of auxiliaries, more than the later
Belgian Congo would. All in all, the members of this group
enjoyed little financial independence, because trade was limited to
such a small area, and all other opportunities were linked to roles
subordinate to Europeans.
The African elite groups suffered from epidemics, but probably
not nearly as much as the general population, which also had to
cope with food shortage. No hard demographic data are extant,
and the population estimates, which range from 27 million to 12
million, in early years are completely unreliable. At first smallpox
was the great killer. It had been present in the Lower Congo for
354

KING LEOPOLD'S C O N G O : SOCIAL L I F E
two centuries at least, and was reported in Katanga before
colonisation. Elsewhere it was more recent. A steamer unwittingly
carried the disease to Kasai in 1893, where it spread like wildfire.
But, so far, there are no reliable data about the frequencies of such
epidemics, their morbidity or their mortality. Nor is much known
about the epidemic spread of other diseases such as jiggers and
some venereal disease. The latter affected reproduction rates and
affected the demography perhaps as much as, or even more than,
the smallpox losses. The whole subject needs research. Indeed we
do not even have a comprehensive list of all the diseases involved
in these epidemics. The government created a vaccine centre for
smallpox at Boma in 1893, and by 1906 no single epidemic of the
affliction was reported in the country. Sleeping sickness may have
been endemic in Lower Congo and certainly was so at the
Ubangi—Congo confluent. The first reported case came from
Leopoldville in 1885, shortly after a physician had settled there.
But the epidemic dates only from 1899. From that year onwards
it spread to Stanleyville and then across to Lake Tanganyika and
beyond. The advance elsewhere was slower. Still by 1908 almost
the whole country was affected. Only parts of northern Katanga
and of the northern Orientale Province remained immune. All
descriptions stress a terrifying mortality, and from data of the
1920s it can be estimated at some 80 per cent of the population
in the worst stricken areas. This was the 'black death' of the
Congo. It must have affected the life-style of every single
community in the stricken areas. In 1906 the government
registered serious alarm. Sanitary brigades, and a research
laboratory at Leopoldville, were created to stop the disease, but
by 1908 no lasting successes had been achieved.
In general the health services were poor. From two physicians
in 1888, the number rose only to thirty by 1908. Apart from the
Lower Congo, hospitals were inadequate. The Commissions of
Hygiene set up to fight malaria, in 1892 (Lower Congo) and 1899
(all districts), were ineffective because of incompetence and
quarrels over rank. Malaria had always been there and did not
upset the population as much as the epidemics. The second major
factor in unsettling the rural populations was food shortage, even
famine in rubber exploiting areas. This was an effect of the tax
system, which lasted beyond the end of the period. By 1908 a few
peoples still remained unaffected by the expansion of the new
355

state; most had submitted and found their way of life profoundly
altered. In most places the population was forced back into a bare
subsistence agriculture, almost no time being left for the pursuit
of customary crafts, so that certain skills (such as metal mining
and smelting, the preparation of lye for soap, etc.) were lost. Only
in the most remote areas was regional trade still flourishing.
Genuine development was arrested, indeed reversed. Often the
taxes in kind did not correspond to the pre-existing regional
specialisations, but consisted of foodstuffs, and thus along the
Congo river, for instance, the whole regional division of labour
was largely disrupted, except for the production of pottery and
fish.
The country as a whole formed a new economic unit in which
populations were assigned production items and goals, while the
products of their own specialisations largely came to be replaced
by imports, as often as not of inferior quality. This was true for
mass products, such as hoes, most textiles, and in part salt, as well
as for luxury goods. And given the narrow base on which the
export economy of the country rested (in 1908, 90.9 per cent of
exports by value were still ivory, rubber and copal), the intricate
spatial systems of complementary production were lost.
Major change also occurred in social organisation. Mission
action stressed the nuclear family, at the expense not only of the
polygynous family but also of the larger kinship structures which
lost many of their functions, even though they did survive.
Existing patterns of solidarity were seriously hurt by the stress
on individuality, whether it was in saving one's soul or making
a living. While this may have benefited a few in the emerging elite,
it hurt the mass of the population. The foundation on which
solidarity rested was eroded, as collaboration in production and
exchange, judicial rights, property rights over land and defence
were all more or less curtailed, once sovereignty was lost. Village
leadership was still flexible and still expressed itself in palavers,
but medalled chiefs, whether imposed or legitimate represent-
atives, became agents of the government and were no longer
bound by customary checks. Meanwhile, the need first to resist
and later to organise against oppression led some acephalous
groups to co-operate in larger numbers than ever before. In
Equateur province different groups rallied together in the area of
Boende to prevent the penetration of the state, and the Budja
356

KING LEOPOLD S CONGO: CONCLUSION
fielded up to 5 ,ooo spearmen during their insurrection, although

their 'customary' structure may never have provided for col-
laboration on this scale. Such coalitions did not survive the
conquest, but they helped, as a subsidiary factor, to establish
ethnicity.
The awareness of encapsulation within the state, and the labels
attached by the Europeans to different regions, helped to formulate
an inchoate sense of tribalism. Once the European labels were
accepted, ethnicity was under way. Examples of' new' tribes were
the Lulua, the Suku, the Bangala. Rare were the labels which did
not give rise to a permanent ethnic group, for example, the
Likwangula, although even this term came to designate a quasi-
tribe — the villages of veterans. Tribalism was to be greatly
fostered by the restrictions on free travel and the linking of every
individual to a 'customary chief, which was decreed after 1910
and rigidly enforced in the 1920s. This was to lead to the
emergence of strong ethnic groups.
Ethics and patterns of cognition were much less affected by the
European impact, especially by the missionary impact, than is
often thought. The fundamental constellations of world-views
remained unaffected or altered only very slowly. Converts reinter-
preted Christianity into the old categories, and traditional religious
movements retained almost all of them. But Europeans and their
lore added to the available range of symbolic representations and
to the phenomena to be explained. Changes in the arts, even the
visual arts, have not yet been sufficiently explored. Clearly the
demand was altered, new products (dyes, for example) became
available, new scenes were depicted (porterage, missionaries, etc.)
but most artists lost their patrons. Hitherto the continuity of
socio-cultural patterns has been over-emphasised in the social,
political economic and artistic spheres, while change was over-
estimated in the religious sphere, but such a generalisation cannot
be substantiated without intensive studies, which are still lacking.
CONCLUSION
The Congo State was unique, in that it constituted a colony

without a metropolis, or a state whose capital lay outside it.
Contemporary observers rightly sensed that the years from 1890
to 1892 constituted a fundamental turning point, because the
357

system of exploitation which was to keep the state solvent, was

then developed. Because of its inherent abuses and economic
limitations, the system began to break down by 1906, but it would
not be until 1919—21 that the transitional regime gave way to the
main structures of the Belgian Congo. The years 1890-92
constitute a more significant turning point than the date of the
foundation of the Congo State, or Stanley's arrival in 1879, as the
earlier years still can be seen as an extension of the nineteenth-
century dynamics of the European commercial penetration. The
structural realisation of European hegemony on the ground was
only beginning to be effected between 1890 and 1892. The legacy
was to be the constitution of a contemporary state: Zaire.
358

CHAPTER 7
SOUTHERN AFRICA, 1867-1886
In 1867, a solitary diamond was picked up by chance near the

appropriately named settlement of Hopetown on the Orange river
frontier of the northern Cape Colony. It was sent to the nearest
magistracy, Colesberg, and from there by post to Grahamstown,
where it was identified. When it arrived in Cape Town, Richard
Southey, colonial secretary to the Cape government, declared in
words both celebrated and prophetic: 'Gentlemen, this is the rock
on which the future success of South Africa will be built.' New
finds were reported daily from alluvial diggings along the Orange
and Vaal rivers and their tributaries, and more importantly by
1870 diamonds were also being found in the open veld around
the area to become known as Kimberley. Within five years it had
become the world's largest producer of diamonds, outstripping
even Brazil. A new era in the history of southern Africa had
begun.
In 1870, the political economy of Southern Africa was
characterised by tremendous regional diversity. African king-
doms, Afrikaner republics and British colonies co-existed in a
rough equilibrium of power, but pursuing widely differing social
and economic goals. Although most Africans lived in largely
self-sufficient agrarian societies, few were untouched by the
coming of the merchant and the missionary. South of the
Limpopo, much of the region was dominated by the operations
of commercial capital derived from the mercantile enclaves of the
coastal Cape Colony and Natal. Trading insinuated itself into the
largely pre-capitalist agricultural economies of African peoples
and into the proto-capitalist agricultural economies of the
Afrikaners on the highveld, while the demands from Cape
merchants for cattle and the firearms they brought in exchange
profoundly affected the pastoral societies of south-western Africa,
transforming the nature of warfare in the region. Nevertheless,
unlike the Swahili traders north of the Zambezi, who began in
359

SOUTHERN AFRICA, 1867—1886
this period actively to conquer African polities and transform their
economies, in South and south-western Africa, merchants were
not concerned actually to organise production.
In the Cape, the southern Free State and Natal, there were large
pockets of commercially oriented agricultural production — wine,
wheat and wool in the west, sugar plantations in the east. Yet even
here, where a growing number of Africans were becoming
incorporated into the settler-dominated worlds of the Afrikaner
republics or the British colonies, the majority of Africans still
retained access to land and control over their own labour power.
The switch to industrial production which came with the mineral
discoveries, first diamonds in Griqualand West, and later gold on
the Witwatersrand, led to swift and far-reaching transformations
in production and in social and economic relations in the
sub-continent. For the first decade, Kimberley was a haphazard
and chaotic jumble of miners, merchants and migrants drawn to
the arid, sparsely populated diamond fields in the hope of making
a quick fortune; with the amalgamation of claims, the formation
of large companies and the beginnings of a technologically
sophisticated form of underground mining, the drive for inter-
national capital investment and a cheap, controlled labour force
led to major changes both in the labour process and in the
demands being made of the heterogenous societies of southern
Africa. The changes went far beyond the earlier impact of South
African societies of merchant capital, settler agriculture or the
existing urban centres, which were themselves drawn into and
becoming dependent upon the economic hub first of Kimberley,
later of the Witwatersrand. Within a dozen years of the estab-
lishment of the dry diggings, a new class of industrial capitalists
had been created and most of the remaining powerful independent
African kingdoms south of the Limpopo had been forcibly
conquered, largely by British arms, as a precondition for the
region's further capitalist development.
In the early days of the mining industry, capital had not only
to transform the weak colonial and republican states; even after
conquest, it had also to come to terms with African societies and
their ruling classes. Labour migration and commodity production
for the colonial markets posed a threat to the gerontocratic order
in pre-colonial African societies. Both carried the implication that
single households could accumulate property without reference
360

to the traditional control of wealth exercised by chiefs and
headmen, while labour migration enabled young men to bypass
the control of elders over bride-wealth. Chiefs and elders
attempted to control this threat in a variety of ways, with varying
degrees of success: some were able to establish a loose alliance of
interest with recruiters, or the colonial agents; others, perhaps
more dependent on the labour power of young men than the
tribute they could extract from returning migrants, ended up in
headlong confrontation with the new expansionist forces.
The creation of a wage labour force was undoubtedly the result
of the demand of the colonial economies; that it took the form
of migrant labour, however, was related to the complex struggles
between and within ruling classes over the labour power of young
men. These struggles were extraordinarily diverse, and were
shaped by ecology and pre-capitalist forms of production, the
power structure in rural societies, the timing of colonial pene-
tration and the influence of Christian missionaries. Structural
relations within African societies acted as powerfully on the
actions and consciousness of migrants as any 'pull' of market
forces, growth of' new wants', or individual choice on the one
hand, or the ' determining role of the South African state' on the
other.
In the lands between the Limpopo and the Zambezi, the
hunting of elephants for ivory, which flowed to the south as well
as to entrepots on the east and west coast, the trade in gold from
the lands of the Shona to the Portuguese at Tete, and quite
considerable slave-trading were the major forms of economic
enterprise that were geared to production for overseas markets.
Increasingly, in this period, however, it was becoming clear that
new opportunities and new dangers were to come from the south.
Further north and east, the period saw a continuation and
intensification of the connections to the east coast. In these
regions, where the ivory trade was inextricably linked to the slave
trade, the results were shattering. Together with the Ngoni
intrusions, the advent of well-organised slave and ivory traders
from the east coast transformed the political and social geography
of large tracts of what is now Malawi and eastern Zambia.
361

THE DISCOVERY OF DIAMONDS AND ITS IMPACT
The discovery and subsequent mining of diamonds and gold in

southern Africa in the 1870s and 1880s was not fortuitous.
Africans in southern Africa had prospected for and exploited the
gold, copper and iron of the sub-continent since the first
millennium AD. By the mid-nineteenth century, European
explorers were aware that minerals existed in abundance in the
Transvaal and adjacent territories. What was lacking was the
compulsion and the expertise to exploit them. In the bitter
economic depression of the 1860s, at a time of increasing
international demand for cheaper raw materials, the pace of
mining enterprise quickened.
The discovery of the first diamond at Hopetown whetted the
appetites of traders and explorers for more. In 1868, Karl Mauch,
the German hunter and geologist, reported that he had found gold
on the Tati river, to the north-west of the Transvaal and only
150 km from Bulawayo, capital of the Ndebele kingdom. The
finds at Tati, followed shortly after by news of discoveries at
Lydenburg in the eastern Transvaal, brought hundreds of
individual miners to the goldfields from as far afield as Cornwall,
California and Ballarat in Australia. Their excitement was short
lived. There was little gold, and the transport and other costs were
enormous, though the finds in the eastern Transvaal created a new
market for agricultural produce and labour.
Despite continued exploration for mineral resources, the first
half of the 1870s was dominated by the complex affairs of
Griqualand West, as the diamondiferous region came to be
known, where a complicated wrangle erupted between the
African polities, the northern republics, the Cape Colony and
Britain for control. Even prior to the mineral discoveries, the
semi-desert territory had been in dispute between the Griqua
under Nicholas Waterboer, the southern Tswana (Tlhaping) and
Boer contenders. By 1870, both the Free State and the South
African Republic had established magistracies in the region,
although Griqua claims were being pushed vigorously by a
Coloured lawyer, David Arnot, and the Tlhaping chiefs continued
to hold their own against the white diggers, who had proclaimed
themselves an independent republic at Klipdrift in the heart of
the disputed lands, under the presidency of a former British
362

seaman, Stafford Parker. Without British intervention, the
Tswana might well have been able to oust the white diggers, but
territorial disputes which were compounded by the schemes of
colonial merchants and land speculators in search of land and
mineral concessions increased the pressure for British action.1
As anarchy threatened to spill over into violence and no
agreement between the contending parties could be found, the
matter was brought before Lieutenant-Governor Keate of Natal,
for arbitration, and in October 1871 he awarded the diamond-
iferous land between the Harts and Vaal rivers to Waterboer's
Griqua. Although both the Tlhaping and the Orange Free State
vehemently rejected Keate's Award, the British governor at the
Cape, Sir Henry Barkly, annexed Griqualand West in the name
of Waterboer, proclaimed the Keate Award boundaries, and
appointed Richard Southey, colonial secretary at the Cape and
an imperial expansionist of long-standing, the first-lieutenant-
governor of the new Crown colony.
Barkly's actions caused an instant uproar; they greatly em-
bittered relations between the British and the Afrikaner Orange
Free State, which was still smarting from the imperial annexation
of Basutoland after the Sotho/Free State war three years before
and soured feelings between the governor and the newly inde-
pendent government of the Cape Colony. A protracted struggle
now arose over land rights in Griqualand West itself, where
Southey was anxious to broaden the economic base of the new
colony, through the development of capitalist agriculture, and to
finance the government through the auctioning off of Crown
lands. The insecurities of agriculture in an ecologically fragile
environment was the first obstacle, but his plans were chiefly
thwarted by the claims of the colonial land speculators, who had
used the five years before the British annexation to 'build up a
mass of rival claims to vast tracts of the most useful land in the
territory',2 and the extensive occupation by Tlhaping and Griqua
peoples of the farms he hoped to dispose of to white settlers.
It was only under Southey's successor, Major W. O. Lanyon,
1
K. Shillington, 'Land loss, labour and dependence: the impact of colonialism on
the southern Tswana, c. 1870-1900', Ph.D. London, 1981, 77-95.
2
K. Shillington, 'The impact of diamond discoveries on the Kimberley hinterland:
class formation, colonialism and resistance among the Tlhaping of Griqualand West
in the 1870s', in S. Marks and R. Rathbone, Industrialisation and social change in South
Africa: African class formation, culture and consciousness ifyo-ifjo (London, 1982), 108.
363

that the problem was resolved, with the issue of land title to white
claimants and the crushing of African resistance. In 1876 a land
commission sat to hear the claims. Although the territory had been
annexed originally in Waterboer's name, the commission virtually
ignored Griqua title to the land. Moreover, although it revealed
conclusively the fraudulent nature of the original Griqua claims
to sovereignty over the Tlhaping in the area, to have admitted
this would have undermined the very basis of the British
annexation. Instead, the Tlhaping were allocated 'native
locations', along the lines of those in the eastern Cape.
For the Tlhaping and Griqua rulers in the area Lanyon's
policies created a crisis of major proportions. Initially, it is true,
some Tlhaping had benefited from the growth of the diamond
fields; by increasing their agricultural production and trading
wood and dairy products to the new urban centre, they were able
to avoid being forced onto the labour market. Yet this economic
independence was both limited and short-lived. The rapid shift
towards surplus production for the market greatly increased
stratification, undermined an already fragile environment and
profoundly disrupted the Tlhaping polity of Jantje Mothibi,
already weakened by the impact of mission-inspired individual-
ism and the liquor trade. At the same time, white farmers began
to compete for the well-watered agricultural lands around the
water sources, and the area became denuded of the wood which
was in such heavy demand for fuel and heating in Kimberley.
Conflict between the Tlhaping chiefs based in and just beyond the
colony and the new authorities increased as chiefs attempted to
exact tribute from incoming settlers, and refused to respect
surveyors' beacons.
For the Griqua, British annexation was even more disastrous.
Although Waterboer had requested imperial intervention in the
hope of retaining control over the Griqua, while gaining British
protection against the incursions of the Boers and the disorders
of the diggers, he rapidly discovered that his former rights were
totally superseded by the new administration. With annexation,
the wealth and authority of the Griqua declined rapidly. By the
1870s, the Griquas' fragile hunting and herding economy was
already in decline. Unlike the Tlhaping, they found the only way
to earn ready cash was through the sale of land to the ever-ready
speculators, and even those with title to land were ignored by the
364

1876 Land Court. After a history of collaboration with the British,

individual Griqua, including Waterboer himself, now found
themselves brought before the colonial courts for minor infringe-
ments of a new and alien law.
In 1878, encouraged perhaps by the prior uprising of their kin
in Griqualand West, the embittered and impoverished Griqua
rose against the high-handed administration; they were joined by
Jantje Mothibi's son, anxious to restore chiefly hegemony amongst
the Tlhaping, and by a scattering of southern Tswana and Kora
peoples both within and on the fringes of the new colony, whose
lives had been similarly dislocated by alien pressures, wrought by
missionaries, traders, settlers and the new government in
Griqualand West. Although the resistance was easily put down
with the despatch of a British military expedition under Colonel
Warren, in the context of the simultaneous uprisings elsewhere
it roused widespread fear, and accelerated the decline of the
Tlhaping and the western Griqua.3
The apparent end of African opposition, the establishment of
locations and the collection of hut tax for the first time now
facilitated the incorporation of Griqualand West by the Cape
Colony. Although Britain had taken over Griqualand West on the
understanding, that it would be annexed by the Cape, in the early
1870s the colonial ministry was reluctant to take on the burden
of an administration that could not pay its way; it was also
unwilling to antagonise its republican neighbours, already in-
censed at the British annexation, especially in view of the large
number of Free State sympathisers amongst Cape Afrikaner
voters. With the payment of an indemnity to the Orange Free State
by the British government, and the crushing of African opposition
in 1878, the Cape government was persuaded to incorporate the
Crown Colony, and in 1880 the first four members for Griqualand
West took their seats in the Cape parliament. Among them was
Cecil Rhodes, who had first made his way to the diamond fields
in 1871, and who was already making a name for himself as a
leading entrepreneur.
By the time the Cape took over Griqualand West, new patterns
3
The section on the uprisings in Griqualand West is based on I. B. Sutton, 'The
1878 Rebellion in Griqualand West and adjacent territories', Ph.D. London, 1975; for
the southern Tswana, see Shillington, 'The impact of the diamond discoveries* in
Marks and Rathbone, Industrialisation and social change and A. J. Dachs, 'Missionary
Imperialism', J. A/r. Hist., 1972, I J , 4, 650-2.
565

of ownership had begun to emerge on the mines, and the question
of the form of control over the state had become a matter of great
moment to the new men of capital, like Rhodes, who were coming
to dominate Kimberley. Between 1867-70, individual prospectors
and diggers had worked the alluvial fields, and the vast majority
of stones were found by Tlhaping, Kora and Griqua north of the
Orange river, and sold to traders at Hopetown, who were already
heavily involved in the interior. They were the agents of the Cape
merchants, most of them connected with British firms, who
handled the gems, which were then sold on the European market.
Port Elizabeth firms, being nearest the diamond fields, secured a
predominant part in the new business. The merchant houses also
supplied the food and materials required by the diggers. Labour
was undertaken by the diggers themselves or, once whites made
their way to the 'dry' diggings in large numbers after 1870, their
servants, many of whom accompanied their masters to the fields.
In general, the merchants of Cape Town and Port Elizabeth had
been loudest in the clamour for British annexation in 1871, while
the majority of white diggers favoured a continuance of the
'diggers' democracy' or annexation by the Orange Free State,
which they felt would be more sympathetic to the master-servant
relations they wished to see established on the fields.4
By the time of the British take-over, there were some 10,000
white diggers in the new colony and about 30,000 black, but
estimates vary considerably. Fortune hunters both black and
white flocked to Griqualand West from all over the world. About
a quarter of the whites were from overseas, the rest from South
Africa. Many left the settled districts of the Cape and Natal: Cape
Town was said to be denuded of population. Cape Coloured
people, Mfengu and Ngqika, set out as prospectors and inde-
pendent diggers, but the majority of Africans went as labourers,
many sent by their chiefs to obtain firearms, the main commodity
on which wages were spent. Between 1871 and 1875, some 50,000
Africans came to work on the diamond fields every year. Many
of them were migrants from Basutoland, but the majority came
from the north-eastern Transvaal and were collectively known as
« Both this section and what follows are heavily based on Shillington, ' The impact
of the diamond discoveries on the Kimberley hinterland', and R. Turrell,' Kimberley:
labour and compounds, 1871-1888', both in Marks and Rathbone, Industrialisation and
social change.
566

Pedi; both groups were following a tradition of working on Cape

farms and public works, which went back a generation.
The ' dry' diggings consisted of four mines, named after the
earlier farming properties: De Beers, Dutoitspan, Bultfontein and
Colesberg Kopje or Kimberley. Of these Kimberley and De Beers
were the rich mines, in both quantity and quality of diamonds
products; collectively, the diggings became known as Kimberley.
With the rush of individuals to the mines, a myriad of claims were
staked out - as many as 3,588 in 1871, some even divided into
halves and quarters. Southey, who feared the narrow dependence
on a single resource which would follow from floating the mines
on the London stock-exchange, limited individual diggers to only
two claims, 30 x 30 ft square. In this way, too, he hoped to
maintain a large population on the fields. So long as the interests
of the small digger remained paramount, small-scale black pro-
ducers were able to compete with white. Mining methods were
crude, based on open-cast quarrying carried out by diggers using
picks and shovels, and needed little capital outlay. While white
diggers depended on an uneasy supply of African labour, the black
digger was able to rely on family labour, and in 1875, when the
number of claim-holders had already shrunk dramatically, there
were still 120 black claim-owners (out of a total of 135), almost
all of them concentrated on the 'poor man's kopje' of
Bultfontein.5
The success of the black diggers formed the focus of the
discontent of the many small white claim-holders and diggers,
whose hopes of instant wealth rapidly disappeared when con-
fronted with the difficulties of diamond production and sale. A
luxury product, diamonds were also dependent on the state of the
international economy and the whims of the wealthy. Over-
production could easily flood the limited market, while a slump
on the European stock exchanges had immediate repercussions
for local diggers.
As crucial for the diggers were the difficulties and dangers of
their primitive technology. As the mining operations went deeper
and deeper underground, the individual activities of the small
digger threatened the prosperity of all; the sides of the closely
juxtaposed pits threatened to collapse, and the bottoms filled with
5
Turrell, 'Black Flag revolt', J. Southern Afr. Studies, 1981, 7, 1.
367

water. The yellow ground near the surface turned blue, which it
was feared marked the exhaustion of the diamond pipes, and many
of the small independent diggers and share-workers (miners who
worked claims for absentee owners, engaging gangs of labour
on their behalf to actually work the claims, and receiving a
percentage of the proceeds) now began to be eliminated. Although
initially share-working offered some advantages to both sides,
enabling men without capital to engage in mining, and diminishing
the costs and uncertainties of development for the owners, the
relationship was a source of constant conflict. It was becoming
clear that there was a need for more expensive and complex mining
techniques and methods, and that the day of the individual digger
was coming to an end. The more affluent claim-owners attempted
to tighten their control over share-workers by contracting them
as servants, and clamoured for a lifting of the restriction on the
number of claims any individual could hold. Although Southey,
with his very different plans for the region, initially resisted the
clamour, by 1874 he was forced to concede ten claims per person.
This satisfied neither the impoverished small digger, nor the larger
claim-holders, who wanted more.
It was Southey's policy on labour which roused the greatest
anger, however. Part of the agitation for a British take-over had
been in the hope that the creation of a colonial state would give
the claim-holders greater control. When Southey failed to accede
to their demand that a Vagrancy Law be introduced and the Cape's
Masters and Servants laws be more rigorously implemented, a
white Diggers' Protection Association was formed. Composed
largely of share-workers, and led by struggling claim-owners
backed by storekeepers, canteen-keepers and small diamond-
buyers, members of the association patrolled African areas,
allegedly to keep blacks 'under proper surveillance'.
Small white diggers and share-workers who were being
squeezed both by the large claim-holders and cheaper African
labour, did not join action with the black small claim-owners and
share-workers in their opposition to the process of mutual
impoverishment and proletarianisation, as one might have ex-
pected, but allied with the more successful claim-owners. From
the first, their grievances were couched in a virulently racist
discourse. This did not originate de novo in Kimberley, although
its instrumentality was clear. While the polarities of' white' equals
368

'civilised' and 'black' equals 'barbarous' were well established

in the European and colonial mind well before the foundations
of Kimberley, they were profoundly reinforced in the 1870s by
the impact of Social Darwinism and the circumstances on the
mines, where white diggers saw little in common with their
unskilled African fellows. The latter were accused of every ill,
including sole responsibility for the large-scale theft of diamonds
on the mines. While IDB (illicit diamond buying) by both black
and white was undoubtedly a problem, its importance was grossly
inflated both then and later by claim-owners and white diggers
for their own purposes.
In 1873-4, as drought followed by floods sent upfoodand
transport costs in southern Africa, the outcry against Southey's
rule mounted; at the same time the world price of diamonds
plummeted as a result of a slump on the Austrian stock exchange,
and the banks began to foreclose on the indebted small diggers.
In March 1875 the so-called Black Flag rebellion broke out.
Although it was easily put down by a show of force brought up
from the Cape, Southey faced a severe crisis; he lacked staff, funds
and support from the local community, and in July 1875 was
recalled in disgrace.6
The Black Flag rebellion marked a major transition on the
diamond fields. Southey, with his Cape landed and mercantile
allies, was replaced by a lieutenant-governor more amenable to
the demands of overseas capital. A commissioner was appointed
to enquire into the financial affairs of the Crown colony and his
reports ' constituted a virtual invitation to companies to take over
the development of the mines'.7 The legal framework of the
colony was changed to permit the formation of joint-stock
companies and all restrictions on claim-holding were removed.
After the commercial crisis in mid-1876, when the price of
diamonds fell by one-third, international capital began to flow into
the mines and by 1881 the number of claims at Kimberley had
been reduced to 71, dominated by twelve joint-stock companies
with a combined share capital of £2.5 million (a third of the total).
Simultaneously, the technology at the fields was transformed with
the introduction of horse whims, washing machines to break
6
The sections on Kimberley and the Black Flag revolt are deeply indebted to
R. Turrell, 'Kimberley: labour and compounds, 51-2; and his 'Black Flag revolt on
the Kimberley diamond fields'.
' J. Flint, Cecil Rhodes (London, 1976), 44.
369

down the diamondiferous ground, and steam engines. The new
phase of company mining had begun.
The next half-a-dozen years saw these 71 claims further
reduced, as the larger companies competed to control the diamond
mines, a process which was accelerated by a crisis of over-
production on the mines in the early 1880s and by the crash in
the share market in 1886 which led to the demise of many of the
remaining smaller companies. By 1887, two giants had emerged
in the struggle to monopolise both production and output: on
Kimberley mine, Barney Barnato, an East End Jew whose life
story seemed to epitomise the rags to riches mythology, had
bought out most of his opponents; on De Beers mine, all the
claims had been taken over by Cecil John Rhodes. By 1887, the
issued share capital of his De Beers Mining Company was
£2.5 million (compared to £200,000 in 1881).
By this time Rhodes was not only one of the wealthiest men
in South Africa, but also a prominent Cape politician. As member
for Kimberley, he had played a crucial role in the British
annexation of Bechuanaland and was poised to launch his thrust
into the interior. Essentially, it was the wealth from the diamond
fields which enabled him, and the imperatives of his mineral
kingdom which impelled him to do so.8
The competition between De Beers and Kimberley Central
Mine for the final monopolisation of the diamond fields has been
largely personalised as a contest between Barnato and Rhodes; in
fact Barnato was relatively easily bought off, although his
shareholders and managers held out rather longer. What was
really crucial in Rhodes's final success, however, was the support
he acquired from powerful financiers in Britain and France, who
appreciated the need for monopoly if the industry was to remain
profitable. With the formation of De Beers, the amalgamation of
the diamond fields was complete; the way was cleared for the
centralisation of both production and marketing. In 1899, the first
diamond syndicate was formed by the dealers who had been most
involved in the creation of De Beers. In just over twenty years
the multitude of black and white diggers had given way to a major
internationally backed monopoly.9
For workers, both black and white, the implications of the
8
Refs re Rhodes to: Shillington, 'Land loss', 214-31.
0
D. Innes, Anglo-American and the rise of modern South Africa (London, 1984), ch. i.
37°

amalgamation of the diamond fields were immense. In 1888-9, 37

per cent of workers lost their jobs - 2,770 Africans and 240
Whites. At the same time African wages, which had risen to 30
shillings a week in the 1870s, was brought down to between 7s.
6d. and 8s. 6d. a week with food. Between 1889 and 1898 working
costs on the diamond fields were reduced by 50 per cent.10 Even
more crucial for the future pattern of industrialisation in southern
Africa, and for both the racial and sexual division of labour, was
the implementation of a rigorous closed compound system for
Africans at Kimberley.
Usually defended as a response to the widespread theft of
diamonds, the compound system can be traced to the falling rate
of profit on the diamond fields in the early eighties, and the
changes in the work process as the new joint companies turned
to underground mining. By 1886, De Beers was able to introduce
closed compounds for its workers, and this provided the model
elsewhere after amalgamation."
It was of fundamental importance for the future of southern
Africa that white workers were able to resist being housed in
compounds; in this they were assisted by the opposition of the
merchants to any form of compounding which they feared would
damage their sales, though they couched their opposition in terms
of 'free trade' and protecting 'native interests'. In relation to
African workers, the merchants were satisfied by promises that
they would provision the compounds, but through their inter-
vention skilled white and Coloured workers escaped the com-
pounds. In fact, both these groups were already in a privileged
position in relation to the 'raw natives from the interior', from
whom they were quick to distinguish themselves. With the
formation of companies, independent white diggers were trans-
formed into overseers bought off with higher wages; they col-
laborated with management in 'disciplining the African labour
force'. They were as carefully, if somewhat more congenially,
controlled in the white working-class suburb of Kenilworth,
effectively tied to the company and isolated from the African
workers. Most of the Coloureds were absorbed as an urban
artisanate and petty bourgeoisie, joining the already considerable
10
ibid.
1
' For a vivid description of the closed compounds by the man who claimed to have
established them for Rhodes, see N. Rouillard (ed.), MatabtU Thompson: an autobiography
(Johannesburg, 1957: reprinted Bulawayo, 1977), 82-3, 88.
371

settled urban black population in Kimberley, or making their way
north to the new gold fields.
The compound system was of immense importance to the
mining industry; it limited the possibilities of united worker
action, prevented desertion and diamond smuggling, and very
substantially reduced the costs of reproducing the labour force.
Yet the compounded migrant was not something decided upon
and simply imposed by De Beers of its own volition. Indeed, the
new industrialists epitomised by Rhodes would probably have
preferred some form of lengthy indentured servitude, if not
outright slavery.12
Nevertheless, the imposition of the compound system marked
a profound transformation in the position of Africans by the
mid-1880s compared to the 1870s, a transformation which can
only be understood against the wider socio-political changes of
the period. In the 1870s, a major preoccupation of every colonial,
republican and imperial interest had been to find and control an
adequate supply of labour. In the Cape and Natal, it is true, the
earlier wars of dispossession and the process of stratification
amongst the peasantry, which had formed around the mission
stations from the mid-century, had already created a pool of
landless labour. However, with the diamond discoveries, and as
local economies began to recover from the depression of the
sixties, an insatiable demand for labour - in the greatest possible
numbers at the lowest possible cost - came from all sides. In both
colonies, new legislation was passed to increase and control local
workers, and both began to look beyond their frontiers for
additional hands.
At the Cape, self-government led to an ambitious development
of railways and harbours; as much for political as for economic
reasons, the Cape ministry procured loans through their Agent-
General in London, and Molteno's Railway Act of 1874 authorised
the construction of no less than four main lines, connecting the
main commercial centres of the Cape to Cape Town, Port
Elizabeth and East London. By 1881, the great drive was to reach
Griqualand West, where diamonds had overtaken all other Cape
exports. Between 1874 and 1885, £zo million had been transferred
from London into Cape government securities, most of it for
12
See, for example, T. Collingwood Kitto cited in R. Turrell, 'Kimberley: labour
and compounds'; 62.
372

expenditure on railways and harbours. The international move-

ment of capital was matched by the mass movement of workers:
between 1873 and 1883, some 23,000 European immigrants were
brought out as workers to the colony. Most of the railway
'navvies' were from Britain, but there were also Germans and
Belgians. At Cape Town docks, black labour was drawn on from
as far afield as Mozambique and housed in compounds at the port,
while elsewhere strenuous efforts were made to induce eastern
Cape Africans to work on the railways and on the farms.
In Natal, indentured labour from India was renewed and an
increasing number of Tsonga from southern Mozambique were
recruited to meet the rising demand from the sugar plantations
and public works. At the same time, Theophilus Shepstone, then
Natal's secretary for Native Affairs, began to look longingly at
the dense concentrations of people in Basutoland, the eastern
Transvaal and even the lands of the Ndebele as fields for Natal
expansion.
With the creation of the new market at Kimberley, which had
become by 1877 the second largest town in South Africa, Africans
found they could earn tax and rent and fulfil their rapidly growing
consumer wants through crop production rather than by working
for settlers. In Natal, the Eastern Cape and Basutoland, this was
the decade of maximum opportunity for African peasant
producers, and many households sent their young men to the
diamond mines to earn the money for the new ploughs, seed and
fertilisers that were beginning to transform the countryside.
White farmers too were beginning to respond to the new market,
and for the first time Afrikaner farmers in the Cape Colony formed
a Boeren Beschermings Vereeniging (a Farmers' Protection
Union) to do political battle on their behalf in parliament. As
undercapitalised farmers, they were most vociferous on the need
for cheap black labour. By the early 1880s, under J. H. Hofmeyr,
they were beginning to make their presence felt in the Cape
Legislative Assembly.
At Griqualand West, mine-owners had blamed Southey for
their labour difficulties, but there was in fact little he could do
as long as Africans had access to land, were able to supply or
withdraw their labour at their own discretion, and had a choice
of occupation. Although Crown colony rule had been declared
over the immediate environs of the diamond fields, there was still
373

no overarching state in the region capable of pulling Africans out
to work through the imposition of taxation or corvees. Labour
recruiters' were soon ferreting out workers from as far afield as
Barotseland and Matabeleland, the eastern Transvaal and
Mozambique, but the demand for cheap unskilled labour con-
stantly oustripped the supply.
Such was the bargaining power of workers in Kimberley that
through the 1870s wage levels constantly fluctuated in response
to the willingness of Africans to offer themselves for labour;
wages were higher than anywhere else in southern Africa, and
there were bonuses when the haul of stones was good. By 1875,
they had gained considerable independence, by insisting that they
should feed themselves. This led to an increase in wages and the
proliferation of black eating houses and canteens. Many Africans
left the informal diggers' 'compounds', and set up their own
camps on the outskirts of the townships. The rate of desertion
and drunkenness was high and the complaints of employers
continuous.
Most important of all was the workers' ability in Griqualand
West to purchase guns in exchange for their wages, a powerful
symbol of their independence and a potent means of ensuring its
continuance. Much of the money earned on the fields was used
to purchase guns, which cost from £4 for an old musket to £11
for a modern breech-loader. The trade in guns reached huge
proportions by 1875, after which time the authorities tried to
clamp down on it. The sale of guns and gunpowder to the
workforce constituted the major contradiction on the diamond
fields; so long as firearms remained the major way of attracting
labour to the mines, neither the claim-holders nor the state could
solve the problems of control. Within two or three years of the
beginning of mining at the 'dry' diggings, there was hardly an
African group from the Transkei to the Limpopo that was not
armed with fairly effective guns.
For both black and white this rearming of African peoples, at the
very moment that fresh encroachment was being made on their
lands, was highly charged with emotion. All over South Africa
it was believed that African rulers under the leadership of
Cetshwayo, king of the Zulu, were conspiring 'to drive the
whites into the sea'. It was in Natal, however, that this conspiracy
374

theory was most ardently held; there the African possession of

guns roused near hysteria.
Ever since its establishment as a British colony in 1845, Natal
had been prone to rumours of an imminent African invasion. A
tiny group of Whites - some 17,000 in the 1870s — was scattered
over a vast area, their settlements surrounded by reserves and
huge tracts of absentee-owned lands, occupied by 300,000—
3 50,000 Africans. Most Africans were still ruled by chiefs, though
these had been carefully chosen by Shepstone, as secretary for
Native Affairs; with only eleven magistrates at his disposal in
1871, he was only too well aware of the limitations imposed by
a lack of men and money. Despite the outcry of settlers and
missionaries against such practices as lobola (bridewealth) and
polygeny, Shepstone recognised that to eliminate them would
radically transform African daily life. In the late 186o's agricultural
competition from the culturally segregated Africans added to
Natal settler anxiety. On their borders were the densest African
populations in southern Africa, all of them still living in powerful
independent kingdoms: Zulu to the north, Sotho to the west, and
Mpondo to the south. A variety of measures — taxation, heavy
tariffs on imports destined for the African market, the enforcement
of European-style dress, fines and fees of court - had failed to
push Africans into working on settler farms. The fears and
frustrations of the struggling colonial farmers, living cheek by
jowl with African neighbours whose land and labour they
coveted, were quickly translated into rumours of African
disaffection.
In 1873, the Hlubi, an African chiefdom settled in the north-west
of Natal, under its chief, Langalibalele, a renowned doctor and
rain-maker, refused to register the guns they had earned at
Kimberley, lest they be confiscated without compensation.
Successful peasant farmers, the Hlubi were also among the first
Natal Africans forced to seek work on the diamond fields. As far as
the officials were concerned, they were already guilty of a number
of offences. When Langalibalele allegedly manhandled the
messengers sent to summon him to Pietermaritzburg to account
for his peoples' actions, Shepstone was enraged, and sent a small
force of white volunteers and African ' scouts' to castigate the
chief and his followers.
Langalibalele did not wait for the humiliation. In November
375

1873 the chief, his large family, and less than a thousand fighting
men, together with a great herd of cattle, fled across the Bush-
man's River Pass into Basutoland, Natal troops in hot pursuit.
When the Hlubi rearguard clashed with their pursuers, and three
whites and two African scouts were killed, the Natal force
panicked and retreated into Natal. News of this minor military
episode 'ran through white Natal like sheet lightening, dis-
charging thirty years' store of accumulated hate and fear'.13
British reinforcements were sent to Natal and Langalibalele's
location was destroyed, as was that of the neighbouring
amaNgwe Ngwane people thought to have been his allies. Two
hundred people were killed, the remaining women and children
driven into settler service, their lands and goods confiscated. At
the same time, Langalibalele and his followers were hunted out in
Basutoland by a large armed force from the Cape and Natal, and
the chief was returned to a highly irregular trial in Natal. Far
exceeding its powers, the court sentenced Langalibalele to life-
long banishment on Robben Island.
In London, Lord Carnarvon, the recently appointed secretary
of state for the Colonies in Disraeli's Conservative cabinet,
recalled the lieutenant-governor, Sir Benjamin Pine, mitigated
Langalibalele's sentence to exile outside Cape Town, granted
compensation to the amaNgwe people, and despatched Sir Garnet
Wolseley, hero of British India and of the Asante war of 1873,
to Natal as special commissioner. For Carnarvon the affair
highlighted the complexity of affairs in southern Africa and
reinforced his conviction that the only solution to its problems
lay in some form of confederation.
CONFEDERATION AND THE ANNEXATION OF THE

TRANSVAAL
The merits of a confederation policy to the man who had piloted
the successful federation of Canada in 1867 seemed self-evident:
the creation of a self-governing, united South Africa within the
British empire was seen as the cheapest and most effective way
of securing metropolitan and colonial interests. A unified
13
N. A. Etherington, 'Labour supply and the genesis of South African Confedcr-
ttion in the 1870s', ]. Ajr. Hist., 1979, ao, 2, 246.
576

CONFEDERATION AND THE TRANSVAAL
administrative structure, communications system and defence
policy would encourage both colonial and overseas investors.
Confederation would resolve local political and financial problems,
such as the continued tension over the Free State's claim to the
diamond fields, the chaotic finances and increasing civil disorder
in Griqualand West, and the refusal of the Cape to take the colony
over; and it would also create a more effective state in the region,
capable of taking concerted measures to deal with relationships
between black and white.
All over South Africa, growing competition for African lands,
itself a product of the rise in land prices and commercialisation
of agriculture which followed in the wake of the diamond
discoveries, threatened to black-white confrontations. At the same
time, the continued independence of powerful and self-sufficient
African kingdoms posed an obstacle to the transformation of their
people into a cheap, controlled and self-perpetuating labour force
for the rapidly evolving capitalist economy of the sub-continent.
The proliferation of firearms in African hands and the inability
of the republican governments to ensure 'a free flow of labour'
to the diamond fields and to other commercial enterprises in the
region could only be dealt with by a form of organisation
transcending existing frontiers and parochialisms.
The imperial problem, which proved insoluble in this period,
was how to persuade or cajole the settler polities to agree to
submerge their local sovereignties. Ever since the annexation of
Griqualand West, imperial officials had hoped the Cape Colony
would take the initiative in such a policy. By the 1870s, it was
the largest and most populous of the settler states in South Africa,
with a white population of some 257,000 (compared to a total of
less than 100,000 elsewhere in South Africa) and an internal black
population of about half a million. Much of the rest of South
Africa was an extension of the Cape's commercial economy. The
spate of railway building and public works in the 1870s had
greatly increased the colony's dependence on British capital and
credit, and major British interests had become involved in the
economic and political changes in the region. They were as
anxious to see the continued viability of the Cape economy, in
the context of expanded opportunities for capitalist development
in the region as a whole. No policy of federation could succeed
without the support of the Cape Colony, and it was hoped that
377

the grant of responsible government would enable it to join with
the Orange Free State as a first step towards such a federation.
Despite the misgivings of Western Cape conservatives and the
continued agitation of easterners for some form of separatism, in
1872 the first self-governing ministry under Sir John Molteno was
elected at the Cape. Reluctant to give up their new independence
and unwilling to shoulder the financial and defence burdens of
the region, however, from the outset the new cabinet was hostile
to the idea of federation. Carnarvon's appointment of the historian,
J. A. Froude, as his personal emissary to South Africa (over the
head of the high commissioner and the Cape ministry), Froude's
encouragement of eastern Cape separatism and the publication of
Carnarvon's first despatch on Confederation in May 1875, which
signalled renewed imperial assertiveness in the sub-continent — all
these actions simply rendered the Cape ministry more suspicious
of imperial intervention.
The Langalibalele affair appeared to offer Carnarvon a way out
of this stalemate. In the hope that it could then be coerced into
a confederation with Griqualand West, Wolseley was instructed
to reduce Natal to Crown colony status, and in six months he had
succeeded through a mixture of flattery, bribery and cajolery.
More importantly, he convinced Carnarvon that he could achieve
his ends in South Africa by direct and forceful means more readily
than through endless discussions with responsible ministries. The
colonial secretary was further encouraged in this view by Natal's
highly respected and influential secretary for Native Affairs,
Theophilus Shepstone, who represented the views of Natal's
expansionists.14
After two abortive attempts to call a conference on confedera-
tion in South Africa, Carnarvon summoned colonial repre-
sentatives to a conference in London in August 1876. By this time,
he was even more concerned to push forward with his plans,
because the 1875 Macmahon Award of Delagoa Bay to Portugal
had given the Republic independent access to the sea, which
would enable it to outflank the hold of the Cape and Natal ports.
At the same time, the attempt by President Burgers to raise money
in Europe to build a railway to Delagoa Bay increased Carnarvon's
apprehension that the republic would soon be beyond his reach.
The conference achieved little, however; neither President Brand
14
Etherington,'Labour supply', 248.
378

of the Orange Free State nor the Cape prime minister, Molteno,
was ready to discuss federation. Through an oversight, President
Burgers of the Transvaal had not been invited, and Griqualaland
West was not directly represented.
In the midst of this, news arrived of the humiliation of the South

African Republic at the hands of the Pedi. This gave Carnarvon
a fresh opportunity for assertive action. The Pedi defeat of
President Burgers and his troops revived the spectre of a general
African uprising, bankrupted the Transvaal treasury and in-
creased internal dissensions. Having failed to convince or coerce
the Cape or the Free State, Carnarvon now secretly instructed
Shepstone to go to the Transvaal and annex the republic, if
possible with local consent.
In January 1877, Shepstone proceeded to the republic and by
April had annexed it to Britain, despite the protests of the
Volksraad and the president. He was urged on by the English-
speaking community and the republic's creditors, bankers and
traders, as well as by mercantile interests in the Cape and Natal,
who feared Transvaal independence and looked forward to a
boom in land prices under British rule.
The British annexation of the Transvaal is frequently regarded
as the outcome of Carnarvon's impatient, personal response to a
'crisis' on the 'periphery of empire'. In fact, however, it came
after almost a decade of renewed imperial intervention in South
Africa, which began with the annexation of Basutoland in 1868
and of Griqualand West in 1871. The chronology confirms the
recent contention that for Britain the 'new imperialism' was
signalled by Disraeli's speeches and policies between 1866 and
1868.15 Behind immediate local exigencies broader economic
considerations impelled imperial advance in southern Africa in the
1870s and 1880s; by then, Britain's relative industrial decline,
falling profits, and the growing competition amongst the great
powers, had led 'industrial interests in Britain... into decisive
support for the acquisition of new markets in Asia and Africa'.16
The initial response of the Afrikaners of the South African
Republic to imperial annexation was muted, perhaps because of
15
F. Harcourt, 'Disraeli's imperialism, 1868-1888: a question of timing', Hist. J.,
1980, 33, 1.
16
P. Cain and A. G. Hopkins, "The political economy of British expansion overseas,
1750-1914', Eton. Hist. Review, 2nd series, 1980, 33, 4, 485
379

Burgers' inability to cope with the underlying tensions of
Transvaal society. Despite the known abundance of its mineral
resources — copper, lead, cobalt, iron and coal were all known to
be present - and its sparse white population in an apparently vast
territory (an estimated 30,000 settlers dispersed over
120,000 sq. miles), the Transvaal state was bankrupt and suffering
from an acute shortage of land. By the mid-1870s, speculation and
land shortage had sent up prices, and the setting aside of 3 million
acres as security for the railway loan precipitated a major crisis.
By the beginning of 1876, the sale of public land had virtually
ceased.17
In the Transvaal there was a traditional solution to land
shortage: expansion into the African domain. With the mineral
discoveries in the Transvaal and Griqualand West there was an
added incentive as the market for agricultural produce expanded.
On every frontier between Boer and African, there were uneasy
murmurings, as settlers translated informal grazing arrangements
into land concessions and private title. The resurgence on the
highveld of African power in the 1860s, as African societies
recovered from the demographic effects of the mfecane and were
able to arm themselves still further through working on the
diamond fields, made the labour shortage particularly intractable,
at the very moment when the mineral discoveries were stimulating
agricultural production and increasing class differentiation
amongst the Boers. Neither an increased hut tax nor the intro-
duction of pass laws, both of which discriminated in favour of
those Africans who lived and worked on white farms, solved the
problem, for the state was incapable of effectively administering
them. Attempts to apply the legislation to weaker African
societies simply led them to seek protection from the more
powerful and independent polities like the Pedi, who could not
be coerced. 18
By the 1870s the Pedi had become 'a major alternative focus
of power' both to the South African Republic and the Swazi
kingdom in the north-eastern Transvaal. From the mid-century,
17
C. W. de Kiewiet, The Imperial factor in South Africa (Cambridge, 1957; reprinted
196)), 14;; S. Trapido, 'Reflections on land, wealth and office in the South African
republic, 18)0-1900', in S. Marks and A. Atmore (eds.), Economy and society in
prt-industrial South Africa (London, 1980), 35 0—69.
18
P. Delius,' Migrant labour and the Pedi, 1840-80', in Marks and Atmore, Economy
and society, 299.
380

the paramount chief of the Pedi had sent his young men south
in order to acquire firearms, and Sekhukhune, who became
paramount chief in 1861, deliberately extended Pedi power, by
attracting Zulu and Swazi refugees as recruits to his army. The
flight of so-called apprentices and dissatisfied tenants from settler
farms to the greater security of Pedi protection further increased
its power. On at least three occasions in the decade before the
outbreak of war, the Pedi had resoundingly defeated Swazi
invaders and forced the Swazi rulers to look to the South African
Republic for assistance, a remarkable shift in the balance of
power.19
Through the 1870s, however, the Pedi kingdom began to suffer
from increased pressure on resources, the result of natural
increase, the influx of refugees and recurrent drought. Pedi
subjects settled across the Steelpoort river, where disputes over
land were compounded by the refusal of Pedi subjects to work
on Boer farms. Finally, in 1876, President Burgers, was forced to
mount a military expedition against them.
For the South African Republic the war was a disaster. After
a brief and indecisive campaign, the Afrikaners antagonised their
Swazi allies but were unable to take the Pedi stronghold. The Boer
commandos, disappointed at their failure to capture Pedi cattle,
returned home. Burgers, with his cosmopolitan background,
liberal religious convictions, and progressive schemes for the
modernisation of the Transvaal state, had never been popular with
the majority of Transvaalers. The fiasco in Sekhukhuneland
completely destroyed his credibility. He was unable to raise his
new war tax; and the struggle against the Pedi was left in the hands
of a band of adventurers. Although by February 1877 they had
forced Sekhukhune to sign a truce, their atrocities increased the
outcry against the Transvaal. Burgers was in no position to
impose the refonns he promised when British annexation appeared
imminent; nor, after the event, were there many ready to spring
to his defence. For the most part, the burghers of the South
African Republic waited to see what the imposition of British rule
really meant.
They were soon to find out. Shepstone had little understanding
of Boer society, and his appointees even less. Although he had
19
Dclius, 'Migrant labour', 295; P. Bonner, Kings, commoners and concessionaires. The
evolution and dissolution of the Swa^t iftb century state (Cambridge, 1983), 114-j, 119-
381

SOUTHERN AFRICA, 1867— I 886
promised the burghers that he would respect their 'laws and
liberties', the Transvaal was governed as a Crown colony, with
a nominated legislature; Shepstone had hoped to win support
through financial generosity, but his treasury was bankrupt, credit
in the territory was frozen and ready cash had all but disappeared,
while his closest financial associates were interested parties of
dubious honesty. The first priority in the Transvaal was to raise
enough revenue to pay for its administration. Yet taxes could not
simply be extorted from the African population through random
raiding as was the republic's wont. Although the machinery of
state was made more efficient, and new departments created
(including the all-important Native Affairs Department), system-
atic efforts to raise taxation were hampered by an insufficient
administration. Attempts to impose taxation on the Boer popu-
lation added insult to injury. Within a short time, 'small-hearted
economy and pitiful tactlessness ruined the British chances of
holding the Transvaal'.20 Profound regionalism and resistance to
central government authority now united with an incipient
nationalism, conjured up and led by the former commandant-
general of the Transvaal, Paul Kruger, who even before the
annexation had been the chief challenger to the president's
authority.
Within four years, the Boers had risen in arms against the
British, and, after the defeat of Major Colley at Majuba in 1881,
Gladstone, who had replaced Disraeli as prime minister in 1880
and had always professed his opposition to the annexation of the
Transvaal and his sympathy with the Boers, withdrew from the
South African interior. Paradoxically, the state which Kruger
inherited had been immeasurably strengthened by the four years
of British rule and the imperial conquest of the Pedi and Zulu
kingdoms, as was shown by the success of republican forces in a
nine months' campaign in inhospitable terrain against the
Ndzundza Ndebele in 1883 (the so-called Mapoch's war).21
WARS OF CONQUEST SOUTH OF THE LIMPOPO

A major justification for confederation and the annexation of the
Transvaal had been the danger of a black—white confrontation in
South Africa. And for the Swazi in the east and the southern
10
de Kiewiet, Imperial factor, 145.
21
P. Dclius,' Abel Erasmus: power and profit in the eastern Transvaal' unpublished
paper (Oxford, 1981). „
382

Tswana in the west, British annexation did win a short respite

from the relentless pressure on their borderlands of the previous
decade. Despite the humanitarian rhetoric, however, Carnarvon's
forward policies respected African independence no more than
they had respected that of the Republic. By the turn of the decade,
a spate of wars had erupted, in which imperial forces took the lead
in conquering those African peoples whom the republican and
colonial states had been too weak to subdue militarily on their
own. By the mid-18 80s, the Zulu kingdom had been dismantled,
the Pedi finally vanquished and a vast tract of territory in the
Transkei incorporated into the Cape Colony. A protracted war
between the Cape and the Basotho had ended with the establish-
ment of a British protectorate over Basutoland, while further
west, most of the Tswana people had also come under British
protection.
At the heart of the crises of the late 1870s was the new imperial
agent in South Africa, Sir Bartle Frere, appointed governor of the
Cape and high commissioner for southern Africa in March 1877.
Frere arrived in South Africa with a formidable reputation as an
administrator and humanitarian, a result of his work in India and
his suppression of the slave trade in Zanzibar in 1872. Brilliant,
ambitious and impetuous, Frere envisaged British expansion as
far as Zambezi and beyond, the strengthening of imperial defences
along the coasts of southern Africa and the delimitation of
Portuguese claims to the east and west. He rapidly determined that
independent African kingdoms constituted the major obstacle to
Carnarvon's schemes and the main threat to settler security in
South Africa. Within months of his arrival, Frere had become
deeply involved in the last of the major wars on the eastern
frontier, the dismissal of the colony's first responsible ministry,
and a rebellion in Griqualand West, although it was his provo-
cation of the Zulu war in 1879 which led to his recall in disgrace
in 1880.
It would be misleading to see these upheavals simply in
personal terms, however. Frere acted on ' behalf of a syndicate'
of interests,22 anxious to use the conflicts resulting from the
intensified pressure on African lands and labour for the creation
of a new form of state in the region, capable of ending regional
22
N . Etherington,' Anglo-Zulu relations, 1856-1878'in A. Duminy and C. Ballard
(eds.), The yingio-ZuJu war: new perspectives (Pietermaritzburg, 1981), 14.
583

BECHUANALAND
Walvis
Bay PROTECTORATE
(1884)
SOUTH WEST
2S°S AFRICA
(Germany 1884)
(Me Mahon Award

o Portugal 1875)
SWAZILAND! _
(Br Prot 1907)
*&™^SJ«°»*»™»
00
30°S 30°S
TEMBULAND.GCALEKALAND
BOMVANALANO (1884-6)
FINGOLANO (to Cape 1879)
4>£

Tort St John's
(toCape1884)
BOMVANALAND
1884-6
vGCALEKALAND
1884-6
BRITISH
KAFFRARIA
(to Cape
"East London 150 km
100 miles
14 Annexations of African territory south of the Limpopo, 1866-97

(based on J. D. Fage, An Atlas of African History (London, 1958), p. 37)

rivalries, subjugating African workers and securing the region's
capitalist development. For colonial and imperial authorities alike,
the wars were both expensive and protracted; nonetheless, they
were an essential precondition for the creation of the new social
and economic order demanded by industrial capital, and the
ultimate unification of the South African polities in a single state,
in 1910.
The first of the wars which punctuated Frere's high com-
missionership erupted amongst the Xhosa peoples on the Cape's
eastern frontier after twenty-five years of uneasy peace. In the
Transkei, decades of warfare and the activities of traders,
missionaries and settlers had undermined even those chiefdoms
as yet formally independent of colonial rule. In the Ciskei,
magisterial rule and taxation had long displaced the power of
chiefs. Yet for Africans, both in the Transkei and Ciskei, the first
half of the 1870s were years of prosperity; the relatively high
wages on the diamond fields and in the greatly expanded transport
and public work projects, together with 'a virtual explosion of
peasant activity' amongst Africans in large areas of Fingoland,
Tembuland, Queenstown and Lady Frere districts and even parts
of the eastern Transkei and Pondoland had given many access to
new opportunities and new resources.23 The success of African
peasants in the 1870s did not go unremarked by the white farmers
and other employers anxious to secure cheap labour. As in the
Transvaal, though with more administrative efficiency, the Cape
assembly passed a multitude of new laws to satisfy their farmer
constituents. Anti-squatting legislation, increased taxation, a
tightening of the pass and vagrancy laws as well as the grant of
individual tenure of land were all intended to force Africans onto
the colonial labour market. None succeeded so well, however, as
the droughts and wars which afflicted the Ciskei and Transkei in
the second half of the decade.
Through the 1870s, the Molteno ministry at the Cape moved
to extend its authority in the territory between the Kei and the
Mthamvuna rivers through the appointment of magistrates. In
1873, it appointed Joseph Orpen, 'an ardent expansionist',24 as
its agent east of the Kei, and within a few months of his
" C. Bundy, The rise andfall of the South African Peasantry (London, 1979), 65-78.
24
C. Saunders, 'The annexation of the Transkei', in C. Saundcrs and R. Derricourt
(eds.), Beyond the Cape frontier: studies in the history of the Transkei and Ciskei (London,
1974), 187-8.
386

appointment, the Mpondomise as well as the Griqua east of the

Drakensberg were brought under colonial rule. In 1875, the
difficulties of the Thembu with the Gcaleka who had defeated
them in 1872, led to the extension of magisterial rule over the
former. It was, however, the Cape-Xhosa war of 1877-8, followed
by the Mpondomise—Thembu uprising of 1881, which enabled the
colony to take over Gcalekaland and Bomvanaland, and led to the
stationing of colonial officials in Xesibe's territory and at Port St
Johns in Pondoland, although the latter was not annexed until
1894. In many districts, the expropriation of African lands and
the forced movement of peoples reversed the gains made by the
peasantry over the previous decade; the hundred-year war for
domination on the frontier was at an end.25
1877 saw the continuation of the fearful drought which had
started as early as 1875 in some parts of the Eastern Cape and
Basutoland. Two-thirds of the crop failed and the number of men
seeking work doubled in a year. Peasant prosperity for some
meant poverty for others, and in some areas, especially on either
side of the colonial frontier, the land was 'overpeopled, over-
stocked and overcultivated'.26 Constant skirmishes over cattle
and trespass were relieved only by the ability of peasants to seek
work on the labour markets of the colony and Griqualand West.
The effects of the drought were exacerbated by depression
amongst farmers and traders, with the collapse at Kimberley and
in the Cape wool market - the result of an economic slump in
Europe. By mid-1877, rumours of war abounded. Friction was
most marked between the Gcaleka (the Xhosa under paramount
chief Sarhili [Kreli]) and the Mfengu, which was the result of a
decision in 1865 by the Governor, Sir Philip Wodehouse, to settle
the latter as a buffer on lands confiscated from the Gcaleka in
the Transkei; this was the Mfengu's somewhat ambiguous reward
for more than half a century of collaboration with the Cape
Colony. A minor scuffle at a beer-drink was followed by raid and
counter-raid between the bitterly antagonistic groups, and by
September, the Cape ministry had decided to intervene.
The first stage of the war against Sarhili in the Transkei seemed
all but over within a couple of months. Frere's plans to offer
Gcaleka lands to colonial farmers, however, led to fresh upheaval.
Within weeks the Gcaleka were joined by the Ngqika in the
15
Bundy, Theriseand fall, 83. ** Dc Kiewiet, Imperial factor, 1 jo.
387

colony, and martial law was declared. Despite the objections of
the Cape ministry, Frere, now obsessed with the notion of a vast
conspiracy against whites being directed by Cetshwayo from
Zululand, requested imperial reinforcements. Disagreements
between Frere and the ministry over the conduct of the war led
to a constitutional crisis, and at the beginning of February the
governor dismissed Molteno and appointed the self-serving
J. G. Sprigg to replace him.27
On the frontier, the war dragged on until the middle of the year
when, defeated by famine, the Xhosa threw in their hand. Frontier
defence was now tightened up, with the establishment of several
new forces and three new Transkeian magistracies; and in May
1878 Sir Thomas Upington, the attorney-general, introduced the
Preservation of Peace Act (the Disarmament Act), aimed at
confiscating African-owned guns. Despite promises of com-
pensation and that disarmament would only be applied in selected
rebel areas, the new act was applied indiscriminately, and simply
provoked further rebellion and war in the Transkei and Basuto-
land, in 1879.
As in the case of the Ciskei and Transkei, the first half of the 1870s
had been prosperous years for Basutoland, as the kingdom
recovered from the wars of 1865-8 against the Free State. The
coming of Pax Britannica and the opening of the diamond fields
rapidly revived the economy. Large amounts of grain and stock
were exported, and equally large quantities of manufactured
goods were imported from the Cape, which had annexed the
territory in 1871. As early as 1873, hut tax more than covered the
costs of Cape administration.
In the long run, however, the high level of agricultural
exploitation proved ecologically disastrous, as increasing numbers
of people and cattle were hemmed into a shrunken territory, the
result of the Treaty of Aliwal North, between the British and
the Free State, which allocated the best farming lands along the
Caledon river to the Orange Free State; by the late 1880s the Sotho
were beginning to colonise the difficult barren highlands in the
interior of the country, with results that can be seen on the
landscape to this day. In the 1870s, however, there were few signs
of Basutoland's twentieth century underdevelopment.
" P. Lcwsen, John X. Merriman: paradoxical South Afritan statesman (New Haven,
1982), 63.
388

Politically, the early years of Cape rule were relatively un-

eventful. Despite the stated policy of the Cape administration,
here as elsewhere, to undermine the power of the chiefs and win
the people over to magisterial rule, the hold of the Koena ruling
lineage was greatly strengthened. Letsie, the oldest son of
Moshoeshoe, was recognised as paramount chief, though in
northern and central Basutoland, his brothers, Molapo and
Masupha, held sway. Despite Cape rule, the Koena chiefs deter-
mined to secure their continued power through military organisa-
tion, and annually thousands of young men were sent by their
chiefs to the diamond mines to earn firearms.
The prosperity and grudging acceptance by the Sotho of the
Cape magistrates was shattered in the late 1870s by drought and
the actions of the Cape ministry. The trouble began in 1877, when
conflicts arose between the Phuti chief, Morosi, and a newly
appointed magistrate in the remote, mountainous Quthing district.
Letsie co-operated with the government in defusing the situation,
but the district remained disturbed. When in 1878 the Cape's
newly passed Preservation of Peace Act was proclaimed in
Basutoland, the prospect of impending disarmament caused
widespread consternation. Morosi now openly defied the white
authorities, incorrectly assuming he had the support of Molapo
and Masupha, and perhaps encouraged by the news of Isandh-
lwana. Full-scale fighting broke out in March 1879, with the
paramount chief collaborating with Cape forces. It was a bitter
campaign in dreadful weather, with Morosi and 1,500 Phuti
beleaguered on their flat-topped mountain stronghold. In
November, the mountain was stormed, large numbers of Phuti
were killed and even larger numbers captured and sent as
prisoners to work on Cape Town's breakwater or on eastern Cape
farms.
Even before the end of the hostilities, plans were drawn up for
dividing the Quthing district into white farms, and the hut tax
was doubled to £1. At the same time, the Cape reiterated its
intention of applying the Disarmament Act in Basutoland. Despite
the protests of local officials, French missionaries and liberal
opponents of the Sprigg government that the men to be disarmed
had in fact fought on the colonial side, Frere supported Sprigg's
aggressive policy. In July 1880 the Disarmament Act actually
came into force, and although the paramount chief, Letsie,
389

vacillated, other members of the house of Moshoeshoe emerged
as leaders of the war which broke out in September 1880.
Casualties were high on both sides, and by the following April,
the war had cost the Cape £$ million. Although the fighting
between the Sotho and the Cape petered out, the imposition of
a cattle fine on the rebels and the shortage of fertile land led to
simmering discontent which turned into civil war. The Cape
government was wholly unable to handle the new situation, and
by 1884 the British government reluctantly reassumed control
over the territory. All legislative and executive power was now
vested in the high commissioner. In effect the Sotho had won their
long struggle for limited independence - or so it seemed. Certainly
most of the senior members of the ruling elite — the ' sons of
Moshoeshoe' — had emerged with enhanced power. Their political
and legal power remained practically unchecked until well into
the twentieth century.
For the Cape ministry the imperial annexation of Basutoland
came as a welcome relief; by this time, they would equally happily
have handed over the Transkeian territories, where fresh resistance
to colonial rule had broken out late in 1879. Sparked off by the
murder of a white magistrate in Mpondomise territory and the
expulsion of magistrates from Thembuland, the struggles even
spread to the Ngqika and Mfengu peoples who had collaborated
with the colonial forces in 1877-8, a result of the extension of the
Disarmament Act and the absence of the small Transkeian
policeforce in Basutoland. The conflicts dragged on into 1881, and
cost the Cape government more than £i£ million to quell.
Although the uprisings were uncoordinated, several thousand
Africans were killed, the Sprigg ministry was forced to resign, and
the Cape lost its appetite for further expansion east of the Kei.
According to the contemporary historian, G. M. Theal, the armed
resistance of the Sotho and the Transkeian peoples together
constituted 'the most formidable attempt ever made by natives
in South Africa to throw off European supremacy'.28
For Frere and his supporters, these wars were an unwelcome

distraction, convinced as they were that the true threat to the new
18
F. Brownlee (ed.), The Iramktian native territories: historical records (Lovedale,
1923), 5 3, cited in C. Saunders, 'The Hundred Years War: some reflections on African
resistance on the Cape-Xhosa frontier' in D. Chanaiwa (ed.), Profiles of self-determination
(Northridge, 1979).
390

order in southern Africa came from the Zulu kingdom to the
north of Natal. Fear of the Zulu was a persistent theme in the
history of Natal, and ever since his succession in 1872, Cetshwayo
was believed to be behind all African unrest in southern Africa,
a view sustained by the undoubted communication between the
independent African polities all the way up to the Zambezi.
Increasingly it seemed as though 'power to control the Zulus
includes that of controlling all the rest'.29 It was, however, with
the British annexation of the Transvaal that the Zulu came to be
seen as the central obstacle to Carnarvon's plans for con-
federation.
Struggling Natal settlers had long eyed Cetshwayo's fertile and
populous lands, while for thirty years, as secretary for Native
Affairs in Natal, Shepstone had toyed with the idea of using
Zululand as a 'dumping ground' for the colony's 'surplus'
African population. With the increasing demand for labour in
Natal, he began to intervene ever more vigorously in the affairs
of the kingdom, in the hope of securing Natal's corridor to the
mineral riches and labour supplies of the interior through the
king's domains. Crucial in his ability to do so were continued
Boer encroachments on Zulu lands in the region of the Blood
river through the 1860s and 1870s. At the same time, the growing
Zulu population was suffering increasingly from pressure on
resources. The Zulu pushed up against Swazi lands, and on
several occasions Cetshwayo seemed on the verge of invading his
northern neighbour.30 Both the Zulu and the Swazi repeatedly
requested Natal intercession, but to no avail, although Shepstone
professed sympathy with their respective causes and blamed the
Boers for the growing tension in the area. Shortly after his
annexation of the Republic, however, Shepstone met a Zulu
delegation to discuss the territory disputed with the republic — and
in his new guise as administrator of the Transvaal backed the Boer
claims to the hilt. Anxious to avoid war, Cetshwayo now
approached the Natal governor, Sir Henry Bulwer, to appoint
a commission of enquiry into the disputed territory.
The commission reported back in the middle of 1878 and, to
Shepstone and Frere's dismay, fully supported the Zulu case.
29
SNA 1/7/6 Shepstone, memorandum, ) Match 1873, cited in Ethcrington,
'Anglo-Zulu relations', in Duminy and Ballard (eds.), The Anglo-Zulu war, jo.
10
B o n n e r , Kings, commoners and concessionaires, 150— 1.
39 1

From this point on, Frere began actively to foment war, despite
the caution of Sir Michael Hicks-Beach, who had replaced
Carnarvon at the Colonial Office. Yet the war was not simply
engineered single-handedly by Shepstone and Frere, with the
object of 'winning the loyalty of the Transvaal Boers and
converting the rest of white South Africa to a more favourable
view of confederation'.31 Before the annexation of the Transvaal,
Frere had needed Zulu co-operation to gain access to that
corridor; 'now he had only to uphold "the Transvaal claims in
order to secure the corridor free of charge . . . ' 32
At the end of 1878, Frere finally published the report of the
Boundary Commission — and demanded the surrender of a
number of Zulu accused of minor border incursions, the payment
of a large fine and the disbanding of the Zulu ' military system'
within thirty days. There was no way the Zulu king could comply
with this ultimatum; what was being demanded was the
dismantling of his state. Persuaded by Shepstone that once the
Zulu were given an opportunity to rise up against' the tyranny of
a bloody despot', the Zulu kingdom would disintegrate through
its internal divisions, Frere believed the Zulu could be conquered
in a single swift campaign; he was sadly misinformed both about
the nature of the Zulu kingdom and the strength of the British
troops at his disposal. Despite the undoubted tensions which had
resulted from population pressure and the development of trading
relations with Natal, ' the invasion, instead of fragmenting the
kingdom, united the Zulu people in support of Cetshwayo and
independence \ 3 3
On 11 January 1879, Lord Chelmsford entered Zululand with
a body of British troops. Eleven days later, they had been
disastrously defeated by the Zulu at Isandhlwana. Even with
reinforcements, Chelmsford was unable to retrieve the situation:
further defeats at Rorke's Drift, Eshowe and Hlobane were only
partly avenged by the British victory at Ulundi six months later.
Although Ulundi was elevated into a spectacular military victory,
after which Chelmsford could honourably resign the command
to Sir Garnet Wolseley, it signalled the exhaustion of both sides
11
C. Webb, 'The origins of the Anglo-Boer War', in Duminy and Ballard, Anglo-
Zulu War, 6.
" N. Etherington, 'Anglo-Zulu relations', 40-1.
13
P. Colenbrander, 'The Zulu political economy on the eve of the war', in Duminy
and Ballard, The Anglo-Zulu war, 78-97; J. J. Guy, Tie destruction of the Zulu kingdom
(London, 1979), 50.
39 2

rather than the destruction of the kingdom. More important in
ending the war was Wolseley's promise that the Zulu people
would be left in possession of their land and cattle, once the king
had been captured and they had laid down their arms.
Nevertheless, it took a further six weeks o f systematic terrorism ' 34
before the king was betrayed to the imperial troops. On the 31
August, Cetshwayo was marched to Port Durnford under heavy
guard, and despatched by sea to join Langalibalele in exile outside
Cape Town.
Wolseley's instructions were neither to annex Zulu territory
nor to extend imperial responsibilities. Having removed the king,
he placed Zululand under thirteen independent chiefs, many of
whom had been men of importance in the pre-war kingdom,
subject to the authority of a British resident. The settlement was
doomed to failure from the outset. By the end of 1881 violence
had erupted between supporters of the royal house, the Usuthu,
and the two appointed chiefs who were most closely tied into the
colonial economy, Zibhebhu and Uhamu. They energetically
forged links with Natal officials and traders, and seized Usuthu
cattle and other property in their desire to build up their private
wealth. Faced with renewed disorder in Zululand, in 1883 the
British government decided to restore Cetshwayo. He was
returned to a severely truncated kingdom, the result of the
machinations of Natal officials, who effectively' turned restoration
into partition, and who divided the country between the king, his
foremost rival Zibhebhu, and... the Colony of Natal'. 35
The true destruction of the Zulu kingdom followed the king's
return. Usuthu protests against the loss of territory went un-
heeded; their unsuccessful attacks on Zibhebhu simply invited
retaliation. The civil wars which ensued were even more devas-
tating than the British victory at Ulundi. In August, the refugee
king was captured, and he died early in 1884, leaving behind a
young and inexperienced successor, Dinuzulu. In an attempt to
recover their lost lands, the Usuthu now entered a disastrous
agreement with Transvaal Boers: in return for their assistance the
Boers claimed one-fifth of Zululand, including the fine cattle-
country north of the Mhlatuze river, and from the Transvaal to
M
Guy, Dtstruttion of the Zulu kingdom, 57-65.
" This paragraph is based on J. Guy,' The role of colonial officials in the destruction
of the Zulu kingdom', in Duminy and Ballard, The Anglo-Zulu war, 148-75; the
quotation is on p. 65.
393

the sea. By 1887, the prospect of fresh disorder in Zululand and
news that the Germans were taking an interest in the coast of
south-east Africa led to further imperial intervention. While the
Boers had to give up their ambitions to an outlet to the sea and
their assertion of sovereignty over the Zulu, their claims to the
area which became known as the New Republic were recognised.
It was soon absorbed by the South African Republic. British
sovereignty over the remainder of Zululand was formally recog-
nised in 1887, when Zululand became a Crown colony. Usuthu
protests against the partition of their country and the return of
Zibhebhu to the north ended with further British military action
and the surrender, trial and imprisonment of the Usuthu leaders,
including Dinuzulu. They were only to return when Zululand was
finally incorporated by Natal in 1897.36
Contrary to Frere and Shepstone's expectations, the conquest
of the Zulu did not in fact lead to peace in southern Africa. Sir
Garnet Wolseley followed up his campaign in Zululand with a
further assault on the Pedi who, perhaps encouraged by Boer
restiveness, continued to resist the Transvaal state's attempts to
extend its jurisdiction over their polity. This time the Pedi were
definitively crushed; Sekhukhune was captured, and imprisoned
in Pretoria. By 1883 one of his more compliant brothers had been
appointed regent; in exchange for supplying labour and military
levies to the local native commissioner, the Pedi were eventually
granted a location, a fraction of their former domain.
Paradoxically, now that their frontiers were secure and their
major enemies conquered, the imperial defeat of the Zulu and the
Pedi enabled the Boers to throw off British control. In 1881 and
1884, peace conventions at Pretoria and London returned internal
self-government to the South African Republic while stipulating
continued British ' suzerainty' over its relationships with foreign
powers, and bound the republic to respect the frontiers with its
African neighbours as established by the Keate Award. As in the
case of Zululand, this proved no bar to continued Boer
encroachments on African lands. The intervention of Boers in the
lands of their southern Tswana neighbours was even more
immediate. There, too, they found a ready-made field for inter-
vention, as competition sharpened between and within chiefdoms
35
The following section is based on Shillington, 'Land loss', 188-226.
394

over scarce resources which resulted from their proximity to the

diamond fields.
Through the 1870s, the Rolong and Tlhaping people in the
Molopo region of southern Bechuanaland had found a ready
market for their produce in Kimberley; unlike the situation in
Griqualand West, where the market undermined the authority of
the rulers, here the chiefs were in the forefront of commodity
production and retained a hold over their followers. Nevertheless,
by the mid-18 70s the Molopo region was the site of conflict over
irrigable agricultural land, and winter-grazing areas, in heavy
demand for the cultivation of winter wheat. The struggle was
intensified as 10,000 southern Tswana, ousted from the western
Transvaal and Orange Free State, arrived in the area. Further
south, too, there was similar friction between the Kora and
Mankurwane's Tlhaping over grazing and agricultural land.
Everywhere the pressure was aggravated by the incursions of
Afrikaners from the Transvaal, seeking the much-valued timber
and cattle resources of the region.
By the early 1880s, endemic friction in both regions had spilt
over into armed warfare. African chiefs were 'assisted' against
their enemies by white volunteers, who by 1883 had carved out
the mini-republics of Stellaland and Goshen. The exhaustion of
land and game resources in the Transvaal coupled with depression
and the amalgamations at Kimberley rendered many destitute and
desperate men only too happy to take advantage of the wars to
find farms and fortune, or take to cattle rustling. At least half of
the volunteers, however, and certainly the key figures in setting
up the new republics, were quite substantial Transvaal land-
owners, officials and traders, tacitly supported by the Transvaal,
eager to control the missionary 'road to the north'. War, with
its possibilities of loot, the trade in guns and ammunition, and
land and timber concessions, provided useful opportunities in
depressed times for speculation and accumulation. Landholdings
in Stellaland and Goshen rapidly changed hands as colonists from
the Cape and the Orange Free State appreciated the speculative
possibilities.
Despite the pleas of the Rev. John Mackenzie of the London
Mission Society (LMS), the Tswana chiefs and Cecil Rhodes (the
new member for Barkly East in the Cape parliament, who was
395

anxious to secure the labour and fuel resources of the interior for
Griqualand West37 and retain the missionary 'Road to the North'
for Cape commerce), the British were slow to intervene in
southern Bechuanaland; Basutoland and Zululand were more
pressing concerns in southern Africa, while the main attention of
the British cabinet was on Egypt and Ireland. Nevertheless, Cape
and LMS agitation ensured that the London Convention of 1884
limited Transvaal expansion to the east of the trade route. At the
same time, on the understanding that it would soon be incor-
porated by the Cape Colony, a British protectorate was declared
over the rest of the territory south of the Molopo river, and
Mackenzie was appointed deputy commissioner.
A 'humanitarian imperialist', Mackenzie believed that imperial
protection would lead to the transformation of African society,
through the undermining of chiefly powers and ' pagan' culture,
the development of a 'class of native yeoman'38 with individual
title to land, and the creation of a wage labour force for white
enterprise. Nevertheless, he opposed the abrasive effects of settler
colonialism; his policies ran directly counter to men like Rhodes
who were pressing for more restrictive labour legislation, and to
the land speculators who hoped to see the incorporation of the
region into the Cape. In little more than six months, he had been
forced to resign and was replaced by Rhodes as deputy com-
missioner.
Rhodes, with his eye on Afrikaner voters in the Cape, quickly
bought the Stellalanders off through the promise that their title
to Tswana land would be recognised by the Cape government.
In Goshen, however, he was less successful: a Transvaal take-over
of the republic in direct breach of the London Convention was
only prevented by the despatch of troops under Major Warren
by the British government, now fearful of German designs in
the area and a possible German—Boer link across northern
Bechuanaland.
There, Mackenzie, who accompanied the Warren expedition,
persuaded Chief Khama to offer land for British settlement in
exchange for British protection and indirect rule. His example was
37
See Shillington, 'Land loss', z 14-16, which argues that Rhodes's demands for
annexation have to be related to his interests in Griqualand West and immediate political
exigencies, rather than representing the first stage of his imperial ambitions to expand
further into the interior.
19
Dachs, 'Missionary imperialism', 6jo.
396

soon followed by the remaining northern Tswana chiefs. By

September 1885, despite strong Cape Afrikaner opposition and
still hoping for Cape annexation, the British reluctantly established
a Crown colony south of the Molopo river, while the region north
of the river up to latitude 220 south was declared a British
protectorate. Sidney Shippard, a friend of Rhodes, was appointed
the new administrator of British Bechuanaland and deputy
commissioner for the Bechuanaland Protectorate.
By requesting a protectorate, the northern Tswana chiefs avoided

the bloody and costly wars which had been the lot of most African
peoples to their south and east, and bought time. Elsewhere, in
much of the territory south of the Limpopo, by 1886, a subjugated
African workforce had been created out of formerly independent
African chiefdoms. Increasingly, young men were forced to seek
work on the mines and farms, railways and harbours of South
Africa, not at the discretion of their families and chiefs but
through necessity, at the behest of colonial authorities and
capitalist employers, and on the terms and conditions they offered.
It is no accident that on the Cape railways, the real cost of African
labour was more than halved between 1877 and 1883; o r that in
Kimberley the bargaining position of Africans was so transformed
that ten years after the Diggers' Revolt, the rigorous closed-
compound system could be established. Yet one should not
exaggerate the speed with which this result was achieved. There
were still considerable areas where Africans could evade the
labour recruiter, while the substantial kingdoms of the Swazi and
the Mpondo remained independent for another decade. The
northern Tswana chiefs, who had negotiated the terms of their
entry into the Protectorate, and Chief Khama in particular, who
was widely regarded by liberal opinion as the model Christian
chief, were able to retain a certain room for manoeuvre. This was
not simply the reward of collaboration: in Griqualand West,
Nicholas Waterboer had discovered too late the penalties of
peacefully offering his land to the British, while in Basutoland the
sons of Moshoeshoe were undoubtedly better off as a result of the
Gun War, though whether this made much difference for the
majority of their people is less clear.
Despite the extent of the upheavals, Africans were not uprooted
from their lands in vast numbers, although almost everywhere
397

land was lost, and people were forced onto the labour market.
The intervention of the British government, aware of the costly
resistance which would be provoked by large-scale expropriation,
meant that in Zululand, the Eastern Cape and the Bechuanaland
and Basutoland protectorates, most Africans retained their lands
(though these had already been severely truncated by the 1880s)
while elsewhere 'locations' on the model of the Eastern Cape and
Natal were set aside. They became the rural base for the migrant
labour system which was to be developed even more systematically
on the Witwatersrand in the next decade than it had been at
Kimberley.
For Africans all over southern Africa, the wars of the 1870s and
1880 caused a great crisis of commitment. They were not simply
fought between black and white; everywhere 'loyal levies'
assisted the colonial and imperial forces - most notably, but by
no means exclusively, the Mfengu in the eastern Cape and the
Swazi against the Pedi. In both cases, ancient enmities overrode
any necessity for a common front against the settlers. In Swaziland,
the 18 70s had been a time of increasing internal and external
uncertainty, with a real possibility of an alliance between the Zulu
and Pedi and other Swazi enemies in the eastern Transvaal. The
Transvaal-Pedi war and then that between the British and the
Zulu came as a great relief to the regents who governed the Swazi
kingdom during the minorities of both Ludvonga (who died in
suspicious circumstances in 1873) and Mbandzeni (who took full
office in 1881). They handled a threatening situation by appealing
to Natal against the belligerence of the Boers and the Zulu, and
to the Transvaal against the Zulu and the Pedi, offering their
services to the Republic and to Sir Garnet Wolseley during both
wars against the Pedi, and sitting on the fence until it was clear
which side was going to win in the Anglo-Zulu war.39
Despite increasing disenchantment with royal authority and the
growing pressures from prospectors, Boer graziers and mission-
aries, Swaziland survived the ' decade of crisis' with its territory
and sovereignty relatively intact, though it lost land in the 1880
boundary delimitation with the Transvaal. Ominously for the
future, however, the republic, deprived of St Lucia Bay by British
annexation, increasingly saw Swaziland as its route to the sea,
" Bonner, Kings, commoners and concessionaires, i j i, 176. Much of this section is based
on Bonner, chapters 8 and 9, and I am grateful to Cambridge University Press for
allowing me to see the page proofs of this work.
398

CHRISTIANITY AND CLASS FORMATION
while Mbandzeni's spate of concession-granting had already

begun. Although his councillors were aware of the danger,
Mbandzeni resisted their attempts to call in a Natal adviser or to
accept a British resident. He was only to recognise the severity
of the threat to his kingdom in 1887.
If the Swazi were able to buy time by opting for collaboration,

in most areas south of the Limpopo, chiefdoms were split on the
issue of resistance. There were many who had profited from the
new prosperity of the seventies and wished to safeguard their
new-found wealth, while the formation of a new class of Christian
peasant farmers, preachers and teachers, small shopkeepers,
artisans and minor civil servants added to the divisions. The great
military kingdoms of the Zulu, the Swazi and the Ndebele, it is
true, managed to keep the missionaries at arms length. Both
Cetshwayo and Lobengula, king of the Ndebele kingdom north
of the Limpopo, feared that Christianity and its new mores would
undermine their authority. Although missionaries were allowed
into all three kingdoms, their activities were strictly controlled;
they made virtually no converts before imperial conquest. Hostile
missionary propaganda was the prelude both to the Anglo-Zulu
war and the British South Africa Company's take-over of Rhodesia
(see next chapter). Given Cetshwayo's hostility to the amakholwa
(believers) it is perhaps not surprising that in the Zulu War, it
was the Christian converts of Edendale and Driefontein who
assisted Chelmsford in Isandhlwana. Already by this time their
lifestyle and values were remote from those of the majority of
Africans.
Yet not all Christians were collaborators. Amongst the Pedi,
for example, the king's half-brother, Johannes Dinkwanyane, and
the Christian community he led on the periphery of the kingdom
became 'increasingly prominent amongst those that rejected
the... demands made by officials and challenged the rights
claimed by landowners \ 4 0 They became increasingly independent,
too, of missionary tutelage. When, in 1876, war broke out
between the South African Republic and the Pedi, Dinkwanyane's
stronghold was the first to be stormed. Amongst the southern
40
P. Delius, The land belongs to us. The Pedi polity, the Boers and the British in the nineteenth
century Transvaal (Johannesburg, 198}).
399

Tswana and the Griqua, also, Christians were to be found

amongst both collaborators and resisters.
The very terms ' collaboration' and ' resistance' are inadequate
to encapsulate the variety and subtlety with which Africans
responded to their rapidly changing world. Violent conquest was
not the only way in which they were incorporated into colonial
society; and it was not necessarily the most successful. Whether
there was direct conquest or the more gradual penetration by
merchants and missionaries, however, the new order was 'but-
tressed' by ideological practices41 — essentially in school and
church and through the law - which shaped and moulded the
life-style and values of an increasing number of Africans. In the
Cape, educational growth in terms of schools and students was
sustained through the 1870s and the first half of the 1880s; the
educational institution which attracted Africans from beyond the
confines of the colony was undoubtedly the outstanding Free
Church of Scotland institution at Lovedale. Its impact was felt as
far afield as present-day Malawi.
In Natal, Lovedale's counterpart was the Amanzimtoti Institute,
which was expanded and separated from the theological seminary
in 1875, while the first African girls' school anywhere in South
Africa was established at Inanda in 1869, 'its object being to train
Christian wives and mothers'. 42 In 1885, there were sixty-four
African schools in Natal, with an average attendance of just under
5,000 pupils and a government grant-in-aid of about £2,500. From
the 18 70s, Wesleyan, Church of Scotland and American Board
missionaries began to ordain African ministers. Increasingly they
came to ' supplant European missionaries as the principal agents
of mission work among Africans', although even before this
happened for many Africans the first experience of Christianity
had come from their fellows who had encountered 'the Word'
on their journeys to the colonial labour markets.
The cleavage between the abantu basesikolweni (school people)
and the abantu ababomvu (' red' or' blanket' people) was particularly
sharp in the Cape, though the distinction between the two was
far from absolute. To Xhosa ' traditionalists', the converts were
traitors, amagqoboka, people who had 'opened a hole which had
41
B. Magubane, The political economy of race and class in South Africa ( N e w York,
•979). 57-
42
M. Horrell, African education: some origins, and development until if]) (Johannes-
burg, 1963), 17.
4OO

allowed their white enemies to gain entrance', but it was
paralleled elsewhere.43 In Zululand, for example, until after the
Zulu war, 'Christianity and Zulu citizenship were mutually
exclusive'.44 To the Victorian visitor African Christians seemed
to embody the ideal independent peasantry. As an Anglican
clergyman enthused of a community in Natal, 'There is not a
village in England corresponding to Springvale, where every man
lives under his vine and hisfigtree.'45 Whatever their image in
the eyes of, others, however, the ' School' people or amakholrva,
as they were known in Natal, saw themselves as the heirs to the
future, 'progressive and civilised'. Everywhere they were imbued
with the values of nineteenth-century laissez-faire liberalism. The
prosperous peasantry settled on the Protestant mission stations of
the Cape and Natal, as well as the petty bourgeoisie which derived
from it, constructed their world out of the mid-Victorian vision
of a 'progressive world order' based on the virtues of free
labour, secure property rights linked to a free market in land,
equality before the law and some notion of' no taxation without
representation'.
It was in the Cape Colony alone that a black elite had full access
to the franchise, and this inspired their peers in the rest of South
Africa well, into the twentieth century. 'Above all...a
strategy... for the political incorporation of part of the African
and Coloured population ',46 the non-racial franchise was bolstered
by Cape Town's liberal establishment - businessmen, politicians
and professionals. Perhaps most significantly and consistently, it
was backed at constituency level by the merchants, lawyers and
officials of the Eastern Cape, whose livelihood was intimately
linked to the prosperity and stability of a black peasantry. Thus,
in a substantial number of constituencies in the eastern Cape,
African voters were able to ensure the return of ' friends of the
native', individuals who could be relied upon to oppose attacks
on African rights and the more coercive government proposals.
In the western Cape, by contrast, the Coloured vote tended to be
41
R. Hunt Davis, 'School v s . blanket and settler: Elijah Makiwane and the
leadership o f the Cape school community', Afr. Affairs, 1979, 78, 310.
44
N. Etherington, Preatbert, peasants and politics in south-east Africa, iS)j-i&So
(London, 1978), 80.
45
Cited in Etherington, Preachers, 173.
46
S. T r a p i d o , ' T h e friends o f the natives: merchants, peasants and the political and
ideological structure o f liberalism in the Cape, 1 9 ) 4 - 1 9 1 0 ' , in Marks and Atmore,
Economy and society, * j o , o n which this section draws heavily.
4OI

in the gift of the local landlord or missionary, and in the towns
'Coloured men were recruited to the electoral machines by a
combination of patronage, corruption and intimidation.'47
The progressive incorporation of large African populations
into the colony in the 1870s and 1880s, the demand for a mass
of unskilled workers and the growth of Afrikaner political
organisations put increasing strains on the liberal tradition in the
Cape. Anti-imperial anti-black populism, as well as identification
with the struggles of the republics amongst poorer Afrikaner
farmers, was represented in the first Afrikaner political and
cultural organisation, founded by the Rev. S. J. du Toit in 1875,
Die Genootskap van Regte Afrikaners (the Fellowship of True
Afrikaners). This aimed at the inculcation of a self-conscious
Afrikaner identity based on Afrikaans (rather than Dutch), an
anti-British interpretation of South African history and the
advocacy of neo-Calvinist educational policies. This was replaced
in 1879 by du Toit's Afrikaner Bond, which achieved its greatest
popularity after the Transvaal threw off British annexation in
1881, spreading rapidly not only in the Cape Colony but in the
republics as well. By 1883, however, the Afrikaner Bond was
dominated by Jan Hofmeyr, an exponent of a broadly pro-imperial
Cape-based nationalism. His wealthier western Cape constituents
were rather more concerned with the protection of Cape farming
interests through parliamentary representation and continued
access to British troops and credit, than with du Toit's pan-
Afrikaner economic nationalism. Apart from moments of
heightened consciousness, as in 1881, Afrikaner nationalism and
pan-Afrikaner sentiment were never as powerful as imperial
officials imagined. For Hofmeyr's following, however, the non-
racial franchise and the independent African peasantry, especially
in view of its propensity to vote for liberal English — rather than
Dutch-speaking - candidates at elections were anathema.48 Thus,
through the 1870s and 1880s the colour-blind franchise was the
subject of bitter attack.
Such was the liberal hegemony, however, that the Afrikaner
Bond never called for total black disenfranchisement, whatever
reductions in black voting numbers it tried to introduce. Ironically,
Carnarvon's confederation schemes, allegedly designed in part to
« Ibid., 267.
48
T. R. H. Davenport, The Afrikaner bond. The history of a South African political
party (iS/o—if/i) (Cape Town, 1966), 50.
40 2

protect Africans against settler rapacity, left the question of the

black vote outside the Cape to the colonists. This outraged liberal
members of the Cape ministry, who thought Carnarvon's South
African (Permissive Federation) Act (1877) involved 'the un-
merited disenfranchisement of the whole of the coloured races,
kafirs, Hottentots, emancipated Negroes'. 49
Beyond the Cape Colony, the incorporating, controlling and
mediating roles of liberals were almost wholly absent. In Natal,
despite the fact that the overwhelming majority of settlers were
of British origin, material conditions had laid the basis for the
ideology and practice of segregation: the allocation of reserved
lands for Africans between 1846 and 1864; the recognition of
chiefly rule, and African marriage practices; the recognition, and
in 1875, the codification of 'native law'; and the attempt to
prevent permanent African urbanisation in the 1870s through the
implementation of the so-called Togt system, whereby African
workers were registered by the day. Under the complex franchise
laws, virtually no African qualified to vote, though theoretically
Christian Africans could be exempted from 'Native law'. In the
republics, matters were simpler: there was to be no equality
between black and white in church or state; Africans in areas
under burgher control were governed by Masters and Servant
laws based on the early nineteenth century Cape practice; the
members of the stronger independent African chiefdoms were
treated warily as the subjects of foreign powers, and there was
no provision made for those Africans who had been' detribalised',
either through conversion to Christianity or through their
oorlam*0 status.
Whatever the legal status of' non-tribal' Africans, in Natal, the

republics and beyond the frontiers of white control, social and
economic change and the work of missionaries had, as in the Cape,
led to class differentiation, the growth of an African peasantry and
the beginnings of an African Christian intelligentsia. While in the
49
BPP C1980. 1877. Further correspondence respecting the proposed confederation of the
Colonies and states of South Africa, Barkly to Carnarvon, 17.4.1877, cited in S. Trapido,
"The friends of the natives', 253.
90
The Oorlams were a group of Africans of mixed ethnic origin who had either
come to the Transvaal with the original trekkers, or had been captured as children and
'apprenticed' to Boer masters. They spoke Dutch, and were culturally assimilated to
trekker society, in which they constituted a class of overseers and skilled workers. For
a history of the Oorlams, see P. Delius and S. Trapido, ' Inboekselings and Oorlams:
the creation and transformation of a servile class', J. Southern Afr. Studies, 1981, 8, 2.
403

early days of mission endeavour, the first converts had been
largely among the oppressed and disaffected, by the 1870s the
missionaries had also been successful in their efforts to win chiefs
to Christianity, hoping thereby to gain the adherence of their
following. By far the most outstanding example was Chief Khama
III of the northern Tswana, the Ngwato people.
Among the northern Tswana, state and church had been
intimately connected since the mid-century. There,' both Sechele
I and Khama III, converted Christian kings, were, in their
espousal of capitalist values of accumulation and consumption,
independent African confirmation of the Protestant ethic'. By the
1870s, Shoshong, the Ngwato capital, had replaced the Kwena
capital as the 'emporium of the north', dominating the still
important trade in ivory, furs and ostrich feathers and increasingly
significant as a supplier of food and cattle for the Kimberley
market. In the late 1870s, it was estimated that ivory worth
£5 5,000 left Shoshong annually, while Khama's personal income
from cattle and ivory sales amounted to between £2,000 and
One response to the expanded opportunities of the interior was

widespread conversion to Christianity, which became identified
with ' progressive' reform. While Sechele of the Kwena trod an
uneasy path between total acceptance of the Christian ethic and
the maintenance of traditional mores, Khama III earned his
reputation as southern Africa's model black Christian by his
refusal to observe traditional practices. This had led to civil war
in 1865 and the bitter hostility of his father, whose office he
forcibly usurped in 1875. In 1872-3, when he briefly replaced his
unpopular Macheng as ruler, his attempts to modify traditional
practices such as royal rain-making and initiation ceremonies, led
to his downfall; on his return to power in 1875 he was more
careful to consolidate a constituency, through building up and
using his personal fortune for national purposes and by renouncing
his rights to royal cattle held on loan by Tswana headmen and
commoners, though the loan system still applied to non-Tswana
vassals. He carefully controlled white traders in his kingdom,
whom he segregated in a separate ward at the capital, and
attempted to conserve natural resources, such as ostrich chicks,
51
N. Parsons, 'Khama III, the Bamangwato and the British, with special reference
to 1895—1925", Ph.D. Edinburgh, 1973, 26.
404

SOUTH WEST AFRICA AND SOUTHERN RHODESIA
once the ivory frontier moved north. Having watched the
disintegration of the Tswana kingdoms further south, he matched
this with a careful policy of non-confrontation against whites and
an expansionist policy further north.
In 1876, Khama, still consolidating his hold over the Ngwato
kingdom, had looked to the British to prevent Boer and Ndebele
encroachments on his kingdoms. By the late 1870s he had
expanded Ngwato control over the Tawana in Ngamiland, and
over less powerful neighbouring peoples, and was pressing up
against the equally expansive states of the Ndebele and the Lozi
to his north. When, in 188 5, the British Protectorate was declared,
Khama III claimed a vast tract of territory up to the Gwaai—
Zambezi confluence, including much of what is now western
Zimbabwe. It was through Khama that Moremi II, the Tawana
king, was converted to Christianity and the Church established
in Ngamiland in the late 1870s; this became an important area for
Ngwato evangelical activity. In 1878, it was again Khama who
persuaded the Paris Evangelical Mission to send its mission to
Bulozi. Although his attempted reforms and close liaison with the
missionaries still roused formidable opposition which focused
around his father, Sekgoma, until the latter's death in 1884, and
around his brothers thereafter, Khama ruled the Ngwato for a
further forty-eight years. By the 1890s Ngwato economic and
strategic significance was bypassed by the building of the British
South Africa Company railway to Bulawayo. Nevertheless,
through the 1870s and 1880s, Khama III was the most powerful
of the northern Tswana chiefs. With his great personal wealth,
his commitment to Christianity, and above all his strategic
position astride the road to the north, he was seen by the
missionaries as holding the key to the interior, and by the British
as a bulwark against Boer expansion.52

By the mid-18 80s, Khama's position as a bulwark against Boer
expansion was of particular significance. For by then British and
Cape alarm at Boer movements westward had been increased by
the German annexation of South West Africa. In August 1884,
52
Ibid, z}-6. N. Parsons, 'The economic history of Khama's country in Botswana,
1844-1930', in R. Palmer and N. Parsons (eds.), The roots of rural poverty in central and
southern Africa (London, 1977), 116-22.
405

spurred by British imperial dilatoriness and Cape sub-imperial
bluster, Bismarck was persuaded to extend German control from
a tiny enclave around Angra Pequena, annexed on the urging of
a Bremen businessman, F. A. Liideritz, who had acquired exten-
sive concessions around the bay, to claim a vast territory from
the borders of the Cape in the south to the Portuguese colony of
Angola in the north. Notwithstanding the chagrin of the Cape
Colony at German intrusion into its hinterland, and the disquiet
of imperial officials, in 1886 an Anglo-German commission settled
the frontiers of the new colony, leaving the Cape only with the
guano islands off Angra Pequena. Conventionally and correctly
regarded as signalling a change in Bismarck's colonial policies, the
annexation accelerated the 'scramble' for southern Africa,
disturbing British hegemony in the region and rousing fears of
an alliance between Germany and the South African Republic.
Through the 1870s, German missionaries and merchants in
South West Africa had agitated intermittently and in vain for a
German take-over of the vast, semi-desert territory. Conflict on
the frontiers between the pastoralist Herero and Nama peoples
over sparse grazing lands and water had been intensified in the
nineteenth century by the advent of so-called Oorlam peoples
(people of Khoisan and mixed descent) from the Cape Colony,
and the proliferation of guns as a result of the activities of Cape
and European traders in search of ivory, cattle and ostrich
feathers. Fierce competition in the 1860s between the Nama and
Oorlam groups in the south was paralleled by seven years of bitter
warfare between the Herero under Chief Maherero and the
Oorlam Afrikaners under Jan Jonker, who had established his
supremacy over the Herero and a number of Nama chiefdoms. By
1870 the Herero had thrown off Oorlam domination, and sheer
exhaustion had led to an uneasy peace which left both sides
looking for outside allies.
Influenced by Cape traders, on more than one occasion
Maherero had appealed to the British high commissioner at Cape
Town for intervention on his behalf; in the mid-1870s, his appeals
were given additional urgency when three separate Boer treks left
the Transvaal in search of fresh lands to the west. Although the
first parties were forced to turn back, the danger of Boer
expansion into an area the colony considered to be its preserve,
together with rumours of its mineral wealth, led to wider Cape
406

Otjiwarongo
* • Okavara
H £
Otjimbingwe .Okahandja
Windhoek AMJAAL
HOTT.
%
Keetmanshoopo
Liideritj Bay
(Angra Paquena) BETHANIE O
HOTT. »
HERERO Bantu-tpHkins group
TOPNAAR Ntma paopls BONOELSWARTS?
S Hunttr-githtrvs
BO Bargdima
15 Peoples of South West Africa before the 1904-j uprising (based on

Wellington, South West Africa and its human issues)
interest in the region. In 1876, the Cape legislature despatched

W. C. Palgrave to the territory, ostensibly to explore its mineral
resources and the disposition of the local peoples towards colonial
rule. Although he persuaded Maherero to accept a protectorate,
however, Palgrave's recognition of the inflated claims of the
Herero to control the pastures around Windhoek, Rehoboth and
Gobabis, were contested by the Nama and especially by Jan
Jonker. The Nama remained loth to come under Cape or British
colonial ' protection', despite the arrival south of the Okavango
river of the so-called ' Thirstland' trekkers from the Transvaal by
407

1878. When in the following year Jan Jonker appealed for Cape
assistance against Herero encroachments on his land, it was too
late.
As so often, while Cape mercantile interests wanted to see the
region annexed, they were unwilling to foot the bill; and while
the British government would have been happy enough to see its
hegemony expanded through local sub-imperialism of the Cape,
they were also unprepared to take on the financial burden of an
uninviting slice of territory. By this time, the cluster of wars
elsewhere in South Africa made even Lord Carnarvon advise
caution. Thus although Walvis Bay was annexed by Britain in
March 1878, Palgrave's wider schemes of settling 400 Boers in a
new British colony in the region came to nothing. The outbreak
of serious war between the Herero and the Nama in 1880, in no
small measure the result of Palgrave's apparent pro-Herero bias,
led to a withdrawal of the resident.
The removal of the British resident, and even the nominal
protection which this had afforded the traders and missionaries,
led, on the outbreak of the four-year-long Nama—Herero war, to
greatly increased pressure on the German government to protect
its citizens in the territory, although in fact virtually no Europeans
were injured. Initially Maherero was successful in driving the
Oorlam Afrikaners from Windhoek, and by 1872 the Rehebothers
had sued for peace. By 1884, however, the remaining Nama had
united behind the remarkable leadership of Hendrik Witbooi to
drive the Herero north of Windhoek and Okahandja. Although
later on, Witbooi was to lead the Nama resistance to the
Germans, when the threat first appeared, the peoples of South
West Africa were bitterly divided. Despite the considerable
German missionary presence, there was no single leader with
sufficient knowledge of the new forces confronting the region
and capable of simultaneously manipulating them and holding his
own against local enemies. Over the decades that followed,
neither resistance nor collaboration was to suffice against what
turned into perhaps the harshest colonial regime in southern
Africa.
North of the Limpopo, Lobengula, king of the Ndebele was well

aware of the potential menace to his kingdom from the whiteman,
and trod a wary if very different path to that taken either by Khama
408

or the chiefs of South West Africa. Throughout this period and

beyond, he was torn between those who advised caution in the
face of the white advance, and those who wanted to use violence
in excluding whites from the kingdom. That the king should have
been suspicious of the intentions of white men from the south was
perhaps not entirely surprising, given the early history of conflict
between the Ndebele and the Transvaal and the particular
circumstances of Lobengula's accession to the throne in 1868. On
the death of Mzilikazi, the Nguni" founder of the kingdom, a
bitter succession dispute had raged between those who believed
that Mzilikazi's ' true' heir, Nkulumane, was still alive and should
be restored to the throne, and those who accepted that he had been
killed in the civil war in 1845—6, and that Lobengula, Mzilikazi's
own nominee, should be recognised. Widespread dissatisfaction
in the kingdom focussed on the succession dispute and erupted
in civil war. The prospectors and traders who had rushed to the
area on the discovery of the Tati goldfields attempted to push the
claims of the faction which promised them the most gain, while
Lobengula attempted to strengthen his hand by granting mining
concessions in the disputed Tati district in the south-west, and
in the lands of the Shona people over whom he claimed sovereignty
in the east.54
The grant of a second mission station to the LMS missionaries
at Hope Fountain in 1870 formed part of the same strategy,
although Lobengula's tolerance of a handful of missionaries in his
kingdom did not imply any encouragement of their evangelical
activity: neither the LMS nor the Jesuits, who sent eleven
missionaries to start work in Matabeleland in 1879, made a single
convert. Fearing the threat to his authority, Lobengula prevented
missionaries from working amongst the subordinated Shona
people and strongly opposed any of his subjects becoming
Christian.5S
In 1870, however, it was the intervention of Theophilus
Shepstone, then still secretary for Native Affairs in Natal, which
prolonged the civil war after Lobengula's victory at the battle of
" For the origins of the Ndebele and other mftcant offshoot kingdoms, see Cambridg)
history of Afrua, v, chapter 9.
54
R. Brown, The Ndebele succession crisis, IM-IIJJ (Salisbury, Central African
Historical Association, 1962).
" J. R. Cobbing, "The Ndebele under the Khumalos', Ph.D. Lancaster, 1976;
N. M. B. Bhebe, 'Christian missions in Matabeleland, 1859-1920', Ph.D. London,
1972.
409

30\°E
Approxareaof Ndebele settlement, c. 1890
- Approx. area of Ndebele raiding and tribute
collection, pre-1890
) Goldfields = = Hunter's road
• Route followed by Pioneer column
Battle-sites.1893 war
20°S
30 "C
16 Mashona and Matabeleland in the late nineteenth century (based on

G. Kay, Rhodesia: a human geography (London, 1970), p. 40, and T. Ranger,
Revolt in Southern Rhodesia (London, 1967), p. 45)
Zwangendaba. In Natal the discovery of gold at Tati had roused

the wildest hopes of vast mineral wealth in the interior, while
Transvaal attempts to claim the region roused fears that Natal's
rightful heritage would be lost. Shepstone, who was closely
associated with Natal prospectors and expansionists, used the
succession dispute in Matabeleland to resurrect one of his
employees, Kanda, into the missing Nkulumane. In 1871-2, he
went further and encouraged Kanda to invade the Ndebele
kingdom from Tati, in alliance with Macheng of the Ngwato. By
that time, however, the civil war had died down and whites in
410

Matabeleland had thrown their weight behind Lobengula. The

invasion was easily repelled, and this contributed to Khama's
victory over Macheng at Shoshong; nevertheless, Lobengula was
to be haunted by the possibilities of insurrection until Kanda's
death in the Transvaal in 1884. Indeed, in 1878, when Shepstone's
emissary, sent to report secretly on possible imperial expansion
across the Limpopo, threatened Lobengula with Nkulumane, he
and his companions were mysteriously murdered.56
Although the rebels had had a certain amount of support,
Lobengula emerged from the invasion with his position
strengthened; however, he never escaped the taint of being a
usurper, or achieved the power of his father. Throughout his
reign, he remained dependent on the faction which had supported
him in the crisis, including diviners of the Mwari cult which
deepened its influence among the Ndebele in these years. By the
1870s, royal herds had been depleted, and the strength of
surrounding peoples was increasing, as they came to arm them-
selves with the more freely available firearms; and although the
Ndebele also acquired guns - indeed the first batch of Ndebele
went to Kimberley in 1877 in response to the recruitment
campaigns of Alexander Baillie — they were never fully integrated
into Ndebele military strategy. As elsewhere, there seemed little
incentive for a successful army to change its methods of warfare.
Nevertheless, for most of the 1870s and 1880s, Ndebele troops
continued to dominate their neighbours and consolidated their
hold over the Shona to the east, despite the latter's mastery of
defensive warfare and their flow of guns, mercenaries and military
know-how from the related Venda people in the Transvaal. In
1879, a l ar ge Ndebele army attacked the stronghold of the Rozwi
ruler, Chivi Marorodze, accused of building a private army, and
was probably victorious, though it suffered heavy losses.57 A
number of Shona chiefdoms, including some Rozwi polities, paid
regular tribute to the Ndebele kingdom for the first time in
56
B r o w n , The Ndebele succession crisis, iz; T . J. C o u z e n s , 'Literature a n d
I d e o l o g y : the Patterson embassy t o L o b e n g u l a , 1878, and King Solomon's mines', Institute
o f C o m m o n w e a l t h Studies, collected seminar papers o n the societies o f southern Africa
in the nineteenth and twentieth centuries, v ( L o n d o n , 1974).
57
There is s o m e controversy o v e r this battle. A c c o r d i n g t o D . N . B e a c h , ' T h e rising
in south-western Mashonaland, 1 8 9 6 - 7 ' ( P h . D . L o n d o n , 1971), 14Z, the Shona w o n ;
according t o J. R. C o b b i n g , ' T h e N d e b e l e under the K h u m a l o s ' , 3 1 8 , the N d e b e l e w e r e
victorious. A s b o t h agree the Chivi w a s captured, taken to B u l a w a y o and skinned alive,
and that the R o z w i paid tribute t o the N d e b e l e , albeit irregularly, thereafter, it d o e s
s e e m as t h o u g h the N d e b e l e were the victors.
411

Lobengula's reign, and formed a definite political relationship
with it. Many Shona adopted Ndebele culture, language and way
of life, and were incorporated into its army, thus avoiding
retaliatory raids.
ACROSS THE ZAMBEZI
By the mid-i88os, however, Ndebele activities were gradually

restricted by colonial and imperial expansion northward. The
declaration of the Protectorate over the northern Tswana in 1885,
partly dictated by Khama's fears of Ndebele expansionism after
Ndebele cattle raids on the Tawana people of the Lake Ngami area
in the preceding couple of years, bolstered the Ngwato against
the Ndebele, while to the north, Ndebele expansion was checked
simultaneously. Here, the consolidation of Lozi power with the
accession of Lewanika and the reassertion of Portuguese ambitions
along the Zambezi ended any hopes the Ndebele may have had
of adopting their traditional response to danger and escaping the
closing noose from the south by migrating yet further north.
Like Lobengula himself, Sipopa who restored the Lozi
monarchy after the Kololo s8 interregnum in 1864, faced a divided
kingdom. After an attempted coup in 1869, he tried to eliminate
all centres of discontent and in so doing alienated the faction that
had put him in power. Losing control over the Lealui valley, in
1874 Sipopa moved his capital to Sesheke, where he hoped to find
support from his Toka and Subiya tributaries, and perhaps to
strengthen his hand through increased proximity to the trade and
gun frontier further south. By moving south, away from the places
of ritual significance on the flood-plain, and by relying on Mbunda
diviners, however, Sipopa further antagonised the' traditionalists'
in his kingdom. By 1876, his despotic and vicious rule provoked
an uprising in which he was shot.
The death of Sipopa did not end the struggles for power in
Bulozi. His successor, Mwanawina, was toppled after only two
years in office and was replaced by the young Lubosi, who was
in turn faced by a number of contenders with equally legitimate
claims to the throne. Lubosi re-established the capital at Lealui
and tried to reconcile the traditionalists. In doing so, however,
he lost the support of those factions that had supported his rivals,
" The Kololo were another Nguni offshoot - see Cambridge history of Africa, v,
ch. 9.
412

ACROSS THE ZAMBEZI
and the early years of his reign were marked by further attempted
coups and invasions, culminating in 1884 in his exile and
deposition. Yet this last rebellion brought to power a man so out
of touch with Lozi sentiment that many feared the collapse of the
Lozi state and domination by the southern Mbunda people whose
interests he seemed to represent. By the end of 1885, Lubosi had
regained the kingship after a resounding victory over his enemies.
From this time, he came to be widely known by his praise-name
(already in use before his restoration), 'Liwani ka la Matunga'
(one who gathers together) or, more commonly, Lewanika, ' the
conqueror'.59
By the time of his restoration, it was clear to Lewanika, too,
that he would have to come to terms with the forces from the
south. For the past decade he had watched the slowly moving
imperial frontier, and was well aware of the strategies of his
immediate neighbours, Lobengula and Khama. Indeed, despite
the bitter animosity between the Lozi and the Ndebele, for a time
in the early 1880s he toyed with joining Lobengula in a defensive
alliance against the whites. In the event, he decided to emulate
Khama by inviting in Francois Coillard of the Paris Evangelical
Mission as an intermediary with the white man and his
technology.
For people south of the Zambezi, and to some extent even for
the Lozi to the north-west, the crucial issue in these years was
increasingly what strategy to adopt towards the intruders from
the south. Although a trickle of white hunters, merchants and
missionaries also made their way into what is now Malawi and
eastern Zambia from the south and east, for most peoples along
the river and to its north, the upheavals they experienced resulted
from the activities of well-armed slave and ivory hunters from
the coast, exacerbated by the raiding and warfare of Ngoni
invaders from the south, who had established themselves in four
major military kingdoms in east-central Africa by the 1870s. Here,
there was an intensification and modification of older patterns of
raiding and trading, rather than any new direction. Although
some of the major chiefs, such as the Bemba ckitimukulu or the
>« This is based on M. Mainga, Buloî under the Layana kings (London, 1973);
G. L. Kaplan, The ilites of Barotseland ifyt-ifff: a political history of Zambia's Western
Province (London, 1970); G. Prins, The bidden hippopotamus: reappraisal in African history:
The early colonial experience in western Zambia (Cambridge, 1980).

mwase kasungu of the Chewa, were able to maintain and even

increase their power in these years, for the majority of the
agricultural peoples of the region, there was little question of
diplomatic manoeuvring for position in the face of the onslaught;
the choice was whether to join the predators and at least share
in the security of their stockaded settlements or take flight to the
swamps and hills.
The nexus of ivory, slave and gun-trading was not new in this
period. Throughout the nineteenth century, 'the increased
European demand for ivory piano keys and billiard balls changed
a continent \ 6 0 It was accompanied by an increase in the slave trade
both for hunters and soldiers and human porterage. At the same
time, the European demand for tropical crops was met through
the local use of slave labour. As the most sought-after commodity
in exchange was guns, those chiefs and adventurers who mono-
polised the traffic immeasurably increased their power over their
followers and their neighbours. Warfare became more lethal and
paradoxically, at the very time that European abolitionists
congratulated themselves on having virtually stamped out the
slave-trade, its destructive impact in east-central Africa may well
have increased.
Although we have no figures for the loss of life involved either
in the slave-raiding campaigns or in the wars of the Ngoni, the
reports of travellers and missionaries who followed in the wake
of Livingstone to trans-Zambezia talk almost uniformly of war
and devastation, death and destruction. While this was partly
propaganda intended to encourage support for mission activity
and engage British intervention, there can be little doubt that these
were years of immense dislocation and insecurity for the majority
of agricultural peoples outside the pools of protection established
by the warlords themselves.
By the 1870s, south-central Africa was criss-crossed by a
number of trade routes to the east and west coast. The reopening
of the Portuguese trading centre at Zumbo on the Zambezi-
Luangwa confluence had led to the expansion of the slave and
ivory trade along the Zambezi where there were no powerful
African rulers to block it. Although both the Plateau Tonga and
the Gwembe Tonga seem to have been able to use the trade to
60
L. Gann, The birth of a plural society under the British South Africa Company: Northern
Rhodesia 1894—1914 (Oxford, 1959)-
414

ACROSS THE ZAMBEZI
arm themselves against their more powerful Lozi and Ndebele

neighbours, they were in no position to defend themselves against
the ruthless slavers who were raiding as far as the Kalomo Gorge
by the mid-i87os.61 The Portuguese government exerted little
control over the new breed of traders, though many of them
operated in their name. Two of the most notorious, 'Kanyembe',
and his brother-in-law, 'Matakenya', were even recognised as
Portuguese officials and granted title to land. Their Chikunda
armies were recruited in the main from slaves of mixed origin
who had escaped from the Portuguese estates on the lower
Zambezi. Kanyembe made his base on the south bank of the
Zambezi and from there raided both along the river and to the
north of it, interfered in local politics and added local Africans
to the ranks of his armies. His men, who were notorious for their
brutality, were armed with flintlocks and soon controlled
communications on the Zambezi. Matakenya raided the Nsenga
and Bisa people along the Luangwa river, and amongst the Lala,
Lamba and Lenje peoples along the upper Luapula.62
In the upper Luapula and upper Luangwa valleys, Matakenya's
raiders met with Bemba and Swahili traders, replacing the Bisa
and Kazembe who had dominated the long-distance trade of the
region over the past century. The rising profitability of ivory may
account for the more permanent Swahili trading settlements and
the increased scale of commercial activity in what is now northern
Malawi from the late 186os. The traders between the southern tip
of Lake Tanganyika and Lake Mweru were linked to the entrepots
of Ujiji and Tabora, and from the east of northern Lake Malawi,
the route went from the Deep Bay crossing to Kilwa. Both areas
still had rich concentrations of elephant in the late nineteenth
century, while the Ulungu-Mweru region was an important
source of salt. Whereas the earlier traders had restricted themselves
to commercial activity, this new 'generation of Swahili traders,
led by Tippu Tip and his associates, brought about... a significant
alteration of the political role of foreigners \ 6 3 Tippu Tip's attack
on the Tabwa, in 1867, was the prelude to almost continuous civil
war after the death of the paramount chief in 1870, which was
* l ; T. I. Matthews, 'The historical traditions of the people of the Gwembe valley,
middle Zambezi' (Ph.D. London, 1976), 305-18.
*2 A. Roberts, A history of Zambia (London, 1976), ch. 8.
61
M. Wright and P. Lary, ' Swahili settlements in northern Zambia and Malawi',
Afr. Hist. Studies, 1971, 4, 3, 5J2; this section draws heavily on this article.

readily exploited by Swahili and Bemba outsiders. Swahili traders
even intervened in the internal affairs of the Kazembe kingdom
and challenged the hegemony of what had been the most powerful
African state in the region at the beginning of the nineteenth
century.
Further south and east, the Senga people seem to have been less
threatened by the Swahili settlements established in the 1870s.
Kota Kota, the settlement of the Jumbe, established at the expense
of the Marenga people, was by the 1870s a well-established
commercial and administrative entrepot recognised by the sultan
of Zanzibar. The Jumbe controlled traffic across Lake Malawi,
and may, like the other Swahili traders in Senga and Henga
country, have afforded the local populace some protection against
the Ngoni and the Bemba.
The Bemba were indeed amongst the few peoples of east-central
Africa to prosper as a result of the ivory and slave trade. Slightly
off the main trade routes, and with a tradition of raiding and
hunting rather than trading or agriculture, their interests com-
plemented rather than competed with those of the Swahili. Their
campaigns against Ngoni invaders in the mid-century meant that
they were well-organised to pursue the ever-diminishing herds of
elephant, and they had little use for the slaves they captured in
their raids. In general, the East African traders were content to
act in alliance with the chitimukulu and never actively intervened
in Bemba internal politics. As a result of long-distance trade, local
production and exchange was expanded and diversified. Bemba
iron weaponry improved, and after the mid-1880s the Bemba used
the new types of guns available to extend their hegemony over
surrounding peoples. In the 1870s and 1880s they were able to
take over much new territory in the south, north and west from
the Bisa, Mambwe, Lungu and Tabwa.
While the long-distance trade strengthened the Bemba against
their neighbours, it led neither to economic independence within
Bembaland, nor to an increase in the power of the chitimukulu.
Access by chiefs to beads, cloth, shells and firearms was a
significant way of attracting and rewarding followers, and tended
to strengthen subordinate Bemba rulers against their own people
and against the chitimukulu who never had a monopoly over the
trade. What fragile unity there was amongst the Bemba' was based
primarily on links between a number of chiefs, each with his own
416

ACROSS THE ZAMBEZI
specific economic interests; and no one of them was clearly

predominant'.64 Between i860 and 1880, the Bemba polity had
probably doubled in extent; yet it consisted of about twenty
chiefdoms which were only loosely connected, mainly through
matrilineal ties. They all recognised the ritual supremacy of the
cbitimukulu, but had no centralised political military or economic
institutions. The chiefs were linked rather by a flexible kinship
and affinal network, and their power 'from the cbitimukulu
downwards, was largely determined by the prevailing constella-
tion of alliances and antagonisms within an expanding and
unstable system'.65
To the west and south of Lake Malawi, the agricultural Chewa
and Manganja peoples, living in dispersed matrilineal villages,
were less fortunately placed. While beyond the main onslaught
of the Bemba and Swahili slave raiders, they were at the mercy
of two other groups of intruders, the Ngoni and the Yao. The
Yao, who had moved into the area that is now southern Malawi
in several waves from the mid-nineteenth century from their base
in Mozambique, had long had contact with the political economy
of the coast. By the 1870s, powerful dynasties had emerged and
had established military and commercial chieftaincies, with a
distinctive culture modelled on that of the coastal elite. Thus,
Mponda and Makanjira's settlements near the lake 'were sub-
stantial townships containing up to 8,000 inhabitants'. 'Pioneers
of the Arab trading frontier' like the Swahili in the north, they
'introduced a new factor into political relations south of the
lake'. 66
The Ngoni, like the Ndebele, originated in the south, in the
wars which accompanied the rise of the Zulu kingdom in the
second and third decade of the nineteenth century. During many
years of migration and warfare they had absorbed a vast number
of ethnically diverse peoples, who had come to identify culturally
and politically with their conquerors. At the same time, the Ngoni
themselves, while retaining their northern Nguni military and
political organisation, had assimilated many of the social norms
of surrounding peoples. By the 1870s, four main groups had come
to settle in east-central Africa: the Maseko Ngoni, at the south
64
A . Roberts, A history of the Bemba ( L o n d o n , 197}), 213.
6
» Ibid., 171.
66
J. L. McCracken, Politics and Christianity in Malawi, tSyj-1940. The impact of the
LJvingstonia mission in the northern Province (Cambridge, 1977), 6.
417

end of Lake Malawi in the modern Dedza district; those of Chief
Mpezeni near Chipata (Fort Jameson) on the eastern Zambian
plateau; in Chewa country, those of Chiwere Ndhlovu, himself
an Nsenga by birth and a former captive; and to the north, the
Ngoni of Chief Mbelwa. While the Ngoni were not primarily
slave-traders, their political economy depended on raiding: to
establish themselves on the land, accumulate cattle and incorporate
new soldiers into their armies.
While contemporaries stressed only the destructiveness of the
Ngoni, more recently historians have been concerned to show the
constructive state-building aspect of Ngoni activity. Within their
realms, order and a certain security prevailed. Yet their impact,
together with that of the slave-traders, whether Bemba or Yao,
Swahili or Chikunda, undoubtedly added to the chaos and
confusion of east-central Africa in these years. It is true that
outsiders brought the region new principles of political organisa-
tion, and new ideas, crops, and technologies. Yet these were more
than outweighed by the destructive impact. The prosperity of
the outsiders was largely bought at the expense of the local
populace, who lost their autonomy, their land and their people.
Even people like the Yao or the Bemba gained little in the
long-term from their unequal relationship with the world
economy. In exchange for luxuries and quickly consumed com-
modities, they traded their people, exterminated their game, and
lost their independence.
Even the slaves who were used internally did little to transform
or even expand the local forces of production. The internal use
of slavery distorted and dislocated local social relations as
' brothers' and ' wives' were sold to traders for the much-coveted
beads and cloth, guns and powder. For a matrilineal people like
the Yao, slave-marriage had very real advantages for the men, who
could establish virilocal lineages, and gain total control over the
labour power and offspring of women who had no protective
kinship ties. For women there were few benefits. As the premium
on slaves rose, even the limited security of a slave marriage was
lost as women acquired exchange value.
The incorporation of captives into the Ngoni regiments also
deprived local societies of their necessary male labour. Whether
the major famine of the 1870s can be traced to this withdrawal
of local labour is not clear. What is clear is that, by the 1870s,
418

ACROSS THE ZAMBEZI
the political geography of large sections of the region had been
transformed as slaves were concentrated in the commercial
entrepots, young men were taken into the Ngoni armies and those
who escaped fled to the hills or huddled for safety in stockaded
villages.67
It was in this complex, fluid and insecure world that the first
Free Church of Scotland missionaries founded the Livingstonia
mission at Cape Maclear at the south end of Lake Malawi in 1875,
to be followed a year later by a Church of Scotland mission at
Blantyre in the Shire highlands. Inspired by Livingstone's death,
they hoped to establish 'practical' Christianity and 'legitimate
commerce' in the heart of Africa. Both parties were accompanied
by lay artisans, responsible for the practicalities of establishing the
missions and bringing industry to the African. In addition, in
1878, the small group of Glasgow businessmen who supported
the Livingstonia mission also founded the Livingstonia Central
Africa Company (renamed the African Lakes Company in 1881)
to establish cash-crop production and control European trade
in central Africa. The company had little hope, however, of
undercutting the Yao and Arab east-coast trade, and was soon
itself dependent on the export of ivory, most of it from powerful
slave-owning traders like the Arabs at Karonga or the Jumbe of
Kota Kota, for whom it provided a convenient new outlet and
whose slaving activities it thus inadvertently advanced.
Initially the missionaries had little success in introducing
Christianity to the more powerful Yao chiefdoms. The close
connection between slavery and the coastal trade on which chiefly
power depended meant that, as in other parts of Africa, the earliest
mission adherents were largely the disaffected and the
dispossesed - freed slaves returning home, refugees from the
slave-raiding and warfare, the children of the Kololo rulers, who
had accompanied Livingstone into southern Malawi. Within a
year, the presence on the settlement at Livingstonia of these
dissidents had led to troubles with neighbouring chiefs as well as
acute problems of internal authority.
At Blantyre and at Livingstonia's outstations around Bandawe
(where Livingstonia was moved in 1881) these problems were
67
M. Vaughan, 'Social and economic change in southern Malawi: a study of rural
communities in the Shire Highlands and upper Shire valley from the mid-19th century
to 1915', Ph.D. London, 1981.
419

compounded by the arbitrary actions of the artisans suddenly
granted uncontrolled responsibilities. Rival headmen and villagers
actively encouraged the mission's intervention in village affairs in
the hope of attracting the assistance of external authority, and in
the early days of both missions the lay artisans in command
flogged and fined Africans on the stations and in the villages
outside it who infringed their code of conduct. When in 1879 news
reached Britain of violent miscarriages of justice at Blantyre, the
entire industrial enterprise was thrown into question. Both
missions were ordered to refrain from civil jurisdiction, though
the change in policy was more successfully implemented at
Bandawe than at Blantyre, which retained its character as the
'nucleus of a state'. 68
Despite its early difficulties, over the next decade, Livingstonia
pioneered a new form of mission, directly engaging with village
life, and seeding dozens of new outposts over much of northern
and central Malawi. At Bandawe the missionaries were welcomed
by local headmen who sought allies against Ngoni raiders and
local rivals. Depleted by years of Ngoni warfare, their economy
undermined and their polities in disarray, the lakeside Tonga
rapidly appreciated the new opportunities presented by the
mission. The capitalist ethic it propounded interacted with the
emphasis on individual achievement in their society, itself perhaps
a response to a decade of profound social disruption. Like the
Mfengu in the eastern Cape, the Tonga responded with alacrity
both the wage labour and the education the missions provided;
by 1894 there was a regular attendance of over 1,000 pupils at
eighteen Free Church schools, while in the same year, 1,400 Tonga
were working for the African Lakes Company, and 4,000 for
settlers in the Shire highlands.69
While the Tonga constituted the most spectacular missionary
success, and the Yao — who turned increasingly to Islam in these
years — their most dismal failure, the missionaries also made some
headway amongst Mbelwa's northern Ngoni, in part the result of
internal tensions amongst his people. They do not, however, seem
to have been able to exploit similar tensions amongst either the
Ngoni to the south of the Lake or the Ngonde to its north. As
elsewhere, the relationship of Africans to the mission initiative —
68
McCracken, Politics and Christianity, 70, on which this account is heavily
dependent. *» Ibid., 80-1.
420

ACROSS THE ZAMBEZI
itself the product of particular historical circumstance — was to

profoundly shape not only their interaction to colonialism, to
which it was the prelude, but also and more significantly the
geography of opportunity and class, of politics and education,
well into the twentieth century.
421

CHAPTER 8
SOUTHERN AND CENTRAL AFRICA,

1886-1910
If diamonds had begun the transformation of southern Africa, the

industrialisation which followed the discovery of vast seams of
underground gold on the Witwatersrand in 1886, followed by the
renewed assertion of British supremacy in the interior of southern
Africa, greatly accelerated the forces making for change over the
entire region, and set the pace for much of the twentieth century.
In the 18 80s the sub-continent was still composed of a cluster of
independent African kingdoms and Afrikaner republics, British
colonies and protectorates; the huge new German acquisition of
South West Africa was still largely unconquered. By 1910, with
the political unification of the South African colonies, British
ambitions of creating a southern African confederation seemed
well on the way to fulfilment, while, to the north, British imperial
frontiers stopped short only at Katanga and Tanganyika. All over
southern African the annexation of African polities meant the
establishment of colonial states, with government departments
and courts, alien soldiers and policemen. By 1910 railroad arteries,
often built at enormous human cost, connected the coast with
mining centres as far afield as Bwana Mkubwa and Elisabethville
(Lubumbashi), opening up new markets and releasing new
sources of labour. Boundaries had been drawn which were to last
beyond the colonial period, and it was accepted by the colonial
rulers that the Zambezi was to be the boundary between the' white
south' and the ' tropical dependencies' of east and central Africa,
although British Central Africa1 uneasily straddled the divide.
Everywhere, the colonial administrator and tax-collector, but
also the trader, prospector and labour recruiter, gave colonial rule
its main practical effect. The scope of mission work, already
1
This was the name applied from 1889 till 1897 to the British sphere of influence
north of the Zambezi, including both that part ruled by the British South Africa
Company ( from 1897 Northern Rhodesia) and that part ruled directly by the British
government through the Foreign Office (from 1897 Nyasaland). The latter was
distinguished from the former as the British Central Africa Protectorate.
422

SOUTHERN AND CENTRAL AFRICA, 1886— I9IO
entrenched to the south of the Limpopo and in the Shire
highlands, was vastly extended, as numerous new societies
appeared on the scene, and introduced schools and, on a very
limited scale, clinics. Despite settler criticism, these societies
played a key role as mediators between rulers and subjects,
encouraging individualism, wage labour and commodity produc-
tion, and fostering the growth of a class of educated Christian
Africans, who were to become the most effective critics of
colonialism. Ineluctably, the traditional societies of the sub-
continent lost their remaining autonomy and were meshed, albeit
still incompletely, into a money economy. In the new dispensation,
whites assisted by the colonial states controlled private property
and the means of production - although not all whites equally;
and blacks were increasingly seen solely as the source of labour
power. Even in South Africa, however, the extent of the
independence of settler colonialism should not be exaggerated:
the Union formed in 1910 was still a poor and divided one,
dominated by international mining capital, with white settlers its
most convenient if somewhat uncomfortable allies.
This is not to argue that the history of the region can be
portrayed homogeneously as a single history. The diversity of the
African societies of the sub-continent, their differing nineteenth-
century experience and the historical interplay of personality and
chance led to complex and distinctive developments within the
region. It is characteristic of capitalism that it should advance
unevenly. In southern Africa this unevenness was not merely
regional and sectoral; it was a function of the diversity of
capitalism itself. Capitalism occurred in many forms - from the
highly industrial complex of the Witwatersrand, to the essentially
rentier and mercantile activities of the land concession companies;
even in 1910, side by side with the most advanced technology in
the world, there were white farms still based on non-capitalist
forms of production, and protectorates and reserves in which the
homestead, communal tenure and the authority of chiefs remained
superficially intact, though increasingly dependent on the colonial
economy for their continued existence.
As we have already seen in chapter 7, the ' scramble' for Africa
had its specifically southern African aspect, which was already
under way in 1886, and which was further spurred on by German
intervention in a region Britain regarded as her sphere of influence
423

SOUTHERN AND CENTRAL AFRICA, 1886-I9IO
and by its known abundance of mineral resources. After 1886 the

African territories south of the Limpopo required only a' mopping
up' operation as they were absorbed into the neighbouring
colonial states; north of the river, however, in 1886 the ' scramble'
had barely begun. From 1887, Cecil Rhodes, who had already
played a major role in the extension of British domination over
Bechuanaland, and who was in the throes of amalgamating the
Kimberley diamond mines, turned his attention to the north.
Within less than five years a vast expanse of territory including
present-day Zimbabwe and Zambia had come under the control
of his chartered British South Africa Company.
By far the most important factor in accelerating the pace of
change was the discovery of gold in the Transvaal. While the
diamond discoveries had begun a series of social and economic
changes, the discovery of gold, and in particular in 1886 the deep
seams of gold along the ridge of mountains between Heidelberg
and Potchefstroom, known as the Witwatersrand, intensified and
carried this revolution far beyond the Limpopo. Within a decade
of the official proclamation of the fields, the economy of southern
Africa had been transformed and its political direction sharply
changed. By the turn of the century, the region had become the
world's largest producer of diamonds and, more significantly,
gold. From a pastoral backwater,' an unprosperous State that had
never known genuine solvency ',2 the South African Republic had
become essential not only to the political economy of southern
Africa, but also to the world money market. Its new wealth
transformed the local balance of power, challenging Cape domi-
nance and ultimately threatening British hegemony in the sub-
continent. The coastal colonies engaged in cut-throat competition
to control the lucrative trade to the Rand, while men and money
flowed to the gold-fields. The ramifications were felt over an
ever-widening area, as capital expanded from this storm-centre in
an accelerating drive to find and exploit additional mineral
resources, labour supplies and land.
It is relatively simple to delineate the demographic changes
wrought by the mineral discoveries, and to cite statistics of gold
production, or revenue and expenditure accounts. These are but
a clue to the impact of capitalist development on the relations of
production all over southern Africa, and to the traumatic and
2
C. VC. de K i e w i e t , A history of South Africa, social and economic ( O x f o r d , 1937), 123.
424

frequently ruthless transformation of societies, both black and
white. Some of the implications of industrialisation for the
relationship of race and class in southern Africa have already been
delineated in Kimberley. The sheer scale of the demand for labour
on the gold-fields, the controls demanded by an acutely cost-
conscious industry over its massive unskilled labour force, and
thus the assistance it expected from the state, set the experience
of the Rand apart. The gold-mining industry was at the heart of
the structure and evolution of modern South Africa - ' it was there
that... the first and most extensive industrial institutionalisation
of racial discrimination in South Africa' occurred.3
For whole societies, as for individuals, both black and white,
the period was one of tremendous upheaval. The conjuncture of
the 'scramble for resources', the imposition of colonial or
company rule, and an industrial revolution based on mineral
extraction, meant that within 'a matter of decades, and sometimes
perhaps within the space of even a single generation, Africans
could successively be pastoralists, peasants, proletarians or pris-
oners' in a traumatic 'downward socio-economic spiral'.4
All over the sub-continent, moreover, this decline was accel-
erated by the disasters which struck at rural society in the mid
1890s: pre-eminently the rinderpest epidemic which devastated
African cattle-holdings in 1896-7, preceded in many cases by
smallpox, and followed even more disastrously by locusts, drought
and further human disease. In what is now Zambia, it has been
remarked,' the 1890s were perhaps most widely remembered not
so much for wars and high politics, as for a series of natural
disasters'.5 And the same was true for much of the rest of the
region. But whereas in the past, African societies had been forced
to recover from natural disaster through falling back on their own
resources, the disasters of the 1890s further locked them into
dependence on the burgeoning colonial economy.
Contrary to much conventional literature, this dependence was
not simply the result of 'natural' processes; it was actively
engendered by violence and deception, even in areas where
imperial expansion took place without initial warfare. Nor can the
1
F. A. Johnstone, Class, race and gold: a study of class relations and racial discrimination
in South Africa (London, 1976), 3.
4
C. van Onselen, Studies in the social and economic history of the Witwatersrand,
1U6-1914, vol. 11, The new Nineveh (Johannesburg and London, 1982), 172.
1
A. D. Roberts, A history of Zambia (London, 1976), 171.
42 5

lines be simply drawn between colonial or company rule, or the
contrasting modes of British, Boer or German domination. The
discourse of Social Darwinism and naked racial domination was
universal and peculiarly well-suited to an era of land and labour
expropriation. Everywhere the priority of administrations was
for revenue and labour, and imperial and company rule were
followed by the imposition of taxation. Attempts to raise tax were
frequently accompanied by hut-burning, flogging and the confis-
cation of crops and cattle for non-compliance. Although in many
areas, state and administration were weak, and fearful of rousing
widespread resistance, this did not prevent much localised violence
in the subjection of African societies, and in the exploitation of
labour on mines and farms. Indeed, the violence was frequently
greatest where administrations were weakest: in longer-settled
areas, legislative intervention and the earlier loss of land and cattle
were in themselves capable of pushing out the much-needed
labour supply.
In some areas, colonial expansion had already created an
available pool of male wage-labour at least at certain times of the
year; with the advent of foreign rule many of the previous
occupations of men — in government, as well as in hunting and
warfare - came to an end. However, so long as Africans had access
to the land, they weighed the' opportunity cost' of different forms
of labour, and had some bargaining power. In most of the
pre-industrial African societies of southern Africa, women did
much of the arable agriculture, and young men could be spared
from the homestead seasonally without disturbing production, so
long as the process was selective. In general, it was the young men
who were the first migrants, very often sent out by the homestead
itself, which in turn tried to control their return and their wages.
For the young men, a spell of labour migration could give them
independent access to the marriage payment called lobola. This was
a process which, as we have seen, had begun earlier,6 but which
undoubtedly accelerated in this period, and had profound rami-
fications for the control by the homestead head over young men
and women, and for colonial administrators over their taxpayers.
This in part explains the increasingly anxious cries of the chiefs
and headmen to commissions of enquiry concerning their lack of
6
For further discussion see S. Marks and R. Rathbone (eds.), Industrialisation and
social change in South Africa: African class formation, culture and consciousness Qjon&on, 1982),
introduction and essays by Guy, Kimble, Harries and Shillington.
426

SOUTHERN AND CENTRAL AFRICA, 1886-19IO
control over their juniors, and the alliances they struck with
colonial administrations and recruiting agencies to secure the
return of migrants.
Yet it is difficult to assess the impact of migrant labour for the
region as a whole. In south-central Africa, the main agricultural
areas probably did not send migrants, and it was from the lands
already devastated by slaving and raiding that theflowwas
greatest. In the early stages, earnings were often used to expand
agricultural production, and single spells of absence probably left
homestead production unscathed. Nevertheless by the end of the
first decade of the century, missionaries in British Central Africa
were beginning to detect signs of strain as a result of the absence
from home of the large numbers of migrants demanded by new
colonial enterprise, and especially transport, while the absence of
young men responsible for burning the bush and cutting down
trees was adding seriously to the ecological dislocation which
resulted from colonial game and settlement policies.7 South of the
Limpopo too, repeated migrancy was becoming the norm, and
numbers of young men, and increasingly even young women,
were beginning to disappear from the countryside to settle
permanently in town, with major repercussions for family and
homestead structure. The division of labour in the countryside
began to change, and increasingly the burden of agriculture fell
on women and children. Once trapped in the migrant-labour
system, it became more and more difficult to escape.
For the migrants, conditions in the towns and mines were often
appalling, with high mortality and frequent violence. Africans
rapidly developed strategies of survival in the towns, and forms
of worker consciousness evolved which included heightened
ethnic identification for the purposes of protection and mutual
assistance, new organisational forms, concerted desertion, and, on
occasions, strike action. Their rural roots did not prevent migrants
from displaying an acute understanding of the labour market and
their position within it; nor did it prevent their taking militant
action to demand wage increases or improved conditions. It
would however be misleading to think, in this period, of a
fully-fledged and self-conscious working class.8
7
L. Vail, 'The political economy of colonialism in northern Zambezia, 1870-1975'
in D. Birmingham and P. M. Martin (eds.), History of Central Africa (London, 198}).
8
Marks and Rathbone, Industrialisation and social change, 26-7; C. van Onselen,
' Worker consciouness in black miners: Southern Rhodesia, 1900-1930',/. Afr. Hist.,
•97}. «4. 2 . 137-56-
427

2O°S
3O°S
17 Labour migration routes in the early twentieth century
For many white Afrikaners the changes were no less profound.

In 1886 the South African Republic was still a pre-industrial
agrarian state ruled by cattle-owning notables; by 1910 it was at
the heart of a rapidly industrialising subcontinent dominated by
mining capital. The dramatic imposition of capitalist relations of
production, the ravages of war and imperial social engineering
had transformed Afrikaner society as painfully and perhaps more
completely than its African counterparts.
Yet, as in most industrial revolutions, the costs were unevenly
distributed. And what brought suffering and deprivation to the
many, also provided new opportunities for the few. The political
order re-established after the war in South Africa in 1902 was
predicated ultimately on an alliance between the mine magnates
and the more prosperous Afrikaner landholders, and this alliance
emerged as the most important political force by 1910. Even
428

before the war, the cleavage between the landless Afrikaners and
the notables in control of the state was in evidence; after the war,
with the increased capitalisation of white agriculture the processes
of class formation were accelerated, and British officials remarked
on the hostility between the old landowners and their increasingly
displaced white clients.
In other areas of white settlement, both company and colonial
power reached an accommodation with the forces of local
capitalist development, whether in South West Africa, Rhodesia
or in the Shire highlands. Elsewhere, different alliances were
struck with local ruling elites, in situations where African political
structures were both partially dissolved and then conserved. In
the reserves and protectorates of southern Africa, Africans were
still governed through chiefs and hereditary headmen, but their
roles were subtly transformed as they became appointees of the
colonial authorities. Again, the process varied from area to area.
Whereas, in the nineteenth century, the Cape and Natal states had
attempted to destroy the overarching powers of Africans kings
or paramounts, the utility of intermediate chiefs in providing a
cheap administrative infrastructure was more easily recognised.
Whether or not an accommodation was reached with the more
powerful African rulers depended in large measure on the number
of whites present in a given area: in British Central Africa, where
the number of white administrators and soldiers available to
Crown or Company was minute, there was no alternative to
finding African allies. In areas with a large settler population,
anxious to expropriate African land and labour rapidly, the
African king was seen to hold a pivotal position in maintaining
the military and social cohesion of the polity - a cohesion which
had to be destroyed; in more marginal areas of settlement,
aristocracies and administrators were able to come to a mutual
accommodation. Thus in this period the Ndebele king had to be
destroyed, as the Zulu king had been earlier. In South West Africa
too, the Germans ruthlessly attempted to destroy the African
social fabric and political hierarchies in the first decade of the
century. On the other hand, in Barotseland, Basutoland, Swaziland
and the Bechuanaland Protectorate, the paramounts played a key
role in the peaceful incorporation of their territories within the
imperial framework, and in the mobilisation of labour for the
colonial economy. The exception was British Central Africa,
429

where, despite or perhaps because of, his military weakness, Harry
Johnston adopted a policy of' weakening, dividing and wearing
down every political system that possessed any capacity to menace
or resist'.9
All over southern Africa, demands by the colonial state for tax
were met, where possible, by increased agricultural production.
The rise of new markets meant that the prosperous peasantry,
already in evidence in the Cape and Natal by the mid-nineteenth
century, became more widespread. Yet new towns and their
markets also opened up possibilities for those with the resources
and skills to exploit them — whether black or white. For
individuals able, whether through their prior historical experience
and education or by good fortune, to take advantage of the new
opportunities, the processes of proletarianisation could be warded
off, if only for a time.
For the up-and-coming black bourgeoisie, many of them the
product of mission education, and imbued with the values of a
new individualism, Kimberley and later the Rand, as well as the
smaller centres of central Africa, provided an arena for the
development and deployment of their entrepreneurial and clerical
skills. If their hopes of inclusion in a wider, more tolerant society
- in which class and achievement, rather than race, would be the
criteria for acceptability - were to be dashed by the provisions
of the Act of Union in South Africa and colonial racism elsewhere,
there was still in their manifest mobility and 'progress', some
room for optimism.10
Yet for the middle-class Coloureds and Indians, the period was
probably one of regression rather than advance. Strident demands
for urban segregation were voiced by settlers all over the
subcontinent in these decades, and increasingly Coloureds and
Indians found themselves classified as 'uncivilised natives'. The
older Cape tradition of social control through class co-option was
overtaken by the high tide of European racist ideology. Even
wealthy Indian merchants found themselves insulted, spat upon
and assaulted in the streets of Durban and Johannesburg; barriers
were imposed on Indian immigration into all the South African
colonies, and the grant of responsible government to Natal was
* E. T. Stokes,' Malawi political systems, 1891-1896'inE. T. Stokes and R. Brown,
The Zambesian past. Studies in Central African history (Manchester, 1965), j6o.
10
For the optimism of the petty bourgeoisie in this period and its characteristics,
see B. Willan, 'The life and times of Sol T. Plaatje', Ph.D. London, 1980 (London,
forthcoming).
430

THE DISCOVERY OF THE W ITW ATERSR AN D
immediately followed by a spate of anti-Indian discriminatory

legislation. The period which witnessed the growth of a class of
educated blacks also witnessed increasing settler antagonism to
the whole notion of ' civilised natives', who might have ' ideas
above their station'.
The frustrations and freedoms of the new town were perhaps
nowhere better evidenced than in the growth of the independent
churches. Although their origins lay in the countryside, with the
activities of a small group of mission-educated preachers, it was
on the Rand in particular in the 1890s that the independent
churches spread like wildfire, especially after Mokone formed his
Ethiopian Church in 1892, and established links with the African
Methodist Episcopal Church (AMEC) in the United States of
America. The whirlwind visits to South Africa of, on the one hand
that turbulent 'missionary hitchhiker', Joseph Booth, and on the
other, of the fiery Bishop Turner of the AMEC with his ' back
to Africa' schemes, added to the ferment of what became known
as 'Ethiopianism'. Fuelled by the rawness of colonial conquest,
the resentment caused by settler racism and the struggle over land
and labour, the proliferating African independent sects came in
many instances to be characterised by a millenarian vision which
roused the anxiety of settlers, missionaries and administrators
alike.
For migrants from British Central Africa, the experiences of the
Rand had explosive force. Many returned home with heightened
expectation and a sharpened sense of grievance, which character-
ised leaders of the Watchtower movement like Eliot Kamwana
or Charles Domingo; the discontent expressed itself in the
millenarianism which found a ready seed-bed in a territory which
had been the scene of so much mission activity since the days of
Livingstone. For many this constituted the answer to the
contradiction between their aspirations and the harsh reality of
colonialism.
THE DISCOVERY OF THE WITWATERSRAND AND THE

END OF AFRICAN INDEPENDENCE SOUTH OF THE
LIMPOPO
Already in 1886, it was well known that gold was to be found

in the Transvaal. From the 1850s there had been rumours of its
existence, and the Republic's first Gold Law dates from 1858. In

1874 alluvial diggings had been started in the Lydenburg district

of the eastern Transvaal. Although these were quite quickly
worked out, ten years later the De Kaap fields were discovered,
also in the eastern Transvaal, while first reports in 1885 from the
Sheba mine in the Barberton district aroused spectacular hopes.
The diamond discoveries meant there was local capital to invest,
while small diggers displaced by the company amalgamations
streamed north. The excitement was short lived, however. The
claims were exaggerated and the Barberton fields were soon
deserted, leaving a legacy of bankruptcies and of scepticism on
the share market.
Thus when a few months later the existence of new gold-bearing
reefs was announced, stretching some 40 miles from east to west
along the Witwatersrand, there were many — including the by now
prominent Cape politician and diamond magnate, Cecil John
Rhodes, and his associate, C. D. Rudd — who remained cautiously
on the sidelines;11 but if Rhodes and Rudd were slow off the
ground, others were not. Very quickly, capitalists who had first
made their fortunes on the Kimberley mines moved on to the
Rand, with all their expert knowledge of the international capital
market and the experience they had gained in systems of labour
control. In September 1886 the Rand was declared a public
digging and by the end of the year land was being sold in the new
city of Johannesburg. Within three years more than five hundred
companies had been floated, though few of them were actually
producing gold.
Some of the feverish excitement which accompanied the
mineral discoveries can be appreciated from the colonial and
imperial literature of the time. Here, one description must suffice
- Lady Bellairs writing of her trip to Johannesburg in 1889, only
three years after the opening up of the fields:
Never in the history of the universe was such an extraordinary city conceived
or carried out as Johannesburg...Day after day comes the news of fresh
discoveries; week by week patience and the pick are teaching us what we may
later expect. We are simply living in a sea of gold...
Johannesburg is barely two and a half years old; but as we drive across the
tops of hills and gaze downwards it seems impossible to believe it. Acre after
acre, mile after mile covered with lordly buildings or the humble shanty. House
room is precious and costly...An ordinary eight-roomed house may total
£3,000 or £4,000 and a rental of £50 a month is by no means uncommon... every
" A. P. Cartwright, Gold paved the way (London, 1967), 39.
452

THE DISCOVERY OF THE WITWATERSR AN D
bit of furniture or everything indeed that comes from the Cape costs threepence
per pound in freight alone... Two hundred wagons a day come into the market
place, each carrying a freight of 7,000 to 8,000 lbs and drawn by twenty patient
oxen. A month they have been on the road.12
Yet within a year of this flurry, Johannesburg was in the depths
of gloom: in 1889 at the height of the boom, the miners hit pyritic
ore. Hitherto, it had been relatively easy to crush and extract the
gold from milled ore, through its amalgamation with mercury.
This new problem threatened to add £1 per ton to costs, and
render many of the mines unprofitable. As the market collapsed,
dozens of companies closed down. The crash resounded across
South Africa. In the coastal colonies, already closely tied to the
fortunes of the Rand, the depression spread. In the Cape, three
banks were forced to close down. The pattern of boom—slump,
which was to characterise the mining industry and the increasingly
inter-connecting regional economy, was already establishing
itself.
It was this crash which disillusioned both Lady Bellairs and the
many individual fortune-seekers who made their way so hopefully
to the Rand: patience and the pick could not exploit the new gold
fields. Mining engineers now came to appreciate that the' so-called
reefs were not reefs... but layers of conglomerate in a vast lake
of sediments which were thus continuous and could be followed
to great depths'. 13 The gold fields picked up as this fact was
appreciated.
As local capital proved inadequate to the task, international
investment came to the rescue. It needed little encouragement. By
the late 1880s the decline in world prices from their 1866-7 levels
was being widely ascribed to the fall in the world's gold output
- estimated by contemporaries as a 20 per cent decrease between
1875-91, compared with the previous quarter century. At the
same time, gold was increasingly becoming the world's monetary
standard. European bankers and financiers greeted the new finds
with enthusiasm, and new technology was brought to bear on the
problems of mining refractory ores at great depth: the discovery
of the MacArthur—Forrest cyanide process made possible the
recovery of pyritic gold and made it viable to mine even low-grade
12
Lady K. F. Bellairs, The Witwatersrand goldfields. A trip to Johannesburg and back
(London, 1889), cited in D . H. Houghton and J. Dagut, Source materials on the South
African economy, 1860—1970 (Cape Town, 1972), 1, 302-3.
13
Cartwright, Cold paved the way, 61.
433

ore; steam power and electricity opened the deepest levels to

exploitation. In 1892 the first deep-level mines were in operation.
In spite of the 1895 stock market crash in mining shares (in part
the result of over-speculation, and the large amounts of capital
necessary for the development of the deep levels), by then the
future of the deep-levels was well appreciated by European
financiers and the leading companies: Wernher, Beit and Company
and their subsidiary, Ecksteins, and Rhodes's Consolidated
Goldfields.
From 1886 until 1913 between £116 million and £134 million
was invested in the Witwatersrand by European shareholders —
three times as much finance as in all Canadian and Australian
mining activity. It came in five major spurts: 1889, 1895, 1899,
1902—3 and 1908—9, although it became far more difficult for the
mines to raise outside capital after 1903.14 In addition, the
profitable exploitation of the newly discovered deep-level mines
was made possible only by the flow of huge amounts of overseas
capital needed for primary development. Finance was necessary
not only for investment in mining: because the minerals were
discovered so far in the interior and in the midst of essentially
pre-industrial societies, an entirely new infrastructure had to be
created to make the mines viable. Given the significance of the
discoveries, however, investment was not slow in coming: British
investment in South Africa rose from £34 million in 1884 — on
the eve of the gold-mining revolution — to £3 51 million in 1911
(the sharpest rise anywhere over the same period of time).15
Crucial for the raising of these vast sums of money was the
amalgamation of the individual companies into a small number
- eight by the century's end - of mining finance 'houses' or
groups, with a high degree of overlapping ownership. Apart from
giving the mines access to the major sources of international
capital, they enabled research services to be centralised, thus
cutting overhead costs. Of these groups, the two most important
were Wernher, Beit and Eckstein, and Rhodes's Consolidated
Gold Fields, both formed in 1892. By the end of the 1890s
14
Van Onselen, Studies, i: New Babylon, 3 ; for detailed studies of investment on the
Rand, see R. V. Kubicek, Economic imperialism in theory and practice: the case of South
African gold-mining finance, 1886-1914 (Durham, NC, 1979) and J. J. Van-Helten,
'British and European economic investment in the Transvaal, with specific reference
to the Witwatersrand goldfields and district, 1886-1910', Ph.D. London, 1981.
15
A. K. Cairncross, Home andforeign investment, 1870—if i) (Cambridge, 1953), table
42, 185.
434

THE DISCOVERY OF THE WITW ATERS R AND
Wernher, Beit alone produced about half of the gold mined on

the Rand. Both these companies, together with Farrar's Anglo-
French group, undertook the risky business of raising capital both
in the UK and on the European stock exchanges to work the very
expensive deep-levels, and their involvement in these levels and
in a wide variety of enterprises connected with the mining
industry and the development of the Rand, was to give them a
long-term stake in the territory and an acute concern with the
actual nature of the Transvaal state which was less felt by the
smaller companies. By 1889 the Transvaal Chamber of Mines had
been formed to act as the spokesman of the gold-mining
companies, and to negotiate on their behalf with the state. Until
1902 the chairman was invariably drawn from Wernher, Beit and
Eckstein.
From the late 1880s gold rapidly outstripped diamonds as the
region's most important export. By 1898 the value of gold
production had risen to £16 million, more than a quarter of the
world output. In the twelve years between 1883 and 1895, the
revenue of the South African Republic increased twenty-five
times, as a result. At last immigration to the South African
colonies needed no official encouragement, as thousands of
fortune seekers of every description made their way to the
'Golden Rand'. In 1870 the total white population of southern
Africa was probably under a quarter of a million; by 1891 it had
increased to over 600,000, while by 1904 it was over one million.
Within ten years of its foundation, Johannesburg's white popu-
lation numbered 102,000, and it ranked 'as one of the commercial
centres of the world... For thirty miles along the crest of the ridge,
the pit-head gears, batteries and surface works of the sixty or
seventy companies in active operation give evidence of the
millions of capital which have been invested.'16 Thus, within a
remarkably short period of time, a highly sophisticated form of
capitalism was implanted on the South African veld.
Nor were the effects of the gold discoveries confined to the
South African Republic. In the Cape Colony, the Orange Free
State and Natal, changes set in train by the diamond discoveries
and the opening-up of the early gold-fields were now sharply
accelerated. In almost every sphere, the mineral discoveries led
to economic growth. Roads, railways and harbours had to be
16
W. B. Worsfold, South Africa (London, 1897), 156-7.
435

built, and enterprises already established in the coastal colonies
rushed to the north in an effort to reap the benefit of the new
opportunities. Manufacturing, though in its infancy even in the
Cape Colony, also responded to the new consumer market on the
Rand, and in the new towns and villages which sprang up in its
wake. The need for coal led to the renewal and expanded
exploitation of the mines in the Cape, and especially in Natal and
the Transvaal. Farming became more market-orientated and land
prices soared. In Natal, both the dairy and the wattle industries
trace their origins to the 1890s, while in the Cape and the Orange
Free State more scientifically based farming practices were adopted
by both white and black farmers. Nevertheless, there was still
tremendous opposition to government-directed attempts at
disease control, such as the Cape government's anti-scab Act in
1894 and the efforts all over southern Africa - often more drastic
than the disease itself - to stamp out the rinderpest epidemic
of 1896-7.17
If the diamond discoveries had led to a vastly increased market
for labour, with these new developments the demand became
ubiquitous and insatiable. By far the greatest number of unskilled,
cheap workers was needed on the Rand. With the development
of the deep-level pits an enormous amount of support work was
necessary, which could not be easily mechanised. The problem for
the mine magnates was to attract a work force to the dangerous
and difficult work underground at a time of expanding demand
from other sectors of the economy, while they had no real leverage
over the Transvaal state and Africans still had independent access
to land.
As we have already seen, the years following the diamond
discoveries saw the conquest of the majority of the African
polities south of the Limpopo. Now the final stages were com-
pleted as the drive for land, minerals and labour, spurred by the
demands of the Rand and the increased availability of speculative
capital, intersected with earlier historical processes of settler
expansion. Within the Transvaal, the defeats of Malaboch's
Xananwa in 1894 and the Venda in 1898 rounded off the
conquests of the African people of the north and east, which had
begun with the wars against Sekhukhune (1876 and 1879) a n ^
17
SeeC. vanOnselen, 'Reactions to rinderpest in South Africa',]. Afr. Hist., 1972,
'3. 3. 47S-88-
436

THE DISCOVERY OF THE WITW ATERS R AN D
Mabogho (1884). In the Cape, Pondoland was finally brought

under formal administrative control after a dramatic show of force
by Rhodes in 1894. In the next year Bechuanaland south of the
Molopo river, since 1885 the Crown Colony of British Bechu-
analand, was incorporated into the Cape Colony, though northern
Bechuanaland remained a British protectorate under imperial
control. Zululand, too, after eight years of destructive civil war
between the Usuthu and their opponents, was annexed by the
Crown in 1887, and handed over ten years later to the white
colonists of Natal, who had achieved responsible government in
1893. In each case the immediate occasion for annexation differed,
though the underlying causes were the same as they had been in
the previous decade. For Britain the' natural' agents of annexation
continued to be the local settler communities. With their increased
wealth and population, and united in their desire to control
additional land and labour, they needed far less urging than they
had before 1886.
Both within and outside the frontiers of the colonial states
concessionaires were spurred by the prospects of further discov-
eries and by the speculative capital available for investment in
southern Africa as a result of the Rand discoveries. Land
speculation had long been a feature of South Africa's political
economy, but with the discovery of the Witwatersrand and the
rise in land prices, huge new areas were staked out by land
companies, mining houses and individual entrepreneurs. In the
Transvaal itself, land prices rose dramatically - and although for
a few this spelt the beginning of good fortune, those without
land-title (usually poor kin and clients of wealthier Boer farmers)
increasingly found themselves landless in the midst of plenty. By
1900 land companies (some owned by mining houses) possessed
some 20 per cent of the land in the Transvaal, of which Wernher,
Beit and Eckstein controlled about a quarter.18
It was in Swaziland that the effect of this land speculation and
concession-hunting could be seen at its most spectacular. From
the 1870s Boer graziers and British mineral prospectors had
sought land and mineral concessions from the Swazi king,
Mbandzeni. With the discovery of the De Kaap and Komati gold
fields on Swaziland's western flank in the eastern Transvaal,
18
War Office, Intelligence department, Ust of farms in... [21] districts, Transvaal, with
the names of their registered owners (Pretoria, 1900); 1 am grateful to Dr Stanley Trapido
for this reference and information.
437
i
concession-seekers flooded in, hopeful that Swaziland would

prove a second Barberton. Perhaps believing that he would be
able to neutralise the increasing political pressure he was feeling
from the Boer graziers and the South African Republic and gain
British protection, Mbandzeni granted concessions to the
newcomers, frequently to land already allocated to graziers, with
increasing disregard for the consequences. The country became
a speculators' paradise, as concessions granted by the king were
then traded, sold and resold, subdivided and sublet. In 1889
Mbandzeni even granted a concession to grant concessions. The
appointment of the Natalian, 'Offy' Shepstone (the son of Sir
Theophilus Shepstone), as official adviser to the King failed to
halt the decline into anarchy, and he was soon ousted by a white
committee which virtually ran the country. The deteriorating
political situation was matched by a deterioration in the health of
the king, and led to a spate of witchcraft accusations and political
murders. After Mbandzeni's death, the ability of the Swazi ruling
group to juggle the complex rivalries of Britain and Boer
concessionaires, and the increasing number of republican and
imperial agents, was limited.
By bribing the local British concessionaires and probably
Shepstone himself, the South African Republic now staked a claim
to a large number of concessions which made the country virtually
ungovernable without republican intervention. Quite apart from
providing land for his subjects, Swaziland had a strategic im-
portance for Kruger in that it lay in the path of the Republic's
independent access to the sea at St Lucia Bay. Although Britain
was determined to block Kruger's attempts to gain a port, by 1889
it was clear she was reluctant to intervene in the increasingly
sordid strife of the concessionaires in Swaziland, and was willing
to hand over Swaziland to the republic in exchange for a free hand
north of the Limpopo. Despite the vociferous opposition of the
Swazi rulers, in 1894, under the Third Swaziland Convention,
Britain finally agreed to Transvaal sovereignty over the territory
- taking care to annex the Tsonga territory between Swaziland
and the sea to Zululand almost immediately thereafter. The
protectorate lasted until the South African war, when Swaziland
came under British control.19
19
This section draws heavily on P. Bonner, Kings, commoners and concessionaires: the
evolution and dissolution of the if th century Svaî state (Cambridge, 1983), 185—207.
438

THE SCRAMBLE FOR ZAMBEZIA
The Limpopo constituted no barrier to the flood of concession

seekers, and the years between 1889 and 1895 saw the dramatic
annexation of all the African territories up to the Congo in what
has been termed 'a gigantic speculation in mining futures'.20 In
south-central Africa, British ambitions competed with those of the
South African Republic, Portugal, Germany and the redoubtable
King Leopold II of the Belgians. In an era of increased inter-
national competition, the rivals staked their claims to the sub-
continent, beguiled by the lure of what Joseph Chamberlain
described as 'those vast auriferous and fertile regions which
stretch almost into the very heart of the African continent'.21 In
east-central Africa, to the west and south of Lake Malawi, the
thrust from the south encountered the less powerful, but none-
theless still significant, anti-slavery missionary 'frontier' from the
east.
From the reports of Karl Mauch and Thomas Baines in the
1860s on the Ndebele kingdom, it was known that there were
'ancient workings' between the Zambezi and the Limpopo,
rumours of which had reached fabulous proportions. By the
mid-1880s Lobengula, king of the Ndebele was surrounded by
concession hunters seeking trading, land and mineral rights. In
1887-8 alarm at possible Transvaal expansion across the Limpopo
led the British high commissioner at the Cape to secure a treaty
with Lobengula declaring the area a British sphere of interest.
It was at this point that Rhodes, by now the most powerful
capitalist and politician at the Cape, decided to enter the arena.
By 1888, having missed out on the most important mining
properties on the central Rand, he was beginning to regret his
initial caution. In central Africa he hoped to find a ' second Rand',
open a new area for exploitation, and out-flank the increasingly
powerful South African Republic.22 At the end of August 1888,
he hastily sent a party under Charles Rudd, to Bulawayo, to secure
the mineral rights to Lobengula's kingdom. Assisted by imperial
officials in South Africa and by the former missionary J. S. Moffat,
20
R. E. R o b i n s o n and J. Gallagher with Alice D e n n y , Africa and the Victorians
( L o n d o n , 1961), 250.
21
In a s p e e c h to the L o n d o n Chamber o f C o m m e r c e , May 1888, cited in J. L. G a r v i n ,
Tie lift of Joseph Chamberlain, 3 v o l s . ( L o n d o n , 1 9 3 2 - 3 4 ) , 1, 4 6 4 .
22
I. R. P h i m i s t e r , ' R h o d e s , R h o d e s i a and t h e R a n d ' , J. Southern Afr. Studies, 1974,
1, i, 79-9°-
459

now the British representative in Bulawayo, Rhodes's agents soon

had the edge over their competitors at the capital. Whether,
however, Lobengula ever signed the concession Rudd claimed
subsequently to have procured is dubious. Very soon after his
departure Lobengula repudiated the concession, which he had
been led to believe had full imperial backing. Moreover, although
the oral agreement between Rudd and Lobengula stipulated that
no more than ten surveyors or miners were to enter the kingdom
at a time, the written document granted the newly formed British
South Africa (BSA) Company full rights to do 'whatever was
necessary' to exploit the mineral riches of the kingdom.
Concession in hand, in London Rhodes was able to outbid or
incorporate his chief rivals, and in 1889 he persuaded the British
government to grant his BSA Company a royal charter for the
exploitation of central Africa. This authorised the company to
extend its administrative control over a huge and vaguely defined
area north of British Bechuanaland, confined only by the Portu-
guese and German coastal colonies. Little attention was paid to
the nature of the company to which these wide powers were being
delegated. An amalgamation of big financial interests in the City
of London, its prospectus bedecked by an array of aristocratic
directors, the BSA Company represented the most naked form of
unchecked capitalist enterprise. Notwithstanding some bland
phrases about protecting 'native interests', it set about realising
a profit in central Africa with ruthless determination.23
Even before his charter was ratified, Rhodes was encouraged
to stake his claims across the Zambezi. Here, the head of the
Conservative government and British Foreign Minister, Lord
Salisbury, was alert to the dangers of advance by the Portuguese,
still dreaming of joining their east and west African territories,
to German ambitions of conquering ' Mittelafrika' and to the
southward drive of King Leopold of the Belgians from the Congo
Independent State. In the Shire highlands in particular, and
around Lake Malawi, Anglo-Portuguese rivalries had sharpened
since the mid-i88os. The appointment in 1883 of a British consul,
with a watching brief on the slave trade in this region, led the
Portuguese to step up their activities both north and south of the
Zambezi, in an attempt to secure treaties with African chiefs.
23
For an account of the British South Africa Company in its formative years, see
J. S. Galbraith, Crown and charter (California, 1974) and his' Engine without a go vernor:
the early years of the British South Africa Company', Khodtsian History, 1970, 1, 9-16.
440

Increasingly, local British missionaries and the African Lakes

Company called for Foreign Office intervention and the declar-
ation of a British sphere of influence in the region. Faced with
Treasury reluctance to take financial responsibility for the expan-
sion of empire, however, there was little Salisbury could do but
send the young Harry Johnston, who had already impressed him
with his imperial zeal, as emissary to Portugal and consul to
Mozambique. A chance meeting with Rhodes on the eve of his
departure in May 1889 enabled the two men to settle 'the
immediate course of events in South and Central Africa'.24
Rhodes agreed to grant Johnston £2,000 for a treaty-making
expedition, and promised more if he would establish the rights
of the chartered company north of the Zambezi. For Salisbury
nothing could have been better. Rhodes and his company were
the ideal agents to further British imperialism north of the
Limpopo, against the encroachments of both European powers
and the South African Republic, at minimal expense to the British
taxpayer.
The next twelve months were occupied by a flurry of treaty-
making in central Africa by Johnston and the agents he appointed,
as well as by additional emissaries engaged by Rhodes. A major
objective was Msiri's kingdom in Katanga (the present Shaba
province of Zaire) where rich copper deposits were already
suspected. But Rhodes and Johnston were concerned to extend
the frontiers of imperial hegemony over as wide an area of
south-central Africa as possible. While Johnston was anxious to
stake out British claims against German and Portuguese expansion
from the east and west coasts, and against King Leopold's moves
south from the Congo Independent State, Rhodes wanted to
secure the rights of his company over the mineral wealth he
believed to run in a continuous reef north from the Rand through
Mashonaland and across the middle Zambezi to Katanga. More-
over, if Africa's potential was to be exploited, the main lines of
communication had also to be secured: for Rhodes, the route to
Beira was to become vital in the early years of his settlement in
Rhodesia, though from the wider imperial point of view it was
Johnston's vision of a Cape-to-Cairo route which was furthered
by the treaty-making. Although none of Rhodes's agents
24
Cited in R. Oliver, Sir Harry Johnston and the scramble for Africa (London, 1957),
152.
441

SOUTHERN AND CENTRAL AFRICA, 1886— I 9 I O
persuaded Msiri to grant concessions to the BSA Company, and

in the following year his kingdom was forcibly annexed to the
Congo Independent State, the agreements signed in 1889-90
claimed the whole of present-day Zambia and Malawi for
Company and Crown.
The precise nature of the treaties and their doubtful legality
was of little importance. In general, they secured mineral and
sometimes land concessions for the company, while the chiefs
agreed to accept British jurisdiction over non-Africans in their
domains and over their external relations. Despite all the paper-
work, colonial rule in what is now Malawi was based on the
declaration of a British protectorate - over the Shire highlands
in 1889, and over the whole of the area west of Lake Malawi to
the Luangwa watershed in 1894; in the company sphere still
further west, the signature of the more powerful chiefs on treaties,
which they were led to believe were with the Crown, became the
basis of BSA Company rule, even over peoples who were not party
to the signing. But the frontiers of the British, German and
Portuguese domains were being decided in the chancelleries of
Europe, rather than in Africa: and here the treaties provided
useful bargaining counters. In 1890-1 conventions were drawn
up between the British and the Portuguese, and the British
and the Germans, which largely established the frontiers of the
modern states of Zimbabwe, Mozambique, Malawi and Tanzania
in the east, and between Zambia, Angola and Namibia in the west,
although the precise boundary line between Zambia and Angola
was only settled in 1905, through the mediation of the king of
Italy — and with the loss of considerable territory claimed by the
Lozi king.
THE ESTABLISHMENT OF COLONIAL STATES: SOUTH

WEST AFRICA AND RHODESIA
Treaties and lines on maps were one thing; giving them substance
was another matter. In South West Africa, Bismarck hoped that
the privileged Deutsche Kolonialgesellschaft fur Siidwestafrika
would effectively administer the territory in exchange for a
monopoly over trade, land and minerals, and at no cost to the
German taxpayer. By 1888, however, this German company had
shown itself totally incapable of raising sufficient capital either to
442

SOUTH WEST AFRICA AND RHODESIA
develop the vast semi-arid territory or to subdue and control the

local inhabitants. Most of the peoples of South West Africa, who
had been persuaded to sign the so-called Schiit^yertrdge (treaties of
protection) with representatives of the company or the German
commissioner soon realised the latter's incapacity to extend the
protection promised. They therefore continued to pursue the
struggles over cattle, grazing and waterholes which had punctu-
ated the history of the area for much of the nineteenth century.
Nevertheless, in 1890, the southernmost Nama chiefs accepted
Schiit^yertrage and, after a brief flirtation with Cape merchant
interests, Kamaherero, chief of the Herero, who was being
threatened again by the Nama chief, Hendrik Witbooi, renewed
his adherence to the Germans. Witbooi remained the German's
most formidable opponent. Fully aware of the implications of
colonial rule, he had prophetically warned Kamaherero as early
as 1889, 'You will... eventually sorely regret... that you have
handed over your country and your governing rights to the white
people',25 and began making peace overtures to the Herero and
other Nama chiefs in the hopes of uniting them against German
rule. The death of Kamaherero in 1891, however, left his
successor, Samuel Maherero even more dependent on the Germans
in the face of serious internal opposition as well as the continued
threat from Witbooi; he was to prove less compliant in the
following decade. By 1892 the imperial commissioner in South
West Africa, C. von Francois, decided to send a military
expedition against Witbooi, although it took two years of fierce
guerrilla struggles before the latter was forced to submit to the
new landeshauptmann, Theodor von Leutwein, in 1894.
Given the continued turbulence, it is hardly surprising that
German industrialists failed to invest in the colony, despite a
flutter of activity in 1888, when it was believed that gold had been
discovered. German authority remained fragile until 1893 when
L. von Caprivi, Bismarck's successor as chancellor, declared that
despite its poverty, he had no intention of abandoning the
territory. He encouraged investment by the grant of lavish
concessions to commercial companies. Almost unnoticed, Rhodes
controlled the two largest, including all the mineral-rich land. By
the early twentieth century, nine companies held about a third of
the land, although none of them had paid a dividend nor carried
25
Cited in J. H. Esterhuyse, South West Africa, 1880-1894 (Cape T o w n , 1968), 159.
443

out much by way of economic development. At the same time

the Germans were drawn ineluctably into local conflicts, as the
government now had to establish the minimal conditions under
which the companies could operate.
With the defeat of Witbooi, South West Africa was rapidly
opened up to white settlement, despite the opposition of Leutwein
(who became governor in 1898) to the more extreme settler
demands. German attempts to enforce frontiers on grazing
peoples by confiscating vast herds of cattle, the banning of the
gun trade from Botswana, and the favouritism extended to the
western Herero under Samuel Maherero as against his chief rival
in the east, Nikodemus, led to a major uprising of the eastern
Herero (or Mbanderu) in 1896. Despite Leutwein's attempts to
avoid war, three Mbanderu chiefs were court-martialled, and
resistance against German rule mounted. The governor had little
doubt about the necessity for capitalist development, before
which the indigenous peoples would be forced to bow. Within
ten years a colonial state had been established, facilitated by the
building of the Otavi railway, the reserves policy of 1897—8, and
the rinderpest epidemic which wiped out 95 per cent of African
cattle-holdings, thus lowering resistance and freeing grazing land
simultaneously. By the turn of the century, the Rhenish mission-
aries were noting with alarm 'the disintegration of tribal life'.26
For much of this period, the Ovambo people, who straddled
the region claimed by the Portuguese and the Germans, remained
outside formal colonial control. From the mid-1880s cattle
replaced ivory as the main item of trade, which the Ovambo,
armed with guns from their slaving days, raided from their
neighbours and traded to the new markets in Botswana and at
Kimberley. Equally hit by the rinderpest epidemic, and the
disastrous drought and flood years which followed, the Ovambo,
like the Herero, became increasingly indebted to European
traders. The war leaders who had emerged in the years of slave-
and ivory-trading now extorted harsh and arbitrary tribute from
their own people on behalf of the chiefs. Already by the turn of
the century a trickle of pauperised Ovambo were making their
way to the labour markets to the north and south.
In Rhodesia, rule by concessionary company lasted rather
longer than in South West Africa. One of the great advantages
26
H. Bley, South West Africa under German rule iff4-1914 (London, 1971), 114.
444

of Rhodes's BSA Company for the British government was its

promise to populate central Africa with white immigrants, who,
it was believed, would not only make British occupation effective
but would also bring the necessary capitalist development to the
interior. Using company shares to buy out opposition, and backed
by both British and Afrikaner opinion in the Cape Colony, where
he was now prime minister, Rhodes sent a 'pioneer column' into
the indeterminate borderlands between the Ndebele and the
Shona in mid-1890. It consisted of nearly two hundred white
settlers, one hundred andfiftyblack settlers drawn from Kimberley
and the eastern Cape, and some five hundred armed police.
Lobengula watched the column make its way to Salisbury with
dismay, but without interfering. By October, three hundred men
were already prospecting, and 465 gold claims had been registered.
At the same time a small body of men under the company
administrator, A. R. Colquhoun, marched into the eastern high-
lands where the gold reef was thought to run, and whence Rhodes
also hoped to forge a route to the coast. Here, however, the
company's claims clashed with those of Portugal, and the last
months of 1890 and beginning of 1891 saw a number of
skirmishes, both sides using and being made use of by local Shona
chiefs and by the ruler of the Gaza kingdom and his chief rivals
in Manica. In 1891 an Anglo-Portuguese convention delimited the
BSA Company's sphere. In all of this, little account was taken of
the prior rights of the African peoples of the area. By and large,
the company proceeded on the authority of the Rudd concession
— even though it was allocating land which had in no way formed
part of Lobengula's kingdom — and which in any case the Rudd
concession had no authority to dispense.
From the outset it is clear, however, that the BSA Company's
real goal was the Ndebele kingdom, where the main gold reefs
as well as the richest lands of the territory were believed to be.
If the lands between the Limpopo and the Zambezi were to be
opened up for capitalist development, the Ndebele kingdom
would have to be destroyed. In 1893 Leander Starr Jameson, who
had replaced Colquhoun as administrator, deliberately provoked
war. A raid by the Ndebele against their Shona tributaries at Fort
Victoria gave him the pretext. In October the company's troops
invaded the Ndebele kingdom, and within weeks Bulawayo was
in flames. The king fled northward towards the Zambezi, where
445

he soon caught smallpox and died. Despite the successful invasion

of the strategic centre of the kingdom, Ndebele administration
and political structures remained intact, however, and Lobengula's
son, Nyamanda, was quickly recognised as his successor. Never-
theless, the war resolved, as Jameson had opined, the' uncertainty'
which was devaluing Chartered shares, removed an obstacle to
the flow of Shona labour, and opened access to the reputedly rich
Ndebele gold-fields.
The conquest of the Ndebele was followed by a speculative
boom in BSA Company shares, the investment of more than
£500,000, and the flotation of several new speculative companies:
these were concessionaries of the BSA Company, which received
a half-share of any profits on mining ventures. By the end of 1894,
however, it was clear to Rhodes's mining engineers that the
second Rand was not in Rhodesia, but on the Rand itself. As their
hopes of instant wealth through gold waned, so the settlers and
the company turned to the other African assets: cattle and land.
In terms of their agreement with the BSA Company, settlers
who participated in the war were granted farms on a lavish scale
and were additionally rewarded with mineral claims. A land
commission perfunctorily set aside two land reserves for Ndebele
occupation on the Gwaai and Shangani rivers, on soils neither
white nor black farmers thought worth cultivating. Now Jameson
set about allocating the rich Ndebele heartlands to the' honourable
and military' element who had formed land syndicates and were
already buying out the 'Pioneers'. 27 Vast quantities of Ndebele
cattle were looted on the fiction that they had belonged to the
king, and forced labour and taxation were prised out of both
Shona and Ndebele communities by the newly appointed and
increasingly unpopular Ndebele police. Inadequate administrative
manpower was made good through the enlistment of settlers into
the newly formed Native Affairs Department, which was mainly
concerned with branding and registering the looted cattle, im-
pressing forced labour and extorting further cattle in lieu of tax.
These activities rapidly led to an explosive situation, as Ndebele
opposition gathered force under Nyamanda.
The flashpoint came in 1896, when to company misrule were
added further despair and room for hope: the outbreak of
27
The phrase was that of the imperial resident commissioner, William Milton,
when he arrived 'to salvage something from the wreck'. See R. Palmer, Land and racial
domination in Rhodesia (London, 1977), 35—6.
446

rinderpest ravaged what was left of African cattle, while the
removal of company police from October 1895 in anticipation of
Jameson's attack on the South African Republic gave Africans
their opportunity. On 24 March, the Ndebele rose, and within a
week more than 145 whites had been killed and the survivors
huddled in laagers in the towns. Late in June, the Ndebele were
joined by the Shona of central and western Mashonaland who had
come out in full-scale war following sporadic acts of violence
against white tax-collectors, labour recruiters and farmers and
their black agents. Only the timely arrival of imperial troops and
the collaboration of a number of Shona polities which either
feared the resurgence of Ndebele power or were more preoccupied
with local rivalries and regional political alignments, saved the
settlers. Nevertheless, continued Ndebele strength induced
Rhodes to seek a negotiated settlement with Nyamanda's advisers
before the BSA Company was completely bankrupted. As a result
of his initiative, the royal counsellors accepted peace in exchange
for personal amnesty, the disbanding of the Ndebele police and
promises of land. Although the war against the Shona dragged
on until almost the end of 1897, the preparedness of the Ndebele
leadership to come to terms with Rhodes probably preserved
company rule in Southern Rhodesia.
Although there is some controversy over the extent of the
co-ordination both between the Shona and Ndebele and between
the Shona polities themselves, amongst the Ndebele and those
Shona polities within their political orbit the uprising was most
probably organised by the still intact royal political authorities.
Amongst the Shona it seems likely that in some areas the
Chaminuka-Nehanda spirit mediums and priests of the Mwari cult
played a political role in advising paramounts and overcoming
the problems of scale, though their uprisings were neither as
preconcerted nor as well-organised as is sometimes thought.28
Nevertheless, chimurenga, as the rising was called by Africans, led
to direct imperial intervention in company affairs for the first time,
with the appointment of a British resident commissioner at
Bulawayo responsible to the imperial high commissioner in Cape
28
The nature and organisation of the 1896-7 rising is a matter of some controversy.
For the debate see T. O. Ranger, Revolt in Southern Rhodesia, 1896-7 (London, 1967);
D. N. Beach, 'The risings in south-western Mashonaland', Ph.D. London, 1971 and
his 'Chimurenga': the Shona risings of 1896-7', / . Afr. Hist., 1979, 20, 4; and
J. Cobbing, 'The absent priesthood: another look at the Rhodesian risings of 1896—7',
J. Afr. Hist., 1977, 18, 1.
447

Town. Inspired by fear of further African resistance, as well as

guilt and compassion, the Colonial Office demanded the reform
of the Native Affairs Department, land to be set aside for African
occupation, and mission and industrial education to be encour-
aged. Inadequate though these changes were, they set some limit
on the extent and speed of African dispossession.
THE COLONIAL PRESENCE NORTH OF THE ZAMBEZI
Despite the initial hopes of the British government that the British
South Africa Company would provide for the effective occupation
of the territory to the north of the Zambezi as well, events to the
south left Rhodes little time, energy or resources to spare until
the late 1890s. Opposition from the missionary lobby and from
certain interests within the African Lakes Company ensured the
separation of the region around Lake Malawi and the Shire valley
from the BSA Company's domain to the north of the Zambezi.
In 1891, after consultations in London between the Colonial
Office, the Foreign Office, the African Lakes Company and
Johnston, the area known as Nyasaland (now Malawi) was
declared a Protectorate.
Johnston returned to his administrative capital at Zomba as
commissioner and consul-general for the territories under British
influence north of the Zambezi. At the same time, the BSA
Company had appointed him administrator of their sphere and
agreed to contribute an annual sum of £ 10,000 towards the costs
of administration, as well as a transport subsidy. For both the
British government and the BSA Company, the ambiguity of
Johnston's role was offset by their mutual advantage. It released
Lord Salisbury from Treasury parsimony; and it was a cheap way
for the company of securing their possessions at a time when they
were unprepared to launch their own administration. Even in
Nyasaland it was anticipated that Johnston would secure the
company's long-term land and mineral interests, while the estab-
lishment of a colonial state on its eastern flank and main
communications route was a further advantage.
Financial insecurity and the necessity to serve two masters soon
soured Johnston's relationship with Rhodes, however, and in
1895 the Treasury took over financial responsibility for what then J
became known as the British Central African Protectorate. The *
448

COLONIAL PRESENCE NORTH OF THE ZAMBEZI
BSA Company appointed its own deputy administrator in Nor-

thern Rhodesia, as its territory north of the Zambezi came to be
called. During the first six years of the protectorate, Johnston
engaged in a series of wars against the Swahili-Arab traders and
the Yao, in particular those who straddled his lines of commun-
ication. His strategy was dictated by his limited funds and troops;
he had less than 200 men at his command, including 71 Indians,
mainly Sikhs. He sought collaborators amongst the weaker
agricultural groups, who were in general happy to have allies
against their raiding and slaving neighbours, and amongst dis-
satisfied elements within chiefdoms. Yet his campaigns in the first
couple of years were not particularly successful. It was only after
he was able to increase the number of his troops with the
additional finances he obtained from Rhodes in 1894, and the
doubling of his resources with the Treasury's first grant-in-aid in
1895, that he was able to step up the campaign against the major
opponents of his regime. By the time he left the protectorate,
Johnston could boast that he had rid the area of slavers.
Although Johnston thus defended his 'little wars' as anti-
slavery campaigns, they also had the virtue of releasing 'slave
labour' for European enterprise. His vision was not dissimilar to
that of Rhodes. From the outset, he was convinced that the
protectorate's future development should be based on the marriage
of white enterprise and black labour, assisted by the Asian
middleman. As he wrote in his final report before leaving the
protectorate: 'The native labour question is almost the most
important question which can now claim the attention of those
administering the Protectorate... All that needs now be done is
for the Administration to act as friends of both sides, and
introduce the native labourer to the European capitalist.'29 From
the African point of view it looked rather different. As Nkosi
Ngomane, chief of the Maseko Ngoni put it: 'I come to ask why
the white man brings war to my country, kills my people and
harms my villages.'30
Notwithstanding the treaties signed with chiefs north of the
Zambezi at the beginning of the 1890s, most of the people living
29
F O print 6851 n o . 142, H . H . J o h n s t o n , Report o n t h e British Central Africa
Protectorate 1 8 9 5 - 6 , 29 April 1896, cited in E . T . S t o k e s , ' Malawian Political S y s t e m s '
in E . T . S t o k e s and R. B r o w n , The Zambcsian past, 372.
30
Cited in R. I. R o t b e r g , The rise of nationalism in Central ylfrica: the making of Malawi
ami Zimbabwe, ii/)-if64 (Cambridge, Mass, 1966), 20.
449

Ethnic group
Town or village
Colonial frontier
Limit of tote 19th C Lozi claims
Border of L021 reserved area, 1899
Lozi f toodplain
390km
2(k) miles
G *\$l. -"o \^
18 British Central Africa in the late nineteenth century

to the east of the Protectorate were drawn under colonial rule
somewhat more gradually. Neither Johnston, fully engaged in his
wars against Yao and the Swahili-Arabs, nor Rhodes, preoccupied
with events south of the Zambezi, had time to pay attention to
the subordination of what were to become the north-eastern and
north-western provinces of Zambia. In the north, Rhodes was
anxious to retain access to the important Lake Tanganyika
waterway, and between 1891 and 1894 Johnston established three
administrative outposts amongst the Tabwa, Lungu and Mambwe
people at Abercorn (Mbala), Fife and on Lake Mweru, to this end.
This was all he could do to keep communications open to the
north. He was certainly in no position to tackle the three
dominant groups in the area, the Bemba, the Ngoni and the
Lunda.
Almost imperceptibly, however, the white presence was
making itself felt. Although the decade up to 1895 was charac-
terised by its continuity with what had gone before, even before
the BSA Company appointed its own officials to take possession
of the region, the balance of power was shifting in response to
white activities on the periphery. Internal tensions over succession
to the kingship among the Bemba, raids and counter-raids and
the continuation of the profitable slave- and gun-trade between
the Congo and the coast may all have seemed of more moment
than the occasional white visitor, but the German defeat in 1893
of a large Bemba raiding party under their king was a frightening
portent of the white man's strength. Increasingly too, African
leaders were having to take account of missionaries, with the
expansion of the London Missionary Society (LMS) among the
Mambwe, and the White Fathers gradually extending their work
into Bemba country.
From 1895 the BSA Company began intercepting slave caravans
in the north-east, after Johnston's successful attack on Mlozi at
Karonga had cut the Bemba slave-trading link to the south-east.
Even then, the abortive attack on Kazembe's Lunda kingdom
on the Luapula river warned Major P. W. Forbes, the newly
appointed deputy administrator, of the need to proceed with
caution. It was only in 1899 that Kazembe's capital was finally
destroyed, and the chief fled to the Congo, surrendering the
following year. Limited attacks on slave and ivory caravans
around company outposts led the East African traders in Tabwa

S O U T H E R N AND CENTRAL A F R I C A , I 886— 19IO
country, who had an eye on what was happening at the coast, to
surrender the slave trade — the main object of company attacks
— in the hopes of preserving their profits in ivory.
To establish its authority in this area, however, the company
had to come to terms with the powerful Ngoni and Bemba
chiefdoms. Amongst the Ngoni, it was only the followers of Chief
Mpezeni who offered open resistance. As early as 1885 Mpezeni
had granted a large land concession to the German adventurer,
Karl Wiese. In 1891, Wiese, aware of the encroaching frontier of
the British South Africa Company, sold his claims to one of its
subsidiaries, the North Charterland Company - taking care to
inflate its value by circulating rumours of the rich gold resources
to be found in Mpezeni's country. From 1896 an increasing
number of prospectors began arriving in Ngoni country around
Fort Jameson (Chipata). Aware of the activities of Jameson and
Johnston to his south and east, the chief eyed them warily, though
his son and army pressed for war. In 1898, on a convenient
pretext, Mpezeni's country was invaded, his son and army-
commander killed, and vast numbers of cattle looted. By 1906 the
Fort Jameson Ngoni had lost 10,000 square miles of land and
five-sixths of their cattle. Fort Jameson became the capital of
North-Eastern Rhodesia.31
The fate of Mpezeni was a salutary lesson to the weaker peoples
of the region. No other group was to feel quite the same
onslaught, though the administration anticipated trouble with the
still powerful Bemba, whose chitimukulu (king) was reported to
have over a thousand guns in 1893. Perhaps as much through luck
as design, the Bemba came under the rule of the BSA Company
without overt struggle. The natural caution of Bemba leaders as
they took the white man's measure was matched by British
reluctance to provoke conflict. The gradual influence gained by
the White Fathers, and in particular the hold which Bishop
Dupont gained over Chief Mwamba in his last days in 1898,
together with the deaths in 1896-8 of the other leading Bemba
chiefs who might have led organised resistance, and the poverty
of resources in their area further contributed to this peaceful
outcome. On the death of Mwamba, the company established an
outpost in Bemba country, and indeed played a role in deciding
31
J. K. Rennie, 'The Ngoni states and European intrusion', in Stokes and Brown,
The Zambesian past, 325-31; J. A. Barnes, Politics in a changing society: the Fort Jameson
Ngoni (London, 1964), 93-6.
452

his successor.32 This, together with the crushing of the Ngoni and
the final defeat of Kazembe, meant that by the turn of the century
North-Eastern Rhodesia was finally under company control. By
1902 taxation was collected without opposition in all but the more
remote areas.
The process of imposing colonial rule in North-Eastern Rho-
desia, once it began, was both swifter and less violent than it had
been either to the south or in Nyasaland. It may be that the natural
disasters of the 1890s had weakened the ability of even the more
powerful peoples to resist, while for lesser groups, company rule
meant a welcome release from the raids and exactions of tribute
by the Ngoni and Bemba and the activities of slavers. Above all,
however, in this region it was the apparent lack of resources which
spared people major confrontations with colonialism: in the one
area where gold was believed to exist, the onslaught was as
dramatic as in Southern Rhodesia, and the expropriation as brutal.
In the north-west, too, little was done to give practical effect
to the treaties obtained by Rhodes's agents until the late 1890s,
despite the desire of the Lozi king, Lewanika, for a British
presence. Lewanika's 'scramble for protection' in the 1890s arose
from the same set of circumstances as his original decision to
invite the white man into his kingdom in 1886.33 As part of a
wide-ranging strategy to restore his power both internally and
externally, ideologically and materially, in that year he renewed
an invitation to the pro-British missionary, Francois Coillard of the
Paris Evangelical Mission, to settle near his capital of Lealui.
Despite the fears of those groups in his kingdom threatened by
the new religion and the increase in royal authority which
Coillard's presence brought, the missionary rapidly established
himself in the king's confidence. Over the next three or four years
Coillard aided and abetted Lewanika in his attempts to emulate
the example of his friend and ally, Khama of the Ngwato, and
gain British protection. For the missionaries in Barotseland, as
31
This section is heavily indebted to A. D. Roberts, A history of the Bemba. Political
growth and change in north-eastern Zambia before 1900 (London, 1973), 230-54.
" For the most recent interpretation of Lozi history, see G. Prins, The bidden
hippopotamus (Cambridge, 1980); this section is based on the above, and also on
M. Mainga, Buloî under the Ljtyana kings. Political evolution and stateformation in pre-colonial
Zambia (London, 1973); G. Caplan, The elites of Barotseland ii/S-if6f (London, 1970);
L. van Horn, "The agricultural history of Barotseland, 1890-1964' in R. Palmer and
N. Parsons (eds.), The roots of rural poverty in Central and Southern Africa (London, 1977),
i44~69,andW. G. Clarence-Smith,' Slaves, commoners and landlords in Bulozi, c. 1875
to 1906',/. Afr. H«/.,i979, 20, 2, 119-34.
45 3

SOUTHERN AND CENTRAL AFRICA, l886— 19IO
earlier in Bechuanaland, imperial protection and the establishment

of communications with the south were crucial.
Partly in response to Lewanika's entreaties, Frank Lochner, one
of Rhodes's agents on his way to Katanga, stopped at Lealui in
June 1890. Taking care to blur the distinction between Company
and Crown, and assisted by Coillard and Khama's emissary,
Lochner persuaded Lewanika to grant the BSA Company the
mineral rights to his kingdom. This was matched by the promise
of an annual salary of £2,000, firearms against his enemies to the
south, the Ndebele, and technical and educational assistance: the
latter a feature of all the company's treaties at this time.
Through the Lochner concession, the BSA Company claimed
the whole of what was to become North-Western Rhodesia.
Between Barotseland and the more powerful people of North-
Eastern Rhodesia, there was a broad swathe of country which was
unvisited by any company agent. Thus, Lochner had every
interest in backing the widest Lozi territorial claims, even to the
point of including the Copperbelt in Lewanika's domains. By
maximising the king's claims, the Lunda and Luvale in the north,
the Kaonde in the north-east, the Ila to the east and the Tonga
and Toka in the territory disputed with Lobengula in the south
were all brought under company rule.
Although these claims also suited Lozi expansionist plans,
bitter internal opposition to the Lochner concession, which even
led to witchcraft accusations against the king, was matched by
Lewanika's own consternation when he discovered he had signed
a treaty with the BSA Company and not with the Crown.
Through the nineties, Lewanika's opponents blamed drought,
locusts, rinderpest and famine on the concession and on the
coming of the missionaries. Yet for the first seven years there was
little to justify their fears of the company. Despite repeated pleas,
even the Resident requested by the Lewanika was not sent.
Indeed, in these years, Lewanika was able to build up his personal
power through the consolidation of the administrative, political
and economic reforms he had instituted in the 1880s, and to extend
his rule over outlying parts of the empire, which supplied the Lozi
aristocracy of the heartland with tribute and slave labour, cattle
and ivory. Nevertheless, by the mid-nineties Lewanika clearly
recognised the fragile basis of his power in relation to the
encroaching Europeans. He welcomed the defeat of the Ndebele
454

LAND, LABOUR AND TAXATION
in 1893; but the advent of white prospectors from the south, the
appearance of Portuguese forts in territory claimed by the Lozi,
and the failure of the French missionaries to provide the firearms,
skills and education he had anticipated, led Lewanika to appeal
once more for British protection.
Finally, in 1897 Major R. T. Coryndon was sent — more because
the British government was anxious that the BSA Company
should uphold its claims in the area against the Portuguese than
because of company guilt over unfulfilled obligations. In 1898,
the Lawley concession gave the company even greater jurisdiction
over Bulozi than the Lochner treaty, though it met with less
popular resistance. Unlike the first concession, however, the
advent of Coryndon, the Lawley concession and the British
Order-in-Council of 1899, which granted ultimate sovereignty to
the British monarch, together with a further concession by
Lewanika in 1900, spelt the end of Lozi independence. Over the
next decade the powers of the Lozi aristocracy, if not of the king,
were whittled away. British insistence on the abolition of serfdom
and slavery in Bulozi in 1906 - in part a response to increasing
slave resistance, in part to release labour for white enterprise in
the south — had serious repercussions for the cultivation of the
flood plain on which Lozi agriculture depended. For the thousands
of emancipated slaves, migrant labour to Rhodesia and the Rand
may have been preferable to bondage to the local ruling class.
Lewanika's 'scramble for protection', the shrewdness with which
he manipulated the benefits brought by the missionaries and the
lack of mineral resources in his domain meant he secured a
protected status from a company weakened by its experiences
further south. Nevertheless, his hopes of controlling the modern-
isation of Bulozi, symbolised perhaps in the brief appearance of
an independent school and church movement in Bulozi under
royal charge in 1904-5, were in vain. Bulozi passed from an
independent kingdom to a protectorate within a protectorate, as
much tied to the southern African political economy as its
neighbours.
All over southern Africa, the imposition of colonial rule meant

the imposition of taxation to raise much-needed revenue for
administration and to force people out to work. This was a
455

SOUTHERN AND CENTRAL AFRICA, 1886-1910
well-worn expedient within South Africa, where in both the

Afrikaner republics and the British colonies, taxation on Africans
had been levied and periodically increased during the nineteenth
century, especially in response to labour demand. The same was
true north of the Limpopo. Even north of the Zambezi, and
despite the small number of settlers and the limited scale of capital
investment, the demand for labour was vast, especially until the
construction of railways relieved the need for human porterage.
In Southern Rhodesia, the Native Affairs Department was
specifically established to collect tax, and BSA Company police
enforced payment even before they had a legal right to do so. In
Northern Rhodesia, too, taxation had been imposed everywhere
by 1902, although it took several years before it was collected
regularly over the entire area. In the Shire highlands, Johnston
had imposed an annual hut tax of 6 shillings as early as 1892,
although he was soon forced to lower it to 3 shillings in view of
the lack of employment and the very low wages paid in Nyasaland.
In 1901 hut tax was doubled once more, except for those Africans
who could prove they had been in white employ for thirty days.
In British Central Africa, Africans working on the plantations
or in the mines were frequently met by tax collectors as they
received their monthly wage packet; although it was illegal, many
employers paid their workers' tax directly to the district
commissioner. Taxation both disguised and facilitated chibaro,
forced labour practices, as tax defaulters were despatched by local
officials to work on the mines in Southern Rhodesia or the
plantations of the Shire highlands. Until the longer-term mech-
anisms of landlessness, growing consumer wants and new work
discipline had been established, collectors extracted tax through
overt violence. While the starkest example of this was in Southern
Rhodesia before the rebellion, in British Central Africa, too,
missionaries were outraged by the frequent trail of hut-burnings
and even shootings that accompanied tax collection. In 1901,
when Alfred Sharpe, Johnston's successor as commissioner of the
British Central Africa Protectorate, precipitately raised the rate,
many people moved across the boundary into Mozambique,
reversing the usual pattern of migration. In Northern Rhodesia,
as late as 1907—8, there was serious danger of tax resistance
amongst the Ila and outright opposition amongst the Lunda.
Perhaps the most striking example of resistance to taxation
456

came in 1906, in Natal and Zululand, where the government had

imposed a ^ i poll tax on all adult males, over and above the
already existing burden of taxation. The tax fell particularly and
deliberately heavily on Africans who were being squeezed by
their landlords, evicted from newly developing wattle plan-
tations, and hit by a combination of drought and depression.
Heavy indebtedness and landlessness were already a feature of the
Natal countryside, and thousands of Africans were being forced
onto the labour market. In a very real sense, resistance to the
poll tax was resistance to proletarianisation. When magistrates
attempted to collect the tax at the beginning of 1906 they were
met by sporadic acts of defiance, which led to the declaration of
martial law, and large-scale troop movements. These may well
themselves have sparked off the armed resistance led, in its main
phases, by the minor Natal chief, Bambatha, and encouraged by
the use of the name and authority of the son of the last Zulu king,
Dinuzulu. When martial law was finally lifted in September 1906,
between 1,000 and 4,000 Africans had been killed and 7,000 held
prisoner. White losses amounted to some thirty: striking evidence,
if it were still needed, of the superiority of white firepower over
unarmed Africans in early twentieth-century South Africa.
While tax provided revenue for colonial administration in the
face of the British determination that colonies be self-financing,
its effects in forcing people out to work were equally important.
To a certain extent, in those areas where Africans still had access
to sufficient land and to markets, they were initially able to meet
tax demands by producing crops and cattle for sale. Yet there was
none of the debate in southern or central Africa about the
possibilities of extensive peasant production which characterised
East and West Africa slightly later on. In South Africa, by the turn
of the century the prosperous eastern Cape peasants were already
'passing through a period of stress', although the demand for
provisions, horses and cattle as well as transport during the South
African war opened up new opportunities both there and
elsewhere for black producers. In South Africa and Southern
Rhodesia, agricultural production for the market, stratification
amongst the peasantry, and the growth of a relatively wealthy
stratum, continued well into the twentieth century.34 Its incidence
34
For the South African peasantry, see C. Bundy, The rise and fall of the South African
peasantry (London, 1979), especially 109-64;forCentral Africa, see Palmer and Parsons
(cds.), Roots of rural poverty; for a critique of the literature and evidence for the continued
457

was uneven, however, and increasingly it was on the fringes of

white agriculture and in the face of the indifference, if not the
outright opposition of the state.
In British Central Africa, despite the paucity of settler agri-
culture, there was no greater encouragement given to African
peasants. An attempt by Sharpe to encourage African cotton
production had little success in this period as a result of planter
hostility and the demand for mine labour in the south. Contrary
to colonial ideology, which held that true progress depended on
individual land-title in male hands, agricultural production
expanded amongst matrilineal people like the Manganja and
Chewa of in the Shire valley, especially after Lomwe immigration
from Mozambique relieved some of the pressure on local peoples
from the plantations. They were amongst the few successful
peasant cultivators in the region, although in Northern Rhodesia
the Ila were able to take advantage of the line of rail to produce
for the market and in the north the opening of the Katanga copper
mines opened new opportunities of food production.
Elsewhere in Central Africa, the administrative attack on
'slavery', taxes on iron hoes, and the prohibition on African
hunting and possession of firearms probably reduced food pro-
duction. In North-Eastern Rhodesia, the natural disasters of the
1890s were compounded by the absence of young men as labour
migrants, the administrative insistence on closer settlement and
gun and game laws which led to the proliferation of wild game
while denying the people the meat. Tsetse-infected wild animals
spread sleeping sickness amongst both humans and cattle, and this
reached epidemic proportions in some areas. Even where there
was no outright expropriation of land, large numbers of young
men were forced to seek work beyond the homestead, and already
in the first decade of the century, widespread social and eco-
logical dislocation was in evidence, further accelerating labour
migrancy.35
Everywhere it was assumed that capital and enterprise were the
prerogative of the white man; the role of the black was to provide
labour. In Northern Rhodesia, a trickle of white settlers was
prosperity of at least a section of the peasantry into the mid-twentieth century, see
T. O. Ranger, 'Growing from the roots: reflections on peasant research in Central and
Southern Africa', / . Southern Afr. Studies, 1979, 5, i, 99-1 j5.
35
L. Vail, 'Ecology and history: the example of eastern Zambia', / . Southern Afr.
Studies, 1977, 3, 2.
458

encouraged by the BSA Company in the hope that their presence

would lead to a rise in land prices, and between 1898 and 1908
the concessions granted by Lewanika enabled it to allocate land
in Ila-Tonga country and along the line of rail. Hopes of copper
at Kansanshi and lead and zinc at Broken Hill (Kabwe) led the
company to extend the railway into the colony, but only attracted
a handful of prospectors. Nor were early plans for cotton and
rubber plantations or cattle-ranching any more successful; by
1911 there were some 1,5 00 white settlers in the territory. The
lack of settler success did not, however, lead to any encouragement
of African agriculture. From the outset, Northern Rhodesia was
seen as an appendage of the south, and any alternative to labour
migration was discouraged.
In Nyasaland the declaration of a Protectorate in 1889 had been
accompanied by land speculation on a large scale, and by 1891 a
number of coffee planters had established themselves to the south
of Lake Malawi, closely tied to major trading interests. Although
Johnston claimed to discourage wanton speculation in land and
to prevent African dispossession, he never queried the right of
chiefs to alienate their people's land. The clause he inserted in
the title deeds granted to planters, guaranteeing that African
cultivators should be undisturbed in their occupancy, was a
dead-letter from the outset. In all, he alienated 3.5 million acres
of land in the Protectorate, 2.7 million to the British South Africa
Company and the African Lakes Company in the north, the rest
in the fertile but already densely settled Shire highlands. Sharpe
allocated a further 360,000 acres for railway construction, and
100,000 acres to settlers — again in the overcrowded south. At
the beginning of the century, only 1 per cent of these lands was
under cultivation.36 Africans who remained on the alienated land
were forced to render labour in lieu of rent in the hated thangata
system, and the planters scoured the rest of the country for
additional workers.
White enterprise was equally unimpressive in Southern Rho-
desia. There as we have already seen, huge tracts of land had been
alienated by the turn of the century, amounting to about one-sixth
of the territory's 96 million acres. Most belonged to speculative
syndicates, and relatively few settlers were actively engaged in
36
B. S. Krishnamurthy, 'Economic policy, land and labour in Nyasaland,
1890-1914', in B. Pachai (ed.), The early history of Malawi (London, 1972), $85-8.
459

agricultural production. In the early days, most continued to

depend on Africans for their food supply, which they obtained
largely through trade. As long as the BSA Company believed the
future lay with mining, it did not pay much attention to the
territory's agricultural potential, although it clearly hoped for a
general rise in land value. By 1903, however, disillusionment with
the ' second Rand' had brought about a collapse in the market for
Rhodesian shares. The mines were scattered and the ore poor, and
although gold mining remained central to the economy for a
further thirty years, the company now decided to encourage white
immigration and farming more actively. In 1906 it set up an
estates department to do so, and developed agricultural and
veterinary services to assist and advise white farmers, while by
1908 settlers were given a greater say in the legislative council
which had been established in the aftermath of the Shona-Ndebele
risings.
Although it is frequently asserted that the touchstone of
imperial trusteeship in this period was the protection of African
land rights against settler encroachment, even in those areas south
of the Limpopo subject to direct imperial control the record is
far from unambiguous. In Basutoland, it is true, whites were
specifically barred from acquiring land, a reflection of successful
Sotho resistance in the Gun war of the early eighties against Cape
colonial rule. In the Bechuanaland Protectorate, too, relatively
little land was lost to white settlers, apart from small areas around
the Tati goldfields and in the Ghanzi district. Fortuitously and
fortunately for the northern Tswana, Rhodes was unable to
establish company rule in the Protectorate — which he claimed as
part of his Charter — before the fiasco of the Jameson Raid, and
the Shona—Ndebele rebellion put paid to his ambitions in the
region. For both Basutoland and the Bechuanaland Protectorate,
these were relatively quiet years in which the ability of richer
peasants to provision South African and Rhodesian markets
respectively, and the alliances forged between the leading chiefs
and the British administration and missionaries, masked the
longer-term processes of growing social stratification and depen-
dence on the South African labour market.
In British Bechuanaland, annexed as a Crown colony in 1885,
the case was rather different. By the time of the British take-over
it was already the site of a complex struggle over land. The
460

southern Tswana had lost three-quarters of their land by the 1884
London convention, while a land commission held in 1886 did
little to resolve their land shortage. Ten reserves were marked out
as Tlhaping territory, while the intervening 14,000 square miles
were made available for white settlement and bought up by
white farmers in the decade between British and Cape colonial
annexation. Cape annexation and the renewed demands from
white settlers for land and labour increased the insecurity of the
southern Tswana chiefs, embittered by their loss of land, and by
the impact of imperial rule. The contrast was particularly galling
for those southern Tswana who had been able to avoid work on
the diamond fields by producing fuel and food for the Kimberley
market. As their tax burden increased, so their capacity to lead
an independent existence diminished. In December 1896, under
the Tlhaping chiefs, Galeshiwe and Luka Jantje, who had led
resistance in Griqualand West in 1878, and Chief Toto of the
Tlharo, the southern Tswana in Phokwane and the Langeberg
rose against Cape Colony rule, incensed finally by the shooting
of their cattle during the rinderpest outbreak. It took over 2,000
well-equipped Cape Colony troops to bring the rising to an end.
In August 1897 the southern Tswana chiefs found their lands
confiscated, their cattle looted and their people dispersed as
indentured labour to the farmers of the western Cape.37
In two other areas the British sanctioned the outright expro-
priation of African land. In 1897, when Zululand was finally
incorporated into self-governing Natal, it was agreed that after
a five-year moratorium a joint Imperial—Natal delimitation com-
mission should set aside reserves for African occupation, releas-
ing the rest for white enterprise. Between 1902 and 1904, some
two-fifths of Zululand was opened to white farmers from Natal
to produce sugar, wattle and cattle - a potent factor in persuading
the people of southern Zululand to join Bambatha in 1906.
Although white administrators in neighbouring Swaziland
feared African resistance to land loss, the Swazi also lost much
land during the first decade of the century. There, as a result of
Mbandzeni's concessions, the question of land ownership was the
thorniest problem facing the British administration which took
the territory over from the Boers after the South African war.
37
K. Shillington, 'Land loss, labour and dependence: the impact of colonialism on
the southern Tswana, 1870-1900', Ph.D. London, 1981; H. SakerandJ. Aldridge.'The
origins of the Langeberg rebellion',/. Afr. Hist., 1971, 12, }.
461

Despite the prolonged struggle of the Swazi, led by the queen
regent, who protested that the king had granted only the usufruct,
in 1907 two-thirds of Swaziland was recognised as belonging to
foreign concessionaires and only one-third to the Swazi them-
selves. There were few colonial officials who did not agree that,
for the territory to be developed, European property rights had
to be secured.
Within the existing British colonies and Boer republics, these
years saw the rapid increase of settler populations and the
consolidation of areas of white settlement distinct from African
reserves or locations. Despite the insatiable land-hunger of white
farmers, there was no dramatic transfer of land. Everywhere there
was strenuous opposition by whites to the maintenance of tribal
lands and communal ownership; it was expressed forcibly before
the Cape Labour Commission of 1893, the South African Native
Affairs Commission of 1903-5 and the Natal Native Affairs
Commission of 1907-8. Yet by and large fears of African
resistance, together with the countervailing interests of land
companies and industry, prevented wholesale expropriation. In
the Cape it was believed by liberals that the introduction of
individual tenure, private land ownership and production for the
market would inevitably push out surplus labour, and this was
given full expression in their support for Rhodes's well-known
Glen Grey Act of 1894, which he significantly termed 'a Bill for
Africa'.38 This Act divided the land of that district into 4-morgen
holdings, imposed a 10-shilling tax on men who could not prove
they had been in employment for three months in the year, and
established local councils in lieu of African participation in the
central representative institutions of the state. Individual tenure
was opposed however by most non-Christian Africans and
chiefs, as well as by white farmers, who feared the creation of a
competitive peasantry, and those colonists who connected land
rights with access to the franchise.
Rhodes and his fellow mining-magnates could tolerate the
existence of a limited number of African peasants: as Rhodes
recognised, the subdivision of African lands into peasant allot-
ments would necessitate the proletarianisation of the less fortu-
nate, a process already well in evidence in the Cape Colony. As
18
For a discussion of the Glen Grey Act which sees it as the basis for later policies
of segregation, see M. Lacey, Working for horoko (Johannesburg, 1981), 14-17.
462

he remarked in his introduction to the Act, '... every black man

cannot have three acres and a cow, or four morgen and commonage
rights... It must be brought home to them that in future nine-
tenths of them will have to spend their lives in daily labour, in
physical labour, in manual work.'39
Yet if there was only partial conflict between the interests of
the peasantry and those of the capitalist sector who wanted labour,
there was total conflict between African peasants and white
farmers as agricultural production became commercialised, and
they competed for land and markets. During this period, however,
this conflict was not always fully evident. In many areas, where
white farmers were still establishing themselves, they depended
on the Africans on their land for labour, and at times for food.
In the early days, settlers, notwithstanding their claims to land,
made their living through trade and extracting rent and were in
no position to eliminate ' squatting \ 4 0 The large land companies
were often equally happy to retain the rents of the Africans living
on their newly acquired acreages; where, as was often the case,
the land companies were the subsidiaries of mining houses, there
was a triple bonus, as the homesteads provided not only rent, but
male migrants and cheap female food-producers.41
In areas of rapid white population growth and commercial
agriculture, however, land prices boomed and companies began
to sell off some of their land: in Natal, for example, especially in
the midlands, numerous rent-paying tenants were displaced. As
agriculture became directed to the markets of the Rand and
Kimberley, small farmers all over South Africa appealed to
governments to attack the 'evils' of'kaffir-farming', and pushed
for labour-tenancies rather than rent-tenancies. This period wit-
nessed a spate of anti-squatting legislation: in the South African
Republic in 1887 and 1895; in the Orange Free State in 1893; in
Natal in 1896; in the Cape in 1899 and 1909, and even in Southern
Rhodesia in 1908. Of all these attempts, it was only the Cape's
Private Locations ordinances which were really successful. This
39
30 July 1894 in ' Vindex', CecilRhodes: bis political life and speeches (London, 1900),
)82.
40
J. K. Rennie, 'White farmers, black tenants and landlord legislation: Southern
Rhodesia, 1890-1950',)'. Southern Afr. Studies, 1978,5, 1; T. Keegan,' The restructuring
of agrarian class relations in a colonial economy: the Orange River Colony, 1902-1910',
/ . Southern Afr. Studies, 1979, 5, 2, 254-54.
41
S. Trapido, 'Landlord and tenant in a colonial economy: the Transvaal,
1 8 8 0 - 1 9 1 0 ' , / Southern Afr. Studies, 1978, J, 1, 26-58.
463

was a reflection of the greater development of capitalist relations
of production in the Cape and the concomitant development of
its state apparatus, for ' anti-squatting legislation made very little
impact unless there was an already existing capacity to compel
labour to work \ 4 2 Outside the Cape the opposition of the more
powerful farmers and absentee landowners rendered much of
this legislation inoperative; nevertheless, it provided precedents
which were of the utmost significance in the decades to follow.
It was in South West Africa that the processes of expropriation

and proletarianisation were worked out with the greatest ferocity
— though the contrast with the rest of South Africa was in degree
rather than in kind. There, through the 1890s, Africans had seen
the arrival of settlers who eyed their lands and cattle greedily, and
whose arrogance was unchecked by legal or civil institutions.
Although the actual amount of land alienated was not large, it was
situated in the better-watered highlands, for long contested
between Nama and Herero pastoralists. The activities of the
speculative companies, too, spread unease, though again it was
the rinderpest outbreak which destroyed African herds and
brought the Herero, at least, to the edge of despair. Widespread
indebtedness was a feature of these years, as Africans attempted
to replace their lost cattle and had to purchase food for the first
time. The misguided attempts in 1902-3 of Governor Leutwein
to place a moratorium on debts, after giving creditors a year in
which to collect outstanding sums, led to savage extortion and
looting and simply exacerbated the situation.
At the beginning of 1904, when the militia was in the south
stamping out a small disturbance amongst the Bondelswartz
people, the Herero rose under Samuel Maherero, and killed over
a hundred German settlers and traders. The Germans retaliated
in full-scale war, under the ruthless General von Trotha, who,
supported by the German High Command, saw the campaign
quite simply as one of extermination. Before his recall by a
horrified German Reichstag in mid-190 5, over 70 per cent of the
Herero had been killed, many of them dying in the desert in an
attempt to escape to the Bechuanaland Protectorate. The death-rate
was almost as high amongst the Nama, who joined the uprising
in October 1904 under the eighty-year-old Hendrik Witbooi.
« Ibid., 44.
464

LABOUR ON THE RAND
Despite Witbooi's death in 1905, the Nama, now led by Jacob

Marenga, were only finally vanquished in 1907; thousands died
in concentration camps. Virtually all Herero cattle perished, while
of the remaining human population, about a quarter were
deported to other parts of the country. At the end of the war, the
peoples who participated were deprived of all their remaining
lands and cattle, which were sold to settlers. All forms of
chieftainship were proscribed, and the people were forbidden to
rear cattle, wear tribal dress or meet in groups: even the Christian
revivalism which followed in the wake of the wars was frowned
upon as evidence of nationalist sentiment. Paul Rohrbach, a
well-known spokesman for settler interests, wrote in 1907 that the
objective was to ensure that' as far as possible [the Herero] should
be stripped of his national identity and his national characteristics
and gradually amalgamated with the other natives into a single
coloured working class'.43 By 1912 only 200 Nama and Herero
males were without paid employment.44
The sheer devastation of the war, which carried dispossession
to its logical conclusion, served as a warning elsewhere in
southern Africa. Ultimately, von Trotha had to be stopped
because he was killing off the settlers' labour supply; indeed to
complete their conquest and build railways and harbours after the
war, the Germans were forced to import labour from the eastern
Cape, and to look for the first time to the Ovambo people to the
north, who had so far been only marginally affected by the German
presence. Although after the Herero revolt the Reichstag
prevented the conquest of the north, by 1908 with the help of the
Finnish and German Lutheran missionaries in Ovamboland,
treaties had been signed with all the chiefs, persuading them to
send migrant labour to the mines and construction works of what
became known as the Police Zone in the south. The Ovambo of
South West Africa were only brought fully under colonial rule
in 1915 after the South African invasion of the colony.
LABOUR ON THE RAND
By the beginning of the twentieth century fierce African resistance

to the alienation of land was a major factor all over southern Africa
in persuading colonial authorities and employers that industry
43
Bley, South West Africa under German rule, 224. •• Ibid., 50.
465

S O U T H E R N AND CENTRAL A F R I C A , 1886—1910
would, in the first instance, have to be based upon migrant labour.
Yet if Africans could not simply be divorced from the means of
production, these years saw a network of laws and institutions
designed to push them on to the labour market, and to keep them
on the treadmill for as long as possible. Everywhere, breach of
contract became a criminal offence under newly designed Masters
and Servants laws. Pass laws, which had originated in the agrarian
slave economy of the Cape, were now revitalised and reformulated
to cope with the problems raised by industrial development. In
the rural areas of southern Africa they were advocated by farmers
anxious to prevent the movement of their labourers into the
towns, and to prevent unauthorised trespass. It was, however, on
the Rand that they received their greatest elaboration as an
instrument of control over the mine labour force. As C. S. Gold-
man told the 1897 Industrial Commission of Inquiry, the Pass
law' which has been condemned as an instance of Boer oppression
of the natives... came into existence at the request of the mining
industry'. 45
At the insistence of the Chamber of Mines, under law 31 of
1896, the Republic's Volksraad (the legislature) provided that
Africans entering the mining areas had to have passes authorising
them to seek employment for three days. Once employed, the
worker handed his pass to his employer until he was discharged,
and was provided with an ' employers' pass in exchange: failure
to produce a pass on demand rendered an African liable to a £3
fine or three weeks' hard labour for a first offence, and a rising
scale of penalties for any offence thereafter.
As the pass laws themselves indicate, even the conquest and
annexation of the remaining independent African polities of
southern Africa did not fully answer the new demand for labour,
or at least not quickly or cheaply enough. Throughout the 1890s
wage levels on the mines fluctuated in response to supply and
demand. As early as 1893 the Chamber of Mines toyed with
schemes for recruiting from Mozambique, which had provided
migrant labour for colonial enterprise from the 1850s. In 1896
it formed its first centralised recruiting agency, which became
known in 1900 as the Witwatersrand Native Labour Association
(WNLA, dubbed by Africans 'Wenela'), to obtain labour from
southern Mozambique on a regular basis. By the turn of the
45
C. S. Goldman, evidence to the 1897 Industrial Commission.
466

LABOUR ON THE RAND
century almost two-thirds of the labour on the Rand came from
Mozambique. There the Portuguese defeat of Gungunyana's Gaza
kingdom in 1895, and the passage in the following year of a labour
ordinance making it obligatory for Africans to work for colonists,
proved admirable recruiters for the mines. The low cost and
longer contracts of Mozambique labour made it additionally
attractive.
If the defeat of the Gaza kingdom was fortuitous, the Chamber
of Mines was not slow to appreciate the beneficial effects of natural
disaster for its labour supply. The equation was simple: in years
of abundant harvest, the flow was slow; in bad years it was more
forthcoming, and wages could be reduced with little immediate
effect on supply. All over southern Africa, the rinderpest epidemic
played a major role in propelling Africans to the mines: in
Pondoland, for example, only annexed to the Cape Colony in 1894,
and long resistant to the appeals of the labour recruiter, migration
to the mines now began in earnest - in exchange for advances of
cattle paid to heads of homesteads.46 Yet without the broader
structural changes taking place, even this would have been only
a temporary windfall for the mining industry, while Africans
earned enough to replace their herds. That it was not so must be
attributed to the longer-term processes advocated by the
chamber and slowly implemented by newly formed colonial
administrations.
By the 1900s, for Africans all over the sub-continent forced to
seek work, the great magnet was the Witwatersrand. In the early
years of the century, WNLA recruiters honeycombed the sub-
continent, while in 1903 the Rhodesian Native Labour Association
was formed for the same purpose. The undercapitalised, scattered
and small-scale mines of Southern Rhodesia found it difficult to
compete with the Rand, and they resorted to a variety of forced
labour practices to ensure labour remained in Southern Rhodesia
and to keep down its costs. The predictable results were high rates
of mortality, disease and desertion. So desperate were the
Rhodesian mines for labour that they even thought of scouring
Asia for miners, following the example of the Rand in 1904.47
Within the South African colonies, Africans had rather more
46
W . Beinart, 'Joyini I n k o m o : cattle advances and the origins o f migrancy in
Pondoland', / . Southern Afr. Studies, 1979, 5, 2.
47
C. van Onselen, Cbibaro. African mint labour in Southern Rhodesia, 1900—rf))
(London, 1976).
467

SOUTHERN AND CENTRAL AFRICA, I 886— I 9 I O
choice about where to work and what work to do, and historical
patterns of labour-seeking had already been built up. For most
people in Nyasaland, however, there was only the poorly paid
work and harsh conditions on the plantations of the Shire
highlands or, until the building of the Nyasaland railway, head
porterage; in Northern Rhodesia there were even fewer local job
opportunities, apart from limited mining at Kansanshi, Broken
Hill and Kasempa. Increasingly central Africans made their way
west to the Katangan (Shaban) copper mines, or south - either
to the mines of Southern Rhodesia or preferably to the Rand itself,
much to the consternation of local employers. Nevertheless,
despite local opposition, by 1903 the government of Nyasaland
had accepted institutionalised labour-recruiting for the export of
contract labour to the mining industry in the south. In that year
over 6,000 recruited workers crossed the Zambezi at Feira to work
in the south, of whom 1,000 were on their way to the Rand; by
1904 the numbers recruited for the Rand had risen to 5,000 — quite
apart from those who made their own way south. By comparison
with the three to six shillings a month offered to workers in the
Nyasaland Protectorate, the 30s. for a thirty-shift month on the
Rand mines was a powerful draw. In 1907, the Protectorate
government, appalled by the high mortality on the gold-mines and
pressed by local planter interests, prohibited further recruiting;
the recruiting posts simply moved across the frontier and the
stream continued unabated. By 1910 there were some 20,000
Nyasalanders in Southern Rhodesia, and a smaller but still
substantial number on the Rand, determined to sell their labour
on the most favourable labour market.
Control over the supply of labour was one problem; reducing
its cost was another — and reducing the costs of labour was of
the essence on the Rand, as in Rhodesia. In addition to the high
operating costs and geological constraints involved in deep-level
mining on the Rand, the mine magnates faced two further
problems. Compared to Australia and Canada, the yield of South
African ores was low. At the same time, because of the ceiling
on the price of gold on the world market, the costs of extraction
could not be passed on to the consumer. The cost of imported
stores was beyond the magnates' control, and constantly rising.
The crucial area of cost reduction was thus seen to be in workers'
wages — and more specifically black workers' wages — which were
by far the largest single item of expenditure.
468
LABOUR ON THE RAND
By 1899, in addition to the one million black workers there were

some 12,000 white workers in the Rand and, ideally, the magnates
would have liked to cut the wages of all workers. The increased
complexity of deep-level mining technology after 1890 had
created a demand for highly skilled labour with experience of
hard-rock drilling underground. While an unskilled labour force
could be prised out of the pre-capitalist social formations of
southern Africa, the immediate need for skilled labour could only
be met by attracting workers from the United Kingdom, Australia
and the United States of America, by the lure of high wages. The
cost of living in the interior was extremely high and most miners
migrated in the knowledge that their working days would be
short. Because of the very high concentration of silica in both the
lode and the ore on the Rand the incidence of miners' phthisis
was perhaps the highest in the world: in 1904 the average age at
death of Cornishmen who had worked on rock drills in the
Transvaal was 36.4 years, with an average period of rock drill
employment of only 4.7 years.48
While most white workers had a knowledge of craft union
organisation and strategy, some came to the Rand with the vivid
collective memory of the assault on the working class in the
diamond mines of Kimberley. Fully aware that skills can be learnt,
and that the mine owners had every incentive to find less
expensive substitutes, the white miners on the Rand — mostly, like
their African counterparts, single men without their wives and
families — organised to protect their position. Indeed, the first gold
mineworkers' union, the Witwatersrand Mine Employees' and
Mechanics' Union was formed in 1892 to protest 'against the
attempt of the Chamber of Mines to flood these fields with labour
by means of cheap emigration' from the United Kingdom.49 But
they perceived the greatest threat as coming from the African
miners working under them. Kimberley had already shown that
- where the magnates had control of the state - the compound
system, labour migrancy and pass laws ensured that black labour
was both cheaper and more easily controlled than its white
counterpart, and that capital had every incentive to try to
substitute black labour for white. Trade unionists clearly recog-
nised the threat, and although one cannot discount the heavy
*8 G. Burke and P. Richardson,' The migration of miners'phthisis between Cornwall
and the Transvaal, 1876-1918", J. Southern Afr. Studies, 1978, 4, 2.
49
H. J. Simons and R. E. Simons, Class and colour in South Africa,
( H a r m o n d s w o r t h , 1969), 53.
469

SOUTHERN AND CENTRAL AFRICA, 1886-1910
overlay of late-nineteenth century racism in their utterances, the

failure of a single, trans-racial working class to emerge at this time
related largely to their structural vulnerability.50
Throughout the 1890s various attempts were made by white
unionists to protect their position through the reservation of
certain jobs for whites only, but with only limited success. At the
outbreak of the South African war, engine-driving was the sole
occupation barred to blacks by statute. In general white, English-
speaking workers still had a monopoly of skilled and supervisory
jobs on the Rand, although the industry attempted, with varying
degrees of success, to limit the number of white underground
workers. The scarcity value of white workers and their experience
of class organisation, as well as the need of the mine magnates
for working-class support in their conflict with the South African
Republic, meant that the full thrust of cost reduction was to be
borne by the black workers. In 1888 mine managers were already
petitioning the Volksraad for a pass system, monthly contracts
and the registration of all Africans in the Transvaal. The first
annual report of the Chamber of Mines, in 1889, complained
bitterly of competition between mine managers for labour which
led to overt attempts to bribe workers to desert their employers.
It deplored the fact that there had been 'a steady rise in wages
all round, which is adding a very heavy additional expense to the
working of the mines'.51
Throughout the 1890s, mine magnates were dissatisfied with
the numbers of black workers they were able to secure, the wages
they were forced to pay, and the degree of control they were able
to exert. At every point, they blamed the failure of the South
African Republic to come to their assistance. Even where the state
passed the legislation requested by the industry, as in the case of
the pass laws, the inefficiency of the police force meant that it
was ineffectively carried out, while the power which large
landowners had over the state apparatus at a time of rising labour
demand in the agrarian sector meant that Africans making their
way to the gold-fields both from within and from outside the
Republic had to ' run a gauntlet offieldcornets set on acquiring
50
Johnstone, Class, race and gold, 49-75; and R. H. Davies, Capital, state and white
labour in South Africa, rpoo—ifio (Brighton, Sussex, 1979), 56-60.
" Transvaal Chamber of Mines, First annual report, Statement of the Chairman,
31 Dec. 1889 (Johannesburg, 1889).
470

LABOUR ON THE RAND
labourers for themselves and their fellow burghers'. 52 Moreover,

once the worker managed to reach the Rand, the state seemed
unable or unwilling to prevent his desertion, or, almost as
disastrously for the industry, his incapacitation through drink.
The liquor question, indeed, stood at the heart of a complicated
wrangle both within the industry and between the Chamber of
Mines and the Kruger government. Initially, both because it
appreciated the social and economic control which the manipu-
lation of alcohol gave it over the black worker, and because of
the involvement of its own members in the distillation and sale
of liquor, the chamber was reluctant to advocate total prohibition.
Nevertheless, by 1895 the demand for a stable and efficient
work-force took precedence over these concerns — crucially, just
as the deep-level mines were coming into production. In that year,
the annual report of the Chamber of Mines complained bitterly
of the 15-25 per cent of the black labour force disabled by drink,
a factor which added significantly to their wage bill. Despite Act
17 of 1896 providing for 'total prohibition', the next few years
saw little improvement: illicit liquor syndicates, with the tacit
connivance of the police force and the President himself, simply
took over where the licensed canteen-holders were forced to leave
off.53 Again, this constituted a major grievance of the magnates
against the Kruger government - and one that was only resolved
with the imperial annexation of the Transvaal.
Observing the outcry of the magnates in the years leading up
to the war on the issue of labour, the British radical, J. A. Hobson,
a close observer of the South African scene at this time, was moved
to comment that 'put in a concise form' the South African war
was 'waged in order to secure for the mines a cheap adequate
source of labour'.54 Important as this insight was, the contradic-
tions between the Kruger state, the mining industry and imperial
Britain which led to the Jameson Raid and the South African war
went beyond the issue of cheap labour. They were related to the
whole cost-structure of the industry, the nature of the state in
52
S. T r a p i d o , ' Reflections o n land, office and wealth in the South African Republic,
18 50—1900', in S. Marks and A. Atmore, Economy and society in pre-industrial South Africa
(London, 1980), 361.
53
This paragraph is heavily indebted to C. van Onselen, 'Randlords and Rotgut,
1886-1903', in his Studies, 1: New Babylon, 44-102, originally published in History
Workshop, 2, 1976.
54
J. A . H o b s o n , The war in South Africa ( L o n d o n , 1900).
471

which it was established, and the stability of the region on which

it depended, as well as Britain's position in the sub-continent and
as a world-power.
THE CONQUEST OF THE SOUTH AFRICAN REPUBLIC
For the burghers of the South African Republic, the sudden

intrusion of the vast quantities of capital and of alien immigrants,
known as uitlanders, constituted a crisis of the severest order. It
was not just that the speculating, gambling, drinking and whoring
of a brash bachelors' town offended the sensibilities of Calvinist
burghers and their fundamentalist president, Paul Kruger. The
usual picture of Kruger and his burghers as obstinately standing
in the way of change is a misleading one: as early as 1883, he
announced a policy of developing the resources of his country
through selective industrialisation based on concessions, and a
number of capitalists like the Hungarian, Nellmapius, and the
East European Jews, Lewis and Marks, made their fortunes by
developing enterprises which were of some benefit to Transvaal
producers.55 The demands of the mining revolution, however,
went well beyond state-controlled concession. Quite suddenly,
burghers found their whole way of life was being threatened, both
at the material level and in all its assumptions. For many it meant
a painful process of proletarianisation, as they lost their land
through drought and disaster compounded by the growing
capitalisation of agriculture; the possibilities of an independent
livelihood as transport riders or petty entrepreneurs also dimin-
ished as mining capitalists themselves expanded into transport and
construction.56 Moreover, despite its heterogeneous origins, the
uitlander community had behind it the resources of the British
state. Even without the previous half-century of friction and
suspicion between the South African Republic and the British, the
situation would have been difficult; under the circumstances,
Kruger's actions in defending his republic were by no means as
reactionary as contemporary antagonists proclaimed.
There were, however, limits to what Kruger could do. Essent-
ially he tried to use the opportunities opened up by the mineral
discoveries to enrich his fellow-notables in the republic, while
" Trapido, 'Reflections on land, office and wealth', 360.
56
Van Onselen, Studies, 1: New Babylon, 19-20, and chapter 4; 11: New Nineveh,
111—24.
472

CONQUEST OF THE SOUTH AFRICAN REPUBLIC
limiting their impact by controlling policy in five main areas:

seeking alternative foreign alliances; encouraging industrialisation
through the grant of concessions; developing an independent
railway and tariffs policy; expanding territorially; and excluding
the uitlanders from effective political power. The problem was that
Kruger's first four strategies brought him into direct conflict with
British imperialism, the mine magnates and the Cape Colony,
while hisfifthprovided an excellent excuse for British intervention
on behalf of its ' rightless' subjects.
Each of Kruger's strategies had its earlier roots, but they now
acquired a new shape and significance. Even before the discovery
of the Rand, the politics of railway construction and tariff policies
had held the centre of the stage in the relationships of the two
British coastal colonies to the land-locked interior. In 188 5 Kruger
had offered the Cape a customs agreement which the colony had
short-sightedly turned down. With the discovery of gold on the
Witwatersrand, the tables were turned and, from 1887, the
competition to reach the new Rand market by rail became
increasingly intense. The competition was not simply between the
coastal colonies and the Transvaal-backed, German-financed,
Netherlands-South Africa Railway Company (NZASM) line
from Delagoa Bay - the natural and nearest outlet for the
Transvaal. The bitterness spilled over between the coastal colonies
themselves, and even between the three main Cape ports. It was
further complicated by Rhodes's attempts to buy Delagoa Bay
from the Portuguese in the late 1880s and again between 1892 and
1894. By 1890, in addition to being one of the wealthiest men in
South Africa, Rhodes had become Prime Minister of the Cape
Colony, in alliance with the Afrikaner farmers' party, the Afri-
kaner Bond, and he was anxious both to secure Cape economic
supremacy and an outlet at Delagoa Bay or Beira for his land-locked
new colony in Rhodesia. He was foiled in his attempts, however,
by the opposition of the Portuguese, the German veto of his
plans, and the refusal of the British government to come to his
assistance.
Although the Delagoa Bay line eventually reached the border
of the South African Republic in 1889, at the same time as the
Cape railway reached the Orange river, the depression on the
Rand in 1890-1 delayed its further progress. Kruger, anxious for
a rail-link and for the capital Rhodes offered the NZASM to
473

construct the section of the railway connecting the Cape line to

the Rand, and mollified by the bait of Swaziland dangled before
him by the imperial government, now granted Cape colonial
produce preferential treatment in the Transvaal, under the Sive-
wright agreement. For the two years between 1892 and 1894, Cape
farmers enjoyed a virtual monopoly of the Rand market, and the
Rhodes-Bond alliance on which the British pinned many of their
hopes for successful imperial and Cape dominance in southern
Africa flourished. At the same time, the Bechuanaland railway was
extended to Mafeking, and the Rhodes ministry renewed the
internal railway construction at the Cape, which brought the three
Cape ports into communication with one another and with the
Rand.
At the beginning of 1895, however, when the Delagoa Bay and
Natal railway lines reached the Witwatersrand, and with the lapse
of the Sivewright agreement, the crisis over railway and tariffs
erupted once more. Disagreement over the division of the rail
traffic between competing lines led the South African Republic
to raise the tariffs once again on the Cape line, and when Cape
traders attempted to circumvent the tariff by using ox-wagon
transport for their goods once they crossed the Vaal river, Kruger
retaliated by closing the river crossings (or drifts). The Cape
ministry appealed for British assistance, and Kruger was forced
to give way before an ultimatum issued by the British Prime
Minister, Lord Salisbury. At this moment, British fantasies,
which had recurred at intervals since Majuba, of a pan-Afrikaner
conspiracy to overthrow its supremacy in southern Africa, could
not have been further from the truth, as Rhodes carried the
Afrikaner Bond behind him in opposition to republican policies.
That Salisbury should have taken such a hard line on the Drifts
Crisis was in large measure connected to the imperial government's
wider concern at the changing balance of power in southern
Africa, the need to protect British creditors and investors in the
Cape Colony who were threatened by the Transvaal's growing
independence, and fears that Britain was losing control over one
of the swiftest growing markets in the world and the largest single
supplier of gold. Kruger's search for independent access to the
sea and his appeal for German diplomatic and financial support
increased imperial apprehension. Germany was warned that
intervention in South Africa could mean war, as Britain deter-
474

mined to secure the whole of southern Africa as her ' sphere of

influence', using the ambiguous terms of the 1884 London
Convention. While the Liberal government in power in Britain
from 1892 to 1895 refused to sanction overt force against the
Transvaal, it hoped for the reassertion of British hegemony in a
confederated South Africa, through the sheer weight of British
numbers in the sub-continent.
For leading companies like Consolidated Goldfields and Wern-
her, Beit and Company, with their main interests in the deep-level
mines, these demographic calculations were of academic interest.
Committed to a long-range mining programme, they were par-
ticularly affected by the obstructive nature of the republican state.
In 1895, as the deep-level mines were just coming into operation,
development costs were high and they calculated the expenditure
of every penny. The state's taxation policies weighed more heavily
on the deep-level than on the outcrop mines; they were more
dependent on state intervention to control their vast labour force
and they complained bitterly that the Netherlands—South Africa
Railway Company's monopoly over the internal railway system
and Kruger's concessions policies led to inefficiency and added
considerably to their finely calculated costs.
The liquor and dynamite concessions were of special concern
to the mine magnates: they alleged these amounted to an
additional tax on the industry of some £600,000. They were
particularly irksome to the deep-level mines, with their tight
working costs and greater need for labour and explosives. By the
beginning of 1895, Beit, Rhodes and other leading mine owners
were plotting a coup d'etat with the Uitlander Reform Committee
in Johannesburg, which had been agitating for increased franchise
rights since 1892. The appointment in June 1895 of Joseph
Chamberlain as secretary of state for the Colonies in the Unionist
government was a crucial moment in the conspiracy. In August
1895 Rhodes persuaded Chamberlain to grant him a strip of
territory along the Bechuanaland frontier, ostensibly to build a
railway, although the colonial secretary well knew it was intended
as a launching pad against the Republic. Then, despite the
manifest reluctance of local Uitlanders to support an armed
invasion and the last-minute attempts by Rhodes and Chamberlain
to call the affair off, on 29 December 1895, Leander Starr Jameson,
at the head of a column of 478 British South African Company
475

police, rode into the South African Republic, hoping to precipitate
an uprising on the Rand.
Four days later, the incursion was all over. On 2 January 1896
Jameson and his men surrendered to the forces of the South
African Republic, Rhodes was forced to resign from the premier-
ship of the Cape Colony, which he had held since 1890, and the
British—Bond alliance he had carefully constructed was irretriev-
ably destroyed. All over South Africa, Boer and Briton became
increasingly divided along ethnic lines, and republican suspicions
of imperial intentions dramatically increased. Far from removing
the grievances of the mine magnates, the Raid precipitated the
prolonged depression of 1896-8 on the Rand, and exacerbated the
tension between Kruger and Britain. Despite the very serious
attempts made by the Transvaal in the wake of the Raid to meet
the demands of the mining industry by tightening up adminis-
tration, introducing a new pass law, reducing the tariffs charged
by the NZASM and reforming the gold law, over the next four
years the political situation continued to deteriorate. Continued
agitation over the franchise issue for Uitlanders and attacks on
the Kruger regime, were directed by the newly formed South
Africa League and covertly financed by Rhodes and his colleagues,
despite their undertaking after the Raid to abstain from politics.
Incidents which would barely have caused comment in mining
communities in other parts of the world were publicised in highly
partisan fashion in the press, largely owned by the leading
magnates.
In April 1897, Sir Alfred Milner was sent to South Africa as
high commissioner and governor of the Cape. It was hoped by
all sides that this 'model civil servant' would resolve the most
intractable imperial problem of the day. Milner was more than
simply a civil servant however: his previous career had been spent
closely connected at every point with the financial heart of the
British state, as private secretary to Lord Goschen, himself no
mean businessman, who was within a year of Milner's leaving his
service to become chancellor of the Exchequer; as under-secretary
in the Egyptian Ministry of Finance; and then as chairman of the
Board of Inland Revenue. It is unlikely that he did not recognise
the significance for the British economy of either South Africa's
gold or her new markets at a time of increasing international,
especially German, competition. Increasingly, the Transvaal came
476

to be seen as the key to continued imperial supremacy, not simply

in southern Africa but more generally.
Within a year of his appointment Milner had concluded that,
notwithstanding the industrial commission held in 1897, there was
no hope that the South African Republic could produce the
reforms demanded by the mining industry and the imperial
government. With the backing of the British cabinet, he intervened
with growing frequency in the Transvaal's affairs. Increasingly
under attack, the republican government began to rearm. Despite
the inclination of many of the capitalists on the Rand to negotiate
with the rather more accommodating Kruger regime, there can
be little doubt about the continued hostility to the Republic of
the leading deep-level owners actually in South Africa: Fitzpatrick,
Phillips, Beit, Farrar and Rhodes. After Milner's deliberate
rejection of franchise proposals made at the Bloemfontein Con-
ference in May-June 1898 by Kruger's new state attorney, Jan
Smuts, which might well have resolved this particular issue, war
became increasingly inevitable. At the end of September 1899,
with the backing of the Orange Free State, the Volksraad
despatched an ultimatum demanding the removal of British
reinforcements from South Africa, and accusing her of unlawful
intervention in the Transvaal. It amounted to a declaration of war.
On 11 October commandos of the Transvaal and Orange Free
State made a pre-emptive strike into Natal, hoping to take Durban
and to cut the Cape—Rhodesia rail link before reinforcements
arrived and in the belief that they would get international support.
Much ink has been spilt by historians trying to understand the
'causes' of the Jameson Raid and the South African war, and a
good deal of the emphasis has been laid on the fears of the British
that the new wealth of the Rand would draw the Cape into a
pan-Afrikaner, anti-imperial federation which would jeopardise
her sea-route to India.57 Yet Afrikaner nationalism in this period
was more a product than a cause of imperial agitation. Much, too,
has been made of the 'imperial ambitions' of Rhodes; the
'obduracy' and 'obstinacy' of Kruger; the aggression of Cham-
berlain and the importance of public opinion (which he was
largely responsible for creating) in finally forcing his hand; and
the determination of the ' half-German' Milner to have ' his' war.
57
See R. Robinson, J. Gallagher with A. Denny, Africa and the Victorians (London,
1961).
477

Undoubtedly, personality and accident played a role in the final
configuration of events; nevertheless, these protagonists can only
be understood if account is taken of the increasingly competitive
late-nineteenth century world they operated in, a world in which
control over the largest single source of new gold internationally
and the most rapidly growing market were seen as crucial to
imperial interests. Whatever their patrician distaste for individual
mine magnates, it was this which made such strange bedfellows
of Lord Salisbury and Cecil Rhodes, Lord Milner and Wernher,
Beit and Company.58
At the end of the nineteenth century, a peculiar disjuncture
existed between the enclave of advanced mining capital, based
upon the most advanced and sophisticated technology and huge
concentrations of European finance capital in the Transvaal, and
political/state power which remained in the hands of Afrikaner
notables, themselves largely dependent on their land-holdings.
The form of the state in the late nineteenth-century Transvaal still
reflected the pre-industrial agrarian society of the Afrikaners.
Although Kruger tried, through his concessions, to create an
independent industrial base, in fact these attempts simply led to
an increase in the possibilities of corruption among the small
bureaucracy emerging out of the class of notables, most members
of which still retained the ideology, life-style and kinship obli-
gations of the earlier agrarian way of life. Although there were
very real signs in the 1890s that the South African Republic was
capable of 'reform', i.e. of 'modernising', and that individual
Afrikaners were undoubtedly making the transition to capitalism,
it was an earlier entrepreneurial and individualistic form of
capitalism which was remote from the world demanded by the
new concentrations of economic power on the Rand, and from
the demand of the mining magnates for a new form of centralised
and effectively coercive state apparatus. In large measure, Jame-
58
For the role of public opinion in 'pushing' Chamberlain into the war, see
A. N. Porter, The origins of the South African war. Joseph Chamberlain and the diplomacy of
imperialism, ifyj-po (Manchester, 1980). The literature on Rhodes, Milner and Kruger
is too extensive to list in a footnote. For a critique of this literature and an elaboration
of the argument in this section, see S. Marks and S. Trapido,' Lord Milner and the South
African state', History Workshop, 1979, 8, 50-81; and S. Marks, 'Scrambling for South
Africa',/. Afr. Hist., 1982,23,1. For a well-argued account of the relationship between
the mine magnates and the imperial authorities, but with which the above interpretation
differs in some respects, see A. H. Jeeves, ' The Rand capitalists and Transvaal politics,
1892-9', Ph.D. Queens University, Ontario, 1971.
478

son's abortive raid in 1895, and the deliberate contrivance of war

in 1899, were designed to eliminate the dangers of foreign
intervention and to substitute a different political authority in a
region which had assumed major importance for both the British
and the international economy.
The outbreak of war was greeted with jubilation by large
sections of the British press and public. Accustomed to measuring
their maxim guns against ill-armed African peoples, the British
thought they would have a swift victory, but in fact Kruger's
ultimatum caught them unprepared. Contrary to Milner's expec-
tations, the war lasted three years and had a devastating effect
on the countryside for another two. The response of the British
military commanders to the guerrilla tactics adopted by the Boers
after the British conquest of Johannesburg in 1900 was to divide
the countryside into a series of fenced areas by means of
blockhouses. Columns were then despatched in a series of great
drives to scour the countryside, burning the farms, scorching the
earth and rounding up the women and children into camps which
soon became notorious; insanitary conditions, poor rations and
gross overcrowding led to major epidemics and frightful mor-
tality. From the beginning of 1900 to 18 February 1901, 28,000
out of a total of 117,000 white inmates had died of disease, the
vast majority of them women and children. The camps left a deep
scar on Afrikaners, and became a potent symbol in the construction
of their nationalism; they led Campbell-Bannerman, leader of the
British Liberal party, itself deeply torn over the war, to refer to
British 'methods of barbarism'.59
Less remarked at the time was the fate of Africans who were
also herded into camps in the attempt to denude the rural areas.
Unlike the Boer women and children, the Africans were forced
to work for their upkeep, and by the end of the war their
percentage mortality in the camps was even higher than that of
whites; by 1901 there were twenty-nine camps with 107,344
inmates - of whom, according to the official figures, 13,315 died.
The true figure was probably much higher.60
Nor were African fatalities limited to death in British concen-
59
S. B. Spies, Methods of barbarism? Roberts and Kitchener and civilians in the Boer
republics: January 1900—May 1902 (Cape T o w n , 1977), 265—6.
60
Ibid. This section is based on P. Warwick, Black people and the South Africa mar,
1899—1902 (Cambridge, 1983); W. R. Nasson: 'Black society in the Cape Colony and
the South African war of 1899-1902: A social history', Ph.D. Cambridge, 1985.
479

tration camps. Afrikaner commandos plundered peasant supplies

and assaulted blacks in an assertion of republican overrule. In the
northern Cape, the Boer invaders were assisted frequently by
impoverished local land-owners, who threatened blacks with total
expropriation and disenfranchisement in the event of republican
victory. During the war, African resistance to this violence was
on occasion as desperate and direct, whether it took the form of
massed attack as at Holkrantz in 1901, or of individual acts of
desertion, the poisoning of wells and the maiming of Boer cattle.
In the Cape, blacks gave evidence against their former masters in
a further pursuit of revenge. After the war, a virtual jacquerie
erupted in the countryside, fuelled by the embittered rural
relationships and the hopes engendered by British promises of
change. These were soon to be shattered as warring Boer
and Briton rapidly united to restore black-white property
relationships.
In other ways, too, the view that the South African war was
purely ' a white man's affair' was manifestly a myth. Everywhere,
Africans provided labour, accompanying the troops as transport
riders, cattle guards and baggage carriers, while on the Rand they
were drafted into service on the railways and gold mines under
martial law and at minimal rates of pay. In some areas they played
a more active role as armed combatants, spies and scouts, despite
declarations to the contrary by both sides. Although the Swazi
watched from the sidelines, in the northern Cape and western
Transvaal, Tswana groups actively assisted the British and played
a key part in the siege of Mafeking, while in the eastern Cape black
levies prevented the entry of commandos and manned town
garrisons. They played a notable part in the sieges of Ladysmith,
Kimberley and Mafeking.
By 1902, however, it was clear that the greater resources of
ammunition and manpower of the British army had worn the
Afrikaners down, and their leaders were forced to sue for peace.
On 51 May a treaty was signed at Vereeniging. In all, between
1899 and 1902 nearly half a million British troops had been
deployed in South Africa against a Boer force of between 60,000
and 65,000. The cost to the British taxpayer was in the realm of
£222 million; some 6,000 British soldiers had died in action — and
another 16,000 of enteric fever and other diseases. The Boers lost
480

THE CONSTRUCTION OF MODERN SOUTH AFRICA
some 7,000 in action. The total number of African dead is

unrecorded. Thus ended one of the costliest and most bitterly
fought of colonial wars.
THE CONSTRUCTION OF THE MODERN SOUTH AFRICAN

STATE
Even before the war was over, Milner had appointed his ' kind-
ergarten ', young men from Oxford who shared his social assump-
tions, to join him in 'reconstructing' the Boer states. At one
level, this was a matter of establishing an efficient bureaucratic
machine, setting up municipal government, and re-establishing
some modicum of order after the war. Within a short time, the
Boer commandos had been dismantled, and the prisoners-of-war
and ex-servicemen resettled. The dynamite concession was abol-
ished and the Netherlands-South Africa Railway Company
bought out, ending two of the most obvious grievances of the
mining houses. An inter-colonial council was established to unify
railway and tariff policies between the four colonies, and to discuss
matters of common concern.
At another level, the changes were more profound. As high
commissioner and governor of the Crown colonies of the Trans-
vaal and Orange River, assisted by lieutenant-governors, Milner
held unprecedented power to intervene in South African society
and shape its future. Fundamentally these years o f reconstruction'
saw the accelerated transformation of South Africa by the social
engineers along the lines demanded by the mining engineers a
decade earlier. At the heart of this reconstruction were the
relationships between agrarian and mining capital and labour,
both black and white. In the new dispensation, there was no doubt
but that the interests of international mining capital were a matter
of much moment. The smooth functioning of the mining industry
was central to Milner's administration for economic and political
reasons. Yet the end of the war saw a major crisis in the structure
of its financing and a major labour shortage, in part provoked by
the actions of the Chamber of Mines itself in reducing black wages
in 1900 to half their pre-war level. The return of Africans to their
war-torn lands, the breakdown of the Mozambique labour supply
during the war, and the labour demands from other sectors of the
481

WIPE
1
/ ANGOLA
T !
j
(
N
NORTHERN
RHOOESIA
'9
i — . . — S
j
l
I
TJ
*!&
~ X. sCsaltsburv t Sana
r"
\ Tsumeb. i T \ '^ j r l Gweloi G
s
\
y/
v
\ Otawi J£ .^Grooi'ontBin | \ i 9 0 2 j / Fort Victoria
20°S
1
>
FrancistownV
/
2O°S
\ GERMAN / • \
|1
\ / * >
J \
\ _i^^-.l902 r- BECHUANALANO / \ Beit Bridoe
Messiria ww** » f
\jS iWindhoek 7 O k
Waives Bay t t PROTECTORATE / / « •\ 5
( SOUTH-WEST }
/
\
Gaberonesj/,y
If
J J
T R^fNSVA>
\ \
CO 1 AFRICA retoria Q ^_
ânne|bur£
\ I 9 0 g Keetmansdorp
luderitiyr- t -*-«-»- fc _ i >*
\ Vi9iO
! ^S ^ Vryburg/^ j / ^ 7
^v"7
\ >cKalkfo 'item ewcastl^^l
Pon NollotrViJ?3
^\?*ftadvs //
C A iUTCH \piet j
30 "S \ *Okiep f\i89o/ y L A flD
/• KSU burbarj 40f>e
\Co ,' ^
Railways constructed by 1910
9
COL o N / Y *Vss M s n ueeni c
Main harbours
Coal Co Copper
B»/ Reined L,__X°, D Diamonds G Gold
JJ 0_ 500 km
CapeTownV> > i jgrcester Oudtshoorn ^ « — ^ i i
to\°£
=J^— <• ~_/—^Porl Elizabeth
f ibo miles
19 Southern Africa, c. 1910

economy in the immediate post-war boom all exacerbated the

crisis. The chamber was in any case anxious for state assistance
to force down all wages and thus make possible the mining of
low-grade ores left unexploited before the war.
Despite the unceasing demands for land and labour from every
side, the forcing out of labour through the outright expropriation
of African land, which was demanded by many farmers, was
clearly too costly and dangerous, in the light of recent experience.
In Natal and the Transvaal land colonisation companies opposed
the uprooting of their rent-paying tenants, while in the Orange
River Colony, share-cropping was particularly crucial to the
restoration of agriculture after the war.61 Moreover, although the
mine magnates were anxious to solve their problems of labour
supply, they were becoming increasingly aware of the utility of
African reserves in subsidising their labour costs. As Howard
Pirn, at that time the accountant for the Chamber of Mines and
an important influence on the 'kindergarten', observed in 1905:
Let us assume... that the white man does turn the native out of one or more
of his reserves... the natives must live somewhere. We will suppose that he
is moved into locations attached to the large industrial centres... In the location
he is more closely huddled together in surroundings in which his native
customs have no place and he is compelled to purchase from the white man
the food which in his own country he raised for himself. What the white man
gains, therefore, is little more than the labour required to pay for the food which
under natural conditions the native raised for himself... The white man has
not yet shown that in South Africa his cultivation of the simple crops which
the native requires can compete with native cultivation... For a time the urban
location consists of able-bodied people, but they grow older, they become ill,
they become disabled - who is to support them ? The reserve is a sanatorium
where they can recruit; if they are disabled, they remain there. Their own tribal
system keeps them under discipline, and if they become criminals there is not
the slightest difficulty in bringing them to justice...As time goes on these
location burdens will increase and the proportion of persons in the locations
really able to work will still further diminish... it is a fair assumption that at
the outside one-fifth of the location population... is able to work. This means
that the wages paid by the employers will have to be sufficient to support four
other persons besides the workman... 6 2
61
T. Keegan,' The transformation of agrarian society and economy in industrialising
South Africa: the Orange Free State grain belt in the early twentieth century', Ph.D.
London, 1981.
61
H. Pirn,' Some aspects of the South African native problem', South Afr. ]. Science,
1905, 4, cited in M. Legassick, 'British hegemony and the origins of segregation in
South Africa', unpublished seminar paper, Institute of Commonwealth Studies,
London, 1973.
483

In the short term, in order to solve its labour and profitability

crisis, the Chamber of Mines decided to import indentured
Chinese labour, a policy which demanded the intervention of both
the imperial and colonial government actually to mobilise the
labour and to secure the necessary international agreements.
Between 1903 and 1907, when the scheme was ended, some 60,000
Chinese labourers were brought to South Africa on three-year
contracts, and at lower wages than those of local African
workers.63 The policy roused a storm of protest, both in South
Africa and in the United Kingdom, where the issue of' Chinese
slavery' was an important feature of the 1906 election campaign.
The importation undoubtedly jeopardised what have conven-
tionally been regarded as Milner's political objectives in South
Africa. It split the English-speaking vote, starkly revealed the
mutual dependence of mining industry and state, and unified
Afrikaner opposition to Milner at an early date.
The dramatic qualities of the Chinese labour 'experiment', and
its political price-tag, should not, however, obscure its longer-term
implications. The bargaining power of African workers was
removed by direct state intervention at the very time that they
were in a relatively strong position to exert it; an explicit colour
bar which was to outlast the use of the Chinese was inserted into
the regulations employing the Chinese in order to allay the fears
of the English-speaking working class; and a definite decision was
taken that the white man would not be employed as an unskilled
labourer at the side of the black, notwithstanding the high rate
of white unemployment on the Rand at the time, and despite
Milner's desire to swamp the Afrikaner demographically.64 Nei-
ther Milner nor the mine-owners were prepared to see the growth
of large numbers of unskilled, non-supervisory white workers on
the Rand. As Milner put it, white labour was 'much too dear': 65
unlike Africans, the unemployed and unskilled Afrikaners who
were beginning to stream into Johannesburg from the rural areas
after the war were wholly landless and required a wage which
would provide for their own upkeep as well as their families'
63
T h e m o s t recent a c c o u n t of the importation o f Chinese labour is P. Richardson,
Chinese nine labour in the Transvaal ( L o n d o n , 1982).
64
F o r the w a y i n w h i c h the imperatives o f the m i n i n g industry t o o k priority o v e r
Milner's purely political objectives, s e e D . D e n o o n , 'Capital and capitalists in t h e
Transvaal i n the 1890s and 1900s', Hist. ]., 1980, 2 3 , 1, 1 1 1 - 3 2 .
65
Milner's reply t o a deputation from the W h i t e L a b o u r L e a g u e , 2 June 1903, in
C. H c a d l a m , The Milner papers, z v o l s . ( L o n d o n , 1 9 5 1 - 3 ) , 4J9.
484

THE CONSTRUCTION OF MODERN SOOTH AFRICA
subsistence and reproduction. Fear that the unskilled workers

would become unionised was a further consideration.
But the white working-class could neither be dispensed with
nor disregarded and this led to a series of interventions by the
state, both to reduce its cost and to prevent its joining up with
black workers. These ranged all the way from municipal
government and fiscal policies, to sewers, transport and housing.
The latter was particularly important, given the growth of
multiracial suburbs in the 1890s. Among the first actions of the
Milner regime was the tightening up of urban segregation: as in
other parts of South Africa, public health measures and the
metaphor of disease provided the ideological justification, as
white working-class housing was separated from that of black, in
much the same way as the ' respectable' working class was being
separated from 'outcast London' in late nineteenth-century
Britain.
The provision of white working-class housing was intimately
linked with further schemes to shape the geography of class on
the Rand and stabilise the skilled working class. Thus among the
first schemes blessed by both imperial administration and mining
industry (which provided the finance) was the importation of
women domestics from the United Kingdom, who, it was hoped,
would release scarce black male labour from domestic service, and
supply wives for white workers.66 Education too was given a high
priority — both for the defeated Boers, and for the children of
English workers on the Rand.
A further way of cheapening the cost of labour was, as mining
spokesmen had recognised even before the war, through improved
agriculture, and here, again, the new administration introduced
far-reaching changes. These were related in turn to Milner's plan
to introduce a' leavening' of British capitalist farmers to the South
African countryside, who would increase the numbers of British
settlers in the ex-republics, transform rural relations of production
and send up the price of land. Milner's immigration schemes
brought too few settlers to have an appreciable political effect,
and the other attempts of the administration rapidly to restore
agriculture were impeded by the devastation caused by the
66
Van Onselen, Studies, i: Nea> Babylon, 17-8; J. J. van-Helten and K. Williams,' The
crying need of South Africa: the emigration of single British women to South Africa,
1901-10',/. Southern Afr. Studies, 198J, 10, 1.
485

scorched earth policy, drought, cattle disease and depression.

Nevertheless, it was during the British administration that state
intervention in agriculture, to become such a feature of twentieth-
century southern Africa, was made effective. Both colonies were
given land boards, which developed into departments of agri-
culture, experts were brought in to advise the government on
scientific farming, and agricultural societies were founded to
advise farmers on crops, animal breeding and disease and new
technology. In 1908 a land bank was established in the Transvaal,
financed by a £5 million British government loan. In that year,
the food shortages of the pre-war era were replaced by surpluses
for export for the first time.67
Although at the end of the war there were those who saw a
future for African peasant production, and compared its produc-
tivity favourably with that of white farming, African hopes that
the British conquest would lead to a redistribution of the land they
had lost were soon disappointed. Pre-war property relations were
swiftly confirmed, in both town and countryside. Popular African
resentment can be traced through the millenarian prophecies
recorded in the colonial archives, the rumours of impending
rebellion - sufficient to secure the almost immediate rearming of
Afrikaner notables — and the proliferation of independent faith-
healing sects. Insecure themselves, especially as the post-war
boom was replaced by a major economic depression, whites
interpreted African restlessness as a result of the effect on 'the
native mind' of the 'white man's war', and agitation by 'Ethio-
pians ' (members of independent black Christian sects), who were
believed to be preaching a doctrine of 'Africa for the Africans',
under the influence of black Americans.
Whatever the dangers from Ethiopianism, the failure of the
British administration to fulfil the hopes that its war-time prop-
aganda had aroused, and the renewed pressures for African land
and labour after the war, were of greater moment in firing
discontent. Milner's emphasis on' well treated and justly governed
black labour' 68 did not mean that he opposed institutional
controls on African labour; his administration took over much
of the legal framework of the Kruger regime. Tax and pass laws
were now administered more efficiently. Arbitrary terror was
67
M a r k s and T r a p i d o , ' L o r d Milner and the S o u t h African State', 6 8 - 7 1 .
68
M i l n e r t o Sir P. Fitzpatrick, 28 N o v . 1899 in H e a d l a m , Milner papers, I I , 35.
486

reduced, but the greatly improved state apparatus drew far larger
numbers of Africans into the network of colonial control. In 1903
a South Africa-wide Native Affairs Commission was appointed
under the chairmanship of Sir Godfrey Lagden, in order to unify
'native policy', in view of an impending confederation of the
southern African territories. Although it took evidence from all
over southern Africa, and included a large number of Africans
amongst its witnesses, the commissioners were largely English-
speaking or imperial administrators. Drawing on existing prece-
dents in the Cape and Natal, its recommendations provided the
blue-print for the segregationist South African state after 1910.
They were directed to limiting the amount of land available to
Africans, establishing territorial ' separation' of land ownership;
the transformation of African 'squatters' into wage-labourers;
and the provision of some form of representation of African
'grievances' outside the central decision-making bodies of the
state.
Much British propaganda before and during the war had been
concerned with the political rights of British subjects, and
educated Africans. Indians and Coloureds all over South Africa
believed that, with the British victory, at the very least discrim-
ination against them would cease and the limited non-racial
franchise which existed at the Cape would be extended to the new
Crown Colonies. These hopes were also to be dashed. Almost the
first action of the Milner administration in the Transvaal was to
remove Indians to locations, thus enforcing Kruger legislation
which the imperial government had been opposing since 1885.
Despite lip-service to the needs of the' respectable' Coloured man,
and fears that if he were not incorporated into the colonial state,
he might join hands with the African, neither Milner nor his
successor, Lord Selborne, took action on his behalf. The small,
formal political organisations which existed to cater for the needs
of the educated middle class amongst Africans, Coloured and
Indians, were equally almost invariably rebuffed, despite their
essentially moderate and pro-British stance. It was his experience
of these years which led Mohandas Gandhi, legal adviser to Indian
merchants in Natal and Transvaal from 1892, to formulate his
techniques of satyagraba, non-violent passive resistance. Mean-
while, in the Cape, Dr Abdurahman, the leader of the Coloured
African Political Organisation, began to advocate black-brown
487

unity and working-class solidarity in response to the failure of the
British to extend political rights. All over South Africa, vigilance
societies and congresses sprang up organised by the growing
black intelligentsia. There were relatively few meeting points at
this stage between Gandhi and the desperately impoverished
indentured and ex-indentured Indian labourers of Natal, between
the African Political Organisation and the increasingly landless
Coloured proletariat, or between the provincial native congresses
and African peasants and workers. Nevertheless, as they found
their attempts at incorporation rejected, middle-class leaders were
on occasion radicalised and strove to speak on behalf of all blacks,
if only to increase their own constituency and credibility.
After the victory at the polls of the Liberal Party in the United
Kingdom early in 1906, new constitutions were devised for the
ex-republics, granting them self-government based on adult male
suffrage. The issue of a vote for blacks was not even subject to
debate. The peace agreement between Britain and the Boers at
Vereeniging had explicitly deferred consideration of the franchise
for ' natives' until ' after the introduction of self-government' in
the Transvaal and Orange Free State: as both Milner and
Chamberlain recognised, this effectively pre-empted the issue. In
both colonies, Afrikaner parties won the elections, led by the Boer
War generals, Hertzog, Botha and Smuts. In the Transvaal the
latter were assisted by English-speakers, who saw in Het Volk
allies against the party of mining capital.
The victory of the Afrikaner parties, the division of the
English-speaking vote, and the abandonment of an imperial
attempt to adopt anti-discriminatory policies, have led many
historians to describe Milner's sojourn in South Africa as a
failure.69 Certainly Milner himself saw the grant of responsible
government to the conquered republics in this light. Nevertheless,
within a very short time, as members of his ' kindergarten' were
among the first to recognise, an accommodation was reached
between the Afrikaner leaders in the Transvaal and the mining
and imperial interests. Dubbed 'the alliance between maize and
gold', 70 it also represented an alliance between mining capital and
the state.
"* The notion of Milner's 'failure' has become a commonplace in the literature; for
the alternative view, see Marks and Trapido, 'Lord Milner', 55-4; and M. Chanock,
Unconsummated union: Britain, Rhodtsta and South Africa (Manchester, 1977), 2-7.
70
The phrase was originally S. Trapido's in ' South Africa in a comparative study
of industrialisation', J. Development Studies, 1971,7; for a recent study of the ' symbiotic'
488

Three issues perhaps epitomise the nature of this alliance: the

Smuts-Botha agreement to allow the gradual phasing out of
Chinese labour, while allowing the mines to increase the length
of African contracts and assisting its recruitment in Mozambique
and Malawi; the readiness of the mining industry to employ
unskilled Afrikaners in supervisory roles during the 1907 mine
strike, during which British troops were called out 'to keep
English miners in order'; 71 and imperial backing for the Land
Bank. Botha and Smuts were keenly aware of the centrality of the
mines for the capitalist development of South Africa and, as
progressive farmers, were as heavily committed to the industry
which constituted their main market as Milner and his kinder-
garten had been.
Within a couple of years of the war, the major political issue
both for colonial politicians and members of the ' kindergarten'
was the unification of the South African colonies. A spur to the
movement came in 1906 with the Bambatha rebellion in Natal,
which revealed the dangers of leaving ' native affairs' in the hands
of small and nervous colonial governments. The demand for a
uniform ' native policy', a synonym for control over labour, was
one of its major objectives. Backed by information and publicity
provided by the kindergarten and financed by the mining industry,
a national convention representing most white political opinion,
thrashed out a constitution in 1908—9. This formed the basis for
the Act of Union, approved by the colonial parliaments, by a
referendum in Natal, and by the British parliament, before coming
into operation on 31 May 1910.
Two issues were the matter of heated and divided discussion
at the convention - and had profound implications: whether or
not the old maxim 'equal rights for all civilised men' would be
adopted by the Union, and how constituencies were to be drawn
up. Although at the time of the grant of self-government to the
Transvaal, the latter issue had been a matter of major concern
(with Progressive Party insistence on equally drawn constituencies
based on their male, white, adult population), after a prolonged
tussle with the Cape delegation, the Act of Union allowed the
relationship between the state and the mining industry, see D. Yudelman, The emergence
of modern South Africa. State, capital, and the incorporation of organised labor on the South
African gold fields, 1902-19)9 (Westport, Conn./London, 1983), 59-78.
71
Lionel Phillips to Julius Wernher, 3 June 1907, in M. Frascr and A. Jeeves (eds.),
All that glittered: selected correspondence of Lionel Phillips 1890-1914 (Cape Town, 1977),
'79-
489

weighting of rural constituencies of up to 30 per cent. This
provision, which granted a certain electoral advantage of rural, and
thus largely Afrikaner, voters, was portrayed at the time as part
of the price of the much vaunted ' reconciliation' between Boer
and Briton. Again, despite the rhetoric, however, the mine
magnates were not averse to a franchise weighted in favour of
rural Afrikaners, especially after the politicisation and mass
unionisation of the white working class which followed in the
wake of the 1907 strike, and which led in 1909 to the formation
of the South African Labour Party. Afrikaner nationalism had not
yet established itself as a monolithic, ethnically mobilisable force,
despite the early efforts of the displaced Afrikaner intelligentsia-
teachers, preachers and professionals — and a burgeoning Afrikaans
literature playing on the recent memory of the war. Afrikaner
farmers, especially in the Transvaal but also in the Free State and
the Cape, saw their future as inextricably linked with the mining
industry and imperial markets, while poorer Afrikaners, them-
selves the product of class-formation in the countryside, were
perhaps as attracted to class-based as ethnically-based associations
in this period.72
Although Cape delegates defended their non-racial, class-based
franchise, which they saw as a useful 'safety-valve' for black
aspirations, they were aware that any attempt to insist on its
extension would jeopardise union. Before the war, Africans in the
Cape eligible for the property-based franchise had already had
experience of Western-type political organisation for a generation
and more. The large increase in black voters in the 18 80s consequent
on the annexation of Transkeian territories and the shift in
hegemony in Cape politics from merchants, interested in a stable
peasantry, to industrialists with an insatiable demand for unskilled
labour had already undermined the basis of the Cape franchise.
In 1887 communally-held land no longer qualified Africans to
vote, while in 1892 the property and educational qualifications
were raised. Nevertheless, with the dramatic polarisation of Cape
politics as a result of the Jameson Raid, the fourteen constituencies
71
For the weakness of Afrikaner nationalism before the South African War see
H. Giliomee, 'Reinterpreting Afrikaner nationalism, c. 1850-1900', unpublished sem-
inar paper, Institute of Commonwealth Studies, London, 198); for the attempt to
organise on class lines, see Van Onselen, ' The main reef road into the working class:
proletarianisation, unemployment and class consciousness amongst Johannesburg's
Afrikaner poor, 1890-1914', in Studies, 11: New Nineveh, especially 125-70.
49O

in which African voters held the balance became an important

electoral factor for white politicians.
Elsewhere in South Africa, blacks were virtually totally ex-
cluded from political rights. In Natal, despite the existence of a
large and vibrant African Christian community, some of its
members being third-generation amakholtva (believers), there were
but half a dozen African voters on the rolls by the turn of the
century. If anything, Natal politicians were more adamantly
opposed to even a qualified black suffrage in the forthcoming
Union than the ex-republicans. In the end a' compromise' proved
acceptable to both the convention and the British government.
While the Cape retained its existing franchise, elsewhere white
adult male suffrage obtained. This, together with the clause
protecting equal language rights, was entrenched under the
constitution. In addition, anyone not of'European descent' was
barred from sitting in either house or parliament.
African and Coloured opposition to the 'colour bar' provisions
in the draft Act of Union was immediate and clear. A joint
delegation was despatched to London in an attempt to get the
British government to veto the discriminatory constitution. It
availed them little. Most British politicians were convinced of the
need for unification in the interests of economic development, and
euphoric about the great reconciliation which had taken place
between the white 'races'. The few lonely voices raised against
the dangers of a constitution which excluded the vast majority of
the population from the political community, were ignored in the
atmosphere of mutual congratulation.
Although representatives of Southern Rhodesia and the high
commission territories were present at the convention, it was
quickly decided that Southern Rhodesia would not enter the
Union until its future was clearer, and that the smaller territories
of Basutoland, Bechuanaland and Swaziland would also remain
outside the Union for the time being, although a schedule attached
to the Act of Union set the terms for their future incorporation.
There was considerable feeling among white politicians that the
schedule represented a lack of confidence in their good faith on
the part of the imperial government and that the territories would
constitute an imperial enclave within the South African state,
derogating from its sovereignty. Nevertheless, the danger of
violent African resistance to incorporation, together with
491

SOUTHERN AND CENTRAL AFRICA, l886—1910
humanitarian opposition and a British desire to maintain an

independent foothold in South Africa meant that the three terri-
tories remained outside the new Union. As in the case of Southern
Rhodesia, however, it was assumed that their ultimate fate was
within the Union, whose frontiers would ultimately extend, as
Milner had envisaged, to the Zambezi.73
Despite the achievement of union a mere eight years after the
South African war, and the enormous changes which these years
had wrought, one should not exaggerate the strength of the
political union established, nor its economic autonomy. Powerful
in its regional context, it was still, as Martin Chanock has
reminded us,
... defenceless in a world of predatory imperial powers; it was believed by many
(with reason) to be inherently unstable; and apart from having within its
borders valuable mineral resources which were foreign owned, it was poor -
a small-time exporter of agricultural raw materials, not self-sufficient in food.
It was a new African country which was weak, divided, poor and owned
overseas, with the majority of its population effectively excluded from its
political processes and distrusted by their rulers... 7 *
The twentieth century has so far transformed all but the last.
73
Milner to Fitzpatrick, 28 Nov. 1899, op. cit.
14
Chanock, Unconsummated union, 2.
49z

CHAPTER 9
PORTUGUESE COLONIES AND

MADAGASCAR
A. ANGOLA A N D MOZAMBIQUE, 187O-1905

ANGOLA AND MOZAMBIQUE BEFORE PARTITION, 1870—85
The inception of the scramble for Africa obliged Portugal to act
on what had been an established ideal for many centuries. Whereas
their contemporaries in other European countries had eschewed
the acquisition of territory, many Portuguese had envisaged the
ultimate conquest and consolidation of the territories in the.
hinterland of their coastal settlements in Angola and Mozambique.
That these ideas never bore fruit resulted from the lack of
urgency, the inadequacy of Portuguese resources, and, above all,
the successful resistance of African peoples to those plans which
actually were initiated. However, when other Europeans began
to show an interest in the lands of central Africa, Portugal chose
to marshal its limited resources in a concerted effort to secure as
much territory as possible. Because of this effort, Portugal
eventually received recognition from the other powers of its right
to exploit the vast land areas of Angola and Mozambique. Yet,
the same lack of resources which had delayed the conquest of these
regions for such a long time played a crucial role in determining
the pace of the establishment of effective occupation and the
nature of Portuguese colonial control.
Even in 1885 Portugal occupied only a small portion of the
territory which would ultimately be recognised as Angola. There
was a degree of irony in the fact that the Berlin Conference of
that year enunciated the doctrine of effective occupation, since
Portugal had recently completed a withdrawal from the interior.
During the 1850s, the Portuguese had penetrated many parts of
the populous western half of Angola in the hopes of profiting from
direct control of these regions. However, when it was realised that
this expansionist policy had not produced the desired results, they
began to retreat to their traditional bases of operation. Thus, by
493

CABINDA*
1 CONGO INDEPENDENT STATE
KONGO 1
Luanda; NDEMBU LUNDA

KASANJE
10 "S
Benguela,
Mozambique
Island
4
\BA
so MASHONALAND '
/
SOUTHERN
2O"s RHODESIA 2O°S
GERMAN SOUTH-WEST AFRICA
/ N D I A N
OCEAN
Boundaries after 1891

— — Area claimed by Portugal on Mapa Cor-de-Rosa
0 SpOkm
0 3 0 0 miles
20 Angola and Mozambique at the beginning of the twentieth century

ANGOLA AND MOZAMBIQUE BEFORE PARTITION
the 18 60s Portuguese activity in Angola was concentrated in and
around Luanda, along the north bank of the Cuanza river, and
in the environs of the southern towns of Benguela and
Mocamedes.
In these regions, however, the Portuguese impact on local
culture had been profound. Over the course of time, a distinctive
and uniquely Angolan society, characterised by its ambivalent
social attitudes, had emerged. In Luanda, Portuguese served as
the official language, but all segments of the population made
equal use of Kimbundu. Colour played a prominent role in the
system of social stratification. Yet it was not uncommon to find
Africans included in the highest stratum of society and mestizos
relegated to the lowest. Although the centre of this nascent Creole
culture was located in Luanda, its impact was not limited to the
capital. In addition to taking root in other country towns, aspects
of Creole culture were spread throughout the southern plantations
by the Mbari slave and ex-slave population, and disseminated
more widely in the highland areas by such peripheral groups as
the coffee planters of the Luanda hinterland and the itinerant
traders of the interior. •
A rather different cultural nexus had evolved in the extreme
south of Angola. In 1849 white settlers began to take up
permanent residence in Mocamedes and the adjacent Huila
highlands. Because of the healthy climate and the weakness of the
neighbouring African states, this nucleus was able to multiply
quite rapidly. During the early 1880s, trek-Boers emigrating from
the Transvaal and peasants arriving from Madeira helped to swell
the ranks of the white colonists. By 1885, their number already
exceeded 1,500, and more were expected from both Portugal and
the Transvaal. In many ways the emerging social and cultural ethic
of the southern population was more similar to that of South
Africa than the older towns of Angola. Unlike the stable convict
populations of Luanda and Benguela, it was a society dominated
by family groups. Moreover, in contrast to the relaxed tolerance
of Creole Luanda, the white elite exercised a strict social and
cultural domination over blacks and mestizos. In this environment
openly expressed racist ideas became the norm of society.
Changes of a different nature had been occurring in the areas
which lay beyond direct Portuguese cultural and political
1
D. Wheeler and R. Pelissier, Angola (London, 1971), 68-71.
495

PORTUGUESE COLONIES AND MADAGASCAR
influence. By 1885, the disintegration of large state systems,

already visible in many areas from the beginnings of the century,
had become general. The Kongo kingdom in the north retained
considerable prestige and ritual importance, but its real political
authority was extremely limited. Power had passed to a multitude
of independent chiefs, who controlled the expanding trade of
the Congo estuary and the adjacent coast. Similarly the once
formidable kingdom of Kasanje, to the east of Luanda, was only
a hollow shell, with power in the hands of the lineage elders.2
Further east, the Lunda empire was rapidly disintegrating, while
in the central highlands the Ovimbundu kings were increasingly
threatened by emergent groups of traders. In the far south, the
ancient kingdom of Humbe on the Cunene river was much
reduced and threatened by the rising power of its former Ovambo
subjects.
This tendency toward political decentralisation did not imply
that African societies were becoming weaker and less able to resist
European encroachments. On the contrary, these changes often
reflected the emergence of new and dynamic groups, more capable
of defending their independence than some of the brittle polities
which they had come to supersede. The Ndembu, immediately to
the north-east of Luanda, served as a focus for runaway slaves and
resisted the Portuguese successfully. Even more striking was the
case of the Chokwe, who raided for slaves, hunted for ivory, and
collected wild rubber. The wealth produced by these endeavours
allowed them to build up a large arsenal of modernfirearmsand
swell their numbers by incorporating alien women through the
system of pawnship. As a result, the Chokwe were able to spread
over eastern Angola and western Zaire, overthrowing the Lunda
empire in the process. The Portuguese were to find it extremely
hard to conquer this elusive foe, particularly because of the lack
of any centralised institution on which to concentrate their
forces.3
The restructuring of African societies mirrored the economic
transformation which Angola had undergone during the nine-
teenth century. The winding down of the Atlantic slave trade
began in 1836 with the abolition decree, and was virtually
2
J. Miller, 'Slaves, slavers and social change in nineteenth century Kasanje', in
F. Heimer (ed.), Social change in Angola (Munich, 197}), 10-29.
3
J. Miller, 'Chokwe trade and conquest in the nineteenth century', in J. R. Gray
and D . Birmingham (eds.), f re-colonial African trade (London, 1970), 175-201.
496

complete by the 1860s. Products of hunting and collecting,

however, took the place of slaves in the African sector of the
Angolan extractive economy. Initially, it was the ivory trade
which assumed increased importance. Stimulated by the end of
the royal monopoly, the ivory cycle reached a peak during the
1850s, before subsiding in the 1860s. Rubber and beeswax,
however, soon filled the vacuum created by the decline in ivory.
Various regions of Angola were also fortunate in finding
indigenous products, which, with only minor attention, could be
profitably grown. Palm products were predominant in the Kongo
area, where the trees grew wild and Africans ran a kind of
semi-plantation system. In the hinterland of Luanda groundnuts
and coffee, both indigenous to the area, provided the staples.
Thus, virtually every region of Angola was able to find a
substitute for the export of human beings.
The Creole sector of the economy also underwent a period of
growth and change. The most important innovation was the
development of a plantation system. Although Africans operated
plantations in many areas, the coastal estates were usually
dominated by Europeans, who often made use of slave labour.
Despite violent price fluctuations on the world markets, both
groups were generally fortunate in finding new cash crops which
couid be profitably grown. Cotton, cultivated on European
holdings in the south, thrived during the period from the end of
the American Civil War until the 1880s, when a steep drop in
prices brought ruin to the industry. Similarly, plantation-produced
coffee and sugar each experienced periods of prosperity. In
addition to the increased attention devoted to the products of the
soil, Europeans in the south also began the intensive development
of fisheries.
The cyclical nature of Angolan production was not without
impact on African society. Since there was no competitive Asian
factor, and European and mestiqo attempts to penetrate the interior
were thwarted, Africans were able to maintain control of com-
merce. Yet, the alterations in the economy served to thrust
forward groups and classes which were in a position to react more
quickly to the changing circumstances. Thus, the Chokwe, whose
residential mobility enabled them to move into areas which
produced a product of contemporary demand, gained in both
wealth and power. Similarly, in these new conditions, successful
497

individuals and corporate lineages of traders were able to challenge

the older order which found it difficult to transform static social
structures based on slaving and defence against raiding. Typical
of these new men were the shrewd Kongo chiefs of the northern
coast whose independence of the monarchy was abetted by their
ability to collect taxes on the ivory, rubber, and palm products
exported from their area. At the same time, in the southern part
of Angola, Ovimbundu trading corporations, whose memberships
were based on dispersed matrilineages, began to challenge the
dominance of the localised patrilineal chiefdoms.
In the mid-nineteenth century, Angola enjoyed a substantial
geographical advantage over Mozambique, which probably ex-
plains the much greater political and economic importance of the
former colony until 1885. Before the building of the Suez Canal
and the development of the mining economy of southern Africa,
Angola was much closer to the important world markets and was
better served by steamers. Although the colony was not without
financial problems and was often in deficit, it was still the prin-
cipal jewel in the Portuguese imperial crown, with larger revenues
than those of any other colony.4
Portuguese influence in Mozambique had been much less

extensive than in Angola. In 1885 it was largely limited to a
number of small settlements along the coast. Starting from
Lourenco Marques in the south, they included Inhambane, Sofala,
Quelimane, Mozambique Island and Ibo in the far north. With
the exception of Lourenco Marques (Maputo), which was
beginning the transformation which ultimately would make it into
a major port, these stations had much in common. Each housed
a very small number of Portuguese soldiers, a somewhat larger
number of African recruits and a few petty traders. Only among
the Tonga in the environs of Inhambane could it be said that the
Portuguese ruled over the local African population. Elsewhere the
weakness of the military stations determined that they had no
impact whatsoever on the neighbouring peoples. Similarly,
Portuguese settlement in the interior was restricted to the two
Zambezian towns of Sena and Tete, neither of which was capable
of influencing events beyond the immediate vicinity.
Portuguese influence, however, was being disseminated in the
4
A. de O. Marques, Historia de Portugal, vol. II (Lisbon, 1975), 149—51.
498

interior by the expansion of the pra^p system of large estates.

During the nineteenth century, praêros (land-holders), often of
mixed African and Asian descent, had extended their control
along the Zambezi and its tributaries. With the aid of their locally
recruited Chikunda armies, they had come to control populations
on both sides of the river. As a group the praêros retained the
independent spirit of the frontiersman and resisted violently when
the Portuguese Crown attempted to intervene in their affairs. As
long as the Portuguese administration allowed free scope to their
activities, however, praêros maintained a nominal allegiance to
it and often acted in its name. Thus, according to M. D. D. Newitt,
'if there had been no "Partition", Central Africa would probably
have been settled by a number of Chikunda tribes under chiefly
families with Portuguese names \ s
Elsewhere, Portuguese influence in the interior of what was to
become Mozambique was virtually non-existent. Although the
vast region bounded by the Zambezi on the south, the Indian
Ocean in the east, the Rovuma river in the north, and Lake Malawi
and the Shire river in the east, was largely unknown to Europeans,
it was in the process of undergoing major transformations. The
incipient spread of Islam was among the most important. Muslim
Swahili sultanates continued to dominate the coast as they had
done for centuries. However, Islam was beginning to make
inroads among the Makua, who occupied all of the eastern half
of the region. It was also being carried to the western portion by
the Yao, whose expansion had taken them as far as the Shire
highlands and the shores of Lake Malawi, where they were
engaged in raiding the local populations. They were often
successful in incorporating the Nyanja, Lomwe and Makua
peoples they encountered into new chiefdoms. In addition to the
expansion of the Yao, northern Mozambique was also the scene
of the aggressive raiding of a number of small Nguni polities. The
populations in such diverse localities as the Luangwa, Shire, and
Rovuma river valleys were all subject to incursions by the groups
which had infiltrated many of the peripheral regions. Thus,
throughout northern Mozambique, aggressive and militant
societies were in the process of initiating important social
changes.
There were a number of factors discernible in the dynamic
5
M. D. D. Newitt, Portuguese settlement on the Zambezi (London, 1975), 506.
499

evolution of society in northern Mozambique which would have

a profound effect on resistance to European encroachment. The
region lacked a dominant state or ruler against whom the
Portuguese could concentrate. Rather, there were a large number
of small chiefdoms, each virtually independent of its neighbours.
Secondly, many groups were accustomed to the use of firearms,
which were supplied in ample quantities by Portuguese merchants
on the coast. Finally, the penetration of Islam provided an
ideological base both for opposition to Christian encroachment
and for a pan-ethnic resistance. Thus, an infrastructure, which
suggested the capability of successful resistance, had evolved in
northern Mozambique.
Conditions to the south of the Zambezi differed greatly
from those prevailing in the north. The whole of the region
beyond the influence oipra^p society continued to be dominated
by the Shangane state of Gungunyana. Since his grandfather,
Soshangane, led a group of Nguni refugees from Natal into
Mozambique in the 1820s, the Shangane had resided in a number
of places. The capital had been located as far south as the Nkomati
river and as far north as the hinterland of Sofala. By 1885,
Gungunyana had transferred his residence from the north and
settled in the Limpopo valley. The itinerant nature of the
Shangane state determined that the amount of influence it
exercised over the peoples of the southern half of Mozambique
largely corresponded to the location of the capital at a particular
point in time. When the Shangane mostly resided in the hinterland
of Sofala, they dominated the kingdom of Manyika and the
Ndau-speaking peoples to the north of the Sabi river. Once they
had shifted southward, it was the Tsonga and Tshopi populations
who were obliged to pay tribute.6
The dominance of the Shangane over so much of southern
Mozambique was to have a profound influence on how the
process of partition would unravel. On the one hand, it provided
the Portuguese with a central focus on which to concentrate their
efforts, while, on the other, the military prowess of the Shangane
posed an obstacle which even most Portuguese respected. By 1885
both factors were already in evidence. The Portuguese claimed
to possess a document signed by Mzila, Gungunyana's father, in
6
G. Liesegang, Beitrdge %ur Geschichtc des Ketches der Gaâ Nguni im siidlichen
Mozambique, 1820-iSfj (Cologne, 1967).
JOO

which the former king had allegedly agreed to become a ' vassal'
of the crown. Gungunyana, however, not only denied the
authenticity of this document, but maintained that he was not
bound by any agreements made by his father. Moreover, because
of the strength of his impis, Gungunyana was able to act in a
defiant manner and treat the Portuguese claims with disdain.
Similarly, the weakness of the forces at the disposal of Mozambican
officials obliged them not to undertake any actions which might
provoke him.7
The lack of military strength and the poverty-stricken nature
of the Portuguese settlements resulted partly from the depressed
state of the economy. Traditionally Mozambique had depended
on the export of ivory, gold and slaves. Yet, by 1885, the
commerce in all three had diminished greatly. The elephant herds
on which the ivory trade had been based had disappeared virtually
throughout Mozambique. Moreover, a great deal of the traffic
which originated further in the interior had come to be exported
by other routes. A similar fate had befallen the gold-mining
industry. Changes in the patterns of marketing the gold produced
in the interior had combined with the exhaustion of the alluvial
deposits, on which the kingdoms of the south had been based,
to bring the trade to a virtual halt. The slave trade proved the
most resilient of the three former pillars of the economy and was
still actively pursued in the northern half of the country. Yet, even
that was succumbing to the forces which were gradually eroding
its importance in East Africa.
Several facets of the economy as it existed in 1885 caused
concern among the Portuguese. Unlike Angola, Mozambique had
proved unable to adjust its economy in accordance with changing
world demand. Although Africans had taken the lead in the
collection of rubber, groundnuts and other products, the new
commodities were subject to sharp yearly variations and usually
produced insufficient revenue for the coffers of the administration.
As a result, the usual annual budgetary deficit had to be corrected
by contributions from metropolitan Portugal. In addition,
Portuguese concerns and citizens played only a minor role in the
commerce of Mozambique. Asian middlemen, usually from
British India, reigned supreme in all transactions in the interior.
7
D. Wheeler, ' Gungunhana', in N . Bennett (ed.), Leadership in eastern Africa
(Boston, 1968), 165—210.
JOI

They dominated the retail trade and also purchased the goods
supplied by African cultivators and collectors. Moreover, the
major commercial houses which exported these goods were
almost exclusively controlled by Europeans of other nationalities
or by Asian financiers. Thus, as the age of accelerated imperialism
was about to begin, Mozambique found itself in a much greater
state of disarray than Angola.
THE ESTABLISHMENT OF THE FRONTIERS, 188 5 —1891
Although neither colony was in a satisfactory state, both contained

the nucleus of a colonial party which advocated expansion. These
colonials had been active on a number of fronts. They had been
instrumental in sponsoring such expeditions as those of Serpa
Pinto and of Brito Capelo and Roberto Ivens. They also functioned
as vocal lobbyists in Lisbon, where they argued the case for
increased metropolitan activity in the colonies. Although most of
their pleas had fallen on deaf ears, the colonials were in a position
to act as the vanguard of Portuguese expansion when Lisbon
changed its policy.
The Berlin Conference of 1884-5 provided the stimulus for a
radical change in Portugal's African policy. On the one hand, it
ruled against Portuguese claims to the control of the mouth of
the Congo river, while, on the other, it suggested that future
questions should be decided by effective occupation. Both these
rulings had a significant impact in Portugal. Colonial matters had
previously commanded little attention either in Parliament or with
public opinion. The outcry against what were widely regarded as
the unjust decisions of the conference, however, brought colonial
considerations to the forefront. Secondly, in the aftermath of the
Berlin Conference, those advocating a pro-expansionist policy
were able to manipulate public opinion to overcome the opposi-
tion of those who felt that the colonies, or at least the unprofitable
ones, should be abandoned. They effectively combined the
sentiments stemming from what was described as a national
humiliation with the sense of urgency suggested by the doctrine
of effective occupation to win the day for vigorous action in
Africa.8
Between 1885 and 1890 the expansionists operated on several
9
G. Papagno, Colonialismo t feudalesimo (Turin, 1972), 94—104.
5O2

THE ESTABLISHMENT OF THE FRONTIERS
fronts. Their goal, as exemplified by the publication of the famous

mapa cor-de-rosa, was to win control of all the territory between
Angola and Mozambique. In terms of diplomacy, they felt that
they scored an early victory by concluding treaties with Germany
and France in 1886. In these documents both nations promised
not to interfere in the territory to the south of the Rovuma river
in Mozambique or to the north of the Cunene in Angola. In
addition to diplomatic initiatives, the pro-expansionists sponsored
a number of expeditions in the African interior, which sought to
explore, sign treaties and prepare the way for effective occupation.
In Angola, the expeditions of Henrique de Carvalho to the Lunda
area, which produced treaties of vassalage with innumerable chiefs
and the Lunda king, and those of Artur de Paiva to the southern
regions to sign treaties with the Ngangela and the Ovimbundu,
were among the most important. Similarly, in Mozambique there
were expeditions to Mashonaland under Paiva d'Andrade and to
the Shire highlands under Antonio Cardoso and Serpa Pinto.9
Although France and Germany virtually exempted themselves
from the contest, there remained a number of questions to be
settled with the British and with Leopold, King of the Belgians.
The Berlin Conference had resolved the issue of the Congo mouth,
but failed to delimit frontiers between Leopold's Independent
State and Angola. The question of the extension of Angola's
borders to meet those of Mozambique remained unanswered,
because in 1888 Cecil Rhodes obtained title to the Lochner
concession, which he claimed gave him exclusive control of
Barotseland. Rhodes and the British South Africa Company were
also at the centre of the controversial issues which were brewing
in Mozambique. He concluded a treaty with Gungunyana and
thus laid claim to Gazaland. Similarly, in an effort to secure a route
to the sea, his agents were active in treaty-making and infiltration
in Manyika. Finally, there was the contested region of the Shire
highlands, where Scottish missionaries had worked since the
18 70s and were equally active as agents of British imperialism.
Although the various participants were parrying for the
advantage on all of these fronts, it was the question of the Shire
highlands which hastened a resolve to the outstanding disputes.
Prior to 1889, it seemed unlikely that a major confrontation would
take place between Britain and Portugal over this area. The
• See E. Axelson, Portugal and tie scramble for Africa (Johannesburg, 1967).
5°3

Portuguese had always considered it as part of their sphere and

had despatched a number of treaty-making expeditions. The
British government, on the other hand, expressed no real interest
in the region and in draft versions of treaties, agreed to recognise
it as Portuguese. Lord Salisbury at the Foreign Office, however,
was aware of the domestic pressures which would occur if the
Scottish missionaries came under Portuguese control against their
will. Thus, he was content as long as the area remained neutral.
Two very distinct events, however, obliged Salisbury to alter his
policy. The first was the discovery in 1889 of the Chinde mouth
to the Zambezi, which, by allowing ocean-going vessels direct
access to the river, made it unnecessary to land goods in
Portuguese East Africa and made the prospect of a viable
protectorate in the Shire highlands much more of a reality.
Secondly, in return for other considerations, Cecil Rhodes offered
to meet the expenses of the proposed protectorate, thus removing
the last major obstacle to a British take-over. It was under these
changed circumstances that Salisbury despatched Harry Johnston
to the Shire highlands, armed with the authority to declare a
British protectorate, if necessary. Since Serpa Pinto was preparing
to take a large expedition to the very same area, a confrontation
between the two powers became more likely. Although the
British government sent increasingly vehement warnings to the
Portuguese, Henrique Barros Gomes, the minister of the Marine,
took little note of these messages. He seemingly hoped that Serpa
Pinto's expedition would provide Portugal with the effective
occupation required by international understanding. Although
the area had already been declared under British protection when
Pinto arrived, the expedition continued in pursuit of its objective.
When this news arrived in London, the Foreign Office despatched
the message which has subsequently been known as 'the
Ultimatum'. In a bellicose manner, the British ordered the
Portuguese to retreat from the Shire highlands. Thus, not only
did Salisbury make it apparent that the British position on
this issue was inflexible, he also hastened the resolve of other
outstanding questions.10
Although the Portuguese ministry resigned two days after
meeting the requirements of the ultimatum, the new government
agreed that a general settlement with Britain was needed. The two
countries drafted a treaty which kept the contra-costa principle alive
10
Ibid., 157-252.
504

THE ESTABLISHMENT OF THE FRONTIERS
by creating a 20-mile zone north of the Zambezi in which Portugal

was to have transit and telegraph rights between Angola and
Mozambique. However, the August Convention contained other
provisions which further incensed the already hostile public
opinion in Portugal to the extent that the Cortes refused to ratify
it. With yet another ministry toppled, Portugal found itself in the
difficult position of being unable to satisfy both an enraged public
and an increasingly impatient Great Britain. Although Britain
agreed to a modus vivendi of six months duration dating from
November 1890, during which a new treaty was to be negotiated
and both sides were to respect the position of the other in disputed
territories, Portuguese politicians realised that further delays
would result in the loss of additional territory. As a new series
of negotiations progressed, this sentiment seemingly filtered
down to a more sober public opinion. By June of 1891, both the
Portuguese government and public accepted a treaty less
favourable than the one which had been rejected six months
earlier. Although the final demarcation of borders could not be
resolved for some years, the treaty of 1891 determined the
essential configuration of both Angola and Mozambique and
brought the European phase of the partition to a close.11
Under the terms of the treaty the outline of Mozambique was
somewhat different from what it would have been had the August
Convention been ratified. It granted considerably more territory
to Portugal north of the Zambezi, but made no provision for the
contra-costa ideal. This increase of territory was compensation for
granting the lion's share of Manyika to Rhodes and the British
South Africa Company. Yet, despite his entreaties and the
provocative activities of his agents in south-eastern Africa,
Rhodes was denied the strip of territory which would have
provided Rhodesia with a direct outlet to the sea. Similarly, it
refused to recognise his claims to Gazaland, although he had
obtained mineral concessions in return for supplying Gungunyana
with 1,000 rifles and other considerations. Thus, despite putting
an end to the idea of linking Angola with Mozambique, the treaty
of 1891 provided European recognition of Portugal's right to
exploit a vast territory in south-eastern Africa.12
The essential configuration of Angola was also determined in
" Great Britain, Admiralty, A manual of Portuguese Bast Africa (Oxford, 1920),
492-503.
12
P. R. Warhurst, Anglo-Portuguese relations in south-central Africa, itfo-1000
(London, 1961), 14—109.
5°5

1891, but the six-month period following Portugal's refusal to

ratify the August Convention resulted in the loss of additional
territory. In the original version Barotseland was divided between
Angola and Northern Rhodesia. During the interim, Rhodes
protested that the Lochner concession entitled the British South
Africa Company to control all of Barotseland. Thus, instead of
being divided along the upper Zambezi, all of Barotseland fell into
the British sphere. However, if the settlement of the south-eastern
border proved unfavourable, the Portuguese were quite satisfied
with the resolution of the north-eastern frontier with the Congo
Independent State. The area had been the scene of intense
competition, and, at one time, Leopold had purchased a gunboat
and actually contemplated delivering his own ultimatum. Yet, in
1891, he was prevailed upon to settle the issue over the conference
table. The resultant negotiations proved successful for the
Portuguese, as they were able to secure the Kasai as the north-
eastern frontier, even though their influence in that region had
been minimal.13 Thus, with the southern frontier having pre-
viously been settled with Germany, 1891 marked the end of the
diplomatic aspect of the partition.
THE OCCUPATION OF THE INTERIOR REGIONS,
1895-1905
Due to a serious financial crisis which hit Portugal in 1890, the
end of the European phase of the partition did not result in the
implementation of the grand designs for colonial development
which had been promised. The crisis was announced by the
collapse of the Baring financial house, but its roots lay in the
serious overspending which had occurred during the 1880s.
Metropolitan Portugal was obliged to abandon the gold standard,
default on interest payments to foreign creditors, and undergo a
period of inflation. It was not long before the effects were felt in
the colonies. Since the large sums of money poured into Angola
had produced no economic gains, measures of financial stringency
were imposed. Similarly, the depression necessitated such serious
retrenchment in Mozambique that virtually all but the most
crucial public services came to a halt. Because of the prevailing
13
E. dos Santos, A qutstao da Ljmda, 1SSJ-1S94 (Lisbon, 1966).
506

THE OCCUPATION OF THE INTERIOR REGIONS
climate of austerity there was little inclination in either colony to
initiate new schemes or even to attempt to establish superiority
over the vast majority of the African population. In fact, it seemed,
once its claims to African territory had been recognised, that
Portugal would again lapse into a period when all major initiative
would be indefinitely postponed.
A poverty-stricken Portugal might have allowed the lull in
African activity to continue had it not been for a revolt in the
neighbourhood of Lourenco Marques. In 1894 the Ronga chief,
Mahazul, took up arms against the Portuguese because his people
were complaining about unremunerated labour and a sudden
increase in the hut tax. Many of the other Ronga chiefs supported
him, and, although no direct assault was made, the position of
the town was described as precarious. Since Lourenco Marques
provided the vital outlet from the Transvaal on which the
economy of Mozambique depended, large numbers of Portuguese
and Angolan troops were imported to deal with the crisis. This
additional firepower soon brought an end to the hostilities around
the town and successfully quelled the rebellion.
The importance of the revolt in Lourenco Marques, however,
stems from the fact that it was directly responsible for initiating
the long-anticipated confrontation between the Portuguese and
Gungunyana. Following his defeat, Mahazul sought refuge in
Gazaland with the Nguni chief. Since the Portuguese would
not have dared to interfere with Gungunyana under ordinary
circumstances, this flight seemingly assured the Ronga chief of his
freedom. However, the presence of thousands of troops in the
territory, commanded by the aggressive high commissioner,
Antonio Enes, represented a distinct break with normality.
Although many observers believed that the Portuguese were
outmatched, Enes was one of the few who felt that Shangane
power was overrated. Acting on this assumption, he ordered
Gungunyana to surrender the fugitive or face a punitive expedi-
tion. Gungunyana had done all in his power to avoid hostilities.
He had attempted to manipulate the various European forces
which surrounded him in an attempt to safeguard his independence
and had demonstrated great restraint in not attacking the
Portuguese when they were at their weakest. However, since he
apparently assumed that not answering this challenge would result
5O7

in the loss of his independence, he refused to comply with the
ultimatum and prepared for war.14
The ease of the subsequent Portuguese victory testified to the
astuteness of Enes's assessment. The power of the Shangane
proved less formidable than many had imagined. It suffered from
the fact that the subject peoples of the Shangane had not been as
thoroughly assimilated as in other Nguni states, from the large-
scale labour migration to South Africa, and from the inability of
their soldiers to make use of the firearms at their disposal. Thus,
the three-pronged Portuguese offensive which moved into
Gazaland found that the size of the force which confronted them
was much smaller than anticipated, that the Tsonga recruits
quickly deserted their Nguni leaders, and that the opposing
rifle-fire invariably was directed over their heads. Therefore, the
taking of Gungunyana's capital, followed shortly thereafter by the
tracking down of the fleeing king, was accomplished by the end
of 1895.«
Although the war itself was somewhat of an anti-climax, its
consequences were important and immediate. Despite a sub-
sequent unsuccessful revolt in Gazaland, it secured Portuguese
possession of what was to be the most important part of the
province. Moreover, with the removal of Gungunyana as a force,
it proved a simple matter to occupy the territories to the north
of the Sabi river which had been in his sphere of influence. In
addition to facilitating the occupation of southern Mozambique,
the victory provided the Portuguese with an enormous psycho-
logical lift. It is not too much to say that it restored pride in a
nation whose confidence had been badly shaken by the Ultimatum
and other sources of embarrassment. The inspirational value was
so great, in fact, that, for one of the few times during the partition
era, the Portuguese began to initiate action rather than respond
to it.
The aggressive spirit which emerged from the defeat of
Gungunyana was largely responsible for a Portuguese attempt to
bring northern Mozambique under control. Unlike many of his
predecessors, Governor-General Mousinho de Albuquerque was
possessed of a vision of what Mozambique should become and
14
D. Wheeler,' Gungunyane, the negotiator: a study in African diplomacy', J. Afr.
Hut., 1968, 9, 4, J8J-602.
15
A. Enes, Ayres d'Ornelas, J. Azevedo Coutinho, J. Teixeira Botelho, and many
others wrote accounts of the campaigns.
508

seemingly had the support to bring these plans to fruition. Before
anything else could be successfully accomplished, he argued, it
would be necessary to defeat the remaining independent African
groups. His first objective was to impose a hut tax on the Swahili
and Makua peoples of the northern Mozambique coast. By 1897,
however, this attempt to interfere with their independence had
provoked a resistance movement which was supported by almost
all of the Makua chiefs, the Swahili sultans, and even Portuguese
merchants. Although large numbers of Portuguese troops were
active in the field during 1897, they were able to make little
headway against the well co-ordinated guerrilla tactics and sharp
marksmanship of an adversary which was well informed in
advance of Portuguese movements. Eventually the necessity of
putting down the 1897 revolt in Gazaland obliged the Portuguese
to suspend their northern operations. They had established a
number of coastal forts and signed some meaningless treaties, but
had fallen far short of Mousinho's goal of bringing the area under
effective Portuguese control.I6 Thus, by 1898, effective occupation
had become a reality only in most parts of the southern half of
the province.
The contemporaries of Enes and Albuquerque in Angola,
however, had not pursued a similar policy of expansion. Between
1891 and 1899 military operations were undertaken only in order
to secure sensitive border regions or to protect a threatened
commercial interest. For example, some small-scale actions were
fought in the Moxico region in order to establish forts along the
extensive eastern frontier and prevent encroachment by Rhodes
from Barotseland. Similarly, 1896 witnessed a number of oper-
ations along the northern border with the Congo Independent
State. More important campaigns resulted from the desire to
secure regions of economic importance. Between 1893 and 1895
troops were sent into the traditionally hostile territory on the
south of the Cuanza river, where the opposition of the Kimbundu-
speaking population threatened the rubber routes. In 1898, a
rebellion caused by an epidemic of rinderpest obliged the
Portuguese to move into the Humbe region, the centre of the
southern cattle trade. When no commercial interest was at stake,
however, the officials of Angola preferred not to initiate military
16
N. Hafkin, 'Trade, society, and politics in northern Mozambique, t. 1755-1913'
(unpublished Ph.D. thesis, Boston University, 197}), 371-98.
509

campaigns. The result was that by 1898 little of the territory

recognized as Portuguese had been occupied.
In 1898, plans for expansion into the interior of both colonies
were dealt a blow by the appearance of yet another financial crisis
in Portugal. Although the unresolved struggle with foreign
creditors was the immediate cause of the collapse, the heavy
spending which had attended the military campaigns in Mozam-
bique certainly contributed to the problem. Critics in Portugal
took special aim at Mousinho de Albuquerque, accusing him of
having wasted enormous sums on unnecessary and glory-seeking
military campaigns. With the value of the currency falling to half
the level of 1891, colonial governors were ordered to desist from
all offensive campaigns. Although Albuquerque resigned in
protest against what he labelled the lack of metropolitan support,
a financially troubled Portugal again found itself unable to put its
plan into operation.
Although the worst of the financial crisis had come to an end
by 1902, the extension of Portuguese control over their colonies
continued to be a slow process. Rather than taking the initiative
in a concerted campaign designed to wrest control of the
independent areas, they continued to react to specific and often
localised problems. In the whole northern half of Angola there
were protracted but small-scale operations against the fragmented
polities lying outside Portuguese control, usually caused by
trading conflicts as Portuguese and mestizo traders penetrated ever
further inland. The largest campaign was against the Ndembu, in
hilly and jungle-covered territory close to Luanda. The Portuguese
lost many men through disease and ambush and wasted vast
amounts of money in order to reduce the Ndembu area to an
apparent state of 'pacification' by 1908. But the area was to
continue to give constant trouble in later years.
In the central highlands and the southern flood-plains, African
kingdoms had retained a much higher degree of centralisation
than in the north, although there were no polities of the size of
the Shangane state in southern Mozambique. At first sight they
appeared to constitute a more formidable foe than the fragmented
northern peoples, but in general they proved easier to conquer
after the first main clash and, above all, easier to administer after
defeat. This was particularly noticeable in the so-called Bailundo
war of 1902. In that year the price of wild rubber fell disastrously
510

on world markets and the growing conflict between Portuguese

and Ovimbundu traders came to a head. Scores of Portuguese and
mestizo traders were assassinated, and the leader of the rebellion,
Mutu-ya-Kavela, was proclaimed king of Bailundo. But attempts
to draw other Ovimbundu kingdoms into the movement were
generally a failure, in spite of the use of various forms of cults
to create a feeling of Ovimbundu solidarity. This failure was due
largely to the growth of slave-raiding between different Ovim-
bundu kingdoms, as the export of labour to the cocoa plantations
of Sao Tome became greater every year. In the neighbouring
kingdom of Wambu, Mutu-ya-Kavela was feared as a terrible
slave-raider and was unacceptable as the leader of a unified
rebellion. The Portuguese were therefore successful in putting
down the revolt in a very short time, and used the opportunity
to extend their control over all the central highlands.17
In the southern flood-plains the Portuguese experienced more
difficulty in subduing the small but highly centralised Ovambo
kingdoms. They thought it necessary to occupy this remote and
sparsely populated area in order to settle a lingering boundary
dispute with Germany and put an end to the regular Ovambo
raids, which were penetrating deep into central Angola. But in
building up their systematic raiding industry, the Ovambo had
accumulated a large stock of modern rifles and had learnt how
to use them with devastating effect. A first Portuguese column
in 1904 was ambushed and lost over 300 men. Two successive
campaigns in 1905 and 1906 were unable to avenge this humiliating
defeat, and in 1907 the largest expedition ever mounted in Angola
to that date was sent to deal with the Ovambo. But the Portuguese
scored only a partial success in 1907. They defeated the Mbadya
people, responsible for the 1904 ambush, but they failed to
attack and occupy the Kwanyama kingdom, the largest in
Ovamboland.18
By 1907 the Portuguese felt their position sufficiently secure in
Angola to reimpose the collection of hut tax. Yet the occupation
of the territory was far from complete. Only the coastal zones and
the densely populated western and central highlands were
considered as occupied. Otherwise, except in isolated areas, the
17
D . Wheeler and D. Christensen,' To rise with one mind' in F. Heimer (ed.), Social
change in Angola (Munich, 1973), J4~92.
18
R. Pelissicr, 'Campagncs militaircs au Sud-Angola, 1885—1915', in Cabicrs d'etudes
africaints, 1969, 9, 54-12 3.
5"

vast and sparsely populated sandy plains which comprised most
of eastern Angola had yet to feel the impact of colonial rule.
Moreover, those territories which had been conquered were far
from secure. Before effective control would actually become a
reality, there would be a series of serious revolts throughout the
highland regions.
Events in Mozambique after the financial crisis followed a
course similar to that in Angola. In 1902 the Portuguese undertook
a major campaign in the Barwe region and in the Tonga pra^ps
of the Gorongoza, the last independent regions south of the
Zambezi. Portuguese influence in these areas had existed only in
the form of Manuel Antonio de Souza, an adventurous praêro
who had collected tribute from the Makombe, chief of Barwe.
Following de Souza's assassination, there was a contest for the
position of makombe between Nchanga, who had disposed of the
adventurer, and Shupatura, the legitimate ruler. Despite the role
played by Nchanga in liberating the territory from the forces of
de Souza, the Barwe continued to recognise the leadership of
Shupatura. However, when the latter seemingly accepted sub-
mission to the Mozambique Company in 1900, most of the Barwe
rallied around Nchanga and his call to resist the white man.
Although Barwe was included in the company's sphere, the fact
that its agents were denied access to the territory obliged it to call
on the government of Mozambique to deal with the situation. In
1902 a force under the command of Azevedo Coutinho was
entrusted with the dual charges of extending Portuguese control
to the Gorongoza region and suppressing the opposition of
Nchanga. Despite determined opposition, the largely unco-
ordinated resistance movement collapsed by the end of the year
and both regions succumbed to Portuguese control.19
The occupation of the remainder of the province, however, was
more protracted. Although Barwe had been the major target of
the 1902 expedition, the force seized the opportunity to crush
resistance in the pra^ps in the lower Zambezi valley. By 1904
expeditions sponsored either by the government or the Zambezi
company had extended control along the upper stretches of the
river and along the tributaries both to the north and the south.
The occupation of the northern districts, however, was not to
10
J. Azevedo Coutinho, MtmSrias de urn velbo marinbtiro e soldado de Africa (Lisbon,
1941), 529-660.

come until a later date. Although the Portuguese scored an

important victory in 1906 when they defeated the Namarral chief,
Marave, other Swahili and Makua groups along the coast
continued to resist. Not until 1910, with the defeats of the Sultan
of Angoche and Farelay, the ruler of the coast opposite
Mozambique Island, was this opposition brought to an end.20
Once the Portuguese had secured the littoral, the Makua chiefs
rapidly succumbed to the government agents who increasingly
penetrated the interior. Events in the territories of the Niassa
Company in the far north followed a similar pattern. For almost
twenty years a succession of Yao chiefs, bearing the title mataka,
had successfully resisted the feeble attempts of the company to
extend Portuguese influence. By 1912, however, after a series of
campaigns had successfully established forts among the Makua
and Makonde, the reigning mataka was obliged to surrender.21
The submission of the last major chief bore testimony to the fact
that a military occupation of the province had been completed.
SOCIAL AND ECONOMIC STRUCTURES, 1895-1905
The advent of colonial rule in Mozambique did not produce many

changes in the colonial social structure. Perhaps the greatest
impact was in the Zambezi valley, where absentee European,
landlords were coming to replace the racially mixed praêro who
had dominated the region during the nineteenth century. The
system by which the lessee of a prazo was granted a monopoly
of commerce also obliged Asian merchants to seek their fortunes
in other parts of the colony. Yet, despite the fact that most
Portuguese considered Indians as a menace to be done away with
as soon as possible, no effective measures were taken to threaten
their still dominant position in the retail trade of the colony. The
inability of the Portuguese to dislodge Asians from control of
commerce combined with the absence of opportunities for
small-scale farming was partly responsible for preventing large-
scale immigration by the illiterate peasantry of Portugal. Thus,
although there was a growth in the European population of
20
M. D . D . Newitt, 'Angoche, the slave trade and the Portuguese, c. 1844-1910',
J. Afr. His/., 1972, 13, 4, 659-72. Namarral was the name given to a confederacy o f
the central Makua.
21
Manual of Portuguese East Africa, 510-12.
513

Mozambique, a significant portion was comprised of apolitical

foreigners employed on short-term contracts. Since the residence
in Mozambique of both the Portuguese and foreign populations
was often transitory, neither group was very active in expressing
its discontent with local conditions.
In Angola, on the other hand, the imposition of a more effective
colonial rule produced conflicts of major proportions between the
recent arrivals and local Creole society. In the post-partition era
poor whites and metropolitan officials began to enter Angola in
increasing numbers. The arrival of each group contributed to a
reduction in the status of the indigenous population. The poor
whites were instrumental in increasing racist feeling against
Africans, whether they were 'uncivilised' or 'assimilated'. At the
other end of the scale the importation of metropolitan Portuguese
to man the colonial administration resulted in the removal of
Creoles from many of the positions they had previously occupied.
The Creole response to the new challenges was somewhat
ambivalent and disunited. All sectors could agree on their
opposition to the growing influence of Lisbon in their affairs.
However, their views on racial matters ranged from the ambivalent
position of the majority who despised the ' uncivilised' African
while decrying the growth of racism within the colonial nucleus,
to the pro-African radical assimilado journalists of Luanda, who
tried to defend Africans from harsh labour policies and glorified
traditional African culture. Although each sector of opinion
wanted a greater degree of autonomy, these divisions within
Creole society weakened its opposition to metropolitan control.22
Despite the increasing presence of metropolitan officials, the
same lack of resources which delayed the occupation of the
African territories obliged Portugal to link colonial rule with
foreign capital. This was especially the case in Mozambique,
where almost a third of the colony was entrusted to the
administration of chartered companies. Subject only to certain
veto powers retained in Lisbon, the Mozambique Company and
the Niassa Company were granted the right to rule over their
respective spheres as virtual sovereign powers. Although control
of the companies was theoretically to remain in the hands of
Portuguese based in Lisbon, in actual fact the major decisions
were made in boardrooms in London, Paris or Brussels. Of the
" Wheeler and Pelissier, Angola, 90-108.

SOCIAL AND ECONOMIC STRUCTURES
two, the Mozambique Company was the more successful, largely

because it possessed a readily exploitable resource in the transit
traffic between its port of Beira and Southern Rhodesia. The
Niassa Company, on the other hand, exemplified all that could
be wrong with a chartered company. The directors were more
concerned with in-fighting and speculation than with laying the
foundation for the exploitation of a colony. As a result, fifteen
years after the granting of the charter, virtually nothing had been
accomplished in its territory.
Foreign capital also provided many of the essential services of
Mozambique. Colonial planners such as Enes and Albuquerque
realised that the limited quantity of Portuguese investment capital
would be insufficient for the needs of the colony. Potential foreign
investors, therefore, were encouraged to aid in the development
of Mozambique. As a result, transportation, communications, the
municipal services, the major commercial houses and commercial
agriculture were all dominated by imported capital. The role of
foreigners was particularly important in railway construction and
the search for export commodities. By 1900 they had completed
the vital rail links between Lourenco Marques and the Witwaters-
rand and between Beira and Southern Rhodesia. They also had
received a significant number of land concessions throughout the
province, on which experiments were proceeding in the search for
viable export crops.
The advent of colonial rule in Angola, however, was not
accompanied by as great an influx of foreign capital. Although
concessions in eastern and central Angola were granted to
foreign dominated companies, the Portuguese Parliament refused
to ratify the proposed charters. The Mocamedes Company, which
was financed by interests connected with Rhodes, was the only
chartered company allowed to function in the colony. Unlike its
counterparts in Mozambique, however, it was denied access to
customs duties and to the coast.23 Although the hostile attitude
of the government with respect to chartered companies did not
extend to all foreign capital, the economy of Angola remained less
dependent on alien sources of revenue than did Mozambique.
Foreign investment was largely confined to such activities as the
provision of services in the principal towns and the construction
11
H. Drechsler, 'Germany and Southern Angola, 1898-1903', Presence Africaine
(1962), 42-3, 51-69.
515

of the Benguela railway. However, it was noticeably absent from

the productive spheres of the economy, such as coffee-planting
or fishing.
The absence of imported capital, however, did not mean that
there was no foreign input in the colonisation of Angola. In fact,
missionaries from several European nations and the United States
played a crucial role in the evolution of the colony. Although the
Portuguese were always distrustful of the large number of
predominantly American and English Protestant missionaries
who flocked to Angola, they were bound by international treaty
agreement to permit their entry. Protestants were accused of
complicity with the supposed machinations of other colonial
powers and of fomenting dissent among the African population.
The Portuguese initially attempted to employ Catholic mission-
aries from their Colonial Seminary of Sernache to combat the
growing influence of the Protestants. When it became apparent
that there was an insufficient number of Portuguese for the
undertaking, the government turned to the French-based Holy
Ghost Fathers, who came to hold a virtual monopoly of Roman
Catholic mission work. By providing them with ample subsidies,
the government hoped to counteract what it considered the
'denationalising' influence of the Protestants. The politics of
religion, therefore, determined that mission influence in Angola
was much more significant than in Mozambique.24
The imposition of colonial rule did not have an immediate
impact on the Angolan economy, which continued to rely heavily
on products of hunting and collecting from the interior and on
a colonial sector afflicted by rapidly alternating booms and slumps.
There was a distinct shift southwards in the economic centre of
gravity of the colony, although this had very little to do with the
imposition of colonial rule. At the turn of the century, northern
Angola was afflicted by a devastating epidemic of sleeping
sickness, which depopulated large areas and lasted many years.
Labour for plantations became very difficult to obtain and
transport was paralysed by the lack of porters. The situation was
aggravated by the slump in world prices for coffee after 1896. By
1906 exports of coffee had fallen to about a quarter of their
value ten years previously, and many plantations were simply
abandoned. The other great pillar of the northern economy, wild
24
M. Samuels, Education in Angola, 1878-1914 (New York, 1970).
516

rubber, was adversely affected by the lack of porters and was

redirected to more southern ports. There was also a significant
shift in the sources of rubber. Whereas previously rubber had
come from creepers and trees which grew mainly in northern
Angola, from the 1890s the majority of Angola's rubber exports
came from a root which grew in the dry sandy plains of the
south-east.
The plantations of central and southern Angola enjoyed a boom
in the 1890s consequent on the tariffs adopted in 1892 in order
to stop imports of foreign spirits. Locally grown sugar cane was
distilled into aguardente (fire-water) and used for the rubber trade
inland. The oases of Benguela were the largest producers in
Angola, although many coffee plantations in the hinterland of
Luanda were also converted to sugar cane. In the far south,
economic prosperity was based on a booming export trade in
cattle and dried fish. These commodities were supplied to many
plantations in the Gulf of Guinea and especially to those on Sao
Tome, where dried fish became an essential ingredient in the diet
of plantation labourers. Benguela was the greatest beneficiary of
all these economic changes, and by 1905 it had become the chief
port of Angola, while Luanda lay 'bankrupt and beautiful'.25
But in the early years of the new century, Angola was
experiencing the beginning of a slump. Rubber prices declined
slowly, and it became increasingly evident that African wild
rubber could not compete with the plantations of South-East Asia
once these came into full production. Angolan plantations were
hit first by the collapse of coffee prices in 1896, and then by
administrative measures taken from 1901 to curtail the production
of aguardente. Portugal undertook these measures reluctantly, as
a result of international pressures, but by 1905 the aguardente
industry was clearly doomed. Attempts to produce sugar instead
failed, as did efforts to replace sugar cane with cotton. In the far
south cattle exports never fully recovered from the terrible
rinderpest epidemic of 1897—9, and dried fish began to face severe
competition from other suppliers. Thus by 1905 the whole colony
was suffering from a slow but steady recession, and this un-
doubtedly goes a long way toward explaining the violence of the
clashes between Portuguese and African traders in the interior,
which were responsible for much of the military activity.
15
H. Ncvinson, A modern slavery (New York, 1906).
517

The advent of colonial rule, however, brought little change in

the nature of the exploitation of labour in Angola. Despite reams
of legislation guaranteeing the rights of labourers, the situation
as it existed on the plantations differed little from slavery. Workers
continued to be secured by purchase from the interior. Moreover,
they were viewed as disposable property and could be sold at the
discretion of a plantation owner. Although the law provided
proscriptions against corporal punishment and guaranteed that
nominal wages were to be paid, these safeguards were often and
easily avoided. Thus, much of Angola's production continued to
be based on a slave system which had been abolished in name
only.26
Since Mozambique did not possess a similar plantation
economy, the exploitation of labour in agricultural pursuits
differed in some respects. Yet, as the conquest of the province
proceeded, the roots of an almost equally pernicious system began
to take hold. Africans residing on lands conceded to various
European interests were obliged to pay taxes in commodities
determined by the concessionaires. Moreover, either a private
police force or government troops could be called on to make
certain that any surplus was sold to the proprietor at rates well
below the market value. Not only did this system deprive African
producers of the opportunity to market their goods advantage-
ously, it often resulted in serious decreases in the production of
the staples used for home consumption. Although forced culti-
vation of this nature was generally confined to the Zambezi pra^ps
in 1905, the system was to be extended to most of northern
Mozambique once those territories had been conquered.
Despite the incipient system of forced agricultural labour, the
productive sector of the Mozambican economy had shown little
development by 1905. Industrial production had always been
limited to the production of aguardente for export to the South
African Republic and local consumption. However, pressure from
certain interests in Portugal and severe restrictions imposed by
the British in the Transvaal after the Anglo-Boer war, caused a
decline in production. The agricultural sector had as yet not
reaped the benefit of the experimental projects which had been
undertaken. Of the various crops only sugar had demonstrated
that it could both thrive in the climate and bear the cost of
26
J. Duffy, A question of slavery (Oxford, 1967).
518

transportation. Yet, in 1905 most of the sugar industry remained
in the developmental or planning stages. Therefore, the produc-
tion of rubber, which reached its peak in that year, was the most
valuable export of the colony. Otherwise, exports were limited
to small quantities of the various oil-producing products. Thus,
in comparison with Angola, the productive sector of the
Mozambican economy remained at a relatively low level.
Although Angola and Mozambique differed in their productive
capacities, both were being manipulated in such a way as to ensure
that they suffered from a dual dependency. On the one hand, their
relationship with metropolitan Portugal was one of classical
colonial dependence. The description of Portuguese imperialism
as uneconomic is certainly an over-simplification.27 The colonies
provided certain segments of the financial establishment with
ample opportunity for profit making. For example, the Banco
Nacional Ultramarino, which functioned in both colonies under
extremely favourable circumstances, provided its Portuguese
stockholders with considerable dividends. Moreover, Portuguese
industry benefited from both its importing and exporting
relationships with the colonies. On the one hand, it ignored the
colonial produce for which it had no need, thus requiring those
commodities to compete on the world market. On the other, it
consumed virtually all of the sugar produced in the colonies. The
increasingly centralised control that Lisbon exercised over the
colonies ensured tariff manipulation in favour of metropolitan
interests. Thus, through selective duties imposed on the colonies,
goods of Portuguese manufacture, such as wines, shoes, clothing
and tobacco, found virtually no competition in the colonies. The
favourable balance of trade with the colonies, therefore, proved
very beneficial to many interest groups in Portugal.
In addition to being dependent in its relationship with Portugal,
each of the colonies found itself in the position of being exploited
by its neighbours. The economy of Mozambique, for example,
was almost entirely dependent on South Africa and Southern
Rhodesia. The transit trade from Lourenco Marques to the
Transvaal and from Beira to Southern Rhodesia was by far the
most significant economic activity in the province. Rail transport
and the attendant activities not only provided the major source
of employment in the colony, but contributed valuable sums to
27
R. J . H a m m o n d , Portugal and Africa, igij-ifio ( S t a n f o r d , 1966), 3 3 5 .
519

the coffers of the colonial government. The export of labour to

the mines of the Witwatersrand, however, was an even more
important source of revenue. The exploitation of migrant
labourers, whose numbers approached 60,000 by 1905, aided the
provincial government in a number of ways. The government
received a fee for each worker recruited in Mozambique, had his
hut tax deducted from his earnings, and was provided with
valuable foreign exchange from the gold which the workers
received - in wages. Thus, in addition to the ever-growing hut
tax, the direct and indirect contributions derived from its
relationship with its neighbours enabled Mozambique to
function.
The exploitation of Angola by its neighbours also involved the
use of black labourers. Yet the export of labour to the islands of
Sao Tome and Principe proved more pernicious than the migrant
labour system in Mozambique, because its dividends to the colony
were insubstantial and caused extreme dislocations among the
African population. The early years of the twentieth century
proved exceedingly productive ones for the cocoa plantations of
these Atlantic islands. Sao Tome, in fact, was the only Portuguese
colony which regularly returned a large surplus of revenue. Its
prosperity, however, was absolutely dependent on the cheap
labour provided by Angola. It was estimated that between 1885
and 1903, 56,189 servtfaes were exported from Angola.28 Unlike
Mozambique, where migrant labourers exercised some freedom
of choice, the serviqal was virtually in the position of a slave.
Although he was allegedly a labourer contracted for five years,
more often than not he was captured in a slave raid, transported
in chains first to the coast and then to the islands, whence he rarely
returned. Thus, because of the needs of the planters in Sao Tome
and Principe, many Africans in Angola lived in a state of
insecurity similar to that which prevailed during the slave trade.
By 1905 new patterns of dependency and production for the
export market had significantly modified the place of Portugal's
two major colonies within the imperial whole. On the one hand,
Angola had progressed slowly and erratically in the economic field
since 1885, so that the total revenue of the colony was now only
half that of Mozambique. In the 1870s Angola had occasionally
produced budgetary surpluses, but by 1905 there was a regular
28
A. Marvaud, Lt Portugal el ses colonies (Paris, 1911), 208-15.
520

and extremely large deficit. But Angola could count on two

pressure groups to maintain expenditure at a high level. The white
settlers of the south exercised a political influence in Lisbon out
of all proportion to their numbers and wealth, while the powerful
textile lobby in Portugal looked to Angola as one of its major
markets. On the other hand, Mozambique, which had earlier been
viewed as the backwater of the empire and produced less revenue
than Goa, was now entering a period of great stability and relative
prosperity. The revenue derived from migrant labour, rail
transport to the Transvaal and hut tax, not only provided the
largest revenues of the whole empire, but also guaranteed a
budgetary surplus. But as there was no vocal settler community,
and as Portuguese commercial interests were much weaker,
Mozambique could not stop its budgetary surpluses from being
used to subsidise other parts of the empire, and notably Angola.
Thus Mozambique became ever more dependent on its neighbours
and lost the opportunity for productive investment, while the
Portuguese state wasted huge sums in Angola on unproductive
projects such as the building of the Mocamedes Railway to pander
to the demands of white settlers.
B. MADAGASCAR AND FRANCE, 1870-1905

RAINILAIARIVONY, 1870-85
With his power henceforth firmly established, Rainilaiarivony,20

the Prime Minister and consort of Queen Ranavalona II, could
govern the country without too many worries, and was able to
turn his attention to reforms. He was motivated to contemplate
these reforms by his desire to strengthen the country and make
his government conform to a European model. Christianity,
which he had just adopted, entailed a certain number of changes
which the ever more numerous and influential English mission-
aries were urging on him. But the prime minister had too much
common sense, and had learned his lesson too well from the
dreadful example of Radama II, to make too abrupt a break with
the customs and traditional feelings of his people. His reforms
were prudent, taking account both of the needs of his subjects
and the requirements of his authority.
"> See Cambridge history of Africa, v: c. 1790—c. 1X70, 4IZ-16.
521

and extremely large deficit. But Angola could count on two

pressure groups to maintain expenditure at a high level. The white
settlers of the south exercised a political influence in Lisbon out
of all proportion to their numbers and wealth, while the powerful
textile lobby in Portugal looked to Angola as one of its major
markets. On the other hand, Mozambique, which had earlier been
viewed as the backwater of the empire and produced less revenue
than Goa, was now entering a period of great stability and relative
prosperity. The revenue derived from migrant labour, rail
transport to the Transvaal and hut tax, not only provided the
largest revenues of the whole empire, but also guaranteed a
budgetary surplus. But as there was no vocal settler community,
and as Portuguese commercial interests were much weaker,
Mozambique could not stop its budgetary surpluses from being
used to subsidise other parts of the empire, and notably Angola.
Thus Mozambique became ever more dependent on its neighbours
and lost the opportunity for productive investment, while the
Portuguese state wasted huge sums in Angola on unproductive
projects such as the building of the Mocamedes Railway to pander
to the demands of white settlers.
B. MADAGASCAR AND FRANCE, 1870-1905

RAINILAIARIVONY, 1870-85
With his power henceforth firmly established, Rainilaiarivony,20

the Prime Minister and consort of Queen Ranavalona II, could
govern the country without too many worries, and was able to
turn his attention to reforms. He was motivated to contemplate
these reforms by his desire to strengthen the country and make
his government conform to a European model. Christianity,
which he had just adopted, entailed a certain number of changes
which the ever more numerous and influential English mission-
aries were urging on him. But the prime minister had too much
common sense, and had learned his lesson too well from the
dreadful example of Radama II, to make too abrupt a break with
the customs and traditional feelings of his people. His reforms
were prudent, taking account both of the needs of his subjects
and the requirements of his authority.
"> See Cambridge history of Africa, v: c. 1790—c. 1X70, 4IZ-16.
521

In 1873 the queen and her consort accompanied by the entire

court travelled to Betsileo, and spent a month at Fianarantsoa,
where they promulgated a code of laws specially formulated for
that area: Imerina was still governed by the Code of 101 Articles.
The other subject peoples retained their traditional customs and
their chiefs under the supervision of Merina governors. The sale
of alcohol was prohibited, at least in Imerina, as had already been
the case in the days of Andrianampoinimerina, but this did not
prevent some members of the upper class, in particular some of
the royal princes, drinking to excess. In 1877 it was decided to
free the ' Mozambique' slaves, that is, slaves of African origin.
This had long been demanded by Britain as part of its campaign
to end the slave trade. The owners kept only their Malagasy slaves,
who were by far the greater number. Some of the freed slaves were
settled in Sakalava territory as colonists. In 1878—9 the army,
which had lost a great deal of its effectiveness since the days of
Radama I and his conquests, was reorganised. Military service was
made obligatory over the age of 18. In principle it was to be for
five years, but in practice it amounted to three months' training,
with a subsequent liability to call-up at any time. This measure
entailed the discharge of 6,500 old soldiers, whom the Prime
Minister distributed amongst the villages of Imerina with the title
of sakaiâmbobitra (friends of the villages). They were given the
task of general supervision and of acting as official registrars for
the state (with the aid of secretaries with school training). They
were paid by means of a tax on the certificates they issued. The
Prime Minister thus provided them with a pension, whilst
strengthening his own power over the country areas and
diminishing to an equivalent extent that of the land-holding
aristocracy.
In March 1881 the Code of 305 Articles was promulgated for
Imerina, replacing the 1868 Code of 1 o 1 Articles. It enacted new
rules concerning civil and criminal law and procedure. Previously
accepted customs not modified by this document were expressly
preserved. Some of these customs had already been abolished, as
being incompatible with Christianity, such as polygamy and the
unilateral repudiation of wives (which was replaced by legal
divorce). Corporal punishments were abolished except for irons,
and fines were fixed in cattle and in money. Three courts of law
were set up at Tananarive. This code was so well adapted to the
5"

MADAGASCAR: RAINIL AIA RI VON Y, 1870-85
country that later on it was retained under the French regime and
extended to cover virtually the entire island.
The sakaiâmbohitra, known from now on as antily (watchmen),
were retained but, having little education and a merely informative
role, they were ill-equipped to enforce the new laws and over-
whelmed the Prime Minister with unimportant details and
requests for directives. To remedy this situation, he took two
kinds of measures. Firstly, he set up in 1881 eight ministries
(Interior, Foreign Affairs, War, Justice, Law, Trade and Industry,
Finance and Education), each having several departments — a
system modelled on those of Europe. It did not function well in
Madagascar, despite the large staff employed. This included 21
high-ranking officials, 215 executive civil servants and 24} clerks.
Some of those holding office were incompetent and most were
afraid to take any decision for fear of angering the authoritarian
Prime Minister. In vain, he urged them to use their responsibility
wisely; it had no effect. Secondly, he encouraged the activities of
the fokorColona (popular assemblies in the villages), which Andria-
nampoinimerina had successfully employed to ensure the main-
tenance of order and the enforcement of law. These bodies could
draw up by-laws, and in Imerina at least were well established and
proved useful.
Although they set up a royal chapel to ensure their own
independence, the Queen and her consort gave encouragement to
the Protestant missionaries, especially the British. The London
Missionary Society (LMS) alone possessed 1,200 churches and an
equal number of schools. Schooling was the responsibility of the
missions, but the Prime Minister had decreed the principle of
compulsory education, and saw that it was applied. The villagers
built the schools and provided the teachers with rice. The LMS
had established three high schools for Theology, Education and
Medicine. A few young men had been sent to Europe, of whom
two returned as doctors. A hospital and several dispensaries were
set up. The Anglican, Quaker and Catholic missions were smaller,
but were gradually increasing in importance. Some subordinates,
trying to curry favour with those in power, made difficulties for
the Catholics, but the Prime Minister (whose daughter-in-law,
Victoire Rasoamanarivo, was a militant Catholic) knew how to
temper their ardour. All these new developments in legislation
and administration, religion and education, had their greatest
523

effect in Imerina, but also spread into Betsileo and to Tamatave.

Elsewhere, and even in the remoter parts of Imerina itself, the
traditional customs (fombandraâna) were religiously preserved, in
deference to the widespread distrust of any change.
Rainilaiarivony, who was quite well informed on European
matters, was above all afraid of France, which had never formally
renounced its 'historic rights' in Madagascar, and which was
being pressed by two groups, the Catholics and the people of
Reunion, to extend its influence over the island. The missionaries
appear to have given him the impression that he might hope for
British protection, but the British government, now launched on
its Egyptian adventure, was anxious not to give France any new
cause for enmity. The various European colonial powers had
begun the process of dividing up Asia and Africa between them,
and in this game of chess, Madagascar was merely a minor pawn.
Jean Laborde,30 who had died in 1878, had left estates that were
claimed by his heirs, whereas the Malagasy government considered
that these had only been granted to him for life. Moreover, in 1840
the French had occupied Nosy-Be; this was after the Sakalava
rulers who had fled to the island had ceded it to France, together
with their rights over those parts of Madagascar itself from which
they had been driven by the Merina occupation. In October 1881
the French consul laid claim to these rights, and to those of
Laborde's heirs. Rainilaiarivony replied that the whole island
belonged to the queen, and that he could not give up any part
of it, 'even enough to plant a grain of rice'. He set up Merina
flags on the disputed stretch of coast, which were torn down in
June 1882 by the French sea captain, Le Timbre. Rainilaiarivony
then sent a diplomatic mission to Europe under the leadership of
his nephew. He obtained from Britain the renunciation of all
freeholds held by British subjects; these received long leases in
exchange. With France discussions dragged on, and eventually
broke down, since the envoys dared not take any decision without
the prior authorisation of the Prime Minister (November
1882-August 1883). Meanwhile, a reasonably favourable French
government was replaced by another, in the event a short-lived
one, in which for a time care of the navy was entrusted to the
deputy for Reunion, Francois de Mahy. He sent a small naval force
to Madagascar, under the command of Admiral Pierre, which
30
See Cambridge history of Africa, v, 409-12.
5*4

MADAGASCAR: R AINI L AIA R I VON Y, 1870-85
occupied Majunga on 16 May, and then bombarded Tamatave and

occupied it on i1 June. The Jules Ferry ministry, which was now
in power, demanded the cession of the northern part of the island
as far as the sixteenth parallel and the right for French citizens
to hold property in freehold. These demands were rejected by the
Prime Minister in a major public speech, which was supported by
a general outcry.
Ranavalona II died on 13 July 1883, and Rainilaiarivony, then
5 5 years old, chose as Queen and wife a 22-year-old princess, who
became Ranavalona III. The war was dragging on. The French
had occupied Diego Suarez, but were unable to advance into the
interior for want of sufficient troops. An Englishman, Colonel
Willoughby, who had been engaged as a military adviser by the
Prime Minister, had helped Rainandriamampandry, the governor
of Tamatave, to organise a strong defensive position on the
Farafaty lines. But the French blockade stopped the delivery of
arms and halted the collection of customs, which were the main
source of revenue. In France the Ferry administration fell on 10
March 1885 over the Tonkin affair, and there was a general desire
to have done with other colonial expeditions still in progress.
Negotiations were opened between the two weary adversaries.
Accordingly, on 17 December 1885 a treaty was signed by
Admiral Miot and the diplomat, Consul Patrimonio, on behalf of
France, and Colonel Willoughby and one of Rainilaiarivony's sons
for the government of Madagascar. Its terms were as follows. The
French government would represent Madagascar in all its external
relations (' toutes ses relations exterieures') (the Malagasy text said
'a Pexterieur' (abroad)). A French 'resident' would 'preside over
Madagascar's external relations, taking no part in the internal
administration of the states of 'Her Majesty, the Queen'. He
would reside at Tananarive and would be supported by a military
escort. An indemnity of 10 million francs would be paid to
France. The queen would 'continue to preside over the internal
administration of the entire island'. France reserved the right to
occupy Diego Suarez Bay.
In a clarificatory letter requested by Rainilaiarivony, Miot and
Patrimonio specified that the resident's escort would not exceed
50 men, that French settlers would accept 99 year leases, and that
the occupation of Diego Suarez Bay would not extend more than
a mile and a half south of the Bay.
5*5

THE PHANTOM PROTECTORATE, 1886—94
Rainilaiarivony had avoided the term 'protectorate', but the

French had the firm intention of imposing one in practice. The
first French resident, Le Myre de Vilers, encountered stubborn
resistance from the prime minister, and was able to gain conces-
sions only on minor points. The Miot—Patrimonio letter, which
had been written after the treaty had been signed, was not
recognised by the French government, and the troops advanced
as far as 20 kilometres from Diego Suarez Bay. Rainilaiarivony
insisted that the recognition of the credentials {exequatur) of
foreign consuls be the sole prerogative of the queen, whereas the
Resident's intention was that these should first be presented to
him. This matter was never settled, even after Lord Salisbury's
British administration had recognised ' the Protectorate of France
over the Island of Madagascar, with all that this entails, in
particular the consuls' exequaturs...' (5 August 1890), in exchange
for the French recognition of the British protectorate over
Zanzibar. The Comptoir d'Escompte established a branch in
Madagascar. A telegraph line was built from Tananarive to
Tamatave. French residents were installed in Tamatave, Majunga,
Fianarantsoa and later at Tulear.
Thus in 1894, although the status of Madagascar was in dispute
at the international level, internally it remained more or less
unchanged. The regime benefited Imerina, whose rice fields were
extended with the labour of a large number of slaves. Tananarive
had 75,000 inhabitants. Its roads were still only stony paths, but
the wooden houses at the top of the hill were gradually being
replaced by two-storeyed constructions of brick and stone. The
ruling oligarchy (composed of important members of the hova and
andriana castes) adopted European fashions, and grew rich on
trade, on political influence and the labour of slaves. The rulers
were entirely christianised: of the 164,000 mission pupils in the
island, two-thirds were in Imerina. The queen was a symbol and
a figurehead. Everything was done in her name, but she had no
real power. The Prime Minister, who was also the commander-
in-chief, ran everything. He was older now, and had to face plots
against his government. He was informed of everything, but
could not prevent the growth of disorder in the countryside.
The machinery of government was rudimentary indeed. State
526

MADAGASCAR: PHANTOM PROTECTORATE, 1886-94
revenues were derisory, and depended on a few archaic taxes, and
the income from customs duties were mortgaged by the need to
pay the 10 million francs indemnity due to France. Officials went
unpaid. The corvee, a form of taxation that was very well suited
to a country without a regular currency (Spanish or French
piastres were used, cut up into small pieces), made it possible to
undertake certain public works such as public buildings and
irrigation dykes, but there were neither roads nor bridges.
Porters' shoulders were the only means of transport. The economy
was based on self-sufficiency: rice, cassava, sweet potato, poultry,
cattle, house-building, the weaving of cloth, the plaiting of mats,
iron work and small industries introduced by the Europeans such
as brick-making and tin-smithing. More and more European and
American cloth was imported. Exports consisted of live cattle, a
little wild rubber, wax and raphia. In the absence of properly
equipped ports, goods were offloaded on the beaches.
Some of the provinces, notably Betsileo and Betsimisaraka,
were administered directly by Merina governors and military
commanders, while others, such as Sakalava and the south-eastern
territories, were under a sort of protectorate, with the local kings
retaining their functions under the surveillance of a Merina
garrison. This colonial system worked reasonably well, but did
not succeed in creating any overall feeling of nationhood.
However ardent and sincere might be the patriotism (tinged with
xenophobia) of Imerina, the subject peoples did not share in it,
and were at times indeed hostile to it. Rainilaiarivony did not
think it worthwhile that these peoples should participate in the
government; on the other hand, he encouraged Merina colonists
to settle in certain provinces, especially Betsiboka, the Sihanaka
country, and Tamatave. A third of the island was still independent
of Imerina: in the west, Ambongo and part of the old Sakalava
state of Menabe; in the south, parts of Bara and Tanala,
Antandroy and Mahafaly. In the southern part of the island,
Prince Ramahatra, a cousin of the queen, seized Tulear in 1890
at the request of French merchants who had been held to ransom
by the local chiefs.
The army was poorly equipped for its duties. It possessed 300
field-guns, some of which were very old, and 20,000 modern rifles,
which were often badly maintained in spite of having cost a lost
of money. The soldiers received neither pay nor rations, and lived
5*7

iegoSuarez
&
ts°s
amatave
20°S 20"S
2S°S Boundaries of the Merina kingdom

after the conquests of Radama I
K % 3 Independent kingdoms and
ethnic regions
I 1 Land over 3000 feet
100 km
3 miles
21 Madagascar
528

MADAGASCAR: FRENCH CONQUEST
at home, assembling only periodically for training exercises. The
Prime Minister recruited a few British officers, who trained some
good artillerymen. The soldiers were mainly Merina; others were
not trusted. This army was not adequate for the maintenance of
order; banditry was on the increase and there were slave raids by
independent peoples in certain areas of the plateau. Rainilaiarivony
was not sufficiently aware of this military weakness, which became
all too obvious in the conflict with France. He still relied more
or less on those two great defenders of the island traditionally
called 'General Hazo' and 'General Tazo' - the forest and the
fever.
THE FRENCH CONQUEST AND THE INSURRECTION,

I 894—96
By 1894, Europe had entered the final phase of the partition of
Africa. In France the opponents of colonial expansion had been
reduced to a small minority, and it was generally recognised that
the ambiguous status of Madagascar could no longer be tolerated.
Le Myre de Vilers was sent to Tananarive, and on 17 October
1894 he handed to the Prime Minister the draft of an agreement,
by which Foreign Affairs would be handled directly by the
Resident, while the French would have the right to maintain on
the island as many troops as might be necessary for security, and
to undertake public works. Rainilaiarivony submitted a counter-
proposal, which granted nothing. On 27 October Le Myre de
Vilers lowered the French standard at the Residency and left the
capital, giving orders that the French nationals should follow him.
This was war. The French Assembly voted supplies, and on 12
December the French forces occupied Tamatave, followed on
15 January 1895 by Majunga. It was from here that the French
expedition to Tananarive set out.
This ill-planned expedition, undertaken in the middle of the
rainy season and involving the construction of a road through the
swamps, opened with the victory of General Tazo: out of 15,000
men, 6,000 died of fever. The Merina army, on the other hand,
inflicted only minor losses. The successive positions it took up
were each evacuated when it felt the weight of modern firearms.
The coming of the dry season allowed the French to advance more
rapidly, and General Ra'nianjalahy made his last efforts at

resistance at Tsarasoatra on 29 June and at Andriba on 22 August.

On 14 September General Duchesne formed a light infantry
column of 4,000 men, equipped with mules for the transport of
provisions and heavy weapons. Rainilaiarivony appeared at a loss.
He ordered an attack on the rear of the column, but it failed. The
Malagasy gunners, who were the corps d'elite of the Malagasy army,
were slaughtered at their guns. On 30 September 1895 the heights
flanking Tananarive to the east were occupied. An artillery battery
fired two high explosive shells on the queen's palace, decimating
the huge crowd that had gathered there. The queen raised a white
flag, and that same evening French troops entered the town.
The treaty brought by Duchesne was signed the following day,
1 October. The queen accepted the French protectorate with all
its consequences. Henceforward, the resident-general would be
responsible for the internal affairs of the kingdom. Rainilaiarivony
was exiled to Algiers, where he lived in a fine house until his
death on 17 July 1896. But the Merina oligarchy survived almost
unchanged. The protectorate formula had produced good results
in Tunisia, and seemed to offer the most economical method of
colonisation.
Only government personnel and the more or less Europeanised,
christianised population of the towns had so far been taken into
consideration by the French. No thought had been given to the
country people, who were still faithful to their ancestors and
hostile to foreigners, even to the defeated government. As early
as 22 November 1895, a band of men 2,000 strong took Arivoni-
mamo, not far from the capital, and assassinated the Merina
governor and also a British missionary and his family. The rebels
demanded the abolition of Christian worship, schools, military
service and labour dues. The subject peoples, too, were beginning
to be troublesome. The peoples of the south-east had their own
revolution, driving out their nobles, who had become the agents
of the Merina. In Betsimisaraka territory there were revolts
against the Merina, who were massacred. The French army
suppressed these insurrections ruthlessly and without distinction.
In December 189; it was decided in Paris that Madagascar
should be transferred from the Ministry of Foreign Affairs to that
of the Colonies, and in January 1896 the queen was obliged to
recognise France's taking of possession. In fact this changed
nothing. The resident-general, M. Laroche, was a former French
5 5°

MADAGASCAR: GALLIENl's FIRST TOUR OF DUTY
prefect, of the Protestant faith, who treated the queen respectfully

and sympathetically, but was completely untrained for colonial
command, and had no authority over the army. In Imerina
nationalistic and xenophobic feelings were gaining ground, iden-
tifying with thefahavalo (bandits). This produced the revolt of the
menalamaba (red togas), who wore on their foreheads the felana
(round shell) of the ancient warriors. The principal leaders were
on the one hand zfabavalo called Rainibetsimisaraka, and on the
other hand two governors, named Rabezavana and Rabozaka.
They operated mainly in the thinly populated regions on the
northern and southern borders of Imerina. The French responded
with flying columns, which drove back the insurgents and burned
the villages. When they left, the insurgents returned, their
numbers swollen by the unfortunate villagers. The rebellion came
to within 15 km of Tananarive. It was said that some of the
ministers were encouraging the insurrection, in spite of the
queen's peace-loving proclamations. Authority was disintegrating
everywhere, except for a few centres held by the French, and chaos
was rapidly spreading to the whole island.
GALLIENl's FIRST TOUR OF DUTY, 1896-99
It appeared that new policies were required in order to bring the

situation quickly under control. The minister for the Colonies,
Andre Lebon, called in General Gallieni who, first in the Sudan
and then in Tonkin, had acquired a well-earned reputation for
his skill in colonial pacification. Although on his return from
Indo-China he was very ill, he accepted this virtually hopeless task.
A law of 6 August 1896 proclaimed Madagascar a French colony;
the Chamber of Deputies, moreover, decided unanimously that
this entailed the abolition of slavery. Ministerial directives,
probably suggested largely by Gallieni in the light of his previous
experience, ordained the 'abolition of the hova hegemony':31 'the
sovereign power must now make its presence felt directly,
through the intermediary of the chiefs of each particular people'.
They prescribed ' a policy of firmness... in respect of certain
persons at the Court of Emyrne' (Imerina), 'moderation in
respect of the native lower classes',' prudence... in the suppression
31
In France, the word bova, though only applicable to certain castes, generally
designated the Merina.
551

of acts of rebellion', and 'equality of treatment for all religions,

except for the banishment of those who had concealed political
designs beneath the cloak of some kind of religious belief. On
28 September, Joseph Simon Gallieni, having reached Tananarive,
was granted total civil and military powers. He was afterwards
given the title of governor-general.
The Merina oligarchy was abruptly reduced to submission.
On 15 October Prince Ratsimamanga and the minister Rainan-
driamampandry were shot, after being condemned to death as
accomplices of the rebels. Ranavalona was henceforward only
'Queen of Imerina'; then on 27 February 1897 she was exiled,
first to Reunion and then to Algiers, where she died in 1917. The
minister Rasanjy, who had supported the cause of France, was
appointed governor of Imerina. On 15 March the royal remains,
which might have become a central focus for sedition if left at
Ambohimanga, were transferred to Tananarive. The festival of
the Royal Bath was replaced by the French holiday of 14 July.
On 17 April the feudal system was abolished. These measures
reflected in no uncertain way both Gallieni's republicanism and
his sense of immediate necessities.
The methods of pacification developed by Gallieni in Tonkin,
and later employed in Madagascar, involved, according to his own
directives, three main points. First, I'action politique, getting to
know the country and rallying the people to the French. Recourse
to I'action vive (military intervention), was to ensue only if political
action proved inadequate. Secondly, la tache d'huile (oil stain), the
spreading of the French presence by diffusion. There were to be
no more flying columns: 'any forward movement of troops must
be accompanied by the effective occupation of the conquered
area'. Under cover of military posts, the territory was to be
brought under control by reassuring the population, protecting
them from the fahavalo and helping them to re-establish their
villages and their crops. To those in charge of the posts Gallieni
gave the order: 'Act in such a way that those you administer
tremble only at the thought of your departure.' Thirdly, unity of
command: a single commander at each level, responsible for
everything to his immediate superior. This principle, giving his
officers a sense of responsibility and putting them on their mettle,
proved very fruitful.
By the end of 1896 links between Tananarive and the ports had
53*

MADAGASCAR: GALLIENI S FIRST TOUR OF DUTY
been established, or if they had already existed, made secure. The

greatest number of troops were concentrated in Imerina. The
various rebel bands were first isolated and then driven back to
arid regions. In June 1897 operations were brought to a close.
Since the French presence was now being acknowledged, no
action was taken against the rebel chiefs. In the provinces
previously ruled by the Merina, the new racial policy was
implemented by replacing the Merina governors with French
officers or administrators directing or supervising chiefs taken
from the local ethnic groups. This measure was welcomed, but
the disorders of the previous years were at times difficult to
resolve. In the arid region to the north of Imerina, the leader of
the Rainitavy bands surrendered only in November 1897. At that
date the entire former kingdom had been occupied and brought
under control. An insurrection in October 1898, caused by the
misbehaviour of some settlers amongst the northern Sakalava,
was suppressed in three months.
The lands that had been independent of Imerina were more
difficult to subdue. There, the French were dealing with warrior
peoples, practised in the art of ambush in uncharted and semi-desert
terrain, broken by swamps and forests, in which distances were
enormous and supply difficult, and where there existed no single
chief capable of making his authority felt over any substantial area.
The troops used were mainly Senegalese and Malagasy, with
French officers and NCOs, but there were also elements from the
Marines and the Foreign Legion. The local Ambongo kinglets
were picked off one by one. Contrary to Gallieni's intentions, the
invasion of the Sakalava of Menabe resulted in the massacre of
King Toera and his warriors, thus prolonging hostilities. The
king of Fiherenana surrendered. The king of the Imamono Bara,
who had given assistance to the French, had his kingdom raised
to the status of a protectorate. Inapaka, king of the Be Bara,
repulsed an attack on his stronghold of Vahingezo, which was not
taken till 1899. Another famous stronghold, the mountain of the
Ikongo, where in former times the Tanala had defeated several
Merina armies, was taken in 1897; but in 1899 a rebellion broke
out among the Tanala, who held out in the forest for two years.
In the same way, the forested cliffs between the Bara and Antesaka
were used as a refuge by numerous rebels. Apart from these two
very restricted areas, the only part still to be subjected was
533

Mahafaly and Antandroy in the extreme south, a land of waterless

thorny bush.
Gallieni had been tireless in giving detailed orders for pacifi-
cation and setting up military sectors, which were constantly
modified according to need. He had personally travelled twice
completely round the island by boat to examine the situation and
to determine the best points at which to disembark the troops.
He was also responsible for general policy and the administration
of the country. The provincial and district boundaries, which had
been largely taken over from the Merina, were frequently altered
as the need arose. The majority of the Malagasy people had
accepted the authority of France in succession to that of the
Queen, and Gallieni was everywhere received with unfeigned
popular demonstrations. The freed slaves had mostly left Imerina
either for their own homes or to settle in Betsiboka, and for this
reason about half the population of Tananarive had left the town.
The high-caste Merina who had lost power, slaves, fiefs and the
virtual monopoly of trade in one fell swoop, must doubtless have
felt some regret, but they took care not to show it. A number of
them, who had sufficient education, found positions in the new
administration, and their skills also found an outlet in the
economic field, either in commerce or on the plantations on the
east coast. Gallieni now favoured the Merina, and relied on them
for the general development of the island. One of his circulars of
1898 even said, with some exaggeration: 'The Hova race appears
to be the only one capable of providing an adequate population
and labour force for the future.'
Gallieni paid great attention to this question of manpower; he
had indeed come to the country with an axiom, doubtless
suggested to him in France: 'Colonies are made for French
colonists.' The establishment of a head tax and of compulsory
labour contributed in some measure to this aim. The tax forced
the Malagasy people to obtain money, either by producing on
their land a little more than they consumed, or by engaging
themselves as wage-earners in the service of the Europeans. The
obligation of compulsory labour, fixed first at 50 days a year, and
later at 30, were a tax in kind. This was the old corvee under another
name, and made it possible to carry out some public works
without making inroads into the meagre budget; the settlers could
redeem their employees' obligation of compulsory labour with
5 34

MADAGASCAR: GALLIENI S SECOND TOUR OF DUTY
money. Between 1896 and 1899, 200,000 hectares of land were

granted in concessions; thus Gallieni hoped to attract capital
investment and launch the process of economic development.
The value of imports doubled in two years, even though
Madagascar, as a colony, had entered a customs union with
France, so that French goods had replaced foreign ones, especially
British and American textiles. Large French import—export com-
panies opened branches: the Compagnie Lyonnaise in 1897, and
the Compagnie Marseillaise a year later. The introduction of
French currency, the establishment of peace, and the development
of roads and markets, all encouraged internal trade and local food
production; but production for export was still at a very low level.
One of the subjects of concern to Gallieni was that of the
Christian missions. As they increased in size, they had to be dealt
with tactfully and even encouraged; but he frequently had to deal
with the rivalries between them: in particular he reminded the
British Protestants that Madagascar was French, and that the
teaching of French should take up at least half the timetable.
Personally he had no difficulty in remaining neutral between the
various denominations; he was a disciple at once of Voltaire and
of Spencer.
G A L L I E N I ' S SECOND TOUR OF DUTY, 1900—05

As a disciple of Voltaire and Spencer, Gallieni was an evolutionist.
' The natural development of the country must dictate its organ-
isation,' he wrote (L'organisation... doit suivre le pays dans son
developpement naturel). He was a broad-brush technician, one
who thought before acting, ever ready to learn from experience,
constantly on the watch against rigidity, and waging war on
routine and bureaucracy. ' I am for ever pushing onward.' He was
always prepared to alter his own ideas if he found that they were
inadequate or out of date, and his second tour of duty was marked
by a number of changes. In 1900 the completion of the pacification
programme was entrusted to a ' Commandement superieur du
Sud' (Southern High Command), with its headquarters at Fian-
arantsoa, under the command of Colonel Lyautey, who had been
Gallieni's assistant in Tonkin and was to be the architect of French
Morocco. In 1900, Ambovombe, the Antandroy centre, was
occupied, and peripheral military posts were set up. The Mahafaly
535

kings surrendered one after the other; in the forested cliffs of the
south-east, the two rebel areas were eliminated. In 1902, with
the pacification programme apparently completed, the Southern
High Command was abolished.
In fact there were still a few local difficulties in Antandroy, but
a major problem arose in 1904 with the revolt in the south-east
amongst the Antemanombondro, a small tribe to the south of the
Antesaka, who killed some Europeans. The movement spread to
northern Antanosy and the 'Falaises' region. These areas were
covered with thick bush and forest, which had never been
thoroughly pacified, and in which a prohibition on bushfireswas
destroying the agricultural traditions of the inhabitants just at the
time when taxation was increasing. Gallieni sent troops with
orders to 'pacify, not exterminate'. The instigator of the revolt,
Corporal Kotavy, was captured in September 1905. That was the
last spasm; the unification of the island, begun by Andrianampo-
inimerina, was now complete.
Hitherto, estimates of the total population had been conjectural.
The first systematic census in 1900 gave a total of about 2.5
million, which was not large for an island larger than France.
Gallieni established an indigenous medical service, with hospitals
and dispensaries in the major centres, and a School of Medicine
to train Malagasy doctors. In 1898 and 1902, the coastal region
was ravaged by the plague, which had arrived from India.
Pacification, road building, and the development of industry,
plantations and mines all helped to facilitate internal migration.
In 1903, 8,000 Antesaka emigrated temporarily to Diego Suarez,
Tamatave and western Madagascar as manual labourers; the
former Merina slaves continued to settle along the Betsiboka,
and Merina colonists spread out westwards and southwards from
Imerina. Bara herdsmen spread into the empty spaces of the west,
close to the plateau, as did Betsileo farmers, who also settled in
Betsiboka. In this way the large demographic gaps began to be
filled in little by little. The non-Malagasy population also increased,
as a result of the arrival of officials, settlers and merchants: in 1905
there were 7,800 French (more than half of them from Reunion),
1,000 Mauritians, 1,000 Indians (mainly from Gujarat) and 450
Chinese.
Education, like the medical service, was one of Gallieni's
constant preoccupations. Without interfering with the rights of
536

MADAGASCAR: GALLIENl'S SECOND TOUR OF DUTY
the mission schools, he began from 1903 onwards to establish

secular education, in accordance with the policy of the French
government of the day and with his own convictions. When he
left there were 343 government schools and 178 mission schools.
There were also three high schools set up to train the Malagasy
in administration and commercial skills, one in Tananarive and
one on each of the two coasts.
Another new measure was the abolition of compulsory labour
in 1901; its place was taken by a new tax, the proceeds of which
were allocated to Public Works. Gallieni extolled the economic
and social role of taxation, and its educational value, in that it
forced people to work, and thus to develop the country. It quickly
became apparent that the tax, though it varied from region to
region, bore too heavily in some areas that lacked resources. But,
although she readily lent capital to foreign countries, France at
that time insisted that her colonies, though still in their infancy,
should find the funds for their own budgets.
During his leave in France, Gallieni had nevertheless obtained
a loan of 60 million francs for the construction of a state-managed
railway line from Tananarive to Tamatave. The first two sections
were completed before he left in 1905. Transport on this essential
route had already been made easier by river boats and by the
construction of a wagon road; in 1898 the route employed 63,000
porters, and in 1904, 8,000 ox-carts. Gallieni brought the first
motor-car to Tananarive; other wagon roads linked Tananarive
to the major centres. These public works, like the construction
of official buildings, were undertaken first by the army engineers,
and then by a public works' service.
A law of 1896 had, in principle, guaranteed traditional Malagasy
property rights. After due declaration and survey, these rights,
together with concessions granted to both Europeans and non-
Europeans, could be entered in land registers. Gallieni had
modelled this system on the Australian Torrens Act, which was
simpler and more suitable for new countries than the archaic
complications of the French system. The enormous concessions
granted in the early days were curtailed. By the time of his
departure, Gallieni had come to the conclusion that Madagascar
was not a colony for settlers. He relied increasingly on the
Malagasy peasantry, establishing services for agriculture, stock-
rearing and forests, and taking steps to revitalise the fokon'ola
537

(village assemblies). At this time 80 per cent of external trade was

with France. Discoveries of gold veins in the region of Diego
Suarez had made gold the most important export product; the
other exports — rubber, skins, raphia, cattle and wax — still came
essentially from the exploitation of the natural resources of the
country.
Gallieni finally left for France in May 1905. Over the previous
nine years, his views had changed considerably. The future now
seemed to depend mainly on the economic, social and cultural
evolution of the Malagasy people. In 1902 he had founded the
Academie Malgache, and recognised the Merina dialect as the
second official language. His authority, his great intelligence and
his appreciation of the role of change had laid the foundations for
modernisation. Raseta, the nationalist Malagasy member of the
legislature, called him 'a positive, constructive person', and
Roberts, the Australian historian, not normally favourably dis-
posed towards French colonialism, has said:' Gallieni, perhaps the
greatest figure in French colonial policy, really made Madagascar.'
538

CHAPTER 10
EAST AFRICA 1870-1905
East Africa in 1870 is best defined as the economic hinterland of

the commercial entrepot of Zanzibar. This area included much of
what is now Tanzania, Kenya and Uganda, as well as parts of
Rwanda, Burundi, Zaire, Zambia, Malawi, Mozambique and
Somalia. The region was criss-crossed by trade routes which
converged on the Swahili towns of the east coast. It was along
these paths of Swahili expansion that Europeans in the later
nineteenth century began to penetrate and occupy the interior. By
1905 Britain and Germany had divided most of East Africa
between them, and they had achieved overall command in military
and strategic terms. But their new colonial governments exercised
a very uneven control over the African population, while the new
colonial economies had only partly deflected local labour and
capital from older systems of production and exchange. New
economic structures were emerging, but their impact was only just
becoming evident. Thus the period under study here is very much
one of transition. To understand it, we need to consider not
only the European innovation but also the local and regional
economies, and the wide-ranging Swahili commercial and cultural
network. We must also acknowledge the importance of individ-
uals. The changes of the period gave much scope for the exercise
of leadership and the pursuit of political rivalry. Initially, African
horizons of statecraft were much enlarged, though often one early
effect of colonial rule was to narrow them. What follows, then,
is a synthesis stressing processes of differential integration in the
immediate pre-colonial period, the dynamics of the decade of
military conquest, 1888-99, and the terms of reconstruction
preparatory to systematic state formation which occurred from
the turn of the century until 1905.
539

EAST AFRICA, 1870-1905
THE COMMERCIAL SPHERE AND REGIONAL PATTERNS

Through the activities of the Swahili commercial system, East
Africa during the early and mid-1870s became a more internally
integrated and geographically specific region of the world than
it had been before. The transport network became denser and
more pervasive, its personnel conveyed cultural as well as material
values, and credit relations hastened the dissolution of old and the
construction of new social relations. But commerce did not act
in isolation. Deployment of land and naval forces by Egypt and
Britain hardened frontiers on the upper Nile and at the coast. In
response, rulers in the interlacustrine region regularised their
armies and the Sultan of Zanzibar also took steps to strengthen
his state apparatus. Such centralisation met with resistance from
coastal community leaders. There is no doubt that the presence
of the well-armed, if not always well-stocked, long-distance
traders entered strongly into the calculations of leaders in all
manner of polities.
Behind much of the expansionism of regional powers lay a
determination to control valuable resources. Jurisdiction over
land and people was jealously sought and defended. Tribute in
labour and produce helped chiefs to concentrate surplus in their
own hands. The numbers of women at their disposal, living at
their headquarters, represented a rough index of many chiefs'
assets. Variant indices of wealth in the northern and inter-lacustrine
regions were numbers of cattle and/or clients, and pages or
apprentice functionaries of state, as in Buganda. Wealth in
followers, clients, or women was potential, while wealth in ivory
was more easily converted. Chiefs and ambitious upstarts alike
sought to acquire ivory, the former by vigilance and insistence
upon the ' ground' tusk of any elephant felled in their domains
and by extension of police power to make certain that trading or
hunting parties entering their country did not deal with under-
lings. Principles of superiority and territoriality thus became
strengthened. The exacting of tolls for passage and of tribute from
subordinated communities was not necessarily accompanied by
elaboration of governmental apparatus. The most evident area of
expansion was military. Such generalisations take on greater
significance in the light of conditions within specific regions,
which contained varied political cultures, social formations and
economic resources.
540

COMMERCIAL SPHERE AND REGIONAL PATTERNS
The region to the west of Lake Victoria commonly known as
the interlacustrine region, where political life was shaped by
kingdoms, underwent an enlargement in military and economic
scale beginning in the early 1870s. As Egypt reached for the
headwaters of the Nile, its garrisons and diplomatic initiatives
became part of the political scene, above all in the kingdoms of
Buganda and Bunyoro. This advance of political and adminis-
trative ambition led to the formation of the Equatoria Province,
which was the first formal manifestation of imperialism in East
Africa. The activities of the Sudanese slave- and ivory-traders, the
'Khartoumers', had already created a frontier of resistance,
however, and among the politically decentralised people of
Lango, defensive measures had fostered a new stratum of regional
military leadership. Egyptian garrisons and officials, therefore,
confronted an increasingly well mobilised and hostile society in
the area north of Lake Kyoga. For their part, the Langi increased
their interactions with Bunyoro, from which traders came with
iron wares and to which men could go to participate in the
struggles between rival claimants to the throne and to obtain
booty.
Bunyoro was the kingdom on the first line of contacts with the
Egyptian agents. This state was based not so much in its political
apparatus as in its economy, which provides a classic example of
iron-age regional exchange. The varied resources included a solid
trinity of iron, salt and cattle. With the increase of external trade
with the Swahili, the abundant sources of iron had brought
wealth to the rulers.* Nyoro boatmen dominated the key river-
ways and waters of Lake Kyoga and Lake Albert (Mobutu), and
were in a position to extend and intensify lines of trade and
communication.
As Kabarega (ruled 1869-99), the newly installed omukama of
Bunyoro, began to take full command of the state after a
protracted succession struggle, he built a new military organisa-
tion, the abarusura, into which many non-Nyoro were recruited,
to be paid in booty. In conjunction, this circulation of men
intensified the interchange of cattle, food and trade goods. While
there is no doubt that the abarusura escalated the means of violence
at the disposal of the state, they were rarely used in outright
conquest but rather for the reconquest of Toro, a breakaway state
to the south-west of Bunyoro, for forays to the southern salt lakes
which were strategic crossroads of trade, and for internal and

EAST A F R I C A , 1870—1905
40
1
\ ) % 26
r 14
\ Brava
^ ~
(/ 3 (] I
..•••2 \i\_5 1
1
S /
of 15
1 Kismayuf' — —
Victoria (
c \
V: 12
I; 10
/
^Lamu
: ) '
r
1
' 24
• 7/
•. I • «
J
^Malindi
•. IT Y(*Mwanza
19/
• 2 8
M i Kilimanjaro* ••.. /Mombasa
{l v 1
V .*
/
•••••
Tang< 'J D
•; Tabora
7 ) I Pemba
Pangani J U
S. V X Tanganyika
[ViZanzibar
3 Bagamoyo
Xpares Salaam
8
) (3
17
/j Uilwa
( /
\
_..••• 20 . i. 10°S
i K3ronga\
ii I 27
i Nytsa
(
S- ) 29
>—
30 \ \ 40
Primary rift valley Major trade routes o | t 390km

Groups, kingdoms and areas 0 2 0 0 miles
1 Bemba 11 Katanga 21 Nkore
2 Buganda 12 Kikuyu 22 Nyamwezi
3 Bunyoro 13 Kimbu 23 Rufiji
4 Burundi 14 Langi 24 Rwanda
5 Busoga 15 Luyia 25 Shambaa
6 Fipa 16 Masai 26 Somali
7 Haya 17 Mahenge 27 Songea
8 Hehe 18 Manyema 28 Sukuma
9 Jie 19 Mijikenda 29 Yao
10 Kamba 20 Ndonde
22 East Africa, late nineteenth century
542

border control. The looseness of administrative control in Bun-

yoro as contrasted with Buganda was permitted by the vitality of
its economy, which was tapped rather than dictated by the state.
Buganda in the early 1870s was ascendant. Kabaka Mutesa
(ruled 1854-84) had firm control after fifteen years of rule, and
a machine of government capable of territorial expansion and
tributory exaction. To the old strata of hereditary and appointed
chiefs, Mutesa added a new dimension. When he mounted
expeditions into new areas, the kabaka assigned commanders,
batongole chiefs who served at the pleasure of the ruler, but whose
functions were more military and colonising than those of the
bakungu, the appointed chiefs in core areas. The motives for Ganda
military activities varied: Busoga remained a preserve for slave-
raiding and ivory acquisition, while pressure on Nkore secured
the southern approaches by which traders entered the sphere of
influence of both Buganda and Bunyoro. To further control the
terms of access by long-distance traders, the small states of
Karagwe and Kiziba to the west of Lake Victoria were kept in
the shadow of Ganda power. Buganda under Mutesa came to
define and dominate a new regional economy built through the
power to commandeer and monopolise. This strategic military
and economic expansion was well exemplified on Lake Victoria,
where a navy of large canoes secured a monopoly of violence
along the littorals and dominated the transport of goods and
commodities. Usukuma at the south end of the lake became a
source area for iron wares, produced by the Ronga and for food.
The Ganda actively participated in the development of this lake
route for long-distance trade, to the neglect of the overland
approach to the west of the lake and the detriment of Bunyoro.
Between 1874 and 1876, Egyptian forces had attempted to
garrison posts south of Lake Kyoga. After overestimating the
success of his diplomacy and occupation, Charles Gordon had
ruefully to confess that rather than annexing Buganda, his soldiers
had been effectively detained by the kabaka. The policy of
welcoming visitors but keeping them under control was consis-
tently followed by Mutesa, who was very receptive to ideas and
technologies that could enhance his power. By the mid-1870s, to
offset the aggressiveness of the Equatoria administration, the
kabaka called upon westerners to come and reside at his capital.
While seeking to strengthen his ties to Zanzibar, he saw Christians
543

as a means of countervailing Muslim power in Equatoria and
among Ganda converts.
The elite of Buganda depended upon the kabaka's patronage
for political advancement. They were also more economically
dependent upon the state than in neighbouring kingdoms and
polities where clan and community retained control over distri-
bution of land. Advancement for youths entailed apprenticeships
at district and provincial courts, from which an aspirant could
move to the kabaka's capital and there serve the leading chiefs or
the kabaka more directly. The more concentrated power became
and the more expansive was the state, the more feverish was
competition among the pages for a place in the meritocracy within
an autocracy. At the same time, as the monarch outgrew and
became abstracted from the localised and traditional religious
systems, Islam and Christianity became available as potential
religions of the state, to be domesticated just as long-distance trade
had been. Mutesa observed Ramadan and encouraged mosque
building in the early 1870s. Literacy was a functional by-product
of the new religion, and pages imbibed not only the theological
precepts but also the legalisms of Islam. The year after initiating
relations with Christians and inviting them to his court, Mutesa
launched a purge of Muslims and authorised the public burning
of pages whose loyalty to the new religion had compromised their
loyalty to the kabaka. By contrast, Bunyoro did not have the same
concentration of upwardly mobile youths; its functionaries were
not too bureaucratised, and the mukama himself did not lead in the
prayers of any foreign religion. Kabarega nevertheless spoke
Arabic, wore a kan%ut and otherwise showed in his costume and
conduct an awareness of the material importance of the northern
and Swahili factors in Bunyoro.
Rwanda was a major expansive state of the interlacustrine
region that was not extensively penetrated by long-distance
trading interests or subject to the manoeuvres of external powers
in the immediate pre-colonial period. Motivated by population
pressures, the Rwandan monarch, Kigeri (ruled c. 1865-95),
embarked upon conquests of areas to the north. The principle of
caste superiority of Tutsi pastoralists over Hutu agriculturalists
was thus extended into new areas. But even in the older parts of
the state the extent and thoroughness of Hutu subordination
remained variable. From the perspective of some Hutu in Rwanda
544

and many Iru cultivators in Nkore, their states were not rigidly
stratified. Tribute often went to the kings directly from cultivating
communities through their own leaders, not through Tutsi or
Hima lords. Court ideology claimed for the Tutsi an intrinsic,
primordial superiority, but the actual process of instituting
unequal relations took place slowly. As pressures of human and
animal populations on the uplands increased, so the competition
for land mounted and with it the potential for deterioration of
the terms of clientship. The Hutu cultivators had to pay more
dearly for the lease of cattle and access to land. Certain Tutsi also
became poorer. Some lost status and joined the Hutu. Others
clustered around wealthy Tutsi as clients. A third option was
emigration, and a number of Tutsi moved to Unyanyembe to
serve as herdsmen for the Nyamwezi.
In sum, the kingdoms of the interlacustrine region, with their
varied social formations and places within the regional economy,
adopted different postures towards long-distance traders. Buganda
and Bunyoro concentrated on ivory and, secondarily, slaves for
exchange in the name of the state. In Rwanda, however, the
penetration and role of foreign trade was slight. Cattle rather than
trade goods, firearms or other imports remained the base of state
power. The regiments in which both Hutu and Tutsi served were
socially stratified with Hutu in the rank and file and Tutsi as
officers. Military units took their collective identity from their
named regimental herds. Very little currency mediated relations
of personal obligation.
To the east of the interlacustrine region a wide area was
occupied by pastoralist people and cattle-owning cultivators, who
did not form a common political organisation and were within
themselves, like the Langi, decentralised. This northern interior
region has continually provided the ingredients for cultural and
economic stereotypes, because it contains such sharp linguistic
and occupational differences. But autonomy and autarchy were
not the rule; Nilotic and Bantu-speaking groups, pastoralists and
cultivators, were in fact involved in complex and changing
interactions. The Masai, in particular, became labelled asfierceand
predatory. This image was sometimes promoted by traders
wishing to discourage rivals from establishing relations with the
Masai who, contrary to the propaganda, served very effectively
as local contacts in the ivory trade. The Masai in fact acted as
545

EAST AFRICA, 187O—1905
middlemen between the Dorobo hunters and the coastmen and
secured their relations with the Dorobo suppliers in a number of
ways including marriage, trusteeship of cattle, and so forth. In
times of distress when the milk and blood diet became impossible
to sustain, many Masai redefined themselves as Dorobo, at least
temporarily changing their livelihood and life style. The status of
Masai was thus contingent upon adequacy of cattle herds. As the
Swahili developed this commercial province in the northern Rift
Valley, they had to adapt to the special requirements of their
pastoral allies, supplying livestock and beads in exchange for
ivory. Slaves, firearms, and cloth had slight commercial value in
Masailand.
The Kikuyu and neighbouring peoples of the forested central
highlands of Kenya in the 1870s were absorbed mainly in the
expansion of their agriculture, but were not impervious to trade,
particularly in food. The Kamba attempted to retain the middle-
man function between the Kikuyu and the Swahili caravans and
had managed to do so before about 1870, when the main arteries
of the ivory trade had run far to the south, striking from
Mombasa, Vanga and Pangani due west toward Kilimanjaro
before turning northward into the Rift Valley. In the 1870s,
however, traffic ran closer and Kikuyu youths in the southern
districts such as Kiambu ranged further afield to make contact
with passing caravans and promote the direct sale of food.
Relations with the Masai deteriorated as Kikuyu warriors defended
agricultural activity in areas that also served as dry season
pastures.
From place to place, mobilising age cohorts, clients or ritual
followings, individuals emerged as entrepreneurs and local not-
ables. The territorial control of these big men remained limited.
Land was vested in local families and clans. Only Chief Mumia,
of the Luyia, who had successfully allied with traders, retained
Masai auxiliaries in order to embark upon expansion and con-
solidation of a tributary state.
The commercial hegemony wrought by demands for ivory and
slaves necessitated very little overt political control. Even the
allegiance of subjects of the Sultan of Zanzibar, especially in the
distant interior but also on the coast itself, was at best situational.
The power of Zanzibar rested above all in the hands of bankers,
brokers and other wealthy figures who dispensed credit and kept
546

accounts that summed up a continuing relationship with the
mobilisers of exports, traders and expediters who dispersed
throughout the Swahili sphere. The primary creditors were
generally Indian, many of the principal up-country traders were
Arab, the petty traders were usually islamised Africans, and
retainers were often slaves, sometimes of the second generation
or more. Slave categories like Nyasa and Manyema loosely
reflected origins in the interior. This pluralism notwithstanding,
the system was an organised whole engendering loyalty as long
as it was expansive and lucrative. The Swahili language was the
lingua franca of mercantile penetration, and to designate the entire
system Swahili is to stress trade and assimilation, rather than race.
In the metropole of the Swahili system before 1873, the sultan's
revenues had been derived to a significant degree from the export
of slaves. He was also a major owner of slaves and plantations,
who benefited fiscally from the general extension of clove
plantations worked by slave labour. Marketing of slaves was an
institutionalised economic activity, with the large majority of
slaves being brought from the Lake Malawi (Nyasa) area by Yao
and Swahili traders. The Yao had operated for over a century on
an overland route from the southern parts of the lake, while the
Swahili were relative latecomers. In the third quarter of the
century, the Swahili spread into the area to the north of the
Rovuma, where the stabilisation of the Ngoni conquest states
provided a new set of political and commercial conditions. Once
delivered to Kilwa, some slaves were shipped to Arabia. The
principal demand, however, came from the clove plantations of
Zanzibar and Pemba and the grain and coconut plantations of the
Malindi coast north of Mombasa, which had expanded rapidly in
the 1860s.
In the first years of his reign, Sultan Bargash (ruled 1870-88)
proved to be a' progressive' ruler in the eyes of the British officials
prosecuting the anti-slavery campaign in East Africa. His positive
policy was grounded in a constellation of conditions, some of
them internal and some external. Even to become installed,
Bargash had to make peace with the British, whose naval and
diplomatic power secured him against rival claimants. Econ-
omically, he was vulnerable, as succession to office was not
accompanied by the automatic transfer of his predecessor's
property, which had to be distributed to heirs. Over the course
547

of his reign, a sultan indeed stood to gain considerable property
in land and profit from trading activities, but the most secure
source of wealth at the outset was the lease of customs.
Customs masters at Zanzibar were invariably Indian, and often
of the Bhadia community. The long-serving Jairam Sewji, who
controlled the customs at the time of Bargash's succession, had
effectively become a state banker, extending credits to the head
of state, to European trading firms, to plantation owners and to
caravan merchants alike. Nearly his peer in wealth was Taria
Topan, who had cultivated a slightly different clientele - American
as opposed to British firms. In 1874—5, Taria was able to outbid
Jairam and took over the lease of customs for the ensuing five
years at a rent of $350,000. Afterward, Jairam Sewji bid the rent
up to $500,000 and resumed the position, while Taria Topan
concentrated on private business.
Great adjustments of the political economy of Zanzibar in the
1870s came on the heels of the 1872 hurricane, which, while
largely sparing Pemba, felled many clove trees on Zanzibar island.
The Sultan's fleet was among the losses to shipping. On the
mainland shore, such ports as Bagamoyo also sustained major
damage to housing and fishing vessels. Energies had to be
diverted to economic reconstruction in the islands and Pemba
emerged as the new centre of clove production. The world price
of cloves, which had declined owing to overproduction, suddenly
rose, giving the Pemba planters a high return on the harvest of
their surviving trees.
The prospects of obtaining new finance for clove development
became poorer in 1873, when Sultan Bargash signed a new
anti-slave-trade treaty with the British, agreeing to close the slave
market in Zanzibar and make illegal the shipment of slaves from
the mainland coast, whether to the islands or beyond. Indians
could no longer grant credit against the export of slaves, and new
lines of credit had to be established. Furthermore, with the agree-
ment of the government of India, the British consul, Sir John
Kirk, in 1874 began to apply increased leverage against British
Indian subjects who owned slaves. Ownership of plantations
became more than ever an Arab preserve. Indian bankers, unable
to operate in their own name plantations on which they had
foreclosed, dramatically raised mortgage interest rates for Arab
and other subjects of the sultan who still had the right to own
548

slaves. Indian capital found its outlet in credit against the delivery
of'legitimate' commodities, of which ivory still reigned supreme.
The outlawing of the slave trade was followed by various
manifestations of opposition and civil conflict in coastal com-
munities: antagonism towards missionaries who gave refuge to
runaway slaves flared into active hostilities near Mombasa; a
contingent of Baluchi soldiers had to be sent to Kilwa; the Mazrui
leader, Mbarak of Gazi, began a period of rebellion and brig-
andage lasting from 1874 to 1884. But these events were neither
unprecedented nor uncharacteristic of the factious coastal
leadership. A revealing episode at Bagamoyo in September 1872,
before the treaty had been concluded, gave evidence that Bargash
faced challenges from those installed by his predecessor, Majid.
A motive for the sultan's visit was a plan to promote Bagamayo
as the premier port of the coast. The development of Dar es
Salaam had been suspended at an early stage owing to Majid's
death, and Bargash intended to develop police posts and wells to
facilitate movement in and out of Bagamayo. The liwali and
commandant of the forces on the Mrima coast was a Baluchi, Sabri
bin Saflr, who, rather than paying homage to Bargash, virtually
snubbed him. Among the tasks facing Bargash quite apart from
the issue of the slave trade was the substitution of his own
appointees, and a new man became the liwali of Bagamayo in
1875.
In 1875 the sultanate of Zanzibar was also threatened by the
Egyptian occupation of ports on the upper coast. The commercial,
financial and political elites saw their common interest and the
need for Great Britain's protection. That Zanzibar did not own
the allegiance of these distant coastal communities was evident
from their warm reception of the Egyptians. The great Indian
merchants of Zanzibar had warehouses and other property of
considerable value in Kismayu, Marca and Brava, for which they
threatened to seek compensation if the ports were awarded to
Egypt. The decision in favour of Zanzibar's claims was quickly
followed by British measures to create a ' new sultanate' capable
of preserving a commercial sphere as well as extending formal
territorial control.
549

EAST AFRICA, 187O-I9O5
THE NEW SULTANATE AND THE SWAHILI COMMERCIAL

SYSTEM
The slave trade from and within East Africa had reached a peak
in the later 18 60s, due to overseas demand, the labour requirements
of an expanding coastal and island plantation sector, and the need
of porters for interior caravans. The British, American, German
and French commercial agents in Zanzibar may have dealt only
in 'legitimate' goods, but these included firearms, powder and
caps used in precipitating wars with their harvests of captives, as
well as cloth and other trade goods that went into the purchase
of slaves. As suppliers, Western traders were participants in the
Swahili system and a symbiosis amongst traders of all colours and
ideological postures continued until the turn of the century. The
western traders not only stocked trade goods, they also signalled
the new wants and rising prices of commodities consumed by the
industrialised countries. With the completion of the Suez Canal
in 1869, transport and communication between Europe and East
Africa became speedier and more regular. The treaties of 1873
and 1876 between the British and the sultan of Zanzibar did much
to constrict the slave trade. The Swahili system and the East
African economy proved to be equal to the task of absorbing
continuing deliveries of slaves. Rubber quickly took the place of
slaves as Kilwa's principal export. But, contrary to the assump-
tions of abolitionists, the prohibition on exporting slaves and the
promotion of a legitimate commodity like rubber did not strike
at the root of slavery.
The severing of Pemba and Zanzibar from their source of slaves
led to a split between the licit and illicit sectors of the economy.
A further blow came with the supplementary anti-slave-trade
treaty of 1876, and the organisation of an armed force trained by
an English officer, Lloyd Mathews. The treaty authorised, and the
army carried out, operations on the mainland coastal belt to deter
the movement of slaves from Kilwa northward overland. The
overland routes had sprung up when shipping became risky and
a glut of slaves developed. Suppliers of slaves had been encouraged
by a sustained demand, fed particularly by Malindi planters taking
advantage of low prices. Slaves were marched north from
Kilwa to the coast opposite Pemba, making Pangani and Sadani
commercially important. Further up the coast, the plantations at
5 5°

THE SWAHILI COMMERCIAL SYSTEM
Malindi absorbed quantities of slaves, and beyond that, Lamu and

its neighbours provided points of embarkation for slaves aboard
dhows headed to Arabia with grain and mangrove lumber. Cheap
slaves were also set to work to increase the area under cultivation
in the Shebele valley as well as reinforcing the economic life of
the coastal towns on the lower Somali coast, recently returned to
Zanzibar. Kirk estimated that 10,000 slaves crossed the Juba each
year during the 1870s.
The licit and illicit sectors of East African commerce cannot
be entirely disentangled, but the public separation had several
important consequences. First of all, an alliance formed among
those committed primarily to the licit trade. The partners included
the sultan, the major Indian banker-brokers, and Swahili mer-
chants primarily concerned with the export of ivory, the most
famous of whom was Hamed bin Mohammed, Tippu Tip. At the
forefront of the organised illicit sector were those engaged in
replenishing supplies of slave labour for the coastal and island
plantations. Whereas credit for ivory traders came from the Asian
capitalists directly, resources for the illicit trade had to come from
secondary credit relations and individual initiative. Pemba planters
were known to use their own vessels and retainers to bring slaves
across the channel. Credit remained the basis of the Swahili
commercial system, but in the illicit sector it came increasingly
to be shared out to operators of small and often unseaworthy
vessels and to a host of other petty traders who mixed slave and
ivory trading. Although a primary creditor might operate entirely
in the licit sector from the point of view of goods directly given
and received, there is ample evidence that the multiple links in
the long chain of credit entailed transactions of many sorts. The
commercial system had roots in localised and regional trade. That
slaves were bought and sold at various moments between the
despatch of goods into the interior and the ultimate emergence
of a recordable commodity at the coast was not a legal issue in
any event. Where the sultan's jurisdiction denned slave trade as
illegal, that is, along the coast, the demand for slaves did not let
up after the supplementary treaty of 1876. When the newly
planted clove trees began to bear, harvesters were badly needed
and the price of slaves rose. The risks of illicit trading were worth
taking and the sector was well organised. By the same time, about
1879, tîe promotion of rubber collection in the Kilwa hinterland

had established a legitimate commodity whose export value
exceeded that of ivory. Credit and energy on the southern coast
became more absorbed in this activity, which supported Kilwa's
economy as the traffic in Nyasa slaves declined.
Expansive, absorbtive, and confident, the Swahili system coped
with and profited from the influx of money to carry out European
exploration and establish missionary outposts. Expediting goods
and restocking caravans increased the scale of operations of the
licit operators, the magnates of trade, who were reported to the
world to be gentlemen who graciously eased the travellers' way,
trustingly took cheques drawn on bankers in Zanzibar, and even
stood ready to advance the cause of colonialism. The lower
elements of the commercial world, on the contrary, were described
as vicious, violent and thoroughly committed to slave trading.
The swelling ranks of Swahili society were indeed stratified. Not
all Omanis among the large numbers coming to East Africa in
the 1870s and 1880s were economically independent. Many took
menial jobs at first, but they had ample opportunities for
advancement. The upper strata of Zanzibar, the plantation
oligarchy and the principal ivory merchants, constituted what may
be called the 'Arab' class, access to which was facilitated by
literacy, a modicum of Muslim learning, and an Arab forebear.
Omanis moved into the class readily, but Swahili freemen could
also be accorded the same respect if they became wealthy. Trusted
slaves who rose as agents for their masters replicated the social
order when they had a chance to assume command. Slave porters,
even those without special privileges or rewards, had freedom of
movement and a sense of professionalism. Like skilled slave
craftsmen, they retained a portion of their earnings. Where ties
of obligation were loose and fairly mutual, a sense of identification
grew between master and slave. Caravan porters vented their
displeasure on those foreign employers who were more concerned
to cut costs than to pay adequately or look beyond a short-term
relationship. The professional porters, pagaî, expressed more and
more proletarian consciousness as colonialism advanced. In the
18 70s, however, the mixture of slave and hired labour tended to
strengthen a common Swahili identity. Assimilation and member-
ship led to cultural identification with Islam in its most outward,
communal features, such as the ritual killing of meat, slaughtering
of animals, and holy feasts and fasts.
552

The cadre of the Swahili system was composed of Wangwana,
a term interpreted to mean the free-born and the civilised.
However ambiguous their freedom or however remote their
personal experience of the coast, the upwardly mobile elements
of the Swahili community coveted this status. In its prosperous
days, the Swahili commercial system gave much employment,
some opportunity for advancement, and to its regular personnel
an aura of superiority. The Islamic element played a role in class
stratification, since owners or shareholders or other great
personages made a display of their fasts and daily prayers and
feasted everyone on special religious days. Banners blessed by holy
men at the coast symbolised a caravan unit, and porters with
elevated status as butchers saw that meat was slaughtered with
ritual propriety. Wherever possible, Swahili depot communities
grew rice; diet too came to set apart the Wangwana.
The incorporation of people into Swahili culture and economy
was nowhere more evident than in Manyema, the trans-
Tanganyika commercial province created in the later 1860s. An
observation made by V. L. Cameron in 1874 succinctly and
accurately summarised the situation:
Very few slaves are exported from Manyuema by the Arabs for profit, but are
obtained to fill their harems, to cultivate the farms which always surround the
permanent camps, and to act as porters.
By the time a caravan arrives at Tanganyika from westward nearly fifty
percent have made their escape, and the majority of those remaining are
disposed of at Ujiji and Unyanyembe, frequently as hire for free porters, so
that comparatively few reach the coast.1
A profound dislocation and reconstruction had taken place in

Manyema in a short period of time, marked by violence and
reincorporation within the new commercial economy. The title
Wangwana was appropriated by Swahilised Manyema, regardless
of the fact that they had often been impressed into the community
rather than freely joining it. Chaotic conditions, and the lack of
control over slaves being transported towards the east, drove
many to become refugees. The masterless Manyema who offered
themselves for hire in Tabora eventually formed a highly exploited
labour force.
The smashing of old communities and creation of Swahili-
dominated ones which occurred in Manyema was an unusual
1
V. L. Cameron, Across Africa (London, 1877), vol. 11, zo.
553

consequence of commercial expansion. More often, the dynamics
of regional politics and economics in this decade reflected more
continuity in the combination of internal conditions and the new
or intensified commerce. Two contrasting examples of how
traditionalist and commercial parties formed within states are to
be found in the Shambaa kingdom and the chiefdom of Unyany-
embe. In the former, commerce prevailed at the expense of the
centralised state, while in the latter, the commercial faction
remained continually involved in dynastic politics, ultimately
reinforcing and preserving the state.
The Shambaa kingdom, with its capital of Vugha high in the
Usambara mountains, stood between the coast and Mount
Kilimanjaro. In the early nineteenth century, relations with the
people of the lower plains country had been fairly amicable, based
on complementarity of hills cultivators and plains pastoralists.
From mid-century, however, the development of regular Swahili
trade routes had altered the political and economic calculus.
Kimweri, the king of Shambaai, emerged as an effective controller
of export trade, whose writ extended over a wide region and who
remained in with leaders at the coast and in Zanzibar. Kimweri's
death in 1869 precipitated a succession struggle, pitting Semboja,
whose orientation and alliances rested in the commerce of the
plains, against a traditionalist party focused on the ritual and
political centre of the monarchy at Vugha. A revolt by the
subjugated Bondei in 1871 escalated the level of violence in the
Pangani valley; captives sold to traders streamed into the coastal
slave depots. For more than a decade, the highland and lowland
Shambaa remained at enmity. In the early 1880s the traditionalists
leagued with Mbarak, the Mazrui chieftain, whose brigandage
kept up a pattern of turmoil, dislodging people from communities
and settled production in the Mombasa hinterland. Slaves were
taken and sold for guns with which to carry on the civil war.
Without question, the failure to develop a 'legitimate' export
product contributed to brutalised social relations, increased
callousness of rulers, and chronic treatment of people in this
region as a commodity rather than as producers of surplus. The
demand for these slaves came particularly from Pemba, where
some planters began to organise the requisition and transportation
of slaves from the mainland. Petty traders also participated in
clandestine slave trading. Their supplies were increased by the
554

famines in Ukambani in 1882—3 an<^ m Mijikenda and Kikuyu

countries in 1884. So hunger and pawnship were added to
violence and judicial harshness as steps to enslavement and
replenishment of plantation labour.
The unresolved civil war which plagued Shambaai and the
Pangani valley did not have a parallel in the Kilimanjaro area,
where caravans were more interested in securing food supplies
and facilities before entering or after leaving the ivory frontiers
in and beyond Masailand, than they were in acquiring slaves.
Mangi Rindi of Moshi gained an international reputation by
entertaining explorers, who accepted uncritically his claim to be
' king' of Chaga. Strong as Rindi was, he had effective opponents
to his expansionist designs, and the Chaga as a whole wished to
be organised neither as a centralised state nor as the core of a
regional sub-system. The independence of the Chaga valleys from
one another bears comparison with the autonomy of the Kikuyu
mbari in this period, in respect to their alliances with pastoralists
and attitude towards trade in food with caravans. The contrast
of Shambaai with Chaga and Kikuyu countries at this point in time
suggests that degrees of centralised state formation, dynastic
feuds, and geographical position with respect to the main zone
of illicit trading must all be considered in analysing the tragedy
of the Pangani valley.
Tabora, a Swahili settlement within the territory of the Nyam-
wezi state of Unyanyembe, became a focus of conflict in the early
1870s, when Mirambo, the ruler of the neighbouring chiefdom
of Urambo, undertook a major military campaign with the object
of dislodging the commercial community from Tabora and
obliging it to relocate at his own capital. The hold-up of ivory
effected by Mirambo as part of this campaign proved that he could
indeed disrupt the flow of trade from the interlacustrine and
trans-Tanganyika trading frontiers. Owing to the resultant scarcity
in 1873, the price of ivory at Zanzibar rose dramatically. The
sultan's revenues, even with the recent rise in the customs tariff
charged on ivory, were endangered. In 1875 Bargash despatched
a large and costly expedition to Tabora to discipline the feuding
elements and pave the way for peace with Mirambo. Peace was
made in 1876, Tabora remained the interior entrepot of the
Swahili system, and Mirambo turned his energies to imposing his
military-administrative regime in the region between Unyany-
555

embe in the south and Lake Victoria and Karagwe in the north.
Within this region, he provided an umbrella for trade, from which
he derived wealth, especially when, in the later 1870s and 1880s,
the Lake Victoria basin became dominated by the Ganda and
traffic through Sukumaland increased.
What is masked by older accounts of Mirambo's struggle for
control of the trade is the degree to which it entailed an attack
upon the Nyanyembe as well as the Arabs. It was increasingly
evident in the 1870s that the merchant faction in Unyanyembe,
the wandewa, had gained enough power in chiefdom affairs to
install their candidate as the paramount chief or ntemi. This party
intermarried with the coast-derived Tabora community and
shared many of its interests — organising caravans, restocking in
Tabora, and maintaining regular trade networks. A traditionalist
party also existed among the Nyanyembe, stressing the old
sources of political legitimation as against the new wealth and
loyalties based in trade. Under certain conditions, candidates who
succeeded to the position of ntemi invoked ancient patriarchy and
drew on the persisting extra-commercial sources of power and
patronage. Control of the highest office indeed swung between
traditionalists and merchants, with consequent shifts in political
mood and levels of tribute exacted from the trading community.
Without doubt, however, by the late 1870s Unyanyembe as
well as Urambo was secure for the operations of the Swahili
system, whose values were matched by those of influential
Nyamwezi.
The Nyanyembe indeed predominated within the Swahili
system as it extended in the direction of Lake Mweru and
Katanga. In Katanga, the Nyamwezi offshoot state of Msiri (ruled
1868—91) had been consolidated in the third quarter of the
nineteenth century and reached a peak in the 1880s. The leaders
and members of irregular military units, rugaruga, and many
transient traders known in this region as Yeke, had their roots
in Tabora-Unyanyembe and returned to it periodically. Before
counting Msiri as fully a part of the Swahili system, however, a
qualification must be introduced, for after 1870 his economic
domain was regularly served by Ovimbundu traders from Bihe
in Angola. Msiri watched with increasing apprehension as Swahili
roughnecks eroded the imperium of Kazembe on the Luapula, and
he eventually preferred the more orderly routes linking him
commercially with the Atlantic.
556

As one Nyamwezi sphere was politically organised by Mirambo

in the north, so was another created in the south by Nyungu ya
Mawe, a prince eliminated from contention in Unyanyembe.
The Kimbu state was organised in what had been a region of
commercial uncertainty in 1870 when the Tabora-based Muslim
Indian, Amran Masudi, had ambitiously aimed to conquer the
Sangu paramount, Merere. With his considerable army, and the
full support of a small community of resident Swahili, whose
interests were also threatened, Merere decisively defeated Amran.
In the following years Nyungu avoided contesting control of
those commercial routes, pasture lands or ivory sources that were
important to the Sangu, and developed his administrative, military
and commercial domain in the bushy plateau country between
Lake Rukwa and Unyanyembe. Beyond Nyungu's political con-
trol, but moving ivory through the Kimbu capital, Kiwele, was
a dissident member of the Sangu royal family, Kimalaunga, who
imposed himself as a warlord in the elephant country south of
Lake Rukwa. The older Namwanga, Fipa and Mambwe polities
therefore came under double encroachment, in the north from
Kimalaunga and his rugaruga and in the south from the expanding
frontiers and raiding activities of the Bemba.
The maturation of the Swahili system was becoming evident
by the end of the 1870s. The coastmen were powers to be
reckoned with, having proven their capacity to reduce weak rulers
or to make alliances threatening or maintaining the status quo of
a locality. The roughneck function, sometimes performed by
armed caravans before and during the 1870s, was taken over
increasingly by regional power-wielders concerned to control
aspects of trade and extend territorial jurisdiction. This trend
toward regional paramountcy was not fully realised everywhere,
and could still be deflected by strong external forces.
The leading coastal traders were divided in their policy toward
the emergent regional paramounts. A considerable segment of the
Tabora community had wished to confront Mirambo, while
Tippu Tip and his family adopted a friendly neutrality. By
preferring to see strong rulers maintain the peace over wide areas
Tippu Tip conserved his own energies for ivory trading and
administrative regulatory activities in the commercial provinces
west and south of Lake Tanganyika, where no indigenous
regional powers emerged to organise resources and provide an
umbrella for trade.
557

As an area economically incorporated into the Swahili system,

East Africa in 1880 was as interconnected as it was ever to be.
Connectedness had a violent aspect, especially in places of almost
institutional violence, where militarised elements with a narrow
economic base preyed upon resilient and self-sustaining cultivating
communities. For purposes of defence as well as offence, firearms
and their suppliers were welcomed far and wide. With extended
lines of communication, traders often needed to cultivate the
friendship of chiefs by including guns and powder as trade goods,
even if credit-giving in cloth, beads and copper remained a
pervasive means of engaging to obtain a range of services and
access to commodities.
The major regional powers, preoccupied as they were with
gathering tribute through subordinate chiefs and with regulating
traders, were not the agents of change in everyday life that chiefs
over smaller areas were. Local lords, whether tributary to a
superior or independent over time assumed more explicit juris-
diction, concentrated greater surpluses, gave refuge to unsettled
persons, accumulated women, dispensed patronage and generally
enjoyed enhanced authority. Police and armed forces were regu-
larised. For lack of a better term the resultant phenomenon can
be labelled 'seigneurialism'. It nourished especially in a context
of mixed regional production and exchange where long-distance
trade contributed additional economic resources and small but
critical quantities of capital through credit from traders. The
seigneurs deserve attention because they formed focal points
within the East African economy. They presided over a delicate
and never fully institutionalised shift in the deployment of labour,
employing strategies of redistribution as well as protective and
coercive powers.
If and where there was prosperity in East Africa which entailed
more than the transfer of assets from one group or class to
another, it was to be found in new levels of productivity and
intensification of effort. Seigneurial management deflected labour
into more intensive and large-scale production, into military
service and into porterage. In addition, regional economies
became more marked by specialisation. The histories of iron
smelting and forging in the Nyasa—Tanganyika corridor and
western Usukuma contain evidence of increased manufacturing
for trade over extended areas. As salt remained a staple of regional
558

trade, salt sources became more and more frequented. Each region
had its own near-currencies based on generally accepted use
values, such as iron, hoes, salt, cattle and goats. Imported cloth
was beginning to figure as a near-currency, although it did not
diminish the call for the domestically produced units of account.
Currencies of a more abstract sort were also regionally specific.
In the market-places of the interlacustrine region, cowries were
commonly exchanged. Throughout the central and south-western
region where the Nyamwezi networks existed, copper wire, often
wrapped around fibres as bracelets, constituted the smaller units
of currency.
A regional effect of heightened caravan traffic in the late 1870s
and 18 80s is exemplified by Ufipa near the south-eastern shores
of Lake Tanganyika. The economy of this area was rooted in
long-standing exchanges between cultivators with different seas-
onalities, above and below a rift escarpment. Lake Tanganyika
and the rivers flowing into Lake Rukwa supplied fish. Iron
smelting and weaving of locally grown cotton complemented
food production. Goods in trade comprised iron, salt, and
tobacco. The ruling family of Ufipa had divided the country into
two separate jurisdictions, the western one, Nkansi, being the
more affluent, in part because it controlled the principal trade
routes north and south from Unyanyembe to Katanga. Of critical
importance was the lake crossing from Kirando to Moliro and the
ruler of Nkansi profited considerably from the tolls he charged
whenever his subjects ferried a caravan across. It was characteristic
of the relationship of ruler and ruled in this particular region,
however, that the boatmen bargained separately for their fees.
Altogether the Nkansi rulers can be said to have drawn lightly
upon the resources of their land and people, and avoided
intervening greatly in its essentially sound and diversified domestic
economy. The apparatus of government nevertheless became
strengthened, in part to assure internal order and to remind
producers of their obligations of tribute, and in part to contend
with the threat posed by Kimalaunga. While the political and
military context of Bunyoro in the 1880s spawned a greater degree
of military organisation and activity over a wider area, the
essentials of the economic bases in Ufipa and Bunyoro were
similar.
At the jugular of the ivory trade in 1880 was the community
559

EAST A F R I C A , 187O—1905
of Ujiji which stood at the main point of embarkation for
Manyema across Lake Tanganyika. There could be no question
of the Jiji contesting economic control of the long-distance trade,
or exercising social and political control over the traders and their
entourages. They made the best of their situation, however, by
intensifying agriculture in the rich Luiche river delta until it
became a granary for settlements on both sides of the lake. As
canoemen, sailors and fishermen, they plied the waters for wages
and independent trade. So production, trade and transport made
for a diversified domestic economy. The important role of the Jiji
as producers, their sharp commercial consciousness and unity in
the market had effect in counteracting inflation and adverse terms
of trade. Just as the rulers of Nkansi confined themselves to
moderate tribute, claims to a royalty on ivory, and profits from
caravan-servicing, leaving the domestic economy to its own
dynamics, so the Swahili leaders in Ujiji did not aim at total
control of the Jiji. Again, as with demographic change and greater
and more varied employment, the Wangwana, the Nyamwezi and
many other social groups became enlarged by absorbing and
acculturating people, so the Jiji swelled their number by incor-
porating strangers and slaves.
A final example of regionalism within East Africa in the early
1880s comes from the southern interior. There the Hehe state
formed the dominant element in a balance of power between three
military organisations, the others being the Sangu and the Ngoni
of Songea. The Sangu and the Ngoni became allied against the
Hehe and occasionally co-ordinated their expeditions against
militarily weaker but intrinsically wealthy areas, such as Unyak-
yusa. An essentially sound mixed agricultural economy sustained
the state and very little is known of Hehe ties with long-distance
traders. The major preoccupation of the rulers was to control the
southern highlands by excluding any military rival, although they
did gain cattle and pasturelands through military campaigns
against the Sangu. Beneath the military ebb and flow of the
struggle between the major powers, subordination, displacement
and reconstruction took place among lesser groups of the region.
In the Mahenge area, the Mbunda people regrouped after being
displaced from the area of Ngoni consolidation in Songea to the
west, organising themselves into military units which copied the
weaponry and tactics of their conquerors. The Bena groups under
560

TOWARDS A CLIMAX OF WESTERN PENETRATION
Kiwanga, expelled from the highlands in the course of the

Sangu—Hehe struggle, also became established in Mahenge,
further rearranging the lines of politics and force in this fertile
and well-watered but heretofore more sparsely populated and
economically less exploited countryside.
Mahenge emerged in the 1880s and early 1890s as the source
area of raiding parties which descended upon the mostly Ngindo-
speaking populations of the Kilwa hinterland. This activity was
triggered by famines, but also aimed to appropriate wealth
derived from rubber production and trade, by then the mainstay
of Kilwa's exports. The southern interior, even with its social
fragmentation, recent political reorganisation and the high pre-
mium on military power, was also a zone of Swahili penetration
of growing importance. Caravan routes became regularised and
traders' settlements sprouted between Kilwa and Lake Malawi.
Slaves came from the western shores of Lake Malawi, trade with
Ngoni rulers in Songea flourished, and rubber, constantly
increasing in volume and price came from this region north of
the Rovuma.
For decades, caravans had been organised and porters recruited

on behalf of European travellers, whose comings and goings
provided the East African economy with a veritable tourist
industry. A basic change took place when the African Inter-
national Association (AIA) and various missionary agencies began
to establish continuously occupied stations. The need to maintain
supply lines for these resident whites remained a good source of
business, but by 1883 the intentions of the AIA were obviously
territorial and economic, not merely humanitarian and explora-
tory. The magnates of the ivory trade perforce treated their
incipient rival, the AIA with a measure of diplomacy. On the other
hand, missionary outposts fitted into local patterns of politics and
economic life, becoming centres of refuge, reconnaissance, and
diplomacy. They were, of course, also harbingers of low-paid
wage labour and spread Western ideologies and cultural values.
Among the most important Christian agencies in East Africa
in the 1880s were the White Fathers, who in 1879 had followed
the Church Missionary Society (CMS) to Buganda and spread
561

around Lake Malawi and Lake Tanganyika, in the latter case

taking over the AIA station of Karema. The London Missionary
Society (LMS) had become established in Unyamwezi and ex-
panded to Lake Tanganyika and to the south, where they built
stations among the Lungu and Mambwe people. The LMS floated
the first steam vessels on Lake Tanganyika, but rather than
challenging the status quo in any marked way the captains tended
to build relationships with the Swahili. A rather different situation
developed in the south, where a Scottish firm began to organise
steamer and overland carriage of goods to service the Scottish and
English Protestant mission stations near Lake Malawi and along
the corridor running between Lakes Malawi and Tanganyika. The
African Lakes Company (ALC), as it was called, did business also
with the general public. Soon after it made Karonga a regular
depot in 1884, many up-country Muslim traders found it con-
venient to shorten their own lines of supply and brought ivory
to exchange for textiles, beads and other trade goods.
Ivory trading and missionary supplying were also combined in
the person of Charles Stokes, who after the death of Mirambo in
1884 became an increasingly important mobiliser of the Sukuma
trade and transportation system. Among those businessmen who
specialised in expediting for Europeans was Sewa Hadji, an Indian
based mainly at Bagamoyo. Other Indians too made their fortunes
in this era, in positive alliance with colonial elements, and were
favourably placed to extend their own business as the reach of
colonial authority lengthened.
The Swahili system also adapted to the new times through an
organisation, the HM Company, under the leadership of Tippu
Tip. 2 This venture has been described by Tippu Tip himself as
one commissioned by Sultan Bargash. From corroborating sources
it is evident that the International African Association had offered
Tippu Tip recognition and support if he would agree to become
their agent, using his commercial networks and strong points to
secure their territorial ends. Whatever circumstances inspired the
formation of the HM company, it emerged in 1883 and 1884 as
an independent operation, although strongly backed by the
coalition of Sultan Bargash and Taria Topan, two central figures
in the 'licit' trading sector. Taria advanced large quantities of
2
HM was Tippu Tip's mark on ivory, after his initials (Hamed b. Mohammed). See
Kaiserlichcs Bezirksamt, Dar es Salaam, 1895. National Archives of Tanzania, G.
z 1/920.
562

goods, with which Tippu proceeded to make an alliance with

Mohammed b. Halfan (usually known as Rumaliza), who had a
very large following but little capital or access to credit. By so
doing, Tippu enlisted in partnership a chieftain of the Muslim
brotherhood, the Qadiriyya tariqa. The third partner in the
company was Tippu's relative and deputy, Mohammed b. Said.
So capital and personnel were harnessed in a programme aimed
at keeping the ivory trade in Swahili hands and passing through
Zanzibar. In the middle 1880s, the HM Company extended itself
from Ujiji, where after 1883 Rumaliza was the supreme figure in
the Swahili community, to Manyema and its surrounding territory
up to Stanley Falls. In the south, the company retained a strong
lieutenant, Abdullah ibn Suleiman, who identified himself ex-
plicitly as the agent of Tippu Tip and Rumaliza, and acted as a
stabilising and disciplining force among the other, more unruly,
traders. Eventually, Abdullah dissociated himself from the anti-
European Swahili, who in 1887 began to conflict with the African
Lakes Company (ALC) at Karonga.
The posture of the HM Company was consistently positive
towards the Europeans and cognisant of the mounting realities.
Tippu Tip finally agreed in 1886 to recognise the Congo Indepen-
dent State and become its local governor in the eastern portion
where he had commercial hegemony. Tensions over the non-
delivery of arms and ammunition and the Congo State's efforts
to deflect the flow of ivory eventually mounted and drove the
Swahili to opposition, but until the early 1890s, the HM Company
continued to facilitate colonial expansion and business interests,
especially those which were eastern oriented. Rumaliza, in 1890,
proved to be willing to consign a large quantity of ivory to the
ALC for transport by their southern routes. At the same time he
advised his agents to render every assistance to Captain Swann,
the former LMS steamer captain who had entered the employment
of the Foreign Office, and sought to make treaties in a corridor
from Lake Tanganyika to Uganda.
The religious complexities of East Africa in the 1880s provide

fertile ground for discussion from various perspectives. In non-
Muslim, non-Christian communities, 'traditional' religious ob-
servances were triggered both by normal seasonal changes and
natural catastrophe, famine and sickness. Kabaka Mutesa's phy-
563

sical decline during 1880-4, coinciding with epidemics at the
capital of Buganda, gave rise to a rhetoric of traditionalism and
renewed respect for the deities and their representatives.
Elsewhere, famine gave rise to seers and prophets, whose
ambiguous message could be construed as being either against or
in favour of the Christian preaching and the new culture of power
soon to become overt colonialism. The early African Christian
leaders, far more than the white missionaries, looked favourably
upon prophecy as a form of legitimation within their localities.
They also interpreted their own dreams to arrive at a statement
of inner conviction. More often, famine and violence generated
purely secular motives for gathering under the wing of mission-
aries. By a combination of employment and teaching, small nuclei
of trusted men emerged at the missions and became community
functionaries and disseminators of the new values. Conformity
and socialisation, rather than spiritual conversion, thus proved to
be the major if not the exclusive means of recruiting adherents
to Christianity as well as to Islam.
Frustrated missionaries, whose evangelical passions were not
reciprocated, sometimes targeted Muslims as objects of concern.
A missionary in Unyamwezi, encountering an unschooled
Muslim, became so agitated at the latter's total lack of theological
grounding that he undertook to teach him the basic tenets of
Islam. Other new cosmologies sometimes fared better than
Christianity. The Bachwezi cult, with its abstract deism, spread
from the interlacustrine region at this time and took root in
Unyamwezi. The missionaries kept their eyes fixed upon Islam,
however, tending to discount African religious systems as en-
during rivals.
Christian communities exerted a degree of social control, with
the goals of modifying family life primarily through the practice
of monogamy. They also endeavoured to rebut traditional expla-
nations of disease and misfortune. In the 1880s, missionaries at
many stations in the central and south-west regions of the Swahili
sphere were still unable to do without Wangwana or Swahili
workers, and thus had in their own employ exponents of Islam
and polygamy. While the White Fathers accepted this situation
as a temporary expedient, the London Missionary Society in 1889
abruptly discharged its Muslim employees, complaining that they
misled the local people by claiming that Islam was the equal of
564

Christianity. The release of the imported personnel was symp-

tomatic of a breakdown in the symbiosis between the resident
Westerners and the Swahili commercial system. Furthermore, the
move was a withdrawal from the labour market. Local men could
be paid much less than the wage- and conditions-conscious
Swahili workers.
Dependency among those clustered around mission stations
had many degrees. There were liberated slaves deposited by naval
or colonial authorities, runaway slaves and refugees. The source
of these dependants was not always humanitarian or voluntary.
Old and disabled people and infants were sometimes delivered to
the missionaries at Karema on Lake Tanganyika by Swahili
traders cutting their losses. In other cases, however, sheer security
was the attraction; the mission station was sometimes the best-
armed and defended of any place in the locality.
The general picture of missionary stations as bastions of
paternalism and refuge, while it must stand, simply does not
encompass the situation prevailing in the special psychological
and political climate of Buganda. The Church Missionary Society
and the White Fathers remained close to the capital, under
scrutiny but also with access to the highly motivated youths
serving the kabaka as pages and apprentices for high office. The
intimate contact between missionaries and youths and among the
youths themselves created a hot-house environment in which
allegiances grew rapidly. The execution of allegedly disloyal
Muslim pages in 1876 indicated that even a ruler at the height of
his power would act radically to remind functionaries that their
first duty was to him. The accession of Mwanga on the death of
Mutesa in 1884 brought to power a young man who had grown
up in an era when court education had become intimately
associated with Islam and Christianity, Protestant and Catholic.
The desire to assert control over external factors, anxiety about
those seeking to penetrate from the unshielded east around the
northern shores of Lake Victoria, and a crisis of loyalty
culminating in a second purge, this time of Christian pages in
1885-6, all set the tone for Mwanga's troubled reign. His actions
had much to do with a new ruler's initial insecurity and desire
to rivet the attention of his political world, but in the context of
advancing colonialism, a process of personal consolidation was
never carried through. Those who escaped execution were
565

afterwards organised into three companies, Muslim, Catholic
and Protestant, thus deepening the division of the elite along
denominational lines. These special companies also had licence
from the kabaka to spread awe and terror.
While missionaries in Buganda promoted the vision of a
Christianised state, the illicit sectors of the East African economy
also received great publicity. In 1884, just as the powers convened
in Berlin to decide on the rules for Africa's partition, Bismarck
recognised treaties made by the colonial adventurer, Carl Peters,
and in 1885 these became the basis of a German protectorate on
the mainland opposite Zanzibar and Pemba. Slave trading had
surged upward in 1884, meeting a sustained demand by planta-
tions. More than before, the supply came from people of the
coastal hinterland, including Zaramo, Mijikenda and Pangani—
Taveta groups from the near interior. Humanitarians denounced
the continuing slave trade and invoked imperial intervention.
Within East Africa, increased ill-feeling poisoned relations
between the Mijikenda and the coastal communities which had
exploited conditions of famine and entrustment by selling pawns.
Under close examination, the theory that there was a Muslim
conspiracy against Christians at the end of the 1880s cannot be
confirmed. The supposed conspiracy was deduced from coinci-
dence. Indeed, Christian traders were engaged in a struggle with
Muslim traders at Karonga, a Muslim' revolution' was under way
in Buganda, and the Germans were locked in war with vested
interests on the Mrima part of the coast. There can be no doubt
that a sense of confrontation and rivalry rippled through the
rumour-prone Swahili system, to the particular consternation of
those committed to the illicit sector. Numerous owners, attempt-
ing to convert their wealth quickly to licit assets, sold slaves and
thus generated the first major increase in supply since the 1884
famine. Others took a decision to stand and fight. At Karonga,
the Swahili traders' powers was in a somewhat weakened condition
as against the solidified Ngonde—ALC alliance. Nervous rhetoric
on the Muslim side and an aggressive military response by the
allies precipitated a bloody conflict.
In Buganda, Kabaka Mwanga had encouraged the Muslim
party at court as a counterpoise to the rapidly increasing numbers
of Protestants and Catholics. A desperate attempt to eliminate
prominent men in all three factions failed in September 1888 and
566

THE INITIAL COLONIAL EXPERIENCE
Mwanga was forced into exile. The subsequent distribution of

offices according to religious identity was followed by a coup
carried out by Muslims who aspired to a more radical revolution
to bring the country under a Muslim theocracy. The subsequent
civil wars, conducted under religious banners, sealed a pattern of
political partisanship which had been brewing in Buganda since
the 1870s and carried on until the mid-twentieth century.
There was no co-ordinated conspiracy to prevent the estab-
lishment of Christianity and Western imperialism. Only in Buganda
was Islam militant. What brought about such widely separated
altercations in the years 1887 to 1889 was nothing less than the
militancy of the new colonialism, often cloaked in theology and
humanitarianism, but nevertheless on the offensive, engaging in
the first instance with opposition which presented itself as
Muslim. The outstanding examples of colonial provocation came
with the German occupation of the Mrima.
After 1885, East Africa became both balkanised and subject to

new generalisation. Germany had precipitated its partition by
declaring a protectorate over a portion of the hinterland of
Bagamoyo on the basis of treaties made by Carl Peters. By 1890,
international recognition was given to Italy's claims to the Somali
coast as far south as the Juba river, Britain's sphere from the Juba
river to Vanga, and Portuguese extension up to the Rovuma in
northern Mozambique. The Congo Independent State hulked an
ungainly giant in the trans-Tanganyika area, acquiescing for the
moment in the necessity of Swahili and even missionary
management. In the later 1880s, indeed, colonial ambitions were
pursued indirectly through the economically feeble chartered
companies, or other organisations already present and able to
operate within well-established infrastructures, even where they
did not coincide with the intended boundaries of the colonial
states. The active conquest of East Africa took place between 1888
and 1900 through sporadic and sustained military campaigns best
called colonial wars. The context for the colonial wars was shaped
by two other kinds of profound dislocation and adversity, an
ecological crisis and a collapse of the old commercial system.
These three elements may be treated separately; their full
567

importance however can only be assessed when they are seen in

conjunction.
The colonial wars in East Africa began in German territory,
for the simple reason that agents of the German East Africa
Company were the first to take over control with a view to
establishing a new political entity. Colonial wars were not
distinguished by the immediate issues or the colour of commanders
and personnel, but rather by their context of the new imperialism
as it came to terms with foregoing conditions. The specific task
of state-building stood behind colonial wars, but the idea of a
colonial state in East Africa was still an abstraction in 1887. When
the Swahili fought the Abushiri war on the Mrima and its
hinterland in 1888, they acted as much to carry out a longstanding
determination to prevent encroachment by the sultanate of
Zanzibar as they did to strike at the Germans in themselves.
Liberty was a slogan of Swahili self-determination not created by
the partition of Africa. Abushiri was the scion of the al-Harthi
clan, whose long enmity with the Busaidi dynasty made his
resistance to any intervention on the coast predictable. Further-
more, his trading operations and large community of followers
in Pangani had involved him deeply in the illicit sector of the
commercial system. The Zigua chief of Sadani, Bwana Heri, who
opposed concessions by the Sultan and the overbearing behaviour
of the new Europeans, joined in the war. The unity of Abushiri
and Bwana Heri was incomplete and the terms of their defeat
different. The Germans executed Abushiri and confirmed Bwana
Heri in office.
The Abushiri war was a highly strategic one, as it affected the
entire status of Zanzibar and the main thoroughfares of caravan
trade. The German victory guaranteed acquiescence of the
Swahili notables and commercial system. At the same time, the
war left a strong residue of essentially African colonial forces,
Swahili raised locally and Shangaan ' Zulu' and Sudanese recruited
abroad. The German East Africa Company's charter was ter-
minated and direct responsibility was assumed in 1891 by the
German Foreign Ministry. Measures of careful accommodation
to the cultural values of Islam followed the pacification of the
coast, and the new administration made efforts to incorporate
Swahili elements into the colonial state. While the coastal districts
became models of future administrative practices, the main towns
anchoring the caravan routes were occupied by garrisons.
568

The crumbling of the Swahili sphere was evident in 1891. Credit

for the interior trade was reported to be sharply reduced and old
networks of trust, both the licit and illicit, were soon in dissolution.
The local commercial crisis reached its depths in 1895. Among
the many contributory factors, the partition of East Africa figured
in several ways. It could be predicted that the handling of ivory
in the extremities of the sphere would become a matter of
contention increasingly to be taken under the control of colonial
regimes. Closer to Zanzibar, profit margins were badly affected
by the double customs imposed in 1891, when both the German
East African and British Zanzibar officials taxed goods passing
through Zanzibar to and from German East Africa. Zanzibar was
soon declared a free port in order to retain as far as possible its
role as entrepot for all of East Africa, but for some months before
the decree of February 1892, dhow traffic had been at a virtual
standstill. The hiatus in commercial affairs hastened a reappraisal
by the principal family firms of banker-brokers.
A prime example of withdrawal followed upon the death of
Taria Topan in late 1891. Settlement of such a complex estate
called for examination of many business relationships. Almost
immediately, Tippu Tip was sued for the balance of his credit
account. The outcome was favourable to Tippu, however, for he
had delivered ivory worth substantially more than the goods
advanced to him. Rumaliza dropped his attitude of resistance to
the Germans, claimed his rights as a German subject, and sued
Tippu Tip for a share of this wealth. The court in Dar es Salaam
found that Rumaliza had been a partner of Tippu Tip and awarded
him property owned by Tippu in German East Africa. The
sequence of events thus illustrates how the new colonial situation
turned the balance of relationships, and it underscores the primacy
of economic interests in fixing attitudes towards colonial regimes.
For the Zanzibar-based ivory traders in Manyema, there was
no possibility of surviving the Congo Independent State's military
campaign against the 'Arabs'. Tippu had written to Rumaliza in
1893 urging him to retire peaceably, but at that time Tippu was
enjoying his estate in Zanzibar and Rumaliza was still at Lake
Tanganyika with a large number of dependent retainers and
followers who had to be found alternative employment.
The old complementarities had become contradictions. The
desire of traders to disentangle their assets from those of chiefs
led to various actions of local consequence in the early and
569

mid-1890s. In the south-west, contracts were promoted by the

drawing of an international boundary. On the German side,
favoured by an administration which supported commercial
claims, the Swahili creditors were able to impoverish the new chief
of the Fipa, who resorted to force to obtain increased tribute from
his subjects. Beyond the frontier toward Lake Mweru, Chief
Chungu was able to stall in making good his debts and kept the
local trader virtually an economic prisoner. Nevertheless Chungu,
like many chiefs, faced economic erosion and a shrinkage of
control over people and resources. By the turn of the century, he
was willing to let several of his discontented wives go off with
other men, receiving a compensation in a quantity of' good cloth'.
Conflicts emerged in Zanzibar itself. The sultans since Sayyid
Sa'Id's time had been among Zanzibar's greatest merchants, and
Bargash and Taria had been prime movers in the licit accom-
modation. Sayyid Sa'Id's successors necessarily reflected the
dissolution of the Swahili system, the sultan being accused, for
example, of diverting to Arabs the ivory that was owed to Indian
firms. In 1893, the personal access of the sultans to the revenues
of the state ended when a prince from Muscat was brought in by
the British and placed on an annual salary of $120,000.
After the islands had come under closer British administration
following the declaration of a protectorate in 1890, officials who
had been trained through experience in Egypt began to design
policies whereby Zanzibar could live upon its own agricultural
production. The clove industry clearly had to be supported, and
to this end an order was issued in 1891 prohibiting the recruitment
of labour in Zanzibar for work outside the islands. Consternation
reigned, for many professional porters who would not have been
clove pickers had considered Zanzibar a home resting place
between journeys and knew well that wages offered for trans-
porters enlisted in Zanzibar were higher than on the coast.
Although the Germans were scrupulous in seeing that slaves were
not forced against their will to move back or forth across the
channel, they favoured the free movement of workers and sought
to preserve, not balkanise, the labour market of East Africa. The
Zanzibar controls ultimately had no major effect on the mainland
supply of labour; by the later 1890s, a new labour system for the
clove industry took shape, in which the ex-slaves resident on the
estates had the privilege both of secure occupancy and disposal
5 7°

of their labour to the highest bidder. Because the pay was

seasonally very good, better than anywhere else in East Africa,
Zanzibar and Pemba once again, by economic suction, drew in
migrants.
The Anglo-German Treaty of 1890 had generally settled the
colonial jurisdiction of the two major powers in East Africa, yet
there was still much fluidity within the British sphere. Momen-
tarily, the northern interior fell entirely under the Imperial British
East Africa Company (IBEA). Much has been made of the
difference between company and formal colonial rule. From the
standpoint of both African perceptions and political economy,
however, the colonial situation unfolded in broadly similar phases,
regardless of the auspices. The first may be called the era of armed
caravans, whose relations with local people were similar to those
of large Swahili expeditions. Thereafter came the establishment
of strong-points, with small but significant arsenals and concen-
trations of uniformed soldiers and auxiliaries, often gaining
strength through alliance with a pre-existing regional power.
From these garrisons, colonial officers engaged in alliances and
feuds, often carving out districts and ruling in a highly personalised
way. An official like John Ainsworth at Machakos in Ukambani
was in effect a colonial seigneur.
Personal rule of all sorts diminished when, at the end of the
1890s, the major indigenous military powers in East Africa had
been neutralised through alliance or campaigns of conquest. The
ending of company rule, of the German East Africa Company in
1890, of the IBEA in Uganda in 1893 and the IBEA in what
became the East African Protectorate or British East Africa
(Kenya) in 1895 occasioned discussions of the ultimate shape of
the colonial state. In the event, however, expediency prevailed.
Nevertheless, direct ties with the metropolitan state did have
important consequences for the availability of capital and the
overall capacity to bring about infrastructural changes, above all
through railway construction.
The Uganda railway, which eventually defined the political
economy of colonial Kenya, was authorised on the basis of
entirely external, strategic considerations. From its start in
Mombasa in 1895 to its completion at Kisumu in 1901, it was the
work principally of imported Indian contract labourers. There
were no planned and specified economic goals. Yet the progress

EAST AFRICA, 1870—1905
of the construction had continuous commercial implications,

above all generating boom markets for produce which encouraged
local agricultural peoples to grow and market a surplus. The line
of rail was also a central nerve governing military reflexes and
conveying a logic of armed conquest.
MILITARY CONQUEST IN A CONTEXT OF ECOLOGICAL

CRISIS
The founders of colonial regimes recognised the utility of existing

military formations and exploited them and the antipathies that
earlier warfare had inculcated: In certain circumstances, therefore,
the military conquest was 'soldier to soldier'. In other cases,
however, the context of military conquest was shaped by con-
ditions profoundly modified by ecological crisis. Locust, rinder-
pest, drought, smallpox, sleeping sickness, and other diseases and
pests invaded the countryside and compounded the turmoil
created by colonial military and political activities.
The most intensive military campaigns in East Africa took place
between 1893 and 1898. Of one type were the wars against the
'Arabs', rising to a peak in 1895 but sputtering on thereafter as
well. Rearguard actions were undertaken by the Swahili in the face
of colonial military operations in the Congo Independent State
and British Central Africa. Also on the coast of British East Africa,
the colonial power made a great victory out of thefinalsuppression
of Mbarak's Mazrui and saw the elimination of Arab resistance
as the turning point in the history of the colony. The turmoil of
early colonial conditions elsewhere on the coast could be seen in
the sustained resistance of the sultan of Witu and the challenge
of Hassan b. Omari in the hinterland of Kilwa. Each occasioned
a strong but transitory military presence before the ligaments of
Swahili commerce were allowed to reassert themselves as the
principal organising factor in these zones of comparative in-
difference to colonial strategists. In this perspective the conquest
of the Mazrui can be seen not as an 'Arab war' but as one of the
early examples of the piecemeal conquest of British East Africa
aimed at clearing the line-of-rail from Mombasa to Kisumu.
The logic of regional power struggles of the 1870s and 1880s
carried over into the strategic thinking of the Germans in the
southern highlands of their colony and the British in the conquest
5 7*

MILITARY CONQUEST AND ECOLOGICAL CRISIS
30 ,7 ITALIAN
\SOMALI-
• \ LAND
CONGO
IDEPENDI
INDEPENDENT A/UGANDA EAST AFRICA
STATE < PROTECTORAT
PROTECTORATE
GERMAN
EAST AFRICA
Colonial boundaries 1905

Uganda railway (completed 1901)
Northern railway German East
Africa (progress to 1905)
3 0 0 km
, NORTH-EASTERN .Q
.'' RHODESIA
! <
I I
PORTUGUESE
EAST AFRICA
23 East Africa, 1905
of Uganda. The Hehe wars (1893-8) and the war against Bunyoro
(1894—8) had marked similarities; major rulers continued in
opposition as guerrillas after the enemy occupied their capitals,
and the colonial forces included an ensemble of pre-colonial rivals
deployed by European dictate but often with their own officers.
The guerrilla phase was combated with scorched earth policies,
using famine as a means of subjugation. Finally, in both cases allies
were rewarded with territory at the expense of the resistant and
defeated state.
Ignorance of the Hehe and of the recent history of the southern
highlands prevailed among the Germans until one of their
expeditions was mounted towards the Iringa plateau and was
demolished. A quick appraisal of the regional balance of power
573

soon led the way to a strategy of encirclement; the Ngoni were
treated with diplomatic care in order to postpone any test of
military strength, while other hostile or displaced rulers were
cultivated actively. The most positive alliance was contracted with
the Sangu, who supplied the principal auxiliaries in an advance
from the west. The prearranged reward was the restoration of
Chief Merere as the paramount African ruler in Usangu and parts
of Ubena and Uhehe. Meanwhile, within central Uhehe, the
colonial authorities identified councillors who would accept the
military administrator as the supreme authority, thus building a
secure core while dealing with the continuing resistance in the
countryside. A wide region became mobilised through requisitions
for food. In Mahenge, for example, this demand set in train a
process of local differentiation and commerce that continued and
became compounded by colonial tactics during the Maji Maji war.
Altogether, the war against the Hehe amounted to a war for
regional control.
In Buganda at the end of the religious-civil wars, commanders
with large followings, such as Samei Kakunguru, turned from
domestic struggle to colonial warfare. F. D. Lugard, the officer
in charge, had signed on some remnants of the former Egyptian
forces in Equatoria, but the Ganda so dominated the colonial
forces that opponents sometimes assumed that the British were
merely mercenaries assisting in Buganda's further expansion. The
principal colonial war in Uganda was a long campaign to subdue
Bunyoro. Little room was left for diplomacy, as Lugard was
determined to punish Kabarega for attacks upon Toro. Strategic
outposts overlooking the salt works at Katwe and focal points
of trade provided a cordon intended to seal off Bunyoro, especially
from supplies of firearms. Surrounded but not helpless, Kabarega
gave sustained resistance from 1894 to 1899.
On the British side, Kakunguru was rewarded with a post as
a semi-autonomous administrator, at first in an occupied part of
Bunyoro, then in a large territory in the east centred on Mbale,
near Mount Masaba (Elgon). Ganda contingents were not only
allowed to colonise new areas in this way in the 1890s, they also
took part in the extension of missionary occupation as armed
retainers and local traders at stations as far away as Rwanda.
Religious affiliation continued to be vitally important in deter-
mining access to power, employment and resources. The Muslims,
574

who on the whole were losers in the competition, formed a

militant group among Kabaka Mwanga's following. Mwanga was
politically disregarded from 1893 onward, and he displayed his
hostility to the British in a number of ways. These protests
culminated with a fracas at Buddu in December 1897, and an active
rebellion a year later, when Mwanga and many Muslim Ganda
joined Kabarega for a last stand. Among the Sudanese units in
the British military service in Uganda, there had always been
awareness of Islam as a common denominator. Some Sudanese
had joined with Muslim Ganda in the crisis of 1893, when the
Muslims tried to force a more favourable distribution of land and
offices. Yet the Sudanese were professional soldiers, not religious
zealots. Their mutiny in 1897 and 1898, creating an emergency
for the weak Uganda administration, was brought on by un-
remitting campaigns, long marches, and erratic pay. The mutiny
complicated and delayed the conquest of Bunyoro. At the final
surrender of Kabarega and Mwanga in April 1899, the colonial
forces were composed largely of Ganda irregulars, some Sudanese
and Swahili regular units, and a new component, Indian troops
who were rushed to Uganda to patch the frayed cloak of imperial
control.
After 1898, the incorporation of Uhehe and Bunyoro took place
in the absence of a reigning paramount chief. That an interregnum
or exile of rulers was expedient for colonial consolidation is well
illustrated throughout eastern and southern Africa. In the Zulu
and Ganda cases, most notably, the absence of a supreme leader
allowed for the encouragement of oligarchies, sometimes raised
from subordinate groups, sometimes reasserting anti-monarchical
ambitions. In Uhehe and Bunyoro, the totality of defeat led to
passivity in economic and political life and it was only after the
First World War that sharp irridentism surfaced with the
restoration of a modified monarchy and in the ideological climate
of 'indirect rule'.
When turning to the ecological crisis as a context of militarism,
it must be noted that natural disaster is a constant theme in the
history of East Africa from 1870 to 1905. Cholera and smallpox
had been alarming events just prior to 1870; drought struck
periodically, with perhaps magnified consequences in areas of
active commercial penetration and indebtedness, and cattle
diseases (such as bovine pleuro-pneumonia in about 1880 and
575

rinderpest in several waves beginning in 1887) decimated herds,

dispersed populations, caused compensatory raids and encouraged
cultivation not only for subsistence but also as a means to generate
wealth to exchange for livestock. Yet the physical assault on the
East African economy and society as it occurred in the 1890s
formed part of a devastating conjuncture. Some diseases were
new; many were endemic but had been encouraged by population
movement and concentration in pursuit of one or another
anti-famine tactic. Colonial manoeuvres and commercial traffic
contributed further to these patterns of concentration and
mobility.
Smallpox was the more deadly and became more epidemic as
famine called for the transfers of people and grain from place to
place. Mortality of victims was higher among the malnourished.
Estimates for the Kikuyu range from 50 per cent to 95 per cent
population loss from smallpox and, even if these figures describe
only some of the more closely observed and densely populated
areas, the tragedy was nonetheless great. One of the two con-
sequences of smallpox was a reduction of labour power at a time
of expanded demand and commercial opportunity. Another was
a compounding of ambiguity about the advantages of the colonial
presence, for vaccination carried out by missionaries and officials
helped to stem the spread of the disease and reinforced a message
about the potency of the foreigners as therapists as well as
militarists.
Cattle-keepers were devastated by rinderpest in particular.
While certain insulated zones were spared, the disease spread
rapidly along the commercial networks of the region, killing up
to 95 per cent of the herds. Depending upon the role of cattle in
the diet and the maintenance of social relations, the consequences
were various. The immediate effect was levelling, in that those
portions of society whose subsistence was tied to cattle were
compelled to starve if they could not shift their source of
nutrition. Such was the case with some Masai, whose situation
was desperate. In many situations, however, the emergency was
met by forgoing milk and blood and surviving on cereals and
other vegetables. The Tutsi became more dependent upon the
Iru, Masai upon their Kikuyu or other farming affiliates, and so
forth. Relationships symbolised by the exchange of cattle were
potentially open to redefinition. Some examples of how cattle
576

losses affected societies can be drawn from the experience of the

Langi of Uganda, the kingdom of Rwanda, the Masai, and the
Jie-
The rinderpest epidemic intervened in telling ways in the
political life of both Lango and Rwanda. In Lango, the regional
leadership bred by the military conditions of the previous decades
had perpetuated itself through ownership of cattle and manage-
ment of unusually large and productive communities. The colonial
war in Bunyoro disrupted regional trade, and rinderpest ex-
tinguished assets in cattle, to the point where the economic focal
points disintegrated and more localised ties were reasserted. In
Rwanda, on the other hand, the sudden loss of cattle herds led
to an exercise of royal power to lay claim to all the surviving cattle.
The terms of clientship governing their redistribution, ubuhake,
specified direct obligation to the king himself. Intermediate
allegiances that had been secured by the alternative umuheto
clientship were depreciated. So rinderpest hastened the decom-
position of regional leadership in the decentralised society of
Lango and furthered centralisation in the kingdom of Rwanda.
The impoverishment of the Masai increased their propensity to
use violent means to provide for daily needs. In German East
Africa, where the government had established a footing among
agricultural peoples, the Masai were cast as enemies, while in
British East Africa they were vital allies in a tenuous colonial
occupation. As a result of the clustering of Masai around the
outpost forts of the Imperial British East Africa Company, they
were available to provide a ready source of military recruitment
in the campaigns against the Kikuyu, which produced a harvest
of enduring enmity between the two peoples that has tended to
obscure the extent of more peaceful social and economic exchanges
in a preceding era. Punitive raids by the British—Masai alliance
were paralleled by Kamba brigandage which became widespread
as a scramble for cattle was prosecuted by every means available
within a relatively weak and conflicted political situation.
Not all societies enduring cattle diseases were at the same time
contending directly with colonial intervention. The Jie, in an as
yet unclaimed part of eastern Uganda, lost virtually all their
herds to rinderpest and another disease tentatively identified as
blackwater. Thereupon, they altered their attitude towards Swahili
ivory traders, whom they had merely tolerated earlier, and became
577

willingly engaged in both transport and hunting. As an induce-

ment, the traders brought in cattle from Turkana country to
exchange for ivory and services. In the process, a new orientation
of Swahili activity was strengthened, as caravans moved from the
Somali (Benadir) coast through the ambiguous frontier zone
between Menelik's Ethiopia and the East African Protectorate. As
the Kamba had lost control of the ivory trade in the decades before
1870 in the Mombasa hinterland, so the Somali in the 1890s ceded
their monopoly of trade between Brava and other ports on the
coast and the Boran country in the interior. The retraction of
Swahili caravan and ivory trade in the areas of strong German
and British occupation in the Tanga and Mombasa hinterlands
seems to have coincided with the new initiative on the northern
peripheries. The ivory that had once been carried out via
Kilimanjaro emerged instead at the 'Italian' ports, and north-
eastern Uganda saw unprecedented commercial activity.
High prices for cattle resulted from their shortage. At the turn
of the century, colonial punitive expeditions carried off booty in
cattle and furthered their redistribution by paying off African
leaders who provided support and also by facilitating a transfer
of stock from African to European settler control. Captured stock
from the Nyanza zone were delivered to the Nairobi cattle market,
where the administration auctioned them to settlers. In a truly
long-distance trade in cattle, agents from Southern Rhodesia
bought cattle in Burundi which they herded south through Ufipa,
across the Nyasa-Tanganyika corridor, down Nyasaland and into
Rhodesia, to be sold in Salisbury (Harare).
Locust and drought, as opposed to epidemic diseases of humans
and animals, did not follow the beaten path of commerce. They
could strike anywhere and did so over extensive areas of East
Africa. Drought of course affected pastoral activities as well as
arable agriculture, while locusts attacked grain crops most
disastrously. Locusts, therefore, struck at cultivators who had
various reasons to be anxious about the condition of their
economy. These very cultivators had sometimes suffered a loss of
labour power through disease, or were under pressure to sell their
surpluses to famine areas or colonial posts, and so on. Aside from
raids and other coercive exactions, there was also a market and
motive to acquire livestock which prompted expanded cultivation.
Locusts deflated the efforts of a whole agricultural cycle in certain
years, but did not have the generalised consequences that came
578

SUB-IMPERIALISM AND COLONIAL HIERARCHIES
from drought prolonged over several years. The 1890s were on

the whole dry. The climax in semi-arid areas of Kenya came from
1898 to 1901, when people in desperate conditions welcomed all
forms of relief. In parts of coastal Kenya where transport of food
could be organised by the colonial government, it was given out
in return for allegiance to missionaries or to those willing to
perform public works, thus advancing the influence and instal-
lation of essential colonial elements.
Conquest amounted to an interplay of military expeditions,
political alliance and contracting for auxiliaries, domestic social
and economic conditions, and crisis-stimulated regional activities
that differed from time to time and from locality to locality. While
interpretations of conquest must entertain all these complexities,
the rendering of the history of conquest within a single area may
depend further upon the social group being held in focus. The
situation on the southern side of Kikuyuland was grave in 1893-5
when cattleless and famine-stricken Masai joined forces with the
British. These incursions, painful though they were, took place
in a time of prosperity and resilience when compared with
1898—1900, when the crisis of famine and depopulation among the
Kikuyu was most severe and armed bands spread terror. The
colonial forces of the time were inadequate to impose order. It
was only after the turn of the century that the police and military
organisations of the East Africa Protectorate became reorganised,
delivered the initiative to the colonial state and enforced systematic
administration and tax collection.
The period from 1888 to 1899 has frequently been discussed
in terms of collaboration and resistance. Resisters were the heroes
of later nationalist writers, while collaborators had to be carefully
explained if not condemned. The era is now coming into its own
as a fuller historical drama, fraught with contradictions within
African societies, overshadowed by economic disaster and filled
with minute calculations of advantage as well as gross adjustments
in balances of power.
At the turn of the century, with pacification nearly complete in

the primary areas of colonial occupation, questions of an institu-
tional nature came to the fore. The necessity of political hierarchy,
a reliable chain of command and effective management of
5 79

productive economic activities was clear to every government,

but the exact outcome of the search for a stable and productive
social order was not entirely predictable. Upon the conclusion of
the 1900 Buganda Agreement, an oligarchy in Buganda, composed
mainly of Christians, with Protestants in the strongest position,
achieved one of the most impressive settlements in the annals of
sub-imperialism. This agreement guaranteed that the Bakungu
chiefs would be the principal class mediating between the colonial
power and the common people. A legislative capacity was given
to a council of chiefs, the lukiko, and the principle of freehold land
tenure was conceded as a veritable endowment of the political
oligarchy. The Ganda continued to be regarded as skilful
administrators to be employed in extending the Buganda model
of county and sub-county government to other parts of the
protectorate as they were brought under political control. The
sub-imperialism emerging in Uganda, then, was an outgrowth of
alliances made in the conquest period, which were subsequently
entrenched. This entrenchment of privilege — through recognition
of the Ganda political agency, landed oligarchy, legislative
powers, and access to state-related wage employment - serves as
a prototype against which the experience of other African and
non-African groups within early colonial social hierarchies can be
measured.
Harry Johnston had not expected to institute a landed oligarchy
in Buganda when he became Special Commissioner in 1899. The
practical situation, together with the strong position sustained by
the elite and their official and missionary advocates, obliged him
to retreat from his determination to vest land in the British Crown.
As early as 1893, British officials had noted a great confusion over
land ownership, with buying and selling in some places and
inalienable clan lands in other places. Estates were tied to offices
and certain major chiefs held them in more than one county.
Lastly, the religious compromise ending the civil war had tended
to parcel out counties as quasi-preserves to each of the contending
denominations.
The 1900 Agreement created uniform freehold rights, called
mailo after the square miles in which estates were to be measured.
It was expected that the claimants would be the 1,000 most
important chiefs. The number soon exceeded 1,000, however, and
580

land ownership became coveted as the seal of enfranchisement in

the political oligarchy. When cotton and coffee spread as cash
crops after 1905, land-owners mobilised their tenants as primary
producers. Before 1904, however, the chiefs were mainly useful
for their ability to command people to move or to perform public
works. Shortly after 1900 the exigencies of the disastrous sleeping-
sickness epidemic and the campaigns of control through
evacuation of the tsetse-infested lake shore and islands were
handled expeditiously by the bakungu chiefs.
Outside Buganda, the foundations of the British—Ganda alliance
continued to rest on expediency. Jealousy on the part of some
British officials and divergence of goals affected the durability of
arrangements. Kakunguru could be charged with seeking to make
himself kabaka of Bukedi. In 1906 his authority was curbed by
reassignment to Busoga. James Miti, the key administrator in
Bunyoro, performed in a more suitably disciplined manner, and
survived the 1907 Nyoro rebellion against 'foreign' Ganda
officials. The Nyoro drive for home rule and oligarchic privileges
was thus temporarily thwarted and its ringleaders were deported.
In more subordinate administrative capacities, too, the Ganda
agency persisted, with the networks of Protestant and Catholic
schools in Buganda constantly renewing the supply. Formidable
figures like Apolo Kagwa, the katikiro and a regent of Buganda,
continued in non-official ways to extend and reinforce the
outward movement, not least by supporting evangelisation
through the (Anglican) Church of Uganda. Luganda thus spread
as the language both of the administration and of the church. The
networks of Protestant and Catholic schools in Buganda constantly
renewed the ranks of proselytisers and political agents.
The Swahili agency in German East Africa followed a somewhat

different trajectory. As in Buganda, so on the Mrima, the colonial
occupiers found a situation of turmoil and faction. Hermann von
Wissmann, as commander and imperial commissioner, built his
armed forces around a premise that African Muslims were the
necessary cadre. Arabs were to be eliminated. Emerging from the
Abushiri war, the German administration concentrated its efforts
on establishing social policy and the rules for land tenancy along
the coast. While the interior went through a period of military
581

administration, the colonial state gathered itself as a bureaucracy
on the coast, acting on the whole conservatively, confirming
freehold titles to land and adopting a very gradualist approach to
slavery and emancipation.
The German officials who took over from the chartered
company, wishing to perpetuate the infrastructure and com-
mercial life of the Swahili system, were not quite so anti-Arab as
Wissmann had been. They did, however, build their alliance with
the Swahili, men of small if any property. The principal means
of promoting this relationship was a secular state school system
which extended from the first and ultimately the highest school
at Tanga, to the other main coastal towns where urban central
schools were fed by district primary schools. By 1902, the
governor was ready to carry the state school system into the
interior both for positive bureaucratic reasons and also to
challenge the Christian missionaries to conform in curricula and
supply candidates for the civil service. The state schools,
missionaries alleged, retained a Muslim bias.
The fostering of Swahili as the language of schools and of
colonial administration had much to do with the integrated
government that became more and more evident as military
districts came under civil rule, garrisons were reduced or replaced
by police, and the sons of local notables were called upon to attend
school and act as clerks in the grass-roots levels of local
government. Even if the Germans in 1902 began to hasten this
process of political localisation and homogenisation, they
retained as strong elements of direct rule through akidas, district
officers and tax collectors who very often had served in the
colonial army (Schut^truppe) or were trained in the state school
system. Abrasion between the African elite and local peoples
was as natural in German East Africa as it was in Bunyoro or
elsewhere in Uganda, where hostility to sub-imperial agents
surfaced during the early years of this century.
The Germans and the Swahili established an alliance in state
service which carried over into commercial life. District officers
at Pangani, once a seat of resistance, became actively involved in
preserving the commercial networks which sustained the port.
Nyamwezi and Swahili traders in Tabora also received the full
backing of the German authorities as they pursued their business,
and German firms employed many Swahili agents. As the church
582

proved to be a complement to state employment for the Ganda,

so for the Swahili commerce continued alongside a much enlarged
employment in government, and formed part of a constellation
of allied interests in the colonial situation.
The discipline of the caravan porters from Unyamwezi and
Usukuma was an invaluable human resource which was appre-
ciated and rewarded in German East Africa. When the early
settlers encountered resistance to their labour demands, in
Shambaai, missionaries advised a system of registration and
obligatory wage labour which was installed at Lushoto as a unique
attempt at labour control for the benefit of private interests.
Slavery continued to be legal, but slaves could go to court, buy
their freedom, or be awarded it because of proven abuse. To be
legal, slaves transferred from owner to owner had to agree to the
change before an official. High tolerance for the retention of
captives as dependents and servants of askari in the Schut^truppe
added to the attractions of military service. Nyamwezi as well as
Swahili were recruited into this elite well-paid armed force. The
Nyamwezi were everywhere sought as workers, at the highest
wages offered, and nowhere more than in the Pangani valley where
they formed a significant part of the labour force constructing the
first German-built railway, and establishing sisal estates. They
were permanently settled on land of their own and exempted from
the ticket system prevailing in Lushoto, which obliged Shambaa
men to work for white settlers. It was as an aristocracy of labour,
rather than as a landed oligarchy, that the Nyamwezi and Swahili
together prospered under colonialism.
In the Zanzibar Protectorate, the diminished but surviving
sultanate provided the core of local government. The Omanis
retained the highest positions, and plantation owners continued
to be closely linked in interests. For the British administrators,
the fiscal and economic interests of the state obviously required
preservation of the clove industry and its labour force, but they
were caught by the anti-slavery rhetoric of high imperialism and
received a parliamentary order in 1895 to institute immediate
emancipation. A host of countervailing measures and patterns
resulted. Slaves were freed, but sharp vagrancy laws followed
soon thereafter. By the first decade of the twentieth century the
former slaves were largely stabilised as statutory tenants on
plantations, free to sell their labour during harvest time to the
583

highest bidder. The compromise in Zanzibar thus assured the

preservation of the clove-planter oligarchy and a comparatively
benign labour system. A social and political consequence of this
reformed labour system was a continuation of under-class attitudes
of dependency and low expectations. Education remained the
preserve of Shirazi, a high-status Swahili group, and the Arabs,
and with education came access to the civil service. The state in
Zanzibar thus confirmed economic and political privilege among
the old upper orders.
The Indian communities had been a key part of the Swahili
commercial system. They continued to operate as expediters for
colonial supplies and, like traders of other communities, withdrew
from unprofitable activities and turned to new areas of oppor-
tunity. This commercial behaviour on the part of Indians, which
led to their increasing settlement throughout the interior, was
taken for granted by the British and German officials. The British
recruited new Indian contingents: the thousands of coolie
labourers who were contracted to work on the Uganda Railway
construction, the companies of Indian Army soldiers needed to
bring colonial forces up to strength, and lastly, the individuals
taken into the swelling ranks of clerks and artisans employed by
the East African Protectorate. Whatever other advantages Indians
enjoyed, they were never allowed to become major land-holders
at this time. Since their effective elimination from ownership of
clove estates in the 1870s, Indians failed to make headway in land
ownership beyond commercial plots and buildings and thus failed
to consolidate their sub-imperialism. When the railway was
completed, most coolie labour was repatriated. The King's
African Rifles became organised at the turn of the century and
gradually replaced the Indian forces, and the East African
Protectorate seemed bent upon supporting a white-settler
oligarchy on the land and in the advisory councils of the central
bureaucracy.
The two other allies in the initial colonisation of the East
African Protectorate were either unqualified or rapidly superseded
in the incipient colonial state. The Masai military alliance did
not persist through the entire conquest period: the economic
objectives and livelihood of pastoralists inhibited both sides from
regular employment or expectations of formal education. The
Masai made a treaty in 1904 agreeing to withdraw from the Rift
584

TOWARDS A COLONIAL ECONOMY
Valley railway zone and live within two reserves to the north and
south of the line, connected by a corridor. Such a land settlement,
whether or not based in a treaty, proved to be more easily alterable
than the freehold rights acquired in Buganda or confirmed in the
old Zanzibari domains. The Swahili and Arab communities of the
coastal strip finally became circumscribed as sub-imperial agents
in 1907 when Mombasa was supplanted as the capital by Nairobi.
Employment in government at Mombasa had in any event been
shared with Goans and Indians, and Swahili recruitment into the
other ranks of the state was not systematically fostered by
educational institutions as in German East Africa. The small,
boisterous community of Englishmen and South Africans, which
formed the nucleus of the future polity and protected economy,
totalled in 1904 only a few hundred persons. Nevertheless, as a
consequence of conquest, the ravages of disease, the policy of
evacuation and the diffuse character of African political culture,
this white settler group was able to lay claim to a special
sub-imperial position as Kenya's landed oligarchy.
The economy of East Africa at the turn of the century was largely
disorganised, owing to the effects of campaigns of conquest,
resurgent rinderpest, drought between 1898 and 1902, the spread
of sleeping sickness and uncertainty about the viability of capitalist
agriculture either on plantations or through peasant production.
Established plantation-grown commodities, such as copra and
cloves, continued to be substantial contributors to export earnings.
Hopes of a new generation of labour-producing plantations had
been premature in the 1890s, when poor international prices for
cotton, jute and other fibres retarded marketing and investment.
By the time conditions improved, the Kenya coast had become
a zone of economic indifference in British East Africa. On the
coast of German East Africa, on the other hand, efforts at
economic development remained vigorous, with private capital
moving into sisal and the state promoting cotton through peasant
production.
The sector of the East African economy that flourished was
commerce, led by the Indians, facilitated by the railway and
steamer service which fundamentally altered the cost factor of
585

transport in the Lake Victoria basin, and sustained in other areas
by further sharp increases in rubber demand and price. Once
again, in a buoyant spirit, credit became a prime means of
engaging middlemen and producers. In the production and trade
of vine-grown rubber, furthermore, many of the commercial
elements of the old Swahili system were recombined. The very
areas of East Africa with an abundance of vine were those of long
and intense commercial contact. In the coast-hinterlands of
Malindi and Lamu in the north and Kilwa in the south, the rubber
boom gave an impetus which renewed relations based on clientage,
particularly in the north, and credit in the south.
The commercial revival had much to do with the overall
improvement of commodity prices, which in 1900 stood at a high
point. In the case of German East Africa, the rapidity of response
was great, because many German firms had adapted to the
commercial environment so moulded by the Swahili system. Not
only Hansing and Company and O'Swald, but also the DOAG
(the reformed and more purely commercial German East Africa
Company), engaged extensively in credit-giving to networks of
commercial operatives. Leading Nyanyembe of Tabora, striving
to maintain trade with the south, stood in this relationship to the
DOAG.
By 1900 vine-grown rubber - harvested and sorted by Africans
in the same way as wax and honey, copal, and other commodities
derived from the bush and forest - had become the leading export
of German East Africa and a prominent one for British East Africa
and Uganda. Some of the highest in quality came from the
Mahenge district, where firms based in Bagamoyo and Kilwa
competed, and from an area more exclusively in the Kilwa sphere,
Ndonde. The label ' Donde', given to the rubber from these two
source areas, was consistently high in quality and price.
In Mahenge, the newest district of the rubber frontier, the
Swahili from Kilwa and the Indians from Bagamoyo, together
with their affiliated local traders, rose in number from 200 in 1902
to between 500 and 700 in 1905. The consequence of this feverish
promotion of rubber was that contact men got rich while
producers resented what seemed to be a lowering of prices, but
what was in effect a very high interest on cash advanced at tax-time
against future delivery of rubber. Disparities in wealth were less
evident in the Ndonde area, but sensitivity over prices and terms
586

of trade nevertheless became sharpened through fluctuations in

the conditions of trade. There had been a commercial panic over
rubber in 1902 and early 1903, when the world price dropped,
drought badly reduced production, and firms began to feel that
credit had been over-extended. In Ndonde, the local officials saw
that popular resentment grew as producers were squeezed to meet
their debts, and they endeavoured to centralise marketing to
assure that producers knew and received the full going price.
More normal relations and terms prevailed as international prices
improved; in 1904 they were 70 per cent higher than in 1902.
While the men of Ndonde were receiving more advances of cash
and cloth for their rubber in 1904, their women were participating
in cotton experiments. Reports on the quality were glowing; a
Hamburg broker crowed that Donde cotton would soon prevail
in the market, at the expense of American cotton. Communal
coercion had not been instituted in Ndonde country, as it had been
in the Rufiji district, where peasant aggravation ignited the Maji
Maji war. The Ngindo of Ndonde did, however, join immediately
in the war and attacked the traders and officials at the district
headquarters of Liwale.
The entire southern region of German East Africa was in
rebellion by the end of 1905. After two successive years without
any return for their labour on communal cotton fields, the Rufiji
people ignited the Maji Maji war by taking action against the
enforcers of the scheme. The call for resistance was spread by
messengers of a prophet, Kinjikitili, who distributed medicine
which, he promised, would turn bullets into water. This message
and medicine were successful in mobilising diverse elements, from
the decentralised peoples of the east to the hierarchical Ngoni of
Songea. Many factors contributed to receptivity and simultaneity,
among them peasant fury at exploitation, charismatic leadership,
overbearing administration, cleavages within Ngoni society, and
the unequal rewards accompanying commercial penetration. Two
factors in its spread deserve special attention. The widespread
drought of 1902 had sent supplicants in increasing numbers to the
Kolelo cult centre when Kinjikitile officiated. The same drought
conditions diminished the harvest of vine rubber and coincided
with a drop in the rubber price to create tension among creditors
and debtors. A link-up of cultivators' anxieties and rubber-
producers' sense of exploitation was not direct, but the atmosphere
587

of uncertainty was pervasive. In 1904, rubber was once again

booming. The outrage of the cotton-growers must have been
deepened by the contradiction between the colonial demand that
they should bear all the costs of production while their neighbours
received payment for rubber in advance. The levels of con-
sciousness in the Maji Maji outbreak area must be seen in terms
of the culture of credit.
In 1905, when the Uganda railway first returned a profit, neither
military transport nor production managed by white settlers
figured very importantly. It was the carriage of goods destined
for African consumption and the export of African-produced
commodities that provided the primary business. Cultivators and
pastoralists were engaged in new ways once the cost of transport
had been radically reduced for bulky and lower-priced products.
In 1904, it had become profitable to export groundnuts from the
Lake Victoria basin, where the market price was so far below that
at the coast that the cost of transport was easily absorbed. Exports
from Lake Victoria at that time reflected ongoing production of
food crops and oil-seeds like sesame, pastoral products such as
hides and skins, and ' natural' gathered products such as sansevieria
fibres and vine rubber. Ivory remained a valuable export of
Uganda. While it cannot be said that the manner of producing
these exports had been altered, the means of inducing production
were more co-ordinated. The goad of tax collection pushed people
to market their surplus, and the railway and steamer services
ensured that commodities would move. Two ' railway' steamers
called regularly at major ports around the lake, and metal barges
were supplied to expedite loading and unloading.
At the forefront of the commercial wave were the Indians. The
firm of Alidina Visram outstripped all others, operated all around
the lake and maintained its own fleet of eight cargo dhows. There
was also a cosmopolitan community of western firms, the Italian,
American and German ones concentrating in Uganda and to a
lesser extent in the southern and western 'German' ports. The
German East Africa Company was the dominant western firm at
Mwanza, yet its gross receipts and volume of business did not
match Visram's in the same place. In some respects, therefore, the
scene was reminiscent of that which had prevailed decades earlier
at Zanzibar. Under formal colonialism, however, the adminis-
588

CONCLUSION
tration could tilt the commercial balance by co-operating with

certain designated nationals in the development of new commodity
production and by regulating trade through the granting of
special concessions. The Uganda administration in 1904 gave
rubber-marketing rights to a consortium of Italians, Americans
and Germans, and placed cotton in the hands of a British
organisation.
CONCLUSION
In reviewing the eventful thirty-five years that brought East

Africa to the threshold of its modern colonial and post-colonial
history, the important economic distinction lies in the pre-
dominance of commercial rather than industrial capital as the local
means of generating products demanded by a wider world.
Rubber epitomises the period. Demand for it was promoted by
western industrial consumption and its production in East Africa
took place through intensified commerce, engaging in established
practices of credit-giving to induce production and use of caravan
transport networks and semi-autonomous agents. The rubber
plantations, which came slowly to maturity in the first decade of
the twentieth century, represented on the other hand a different
kind of capital, a new mode of production, with a more powerful
claim on the colonial state. In the plantation sector, sisal in
German East Africa had proven to be a viable forerunner of
industrial agriculture. Dimly, there was an outline of the ensuing
colonial pattern of regional specialisation, with its characteristic
zones of atrophy and growth. The most dramatic shift in the
economic organisation of the region before 1905 resulted from
the completion of the Uganda railway. This revolutionary infra-
structural change, together with an attendant commercial
penetration in the Lake Victoria basin, contributed to the
preconditions for successful peasant production of such com-
modities as cotton.
Taxation has often been seen as a major means of generating
a colonial labour force, and while it doubtless played a role, a
closer analysis of labour in the pre-industrial era indicates that
economic participation by young men, the target population of
labour recruiters, had been defined and redefined in the 1870s and
1880s by the exigencies of military and porterage service. Many
589

other factors, of customary mobility during dry seasons, scarcity
of labour owing to depopulation during the ecological crisis, the
strenuous efforts to retrieve the base of subsistence and many
subtle changes in regional economies, must be taken into account
in evaluating readiness to work for wages. Taxation, rather than
reaching the potential worker, was still mainly a political
instrument for establishing the sovereignty of the colonial state
and the authority of its uniformed forces and political agents. Like
forced labour, taxation heightened popular aggravation and
figured prominently in movements of protest and rebellion.
The stabilisation of political conditions took place through
alliances recognising the privileged position of indigenous leaders,
both hereditary and non-hereditary. Colonial administrators were
likely to be tribalist and elitist in their conception of African
society, but they did not unilaterally shape the neo-traditional
settlement. The collapse of the pre-colonial commercial system
had taken the dynamic out of many a local economy and
undermined the revenues and other assets of kings, chiefs and
upstarts alike. They and their associated social elements did
everything possible to rebuild wealth and prestige. A particularly
telling example is that of the omugabe of Nkore, who managed
to get more absolute control of cattle than his predecessors had
had. The Germans in Buhaya, Rwanda and Burundi fostered
aggrandisement among indigenous rulers too, once political
compliance was won.
Expediency in alliances gave way to rationalised criteria for
defining local political units within the colonial state and in
effecting this closer administration, the sentiments of the elders
carried increasing weight. This tendency indicated a convergence
of interests between colonial administrators and elders, especially
in predominantly agricultural societies. The roots of modern
ethnicity thus became strengthened as claims to control land,
women and youth were made by seniors in the wave of tribal
custom and legitimated by administrators.
The contests of the later nineteenth century, for regional and
local power, for lucrative niches in the commercial system, for
religious and cultural hegemony, had in 1905 become subdued and
subsumed by a complicated, comprehensive conquest. The period
from 1870 to 1905 had been an age of improvement for some and
impoverishment for others, just as the period of stable, systematic
590

CONCLUSION
colonisation was to be. The dimensions of impoverishment,

changing sex roles, and belief systems that demand attention have
thus far barely been touched upon. Indeed, the study of social
discontinuities and continuities and the cross-currents of com-
mercial and industrial capital in East Africa during the transitional
period is only at a beginning.

CHAPTER 11
T H E NILE BASIN AND THE EASTERN

HORN, 1870-1908
EGYPT: THE END OF ISMA'IL'S REIGN, 1876-1879

In the 1870s Egypt had a population of about 7 million, some 90
per cent of it rural. Its economy was dominated by the production
and export of cotton, which had increased fourfold in the 1860s,
and by the later 1870s represented 75 per cent of Egypt's exports
by value. Capital inputs, notably Khedive Isma'Il's enormous
foreign borrowings, tended further to strengthen the dominant
export sector by improvements to water-supply, internal transport
and harbours; and by providing this sector with relatively
sophisticated banking, brokerage and marketing services. Other-
wise, capital flowed into Isma'Il's quest for African empire; and
into the development of the new, European-style quarters of Cairo
and Alexandria. A little went into the primary processing of
agricultural products (cotton-ginning, sugar-refining). But build-
ing and primary processing could not initiate an industrial
revolution.
Nor had there been a revolution in the actual processes of
agricultural production. Egyptian agriculture remained highly
labour-intensive. Its equipment was simple, even primitive, and
extravagant in its use of human and animal muscle-power.
Especially in the corvee whereby the irrigation system was
maintained, the falldhin often worked virtually with their bare
hands. In contrast, the financing of cotton production and the
marketing of the product had created an exotic and precociously
developed financial and commercial superstructure, dominated by
foreign residents who were, thanks to the Capitulations, largely
exempt from Egyptian taxation and from Egyptian administrative
and legal control. By 1880 there were perhaps 70,000 resident
foreigners in Egypt.
The land of Egypt was no longer, as it had been in Muhammad
'All's time, virtually the private property of the ruler; by the later
592

EGYPT: THE END OF I S M A ' I L ' S REIGN, 1876—79
1870s land had become a valid security for agricultural loans. But
in 1878 the khedive and his family still held directly about 20 per
cent of the cultivated area. Rather more than another 20 per cent
was held in large estates of 5 o feddans or more (sometimes much
more) by a small elite of perhaps 10,000 individuals. Many of these
large estates had originated as fiefs (Jiflik) granted by the ruler in
return for financial or administrative duties. Such land was still
taxed at a preferential rate — theoretically a tenth ûshr). Most of
the smaller holdings were still, in principle, part of the ruler's
domain; they paid a tribute (khardj) which was in 1881, on
average, over twice as heavy as 'usbr.
From the 1850s communal village tenures had given way to
individual peasant tenures, a process hastened by the growth of
the export economy and the monetisation of agriculture. Less than
20 per cent of the land was held by falldhin in holdings of under
five feddans; and the average size of these holdings by the 1880s
cannot have exceeded two feddans and was tending to shrink. The
natural processes of economic differentiation were in the 1870s
intensified by ruthless fiscal pressure as Khedive Isma'Il strove to
satisfy his foreign creditors. During Isma'Il's reign (1863—79) the
poorer falldhin may have lost some 300,000 feddans; and by 1880
only about 10 per cent possessed the four feddans needed to
support a family. Some 20 to 30 per cent were completely landless.
Yet thefalldh remained a member of the village community, which
in spite of its decay as an economic and fiscal unit, retained its
own customary code of ethics and behaviour; and a solidarity
which made it 'avant tout refuge contre la legalite'.1 This closed
world was penetrable to the ruling institution only through its
headman (shaykh or 'umda), who had every interest in keeping it
closed, for he used his indispensable intermediary position to
enrich himself at the expense of the villagers. Village headmen
were well represented among the 'middling holders' of 5 to 50
feddans, who held nearly 40 per cent of the cultivated area.
At the pinnacle of society Khedive Isma'Il was an autocrat
against whose arbitrary will even the greatest of his subjects had
no legal protection. His ministers were his creatures; he used his
power quite ruthlessly to exile or even to assassinate any who
attempted to thwart him. His ministers, senior administrators and
higher military commanders were still drawn almost exclusively
1
J. Beique, L'Egypte: imperialist/it tt revolution (Paris, 1967), 47.
593

THE NILE BASIN AND THE EASTERN HORN
from a small, and by now hereditary, elite of non-Egyptian ethnic

origin, conventionally known as Turco-Circassians or mamluks.
Although many of this group had become second-generation
residents in Egypt, they still regarded themselves, not as
Egyptians, but as a branch of the Ottoman governing elite and
as the rulers of a conquered people. Their training was primarily
military; but they were regarded as omni-competent alike in civil
administration and military command. In contrast, the very few
native Egyptians who achieved high office were restricted to civil
administration; and even there often to a limited field (finance,
education, public works) in which they had some specialised skill.
High office attracted great wealth, above all in land. The leading
Turco-Circassians, collectively known as the dhawat ('grandees'),
were predominant among the largest landowners, although by the
1870s a handful of native Egyptians had acquired holdings on a
similar scale.
Under Muhammad Sa'Id, Isma'Il's immediate predecessor,
native Egyptians had been admitted to commissioned rank in the
army. In the earlier years of Isma'Il's reign, when the khedive was
attempting to reduce his dependence upon his Ottoman suzerain,
Egyptians had been appointed as province governors. But after
about 1870 province administration reverted largely to Turco-
Circassians. Egyptian officers were treated with hostility and
contempt by a Turco-Circassian high command which demon-
strated total incompetence in the Ethiopian campaigns of 1875-6,
and which usually blocked the promotion of Egyptians beyond
the rank of company commander.
Other than the military officers, the most important Egyptian
groups were the 'ulamd' and the so-called 'provincial notables',
ranging from prosperous 'umdas to very substantial landowners.
The 'ulamiT still enjoyed considerable influence among the
population, but very little within the ruling institution. The great
offices open to 'ulamd' (grand qadl, shaykh of al-Azhar) were in
the khedive's gift: those appointed were well rewarded for their
complaisance in clothing the regime with the necessary minimum
of Islamic legitimacy. The political influence of the provincial
notables, insofar as it existed at all, was purely local, in spite of
their heavy representation in the majlis al-nuwwab (Chamber of
Deputies) which had been set up by Isma'Il in 1866. This body,
chosen by indirect and open 'elections' heavily influenced by the
594

EGYPT: THE END OF I S M A ' I L ' S REIGN, 1876-79
administration, was in no sense a check upon the ruler but rather
an additional instrument of his administration, a more direct link
to subjects thoroughly conditioned to defer to an arbitrary and
ruthless despot. The presidency of the majlis was always held by
a trusted Turco-Circassian. Important legislation was sometimes
promulgated when it was in recess. Until the crisis of 1879'linever
failed to grant Isma'U's financial demands, however exorbitant;
and its members took no overt political initiative independently
of, still less in opposition to, the khedive.
In 1876 Isma'Il was forced by impending bankruptcy to give
his European creditors oversight of his debt service, and to accept
an Anglo-French 'dual control' of his current finance. Isma'U
undoubtedly hoped to evade effective control by exploiting the
complexity and confusion of Egyptian finances. But a catastroph-
ically low Nile in 1877 and a destructive flood in 1878, followed
by disease and famine, reduced him to financial desperation. In
January 1878 he was driven to set up a commission of enquiry
which should in his own words ' faire la lumiere la plus complete
sur la situation financiere'.2 All the members of the commission
but one were European: its leading members were Rivers Wilson,
Isma'U's British financial adviser; and, representing the 'native
element', Mustafa Riyad. Riyad, an important member of the
dhawdt, was Isma'Il's nominee, charged with the protection of the
khedive's financial sovereignty. But Riyad concurred in the
commission's recommendations, which were unwelcome not only
to the khedive but to large landowners generally. There was to
be a surtax on the privileged 'ushuriland and an end to the private
use of corvee labour. The khedive's vast holdings were to be
nationalised and he himself placed on a civil list; and his autocracy
was to be limited by the creation of a 'responsible ministry'.
In August 1878, as a safeguard against further European
pressure, Isma'Il appointed as prime minister Nubar Bughus, a
Christian Armenian of a family which had often been useful to
Egypt's rulers in their dealings with Europeans. Riyad became
minister of the Interior; Nubar insisted on Rivers Wilson as
minister of Finance; the French 'financial controller' became
minister of Public Works. Isma'Il publicly renounced his personal
1
Ministere des Affaires Etrangeres (M.A.E.), Correspondance Politique (C.P.),
Egypte 60, Isma'B to Waddington, 26 Feb. 1878: cited A. Scholch, Anpten den Agjptem !
(Zurich and Freiburg i. Br., n.d. P1972]), j8.
595

power, and agreed to govern 'with and by' this so-called

' European Ministry', which promptly adopted the commission's
programme of reform. The commission had proposed the repeal
of the muqdbala law of 1871, whereby land-tax was halved in
perpetuity on land for which six years' tax had been paid in
advance. In emergencies (such as the severe natural disasters
which had afflicted Egypt) these arrangements could be modified
'after consultation' with the majlis al-nuwwdb. The ministry
therefore advised the khedive to summon the majlis for January
1879.
This was a miscalculation. By the end of 1878 the rural
economy had virtually collapsed, and the falldhin had often been
reduced to starvation, by natural disaster and the ruthless extortion
of a government living from coupon to coupon. The usually
docile majlis protested vigorously against any increase in taxation;
when Isma'Il let it be known that he was personally opposed to
his European ministry's programme, there were loud demands for
a general decrease. Isma'Tl used this opposition to give a ' national'
and ' constitutional' colour to his opposition to European control
and his struggle to regain personal power. He also mobilised
sections of Egyptian opinion not directly represented in the majlis:
the 'ulama", uneasy at the proliferation of infidel officials in
positions of authority; the small but already vocal coteries of
conscious nationalists and Islamic reformers. Some of the latter
had already developed rudimentary political organisations in the
form of secret societies. So too had the numerous military officers,
mainly native Egyptians, whose careers were threatened by savage
economies in the military establishment. In February 1879 Nubar
and Wilson were mobbed by a gathering of disgruntled officers.
Isma'Il, who had covertly encouraged this demonstration, was
able to pose as the saviour of law and order by dispersing it in
person. The outbreak also gave him a pretext to dismiss Nubar,
who was replaced as prime minister by the khedive's son,
Tawflq.
The majlis, which was now demanding full participation in the
management of Egypt's finances, successfully resisted a rather
half-hearted attempt to dissolve it at the end of March. Isma'Il
thereupon drafted his own plan offinancialreform, the so-called
National Programme (la'iba al-wataniya) which was adopted by
the majlis and also supported by representatives of the 'ulama and
596

EGYPT: THE E N D OF I S M A ' l L ' S R E I G N , 1 8 7 6 - 7 9
the Turco-Circassians. In theory, the National Programme
' constitutionalised' Egyptian finances by Isma'Il's promise to act
only through a council of ministers responsible to the majlis. In
practice, it mobilised the notables in support of an attempt to
restore the khedive's personal control. No more was heard of a
civil list or the nationalisation of khedivial property; measures
unwelcome to notables, such as the surtax on 'ushuri land, were
also dropped. Instead, the interest rate on the funded debt was
to be unilaterally reduced to 5 per cent. On 8 April Isma'Il
replaced his European ministry by a cabinet of his own choice
under Muhammad Sharif.
Isma'Il had shown great skill in unifying diverse elites, both
Egyptian and Turco-Circassian, in opposition to foreign inter-
ference with their particular interests. This unification did not
however imply any loss of political preponderance by the dhawdt,
whom the khedive evidently continued to see as his most reliable
supporters. The new ministry of 8 April, loudly touted as ' purely
Egyptian', was in fact purely Turco-Circassian. Muhammad
Sharif was an able Turk with more than a veneer of French
culture, who combined pride of caste with considerable political
sophistication. Like Isma'Il, he was flexible enough to see how
constitutional forms could be manipulated to preserve the existing
power-structure. But although the new government was a Turco-
Circassian monopoly, Isma'Il did not neglect his Egyptian
supporters. The majlis was allowed to continue its sittings and to
discuss a draft constitution. The military establishment was
increased and half-pay officers recalled. Three 'fai/ab' colonels,
among them the future national hero Ahmad 'UrabI, were
appointed khedivial aides-de-camp.
On 22 April Isma'Il brought the 'National Programme' into
force while affecting to welcome continued European co-operation
in the management of his finances. But the powers did not confine
themselves to protests against the unilateral reduction in the
interest rate; by May they were pressing the sultan to depose
Isma'Il in favour of his son, Muhammad Tawfiq. Sultan Abd
iil-Hamid was not averse to a drastic demonstration of his
suzerainty against a wall who had often treated that suzerainty as
a formality. But he wished to replace Isma'Il not by Tawfiq but
by Muhammad 'All's youngest son, 'Abd al-Hallm, who had a
small clandestine faction in Egypt. However, by 26 June the sultan
597

had been brought to nominate Tawfiq, in a telegram accompany-

ing that addressed to 'the ex-Khedive, Isma'lP.
So long as Isma'Il seemed likely to retain effective power the
Turco-Circassians, led by Sharif, showed considerable solidarity
in supporting him, probably seeing in his personal power the best
guarantee for their own privileges. Faction among the dhawdt
certainly existed; but until the very end of Isma'Il's reign it
evidently had little political significance. Certainly, the' defection'
of Nubar and Riyad was of major importance, but neither was
at this stage the leader of an important faction. Both were
anomalous members of the ruling elite: Nubar, as a Christian, was
a tolerated associate rather than a full member, and had few
followers outside the Armenian Bughiis clan; Riyad, though a
pious Muslim and a patron of Islamic reformers, was widely
believed to be of Jewish extraction. But once there appeared to
be three rivals for the khedivate (Isma'Il, Tawfiq and 'Abd
al-Hallm), Turco-Circassians far more representative than Nubar
or Riyad found themselves, almost willy-nilly, at the head of
politically important factions. By May 1879 Sharif had aligned
himself with the powers in pressing for Ismi'Il's deposition; while
Shahln Kinj, a 'grandee' of hardly less distinction, was organising
petitions in support of Isma'Il. After Isma'Il's deposition, factional
disintegration accelerated. Mustafa Riyad was for two years
(September 1879-September 1881) the real ruler of Egypt. Power
attracted followers; Riyad became a major faction leader and a
bitter political enemy of Muhammad Sharif. Mahmud SamI
al-Barudl, as distinguished a 'grandee' as any, ultimately sought
to augment his power-base by links with the increasingly powerful
faction of disgruntled Egyptian officers. Within eighteen months
of Isma'Il's deposition, the old ruling establishment had been
broken into jarring fragments.
The disintegration of the dhawdt-WAS perhaps the most important
aspect of the collapse of Isma'Il's broad anti-foreign front of
Turco-Circassian and Egyptian elites. There followed a complex
struggle of factions and interests, which neither the new khedive
nor the dhawdt were able to control. The Egyptian officers were
increasingly politicised by their bitter struggle with a Turco-
Circassian high command clinging to its military privileges
perhaps all the more tenaciously because of the loss of dhawdt
hegemony elsewhere. The army, under Egyptian leadership,
598

TAWFIQ, 'URABI AND BRITISH OCCUPATION
became a virtually autonomous force, volatile and unpredictable.

The vacuum of effective authority could for a time be filled by
European pressure and influence, operating through individuals
or factions prepared - not necessarily for discreditable reasons —
to follow policies acceptable to the creditor powers. But this was
an inherently precarious and unstable system; and when it
collapsed in September 1881, there was nothing to replace it.
'C'est impossible de dire', wrote the French consul in October
1881, 'oil reside le pouvoir effectif.'3
TAWFIQ, URABI AND THE BRITISH OCCUPATION,

1879-1883
Almost immediately after Isma'Il's deposition the majlis al-nuwwdb
was dissolved. An attempt by Muhammad Sharif to remain in
office as the leader of a ' constitutionalist' faction was frustrated
by the united opposition of the European representatives and of
the new khedive. In September 1879 Britain and France compelled
Tawflq to accept as prime minister Mustafa Riyad, their most
effective political collaborator. Riyad's ministry included two
distinguished native Egyptians, as well as Turks not of
Muhammad Sharif's faction. Riyad strove with considerable
success to satisfy Egypt's foreign creditors and at the same time
to restore productive capacity by easing the intolerable weight of
taxation on the falldh, at the expense of the rich and privileged.
The muqabala was annulled and the surtax on 'ushurf land intro-
duced. The clumsy and confused machinery of collection, which
virtually invited evasion by influential tax-payers, was rationalised;
tax-collectors were given firm central support against recalcitrant
grandees and notables. In July 1881 Riyad's government reached
a settlement, embodied in the 'Law of Liquidation', with the
European creditors. About half of Egypt's normal revenue was
earmarked for debt service; the Europeanfinancialcontrollers had
a consultative voice in the cabinet. These safeguards induced the
creditors to agree to a reduction to 5 per cent of the interest on
the unified debt.
Riyad's reforms were of genuine benefit to the fallahin. There
J
M.A.E., C.P. Egypte 70, Sienkiewicz to Barthelemy St-Hilairc, 4 Oct. 1881: cited
A. Ramm, 'Great Britain and France in Egypt, 1876-1882', in P. Gifford and
W. R. Louis (eds.), France and Britain in Africa (New Haven and London, 1971), 92.
599

was a sharp fall in the interest rates exacted by village usurers;

and the new khedive enjoyed a brief honeymoon of popularity
in the rural areas. Egyptian specialists and experts trained in
Western skills collaborated whole-heartedly with Riyad. None of
them appears in a revolutionary role during the crisis of 1881—2;
and their doyen, the engineer and educationist 'All Mubarak, was
a member of Riyad's cabinet. The opposition came from
elsewhere: the wealthy and privileged, at first the dhawdt, later the
rural notables; the nationalist and militantly Islamic litterateurs and
propagandists; Egyptian officers, their careers once more at the
mercy of economies and of their Turco-Circassian superior
officers. Riyad's methods were authoritarian, even dictatorial: a
considerable section of the dhawdt, led by Muhammad Sharif,
therefore embraced 'constitutionalism' as a weapon against his
policies. The coteries of nationalists and Islamic reformers,
cautiously patronised by Isma'Il towards the end of his reign, now
attacked Riyad as they had attacked Isma'H's European ministry,
in the name of patriotism, constitutionalism and religion. The
principal mentor of these groups was that enigmatic but extremely
influential figure Jamal al-Dln al-Afghani, who had connections
not only among nationalists but also among able young Azharites
repelled by the lifeless formalism of their official teachers; and
even in the world of freemasonry. Isma'Il had tolerated Afghani,
probably hoping to find a use for him: the less confident Tawfiq
had him summarily arrested and deported in August 1879. Riyad
had once been Afghani's patron; but he soon silenced Afghani's
nationalist and reformist disciples by suspending, and threatening
to suppress, publications which printed their views. More con-
structively, he converted into active supporters of his policy some
of the ablest of these disciples, notably Shaykh Muhammad
'Abduh and Sa'ad Zaghlul.
The 'constitutionalist' opposition of the dhawdt had rather
more substance than that of Afghani's circle of intellectuals. In
November 1879 Muhammad Sharif and some of his friends
published anonymously a ' Manifesto of the Egyptian Nationalist
Party', attacking Riyad's government as an agent of foreign
interests and demanding the restoration of financial sovereignty
to Egypt. It seems very doubtful whether the 'Egyptian
Nationalist Party' had any real existence before its appearance late
in 1879 as a constitutionalist facade for Sharif's political
600

connection. In May 1880 a petition signed by Sharif and eighty-

three others, all wealthy and mostly Turco-Circassian, attacked
Riyad's government as representing not Egypt but the house of
Rothschild, in whose interests it had annulled the muqdbala and
increased the tax on 'ushurtXznd. The signatories were intimidated
by police surveillance so strict that they began to fear for their
lives; Sharif's French-language newspaper, La Re/orme, was
suppressed; sentences of exemplary severity were passed on less
distinguished and more vulnerable petitioners against the govern-
ment. Sharif thought it wise to make an extended tour of his
cotton plantations; some of his associates hastily fled the country.
Given the individual and collective economic strength of the
dbawdt, this rather ignominious collapse may seem strange. But
in khedivial Egypt wealth had never been the slightest protection
against the arbitrary power of the ruler. Riyad, with his foreign
backing, was virtually a dictator; he seems to have convinced the
opposition that he was able and willing to use state power no less
ruthlessly and lethally than Isma'Il himself. The grandee
opposition was also very narrowly based. It did not gain support
from the Egyptian provincial notables, though they were also
aggrieved by Riyad's policies. To many Egyptians, the emergence
and activity of the ' Egyptian Nationalist Party' seemed a mere
intrigue of the Turks.
Nor did the dhawat invoke the assistance of their Turco-
Circassian colleagues who still monopolised the army high
command; and who were, under the leadership of the War
minister, 'Uthman Rifql, using the renewed economy drive to
force jallah officers out of operational command and into retire-
ment. In January 1881 Colonel Ahmad 'Urabl, as the spokesman
of the aggrieved officers, demanded the dismissal of the War
minister. Riyad attempted to placate the officers without sacrificing
'Uthman Rifql. But 'Uthman and the khedive planned to crush
the officers' movement. On 1 February 1881 'Urabl and two of
his colleagues were ordered to the War Office on routine business
and, on arrival, were arrested and brought before a court-martial.
But their regiments, forewarned of the trap, stormed the War
Office and released the arrested colonels. Other troops refused the
khedive's orders to intervene. Tawflq had no alternative but to
release and reinstate 'Urabl and his colleagues and to dismiss
'Uthman Rifql.
601

The officer corps then renewed its oath of loyalty, and the
incident was officially closed. But the falldh officers knew that
Tawflq would seek revenge for his personal humiliation: they
believed that not only their careers, but their lives, were now in
danger. Having advanced too far to withdraw safely, they sought
to protect themselves by pressing more strongly than ever for the
destruction of Turco-Circassian preponderance in the army. Riyad
saw the danger of this situation, and strove for a peaceful solution.
Mahmud SamI al-Barudl, who had replaced 'Uthman Rifql as War
minister, was also a Turco-Circassian; but he sympathised with
the Egyptian officers, probably hoping to use their support to
further his own political ambitions. Already the officers' demands
for increased pay and an increased military establishment
threatened conflict with the European financial control. The
French occupation of Tunis in May 1881 seemed to some officers
to foreshadow a British occupation of Egypt. With Isma'U's
deposition Egypt had lost her financial sovereignty; and now
Riyad was dismantling the army so as to leave her defenceless
against the foreigner and the infidel. The officers' movement
developed overtones of militant patriotism and even of jihad.
During the summer of 1881 links were forged between the
officers' movement and the provincial notables led by Muhammad
Sultan, the richest of the native Egyptian land-owners. The
notables saw in the officers a valuable weapon against the Riyad
government: the officers looked to the notables and the recall of
the majlis al-nuwwdb for political protection against the khedive's
wrath. The dismissal in August of Mahmud SamI and his
replacement by the khedive's brother-in-law was followed in
September by orders posting away from Cairo 'Urabl's regiment
and two others commanded by falldh officers. 'UrabI and his
friends saw in these events a design' to disperse the military power
with a view to revenge on us'. 4 On 9 September the three
regiments marched on the khedive's palace and demanded the
dismissal of Riyad's ministry, the recall of the majlis and an
increased military establishment. Tawflq, again lacking all support
from other troops, again had to comply.
A new government was formed under Muhammad Sharif,
« Public Record Office (P.R.O.), Foreign Office (F.O.), 78/5324, translation of'Uribl
to Egyptian War Minister, 9 Sept. 1881, in Malet to Granville, 11 Sept. 1881. Cited
Afaf Lutfi al-Sayyid, Egypt andCromer (London, 1968), 11; cf. Scholch, Agypten, 147,
325 n. 137.
602

whose reputation as a constitutionalist appealed to the officers.

Sharif recognised the weakness of his position and demanded
guarantees, which were given by Muhammad Sultan, of the
army's loyalty to the new government. Mahmud SamI was
reinstated as War minister. The leading provincial notables
successfully petitioned for the recall of the majlis. It met in
December 1881, and at once began to discuss a draft constitution.
'UrabI was given credit for these developments, seen as the dawn
of a new era, and rapidly became the hero of most politically
conscious Egyptians. British representatives toyed with the idea
of Ottoman military intervention; but the French promptly
rejected this proposal, fearing its repercussions in Tunisia. British
alarm gave way to cautious optimism when it became clear that
'Urabi and the majlis were anxious to avoid a collision with the
financial control. But the situation remained dangerously unstable.
The khedive was by now almost powerless, but had not given
up hope of regaining power, either by political intrigue or by
Ottoman military intervention, which he had vainly requested in
September. The army had a monopoly of organised force; but its
spokesman, 'UrabI, by now the idol of the populace, held no
ministerial office until February 1882.5 The majlis was influential
but had no authority over the government. Sharif was a com-
promise leader with no effective following; his authority depended
on the maintenance of a complex and precarious balance of
political forces.
In January 1882 this balance was destroyed by the Anglo-French
'joint note', whereby Britain and France offered (or threatened)
their total support for the khedive's authority. The note was
primarily a French initiative, probably an attempt to enhance
French influence in an Egypt where British advice and advisers
were becoming all too preponderant. In the interests of Anglo-
French unity of action, Gladstone and Granville concurred,
without enthusiasm and with no inkling of the note's likely
repercussions in Egypt. The nationalists in fact saw the note as
a thinly-veiled threat of military intervention; but they were not
intimidated. The majlis promptly demanded control over that
portion of the Egyptian budget not earmarked for debt service
— an issue on which they had previously seemed willing to
compromise. Sharif, at heart a conservative Turk sceptical and
5
However, he became vakil (under-secretary) at the War Ministry in January 1882.
603

contemptuous of the political capacity of Egyptians, refused to

support this demand. On the petition of the majlis, Tawflq
dismissed Sharif and appointed in his place Mahmud SamI, with
'UrabI as War minister (February 1882). Mahmud Saml's ministry
was the first ever to contain a majority of native Egyptians -
mostly technical experts, with one or two rural notables.
The new ministers were neither militarists nor fanatics; but
their stand over the budget gave notice of their refusal to accept
foreign control outside the special sphere of the debt-service.
France and Britain now had to decide whether it was safe to leave
the complex of European interests in Egypt to a government
dominated by unfamiliar and unpredictable political forces.
Gladstone had not been averse to this experiment; but the real
or supposed imperatives of Anglo-French relations had led him
to acquiesce in the joint note - a policy of 'bluff and swagger'
which had not intimidated the 'Urabists but had provoked them
into an open confrontation with the foreign financial control.
After January 1882 the British (and often the French) repre-
sentative in Egypt began to see armed intervention as the only
solution to this crisis. Auckland Colvin, the British financial
controller, now began to press this view very strongly; and he
converted to it Edward Malet, the consul. Henceforth British
official reporting on Egypt increasingly (and misleadingly)
emphasised the threats of'anarchy' and 'fanaticism'.
In London the Gladstone government strove to avoid an
occupation, either alone or in partnership with France; and at the
same time to keep British policy in step with the French. This
involved a policy of intimidation which provoked precisely the
revolutionary instability which the British wished to obviate. In
June 1882 the French abandoned this policy; but by that time
British interests seemed so seriously threatened by the upheaval
in Egypt that London saw no alternative but to persevere in
intimidation, even in isolation from France, and when intimidation
failed yet again, to launch an isolated military occupation. The
British tried to escape from this rake's progress by sponsoring an
Ottoman intervention. This course was always thwarted by
French opposition; but it was in any case based on a total
misunderstanding of the relationship between the sultan and
Egypt. The sultan would willingly intervene only to draw tighter
the links between Istanbul and Cairo, perhaps through the
604

appointment of 'Abd al-Hallm as khedive. He had no intention

of being 'the mere instrument of European action in Egypt'. 6
'UrabI, for his part, assured the sultan that 'Egypt for the
Egyptians', for all its opposition to Turco-Circassian hegemony
in Egypt, did not imply any weakening of proper Islamic loyalty
to the Commander of the Faithful. The sultan replied that he
would support any leader who could break the infidel grip upon
Egypt; and until 'Urabl's cause was virtually lost he tended to lean
towards 'UrabI rather than towards Tawflq who was, after all, the
nominee of the European powers.
Early in 1882 'UrabI, now War minister, had begun to break
the Turco-Circassian monopoly of high military command.
Aggrieved officers protested; 'UrabI believed that there was a
' Circassian conspiracy' to assassinate him. Suspects were arrested
and court-martialled; on 30 April forty Circassian officers were
sentenced to be cashiered and deported to the Sudan. The sultan
refused to confirm the sentences; the khedive then rejected his
ministers' proposal for milder penalties not requiring the sultan's
confirmation, but accepted a similar proposal from the British and
French consuls. The khedive had preferred foreign advice to that
of his own ministers upon an issue which was very remote from
finance — ostensibly the sole legitimate foreign interest in Egypt.
The ministers broke off relations with the khedive and, usurping
his prerogative, reconvened the majlis al-nuwwdb. Although there
was no general disorder, and no threat to foreigners, Malet and
his French colleague called for the despatch of an Anglo-French
naval squadron to Alexandria. The ships arrived on 17 May; the
consuls now demanded the dismissal of the SamI-' UrabI ministry;
a further joint note from London and Paris suggested that 'UrabI
should go into' voluntary exile'. Thereupon the ministers resigned
in protest; but Tawflq was compelled by popular pressure to
reinstate 'UrabI as War minister with responsibility for public
security (28 May).
On 11 June riots in Alexandria caused loss of life among both
Egyptians and foreigners. The riots were probably a spontaneous
result of tension and excitement; but to Gladstone they appeared
as a 'massacre' of Christians, probably instigated by 'UrabI.
Gladstone could now move with an easier conscience towards
armed intervention, to which he was being pushed by anxiety for
6
Ratnm, 'Great Britain and France in Egypt', 103.
605

the security of the Suez Canal. He was also troubled by a marked
rapprochement between the French and the Urabists, probably in
reaction to the increasingly exclusive British influence over the
khedive; he feared a Franco-Urabist entente which would
effectively exclude British influence from Egypt. On 7 July Tawflq
hinted that he would welcome action from the European warships;
the British admiral, Seymour, was authorised to demand under
threat of bombardment the disarming of the Alexandria forts. The
mere presence of the ships had been worse than useless: by 14
June even Malet had admitted that' the fleet [was] a menace likely
to lead to disturbance, not to protection \ 7 Perhaps a bombardment
(even though the French refused to participate) would at last
topple 'UrabI and obviate the necessity for full military
intervention.
The bombardment took place on 11 July, and was followed by
a collapse of public order in which much of Alexandria was looted
and burned. The Urabist army retired from the city; Tawfiq took
refuge with the British squadron; and on 17 July Admiral
Seymour, having landed British troops, publicly assumed respon-
sibility for law and order ' with the permission of the khedive'.
Elsewhere in Egypt the Urabists were in control; and the
khedive's authority, and British influence, could now be restored
only by a military occupation of Egypt. The alternative was simply
to abandon Egypt to the Urabists, and quite possibly to the
French. On 25 July Gladstone secured a vote in the House of
Commons of £1.3 million for military operations in Egypt.
'UrabI now proclaimed jihad against the British. Immediately
after the bombardment there had been a pogrom of Christians in
some delta towns, but there was no general collapse into' anarchy'.
As the khedive and most of his ministers were now in the enemy
camp, in Cairo a council of notables (majlis a/-'ur/i) assumed
responsibility for civil administration. 'Urabl's jihad showed no
tendency whatever to develop into a social revolution. The
populace was mobilised mainly by its traditional leaders - village
shaykhs, 'uiama', local notables - who invoked traditional Islamic
loyalties. Even the more sophisticated nationalist orators spoke
of defence of faith and homeland, hardly of Egypt as a nation.
Indeed, the Urabist slogan 'Egypt for the Egyptians' did not
7
P.R.O. 50/29/160 (Granville Papers), Malet to Granville, 14 June 1882, cited in
R. Robinson and J. Gallagher, Africa and the Victorians (London, 1961), IOJ.
606

imply the creation of a totally independent national state. The

Urabists still saw Egypt as part of the Ottoman Empire, the sole
legitimate Islamic polity. 'Urabl commissioned a board of 'alamo"
to pronounce on Tawflq's legitimacy as a ruler for Muslims; but
only the sultan, as the highest earthly judge of that legitimacy,
could declare the khedive deposed.
Egyptian ' national unity' was moreover still deeplyflawedby
ethnic and social cleavages. The Turco-Circassians, including
Sharif and his 'Egyptian Nationalist Party' of 1879-80, defected
to the khedive and the foreigner. So did many of the wealthier
provincial notables, led by the enormously rich Muhammad
Sultan, as soon as the popularisation of 'Urabl's movement
seemed even remotely to threaten the social order. The major
revolutionary effect of the movement was to damage beyond
repair the authority and solidarity, already severely shaken, of the
old ruling elite. British military intervention destroyed any early
possibility of its effective replacement by a new elite.
'Urabl had concentrated his better troops inland from
Alexandria, and was slow to recognise the threat from British
forces based on the Suez Canal. The Egyptian force routed by
Wolseley at Tell el-Kebir on 13 September consisted largely of
untrained falldh recruits. Immediately after the battle the majlis
al-'urft announced its capitulation, and Cairo was occupied
without a fight. As a military operation, British intervention had
been a brilliant success; but its political objectives were quite
incapable of realisation. It was impossible to restore the pre-'Urabl
situation, and then quickly to withdraw the British troops. The
old regime was an irreparable ruin. The British now found
themselves taking over political and administrative functions
which demolished it further. Even before Tell el-Kebir, Malet had
imposed upon the khedive a' reliable' ministry under Muhammad
Sharif. Immediately afterwards, the British took over the railway
administration and the training of a ' gendarmerie' to replace the
Egyptian army, dissolved by khedivial decree. The convenient
absence of the senior French officials enabled them to monopolise
control of Egypt's finances. Major administrative reforms on
British-Indian lines seemed necessary before Egyptians could be
entrusted with political responsibility; and much of this pro-
gramme was anything but welcome to the khedive and the
' pashas' - the crumbling pillars of the edifice which the British
607

hoped to restore. Tawfiq's puppet status was perhaps most
glaringly revealed, especially to Egyptian eyes, by the British
'arrangement' of 'Urabl's trial and sentence so as to spare his life.
In May 1883 the British introduced a dummy constitution
evidently intended as a plaything for political children: a mainly
nominated legislative council with no powers of initiative, or even
of disallowance except for proposed new taxes; and an elected
assembly which met briefly and infrequently to 'emettre des
voeux'.
Evelyn Baring, who replaced Malet as British consul in
September 1883, soon discovered that the old regime had collapsed
and that the British programme embodied a fundamental contra-
diction. To withdraw before reforms had time to take effect would
be to invite a relapse into 'anarchy', so that the British would
either have to return or to ' stand aside while others, probably the
French, take up the work'. 8 Nevertheless, until December 1883
Baring still hoped for evacuation within two years. News was then
received that the Sudanese Mahdi had annihilated an Egyptian
expeditionary force under a British commander. This expedition
had drained the last resources of Egyptian military manpower.
There seemed a real danger of the invasion of Egypt by Mahdist
revolutionary 'fanatics'. The British suddenly found themselves
saddled with responsibility for Egyptian defence, which destroyed
any hope of early evacuation, as well as for Egyptian finance.
Finance was problem enough. The 1883 budget was a million
pounds in deficit. The unfunded, 'floating', debt exceeded £5
million and was rising. Reforms, or even solvency, were impossible
if Egypt continued to pour resources into the bottomless hole of
the Sudan. Baring persuaded London to insist on the complete
abandonment of the Sudan, except for the Red Sea ports which
were of strategic importance to Britain. To strengthen Baring's
hand against the reluctant Egyptian ministers, London asserted
—in a despatch (4 January 1884) that became the foundation-charter
of the 'veiled protectorate' - the doctrine that in 'important
questions' Egyptian ministers must either accept British ' advice'
or resign. But Muhammad Sharif preferred to resign rather than
accept so bitter a humiliation. Baring had doubted whether any
Egyptian would be willing to succeed him, and was preparing to
8
Baring to Granville, 28 Oct. 1883, in Cromer, Modem Egypt (London, 1908;
reprinted London, 1911), 738-41 (of reprint).
608

MAHDIST REVOLUTION IN THE SUDAN, 1874-85
impose upon the khedive a ministry of British officials. But Nubar
Bughus agreed to form a government. Although a Christian
Armenian, he was at least preferable to a British official; and his
very marginal status as an Egyptian relieved ' genuine' Egyptians
of responsibility for the abandonment of the Sudan.
THE MAHDIST REVOLUTION IN THE SUDAN, 1874-1885
The destruction of the Egyptian expedition under William Hicks

in November 1883 was the crucial event in the successful
insurrection of the Sudanese Mahdi, Muhammad Ahmad ibn
'Abdallah. Belief in a Mahdi, as a divinely guided restorer of
justice and equity, had existed for many centuries in popular Islam,
though it had never been an official tenet of the sunni faith.
Muhammad Ahmad's own conception of his mission is well set
out in a letter which he wrote, shortly before his death in June
1885, to the Christian Yohannes IV of Ethiopia. He asserts that
Islam had not paused in its universal expansion until
... religion fell into the hands of the Turks, w h o . . . replaced it by kufr. They
annulled the laws of the Merciful and revived the ways of Satan after their own
inclinations. When God determined to cut short such a state of things, he called
me forth as Mahdi... God has manifested His goodness to you in causing you
to be present at this age of prophecy in which we have appeared as a khalifa
to our Prophet Muhammad... Become a Muslim [or] you will undoubtedly fall
into our hands, as we are promised the possession of all the earth.9
The Ottoman empire by its apostasy (which venal 'ultima" had

condoned) had lost its Islamic legitimacy not merely in the Sudan
but universally. Muhammad Ahmad was the sole existing
legitimate successor {khalifa) of the Prophet. It was his mission
to restore the Islamic community {umma) to its pristine purity, as
it existed in the days of the Prophet and his Companions. The
restored umma would be extended, by the sword if necessary, not
merely throughout Dar al-lsldm but throughout the world. The
Mahdi was indeed a cosmic figure. His manifestation was one of
the 'signs of the Hour' - a prelude to the end of the created
universe and the Last Judgement. These eschatological overtones
seem, on the whole, to have been rather lightly emphasised in
Sudanese Mahdism. But the role of the Mahdi as the ' Expected
• Central Record Office (C.R.O.), Khartoum, Mahdia 1/54/11, the Mahdi to
Yohannes, 1502 A.H. (can be dated to Apr.-Junc 1885).
609

Deliverer', who would break oppression and inaugurate a Golden

Age of equity and righteousness, was very strongly emphasised,
and of obvious revolutionary potential.
The Mahdiyya is hardly explicable simply as a belated protest
against two generations of Turco-Egyptian oppression and fiscal
extortion. The sedentary cultivators of the northern riverain
Sudan, who had been longest under Egyptian rule and were most
accessible to the tax-collector, were as a group tardy and
unenthusiastic Mahdists. Moreover, they had often reached a
satisfactory modus vivendi with the ruling institution. By the 1870s
many of them had become members of the bureaucracy, at all
levels up to and including province governor; and this relationship
still survived in spite of Isma'U's harsh fiscal pressure. The early
centres of revolt were on the southern and western periphery of
the Arab and Muslim Sudan, in areas only recently brought, or
still being brought, under effective khedivial administration. Nor
were the insurgents in these regions particularly pious Muslims;
rather the contrary. Nor were they protesting at inveterate
generalised 'oppression', but rather against specific grievances of
comparatively recent origin. The Mahdist message gave these
protests a unity and a holy legitimacy which would otherwise have
been lacking.
The core of the Mahdist movement was of course the group
of dedicated puritanical pietists who were the Mahdi's immediate
disciples. This group was indeed drawn mainly though not
exclusively from the riverain Sudan. Devout and ascetic Sudanese
had long been scandalised by the luxury and laxity of the ' Turks'.
But here again in Isma'lTs reign new factors appear, factors which
gave special point to the charge of apostasy from Islam. The
increasing appointment of European infidels to positions of
authority over Muslims, culminating in Gordon's first governor-
generalship (1877—9), w a s n o t m e r e t y objectionable but un-
canonical. Another inadmissable innovation was the establish-
ment of a Catholic mission in Khartoum itself. Direct
proselytisation among Muslims was of course forbidden, but the
missionaries were active in medicine and education. The ' Turks'
went beyond mere tolerance; the administration maintained
friendly, even cordial, relations with the missionaries. The mission
building was the finest in Khartoum, more imposing even than
the governor-general's palace. The Mahdi was also well aware of
610

Khedive Tawflq's subservience to unbelievers. In these circum-

stances, the Mahdi's ' call to assert the unity of God,... in all ages
the rallying-cry of Muslim rigorists and reformers', was
particularly necessary and apposite. And the conclusion that
'preaching will not purify the Turks; only the sword will purify
them' was to Muhammad Ahmad inescapable.10
The more material grievances which recruited support for the
Mahdi on the periphery of the Muslim Sudan were also generated
by developments in the later years of Isma'Tl's reign. The conquest
and occupation of Darfur (1874) led to resistance not only from
the Fur, but more importantly from the nomad cattle-Arabs
(Baqqara) of southern Darfur, who found that they had exchanged
the light and intermittent suzerainty of the Fur sultan for the
detested encroachment of a tax-collecting bureaucracy. To the
Baqqara, the Mahdi's call was simple:' Kill Turks and stop paying
taxes.' It transformed these notoriously easy-going Muslims into
the unlikely shock-troops of a puritan revolution.
Isma'Il's attempt to assert effective control over the non-Muslim
southern Sudan was unwelcome to the diaspora of northern
riverain immigrants who dominated this region economically and
politically. Yet more unwelcome was his attempt to suppress the
slave trade, the economic base of this community of tough and
enterprising frontiersmen. Khedivial troops and officials, many of
whom also profited by the trade, sympathised with the traders.
Opposition to khedivial policy was given a religious colour by
the employment of Christians (Charles Gordon, Romolo Gessi)
to enforce an alien and ruinous code of business ethics. By the
late 1870s the Bahr al-Ghazal, now the main centre of the trade,
had become ungovernable. In 1878, as a measure alike against the
trade and its insurgent organisers, Gordon encouraged the
Baqqara to expel the northern Sudanese merchants (Jalldba) who
supplied the rebels with arms and ammunition in return for slaves.
The plight of the jalldba, not merely expelled but plundered,
aroused further resentment among the northern diaspora.
The administration of the Sudan was, in the late 1870s, in poor
shape to cope with a major emergency. Isma'Il's grandiose
programme of conquest and expansion had grossly overloaded it
after a period of neglect and deterioration under his immediate
10
P. M. Holt, The Mabdist State in the Sudan, 2nd edn (Oxford, 1970), 117; n o , citing
the Mahdi to Muhammad al-Amln al-Darlr, 28 Sha'ban 1299/1; July 1882.

3O°N
2O°N
El Fasher .-''
DARFUR Shavkanx
Sna*anx } - ..„ ^ . N^.Cjondar
KORDOFAN "Taqali I" \ ^ / Q, T
B a a FaZU9R
0" * ' -j.Qad.r' h / ^GOJ
10° N
SHANQUL
<oj»g
Approximate boundaries of
the Mahdist state.c 1887
A Major battles
TAKA Regions or provinces
Beja Ethnic or tribal groups
9 590km
6 ' ' 300miles
24 The Nile Valley in the Mahdist era, c. \%%o-c. 1900

6iz

predecessors. The appointment of European officials to stiffen the

administration in frontier areas was not an unqualified success.
Whatever their virtues of honesty and efficiency, they were
ignorant of local languages and customs and often (like Gordon)
illiterate in Arabic, the language of administration. They were
resented, not only as aliens and infidels, but as interlopers who
blocked the promotion of Egyptians and Sudanese. Even Gordon
was a mixed blessing. His harrying oftheja/Zaba left the south-west
ready to explode at the slightest shock. His capricious appointment
and dismissal of senior officials did not improve morale. And the
collapse of the khedivate under European and Urabist pressures
damaged the authority of the administration as a whole.
Muhammad Ra'uf, who succeeded Gordon as governor-
general, was neither an ally of the slavers nor a mere incompetent
faineant. He knew the Sudan; his previous career had been
creditable; his taking of Harar in 1875 had been the solitary
success of Isma'Il's invasion of Ethiopia. After Gordon's dynamic
but disturbing regime, there was a case for ruling the Sudan with
a lighter rein and avoiding provocatively energetic policies. In
other circumstances, this strategy might have been successful. But
Ra'uf was certainly slow to appreciate the danger of Mahdism,
perhaps because his attention was first drawn to it by a rival
religious leader whose testimony was discounted. When he did
move to action, his tactics were sensible but over-deliberate and
poorly co-ordinated. He thereby enabled the Mahdi to win
'miraculous' victories which rapidly transformed him from a
voice crying in the wilderness to the leader of a formidable
insurrection.
Muhammad Ahmad al-Mahdl was the son of a boat-builder
from Dongola whose family migrated first to the Khartoum
region and then in 1870 to Aba island in the White Nile, a region
which had been beyond the' frontier of Islamisation' twenty years
earlier. Like so many of his followers, Muhammad Ahmad too
was a frontiersman. After a religious education in the Sudan, he
became the disciple of Shaykh Muhammad Sharif Nur al-Da'im
of the Sammaniyya tariqa (religious brotherhood). In 1878 the
disciple found his master wanting in puritanical rigour, and
remonstrated with him. Shaykh Muhammad Sharif thereupon
dismissed his disciple, in spite of Muhammad Ahmad's repeated
pleas for forgiveness. After this deeply disturbing crisis (for the
613

bond between master and disciple was normally inviolable),

Muhammad Ahmad visited Kordofan, preaching the renunciation
of this world and forming links with notables and holy men,
especially those of the faction opposed to the province governor
and his local Sudanese supporters. He then returned to Aba, where
he built up his own circle of disciples.
Early in 1881 this circle was joined by a novice from south-
western Darfur, 'Abdallahi ibn Muhammad. 'Abdallahi, a diviner
and soothsayer of the Ta'alsha tribe of Baqqara Arabs, had for
some years been seeking the 'expected Mahdi'. 'Abdallahi's faith
and devotion reinforced Muhammad Ahmad's growing convic-
tion of his own divine election, which he confided to his disciples
in March 1881. He then revisited Kordofan, privately revealing
himself as the Mahdi to adherents both old and new, none of
whom betrayed his secret. His public manifestation followed in
June, in letters from Aba to notables and to his adherents, urging
them to join him in hijra. Hijra (emigration, flight) implied a total
repudiation of the corrupt society of infidels and apostates; and
the creation, beyond their reach, of a militant stronghold of the
purified Faith.
In August 1881 an attempt to arrest the Mahdi was so
mismanaged that two companies of Muhammad Ra'uf's troops
were routed by the Mahdi's ansar ('helpers', adherents). The
Mahdists then made their hijra to Jebel Qadir in the south-eastern
Nuba mountains, a turbulent frontier region of Kordofan not
under settled administration. Travelling light, they out-paced the
troops which attempted to intercept them. In December 1881 and
again in May 1882 the Mahdists defeated ill-organised military
expeditions launched against them from Fashoda. Inspired by
these miraculous victories, visionaries and malcontentsflockedto
the Mahdi's stronghold. Local and tribal risings were fomented
virtually throughout Kordofan; and in August 1882 the Mahdi,
now commanding an organised army, advanced upon El Obeid,
the province capital. In September, however, a Mahdist assault
upon El Obeid was repulsed with heavy casualties. Breaking with
Prophetic precedent, the Mahdi now authorised the use of
firearms. He also embodied in his own army the government
troops taken prisoner. These were trained professional soldiers,
usually islamised Blacks. Known as the jihadtyya, they formed a
614

third element in the Mahdi's forces, neither tribal warriors nor

religious enthusiasts. El Obeid finally fell to the Mahdists in
January 1883.
During 1882 there had been Mahdist risings on the west bank
of the White Nile and in the Gezira region between the White
and Blue Niles. The former were contained, and the latter
effectively suppressed, by a new and more energetic governor-
general, 'Abd al-Qadir Hilml. 'Abd al-Qadir recommended a
Fabian policy: consolidation in the Gezira and the riverain Sudan;
and a defensive screen, based on the White Nile, against the
Mahdists in the west. Instead, Muhammad Sharif in Cairo insisted
on a major offensive - an all-or-nothing gamble with Egypt's last
military resources, a scratch force of ill-trained and reluctant
soldiers. The expedition was annihilated at Shaykan, south of El
Obeid, on 5 November 1883.
It was now no longer possible to isolate the Mahdi in
Kordofan. Darfur, already cut off from Khartoum since the fall
of El Obeid, promptly fell to the Mahdists in December 1883,
thanks to Mahdist-inspired risings and the defection of the
government troops. After Shaykan, the hasty withdrawal of the
dangerously exposed White-Nile garrisons left the south isolated.
In Bahr al-Ghazal there were risings both of the northern diaspora
and of the southern Sudanese peoples. The latter were not usually
Mahdists or even Muslims, but made common cause with the
Mahdists against a government seen as a mere predator. The
province finally fell in April 1884 to an invading Mahdist army
from Darfur.
Meanwhile, towards the end of 1883 Khartoum had been cut
off from the coast by the insurgent Beja of the Red Sea Hills. The
Beja were camel nomads who did not speak Arabic, were not
remarkable for their religious fervour and had reached a modus
vivendi with government as transport contractors on the Berber—
Suakin route; but to the Beja, as to other nomads, all settled
government was fundamentally unwelcome. They were raised for
the Mahdiyya by 'Uthman Diqna, a Suakin merchant of part-Beja
descent who had an old grudge against government. But
'Uthman's Beja blood was less important for the success of his
mission than the endorsement of his call to jihad by the religious
leader al-Tahir al-Majdhub of El Damer, the head of a holy family
615

to which the Beja attributed supernatural powers. By early 1884

Suakin itself was under such heavy pressure from 'Uthman Diqna
that British troops had to be committed to its defence.
After Shaykan, Baring had suggested that a British officer
should be sent to Khartoum to organise the withdrawal of the
Egyptian garrison and if possible to set up a successor government.
British opinion pressed for the appointment of Gordon, whose
influence and popularity in the Sudan it over-estimated. Baring
stifled his misgivings about Gordon's suitability; and in January
1884 Gordon was commissioned by the cabinet primarily to report
and advise on the Khartoum situation and not to organise an
evacuation. He was however authorised ' to carry out such other
duties as... may be entrusted to him by Sir Evelyn Baring'. Baring
thereupon instructed Gordon to withdraw the troops and, if
possible, ' to establish some sort of rough government under the
tribal chiefs'. Gordon was permitted temporarily to retain such
troops as he needed to organise an efficient withdrawal; but
emphatically warned that they were not to be retained for the
support of the successor government." Baring had however
significantly changed the emphasis of the original cabinet directive.
Moreover, if Gordon was not merely to report, but to organise
the evacuation, he needed overriding authority; he was therefore
appointed governor-general. But he undermined his own position
and encouraged waverers to defect to the Mahdi by prematurely
divulging the khedivial decree of abandonment, ' not knowing
well its contents'. 12
On arrival at Khartoum in February 1884, however, Gordon
decided, contrary to his instructions, not to withdraw the troops
until the Sudan had been made safe for a successor government,
and he urged Baring to send a British expedition to 'smash up
the Mahdi'. By March the Mahdists controlled the country
immediately to the north of Khartoum; by the end of May they
had taken Berber. Evacuation was now impossible without
British military assistance, which London refused to provide.
Gordon attempted, at first with some success, to expel the
Mahdists from the vicinity of Khartoum; but he had to abandon
a plan to retake Berber. He then resolved to defend Khartoum
11
P.R.O., F.O. 78/3662, Granville to Baring, no. 40, 18 Jan. 1884; F.O. 78/3666,
Baring to Granville, no. 100, 28 Jan. 1884; F.O. 633/6 (Cromer Papers), no. 23, Baring
to Granville, 21 Jan. 1884. Cf. Cromer, Modem Efypt, 343-6.
12
Elton (Lord), General Gordon's Khartoum Journal (New York, 1961), 176.
616

until relief arrived or the city fell. In September he learned that

a relief expedition was after all being organised. Khartoum was
by now under heavy Mahdist pressure; by the end of October it
was closely besieged, and in growing distress for lack of supplies.
Gordon organised a defence of legendary resource and valour, but
Khartoum fell on 26 January 1885 and Gordon was killed, 'a
brave soldier who fell at his post'. 13 Advanced elements of the
relief expedition arrived in two steamers two days later. The
approach of this force had caused the Mahdists considerable
uneasiness; but it was too weak to have effected anything but the
personal rescue of Gordon — to which he would certainly not have
consented.
By January 1885 the Mahdi had already laid the administrative
and legal foundations of his 'new order'. Perhaps even before the
hijra to Jebel Qadlr he had appointed four khalifas — the
equivalent, in the forthcoming ultimate universal triumph of
Islam, of the four ' orthodox Caliphs' who had presided over its
early expansion. 'Abdallahi ibn Muhammad of the Ta'alsha
Baqqara, the Mahdi's closest confidant, took the place of the
Prophet's immediate successor Abu Bakr; the pious and ascetic
'All wad Hilu that of 'Umar; the Mahdi's own kinsman
Muhammad Sharif, that of the fourth Caliph, the Prophet's
son-in-law, 'All. The place of the third orthodox Caliph, 'Uthman,
was offered to Muhammad al-SanusI, the leader of a powerful
religious brotherhood in North Africa; but the offer was ignored
and 'Uthman's place remained vacant. 'Abdallahi soon became by
far the most important of the khalifas. He held overall command
of the Mahdist army, as well as 'divisional' command over his
own fellow-countrymen, the Baqqara levies embodied in the
'Black Flag' regiment. In January 1883, immediately after the fall
of El Obeid, the Mahdi had conferred plenary powers upon
'Abdallahi in all 'matters of the Faith' - which under the Mahdist
dispensation included all public affairs. The ansar were strictly
enjoined never to question 'even silently' 'Abdallahi's orders and
decisions, on pain of punishment in ' this world and the world to
come'. 14
The Mahdi administered his territories through machinery
which did not differ greatly from that of the Turco-Egyptian
13
Holt, Mahdist State, 103, citing Slatin's words.
14
Proclamation of the Mahdi, 17 Rabf I 1 300/26 Jan. 1885: cited Holt, Mahdist State,
122-3.
617

regime - which was indeed the only model known to him. Many
of the changes were of nomenclature rather than of substance. The
function of the Mahdist treasury {bayt al-mdl) was not however
merely to create a revenue by collecting - always with difficulty
from the Baqqara - a share of the booty seized from unbelievers.
The early bayt al-mdl had an important role in maintaining the
solidarity of the Mahdist urn ma by relieving needy brethren. ' He
who has but little... shall be allotted sufficient from the
Treasury... If the Treasury is empty, have patience until God
gives the Treasury sufficiency.'15 Taxation, stigmatised as jizya
(canonically payable only by unbelievers) when levied by the
Turks, could now be sanctified as vyikat, the primitive Islamic
alms tax. The Mahdi's legislation on the relations between the
sexes was also concerned to promote stability and solidarity by
removing the occasion of fitna (dissension, mischief).16 All
communication between the sexes, even formal greetings, was
forbidden outside the family and the marriage bond. Unmarried
women were an important source of fitna: the Mahdi therefore
encouraged early marriage by strictly limiting both the dower paid
by the husband and the traditionally lavish expenditure on
marriage feasts.
To the Mahdi, the sharVa was of course the sole acceptable body
of law; but he did not regard himself as bound by the jurisprudence
of the 'ulama1, which had been superseded by his own special
revelation. The authoritative sources of law were therefore
limited to the Koran, the sunna of the Prophet, and the Mahdi's
own proclamations. In this and other ways he created wellnigh
insuperable difficulties for Muslims conscientiously unable to
accept him as Mahdi. He added to the Muslim confession of faith
the words ' Muhammad Ahmad is the khalifa of the Apostle of
God'; and he blazoned this amended shahdda on his banners. He
at least implied that hijra to the Mahdi had superseded pilgrimage
to the Muslim holy places; and, by equating with infidels Muslims
who denied his mission, he created in Sudanese Islam a schism
which still exists in a muted form. It has been muted by a tacit
consensus which accepts Muhammad Ahmad as a great and holy
mujaddid (' renewer' of Islam), and as a Sudanese national hero —
indeed as abff l-istiqldl, the father of independence. In fact, there
15
Undated proclamation (May 1884-June 1885), cited ibid., 127.
16
Fitna can mean both 'dissension' and 'infatuation'.
618

CROMER'S EGYPT, 1883-1907
is no trace of Sudanese nationalism in the Mahdi's propaganda
or policy; on the contrary, he is at pains to emphasise the universal
claims of his mission. The origins of Sudanese nationalism may
more profitably be sought in the later transformation of the
Mahdiyya from a universal movement to a territorial state.
In June 1885 the Mahdi died, probably of typhus. A few weeks
later, the last Egyptian garrisons in the Sudan, except at Suakin
and in Equatoria, had either capitulated or been withdrawn. The
expedition for the relief of Khartoum retired to the Wadi Haifa
region, where it soon assumed a purely defensive role. A
half-hearted British attempt to advance from Suakin to the Nile
was quickly abandoned. The British turned their backs on the
Sudan in order to grapple with the problems of Egypt.

The least of the problems for the British was Egyptian resistance.
There is no trace of the 'fanaticism' which Baring frequently
invoked to justify continued British occupation. There was little,
if any, popular support for Mahdism. Some have seen a quasi-
political protest in the rural 'brigandage' of the 1880s; if so, it
was a pretty feeble one, compared with the falldh risings of
Isma'H's reign. The elites were as quiet as the masses. By 1885
Khedive Tawftq had tacitly accepted his puppet status. The
Turco-Circassian dhawdt were by now a mere debris, discredited as
a ruling elite and shattered as a coherent political force. Baring's
conviction that the native Egyptian notables were incapable of
replacing them was of course self-interested; but it was not
groundless. Not only were most of these notables politically and
administratively inexperienced; their sorry record of desertion
and delation, as the sands had run out for 'Urabl, gave point to
Baring's strictures on their lack of political stamina and ' civic
virtue'. Western-influenced intellectuals had seen themselves
largely as Isma'H's auxiliaries in a drive for the regeneration of
Egypt by selective borrowing from the West; and the collapse of
the khedivate had shattered their still frail self-confidence.
The silence was broken only by the voice of Shay kh Muhammad
'Abduh, who had ultimately thrown in his lot with 'Urabi and
been exiled from Egypt after 'Urabl's fall. In collaboration with
his old mentor Jamil al-Din al-Afghani he published in 1884 the
619

short-lived but immensely influential periodical a/-' Urn/a al- Wuthqd

('The Indissoluble Bond'). Muhammad 'Abduh believed that
foreign, and especially British, domination could not be effectively
resisted by Egypt alone, but only by the Islamic world as a whole.
But the Islamic umma must first be regenerated by a return to the
true and eternal principles of Islam, long obscured by the static
formalism of the 'ulama" and by the 'corruptions' of popular
devotion. Islam demanded, not the passive acceptance of the faith
in its current imperfect form, but the active exertion of believers
for the rediscovery and dissemination of these true principles,
which could alone give the umma strength and vitality to resist
the betrayal of Islam by worldly and venal rulers — such as Tawflq.
Baring took aJ-'Urwa seriously enough to forbid its circulation in
Egypt. He always took 'Pan-Islam' far more seriously than
' secular' nationalism, which he dismissed as exotic and artificial,
and therefore harmless.
Before 1892, however, the most pressing problems faced by the
British were international, not internal. Britain's position in Egypt
was ultimately dependent upon the absence of German opposition;
and until 1888 Bismarck exacted a high diplomatic price for this
favour. By the London Convention of 1885 Egypt's 'floating'
debt was restricted to a million pounds, and full international
control would be restored unless Egypt were solvent by 1887:
provisions which the French hoped would lead to loss of financial
control by the British and so to early evacuation. The 1887 budget
was in fact 'balanced' only by ingenious accounting expedients.
But by 1889 Egypt was genuinely solvent, and thereafter Baring
built up a considerable reserve. This was achieved by rigid control
of expenditure, not by heavier taxation; the land tax was in fact
reduced in the less prosperous provinces. The shortfall was made
good by a heavy customs duty on tobacco, combined with
prohibition of its cultivation in Egypt. After about 1895 the
revenue benefited from the general buoyancy of the economy,
although the land tax was not increased.
Until 1892 internal politics amounted to little more than the
adjustment of relations between the British and their more
distinguished collaborators. Nubar Bughus had been indispens-
able during and immediately after the Sudan crisis of 1884—5. He
used his unique position to prevent what he called an 'adminis-
trative occupation'; and in particular to force the resignation of
620

Clifford Lloyd, the 'inspector-general of reforms', who was

attempting to bring the Ministry of the Interior and the provincial
administration under direct British control. But Nubar over-bid
his hand when in 1887-8 he attempted, in collusion with the
khedive, to undermine Baring by direct complaint to Lord
Salisbury in London. Tawflq, who had more to lose than Nubar,
was warned that his position as khedive depended upon his
willingness to accept British advice. He took the hint, and
dismissed Nubar at a time diplomatically convenient to London
(June 1888).
Niibar was replaced by Mustafa Riyad; a genuine 'Turk' could
now accept office without incurring the stigma of having
surrendered the Sudan. Riyad was a less unrepresentative figure
than Nubar and a man of generally admitted ability. He was
therefore a useful veil for the ' veiled protectorate', in spite of his
aggressive and autocratic personality and the inclusion in his
political circle of some comparatively outspoken opponents of the
occupation. Like Nubar, he opposed the development of an
'administrative occupation'. But there was no room for two
autocrats in Egypt. In April 1891 he resigned, having fallen foul
of Baring over the appointment of a British judicial 'adviser' with
administrative powers. He was succeeded by Mustafa Fahml, a
wealthy and politically colourless Turk who had held the dummy
portfolio of foreign affairs in almost every ministry since the
deposition of Isma'Il. Mustafa Fahml was, in Baring's words, a
'very facile instrument'.17
So long as the British retained any serious intention of
withdrawing their troops from Egypt it was obviously undesirable
to rely upon mere nullities as Egyptian ministers. But from about
1888, as a result of changes in the European and Mediterranean
power-balance, the British ' official mind' became convinced that
the diplomatic disadvantages of remaining in Egypt were out-
weighed by the strategic dangers of withdrawal. Egyptian
ministers were still of course necessary, lest the powers be
affronted by too blatant a demonstration of British control. But
their competence was now of minor importance, provided they
were prepared to be ' facile instruments'. Mustafa Fahml accepted
this role; but not 'Abbas Hilml, who succeeded his father as
Khedive in January 1892 at the age of seventeen. 'Abbas was
" Cited al-Sayyid, Egypt and Cramer, 78.
62!

determined to assert his independence against Baring (now Lord

Cromer). He could expect some support in Egypt, for the voice
of opposition was at last becoming audible. There was of course
as yet no organised 'national party'; but there were oppositional
groups among certain members of the old elites, the more active
elements of the 'ulama', and the younger generation of the
intelligentsia. The newspaper al-MWayyad (' The Supporter') acted
as a mouthpiece for these circles. Founded in 1889 by the Azharite
shaykh, 'All Yusuf, a protege of Riyad, it was the first newspaper
to be edited by an ethnic Egyptian and consistently to present an
Egyptian point of view. By 1892 al-MWayyad had become a
well-informed and effective critic of the occupation regime.
For 'Abbas Hilml, as later events were to demonstrate, the
embryo nationalist movement was little more than an instrument
whereby he hoped to restore his own personal power. But his
patronage of the movement brought him a popularity which
disconcerted Cromer, who was soon forced to revise his view that
it was 'absurd to suppose that any lad of nineteen could seriously
influence native opinion'. 18 'Abbas took his stand on his legally
unassailable right to choose his own ministers. In January 1893,
without consulting Cromer, he dismissed and replaced Mustafa
Fahml. Cromer reacted by a 'tumultuous storm of sinister
telegrams'19 to the Foreign Office. If'Abbas succeeded, ten years
work would be thrown away, and a new and more serious
Egyptian question would emerge. More, 'Abbas was the agent of
an international conspiracy, involving France, Russia, the sultan
and the ex-khedive, Isma'Il. Cromer demanded authority to use
the British troops to mount a putsch and install the British
advisers as ministers. All this was too much even for Rosebery,
Cromer's one reliable supporter in Gladstone's Liberal govern-
ment. However, the Cabinet supported an ultimatum to the
khedive giving him twenty-four hours to reconsider his choice.
After consultation with Cromer (this was, of course, the crucial
point) the khedive agreed to appoint Mustafa Riyad. The results
were not to Cromer's liking. Egyptians knew that Riyad, whatever
his faults, was too spiky a character to be the mere puppet of
anybody; and Riyad found himself for a moment in the unfamiliar
role of a popular hero. There was a hostile demonstration outside
18
F.O. 141/297, Cromer to Rosebery, 9 Jan. 1893, cited ibid., 105.
19
Rosebery to the Queen, 18 Jan. 1893: Tie letters of Queen Victoria, 3, ii (London,
1931), 107-8.
622

the offices of the British-sponsored newspaper al-Muqattam. In
Egyptian eyes, 'Abbas had won.
Cromer then threatened to resign unless the British garrison
was reinforced: he alleged that Riyad had ' recently taken a strong
religious turn', and was hatching a 'fanatical' movement on
Urabist lines.20 Cromer got his reinforcements, but the Egyptians
were not impressed, and Cromer was dismayed by the rapid
development of anti-British feeling in the course of 1893. Even
the hitherto docile legislative council was coming to life. In
December 1892 it had refused to discuss the budget on the ground
that it had been given insufficient time to do so. A year later it
was to oppose the appropriation for the expenses of the occupying
troops. Indeed, as early as January 1893 Cromer feared that
control had already 'slipped out of [his] hands'. 21 But 'Abbas
over-reached himself by his efforts to undermine the authority of
British officers over their Egyptian troops. In January 1894 this
campaign culminated in a public rebuke to the British Sirdar, Sir
H. H. Kitchener, at a review of the Wadi Haifa garrison. The
khedive was unwise to raise the spectre of an insubordinate army;
on this issue Cromer could expect unconditional support even
from a Gladstone government. 'Abbas was forced to publish an
' order of the day' praising the British officers. Riyad was faced
with the thankless task of persuading 'Abbas to comply with
Cromer's demands. His relations with the khedive deteriorated,
and when he resigned in April 1894, 'Abbas was unable to resist
Cromer's nomination of Nubar Bughus. An attempt by the
khedive to dismiss Nubar in February 1895 was frustrated by an
almost open threat of deposition. By this time even the most
high-handed British action would have carried no serious dip-
lomatic penalty. French opposition was already so strong that
(short of war, which was not seriously contemplated) it could
hardly be stronger; other powers, including Russia, now saw in
the occupation their best guarantee against an Anglo-French
entente. When in April 1895 Nubar resigned on (genuine)
grounds of ill-health, the total defeat of'Abbas was demonstrated
by the return of the pliant Mustafa Fahml as prime minister —
a position which he continued to hold until his death in
1908.
20
F . O . 78/4517, Cromer t o Rosebery, I 9 j a n . 1893 : cited al-Sayyid, Egypt and Cromer,
110.
21
F.O. 141/297, Cromer to Rosebery, 30 Jan. 1893, cited ibid., 11 j .
623

In March 1896 Cromer was taken aback by London's sudden

decision to advance in the Sudan. In 1890 he had believed the
Mahdist state to be crumbling; but in 1891 he had been
unpleasantly surprised by the strength of Mahdist resistance at
Tokar on the Red Sea coast. Thereafter he had consistently
opposed any early attempt at reconquest. Lord Salisbury's
decision of March 1896 was essentially a move in his European
policy, intended to achieve closer relations with the Triple
Alliance and Germany by a gesture of assistance to Italy,
supposedly under dangerous Mahdist pressure in Eritrea after her
shattering defeat by Menelik of Ethiopia at Adowa. The advance
was planned as no more than a gesture, a ' demonstration' in the
Nubian desert and a thirty-mile advance up the Nile to an
objective of no value or importance. The published objective of
Dongola was a mere bluff, intended to alarm the Mahdists and
distract them from Eritrea. Cromer protested that ' native anglo-
phobe opposition', including the khedive, would denounce the
operation unless territory of some value were recovered for
Egypt; any open expression of the khedive's disapproval would
have a disastrous effect on the Egyptian troops. In June 1896, after
a sustained struggle, Cromer extorted from Salisbury permission
for Kitchener to advance to Dongola.
Dongola was taken in September 1896; Kitchener was then
permitted to advance as far as he could without incurring any
military risk. His advance was brought to a halt at Berber in
October 1897. There was at first no question of his resuming the
advance with British reinforcements. Cromer and Salisbury were
agreed that 'no sufficiently important British interest is involved
to justify the loss of life and money'. 22 British troops were
nevertheless committed in January 1898 because it was believed,
quite erroneously, that Kitchener's Berber position had become
so insecure that he risked sharing the fate of Hicks and Gordon.
The political consequences — international, Egyptian, domestic —
of yet another 'Sudan disaster' were obviously unacceptable. In
June 1898 the presence of British troops enabled Salisbury to
extinguish Egyptian rights in the Sudan arising out of previous
possession: he now asserted a joint British and Egyptian right of
conquest over a sovereign 'Mahdi State' extending 'from Haifa
22
Salisbury Papers, Cabinet memoranda, Cromer to Salisbury, 20 Oct. 1897.
624

23
to Wadelai'. In January 1899 he forced upon Egypt a Sudan
settlement whereby the Egyptian 'right of conquest' was trans-
formed into the liability to provide the military and financial
resources which enabled the British to rule the Sudan. This was
unforgivable, even to Egyptians who did not take a wholly
negative view of the British presence in Egypt itself.
The British take-over in the Sudan was facilitated by the
disarray of the Egyptian opposition. After his defeat by Cromer
in 1894—5, 'Abbas relied heavily on support from France. He
sponsored the propaganda in France of the young nationalist
orator and journalist Mustafa Kamil (1874-1908), who held a law
degree from Toulouse. Mustafa's literary and social success in
certain anglophobe French circles led to dangerous delusions.
There were indeed French publicists who outdid any Egyptian
nationalist in indiscriminate denigration of the British in Egypt;24
but no serious French statesman was willing to put his policy 'a
la remorque des cafes indigenes du Caire'. The collapse of the
French position in the Nile valley in 1898—9, followed by the
Anglo-French entente of 1904, left the khedive politically bank-
rupt. He was also disconcerted by his increasing inability to
manipulate the young nationalists whom he had covertly
encouraged in the later 1890s. Admitting defeat, he moved
towards a rapprochement with Cromer, thereby further alienating
the nationalists.
Of these the most prominent, until his death in 1908, was the
khedive's former protege, Mustafa Kamil. An orator and publicist
of genius, Mustafi's intensely emotional' patriotism of the heart'
moved Egyptians deeply; there were overtones of stiff language
in his call for the ' self-annihilation' of the patriot in love of his
country. In 1900 he founded the newspaper al-Liwd' ('The
Standard') which rapidly achieved the relatively enormous
circulation of 10,000 copies. His was the first influential voice
openly to attack British policy in the Sudan and to proclaim the
'unity of the Nile valley'. His organisation in 1906 of a student
strike at the School of Law inaugurated half a century of militant
student politics. His following was above all among the increasing
number of educated, or at least literate, Egyptians who had been
23
F.O. 78/5050, Salisbury to Cromer, tcl. no. 47, 3 June 1898.
24
E.g. E. Chesnel, Plaits d''Egyptt: Its Anglais dans la vallet du N/7(Paris, i888);Octave
Borelli in Le Rosphort igyptitn.
625

exposed to, but not deeply influenced by, western ideas. For this
constituency, still fundamentally Islamic in its modes of thought,
there was no inconsistency in Mustafa's combination of fervent
Egyptian nationalism with support for the Ottoman Empire as
the symbol of Islamic solidarity.
In the 'Taba Incident' of 1906 — a Turco-Egyptian frontier
dispute in the Sinai peninsula - Mustafa Kamil strongly supported
the sultan, as a legitimate Muslim ruler, against an Egyptian
government upheld by the armed force of infidels. Cromer was
alarmed by this resurgence of' pan-Islamic fanaticism'; this and
other symptoms of 'unrest', including the Law School strike,
seem to have convinced British official opinion that drastic action
was needed to quell the 'insubordinate spirit' of Egyptians. A
fracas between falldbin and pigeon-shooting British officers at the
delta village of Danishway (Dinshawai) led to the more or less
accidental death of one officer and the injury of several others,
and became the occasion for a display of judicial terror. After a
scandalously summary trial before a special tribunal, four sentences
of hanging, and fourteen of flogging, were carried out in public
at Danishway. To all politically-conscious Egyptians, Danishway
was profoundly shocking and humiliating: Mustafa Kamil
thought that it did more to arouse anti-British feeling than ten
years of agitation.
After Danishway, nationalism ceased to be the virtual monopoly
of Mustafa Kamil. It was no longer possible for any Egyptian
group overtly to support the occupation. The well-to-do and the
well-educated also became nationalists, in principle at least.
Disillusioned with the khedive, and uneasy at Mustafa's dema-
gogic style, they formed an independent group whose principal
ideologist was the scholar-journalist, Ahmad Lutfl al-Sayyid.
Ahmad Lutfl had in the mid-1890s been a colleague of Mustafa
Kamil in the clandestine nationalist circle sponsored by 'Abbas
Hilml; but his dislike of the khedive's autocratic tendencies soon
led him to break with this group. Ahmad Lutfl and his friends
were strongly influenced by Muhammad 'Abduh, who since his
return to Egypt in 1888 had begun to see some at least of Cromer's
reforms as helping to create the 'civic virtue' without which
Egyptians could never liberate themselves. Ahmad Lutfl took a
similar view, and he and his friends were often called 'the Imam's
party'; but Ahmad Lutfl's political thought, as opposed to his
626

political practice, was in an Islamic society far more radical than

that of 'Abduh or even of Mustafa Kamil. He did not see Islam,
even in Muhammad 'Abduh's ideal revitalised form, as either a
necessary or a sufficient basis for a just polity. Islam did not give
Muslims any privileged access to the principles of political justice,
which were similar in all civilised societies and could be discovered
by empirical investigation. Ahmad Lutfl concluded from his own
investigations that the foundations of a healthy political life were
freedom, interpreted as the absence of arbitrary rule; and the rule
of law — in the western rather than the Islamic sense. In Egypt,
arbitrary rule had bred servility and stifled initiative; Egyptians
were alienated from government, yet excessively dependent upon
it. He attacked British control not because it was alien or infidel,
but because an autocratic regime based on armed force could only
aggravate the political diseases from which Egypt suffered.
Ahmad Lutfl saw Mustafa Kamil's pan-Islamic Ottomanism as
the negation of true nationalism. His opposition to Mustafa's
policy during the ' Taba Incident' induced him in 1907 to found
a newspaper {al-Jartda - ' The Newspaper') and to organise his
political connection as the Hi%b al-Umma (People's Party), to
which Mustafa promptly riposted by organising the Hi^b al-Watani
(National Party). In spite of Ahmad Lutfl's radically' modernising'
principles, his political practice was profoundly unrevolutionary.
Mustafa Kamil called for mass action, and is the father of popular
militance in modern Egyptian politics. Ahmad Lutfl called for
cautious constitutional advance under the guidance of property
and education. It is not surprising that the 'list of [al-Jarida's]
shareholders read like a Who's Who of the notables of Egypt'. 25
The impact of the British occupation under Cromer is some-
times discussed in terms of ' modernisation'. This seems to pay
Cromer an undeserved compliment. Politically, the occupation at
first halted, then stunted and distorted, development of any kind.
The economy expanded, but did not develop. Legal and
educational modernisation long pre-dated the British occupation,
and was not significantly accelerated by British control. Cromer
of course did not see himself as ' developing' or ' modernising'
Egypt. To Cromer, Egypt was simply a corrupt and inefficient
despotism, recently threatened by anarchy, whose stability
happened to be important to Britain. So long as early evacuation
25
al-Sayyid, Egypt and Cromer, 168.
627

was seriously contemplated, reforms were necessary so that

evacuation could take place without an immediate relapse into
anarchy. Later, reforms were still necessary to allay material
discontents and inhibit the growth of disaffection. Hence the
crucial importance of hydraulic improvements. Cromer believed
that, in the short run at least, the irrigation engineer could do more
than the judge or the administrator to gain Egyptian goodwill.
In 1885, at a time of desperate financial stringency, he earmarked
a million pounds for irrigation. Technological improvement
especially in hydraulics, and administrative improvement
especially in taxation, were the most important and positive
aspects of modernisation.
Cromer ultimately gave cautious encouragement to Ahmad
Lutfi's Ht^b al-Umma, mainly as a counterpoise to Mustafa
Kamil's more radical opposition; but he remained sceptical of the
possibility of political modernisation. In his view 'Islamism as a
social and political system [was] moribund'. 26 It could therefore
no more give rise to a genuine nationalism than it could generate
'civilised principles' of public administration. What it could
generate was 'fanatical' pan-Islamic movements, which were
dangerous and had to be suppressed. But nationalism was a mere
sham, a front for the machinations of 'Abbas and the ambitions
of self-seeking poseurs. Cromer was almost as hostile to political
modernisation as the most reactionary 'Turk'. Nor did Cromer's
regime provide any training in administrative responsibility for
Egyptians. The early system of British 'advisers' acting through
Egyptian agents was steadily replaced by direct British bureau-
cratic control; the British bureaucracy increased from 170 in 1883
to 662 in 1906. Expansion was coupled with a decline in quality,
or at least in experience. Many of the earlier British officials had
been distinguished members of the Indian Civil Service; by 1906
youths not long down from university were being placed over
experienced Egyptians.
During Cromer's period access to western-style education
remained very limited. In 1910, with the population of Egypt at
about 11.5 million, there were still fewer than 60,000 pupils in
western-type schools. And of these less than 20 per cent were in
government schools; the remainder were in mission, community
and other 'private' schools. The government system was essen-
26
Cromer, Modern Egypt, 727.
628

CROMER'S EGYPT, I 883—1907
tially a machine for producing junior bureaucrats. The official title
of the department - Public Instruction - reflects its objectives and
ethos. Its British teachers were mostly non-graduates, poorly paid
and regarded as social inferiors by other British officials. Its
director, Douglas Dunlop, became a by-word for authoritarianism,
mechanical rigidity and lack of educational vision. Quantitatively,
however, the neglect of education was not quite so complete as
is sometimes supposed. Between 1881 and 1907 educational
expenditure increased five-fold, and from under one per cent to
2.7 per cent of the budget. Between 1890 and 1910 enrolment in
all government schools doubled; secondary enrolment trebled.
But with only 11,000 pupils in all, government education was
hardly a major modernising force; and girls' education was left
almost entirely to the private schools. Nevertheless, the cumulative
effect of westernised education did lead to social change. Its
products, touched but not deeply influenced by western ideas, and
employed above all in the middle and lower bureaucracy, became
a new social group, whose political consciousness and aspirations
gave it great importance in spite of its small absolute size.
Cromer's policy of economic laissez-faire, combined with
technological improvement and greater administrative efficiency,
gave free play to the economic forces that had generated the
export-dominated economy and made it more dependent than
ever on cotton. From the later 1890s, high agricultural prices and
reduction of the land tax inhibited urbanisation and industrialisa-
tion, and would probably have done so even without Cromer's
rigid refusal to protect 'infant industries'. Thanks to the boom
in cotton prices and the rise in population, Egypt was by 1907
sometimes a net importer, rather than an exporter, of cereals and
pulses: the contribution of cotton to exports rose from 76 per cent
to 93 per cent between 1880 and 1910.
The resident foreigners, whose numbers had doubled from
about 70,000 to over 150,000, still dominated the commercial and
financial superstructure of the economy. Out of a hundred
directors of companies listed in 1901, only fourteen can be
identified as certainly Muslim and probably Egyptian. Foreign
ownership of land had increased from under 5 per cent in 1884
to 12 per cent in 1907. Otherwise, the general pattern of
land-ownership showed little change. British attempts to create
a class of prosperous smallholders, by (for example) providing
629

small loans on easy terms, were necessarily stultified by population

increase and total respect for the existing distribution of property.
The overall share of the falldhtn had indeed increased (but at the
expense of ' medium', not of large, holdings) from under 20 per
cent to 25 per cent by 1910; but the increase in population had
halved the average falldh holding to scarcely more than one
feddan. Although down to 1907 the proportion of totally landless
falldhtn does not seem to have much increased, the proportion
holding sufficient land to support a family had now shrunk to well
below 10 per cent.
In spite of continuing heavy debt service, which more than
cancelled out the very favourable balance of trade, Egypt was a
prosperous society in the early 1900s. The sustained rise of
agricultural prices, above all of cotton, coincided with the effects
of technological and administrative improvement. Between 1884
and 1898, greater efficiency in the collection of the land-tax
enabled government to reduce the amount collected per feddan
by nearly 20 per cent without damage to the revenue. Even
Mustafa Kamil, warning Egyptians of the weight of 'golden
chains', 27 had to admit the material benefits of British control. But
these benefits were very unevenly distributed. The falldh, unless
he was completely landless, got his tiny share; his lot was
improved by the gradual abolition of the corvee; his life was
certainly less hard — and less precarious - than in the last appalling
years of Isma'fl's reign. The main Egyptian beneficiaries were of
course the rich and well-to-do notables. In spite of Ahmad Lutfi's
hatred of autocracy, so long as prosperity lasted the Hi%b al-Umma
was less than militant in opposing the occupation.
The Egypt Cromer left on his retirement in 1907 was still
essentially Isma'Il's Egypt, managed more efficiently and - on the
whole — more humanely. British 'modernisation' was largely
confined to technology and the all-important fiscal administration;
its not unworthy monuments are the Aswan dam and the great
cadastral survey of Egypt. Otherwise, Cromer was a more
cautious moderniser than Isma'Il. ' Modernisation' had no impact
whatever on the attitudes and mentality of the masses, and quite
failed to penetrate the ' closed world' of the village community.
The sole significant social change, the emergence of the ' western-
" Mustafa Kamil, open letter to Campbell-Bannerman, Le Figaro, 14 Sept. 1907:
cited al-Sayyid, Egypt and Cromer, 161.
630

MAHDIST STATE UNDER KHALIFA ABDULLAHI
educated', was certainly unwelcome to Cromer; his policies

tended, perhaps consciously, to inhibit it by feeble educational
expansion and the imposition of school fees. In the early 1880s
the British had in effect filled the gap created by the collapse of
an old ruling elite. But the rapid development of a quasi-colonial
British administration frustrated the development of any new
ruling elite, above all by depriving its potential members of any
significant political or administrative responsibility.
THE MAHDIST STATE UNDER KHALIFA 'ABDALLAHl,

1885-1898
The revolutionary phase of the Mahdiyya ended with the Mahdi's
death in June 1885. His successor, Khalifa 'Abdallahi, created a
new established order in the very process of securing and
consolidating his own succession. 'Abdallahi's authority was
challenged not only by a dissident Mahdist faction who wished
to replace it by their own; but also by those groups who had, like
the Baqqara, seen the Mahdiyya as a protest against all ' govern-
ment'. He responded by the creation of a strongly centralised
autocracy, supported by an elaborate bureaucracy employing the
techniques, and sometimes even the personnel, of its Turco-
Egyptian predecessor. No less importantly, it was supported by
an army which was no longer an amalgam of religious visionaries
and tribesmen hungry for loot, but an increasingly specialised and
professional defence force for a territorial state, and an instrument
of 'Abdallahi's autocracy. Its senior officers ultimately wore the
Mahdist patched jubba, the symbol of holy poverty, in an
elaborately decorated and stylised form - a hybrid of military
uniform and ecclesiastical vestment.
After 1889, offensive jihad was tacitly abandoned for the
defence of' Mahdism in one country'. Indeed, a defensive entente
was ultimately reached with Christian Ethiopia. Its existence was
widely known, but there is no evidence of' fanatical' opposition
to it. Officially, external enemies of the state were still of course
' the enemies of God'; but by the 1890s they were, to many Muslim
Sudanese, also the alien assailants of an indigenous regime which
for all its shortcomings had achieved a considerable degree of
acceptance, at least as a lesser evil than conquest and domination
by outsiders. The Mahdist state claimed an allegiance which
631

supetseded that of tribe or region; in his later years the khalifa
had striven, not entirely without success, to hold the balance
evenly between rival tribal groups. It is in the shared service and
defence of the Mahdist state, rather than in the propaganda of the
Mahdi himself, that the origins of modern Sudanese nationalism
may perhaps be sought.
At the Mahdi's death the great offices of state and the provincial
commands were held mainly by the awlad al-balad, Sudanese of
riverain origin who included the Mahdi's kinsmen, the ashraf.
These groups hoped to secure the succession for the Mahdi's
young kinsman and fourth khalifa, Muhammad Sharif. But since
the proclamation of January 1883 'Abdallahi had been pre-eminent
among the ansdr; moreover the Mahdi had just before his death
publicly denounced and disowned the ashraf as incorrigible
worldlings. The pietists in the Mahdist inner circle took the lead
in acknowledging the succession of 'Abdallahi; the ashraf
reluctantly followed their example. Armed resistance to 'Abdal-
lahi's successison was impossible. Troops loyal to him garrisoned
Omdurman; those of the awlad al-balad were widely dispersed.
'Abdallahi's authority as the ' guidance ' 28 of the movement was
authenticated by his mystical colloquies with the Mahdi. He was
also able to secure the allegiance of the army of the west and
replace its commander, the Mahdi's cousin, Muhammad Khalid
Zuqal, by his own client, Hamdan Abu 'Anja. The junior khalifas
were then compelled to surrender their military resources to
'Abdallahi's Black Flag regiment, and most of the Mahdi's
province governors were replaced by 'Abdallahi's nominees.
Down to 1889 'Abdallahi attempted to maintain the Mahdist
jihad against Egypt and, with some reservations, against Ethiopia.
Quite apart from his own fundamental Mahdist convictions,
successful jihad would distract the ansdr from their internal
dissensions and would be the strongest possible validation of
'Abdallahi's authority as Khalifa of the Mahdi. By November 1886
a Mahdist expeditionary force under 'Abd al-Rahman al-Nujuml
had been assembled at Dongola, and early in 1887 indhdrdt or
letters of warning were dispatched to the sultan, the khedive and
Queen Victoria. But active jihad against Egypt was delayed until
May 1889 by problems on the western and eastern frontiers of the
Mahdist state.
28
Hidya, bidaya: from the same root as the word mahdi; cf. Holt, Mahdist State, i 59-40.
6}2

MAHDIST STATE UNDER KHALIFA 'ABDULLAHI
The problems on the western frontier were essentially a

continuation of those experienced by the Egyptians. The Fur, and
those Baqqara who had remained in their homeland, resisted the
new government as they had resisted the old. By 1888, however,
this 'tribal' resistance had been suppressed. Among the most
troublesome of the Baqqara were the Khalifa's own kinsmen, the
Ta'alsha. Immediately after the crushing of the Fur, 'Abdallahi
ordered the Ta'alsha to migrate en masse to Omdurman, on pain
of harrying by the Mahdist army. In May 1888 they bowed to this
ultimatum; early in 1889 they were beginning to reach the
metropolitan area, where they were settled in Omdurman and the
northern Gezira, with important consequences for the develop-
ment of the Mahdist state. Other dissident or suspect nomad
groups were treated more harshly; the Kabablsh of northern
Kordofan were harried and broken in 1887. By mid-1888 tribal
dissidence had been crushed virtually throughout the Muslim
Sudan; but in Darfur the Mahdists were now being challenged
by a messianic religious leader, known as Abu Jummayza, who
claimed to be the Mahdi's rightful third khaltfa. He gained a large
and enthusiastic following of Fur and Baqqara dissidents. By the
end of 1888 he had driven the Mahdists from western Darfur; but
the movement collapsed when the rebel army was routed outside
El Fasher in February 1889.
The jihad against Ethiopia was constrained by respect for
Ethiopian military strength, most recently demonstrated by the
victories of Yohannes IV over the 'Turks' in 1875 and 1876. In
June 1884 Egypt had agreed to retrocede the Bogos (Keren)
region to Yohannes in return for his assistance in extricating the
Egyptian garrisons in the eastern Sudan. These operations led to
incidental clashes between Ethiopian troops and the Mahdists;
but Yohannes recognised the Mahdist governor of Gallabat and
attempted to establish diplomatic relations with the Mahdi. The
Mahdi replied in the indbara already cited (above, p. 609); but the
Mahdi's death and the succession-crisis of 'Abdallahi limited
immediate Mahdist action to small-scale raiding, or at most to the
penetration of debatable borderlands. In January 1887 Negus
Takla Haymanot, the ruler of Gojjam, drove the Mahdists from
Gallabat, plundered the surrounding country, and retired.
'Abdallahi despatched an army to re-occupy Gallabat; but in
February 1887 he sent to Yohannes a letter which, though an
635

indhdra in form, was in substance an offer of peace on the frontier

provided the Ethiopians ceased to harbour defectors from
Mahdism. Receiving no reply, in October he posted to the frontier
reinforcements under his ablest general, Hamdan Abu 'Anja.
Religious dissidence in the Mahdist army delayed the offensive
until January 1888. Jihad was then formally proclaimed in
Omdurman: the Khalifa announced a vision in which the Prophet
himself had promised victory over the Ethiopians. The jihad was
successful as a large-scale raid rather than as a war of conquest;
the Mahdists took and sacked Gondar and then retired with their
booty. The presence in the Lake Tana region until May 1888 of
a large army under Menelik of Shoa deterred the Mahdists from
further raiding until June. A second raid then ravaged the Gondar
region: it withdrew in August as the rains brought disease to the
Mahdist forces.
In December 1888 Yohannes addressed to Hamdan Abu 'Anja
an eloquent plea for peace and co-operation against' our common
enemies the Europeans'. Abu 'Anja replied, probably without
consulting the Khalifa, in provocative and insulting terms.29
Yohannes thereupon decided to eliminate the Mahdist danger
before moving to a major internal confrontation with Menelik
(below, pp. 654—5). The Mahdists made no offensive move, but
strengthened their fortifications at Gallabat. On 9 March 1889 the
Ethiopians stormed and took Gallabat, but victory was turned
into defeat by Yohannes' death. This providential victory over
the dreaded Ethiopians was greeted with relief and rejoicing at
Omdurman, but 'Abdallahi did not resume the jihad. Instead he
reduced the Gallabat garrison and permitted the development of
an 'undeclared peace' on the frontier, when Menelik, Yohannes'
successor, made it clear that he wished to avoid conflict.
The jihad against Ethiopia was a complex phenomenon. It was
in part a continuation of the struggle for debatable borderlands
which had begun in Muhammad 'All's time. Some Mahdist
commanders, especially Baqqara, seem to have seen in it little
more than a sanctified version of traditional raiding. Even the
major invasions of 1888 look more like ghavyva than jihad, apart
29
C.R.O. Khartoum, Mahdia 1/34/16, Yohannes to Abu 'Anja, 17 Kiyahk
( = Tahsas) 1881 (E.C.)/*5 Dec. 1888. Mahdia 1/55/13. Abu 'Anja to Yohannes,
Jumada I 1306/Jan. 1889: cited S. Rubenson, The survival of Ethiopian independence
(London, Stockholm and Addis Ababa, 1976), 384. There was evidently hardly time
for Abu 'Anja to consult the Khalifa; moreover, the style of his reply is very different
from that of 'Abdallahi's chancery.
634

from the Mahdist zeal in destroying churches. 'Abdallahi was

evidently reluctant to engage in full-scale war with so formidable
an enemy. In February 1887 he had quoted to Yohannes the hadith:
'Avoid the Abyssinians so long as they avoid you'. 30 The
Mahdists always retreated, or remained on the defensive, when
confronted by the armies of Yohannes or Menelik; and 'Abdallahi
used his victory at Gallabat simply to escape from the Ethiopian
adventure without loss of reputation.
Gallabat was followed by the long-delayed jihad against Egypt.
An advance by al-Nujuml's force was indeed by now the only
alternative to its withdrawal from the Dongola region, where the
low Nile of 1888 had brought famine, and the starving troops were
plundering the local inhabitants. In May 1889 the advance into
Egypt began, without rations and along a route remote from the
Nile. The Mahdists quite failed to find the support they had
expected in Egypt. On 3 August the expedition was virtually
annihilated by an 'Anglo-Egyptian' force at Tushkl: over 5,000
Mahdists were taken prisoner or gave themselves up as refugees.
The Khalifa responded by an attempt to intensify the jihad on the
Suakin front. But the Beja now lacked their old offensive spirit,
and 'Uthman Diqna failed to revive it. In February 1891 an
Anglo-Egyptian force took Tokar, on the coastal plain south of
Suakin. Thereafter, jihad against Egypt was confined to hit-
and-run raids on isolated oases and desert outposts.
After the military disasters of 1889-91, jihad was tacitly
abandoned in favour of the territorial defence of 'Mahdism in
one country'. 'Abdallahi attempted to foster internal solidarity by
making his rule more generally acceptable to his subjects. In this
he was initially frustrated by the fatal coincidence of his' calling-in'
of the Baqqara with the catastrophic famine of 1889—90. Security
demanded that the Baqqara should have the first call on supplies:
the commissioner of the bayt al-mdl was imprisoned and executed
when he showed insufficient zeal in stripping the famine-stricken
Gezira of corn to feed them.' Our Lords the Ta'alsha', as the awldd
al-balad sardonically called them, were both feared and despised
as violent and uncouth barbarians. Yet, as a power-base for the
Khalifa, they were far from satisfactory. Undisciplined and
unreliable, in a crisis they were more likely to devote themselves
30
'Abdallahi to Yohannes, Jumada I i3O4/Jan.-Feb. 1887. Translation (with an
erroneous Gregorian date) in F. R. Wingate, Mabdism and the Egyptian Sudan (London,
l8
9 ' ) , 33*-}-
635

to plunder than to give 'Abdallahi effective support. In 189c

the Khalifa attempted to conciliate the awldd al-balad by reinstating
their leaders as province governors in the north, and by
encouraging agriculture and foreign trade. In November 1891,
however, the ashrdf attempted a coup against 'Abdallahi. In
Omdurman, their leaders openly defied the Khalifa's authority.
The Khalifa wished to avoid an armed conflict in which the
Baqqara might get out of control. 'All wad Hilu, the leader of
the pietist group, in his usual role of mediator and peace-maker,
persuaded the ashrdf to surrender their arms in return for the
political and military reinstatement of the khalifa Muhammad
Sharif. Once the immediate crisis was over, 'Abdallahi invoked
the Mahdi's authority, vouchsafed in a vision, to break the terms
of this agreement, and to execute or imprison suspect notables.
Once again the awlad al-balad were removed from high office;
but the Khalifa persisted in a conciliatory policy towards
merchants and cultivators, and his public speeches emphasised
peace and agricultural production. In spite of heavy taxation, the
inevitable result of bureaucratic elaboration and an enormous
unproductive army, the Khalifa's rule was by now widely
accepted, even by the awlad al-balad. Refugees from Egypt
returned to the Sudan. Moreover, at all levels except the very
highest, the awlad al-balad continued to play a major part in
administration. Indeed, without their clerical and technical skills
the Mahdist state could not have operated. A Ja'all, Muddathir
Ibrahim al-Hajjaz, served the Khalifa from first to last as con-
fidential secretary and seal-bearer. Aw lad al-balad, and domiciled
Egyptians, were prominent in the administration of the general
and specialised treasuries, and as clerks to the major military
formations. The regime was comparatively broadly based; the
ordinary cultivator was not molested and was reasonably content
with his lot. Acquiescence was widespread; and some Sudanese
are reported as feeling that an attempted reconquest would be ' an
interference with their independence'.31
The Khalifa's supreme authority and his independence of
specific tribal support were emphasised by his increasing isolation
from his subjects. He now attended the great Friday 'mosque
parade', a major symbol of the militant Mahdist state, only four
11
Intelligence reports, Egypt, no. 9 (1892), statement by Mustafa al-Amln. Cf. Holt,
Mahdist state, 203-5.
636

times a year. The Baqqara, as the main military instrument of his
autocracy, were superseded by the muld%imiyya, which from a
group of' orderlies' had grown by 1895 to a corps d'elite of 9,000
men, a judicious amalgam of jihadiyya, Baqqara and Ja'aliyyun,
financed from its own special treasury. This force, from which the
Khalifa's bodyguard was drawn, was stationed close to his
residence and to the major offices of state; and in 1893 the whole
complex was surrounded by a great wall. 'Abdallahi evidently
aspired to found a dynasty. His brother Ya'qub became his closest
confidant; his son 'Uthman was groomed for the succession and
dignified with the honorific Shaykh al-Dm (' Elder of the Faith').
In the southern Sudan the Mahdiyya is important not for any
positive achievement but for the consequences of its failure to
impose any effective overrule after the collapse of the Egyptian
administration. The Bahr al-Ghazal, conquered in 1884, was lost
in 1886 when the Mahdist army was called away to deal with
insurgents in Darfur. The Mahdists were entirely unsuccessful in
their attempts to re-establish themselves in the Bahr al-Ghazal
between 1893 and 1895. Their forces melted away from disease
and constant skirmishes with southern Sudanese: one Mahdist
expedition was virtually annihilated by the Dinka. Equatoria,
though isolated from Khartoum since 1883, remained in the hands
of its governor, Emin (Eduard Schnitzer), until 1888. A Mahdist
invasion from Bahr al-Ghazal in 1884-5 w a s frustrated by the
vigorous resistance of the garrison and a mutiny of the Mahdist
jihadtyya. It was H. M. Stanley's ' Emin Pasha relief expedition',
rather than Mahdist pressure, which ultimately destroyed Emin's
authority in Equatoria and sowed active dissension among his
troops. 'Umar Salih took Rejaf for the Khalifa in October 1888,
but was forced to retire to Bor in 1891 by southern resistance.
Southern Sudanese were by now no less hostile to a Mahdist
'government' than they had been to the Egyptian administration.
As 'Umar Salih told the Khalifa: 'This religion of ours is very
difficult for them to understand and follow, and so they desert
us'."
In 1892-3 a Belgian expedition from the Congo, with the
co-operation of a remnant of Emin's troops, precariously occupied
some of the former Egyptian posts in western Equatoria. It, too,
" C.R.O. Khartoum, Mahdia 1/54/58, 'Umar Salih to the Khalifa, Muharram
ijO7/Sept. 1889: cited R. O. Collins, The southern Sudan, rtSj-rfyS (New Haven and
London, 1962), 77.
637

was forced to retire by local resistance, especially that of the

Azande. In 1894 the able Mahdist commander, ' Arabl Dafa'allah,
who had now taken over in Equatoria, destroyed the remaining
Egyptian troops. He then followed the retreating Belgians over
the Nile—Congo divide, and until the end of 1894 gravely
threatened the whole Belgian position in the Uele valley. He then
fell back on Rejaf, now almost completely isolated from the north
by the closing of the Bahr al-Jabal sudd. In February 1897 his
dwindling garrison was driven from Rejaf by the Belgians, who
had by now overcome Zande resistance. He retired to Bor, from
where he continued indomitably to harass the Belgians, nearly
succeeding in recovering Rejaf in June 1898.
In spite of 'Arabl Dafa'allah's skill and valour, the Equatoria
Mahdists remained essentially a beleaguered garrison, capable
sometimes of major offensive sorties, but not of government or
'settled administration'. The garrison had to raid for its supplies;
foraging parties were usually harassed, and sometimes destroyed,
by local resistance. Like the intermittent Mahdist incursions into
the Bahr al-Ghazal, its presence intensified, and was quite unable
to control or even moderate, the extremely violent local com-
petition for power and resources unleashed by the collapse of the
Egyptian administration. Firearms, in the hands of jihddtyya
deserters and Zande irregulars armed by the Belgians, cata-
strophically disrupted the local 'balance of power' in many
regions. Some of the smaller peoples were reduced to a mere
handful of dispirited survivors. Other groups, displaced by the
aggression of the Azande and the Nuer, struggled desperately to
recover their vital lost ground.
One important casualty of this violence was the development,
sometimes even the survival, of the Arab and Islamic influences
discernible in certain southern regions in the 1870s. The most
powerful islamising agency had been the Egyptian garrison,
consisting mainly of locally recruited black troops who became
Muslims as a matter of military routine and who of course married
locally. The Mahdist garrison, as 'Umar Salih pointed out, was
quite ineffective, indeed counter-productive, as an instrument of
islamisation. Moreover, in 1880 Muslims had been the effective
power-holders over wide areas of the south; thereafter Muslim
political control was confined to the immediate vicinity of the
Mahdist garrison, ultimately reduced to the single isolated station
638
A
at Bor. Except in the north-western Bahr al-Ghazal, where the

Feroge (Faruql) and other groups had been under Islamic
influence from Darfur for two generations, the Mahdiyya saw
a sharp set-back to the expansion of Islam in the southern
Sudan.
By the mid-1890s the Mahdist state was under considerable
European pressure. Defeat at Italian hands at Agordat in
December 1893, followed by the loss of Kassala to the Italians
in June 1894, caused alarm and foreboding at Omdurman. The
Anglo-Egyptian advance of March 1896 was perhaps less of a
surprise to the Khalifa than it was to Cromer. Once Kitchener
had been authorised to advance to Dongola, he had little difficulty
in doing so; shortage of supplies strictly limited the strength of
the Mahdist forces on the northern frontier. Early in 1897, as
Kitchener continued his advance up the Nile and began his
railway across the Nubian desert, the Khalifa moved to Omdurman
the army of the west under Mahmud Ahmad; and he began to
respond less coldly to the proposals of Menelik II of Ethiopia for
a formal peace on the frontier and co-operation against Europeans.
Ethiopian diplomatic missions were honourably received in
Omdurman; and in 1898 'Abdallahi ceded to Menelik the gold-
bearing BanI Shanqul (Bela Shangul) region, which had however
already shaken off Mahdist control. In June 1897 the Khalifa
stationed Mahmud Ahmad's army at Metemma, the capital of the
Ja'aliyyun. The Ja'aliyyun resisted; their revolt was crushed in
blood, the last and most disastrous clash between 'Abdallahi and
the awldd al-balad.
At Metemma, the morale and discipline of the Mahdist army
was eroded by an increasingly severe shortage of food. Mahmud
Ahmad's lack of decision and initiative as a commander lost the
Mahdists Abu Hamad and Berber, which fell to Kitchener in
August and September 1897. In December, 'Abdallahi mobilised
the central Mahdist army at Omdurman. He may have intended
to despatch it to the north; if so, he was deterred from doing so
by reports of Mahmud Ahmad's desperate supply situation. But
this mobilisation caused the utmost alarm in Cairo and Berber,
where Kitchener was doubtful of his ability to handle even
Mahmiid's force. The hasty commitment in January 1898 of
British troops, scarce resources which could not be retained
indefinitely in a merely defensive role, prescribed a new objective
639

for Kitchener's campaign - the early and total destruction of the

Mahdist state.
In January 1898 Mahmud Ahmad rejected the Khalifa's
proposal for withdrawal to a defensive position at the sixth
cataract, and was permitted to advance against Kitchener.
Mahmud's famine-stricken army reached the Atbara in March;
after considerable hesitation, Kitchener attacked and destroyed it
on 8 April. The final advance upon Omdurman was then delayed
until August, when the Nile flood permitted Kitchener to pass
his gunboats and river transport through the sixth cataract. On
2 September 1898 the main Mahdist army was destroyed outside
Omdurman. The battle was a triumph for technology over
heroism: the ansar were martyred in thousands as they repeatedly
strove, with superb courage and devotion, to pierce Kitchener's
lethal curtain of musketry and fight at close quarters.
THE BIRTH OF THE ANGLO-EGYPTIAN SUDAN, 1898-1907
After Omdurman, the victors had to suppress the terminal

resistance of Ahmad Fadll on the Blue Nile and of the Khalifa
himself in Kordofan. These operations were not complete until
November 1899; meanwhile it was necessary to occupy and
garrison very large tracts of territory, and to open up and maintain
lines of communication and supply. Until 1900 civil administration
was merely incidental to military operations; it was a part-time
responsibility of the Egyptian army intelligence branch, as the
military agency best acquainted with the geography and politics
of the Sudan. Its finances were hardly distinguishable from the
general expenses of the occupying army. The British officers
commanding the province garrisons became mudirs or province
governors; in March 1899 Kitchener's 'memorandum to mudirs'
had prescribed a rather sketchy scheme of province and district
administration. District administration was entrusted to Egyptian
captains and subalterns with the title of ma'mur. In each province
two British officers with the title of Inspector (mufattish) were to
inspect and report on the administration of the ma'murs, who were
however directly responsible to the British mudir; inspectors were
explicitly forbidden to act as their 'channel of communication'.33
33
Kitchener's Memorandum to Mudirs, printed Muddathir 'Abd al-Rahim, Imperialism
and nationalism in the Sudan (Oxford, 1969), 2)7—40.
640

THE BIRTH OF THE ANGLO-EGYPTIAN SUDAN
And that was all. It was left to Kitchener's successor, Reginald

Wingate, to grapple with an increasingly chaotic situation and
provide an effective central organisation for the 'civil side', as it
was called.
Kitchener as governor-general was a capricious despot, con-
cerned almost exclusively with his career and personal prestige,
and often callously indifferent to the welfare of his Sudanese
subjects and indeed of his Egyptian troops. The Condominium
Agreement of January 1899 had originally been devised by
Cromer not only as a means of excluding Egyptian political
influence from the Sudan, but as a means of controlling the
governor-general from the Cairo Agency. But Kitchener exploited
his position as a national hero to remove from the draft agreement
those provisions limiting the governor-general's legislative
omnipotence and financial discretion, and then to ignore the
supplementary financial and other regulations whereby Cromer
attempted to reassert his control. Kitchener's departure for South
Africa in 1899 was greeted with relief by Cromer: and indeed by
some of his own senior officers, who feared that Kitchener's total
lack of system would lead to administrative collapse.
Cromer regained control with the appointment of Wingate,
whom he supervised very closely and treated as a mere executive
subordinate. Although Cromer professed to be concerned only
with 'big questions' of general policy, in practice he constantly
intervened in day-to-day administrative detail. The Sudan budget
was now drafted in minute detail in Cairo; and the 'financial
secretary of the Sudan', stationed in Cairo, was Cromer's agent,
not Wingate's. Yet in relation to his own subordinates, Wingate's
administrative practice was essentially autocratic. Very few of his
colleagues shared the extensive knowledge of the Sudan that
Wingate had acquired as director of military intelligence. Province
governors were seldom consulted on major questions of policy;
and Wingate trained them to keep him very fully informed. Only
Rudolf Slatin, appointed inspector-general of the Sudan in
September 1900, had any significant share in policy-making.
Slatin's Sudan experience, as governor in Darfur from 1879 t o
1884, and then as an unwilling muldîm of the Khalifa until 1895,
gave him unique knowledge and authority.
By 1900, except in the southern Sudan, there was emerging a
fairly clear distinction between military operations and civil
641

administration. Wingate's early administration was essentially that

of an improvised 'civil affairs branch' of the Egyptian army. The
'civil secretary', a British military officer responsible for the
routine functioning of this branch, though not yet for its policy,
became a key man in the system. In spite of the - very sparing
— recruitment of British civilians as administrative probationers,
British officers continued regularly to serve in civil administration.
For most of them, this work was merely a brief interlude in a
normal military career; but a few carefully selected officers were
retained on quasi-permanent secondment to the 'civil side', as a
cadre of skilled administrators holding senior appointments.
In the northern Sudan, the administration seemed by 1900 to
be taking hold. The Omdurman and Gezira Baqqara had been
repatriated to the west. All the leading Mahdist notables, including
all the Mahdi's sons save 'Abd al-Rahman, a youth of fifteen, were
either dead or in prison. Darfur, a perennial source of resistance
and 'fanaticism' which government lacked the resources to
occupy effectively, had fallen into the able hands of a reliably
anti-Mahdist sultan, 'All Dinar, who permitted no interference
with his internal administration but acknowledged the govern-
ment's suzerainty. Yet security remained fundamentally pre-
carious. The Egyptian army was not an unconditionally reliable
instrument. When Wingate arrived in Khartoum, some units were
mutinous, the result largely of Kitchener's neglect. Wingate had
to overcome this crisis by tact and diplomacy rather than by force;
his British garrison then consisted of a mere 250 men, who were
withdrawn to Egypt in summer. Not until 1902 did Wingate
obtain a permanent British garrison of about 800 men.
It was, moreover, soon evident that the destruction of the
Mahdist state had not extinguished popular Mahdism. It was
widely believed that the Mahdi's dispensation would be followed
by that of the 'Antichrist' {al-dajjat), who would be overthrown
only by the second coming of the Prophet Jesus (al-nabi 'Isa).
' False prophets' dedicated to the overthrow of al-dajjal, easily
identified as the new government, appeared almost annually
between 1900 and 1912. Most of these were quite ineffective; but
two of them, down to 1907, seem to have been a real danger to
government. In 1903 a miracle-working jaki from Bornu,
Muhammad al-Amln, built up a following at Taqall, an islamised
statelet in the still unadministered Nuba mountains. His movement
642

THE BIRTH OF THE ANGLO-EGYPTIAN SUDAN
remained, disturbingly, quite unknown to the government until

he was denounced by the qadioi Taqall. Muhammad al-AmIn was
arrested by a military patrol and summarily hanged at El Obeid.
There were fears of a general rising of the Baqqara, with whom
he had been in contact, should he 'miraculously' escape from
custody. In 1904 Adam Muhammad, a Dunqulawl of the diaspora
living at Sinja on the Blue Nile, came out openly against the
government as al-nabt 'Tsd. His movement attracted considerable
local support; but before it could gain momentum it was crushed
by the prompt action of the Egyptian ma'miir of Sinja.
The nabi'Tsdcult probably never enjoyed much support among
the more sophisticated Mahdists, who increasingly looked for
politico-religious leadership to the Mahdi's posthumous son,
'Abd al-Rahman, now growing to manhood. Meanwhile, many
former Mahdists were willing temporarily to accept the de facto
authority of a powerful government that showed respect for
Islam; some were not averse from tactical collaboration with it.
Wingate did not initiate a heresy hunt against Mahdism simply
as a set of beliefs. He was content to suppress its outward
manifestations, such as the sale, circulation and ' congregational'
use (but not the mere possession) of the Mahdi's rdtib or manual
of devotions. But Wingate found it almost impossible to assess
the real strength of militant Mahdism. The persistence of al-nabi
'Tsd movements convinced him that the bulk of the Muslim
population were clandestine militants; and that any visionary who
could gain an initial' miraculous' success against police or troops
might well ignite an uncontrollable conflagration.
Wingate's abiding conviction of fundamental insecurity pro-
foundly influenced his administrative policy. His vigilance against
'fanaticism', and his determination to crush it ruthlessly 'in its
inception', became almost obsessive. But Wingate also sought to
blunt the edge of fanaticism by ostentatious deference to
'orthodox' Islam, by gaining the goodwill oi'ulamd' and secular
notables, and by promoting the material well-being of the mass
of the Muslim population. A semi-official board of 'ulama',
constituted in 1901, advised the government on policy towards
Islam. Christian proselytisation was rigidly forbidden in the
Muslim Sudan; even in the pagan south, where Wingate was
opposed to islamisation, he was very careful to avoid any public
alignment of government policy with missionary objectives. In
643

the northern Sudan, government also supported the influence of

' tribal' shaykhs and other secular notables and offered them a role
as administrative auxiliaries. Anxious to reassert their authority
after its eclipse under the Mahdiyya, the shaykhs and notables
were usually willing to co-operate.
Very light taxation was seen as the best guarantee of a
contented populace. All military expenditure, and about half the
civil expenditure, was in the early 1900s defrayed by Egypt.
Wingate could therefore afford to be lenient where strictness
might lead to disaffection. Unpopular forms of taxation were
sometimes abandoned. In 1901 the herd tax on nomads was
replaced by a light 'tribute' - a symbol of submission to
government rather than a serious financial levy. A progressive tax
on uncultivated irrigable land was introduced in 1906, but
abandoned in 1908. Wingate was also concerned to protect
Sudanese rights in land, and to restrict the alienation of land to
foreigners. In 1905 sales of rural land to non-Sudanese were
forbidden save with the consent of province governors, which
was very sparingly granted; Wingate refused to register the title
of unauthorised purchasers. Meanwhile, scarce resources were
lavished on the investigation and registration of Sudanese freehold
(muJkf) titles. Wingate's land and taxation policies meant that
Sudanese were the principal beneficiaries of the expansion of the
economy which followed the extension of the railway to the Red
Sea and the construction of a modern harbour at Port Sudan. The
capital for these improvements, amounting to nearly £4 million
by 1907, came as interest-free loans from Egypt.
' Progress' and material improvement soon became for Wingate
a very positive commitment and not a mere strategy for ' killing
fanaticism by kindness'. Unfortunately, this commitment was
virtually confined to the northern, Muslim, Sudan. The southern
Sudan, except for its contribution to the Nile waters, was seen
as 'useless territory'. 34 British policy-makers saw an urgent need
to defend its still debatable frontiers against the encroachments
of Ethiopia and King Leopold's Congo; but only exceptionally
any need to ' waste' precious funds on the administration of its
'savage' inhabitants. Government expeditions, hastily pressing
on towards distant frontiers, burst into and further exacerbated
34
'Large tracts of useless territory which it would be difficult and costly to
administer properly' - Salisbury Papers, Cabinet memoranda, Cromer to Salisbury, 5
Nov. 1897.
644

E T H I O P I A : REIGN OF YOHANNES IV, FROM 1875
an extremely violent and confused situation of which British
officers had no knowledge or understanding. Their presence, and
their demands upon southern Sudanese, soon provoked resistance
from the more powerful and confident peoples.
The strategic imperatives of this local epilogue to the scramble
for Africa implied the immediate use of force in crushing this
resistance. But of those who resisted, only the Azande, situated
near the sensitive meeting-point of Sudan, French and Congolese
territory, were effectively conquered and, after 1905, brought
under full administrative control. An attempt was made to control
other recalcitrant groups, especially the Nilotic Dinka and Nuer,
by punitive expeditions which were not usually followed by
settled administration. This attempt to use fire and the sword as
a cheap substitute for administration was a disaster. Punitive
expeditions, though sometimes of almost genocidal severity, quite
failed to crush resistance permanently; they tended rather to
extend it and to increase its scale, with a corresponding increase
in the scale of' punitive' violence. From this vicious spiral, which
long survived the final settlement by 1906 of the southern Sudan's
international frontiers, there was no escape save by settled
administration, for which Khartoum would not pay. In 1907, and
indeed for nearly two decades thereafter,' administration' in much
of the southern Sudan was still virtually synonymous with
military operations. The peculiarities of southern administration
in this period do not of course reflect any long-term plan to
separate the south from the northern Sudan. They reflect rather
the almost complete dereliction by Khartoum of its administrative
responsibilities in the south; and its refusal to provide resources
for anything but 'punishment'.
ETHIOPIA: THE REIGN OF YOHANNES IV, FROM 1875
The military successes of Negus Yohannes IV against Egypt in

1875 and 1876, spectacular though they were, did not solve his
political problems. The Egyptians remained in control of the
Bogos (Keren) region and above all of the sea-board, where at
Massawa they obstructed Yohannes's external trade and the
import of arms. Yohannes and his great general Ras Alula - who
had risen from peasant stock by sheer military ability — continued
to harass the Egyptians in Bogos; but Ethiopian pressure was
645

X1896 AGAME
S E MIE N :.\
ondar A \
WAG
7O°A>
/ 35
"I '
Approximate frontier with Egypt

immediately after the war of
1875-6
Italian territorial claim of 1895
LASTA Provinces. regions
Oromo Peoples, ethnic groups
X Bat ties involving Ethiopian troops
100 km
50 miles
25 Northern Ethiopia under Yohannes IV and Menelik II

646

ETHIOPIA: REIGN OF YOHANNES IV, FROM 1875
never strong enough to expel them from Massawa; and

Yohannes's pleas for assistance from the 'Christian Powers' of
Europe, especially Britain, were for long quite fruitless. Inter-
mittent and indecisive negotiations continued until June 1884,
when Britain constrained Egypt to retrocede Bogos (but not
Massawa) in return for Yohannes's assistance in evacuating the
Egyptian garrisons from the eastern Sudan.
Nor did Yohannes's victories put an end to dissidence in
Ethiopia. The response to his call for holy war against the ' Turks'
had been impressive, but by no means unanimous. Some of the
northern magnates continued in active rivalry. Ras Wolde Mikael
of Hamasien defected to the Egyptians; Wolde Sellassie, the great
magnate of Begemder, whom Yohannes had made ras and
governor in 1875, wavered in his support and his son rose in open
rebellion. In the south, Menelik of Shoa, who had never dearly-
recognised Yohannes's imperial title and regarded himself as a
rival 'King of Kings', had given no more than token support
against the Egyptians. In April 1876, Yohannes deposed Wolde
Sellassie, and replaced him as governor of Begemder by Ras Adal
Tessama of Gojjam, a former opponent who now became a
reliable ally, not least because he aspired to challenge the
hegemony of Menelik in the south.
Begemder rose in rebellion on the deposition of Wolde
Sellassie. Menelik seized this opportunity to consolidate his
position in the Oromo-dominated and heavily islamised region of
Wollo, immediately to the north of Shoa. In September 1876 he
publicly repudiated hostility to Islam in Wollo, and appointed the
most powerful Wollo magnate, Muhammad 'All, as his governor
of Wollo and Amhara. He also appointed a governor to Yejju,
well within Yohannes's sphere of influence. Early in 1877 Menelik
and Muhammad 'All together invaded Gojjam and Begemder. But
Ras Adal was supported by Yohannes; and Menelik's army, which
had recently suffered severe casualties in operations against the
Gurage to the south of Shoa, had little enthusiasm for the
northern campaign. The Shoans received little local support,
accompanied as they were by the Muslim Oromo (Galla) of
Wollo. In May, however, Muhammad 'All and his troops returned
to Wollo, ostensibly to put down an anti-Shoan rising - of which,
on arrival, he promptly made himself the leader. Meanwhile in
Shoa there was a conspiracy, fomented by Menelik's wife Bafana,
647

to replace Menelik by one of Bafana's sons. In June 1877

Yohannes formally proclaimed Menelik a rebel; but Menelik was
already in full retreat from the north. By November he had
regained control in Shoa; but his position in Wollo had collapsed.
Menelik's attempt to assert himself in the north had been
dangerously premature.
In January 1878 Yohannes invaded Shoa; and Menelik decided
to treat for peace. Yohannes was not unwilling; his troops were
ill-supplied and far from their base. But his terms were intended
to demonstrate, and consolidate, his own hegemony. Menelik was
to recognise Yohannes as king of kings, and to perform a personal
act of submission; he was to pay a very heavy tribute, evidently
designed to drain the resources of Shoa; he was to renounce all
claim to sovereignty over Wollo. He was however to assist in the
christianisation of Wollo, and in the imposition of doctrinal unity
upon the Ethiopian Church. He was to expel the Catholic
missionaries, whom he had permitted to proselytise among pagans
and Muslims; and generally to discourage the residence of
Europeans. In return, Menelik was to be formally invested by
Yohannes as negus of Shoa, and to be granted the governorship
of Wollo.
The ceremony of submission, reconciliation and investiture
took place at Liche in Shoa on 26 March 1878. Thereafter,
Yohannes emphasised particularly the imposition of religious
uniformity, which he possibly saw as a means to political
unification. The chiefs and people of Wollo were compelled to
convert to Christianity: Muhammad 'All became Ras Mikael. In
May 1878 the Shoan clergy were compelled at the Council of Boru
Meda to accept the 'two births' variant of Ethiopian mono-
physitism. Menelik could not avoid the expulsion of the Catholic
missionaries; but he evaded Yohannes's frequent demands for the
expulsion of all Europeans, whom Menelik regarded as the
bearers of useful knowledge and techniques. The punctual
payment of tribute he dared not evade - he usually proffered it
personally - and in February 1879 n e performed a further act of
personal submission to Yohannes. By 1881 the burden of tribute
had led to a severe shortage of currency in Shoa and a consequent
depression of the trading economy.
After 1878 Menelik disengaged himself as far as possible from
the politics of northern Ethiopia, and concentrated upon the
648

E T H I O P I A : REIGN OF YOHANNES IV, FROM 1875
conquest and exploitation of the Oromo and Sidama territories
on the ' open' southern frontiers of Shoa. Only in this way could
he hope to escape permanent impoverishment by the payment of
tribute and to procure sufficient firearms, for use when the time
came for a more powerful Shoa to re-enter the competition for
hegemony throughout Ethiopia. But Menelik's penetration of the
Oromo and Sidama kingdoms of the south-west brought him into
conflict with Gojjam — a conflict which Yohannes seems to have
encouraged as a means of distracting his two most powerful
vassals. In 1880 Gojjam still dominated the southern trade: Basso
in Gojjam was the great mart where southern products (gold,
ivory, civet, skins, coffee and above all slaves) were exchanged
for salt (in the bars known as amok which were used as currency)
and for manufactured goods of European origin — textiles,
hardware, glassware and of course arms and ammunition. Gojjam
controlled an established trade; Menelik's forcible intervention
inevitably disrupted it and initially reduced its profitability.
Moreover, Menelik's financial difficulties compelled him to levy
higher tolls upon merchants than Ras Adal. By 1880, Menelik's
Oromo general, Ras Gobana, was asserting Shoan control over
Jimma Abba Jifar, the most important of the Muslim Oromo
kingdoms which lay between the Gibe (upper Omo) and Didessa
rivers; but further to the south, beyond the Gojeb river, Gojjam
was already dominant in the kingdom of Kaffa, the richest and
most powerful of the non-Muslim Sidama states of the south-west.
Kaffa was a sacral kingship of respectable antiquity, and had been
an imperial power, ruling or receiving tribute from the smaller
Sidama groups. In January 1881 Yohannes crowned Ras Adal,
under the royal name of Tekla Haymanot, as negus of both Gojjam
and Kaffa.
However, early in 1882 Ras Gobana stripped a Gojjamese army
of its loot from Kaffa, whose king, Gaki Sheroko, he placed under
tribute to Shoa. Menelik then annexed Oromo territories
immediately to the south of Gojjam; and in June, at Imbabo to
the south of the Abbai, he defeated the main Gojjam army under
Tekla Haymanot, who was taken prisoner and compelled to make
an act of submission to Menelik. Yohannes riposted by supporting
Shoan dissidents, and by depriving Menelik of his governorship
of Wollo. But Yohannes's policy of balance between Gojjam and
Shoa was in ruins. He now recognised Menelik as king of Kaffa
649

(which was not however brought fully under Shoan control until
1897); and he accepted a virtual partition of hegemony between
himself in the north and Menelik in the south. The agreement was
sealed by the betrothal of Menelik's daughter Zawditu to
Yohannes's only acknowledged son, Ras Araya Sellassie, and by
Menelik's recognition of Araya as Yohannes's imperial heir. As
a means of strengthening 'northern' influence at Menelik's court,
Yohannes also insisted that Menelik should marry Taitu, an able
and ambitious princess of the Yejju ruling family.
In spite of Menelik's major success in 1882, Shoan expansion
southwards, especially in its early stages, was difficult and
sometimes hazardous. Immediately to the south of Shoa, the
Arussi Oromo maintained a vigorous resistance until 1886, and
the Gurage until about 1889. These were not in themselves
particularly rich areas (they provided loot in the form of livestock,
but apparently not much else), but their subjugation was essential
to the security of more remunerative conquests further afield.
Between 1880 and 1885 the Muslim Oromo kingdoms between
the Gibe and the Didessa were subjugated. Jimma, by far the most
wealthy and powerful of these, was maintained as a client-kingdom
under its young ruler, Abba Jifar II, who paid a heavy tribute with
exemplary regularity, and assisted Shoa in the conquest of the
smaller Oromo kingdoms, which were subjected to plunder and
to direct Ethiopian rule and settlement. Conquest of these
kingdoms provided not only plunder and tribute, but a permanent
source of wealth from tolls on their external trade, now channelled
through Shoa. Gold and ivory were Menelik's personal mono-
polies ; otherwise the most profitable commodities were civet, furs
and skins, and above all slaves. In Jimma particularly, the slave
trade, whose victims were mainly the neighbouring negroids and
pagan Oromo, had long been highly organised. Abba Jifar
promoted and protected it to the ultimate profit of Menelik, who
took a toll on every slave entering Shoa or sold there. Menelik's
refusal to subject Jimma to the usual destructive processes of
direct Ethiopian rule and settlement was a far-sighted and very
rewarding departure from tradition.
As well as new resources with which to purchase arms, Menelik
needed a secure route to the coast for his trade. At Zeila, Shoa's
traditional coastal outlet, the Egyptians did not encourage the
arms trade with Ethiopia; would-be traders suffered plunder and
650

extortion at the hands of the governor Abu Bakr. Menelik

therefore attempted to open up alternative routes to Assab and
Obok. Both routes were however dependent upon the goodwill
of Muhammad Anfari, the Danakil sultan of Awsa - who had
demonstrated the need for his co-operation by permitting the
massacre in May 1880 of an Italian trading caravan from Assab.
By 1883, however, satisfactory arrangements had been reached
with Muhammad Anfari; and in April 1883 Pietro Antonelli
delivered to Menelik 5,000 rifles from Assab. Assab had meanwhile
(June 1882) been proclaimed an Italian colony, thereby giving
Italian merchants a secure coastal base. The need for such a base
was demonstrated when in 1883 Abu Bakr of Zeila forcibly
removed the French merchants at Obok who had opened up the
route to Shoa. However, in 1884—5 France acquired the Obok—
Tajura region in full sovereignty, primarily as a staging-post on
her route to Madagascar and Indo-China. Paradoxically, these
European acquisitions were in the short run very advantageous
to Menelik; and Franco-Italian rivalry virtually guaranteed that
one of the alternative routes would always be open to him.
Until 1885, however, the volume of trade between Shoa and
the coast was not large. Only comparatively small quantities of
firearms had so far reached Shoa; and Menelik's keen demand
forced up prices to a point where he had difficulty in paying for
what he did receive. In spite of Abba Jifar's sumptuous presents
at the wedding of Menelik and Taitu in April 1883, Menelik had
not yet effectively mobilised his newly conquered resources. A
consignment which Menelik sent to the coast in 1882 was worth
only about £15,000; and of this more than 80 per cent consisted
not of commodities, but of scarce Maria Theresa dollars. In
1884-5 Menelik evidently faced a severe cash-flow problem. In
June 1884 he had to ask Antonelli for extended credit; a year later
Leon Chefneux, trading from Obok, was paid in ivory instead of
coin. But in 1891 Menelik sent commodities worth some £200,000
to Obok. By this time the massive influx of firearms into Shoa
had created a buyers' market, which Menelik exploited to force
down prices and to buy only up-to-date rifles.
Nor were Menelik's problems exclusively economic. As late as
1886 Arussi resistance compelled him to reinforce personally the
army of his uncle, Ras Dargie. Indeed, during this period the bulk
of Menelik's forces, and his most reliable commanders — close
651

relatives like Dargie, Wolde Giorgis and Makonnen, and ' king's
men' like Ras Gobana — were heavily committed in the south;
and he was potentially very vulnerable to pressure from Yohannes,
whose military superiority was regarded as axiomatic in Shoa.
'Northern' influences, headed by Taitu, were strong at the Shoan
court; nor could Menelik expect whole-hearted support against
Yohannes from metropolitan Shoan magnates who would be the
first to suffer in an armed conflict. Down to 1887, he was very
careful not to give Yohannes cause for alarm or suspicion. He paid
his tribute punctiliously and usually in person, and gave Yohannes
considerable military assistance, notably in Wollo.
In sharp contrast to Shoa, remote from foreign interference,
Yohannes's base in Tigre was an exposed salient of the Ethiopian
polity. Its immediate neighbours were not small African kingdoms
or loosely organised tribal communities, but hostile major powers
- Egypt, Italy, the Mahdist State. Preoccupied with external
problems, Yohannes was unlikely to intervene in the south unless
provoked by a direct challenge that Menelik was still careful to
avoid. Instead, he sought to keep Shoa at arm's length by
maintaining his position in Wollo and by a close and supportive
relationship with Tekla Haymanot of Gojjam. The 'Hewett
treaty' of June 1884, whereby Yohannes recovered Bogos from
Egypt, was no solution to his external problems. Egypt retained
Massawa; and Yohannes was right to distrust the British guarantee
of free transit for arms and ammunition. In February 1885 London
permitted the Italians to occupy Massawa. Meanwhile, Yohannes's
fulfilment of his Hewett Treaty obligation to extricate the
Egyptian garrisons had involved him in unwanted hostilities with
the Mahdists; and about May 1885 the Mahdi rejected Yohannes's
peaceful overtures in an elaborate and intransigent indhdra (above,
p. 609). By this time the Italians were obstructing Yohannes's
imports of arms at Massawa and were advancing to localities (e.g.
Sa'ati) within the territory retroceded by Egypt in 1884. At the
end of 1885 Yohannes called upon Menelik for unity against the
Italians. Menelik answered evasively; his secret treaty of friendship
and commerce with Italy, concluded through Antonelli in May
1883, was now bearing fruit in a steady stream of arms to Shoa.
Yohannes's relations, both with Menelik and with the Italians,
seemed to improve in 1886. In January 1886 Menelik personally
proffered a heavy tribute; and thereafter assisted Yohannes in
652

suppressing a major Muslim revolt in Wollo. Meanwhile, the

Italians seemed prepared to negotiate a settlement, and Yohannes
appears to have contemplated a major offensive against the
Mahdists. Indeed, Tekla Haymanot may have been acting with
Yohannes's encouragement when in January 1887 he replied to
the local depredations of the Gallabat Mahdists by storming
Gallabat and ravaging the Gedaref district. The Khalifa responded
by a letter (February 1887) which was indeed an indhdra in form;
but in substance an offer, under reasonable conditions, of an
armistice on the frontier (above, p. 633—4). Yohannes appears to
have ignored this communication. Meanwhile, the negotiations
with Italy collapsed when the Italian commander at Massawa
launched a further forward movement. Ras Alula, impotent
against the Italian fortified posts, riposted by destroying an Italian
detachment. This so-called 'massacre of Dogali' (January 1887)
evoked a cry for 'revenge' in Italy and general rejoicing in
Ethiopia. But neither Yohannes nor Menelik rejoiced at the
prospect of full-scale war with Italy. For Yohannes, faced with
Mahdist hostility and Menelik's evasiveness, it carried formidable
risks; for Menelik, it implied either the abandonment of his
valuable Italian connection or an open breach with Yohannes
which would be far from popular in Shoa. Yohannes appealed for
British mediation, but began to mobilise a large army. Menelik
offered assistance to Yohannes but almost simultaneously
(October 1887) secretly concluded a 'convention of neutrality'
with Italy, which was in all but name an alliance against Yohannes.
Menelik's freedom of action, in contrast to the increasing
constraints upon Yohannes, is well illustrated by Shoan expansion
both eastwards and westwards between 1886 and 1889. The
subjugation of the Ittu and Arussi Oromo in 1885 and 1886 was
followed by the capture of Harar in January 1887. Since its
evacuation by the Egyptians in May 1884 the previously brisk
trade of Harar had stagnated under its local ruler, 'Abdallah
Muhammad 'Abd al-Shakur, who combined hostility to Christians
(both Ethiopian and European) with a recklessly predatory fiscal
policy. Menelik now controlled (he junction of the routes to
Assab, Obok and Zeila; Harar once more became a major centre
of trade, especially with Obok, which Menelik increasingly
favoured in the 1890s. Harar also became the base for Ethiopian
expansion yet further eastwards, into Ogaden and other Somali
653

regions. In the far west, Ras Gobana had occupied Wollega, where
in October 1888 his army, supported by the local ruler Moroda
Bakari, routed a Mahdist invasion from BanI Shanqul.
Meanwhile, Yohannes was further than ever from solving his
Mahdist and Italian problems. British 'mediation' through the
Portal mission at the end of 1887 was simply an attempt, for
' European' reasons, to persuade Yohannes to cede to the Italians
territory to which his title had been recognised in 1884. Yohannes
scornfully refused; and prepared to attack the Italian coastal
positions. However, in January 1888 Khalifa 'Abdallahi, probably
aware of Yohannes's preoccupation with the Italians, launched
Hamdan Abu 'Anja into north-western Ethiopia. The forces of
Negus Tekla Haymanot were overwhelmed; Gondar was sacked,
and the provinces of Dembya and Begemder were ravaged. As
Menelik, on Yohannes's instructions, moved with a Shoan army
towards Begemder, the Mahdists retired (above, p. 634); but
Menelik was accused of having deliberately delayed his advance.
In March 1888 Yohannes moved his armies to the coastal plain
against the Italians; but his losses from starvation and disease were
so severe that he was soon forced to withdraw, having
accomplished nothing.
These military disasters seriously undermined Yohannes's
political position. By April 1888 Tekla Haymanot was attempting
to negotiate a separate peace with the Mahdists; while Menelik,
instead of attacking them, was offering to 'mediate' between
Yohannes and Abu 'Anja. Menelik's military presence in the north
now seemed a threat rather than a support to Yohannes, who
curtly ordered him to return to Shoa. In June, on his way back
through Gojjam, Menelik concluded an alliance with Tekla
Haymanot, evidently directed against Yohannes, at the very
moment when the death of Yohannes's son, Araya Sellassie,
reopened the question of the imperial succession. Yohannes
retaliated by a merciless ravaging of Gojjam and prepared for
war against Shoa. In November Menelik proclaimed general
mobilisation against Yohannes, but the Shoans showed little
enthusiasm for offensive war, and Menelik himself feared to take
the offensive unless his Italian allies also moved against Yohannes.
But Yohannes found the Shoan defences unexpectedly strong, and
his own troops were suffering from smallpox. A major war with
Shoa, in the face of Italian and Mahdist hostility, seemed an
654

unacceptable risk. The Mahdists had demonstrated their hostility

by a second raid into Begemder in the summer of 1888. In
February 1889, in response to the eloquent plea for peace and
co-operation against aggressive 'Turks' and Europeans which he
had addressed to the Mahdists in December 1888, Yohannes
received Abu 'Anja's intransigent and contemptuous reply: ' You
are a great fool to ask peace from us before you enter the religion
of God.' 3s (Above, p. 634.) Yohannes decided to attack the
Mahdists, counting on a victory that would disable one of his
external enemies and restore his damaged military reputation. He
gained his victory at Gallabat on 9 March 1889; but it was
transformed into defeat by his own death from wounds, which
opened the way to Menelik's succession as king of kings and an
Italian occupation of northern Ethiopia.
With Menelik's accession, the political centre of Ethiopia
shifted far to the south of its earlier location. But Shoa was now
ceasing to be a frontier territory. Thanks to Menelik's conquests,
his new capital at Addis Ababa, founded in 1887 and the first
permanent capital since Gondar, was by 1900 approximately at the
geographical centre of his empire. The new territories were not
merely conquered but colonised, by the settlement in them of
Amhara and assimilated Oromo as a military garrison, and as great
and small land-owners to whom the local cultivators were
enserfed as tribute-paying gabbar. Much of the conquered south
was more agriculturally productive than the historic northern core
of Ethiopia; its products, agricultural and other, sustained a
lucrative external trade to which there was no northern parallel.
Although Shoan conquest was often initially destructive, the
naturally poorer north had suffered even more severely from the
ravages of almost continuous warfare.
From about 1880 famine became widespread in northern
Ethiopia. In the provinces of the Lake Tana region - Begemder,
Dembya and Gojjam — famine was aggravated by the repeated
depredations of Ethiopian and Mahdist armies alike; and these
provinces never regained their earlier political importance. By the
mid-1880s famine, accompanied by rinderpest and devastating
human epidemics (dysentery, smallpox, cholera) was spreading
southwards; between 1889 and 1892 it was general throughout
" C.R.O. Khartoum, Mahdia 1/55/15. Abu 'Anja to Yohannes, Jumada I 1306/Jan.
1889, Cited Rubenson, Survival of Ethiopian independence, 384.
655

Ethiopia. But it was most prolonged and severe in the already

impoverished north, of which many regions could pay only
nominal tribute to Menelik in 1893; while the richer territories
of the newly conquered south escaped comparatively lightly.
These catastrophes accelerated and confirmed the economic
preponderance of the south; they also confirmed Menelik's
control of it, by driving southwards a diaspora of 'Amhara'
peasant refugees.
ETHIOPIA: MENELIK AS EMPEROR, 1889-1907
After the death of Yohannes in March 1889, Menelik was quickly

recognised as king of kings virtually everywhere except in Tigre,
where Ras Mangasha Gugsa, whom Yohannes on his death-bed
had recognised as a legitimate son, maintained a rival claim. In
May 1889, by a treaty signed at Wichale, Menelik ceded Asmara,
Massawa and the Bogos to Italy in return for Italian recognition
of his imperial title. Wichale was an insurance against an Italian
recognition of Mangasha; it also made Mangasha and the Italians
virtual rivals for the control of northern Ethiopia. In October a
further convention was signed, fixing the Italian-Ethiopian
border on the basis of actual possession; but in the next few weeks
the Italians advanced southward into Tigre up to the river Mareb,
which they then claimed as their frontier. Menelik protested; but
not very vigorously. In his famous circular of April 1891, in which
he notified to the powers 'who come forward to partition
Africa',36 frontier claims which extended to the White Nile and
Lake Turkana (Rudolf), he did not contest the 'Mareb line'. The
Mareb advance was an immediate problem for Mangasha rather
than for Menelik; indeed, it strengthened Menelik's position by
increasing the difficulty of any sustained and whole-hearted
collaboration between Mangasha and the Italians.
Much more serious was the threat to Menelik's independence
created by Italian deception in the notorious Article XVII of the
Wichale treaty. The Amharic version of this article empowered
Menelik, if he so wished, to deal with European Powers through
the King of Italy. In the Italian version Menelik bound himself
36
Menelik II, Circular, 14 Miyazya 1883 (E.C.)/2i Apr. 1891 (the Gregorian date
usually cited, 10 Apr., is erroneous). Translation in F.O. 1/32, Rodd to Salisbury, no.
15, 4 May 1897; printed, R. Greenfield, Ethiopia: a new political history (London, 1965),
464-5.
656

to the exclusive use of this procedure; and Rome therefore

claimed a protectorate over Ethiopia. Menelik resisted this claim
as vigorously as he could short of an open breach with Italy, which
he could hardly risk under the appalling famine conditions of the
early 1890s and so long as Ras Mangasha remained available to
the Italians as a rival imperial claimant. Down to February 1891,
when Antonelli finally lost patience and quitted Shoa, Menelik
showed great skill and tenacity in avoiding the traps set for him
in various Italian proposals for the 'modification' of Article XVII.
He mastered the 'code' of European diplomatic language so
rapidly and completely that by the later 1890s he could beat the
Europeans at their own game — not the least of his ' modernising'
achievements.
Menelik's protests to other European powers were predictably
rejected by Italy's informal ally Britain and her formal ally
Germany. France and Russia supported Menelik, but with different
nuances. 'Orthodox Moscow' explicitly refused to recognise
Wichale; but was not at this stage in close touch with Menelik
and gave him little practical help. Paris silently ignored Wichale,
but facilitated the massive import of arms through Obok and
maintained close relations with Addis Ababa. The Italians, finding
Menelik intractable, gave rather half-hearted support to Mangasha
in 1891-2; then, in 1893, renewed their attempt to 'reason' with
Menelik, who replied by a formal denunciation of the Wichale
Treaty. In June 1894 Menelik's position was greatly strengthened
when Mangasha 'came in' and submitted in person; while Ras
Alula, the real power behind Mangasha, transferred his allegiance
to Menelik.
During these years there was, apart of course from Tigre,
comparatively little internal dissidence. Menelik maintained close
and friendly relations with Negus Tekla Haymanot, who
dominated the north-central provinces. This relationship seems
to have been temporarily disturbed late in 1892, when Menelik
began to impose his own taxes throughout the empire; but it had
evidently been completely restored by February 1894, when he
ceremonially recrowned Tekla Haymanot as king of Gojjam.
Shoan expansion maintained its momentum. There was deep
penetration into Somali and southern Oromo country; the Sidama
kingdoms of the Rift Valley and Gibe-Gojeb basin were eaten
up one by one. In 1894 the powerful Sidama kingdom of Walamo
657


was subjugated. Only Kaffa now maintained a precarious inde-

pendence ; but even Kaffa sometimes paid tribute, and had by 1892
been compelled to export all its coffee through Shoa.
Early in 1895 there was a clash between the Italians and their
former protege, Mangasha of Tigre. The Italians advanced
beyond the Mareb; they now began to demand the cession of all
Tigre and an explicit recognition by Menelik of their protectorate
over Ethiopia. In September, after careful preparation, Menelik
proclaimed a general Ethiopian mobilisation against Italy. He
enjoyed no direct European support. In July 1895 France had
rejected his offer of an alliance, hoping for a prolonged conflict
'pour degager d'autant la politique europeenne'.37 But down to
the end of 1895 (at least) the French continued to supply Menelik
with arms. In this and other ways Paris, for its own reasons,
sought to create difficulties for Italy while avoiding an open
alignment with Ethiopia. For this Menelik evidently owed the
French no thanks; but their policy nevertheless served him well.
Menelik's position was also strengthened by improved relations
with Omdurman. Ever since 1889 he had been at pains to promote
and safeguard the 'undeclared peace' which the Khalifa had
accepted. By 1894 there was a considerable trade across the
frontier at Gallabat; the local Mahdist and Ethiopian governors
even discussed concerted action against Kidana Mariam, a shifta
leader who preyed on the caravans. In the summer of 1895
Menelik sent an envoy to Omdurman with verbal proposals for
the formal conclusion of peace. 'Abdallahi's reply, though cold
and discourteous in tone, was no indhdra; it was a request for
detailed proposals in writing. Unlike Yohannes in 1887-8, Menelik
could challenge the Italians without fear of hostile Mahdist
intervention.
In the Adowa campaign Menelik was able to concentrate forces
from all over Ethiopia; and, thanks to his establishment of
imperial granaries in the central provinces, to keep this enormous
army adequately victualled for just long enough to strike its blow
— a telling contrast to the Khalifa's almost complete logistic failure
during his defensive campaign of 1896-8. Every important
Ethiopian magnate brought his troops to the support of Menelik.
Some of the great men of course reinsured, in time-honoured
37
Documents Diplomatiques Franfais, Ie serie (D.D.F.), x n (Paris, 1951), no. 99, note
by Hanotaux, 'Question du Harrar et Obok', 23 July 1895.
660

Ethiopian fashion, by private contacts with the Italians; but there

can be no doubt where their primary loyalties lay, or that Adowa
was an impressive demonstration of Ethiopian national solidarity.
Although the Ethiopians greatly outnumbered General Baratieri's
force, Adowa was not simply a victory for numbers and' barbaric'
courage. The Italians, although only ioo miles from their base,
were forced to give battle as the only alternative to a humiliating
retreat for lack of supplies. Menelik, through Ras Alula as his chief
of staff, maintained some overall control of the battle - which is
more than Baratieri did. The Italian troops, advancing by night
through tangled mountainous terrain, lost contact with one
another; and each of Baratieri's brigades fought an isolated action.
Menelik's crushing victory (i March 1896) was followed by
intense Anglo-French competition for his goodwill. Even the
Russians, though never quite sure what they wanted in Ethiopia,
now became very active - but in competition rather than
collaboration with the French. The French were first off the mark:
in June 1896 President Felix Faure offered to negotiate on the basis
of Menelik's 1895 alliance proposal. Menelik was now in no hurry;
but agreement was reached in March 1897. The French accepted
very restricted frontiers for their Obok-Jibuti enclave. They were
not interested in territory, but in gaining' influence' based on their
control of Ethiopia's main outlet to the sea; and above all in
securing Menelik's collaboration with the Marchand mission to
the upper Nile. The 'Convention pour le Nil Blanc' of 20 March
189738 partitioned the southern Sudan between France and
Ethiopia along the line of the White Nile; and Menelik promised
to assist French missions to reach the upper Nile from Ethiopia.
The British, who arrived in Addis Ababa in April 1897, were
anxious to dissuade Menelik from supporting the Mahdists, and
to negotiate favourable frontiers in the Nile valley and Somaliland.
Although Menelik was by now in friendly negotiation with
' Abdallahi, he was pleasingly willing, in a treaty of May 1897, to
denounce the Mahdists as 'enemies of his empire'. 39 On frontiers
he was less forthcoming. He exhibited his circular of April 1891
and convinced the British negotiator, Rennell Rodd, that his
'effective occupation' already extended almost to the White Nile.
'* D.D.F., xin (Paris, 1953). no. 159.
« Anglo-Ethiopian Treaty of 14 May 1897, Art. VI: E. Hertslet, The map of Africa
by treaty (revised edn, London, 1909), no. 99.
661

In fact, Menelik's occupation hardly extended beyond the western
ramparts of the Ethiopian massif; but Rodd was so disconcerted
that he preferred to postpone discussion, and Menelik was later
to assert that the British silence in 1897 implied the acceptance
of his 1891 claims. But in Somaliland a more substantial Ethiopian
occupation forced Rodd to concede the Haud.
Menelik fooled the French no less completely than he fooled
the British. His support for the French missions to the upper Nile
was purely verbal; he secretly encouraged his agents to put every
possible obstacle in their way. The arrival of Europeans from
Ethiopia on the upper Nile would have cost Menelik the
confidence of Khalifa 'Abdallahi, which since July 1896 he had
been at great pains to gain. Menelik's objective was ' to strengthen
the Khalifa against the [Anglo-Egyptian] troops, whom he feared
as neighbours, preferring the Dervishes \ 4 0 Perhaps prompted by
the French, Menelik became convinced in 1896 that the Anglo-
Egyptian offensive in the Sudan was ultimately directed against
Ethiopia; his fears were increased by the movement of the
Macdonald mission from Uganda towards Lake Turkana in 1898.
But if 'Abdallahi could be relieved of all apprehension on his
eastern frontier by afirmpeace with Ethiopia, he could concentrate
exclusively on Kitchener; and if he could then deal with Kitchener
as the Mahdi had dealt with Hicks, Ethiopia's' European problem'
would indeed be solved.
Throughout his arduous but ultimately successful wooing of
'Abdallahi, Menelik hammered away at a theme which had been
introduced by Yohannes IV when he had offered peace to the
Mahdists in December 1888: that all the European Powers were
the common enemies of Ethiopia and the Sudan; and that
religious differences and traditional enmities should not be
allowed to frustrate co-operation and solidarity against the
European menace. Earlier Ethiopian rulers — Tewodros, and even
Yohannes before his bitter disillusionment in the mid-1880s - had
called upon, and apparently expected, the assistance of the
Christian powers of Europe against 'Turks' and 'Ishmaelites'.
Menelik's correspondence with 'Abdallahi bears witness to a
revolution in attitudes which (as other evidence shows) was no
mere personal idiosyncrasy; and which was important for the
survival of Ethiopian independence.
Meanwhile, Menelik was rounding off his empire. In 1897
40
Sudan Intelligence report, no. 60 (1898), p. 20, appreciation by Wingate.
662

Kaffa, the greatest and richest of the Sidama kingdoms, was

subjugated after a bitter struggle, and became thefiefof Menelik's
cousin and close collaborator Ras Wolde Giorgis. In 1898 Ras
Makonnen conquered the BanI Shanqul; and a force under Wolde
Giorgis was sent to the Lake Turkana region to assert Ethiopian
authority — and also to look for Macdonald and 'check his
passport'.41 Dejazmatch Tessama Nado occupied the upper Sobat
(Baro) valley and subjugated the Shanqalla (negroid) groups
south-west of Kaffa. In June 1898 a detachment of Tessama's
force, accompanied by a Frenchman, a Swiss and a Russian, made
aflyingvisit to the Nile-Sobat confluence, where they planted and
immediately abandoned French and Ethiopian flags. This was not
however a serious Ethiopian prise de possession: the Sobat marshes
were of no value to Menelik except as a bargaining counter to
be exchanged for the territories that he really coveted in the
Gedaref, Roseires and Gallabat regions. But the bargaining
counter lost all value when in 1899 the British discovered for
themselves the hollowness of Menelik's claim to effective occu-
pation. Meanwhile the British had occupied the more desirable
frontier areas where Menelik, to avoid complications with the
Khalifa, had refrained from military occupation. Instead, he had
sent emissaries bearing the Ethiopian flag — which was usually well
received by the local Sudanese, but which promptly disappeared
on the approach of the Anglo-Egyptian troops. In the frontier
settlement of 1902 Menelik lost to the Sudan all the debatable areas
except the BanI Shanqul, where he had an army and not merely
a flag.
In 1898—9 Mangasha, dissatisfied with his status (Menelik
refused to make him negus), revolted for the last time and was
imprisoned after his defeat and submission. Tigre remained
turbulent - not even Makonnen could control it effectively; but
it was no longer any danger to Menelik. Externally, Britain and
France, both of whom preferred to maintain the status quo in
Ethiopia, were able — especially after the entente of 1904 — to
restrain Italian aspirations to territorial control. Menelik, in taking
note of the Tripartite Treaty of July 1906, which institutionalised
this balance, was careful to declare that it ' in no way limits what
we consider our sovereign rights'. 42
41
D . D . F . , x i i i , n o . 386, Lagarde ( A d d i s A b a b a ) to H a n o t a u x , 24 D e c . 1897, q u o t i n g
Menelik.
42
F . O . 4 0 1 / 1 0 , statement by Menelik, 10 D e c . 1906: cited H . G. Marcus, The life
and times of Menelik II ( O x f o r d , 1 9 7 ) ) , 2 1 1 .
663

From 1900 until the collapse of Menelik's health in 1908-9, the

history of Ethiopia is comparatively uneventful. The emperor
worked steadily and systematically at the consolidation of his
administrative control. His power was still based largely on the
conquered territories, where previous hegemonies had been
destroyed and which therefore lent themselves to comparatively
close administration and systematic exploitation by trusted' king's
men'. But after 1900 Menelik was strong enough to destroy or
cripple old hegemonies even in historic Ethiopia. Tigre was
divided between two rulers in 1900; so was Gojjam after the death
of Negus Tekla Haymanot in 1901.
Menelik's position was strengthened, and Shoan man-power
reinforced, by the intensive assimilation on favourable terms of
those Oromo willing to embrace Christianity and to adopt the
Amharic language and Amhara social customs. This assimilation
took place at every level, from grandees (like Ras Gobana) to
simple peasants or, in the south, soldier settlers. In 1904 Menelik
decreed the restitution of their ancestral lands to Oromo notables
who had submitted. From peasant to ras, the interests of this
Shoan—Oromo amalgam were bound up with the maintenance of
Shoan hegemony and the imperial power of the Shoan ruler. The
socio-political system created by Menelik therefore had its own
inherent stability which enabled it to survive his incapacity and
death.
There was little specifically 'modern' about the means
whereby Menelik consolidated his administration. They were the
usual methods of a strong ruler in a feudal (or 'quasi-feudal')
society. Detachments of imperial troops under trusted comman-
ders garrisoned strategic points throughout the empire. Taxation
payable direct to the emperor, notably the asrat or agricultural
tithe, was introduced during the 1890s; and fiscal agents were
stationed in the provinces to collect and account for the imperial
dues and revenues. The primacy of the emperor's justice was
asserted by the establishment in each province of an imperial court
of appeal. There was little in this process of 'efficient
medievalisation' which would have surprised - say — Henry II
Plantagenet. Of course, Menelik faced far greater problems of
distance, of terrain and of ethnic diversity, than a ruler in the
medieval West; and he was quick and discriminating in discerning
the usefulness of modern western technology, and up to a point
664

SOMALIA, C. 187O-I9IO
of western organisational techniques, in offsetting these dis-

advantages. Obvious examples are firearms; and, for a corps d'elite,
more systematic military organisation and training. Less often
cited, but invaluable to the ruler of a vast empire with poor
communications, are the telegraph and the telephone. By
December 1902 there was a telephone link between Addis Ababa
and the main provincial capitals; in the last years of his effective
reign Menelik used it as a routine instrument of administration.
Menelik was primarily interested in those ' modern' technolo-
gical devices which enabled him to rule more effectively. He was
however also interested in ' modernising' the image projected by
Ethiopia to the West. This ' cosmetic' modernisation had a serious
political objective: to strengthen Ethiopia's claim to 'parity of
esteem' as a fully sovereign state - a major preoccupation of every
Ethiopian ruler since Tewodros. Cosmetic modernisation and
practical utility were sometimes combined, as in Menelik's
western-style school and his ' cabinet', both inaugurated in 1907.
The school emphasised the teaching of European languages; and
the ' cabinet' enabled a ruler in failing health to shed some of his
crushing burden of work. In administration, as in 'high politics'
and diplomacy where he scarcely made a false move after 1878,
Menelik understood the limits of the possible, kept his objectives
clearly in sight, and had an almost unerring eye for the means best
suited to attain those objectives. He left behind him a socio-political
system of remarkable resilience, which in spite of catastrophic
political vicissitudes survived unchanged in its essentials for two
generations.
SOMALIA, C. 1870—I9IO
In the later nineteenth century the Somali, especially the four great
pastoral clan-groups of the north and centre (Dir, Darod, Hawiya
and Isaq) were strongly and proudly conscious of their ethnic
identity. Ethnic solidarity was reinforced by common socio-legal
institutions and by a common language, the skilful use of which
was at once a powerful weapon in Somali politics and the major
Somali art-form. These unifying forces had not, however, gene-
rated any form of political unity, or indeed of political authority,
above the level of the individual clan; and even clan-leaders were
usually no more than respected dignatories, not 'chiefs' or rulers
with effective institutionalised power. Only in the coastal' sultan-
665

OGADEN Regions, localities (provinces in Ethiopia)

DIR Somali dan-groups
\cldagale Somali clans or lineages
| (other ethnic groups outside Somalia)
Ethnic/linguistic limits of Somali
aoaooa Approximate southern and eastern limits
of DIR and ISAQ clan-groups
Coastal territories of Somali clans which con-
cluded ' Treaties of Protection' with Britain. 1884-6
Menelik's territorial claim of 1891
Frontier of British Somaliland from 1897
" Battles fought by Muhammad Abdallah Hasan
P 300 miles
•'Assab,
(Hal. 1882)'-'
G u l l ol Ada
\ SOUTHERN HAUD
Daratoleh. .
1903 X \
OGADEN # Gumburu
iJ"?4A(p _ D A R o'o
27 Somalia, 1884-f. 1910

666
SOMALIA, C. 187O-19IO
ates' of the Majerteyn and Hobyo (Obbia) was political authority

rather more effective, probably supported by resources derived
from sea-borne trade.
Competition for water and grazing, combined with Somali
intolerance of subordination, therefore generated major violence,
not only between clans but between lineages of the same clan.
The losers in any particular conflict kept careful tally of their
' grievances' (godob); and the (usually violent) rectification of these
grievances was an hereditary obligation of honour. Procedures of
course existed for the regulation and composition of feuds; but
these procedures were too fragile to resist exceptional political or
ecological pressures. Such pressures did not therefore tend to
promote solidarity; they tended rather to unleash a ' war of all
against all' as clans and lineages struggled for survival or strove
to exploit disturbed conditions to settle outstanding grievances.
The most stable of Somali ' peace-restoring' institutions was the
'd/ya-paying group', whose members (drawn from the same major
lineage) contractually assumed a collective responsibility to pay
blood-compensation {diyd). But the drya-paying group was a
comparatively small unit of a few hundred adult males. It could
contain violence within lineages; but not 'war' between lineages,
still less between clans.
Somali identified very strongly as Muslims; in the local context,
'Somali' and 'Muslim' were often used almost synonymously,
perhaps because Islam was the one pan-Somali institution which
tended positively to unify the Somali rather than merely to contain
their conflicts. The holy man (waddd), distinguished by his greater
piety and Islamic learning, was normally very sharply distin-
guished from the waranleb or warrior. Unlike many Muslim holy
men, he was almost exclusively a religious leader and teacher; his
sole political function was to exhort conflicting groups to peace
by dwelling on the heinous sin of violence between Muslims. He
could normally have no role as a political leader, for 'politics' by
its very nature implied bloodshed and loot, which were forbidden
{hardm) to a holy man.
An important development in Somali Islam during the nine-
teenth century was the curiously belated introduction, but very
rapid expansion, of religious brotherhoods {tartqas) offering a
specific 'way' to God. The first to be introduced was the
Qadiriyya, originally founded in the twelfth century AD. The
Qadiriyya emphasised its collective devotions {dhikr) as the ' way'
667

to God; it accommodated itself to Somali customs and even

accepted the Somali habit of chewing qdt. Its rival the Ahmadiyya,
introduced a little later, had been founded at Mecca by Ahmad
Idrls al-FasI (1760—1837), who had been strongly influenced by
the revival of Islamic rigorism during his life-time. The Ahmad-
iyya insisted on active piety, and on the formal marks of such
piety: regular prayer, the veil for women, the turban for men. Its
members were pledged to material austerity; in particular, it
forbade the use of qdt. Its dhikrs (devotions) were more austere
and restrained than those of the Qadiriyya. The Ahmadiyya had
two branches in Somalia: the small and comparatively quietist
Dandarawiyya; and, introduced as late as the 1890s, the Salihiyya,
founded by Muhammad Salih, the nephew and successor of a
Sudanese disciple of Ahmad Idrls. The Salihiyya, under the
leadership of the waddd Muhammad 'Abdallah Hasan, became
after 1899 exceptionally rigorist and intransigently militant. From
about 1885 Somali tariqa leaders had begun to found cultivating
settlements, whose members withdrew as far as possible from the
conflicts of their lineages, bound themselves to strict observance
of the rules of their order and — in sharp contrast to the normal
Somali rejection of subordination — submitted to the authority,
in principle absolute, of the shaykh of the settlement.
Other features of Islamic revival were greatly increased Somali
participation in the pilgrimage and the migration to Somalia of
holy men and religious teachers. The Egyptians during their
period of occupation (1870-84) actively fostered Islamic devotion
and learning. Somali eyes were opened to the crisis of Dar al-lsldm
under infidel rule or the threat of it. Some Somali began to take
an exceptionally rigorous view of Muslim relations with infidels.
A Muslim who associated with infidels or served them in any way,
even, for example, as a commercial agent, was not merely a lax
Muslim, but himself an infidel (kafir), with whom the rigorists
refused to pray. The holy man began to emerge from his purely
religious and eirenic role and to play a greater part in worldly
affairs: Muhammad 'Abdallah Hasan is merely the most spec-
tacular example in an era of politically influential 'shaykhs'.
668

SOMALIA, C. 187O-I9IO
Egyptian occupation and European partition, c. 1870-1899

Until the 1870s northern Somalia was nominally a direct depen-
dency of the Ottoman empire. The Benadir (Indian Ocean) coast
as far north as Mogadishu remained until 1889 a dependency of
Zanzibar. With the opening of the Suez Canal in 1869, Khedive
Isma'Il began to assert Egyptian authority along the northern
coast. In 1875, when the Egyptians also occupied Harar (above,
p. 613), the Egyptian administration of Zeila and Berbera was
confirmed by Ottoman firman. Britain, at first hostile to this
occupation of a strategic coast opposite Aden, had by about 1875
become reconciled to it as a safeguard against occupation by a
rival European power. In 1877 London recognised Egyptian
sovereignty as far as Ras Hafun on the Indian Ocean, on con-
dition that Egypt should never alienate these territories to
another power. Northern Somalia seems to have prospered under
Egyptian administration. Harbour and other facilities were im-
proved at Zeila and Berbera. Overseas trade was active. Livestock
and exotic African products, the latter now channelled through
Harar, were exchanged for manufactured goods and vegetable
foodstuffs. Liaison with the interior was maintained by elders
('dqils), who represented the major <//ya-paying groups at the
centres of administration.
From 1877 to 188} London actively supported Egyptian
authority in northern Somalia; but in 1884, as a result of the Sudan
debacle (above, pp. 608-9,616-7), ^ e British insisted on Egyptian
evacuation; and British—Indian troops were landed to cover the
Egyptian withdrawal. As a substitute for Egypt, London would
have accepted an Italian occupation, as at Massawa. But the
Italians hesitated; and between May 1884 and March 1886 the
British concluded 'treaties of protection' with the northern
coastal clans from the Isa (Dir) on the Gulf of Tajura, eastwards
to the Warsangali and the Majerteyn (Darod), the latter extending
to the Indian Ocean. These treaties bound the clans never to
alienate their territories to another power. Meanwhile, the French,
who had c bought' the Obok region as early as 1862 but had never
performed any act of sovereignty there, suddenly revived this
dormant claim (above, p. 651). By March 1885 Leonce Lagarde,
sent out as governor in July 1884, had concluded treaties of
cession with all the groups surrounding the Gulf of Tajura,
669

including the Isa Somali who already had a treaty of protection

with Britain. An Anglo-French boundary (which partitioned the
Isa) was agreed in February 1888. In December 1889 British
jurisdiction was formally extended to the coastal clans eastward
of the new Anglo-French frontier, other than the Majerteyn where
Italy had established a protectorate in April 1889. Italian
protectorates were also established in 1889 at Hobyo (Obbia) and
on the Benadir coast; but until the early twentieth century there
was virtually no Italian administration on the coast north of
Mogadishu.
After Menelik's occupation of Harar in 1887 (above, p. 653),
the Dir and Isaq clan-groups of the Zeila—Berbera hinterland
suffered increasingly from Ethiopian raids upon their stock.
Indeed, in his famous circular of April 1891 (above, p. 656)
Menelik had claimed as Ethiopian the territories of three of these
clans. The aggrieved groups appealed in vain for the British
protection to which they were theoretically entitled. Having
recognised the Treaty of Wichale, London handled the problem
by negotiation with Italy. An internal frontier was agreed with
Rome in May 1894: it ceded to Italy (and therefore ultimately to
Ethiopia) much Dir and Isaq pastureland. A further slice of
territory in this region was abandoned directly to Ethiopia by the
'Rodd Treaty' of June 1897 (above, p. 662). In the east, however,
the British Somaliland Protectorate as finally delimited included
the territory of a large and powerful inland clan, the Dulbahante
(Darod), with whom there was no treaty of protection an i indeed
no official contact whatever.
Even on the coast, administration was rudimentary. British
vice-consuls, responsible to the government of India through
Aden, resided at Zeila, Berbera and Bulhar. In 1898 the Foreign
Office took over the administration. Sir James Hayes-Sadler,
appointed as consul-general, reduced expenditure by replacing
most of the Indian garrison by an armed Somali constabulary
{Halo), who policed the internal trade-routes more effectively.
These measures, together with more systematic management of
the 'aqils and friendly relations with the commercial and religious
notables of the coastal towns, led to a considerable expansion in
trade. This tranquil prosperity was however deceptive. Especially
in the Ogaden region south of the locally meaningless Protectorate
frontier, destructive Ethiopian pressure steadily increased. The
670

SOMALIA, C. 1870— 19IO
Ethiopians seem to have destroyed tariqa settlements as deliberate

policy. Clans and lineages were displaced from their traditional
ranges; they sought compensation and indeed survival at the
expense of their neighbours. By about 1900 'more than twenty
lineages were at one another's throats>43 in northern Somalia. A
large back-log of 'old outstandings' was built up. Any major
disturbance which weakened the administration, or the emergence
of a new power-factor whose support might be enlisted, would
be eagerly seized upon as opportunities to settle these godob.
The resistance of Muhammad ' Abdalldh Hasan to

In 1899 a new, and profoundly destabilising, power-factor
appeared in the person of the waddd Muhammad 'Abdallah Hasan.
Muhammad 'Abdallah was of an Ogaden Darod lineage (the
Bah-Geri); but his immediate ancestors had settled among, and
intermarried with, the Dulbahante. Sayyid Muhammad's early life
is obscure; even the date of his birth (1856 or 1864?) is contested.
The tradition that he learned the Koran by heart within three years
is an indication, among many others, of his outstanding intellectual
ability. For some ten years between about 1880 and 1890 he
travelled widely within and beyond Somalia in search of Islamic
learning. During these years, he evidently became deeply con-
scious of the infidel threat to Islam. He also gained, either then or
later, a shrewd insight into the more material pressures of colonial
rule. But in 1899 not even Muhammad 'Abdallah could plausibly
stigmatise the 'aloof and benign' British administration as
materially oppressive. He attacked it simply as infidel rule and
therefore ipso facto an oppression; and lashed with biting scorn,
as cowards and virtual apostates, those Somali who ' grovelled to
the hell-ordained and the Christians'.44
The Sayyid had performed the pilgrimage in 1894, and at Mecca
had fallen deeply under the influence of Muhammad Salih. Under
Muhammad Salih's spiritual guidance he underwent a profound
religious experience, whereafter he was commissioned as khalifa
of the Salihiyya brotherhood in Somalia. Between 1895 and 1897
he vainly endeavoured to convince the Qadiriyya religious
43
S. S. Samatar, Oral poetry and Somali nationalism (Cambridge, 1982), 114.
44
D . Jardine, The Mad Mullah of Somali/and (London, 1923; reprinted N e w York,
• 9 6 9 ) . 49- Samatar, Oral poetry, 154, citing M u h a m m a d 'Abdallah Hasan.
671

establishment of Berbera of their reprehensible laxity; and above

all to persuade them to accept Muhammad Salih as qutb al-t(aman
- the spiritual 'axis of the age'. At this stage the British
administration found nothing politically objectionable in the
Sayyid's teaching; but probably under Qadiriyya pressure it
ultimately' closed down' (?i 897) the Salihiyya mosque in Berbera.
The Sayyid then retired to the Dulbahante country, the land of
his maternal kinsmen, where he founded a tariqa settlement; and
he soon gained a strong following, not only among the Dulbahante
but also among two Isaq clans. Feuds between the groups which
joined his movement had necessarily to be composed, or at least
suspended; the administration therefore considered him a
supporter of 'law and order'.
Early in 1899 Ethiopian pressure reached a climax with a
devastating raid into the Ogaden. The Sayyid now began to claim
a direct commission from God to deliver the Somali from the
destructive and infidel 'Amhara'. In April 1899 he replied in curt
and discourteous terms to an official request for the return of a
government rifle believed to have fallen into his hands. In August
he appeared with a following of perhaps 5,000 men at Burao, an
important watering-centre for the Isaq clans. Some of his followers
had visions of taking and looting Berbera, which the Sayyid knew
to be impossible; it was perhaps the need to maintain their
enthusiasm that decided the timing of his famous letter of defiance
to the administration (August 1899). He accused the British of
oppressing Islam without cause; and in effect claimed political
sovereignty by his demand that they should choose between war
and the payment ofjizpa, the tax canonically due to a Muslim ruler
from tolerated infidels.45 He was at once proclaimed a rebel.
The Sayyid's earliest call to jihad seems to have been against
the Ethiopians rather than the British; and until September 1899
the administration itself believed that his movement was essentially
anti-Ethiopian. However, Muhammad 'Abdallah's first act of war
was neither against the Ethiopians, nor the British, nor even the
Qadiriyya; but against his own 'half-brothers' in religion, the
Dandariwiyya section of the Ahmadiyya, whose cultivating
settlement at Shaykh he looted and destroyed in September 1899.
Some dramatic action was doubtless needed to impress his
45
Facsimile and translation, Jardine, Mad Mullah, 42-3.
672

followers; booty was needed both to satisfy their expectations and

as currency with which to purchase firearms. Shaykh seems to
have been chosen simply as a soft target. But thegarad(clan-leader)
of the Dulbahante, 'AIT Mahmud, now took alarm at the Sayyid's
rise to power, and offered his collaboration to the British. He was
promptly assassinated by agents of the Sayyid. The Dulbahante
thereupon split into supporters and opponents of Muhammad
'Abdallah; his opponents were powerful enough to force him to
retire to his paternal Bah-Geri lineage among the Ogaden Darod.
Here he offered his assistance to the powerful Muhammad Zubayr
lineage, which had recently suffered crippling stock-losses at the
hands of the Ethiopians. The Sayyid and his followers recovered
much of the looted stock in an action at Jigjiga in March 1900.
He followed up this success by a major and very profitable raid
upon the Idagale Isaq. Thereafter, certain individuals apart, most
of his Isaq followers deserted him; the Isaq clans became his
consistent opponents, and therefore the targets for some of his
most vitriolic invective.
Meanwhile, the Ogaden Muhammad Zubayr lineage and its
leader Husayn 'Iljeh had, like the garad 'All Mahmud of the
Dulbahante, begun to fear and resent the Sayyid's preponderance.
However, their attempt to assassinate Muhammad 'Abdallah and
his inner circle of followers misfired, though they killed the
Sayyid's principal adviser. The Muhammad Zubayr then decided
to end the feud and sent a delegation to offer diya. But the Sayyid
made exorbitant demands quite unjustified by Somali custom; and
when these were rejected, he put to death the entire delegation
of thirty-two elders. Pressure from the outraged Muhammad
Zubayr then forced the Sayyid to return to the Dulbahante
country. The 'poor man of God' had by now acquired sufficient
stock to buy peace by paying full compensation for the killing of
garad 'All Mahmud. But the schism among the Dulbahante
remained unhealed.
By the late summer of 1900 Muhammad 'Abdallah was based in
the Nugal valley, the most favoured part of a region rich by Somali
standards. From this base, with Dulbahante support prompted by
prospects of loot as well as by religious zeal, the Sayyid could
threaten the interior pastures of the ' British-protected' Isaq clans
in the west. Indeed, the Isaq at once began to withdraw from their
673

summer ranges and prematurely to consume their winter grazing.
The consequent dislocation and over-crowding was a threat that
the administration could not ignore.
The Sayyid had by now created a fighting force which
completely outclassed the fragile clan and lineage groupings. He
organised his 'dervishes' by unprecedented emphasis on the
autocratic and totalitarian features of the tariqa settlement; and
by combining the functions of the waddd with those of the
waranleh, thereby completely repudiating the waddd's traditionally
pacific role. The tariqa settlement tried to be a 'total institution',
but in practice its members could rarely if ever escape completely
from the obligations of lineage and the <#)w-paying group. But the
Sayyid's adherents had 'thrown over father and mother and
tribe ' 46 to follow him. They had in effect cut themselves off from
Somali society, as had indeed the Sayyid himself by his deliberate
flouting of some of its most binding conventions. Their solidarity
was encouraged by a distinctive uniform (the white turban) and,
in action, by the continuous chanting as a battle-hymn of the
Sayyid's poem in praise of Muhammad Salih. Solidarity was
enforced by harsh, even terroristic, discipline. In a tariqa settle-
ment, the shaykh's ultimate sanction was merely expulsion; but
the Sayyid was absolute master of life and death, and he used his
powers freely and arbitrarily.
The hard core of his followers, the dervish ' regulars', were
religious enthusiasts, pledged to embrace martyrdom for Islam as
defined by Muhammad Salih and Muhammad 'Abdallah; encour-
aged and if necessary compelled to repudiate clan and lineage; and
stiffened by a discipline harsh enough to crush even Somali
insubordination. No wonder the dervishes were quite invincible
in the context of traditional Somali warfare. Besides these
'regulars' — perhaps about 6,000 men — the Sayyid had a
fluctuating following of part-time irregulars, often attracted
mainly by prospects of booty. The Sayyid's personal bodyguard
was however drawn from semi-servile cultivating clansmen or
even from former slaves, whom the Sayyid provided with
property and wives. ' Slaves on horses' seem to have been a very
early development in the Sayyid's movement. Otherwise, it
developed very little institutional apparatus apart from the ' inner
council' (majlis khusiisi) of the Sayyid's chief advisers; and no
46
Ibid., 71, citing Col. E. Swaync (who commanded the first and second
expeditions). ,
674

SOMALIA, C. 1870—1910
machinery for its own perpetuation. The contrast with the

Sudanese Mahdist state is striking; but in Somalia there was of
course no ready-made state administration for the Sayyid to take
over and adapt to his purposes.
Although the British soon learned to respect their fighting
qualities, the dervishes did not present an insoluble military
problem; sketchily-trained Somali levies were able to defeat them
in 1901. They were of course bewilderingly mobile, the result of
their ability to survive on camel's milk; while the movements of
government troops were always constrained by water-supply. But
the Sayyid was a serious menace to the Protectorate only when
based in the Dulbahante region, from where he could not only
threaten all the British-protected Isaq clans, but also trade
livestock for firearms through the Majerteyn sultan and the
'British-protected' Warsangali, who were in fact quite outside
British control.
Outside the Dulbahante area the Sayyid had no convenient
alternative base. The Mudugh oases, separated from British
territory by eighty miles of the waterless southern Haud, were too
remote. After the Sayyid's unappeased feud with the Muhammad
Zubayr, the Ogaden could never be more than a temporary
refuge; and by 1903, with increasing Ethiopian pressure, hardly
even that. There was never any serious difficulty in ejecting the
Sayyid from the Dulbahante country. Even the 'unsuccessful'
second expedition of 1902 succeeded in this. The difficulty lay in
preventing his return as soon as the expeditions had retired. The
men on the spot repeatedly pressed for the occupation and the
administration, however lightly, of Dulbahante territory. This
solution was always rejected by London, ostensibly on the ground
that, having no treaty with the Dulbahante, the British had no
duty to protect them from the dervishes - even though the
Dulbahante had since 1894 been within the internationally recog-
nised frontiers of the Somaliland Protectorate. In fact, of course,
any extension of so troublesome an administration seemed utterly
absurd to London. Yet many, perhaps most, Dulbahante were by
now strongly opposed to the Sayyid; but they were quite unable
to resist him without assistance.
The Sayyid's move to the Nugal valley late in 1900 prompted
the first expedition against him, which brought him to battle by
capturing much of his livestock, defeated the dervishes in June
675

and July 1901, and forced him back to the Mudugh oases. But
by October 1901 he had returned to the Nugal valley. The second
expedition again drove him out of Protectorate territory (June-
August 1902), but it was ambushed and forced to retire when it
attempted to pursue the Sayyid into the southern Haud bush. This
time the Sayyid remained at Mudugh; but, as his relations with
the Majerteyn and Hobyo sultans had by now become precarious,
he acquired a port of his own by occupying Illig in Italian-claimed
territory on the Indian Ocean.
The third and fourth expeditions (February 1903—early 1904)
were elaborate and expensive operations, in which the Somali
levies were reinforced by a motley collection of troops from
Britain, India and various African territories. The third expedition
attempted to destroy the Sayyid by pincer-movements from
Berbera and (by arrangement with the Italians) from Hobyo,
combined with an Ethiopian advance into the Ogaden. The
British columns met in the interior without encountering the
Sayyid, who had withdrawn to the western Ogaden. Two
detachments despatched in his pursuit were ambushed and
roughly handled in April 1903. But in May an Ethiopian advance
northward from the upper Webe Shebele river defeated the
Sayyid's forces at Jayd and denied him the Ogaden. In a typically
brilliant and unexpected manoeuvre Muhammad 'Abdallah then
doubled back to the Nugal valley across the British line of
communication, unimpeded by the troops stationed there (June
1903).
The Sayyid improved this occasion by his second well-known
letter of defiance ' to the English', in which he demanded that they
should return to their ' own country' and taunted them on their
recent discomfiture. He added that they could expect nothing
from him but war, because he was a ' poor man of God' without
possessions (hardly an accurate statement) and because his
' country' was worthless. ' If you want wood and stone you can
get them in plenty. There are also many ant-heaps. The sun is very
hot.' 47 But the Sayyid's position had in fact seriously deteriorated
by the second half of 1903. The Ethiopians had denied him the
Ogaden even as a temporary refuge, and indeed threatened his
fall-back base at Mudugh. 'Uthman Mahmiid, the Majerteyn
sultan, was now hostile to the dervishes, whom he doubtless saw
47
Ibid., 122—3, translation of the Sayyid's letter.
676

SOMALIA, C. 1870-1910
as a threat to his own position; he would welcome a favourable

opportunity to attack them. Thus hemmed in, the Sayyid ventured
a pitched battle with the fourth expedition (October 1903-early
1904), at Jidbale in the upper Nugal valley (January 1904). The
dervishes were defeated and put to flight. Constantly harassed by
the Majerteyn, the dwindling band of dervish fugitives retreated
through Dulbahante and Warsangali territory. Now too weak to
impose themselves by force, they found no refuge there; and
ultimately fell back on Illig, from where they were ejected by a
British naval landing in April 1904.
The Sayyid, his movement apparently in ruins and himself a
butt for the gloating taunts of his many Somali enemies, still
negotiated as if he were victorious; but he could not escape the
humiliation of accepting an Italian proposal, in which the British
concurred, to settle him at Illig 'under the Italian flag' (March
1905).48 The British, believing that Muhammad 'Abdallah's
movement had collapsed, were content simply to let him go and
thank God they were rid of a knave. The Italian motives are less
obvious. Perhaps they hoped to use the Sayyid as a buffer against
the Ethiopians, who were by now pushing eastward almost to the
coast; or as a weapon against their nominally subject sultan,
'Uthmln.Mahmud, over whom they had in practice no control
whatever. But Muhammad 'Abdallah simply refused to admit
defeat; and at Illig between 1905 and 1908 he recreated the dervish
movement almost entirely by his gifts as a poet and propagandist.
Previously, most dervish poetry had been composed by the
Sayyid's court poets. But from 1905 Muhammad 'Abdallah
himself poured forth poems of defiance against the British, and
of insult and invective against the British-protected Somali, which
have earned him his reputation as the greatest of Somali poets; the
power of these poems is discernible even in the dim reflection of
translation.49
He was soon able to encourage some Dulbahante and War-
sangali lineages (who doubtless needed little encouragement) to
raid the protected Isaq clans; he almost guaranteed success by
detailing dervish detachments to organise and lead these raids.
There was still no British administration or occupation among
the Dulbahante to offset the Sayyid's rapidly reviving influence
48
For the terms o f the Agreement, Hertslet, Map of Afrit a, n o . 347.
49
Samatar, Oral poetry, 1 4 5 - 8 1 ; B. W. Andrzejewski and 1. M. Lewis, Somali poetry
(Oxford, 1964), 66-102.
677

there. He could not indeed directly threaten the Protectorate

administration; but his propaganda and the raids he sponsored
kept the government clans in so unsettled a state that the attempt
to administer the interior, where Britain had no direct interests,
seemed increasingly pointless and financially wasteful, especially
to a Liberal government in London. After the failure of the 1909
Wingate mission to reach an accommodation with the Sayyid, in
November 1909 London ordered complete withdrawal from the
interior. Only the strategically important coast was to be retained.
By April 1910 the evacuation was complete. The results were
catastrophic. It was hoped that the Isaq would combine in
self-defence against the Sayyid. Instead, they used their
government-issue firearms to settle outstanding godob on an
enormous scale, in an orgy of uncontrolled violence and stock-
destruction which led to general famine — the ' time of eating filth'
— and appalling mortality. Muhammad 'Abdallah soon returned
from Illig to the Dulbahante country. Here he built at Taleh a
massively fortified headquarters from which he dominated the
hinterlands of both northern and central Somalia. Not until
early 1920 was the Sayyid dislodged from Taleh by the British.
A harassed fugitive, but defiant and undaunted to the last,
Muhammad 'Abdallah Hasan died of influenza about December
1920.
Muhammad 'Abdallah Hasan never claimed to be the Mahdi,

although the British believed him to have done so and his Somali
opponents sometimes accused him - perhaps mockingly - of
Mahdist pretensions. True, he sometimes saw in European and
Ethiopian pressure a presage of the ' last days'; but so did other
Somali quite unconnected with his movement. What Muhammad
'Abdallah offered to the Somali was not a ' golden age' of perfect
equity as a prelude to the Last Judgement; but membership of
a purified, rigorist Islamic community under his own absolute
sway and totally dedicated to the expulsion of the infidel. For those
Somali who rejected this offer, he had nothing but total hostility
and contempt.
The immediate effect of the Sayyid's movement was to divide
rather than to unite the Somali. The Isaq clans, who had either
never supported him or had deserted him in 1900, feared and hated
him. They preferred British protection; and not only because their
trade was conducted through British-controlled ports. He
678

exploited Somali internal divisions for his own ends; he created

new and bitter internal dissensions among Ogaden Darod and the
Dulbahante. He prejudiced for two generations the peaceful
co-existence of Qadiriyya and Salihiyya in Somalia. The disruption
of 'clan balances' caused by dervish raiding, and the Sayyid's
deliberate flouting of the conventions that normally regulated
conflict, certainly contributed to the uncontrolled, almost geno-
cidal, violence of the 'time of eating filth'. Many Somali were
profoundly shocked by Muhammad 'Abdallah's total repudiation
of the pacific role of the traditional wadad; to them, a ' learned
shaykh [who] raids the people ruthlessly' was an evil aberration.
'All Jama' Habll (admittedly an Isaq poet) denounced the Sayyid
as a butcher of pious Muslims and a robber of orphans; but even
the Dulbahante 'All Duh condemned him as 'a worthless ruffian'.
Another Isaq poet, Jeni 'Ade, dared to say the unsayable: 'Better
than a murderous Muslim is a Christian who protects your woman
and child.'50
Yet the Sayyid intended that, under his inspired leadership, the
Somali should forget their divisions of clan and lineage and
become ikhrvan (brothers) united by jihad and an uncompromis-
ingly rigorous version of Islam. True, his concept of Somali
nationalism hardly went beyond acceptance of his own leadership.
He insisted that his followers call themselves dervishes, not
Somali. ' Somali' he used as a term of opprobrium for opponents
and doubters, mere dead wood that had to be destroyed. But his
organisation of the Salihiyya as a militarised religious autocracy
gave Somali the unprecedented experience of a powerful central-
ised authority that transcended all the traditional divisions of
Somali society. Above all, he incarnated three qualities deeply
admired by Somali: genius as a poet; total refusal to admit defeat;
and the superbly arrogant panache with which he defied and
taunted the mighty British. His twenty years of defiant and
successful resistance were undoubtedly an example and an in-
spiration for modern Somali nationalists. Retrospectively at least,
he has been very widely accepted as a Somali national hero. But
Somali have rejected the Sayyid's claim to unique religious
authority: today few Somali invoke his spiritual aid (madad) as
a 'saint of great blessing'.SI
50
S a m a t a r , Oral poetry, 1 1 9 , 1 3 8 , 1 6 3 - 7 .
51
I. M. Lewis, The modern history of Somaliland (London, 196)), 84.
679

CHAPTER 12
THE EUROPEAN SCRAMBLE AND

CONQUEST IN AFRICAN HISTORY
African history has been too much dominated by blanket terms,

generalisations which prompt comparisons rather than contrasts.
For the sake of clarity Africa's historians have dealt in universal
currencies, African despotism, an African mode of production,
African nationalism, the African worker, and so on. Without
common themes, it is true, differences would be impossible to
define, and advances in understanding come from realisations that
existing general models make an awkward fit for particular cases.
Equally, historical shorthand can be a substitute for thought, and
whole decades be passed off in a phrase. That has long been the
case with 'the scramble for Africa' or 'the colonial conquest'. It
is the period of Africa's history which has been most written about
and, almost for that very reason, perhaps the least actually
understood, smothered in high abstraction and wide generality.
Alien rule seemed to impose on Africa a crushing uniformity of
rulers' intentions. But the querulousness of the particular has now
been pressed long and insistently enough for this volume, and this
chapter attempts a new synthesis more accommodating to the
complex processes of time and the local spirits of place. Few
Africans were easily crushed by conquest; and they proved
thereafter to be but awkward subjects.
Partition and conquest together took a long time, much longer
than the period covered by this volume. Almost a century elapsed
between the French capture of Algiers in 1830 and the final defeat
of Abd al-Karim in the Rif mountains of neighbouring Morocco.
The French employed wooden-walled sailing ships in the first
enterprise, aeroplanes and tanks in the second. There was likewise
a hunded years' war between colonists and Xhosa on the eastern
frontiers of Cape Colony which started in 1799 - even if, like the
hundred years' war between England and France, it was
punctuated by periods of peace. Ethiopia was not conquered until
1935; and parts of northern Mozambique were not finally
680

SCRAMBLE AND CONQUEST IN AFRICAN HISTORY
occupied by the Portuguese until they were forced to take steps

against Frelimo's nationalist guerrillas in the 1960s. Conversely,
Angola's independence in 1975 came on the fourth centenary of
the first Portuguese military incursion on its territory. But even
if the conquest is confined, as largely it was, between 1874 and
1905, that is still three decades, a space for processes rather than
events.
Two complementary processes, the obvious ones, shape the
argument of this chapter. African authorities lost the race for
power and, as they did so, became increasingly divided. Europeans
accumulated power, but were not much less divided over how to
convert it into authority. The perplexities of rulers, conquered and
conquerors alike, stemmed in large part from their shared
difficulties in securing the more profitable obedience of those
whom they tried to rule. Many Africans resisted conquest, but
more Africans evaded the forms of work which were then devised
for them.
In the mid-1880s the first scramble for Africa took up a number
of threads of African history. Along Africa's coasts the men on
the spot, black and white, intrigued against each other, tested each
other out as allies and made informed calculations about each
other's intentions from within a long history of commerce and
conflict. Some European claims, those of Leopold or Bismarck
for instance, had no sort of local connection. But the second
scramble in the 1890s, which combined steeplechases into the far
interior with the forcible conversion of existing European
predominance into new control, was a more fantastic invasion
altogether; from most local perspectives it was' incomprehensible,
firing into a continent'.1 It is true that the sternest conflict of all,
between Boer and Briton, was also the least blind, bringing to a
barbarous conclusion what Boer propaganda did not greatly
exaggerate in describing as a 'century of wrong'. But elsewhere
the knowing African compromises of the past era of informal
empire could be as aghast as previous African ignorance of the
Europeans at the new white impatience for formal rule which was
armed now, as it had not been a decade earlier, with field artillery,
moral certainty, and some notion of how even tropical Africa
might, with railways, be turned to profit. This rupture with the
past, with its expensive wars and punitive expeditions, its tax
1
Joseph Conrad, Heart of darkness. Penguin edn (Harmondsworth, 1975). xo.
681

demands and enforced cultivation, its slave flights and its

rebellions, its epidemic mortality, social misery and disorder,
ruined both African production and European trade. In the first
years of this century the new colonial rulers everywhere tried to
repair the damage with civil reconstructions of government,
seeking, if with often unintended consequences, to make their
peace with the lords of the land.
FRONTIERS OF CHANGE
This common reciprocal process of the diminution and accumula-

tion of power was anything but a uniform experience. For it was
not the only process which was unfolding at the time. At least
three other frontiers of change were on the move: the frontiers
of trade and belief, of white settlement, and the linked scourges
of famine and epidemic disease. These processes divided Africans,
politically, socially and regionally. Their divisions sharpened in
face of the trauma of conquest and alien overrule, with its
combination of opportunity and menace.
Many African rulers had lived with external trade for centuries;
its control was one of their sources of wealth and power. It is
nevertheless useful to distinguish North and West Africa from the
rest of the continent by the intensity of the foreign commercial
involvement, the extent to which merchant capital and outside
finance rivalled political power in shaping relations of production,
the degree to which, in short, local sovereignty was already
compromised by extra-territorial privilege. By 1870 Mediter-
ranean Africa's rulers were deep in debt; their wealthiest subjects -
and, in Egypt, even the poorest - were beholden to European
merchants and bankers; their advisers and clerics were tormented
by the question how far Islam was compatible with Western
science and modes of government. Western Africa was divided
between two contrasted frontiers of trade and belief. The savanna
states of the Sudan, recently revolutionised by Islam, were
orientated to internal trade and the northern outlets over the
Sahara. Their respective statehoods were challenged by local
rivals, not by European power, capital or ideas. The Guinea forest
states on the other hand were open to the Atlantic, European
traders, gunboats and the subversive moral community of
Christianity. Their internal social inequalities differed accordingly,
682

FRONTIERS OF CHANGE
and so did their forms of expression. In the savanna political
opposition was couched in the language of Islam, in competition
between brotherhoods, in the apocalyptic hopes of Mahdism.
Slave populations, so greatly expanded in the warfare and drives
for intensified production of the nineteenth century, were some-
times unable to communicate in the language of their African
masters, let alone in the idiom of the Europeans who were to
come. They were baibayi, pagans, who, as captive outsiders, are
remembered as speaking 'by gesture as if they were deaf'.2 In the
forest kingdoms and coastal city-states on the other hand, political
dissent was beginning in places to be tinged with Christian doubt.
In the English city-state of Lagos, wives of the African Christian
elite taught, sewed and held soirees, while some of the labouring
poor had by the 1880s subscribed to a Mechanics Mutual Aid
Provident and Mutual Improvement Association; others in 1897
defeated the colonial government in a concerted strike.3 The first
colonial rulers faced therefore quite different forms of African
opposition in West Africa - external, culturally alien in the
savannas, but on the coasts often expressed in the language of the
conquerors, internal, legitimate, and difficult to ignore, however
much the African intelligentsia might be increasingly despised.
Such differences made a lasting imprint on the ways in which
men and women interpreted their world, their obligations towards
and their rights over their fellows. Even within the forest trading
frontier, one can distinguish between those societies which for
centuries had been penetrated by outside credit, and inland
peoples whose rulers managed for longer to control the terms of
exchange. The coastal Akan or Fante people to this day hold a
concept of honour clean contrary to that of the Akan of the
interior, the Asante. By the mid-nineteenth century the Fante were
already known as ' a nation of pedlars', and they still admire the
independently wealthy and educated gentleman. But for the
Asante, whose intimate contact with the coast dates only from the
18 30s, the model of excellence remains the chief, a wealthy man
certainly, but also a status inseparable from its eighteenth-century
2
Poly Hill, Population, prosperity and poverty: rural Kano, ifoo and 19/0 (Cambridge,
•977), i'4-
3
Kristin Mann,' The dangers of dependence: Christian marriage among elite women
in Lagos Colony, 1880-1915', J. Afr. Hist., 198$, 24, 1, 37-56; A. G. Hopkins, 'The
Lagos strike of 1897: an exploration in Nigerian labour history', Past and Present, 1966,
35. MS-ii-
683

4
origins in military service for the state. By the late nineteenth
century, Fante resistance to British claims took the form of
political argument; the Asante knew how to argue, but they still
had recourse to guns.
A similar contrast between coast and interior, but with quite
opposite implications for African reactions to conquest, can be
observed in East Africa. Earlier in this volume it has been argued
that the Maji Maji rising of 1905 behind the coastlands of German
East Africa was fired by the indignation of cotton-growers who
received worse market terms than rubber-tappers. Their sense of
injustice is incomprehensible save within a social culture
dominated by traders' credit.s But in the far interior, when the
peasants of Buganda started paying cash rents to their landlords,
they persisted in seeing the transaction as okusenga, rendering
service to their chief, a term linguistically akin to ' forming a line
of battle'. 6 Three paradoxes emerge from this disparity of
perception, each illustrating the complexities of conquest. The
intensity of East Africa's export trade in ivory, slaves and latterly
in rubber, was a quite recent phenomenon; its commercial
morality was provided by Islam, not Christianity; and its network
of goods and alliances had provided the means for new men to
assemble power over the interior's stateless societies more often
than it afforced old authority. Trading relationships therefore, for
all these reasons and more, stimulated frontiers of resistance
against European intrusion rather than, as on many of West
Africa's coasts, possible channels of accommodation. Moreover,
the Europeans with whom the coastal African traders had to come
to terms were not the English or French, with their long-standing
connections with Zanzibar, but newcomers, first the emissaries of
the Belgian King Leopold and then the Germans, who had to
break rather than exploit existing relationships if they were to
make good. The 'scramble' cannot everywhere be explained in
terms of a forcible European solution to a local crisis of informal
empire; at this local level it was by no means necessarily a part of
African history. The third paradox is that in East Africa, pre-
4
Kwame Arhin, 'Rank and class among the Asante and the Fante in the nineteenth
century', Africa, 1983, 53, 1, 2-22; T. C. McCaskie, 'Accumulation, wealth and belief
in Asante history, 1: To the close of the nineteenth century", ibid., 25-45.
5
See chapter 10, above, pp. 586—8.
6
John Iliffe, The emergence of African capitalism (London, 1985), 53. I am indebted
to Dr Iliffe for saving me from much error.
684

FRONTIERS OF CHANGE
colonial Christianity was associated less with the expansion of

trade than with the intensification of war. The first missionaries
arrived in the kingdom of Buganda, four months' march into the
interior, in 1877-an event of which the centenary was to be
marked, shockingly, by the martyrdom of Archbishop Janani
I.uwum at the hands of President Amin. After little more than a
decade their adherents had been transformed from the fire-armed
agents of royal despotism into the arbiters of power. A further
decade of warfare, this time in alliance with British conquest, saw
them transformed again, into the landlord chiefs to whom rent-
service was due. Insofar as Christianity in Buganda was linked
with trade, it was through the import of guns, ten thousand at least
by the 1880s, and so with the rearmament of African statehood
rather than with the accumulation of European power. The ad-
vancing frontiers of external trade and new belief were not
always harmoniously linked.
To the north and south of the tropics, white settlers were by the
1870s hostile to all forms of African belief, whether indigenous
or universal. It was not simply that the settlers' need to justify
their seizure of African land and exploitation of African labour
made them peculiarly receptive to the general hardening of
European racial attitudes which, except fora still liberal evangelical
minority, converted hopes for a civilising mission into the
nightmares of a defensive domination.7 For settlers, whether in
Algeria or South Africa, were not mere farmers, traders or
artisans, individuals facing Africa. They were at once the strategic
core and the most dangerous opponents of the colonial states
which protected them. The treatment of African Islam or
Christianity was subject as much to the changing structures of
local states as to the changing currents of overall European
perceptions of the non-European world. The settlers of Algeria
had from the start been the auxiliaries of conquest. They were
there to hold ground won by the military. And by the 1870s the
French army had afixedimage of the hostility of all those Muslims
who were not subject to a strict administrative patronage. It was
not possible to admit Muslims to citizenship of the Algerian state,
which was an overseas province of France, unless they renounced
their culture by a public act of apostasy. And yet in 1870, when
7
Ronald Hyam, Britain's imperial century, igij-1914 (London, 1976), 78-85.
685

metropolitan authority was smashed by defeat in war and the

deposition of the emperor, it was the local citizens, the French
settlers, who proved themselves able to paralyse the local
government — something which the Muslim revolts of the
following year could never do. The internal rebels were moreover
appeased by being given a free hand to plunder the lands of the
external ones. Muslims therefore suffered a twofold exclusion
from power. They were feared while they remained beyond
the pale of citizenship; they faced still stronger rivals within
it.
It was much the same for Africans in South Africa. By the 1870s
'the Christianity of free trade', an alliance for progress with
property-owning Africans, was being questioned from within the
Cape Colony by whites who were beginning to envisage a future
in which Africans were no longer a differentiated peasantry
amenable to co-optation but, rather, a proletariat, a sullen alien
mass.8 The Boer republics and the English colony in Natal had
never had any use for educated blacks. Nor had independent
African rulers, except where, as among the Tswana or in
Basutoland, the external threat of white conquest was yet more
dreadful than the internal wound of religious strife. Mission
Christianity did not offer a convincing model of political assimila-
tion on either side of the many frontiers between black and white
in the late nineteenth century; to rulers both black and white,
African Christians were amagqoboka, men and women with a hole
in the heart, full of treasonous thoughts. Where missions did
flourish, it was only as the patrons of minorities which had been
broken up by conflict, either in the wars of European conquest
on the eastern Cape frontier, or else in the wars of African
state-building beyond the northern border of Natal.9 South Africa
resembled Algeria also in the fact that its white settlers had shown
themselves to be the more formidable antagonists of the local
agents of imperial authority, whether as in the 1830s they had
trekked beyond its grasp or, as in 1849, infected by the ' revolu-
8
Stanley Trapido, '"The friends of the natives": merchants, peasants and the
political and ideological structure of liberalism in the Cape, 18)4-1910', in Shula Marks
and Anthony Atmore (eds.), Economy and society in pre-industrial South Africa (London,
1980), 247-74.
* Philip Mayer, Townsmen or tribesmen (Cape Town, 1961), 30; N. Etherington,
Preachers, peasants and politics in southeast Africa, iijj~tffo: African Christian communities
in Natal, Pondoland and Zululand (London, 1978).
686

FRONTIERS OF CHANGE
tionary genius of the age', they had tried to prise it open, at the
Cape, from within.10 African Christians were, by contrast, neither
prospectively loyal enough to be rewarded with, nor demonstrably
dangerous enough to be placated by, the construction of con-
stitutional bridges of accommodation across the gulf separating
black and white.
Nor was there any possibility that the local version of the
informal imperialism of free trade would long continue, by which
independent African rulers provided an external proletariat to
white farmers, planters and mine-owners. African labour was
certainly forthcoming, and it was supplied in as diverse ways as
the vegetable products of the West African traders' frontier. Some
young men (but no young women) walked hundreds of miles to
work in order to escape parental and chiefly authority, others went
as loyal citizens to earn the price of a rifle for their king's armoury,
but most were probably despatched under custody as war captives,
in transactions difficult to distinguish from the old Atlantic slave
trade." By the 1870s supplies were beginning to fall short of what
expanding colonial capitalism needed, and Boer farmers often
poached labourers whom planters and miners had hoped to hire.
But it was almost certainly for another reason that African
independence became intolerable. As late as 1879 it was still
possible for black South African armies to slaughter European
troops in open battle; the extraordinary number of Victoria
crosses then issued to the British expeditionary force paid indirect
tribute to Zulu valour. And African rulers were devoting their
earnings as labour recruiters to the purchase of guns; the Zulu
alone bought twenty thousand rifles in five years in the mid-1870s.
They, and others like them, had to be subjugated not because they
were predatory systems incompatible with white purposes,12 for
10
Stanley Trapido,' The origins of the Cape franchise qualifications of 18 5 5', ]. Afr.
Hist., 1964, j , i, 37-54.
11
Peter N. St M. Delius, 'The Pedi polity under Sekwati and Sekhukhune,
1828-1880', Ph.D. thesis, London, 1980, chs. ) and 6; Charles Ballard, 'The role of
trade and hunter-traders in the political economy of Natal and Zululand, 1824-1880',
Afr. Econ. Hist., 1981, 10, 3-21; Patrick Harries, 'Slavery, social incorporation and
surplus extraction: the nature of free and unfree labour in South-East Africa',/. Afr.
Hist., 1981, 22, 3, 309—30; Philip Bonner, Kings, commoners and concessionaires: the evolution
and dissolution of the nineteenth-century Stuart State (Cambridge, 1982), 80—4, 90-2.
11
As might be inferred from R. E. Robinson and J. Gallagher, 'The partition of
Africa' in F. H. Hinsley (ed.). The new Cambridge modern history, x i : Material progress and
world-wide problems rfyo-ifyS (Cambridge, 1962), ch. 22, especially pp. 617-20.
687

their pillage was compatible enough, but because they were feared
to be growing too strong.13
The third frontier of change which moved across Africa in these

years, that of famine and disease, was certainly more complex and
perhaps also more profound in its effects than the frontiers of
either trade or white settlement. Research on the history of famine
and disease, especially on their intellectual, therapeutic and
political history, is in its infancy. It is plain that for many parts
of Africa the era of colonial conquest was also an era of ecological
catastrophe; but it may not have been a uniquely catastrophic
time. In many African societies famine is an aide memoire to past
history, more regular than the cycle of age-sets, as certain as the
death of kings. Rituals of chiefship warded off drought precisely
because it was to be expected, and drought-famished people, all
Africans knew, were particularly susceptible to disease. Refugees
from drought might take their afflictions with them; they might
equally encounter new disease environments against which they
had built up no immunities. As the Himba of Angola observe,
'hunger does not kill, it is sickness that kills'.14 Armies and
caravans were hospitable to disease, and their movements were
not restricted to the late nineteenth century. The demographic
disaster which accompanied conquest may therefore not have
been unprecedented but it was not for that reason any the less
terrible.
Smallpox was a well-known camp-follower, so were typhoid
and cholera. It was not only white troops who succumbed to
malaria. Even measles was a killer for the young and starving.
Quite new scourges could appear from nowhere. The sandfleas,
called jiggers, which were inadvertently imported to the Congo
coast from Brazil in the 1870s, spread right across Africa in little
more than two decades. In the brief interval between their first
appearance in a new area and the improvisation of simple defences
against their burrowing and egg-laying beneath the skin, they
could literally cripple seedtime or harvest, maiming those culti-
vators who did not die of gangrene. It is not possible to draw
up any general rules about the causal connections between
13
C. F. Goodfellow, Great Britain and South African Confederation 1870-1881 (Cape
Town, 1966), 159.
M
Joseph C. Miller, ' T h e significance of drought, disease and famine in the
agriculturally marginal zones of West-Central Africa',/. Afr. Hist., 1982, 2}, 1, 23.
688

FRONTIERS OF CHANGE
ecological or epidemiological affliction and political action. In

Angola alone, drought could variously inflate political power
through the voluntary enslavement of the famished to the well
fed, destroy its credibility by dearth, or brutalise it by encouraging
raids on the grain or livestock of neighbours. Political opposition
likewise could be stifled as well as goaded by famine.15
It is therefore rash to suggest that there is nonetheless an
illuminating correlation to be seen between disaster and politics
if one measures the southward sweep of the great rinderpest
epizootic of the 1890s against the final forward push of European
conquest. Rinderpest came to Africa in the late 1880s, almost
certainly with European quartermasters. It seems that infected
cattle from Russian or British Asia were disembarked at Red Sea
ports to feed the Italian forces invading Eritrea.16 Entering Africa
at the Horn, the hinge of Africa's savanna belts, rinderpest struck
both west, along the Sudanic corridor between the desert and the
forest, and south, down eastern Africa's pastoral highlands; it
reached the Atlantic in 1892, the Cape in 1897. Rinderpest can
literally devastate not only herds of domestic cattle, killing
•anywhere between 30 and 90 per cent, but also many varieties of
game animals, antelope and buffalo especially. Domestic small-
stock, sheep and goats, are unaffected. It moves like a forest fire,
consuming almost everything at the first encounter, but patchily,
allowing free room for rapid regeneration by any livestock left
standing in the wake of its destruction. Cattle have been observed
to recover their numbers at an annual rate of 15 per cent, at least
for the first few years after devastation.17 If one assumes a 90 per
cent loss and a constant recovery rate uninterrupted by drought,
other cattle pests or raids from neighbours, up to twelve years
would still be needed for herds to recoup half their original size,
15
Jill R. Dias, 'Famine and disease in the history of Angola, c. 1830-1930',/. Afr.
Hist., 1981, aa, 3, 349-78.
16
John Ford, The role of the Trypanosomiaus in African ecology: a study of the tsetse-fly
problem (Oxford, 1971), 138. Ford later (p. 394) gives support to speculation that
diseased cattle may also have been imported to supply the British expedition of 1884-;,
in relief of General Gordon besieged by the Mahdists at Khartoum. The official history
of the campaign shows that the main force, up the Nile, was rationed with tinned beef.
The secondary expedition, based on the Red Sea port of Suakin, was supplied with
Russian (and therefore potentially infected) cattle. However, while the army's veterinary
officers were concerned about sickness in camels and horses, they appear to have noticed
nothing amiss with the store cattle. See H. C. Colvile, History of the Sudan campaign
(London, 1889), especially vol. 11, 189-94.
17
Ford, Trypanosomiases, 165.
689

and seventeen for their full restoration. For Africa's savanna

peoples, whose cattle furnished much of the means of subsistence
and still more of the working capital of family alliance and
political power, that was too long to sit out the recession in
equality of misery. Recovery was sought at other people's
expense, if not by stealing then by borrowing. This was every-
where the intention. The outcome had at least four variations,
depending upon the sources of capital available and its terms.
Where the disaster struck beyond the frontier of European
power, rulers of people whose herds were not totally destroyed
could exploit or redefine their prerogative claims on the remaining
cattle, redistributing them as the working capital of sovereignty.
The rulers of Bulozi and Rwanda did as much. Menelik of
Ethiopia did the same, and more; his continuing conquest of the
surrounding Oromo plains, often noted as pre-emptive of
European empire, was equally a reaction to rinderpest and
drought. Such rulers could face the European invasion with
rejuvenated authority, their actual power commensurate with its
ritual claims to ecological management. But there were other
rulers and sectional authorities who were too enfeebled to recover
on their own. If the rinderpest coincided with invasion, then an
astute alliance could give them borrowing rights on European
military resources. Cattle then became the working capital of
white conquest, seized from opponents and redistributed to allies.
The recovery from disaster of the northern Masai, the chief
auxiliaries of the British conquest of Kenya, provides the classic
example. Further south, however, the European conquest came
first. Then any vigorous African effort to claw back well-being
lost to locusts, drought and rinderpest could only take the form
of rebellion, as in Rhodesia's chimurenga. It was the irony of
rebellion here, as with the Kabylia uprising of 1871 in Algeria,
itself in part an attempt to recover from drought, that defeat
resulted in a still more ruthless denial of resources to the rebels.
Finally, in South Africa, rinderpest devoured pastoral economies,
white as well as black, which were in any case being displaced to
the margins of power by the rapid growth of mining. With their
vastly superior resources entirely unaffected by rinderpest, the
mining magnates could convert the remaining cattle into the
working capital of the labour market, as their recruiters paid out
690

FRONTIERS OF CHANGE
livestock in advances of pay to migrants' families.18 Industrial

capitalism expanded like chiefship, by rewarding the voluntary
submission of the poor.
The relationship between rinderpest and trypanosomiasis or
sleeping-sickness - the Black Death, as it has been called, of
colonial Africa19 — remains controversial. Within little more than
a decade of conquest, sleeping-sickness had wiped out terrifyingly
large numbers of the population in a ghastly swathe from the
mouth of the Congo to the eastern shores of Lake Victoria. The
tsetse-fly, the vector of the trypanosomes, was first reported in
Africa in the days of the prophet Isaiah.20 What caused the new
and deadly expansion of contact between theflyand man is
uncertain. Each of the rival explanations has difficulties. It could
be that the disruptions of conquest, and the freedom of movement
encouraged by the colonial pax or demanded by employers all
upset the caution with which men had previously insulated
themselves fromflyzones - but then war, long-distance trade and
large-scale population movements were scarcely new in Africa. It
may be, alternatively, that rinderpest destroyed the previous
balance between the civil, humanized environment and the wild,
since bush and game, hosts to the tsetse, moved faster than man
and cattle to reclaim the vacant pastures.21 This looks convincing
for the savanna but can scarcely be true of the forest. Whatever
the explanation, sleeping-sickness was the culminating disaster of
the early colonial period. African societies had been divided
against themselves and against their neighbours by an apparently
unprecedented concatenation of change. New sources of wealth
and power overturned or strained existing constitutions while
new beliefs questioned their cosmological bases; new misfortunes,
aweful in scale, mocked conventional explanations of evil and
18
Gwyn Prins, The bidden hippopotamus. Reappraisal in African history: the early colonial
experience in western Zambia (Cambridge, 1980), 85-7; Richard Pankhurst, 'The great
Ethiopian famine of 1888-92: a new assessment', J. Hist, of Medicine, 1966, 21, 2 and
3,95-124, 271-294; Richard Waller,'The Maasaiand the British 1895-1905: the origins
of an alliance', / . Afr. Hist., 1976, 17, 4, 5 29-5 3 - I am indebted to Dr Waller for his
advice in the writing of this chapter; T. O. Ranger, Revolt in Southern Rhodesia 1896-7
(London, 1967), 113, 147; C. van Onselen, 'Reactions to rinderpest in Southern Africa
1896-97', J. Afr. Hist. 1972, 13, 3, 473-88; William Beinart, The political economy of
Pondoland 1860—19)0 (Cambridge, 1982), 54—63.
" See above, chapter 6(B), p. 355.
10
Ford, Trypanosomiases, 466.
21
Both explanations are presented in ibid., which favours the latter one.
691

their therapies; Europeans everywhere became uncontrollable. It
was an age of anxiety,22 of sudden witchcraft panics, a time when
the politics of survival seemed to demand desperate tyrannies. It
was also an age of enquiry, when the excesses of old regimes raised
hopes for the new. Europeans were apprehensive too; they would
not otherwise have embarked on the unknown territory of the
scramble. Nor did they find security there, for military conquest
bred its own disorders. Civil administrations only shifted the
source of threat; early colonial regulations on everything from
rural trade to urban sanitation could look very much like
witchcraft accusations, treating the poor not so much as objects
of misery but as the baneful agents of affliction. Rituals of
cleanliness segregated the rulers from the ruled.23
THE CAUSES OF CONQUEST
An examination of these several frontiers of change can help to

illuminate the wide variety of the African experience of colonial
conquest; it can begin to suggest how much was changed by
conquest in each particular case, and how little. It can do nothing
to explain the causes of conquest, the main concern of most
previous analyses of the scramble. The chief difficulty has always
been to devise answers of wide explanatory power which also
tread delicately the maze of special cases. Plainly, the scramble for
Africa occurred because the European world was changing;
Africa's place in it was changing too. Economic explanations,
which insist that the world powers of the time were less
specifically European than characteristically capitalist, fail, in most
of their classical Marxist forms, to explain why capitalist
imperialism, a primarily economic phenomenon, should have
given way to European colonialism, an essentially political one;
nor do they have anything at all to say on African history.
Explanations which stress not the long-term tendencies of capital
but the short-term predicaments of capitalists get us much further,
particularly over the matter of timing. Hopkins' analysis of the
ways in which European traders tried to cut their costs during
22
Elizabeth Isichei, The Ibo people and the Europeans (London, 1973), 105.
21
John Dunn and A. F. Robertson, Dependence and opportunity:political change in Ahafo
(Cambridge, 1973), 96-121; Maynard W. Swanson, 'The sanitation syndrome: bubonic
plague and urban native policy in the Cape colony, 1900—1909', J. /\fr. Hist. 1977, 18,
3. 387-4>°-
692

the Great Depression of the last quarter of the nineteenth century,

and the severe limits to any remedial measures which were not
reinforced by political action, was very persuasive; moreover, it
showed how African authorities were similarly put under strain.
It was a regional analysis, addressed to the well-established
trading frontier of the Guinea coasts. It is employed more
generally in this chapter.24
Robinson and Gallagher's earlier explanation was continental
in scale and attracted attention because they took changes in
Europe and in Africa equally seriously. They showed how
increasing flows of trade and credit could induce insecurity on
both sides. African rulers were challenged from within; locally
established Europeans alarmed their rivals from without, since
any steps which they might take to repair their African relations
could only look like measures to exclude other European powers.
Their argument has been proved wrong in its single-minded
emphasis on the British occupation of Egypt in 1882 as the
starting-pistol for the scramble north of the Zambezi. More
generally, their explanation of the different African reactions to
conquest was functional rather than historical. Africans, they
believed, confronted or accommodated European initiatives by
reason of their contrasting societal constitutions rather than as the
outcome of a conflict between political classes who had been
divided by their experience of the moving frontiers of change.
They failed to appreciate that such conflicts might be sharpened
by commercial depression. A large gap in the Robinson and
Gallagher thesis was repaired by Sanderson, who showed that the
defensive aggression of British imperialism, their main interest,
was symptomatic of the crumbling of the global British hegemony.
British aggression in the 1870s was the spur, Britain's weakness
the opportunity for the imperialism of younger industrial states,
although one of them, France, was an old colonial rival. More
recently, the growing literature on the particular, and crucial,
South African case has stressed the growing importance which
almost all contemporaries of any influence, even in a Britain bound
by popular prejudice to laisse^faire, attributed to the state in the
management of the ecology of capitalism.25
24
A . G . H o p k i n s , An economic history of West Africa ( L o n d o n , 1973), c h . 4 .
25
R o n a l d R o b i n s o n and J o h n Gallagher, with Alice D e n n y , Africa and the
Victorians: the official mind of imperialism ( L o n d o n , 1961); G . N . Sanderson, ' T h e
E u r o p e a n partition o f Africa: coincidence o r c o n j u n c t u r e ? ' J. Imp. Cwealth Hist. 1 9 7 4 ,
693

All these analyses have their own power. In their combination
there lies an explanation of the scramble or, preferably, scrambles,
for Africa. But even when combined they do not adequately
consider the growing intensity of conquest as the nineteenth
century ended, nor do they ask how far Africans were competing
in the same field as Europeans, rather than merely reacting to a
series of crises in their mutual relations. There is also a hiatus
between explanations of the scramble and analyses of early
colonial government, which is often seen as a makeshift solution
to the unanticipated problems of rule, the problems, especially,
of making Africans pay taxes and making Africans work for
wages.
It is possible to put forward a more connected view than this
of the thirty and more years of partition and conquest, one which
does away with the false distinction between politics and
economics that has baffled so much of the historiography of the
scramble. The international history of the later nineteenth century
was shaped by what can only be called a revolution in power. It
was a European phenomenon, with global ramifications. European
nations became industrial states; they underwent a competitive
and uneven industrialisation of statehood. Non-European powers
were obliged to try to follow suit. Japan was the most con-
spicuously successful in this world-wide competition to meet the
rising costs of sovereignty. The Chinese gave it a name, self-
strengthening, but found it altogether too subversive of estab-
lished authority to give it substance. African polities failed too,
for this and for more fundamental reasons as well. It was an
enormously complex process, difficult enough to grasp today,
impossible to understand then. Statesmen were constantly
bewildered by the unprecedented expense both of diplomatic
bargains and, even more so, of war, except when it was war against
non-European, pre-industrial peoples. Of the many aspects of
industrialisation the most relevant to our theme seems to be the
contrast between the rather unsystematic interventions which
states undertook in production or trade, in order to give specific
assistance to particular interests in their moments of difficulty, and
the growing coherence of public policies for coping with the
3, i, 1-54; and chapter 2 of this volume; Shula Marks and Stanley Trapido,' Lord Milncr
and the South African state', History Workshop, 1979, 8, 50-80; Norman Etherington,
'Theories of imperialism in southern Africa revisited', African Affairs, 1982, 81, 324,
385-407.
694

domestic dislocations and international rivalries of capitalist

expansion. The lack of co-ordination in economic policy reflected
the difficulty of balancing claims from the competing proprietary
interests which were vocally represented in new liberal institu-
tions. The growing purpose in social policy acknowledged the
interest of the proprietary and professional classes as a whole in
maintaining the allegiance of the rapidly growing and potentially
dangerous mass of urban workers.
The first characteristic of industrialising states had two implica-
tions for imperialism. First, to protect particular industrial or
commercial interests in overseas markets was an easier political
option than any partisan intervention in the workings of the
political economy at home.26 This observation reinforces the
significance of the Great Depression in stimulating calls for
imperial measures. The second implication made it all the more
likely that such calls would be heard. For the industrialisation of
land and naval warfare since the 1840s had been largely the
business of private enterprise. Internationally competitive manu-
facturers were principally responsible for the development of
mass-produced infantry weapons of growing accuracy, reliability
and speed, and of ever more advanced means of propulsion and
destruction at sea. National arsenals were continually outstripped
in the techniques of killing, and the international arms trade could
not be subjected to national controls. The nerves of international
diplomacy were increasingly raw. There was a succession of
infantry, artillery and naval scares from the 1860s onwards. The
French decision in the late 1870s to outdo Britain in new naval
technology ensured that any new French empire would not be
defenceless; and the diplomatic destruction of Britain's African
paramountcy at the Berlin Conference of 1884—5 w a s completed
against the background of British public alarm at the obsolescence
of the Royal Navy.27
The European powers found that they had to come to terms
not only with the competitive demands of their domestic capitalists
and the mechanisation of warfare, but also with the social
corrosion of industrial life. This second feature of the industrial-
26
W . G . H y n c s , ' British mercantile attitudes towards imperial e x p a n s i o n ' , Hist. J.
1976, 19, 4, 969—79; P. J. Cain and A . G . H o p k i n s , ' T h e political e c o n o m y o f British
expansion overseas, 1 7 5 0 - 1 9 1 4 ' , Eton. Hist. Review, 1980, 2nd series 33, 4 , 485.
27
William H . M c N e i l l , The pursuit of power: tecbnolog, armed force and society since A.D.
1000 (Oxford, 1983), chs. 7 and 8.
695

isation of statehood meant that states and municipalities assumed
increasing powers of social management, while middle classes
took on increasing skills in managing. New, or newly expanding
professions — factory inspectors, teachers, civil engineers, civil
servants, military and naval officers - acquired a self-interested
faith in enlarging the sphere of public competence. It was
symptomatic that one of the earliest dreamers of formal African
empire, Charles de Freycinet, was both an engineer and a minister
of Public Works. He had reorganised the French railway system;
railway building in French Africa would merely extend it. More
than any other innovation, railways symbolised the links between
profit and power. By the 1870s they had begun to convert the
howling wildernesses of prairie and pampas into farms and
ranches producing cheap food for Europe's millions; in Asia they
were transporting grain from the populous Deccan. But railways
had also carried Prussia's armies to a series of stunning victories.
There was no reason why they should not have the same twin
effects in Africa, defended by but also defending European
control.28 The engine of empire, on land and sea, was driven by
steam. Medical science was another servant of military power.
Government-sponsored economic botany, botanical burglary
indeed, had broken the South American monopoly on the
cinchona bark used for quinine, the victor over the malaria which
laid colonial armies low.29 If conquest and its ecological
disturbances made Africa a more dangerous environment for
Africans, white men's graves at least would not be dug so often.
Repeating rifles meant that not so many had to be put at risk.
Europe's industrial states were acquiring the capacity to reduce
not only their own populations to order but the outside world
as well. Their official classes, if not yet their politicians, had long
had the appropriate attitudes. From the 1850s humanitarian hopes
for the free adaptation of Africans to European conceptions of
civil order began to fade. Thereafter, the view that Africans were
doomed to barbarous poverty unless rescued from without gained
an increasing hold upon the conventional official wisdom. Consul
2
" A. S. Kanya-Forstner, ' Military expansion in the western Sudan - French and
British style', in P. Gifford and W. R. Louis (eds.), France and Britain in Africa: imperial
rivalry anid colonial rule (New Haven and London, 1971), 409-41; G. N. Uzoigwe, 'The
Mombasa-Victoria railway, 1890—1902: imperial necessity, humanitarian venture, or
economic imperialism?', Kenya Hist. Keviev, 1976,4, 1, 11-34.
29
Lucile H. Brockway, Science and colonial expansion: the role of the British botanicgardens
(New York, 1979).
696

William Plowden's blast against Ethiopia's wretched rulers was

typical.
The wasteful government of a military oligarchy, the incessant struggles for
mastery, and uncertain tenure of all power, the careless sensuality of the chiefs,
the wretched administration of the law, the utter decay of learning, and the
selfish corruption of the priesthood have ruined a nation... Individuals are
found who feel that their nationality is lost... But it is to be feared that this
decay cannot be checked by any efforts of their own.'
Ten years later, on the opposite side of the continent, Faidherbe,

governor of Senegal, similarly called for reforms to end the
misrule of warrior chiefs, the tyeddo ' who under the influence of
alcoholic drinks commit all sorts of crimes, and who are a very
real calamity for the country they dominate... a country daily
destroyed by its chiefs \ 3 0 Similar censures could be read in almost
any report reaching Paris or London over the following twenty
years. Officials became convinced that Africa's latent productivity
could never be fully released until it was made more directly
subject to their own efficient probity. But it was not these
self-interested beliefs of the men on the spot, however endlessly
reiterated, which caused the scramble and the growing intensity
of conquest thereafter. For that one has to return to the more
general theme of the internationally competitive industrialisation
of statehood.
Two main trends are immediately noticeable over these three

decades: first, the growing purposefulness of European inter-
vention in African affairs; next, the changing nature of the
perceived African threats, from military confrontation to social
disorder and the collapse of export production. Three interrelated
considerations help to explain these trends: changing political
attitudes and capacities in the industrial states, their sharpening
international rivalry, and the consequences for Africans of their
own largely unsuccessful attempts to emulate the European
revolution in power.
European purposes and methods in the 1890s were very
different from the earlier phase of the scramble in the 1870s and
early 1880s.31 There were really two successive scrambles. The
30
Report on northern Abyssinia, 9 July 18)4 ( F o r e i g n Office confidential print, n o . 4 8 5 ,
1854), 2 6 - 7 ; Faidherbe t o Ma Ba, 23 May 1864, q u o t e d in Martin A . Klein, Islam and
imperialism in Senegal: Sine-Saloum, 1847-1914 ( E d i n b u r g h , 1968), 8 i .
31
For a full discussion see chapter 1 by G . N . Sanderson.
697

first scramble may well have gathered up otherwise unrelated

reactions by special capitalist interests to the onset of the financial
and trading slumps of the mid-1870s. But it is the relative
disconnectedness of commercial interest and political action
which remains striking. British political action anywhere, it is
true, was bound to defend material British interests, since there
was scarcely any part of Africa in which they were not already
privately engaged; and British attempts to repair the security of
existing colonies and African client states were undoubtedly the
first forward moves in the 1870s. But the new European forces
which then joined in were anything but sensitive to threatened
capitalist interests. Belgian capitalists, institutionally the most
advanced in Europe, were quite indifferent to the imperial dreams
of their king, Leopold II, who had been shopping unsuccessfully
for colonies in the Far East since the 1860s before turning,
thwarted, to Africa.32 There was almost constant hostility between
French traders in West Africa and the soldiers who took it upon
themselves to advance French power. Merchants loathed the
disruption of war, especially when they then had to pay for it. The
military had an altogether loftier self-interest. They believed
themselves, like the new professional classes all over Europe, to
be the ' vigilant guardian of the general interest', opposing their
'calm and equity' to the 'passion and greed' of commerce.33
Once all the diplomatic alarms had been soothed by the Berlin
Conference colonial officials did almost nothing to organise Africa
for capital, except to protect their metropolitan exporters with
discriminatory tariffs. Even this minimal external obligation was
banned by treaty from the Congo basin and East Africa. Outside
South Africa, where capitalists were in any case shaping the local
political institutions, the only vigorous economic policy to be
pursued was in Egypt, and in the circumstances that was the
doctrinaire and deflationary liberal policy of solvent government
and sound money. In tropical Africa, Britain and Germany tried
to banish the awkward problems of the proper relationship
between government and business by putting public power into
private hands, getting capitalists to organise their own salvation
through chartered companies. The French and Belgian conces-
sionary regimes in equatorial Africa showed the same desire to
32
See chapter 6(B) a b o v e , pp. 3 2 1 - 2 .
33
G o v e r n o r Briere d e I'lsle, 21 April 1878, q u o t e d in Klein, Islam and imperialism,
12411.
698

disown any public responsibility for private profit or loss, while

the hostility of the British government in Lagos for its over-mighty
neighbour the Royal Niger Company was notorious.
This relative distance between colonial governments and local
business interests was scarcely surprising in the economically
depressed conditions of the time and is in any case not the only
relevant area in which to look for connections between state and
capital. Official coolness towards commerce had two sources.
Local colonial governments were obliged to be deaf to traders'
protests as they looked to fresh duties to maintain their own
depressed revenues; and at home the traders' distress was music
to metropolitan manufacturers. The depression lowered the costs
of their raw materials; and African suppliers now had to compete
with rivals all round the globe, who were brought into production
by Europe's investments in shipping and foreign railways. In this
first scramble then, statesmen had no need to safeguard supplies;
they were plentiful enough; it was markets that were threatened.
During these same years governments were beginning to protect
their own domestic markets; and overseas protection needed little
enough of a state machine, one's own port officials rather than
those of some industrial competitor, and African promises of open
roads. Rivalry over the disposal of the crumbling Ottoman empire
in the Balkans had already given European powers the taste for
diplomatic bargains in other people's property. There was little
new in the first scramble for Africa.
The second scramble in the 1890s was a different matter; it was

altogether a new imperialism. The world was being transformed.
The Great Depression was ending, but not before it had created
a swelling flight from the land into Europe's cities. Urban
populations needed cheap food, and that came increasingly from
overseas. Industrial employers also required ever larger quantities
of raw materials. Commodity prices rose quite sharply from the
middle of the decade; trade was picking up and America's young
industries were buying raw materials which the New World had
previously exported to the Old. Europe had already suffered one
cotton famine, during the American civil war. There had been
surprisingly little worker unrest then. But there were millions
more workers now and they were becoming organised. Their
proletarianisation caused international diplomacy to become
699

obsessed with safeguarding supply. This final consequence of

industrialisation was responsible for the 'geopolitical claustro-
phobia' of the time, the 'feeling that national expansion was
running out of world space',34 and for the colonial scrambles and
naval races which that claustrophobia evoked.
The exploitation of Africa now required overseas production as
much as overseas markets. European officials had previously
denounced the oppressions of Africa's rulers; they now asked
themselves what other than slavery would make Africans work.
Masterless Africans, they anticipated, would be as dangerously
demoralised as the urban unemployed at home. Conquest turned
into a struggle for the control of labour. With textiles everywhere
the leading sector of Europe's industrialisation, by 1900 almost
all colonial governments thought that economic development
meant making Africans grow cotton for export.
THE AFRICAN RACE FOR POWER

Africans had long grown cotton, whether as free villagers or
slaves; African handloom weavers sold their cloth in bulk to
merchants, many of whom had sunk capital in indigo pits; rural
Kano was the Lancashire of West Africa." But unlike Europe, no
nineteenth-century African kingdom underwent an industrialisa-
tion of statehood. Many attempted self-strengthening. That
seemed essential in order to keep up with Europeans, so as to keep
them out. None succeeded. Few got as far as the Chinese in
discovering just how corrosive of political order the process could
be.
On all the frontiers of European rule and trade, Africans
attempted ' to maintain independence by borrowing and adapting
western institutions'. 36 African peoples in the interior had long
employed the same political and cultural smuggling in order to
confront or, failing that, to domesticate the power of their own
stronger neighbours.37 From Europeans they borrowed with a
34
Robinson and Gallagher, 'The partition of Africa', 626; see also C. C. Wrigley,
' Neo-mercantile politices and the new imperialism', in Clive Dewey and A. G. Hopkins
(eds.), The imperial impact: studies in the economic history of Africa and India (London, 1978),
20-34.
" P. Shea, ' T h e d e v e l o p m e n t o f an export oriented d y e d cloth industry in K a n o
emirate in t h e n i n e t e e n t h century', P h D . thesis, W i s c o n s i n , 1 9 7 ; .
36
T h o m a s H o d g k i n , Nigerian perspectives, an historical anthology ( L o n d o n , i 9 6 0 ) , 4 9 .
37
A i d a n W . Southall, A/ur Society (Cambridge, 1956), c h . 8.
700

canny eclecticism. Kings and chiefs well knew the power of ideas;
they themselves continually manipulated the ideas of power. But
literacy, one of the most obvious advantages of Muslims or
Europeans, had its own ' great and hidden powers' unlicensed by
chiefly authority. So Tlhaping chiefs in South Africa were
prepared to welcome resident missionaries as conduits of
knowledge of the outside world only on condition that they
neither taught nor preached religion. King Eyo of Calabar's Creek
Town became quite fatherly in comforting missionaries for their
failures in an evangelism he was determined to prevent. The
asantehene, Mensa Bonsu, for all his passion for self-strengthening,
refused to select prospective school pupils in 1876, 'for Ashantee
children have better work to do than to sit down all day idly to
learn hoy! hoy! hoy! They have to fan their parents, and do other
work, which is much better.' Moshoeshoe of Basutoland,
diplomatic to the last, managed to die just two days before his
baptism in 1870, putting off for ever what he had been postponing
for years.38
Apart from Liberia, there were only three attempts at
christianised African independence beyond the European
frontier - the Methodist-inspired Fante Confederation on the
Gold Coast, the Anglican Egba United Board of Management in
Yorubaland and the Congregationalist Griqua captaincies in
South Africa. The first attempted a federation of established
chiefdoms, the other two were refugee settlements. All of them
were foredoomed to extinction by the advance of white rule, but
none of them looked like resolving their own conflicts between
centrally financed authority and the individual commercial or
farming interests of their most prominent supporters. King
Khama's apparently reckless Christian renovation of his Ngwato
people was rational only because he was a factional leader
dependent upon external Christian support before he became a
national one. And his reform programme succumbed to the
danger that his more cautious contemporaries all feared. In
seeking to control intrusive white power by harnessing it to his
38
T. C. McCaskie, 'Innovational eclecticism: the Asante empire and Europe in the
nineteenth century', Comp. Studies in Soc. and Hist., 1972, 14, 32; Anthony J. Dachs,
'Missionary imperialism - the case of Bechuanaland', J. Afr. Hist. 1972, 13, 4, 648;
J. F. A. Ajayi, Christian missions in Nigeria, 1X41-1891: tie making of a new elite (London,
1965), 100-5; David Kimblc, A political history of Ghana: the rise of Gold Coast nationalism,
lijo-ifiS (Oxford, 1963), 75; Leonard Thompson, Survival in two worlds: Moshoeshoe
of Lesotho, rji6-iSjo (Oxford, 1975), 323.
7OI

own interests he was using a lever which turned in his hands.
White concessionaires moved in. More Africans with ploughs,
waggons and education made their own private contracts with
them. The refusal of self-strengthening in the outsiders' idiom
yielded no better results. Protected by distance in the far interior
from the most palpable effects of white power, King Lewanika
of Bulozi might seem to preserve his own hidden resources of
authority in the ritual allegiance of his people by a carefully
preventive accommodation with the Europeans; but the inde-
pendence which he retained by working on the outsiders' mis-
understandings was itself no more than a conceptual illusion.39
The first demand made of most missionaries was that they

should supply guns. Ethiopia preserved her independence into the
twentieth century because she had guns, not because she was a
Christian nation. Menelik's victorious army at Adowa in 1896, one
hundred thousand strong, was composed largely of firearmed
infantry. Most of the soldiers doubtless carried muskets, but as
many as twenty thousand may have been issued with breech-
loading rifles. Africans imported enormous quantities of guns in
the nineteenth century, over one hundred thousand per annum
by the 1860s, and a scarcely credible total of 16 million for the
continent as a whole by 1907.40 Comparatively few of these were
rifles, a new tool of destruction which became available to
Africans only as European armies discarded earlier models for still
more lethal versions. African rulers had armouries rather than
arsenals. Their military technology was improved, not as part of
an internal process of the industrialization of statehood, but as a
rearmament of statehood by imports from without. They strove
to emulate the most obvious effect of industrialisation, without
undergoing its ordeal. There were exceptions. Muhammad 'All
came nearest to being an African Peter the Great, mobilising all
Egypt's resources in his attempt to industrialise by forced
39
A d u Boahen, 'Politics in Ghana, 1800-1874', in J. F. A. Ajayi and Michael
Crowder (eds.), History of West Africa, vol. 11 ( L o n d o n , 1974), 2 4 2 - 5 9 ; Saburi
0 . Biobaku, The Egba and their neighbours, 1X42-1/72 (Oxford, 1957); Robert Ross, Adam
KoM's Griquas: a study in the development of stratification in South Africa (Cambridge, 1976);
1. Schapera, Tribal innovators: Tsvana chiefs and social change 1791-1940 (London, 1970);
Neil Parsons, 'The economic history of Khama's country in Botswana, 1844-1930',
in Robin Palmer and Neil Parsons (eds.), The roots of rural poverty in Central and Southern
Africa ( L o n d o n , 1977), 113-43; Prins, Hidden hippopotamus, 3.
40
R. A . Caulk,' Firearms and princely power in Ethiopia in the nineteenth century',
J. Afr. Hist. 1972, 13, 4, 6 2 6 ; Gavin White, 'Firearms in Africa: an i n t r o d u c t i o n ' , / .
Afr. Hist., 1971, 12, 2, 182.
702

marches. Successive sultans of Zanzibar tried without much

success to foster sugar-refining, shipbuilding and coconut oil-
milling.41 'Abd al-Qadir in Algeria, Samory in the western Sudan,
Tewodros in Ethiopia, the kings of Imerina in Madagascar,
Ibadan's war captains, all created the beginnings of independent
arsenals. None got very far. Africa's relative technological back-
wardness was rooted in underpopulation and the obstacles which
this presented to political exploitation.42 More fundamentally, its
metallurgy, capable enough of producing high-quality iron, even
steel, was quite unable to break through the ecological constraint
on its development which Europeans in a similar predicament
overcame in the eighteenth century by turning from charcoal to
coal as the source of energy. Africa's iron-working remained tied
to small units, immensely costly in the labour which they
employed, not only in smelting and smithing but also in felling
trees and charcoal burning.43 Imported iron bars were increasingly
used to substitute for labour, but they were of inferior quality.
They were no answer to the Bessemer process or mass production
precision-engineering, nor yet did they multiply backward
linkages in domestic production. Africans had only two options
in rearming. They could continue to import muskets, what today
would be called an ' appropriate' or ' intermediate' technology of
destruction, inefficient but able to be repaired locally, and for
which powder and shot were also easily made. Rifles were not
only more expensive, they implied a greater technological
dependence. They were not easily repaired, their cartridges had to
be imported. The military rearmament of African statehood
turned out to be as vulnerable as its constitutional and cosmo-
logical renovation to its external suppliers. It had four
varieties of consequence for African societies before the conquest,
three of them — debt, disintegration and tyranny — being
disastrous to self-strengthening.
The first consequence was external debt and political dissolution

within. Egypt and Tunis provide the classic cases, but they were
very different ones. Nor were their troubles due entirely to their
41
C. S. Nicholls, The Swahili coast: politics, diplomacy and trade on the East African
littoral r/fS-it/i (London, 1971), 82-3, 366-9.
*z Jack G o o d y , Technolog/, tradition, and the state in Africa ( O x f o r d , 1971); H o p k i n s ,
Economic history of West Africa, 8-77; William Ochieng', ' Undcrcivilisation in Black
Africa', Kenya Hist. Review, 1974, 2, 1, 45—57.
*3 Candice L. Gouchcr, '"Iron is iron 'til it is rust": trade and ecology in the decline
of West African iron-smelting',/. Afr. Hist., 1981, 21, 2, 179-89.
7°3

rearmament; their governments had wider ambitions, forced on

them by proximity to Europe. Both were hit by the general
commercial and financial crises of the 1870s and, in Egypt's case,
by the collapse of the cotton boom when American suppliers
re-entered the market after the end of their civil war. Investment
in government reforms and public works toppled into over-
capitalisation. External borrowing and the penetration of
European merchant capital put growing sectors of economic
activity under consular protection, exempt from fiscal controls.
Both countries suffered political crises, but of divergent sorts.
Tunis presents the purest example of the inability of a pre-industrial
but increasingly open political economy to bear the weight of
increased government spending in face of the European threat.
Its producers were not only increasingly tied to European rather
than indigenous capital, they suffered extraordinary demographic
reverses in the first half of the nineteenth century with a
succession of droughts, famines and epidemics of cholera and
plague. Typically, the articulation of two forms of economic life,
non-capitalist and capitalist, did not eradicate old hazards but
added a new one, the trade cycle. The beylical government then
submerged its subjects in a 'tide of taxes' and judicial reforms
which removed decisions from local competence. The doubling
of the deeply regressive majba or poll tax in 1864 provoked all
the parochial divisions of the unreformed state to rise in arms
against the new tax — but also, as the revolt collapsed, to fight
against each other.44 Egypt's opposition was based not on the old
but the new, army officers and larger landowners, some of them
attracted by the ideas of Islamic reformers. Egypt's army was no
longer especially large. It took up less than 10 per cent of
government expenditure in the mid-1870s, by comparison with
the 60 or 70 per cent devoured by the external and internal debt.
But the army was more susceptible to retrenchment than the debt,
and when colonels protested and the government split over how
to apportion taxes, then something like a national movement was
born.
None of this explains the European invasions of the early 1880s,
the French in Tunis, the British in Egypt. The causes of conquest
** Lucettc Valensi, Fellahs Tunisiens: I'e'conomie rurale et la vie dts campagrus aux r/e et
if1 siecles (Paris, 1977), 281—347, q u o t a t i o n from p . J J 8 ; Bice Slama, L'insurrection de
1S64 en Tunisie ( T u n i s . 1967).
704

are to be found more in Europe's Mediterranean squabbles for

the rotting scraps of the Ottoman empire and the self-serving
panics manufactured by European officials on the spot.4S The
point is, rather, that governmental reform and its collapse had
created increasingly self-conscious coalitions of new men, Islamic
modernists and army officers in Egypt, the westernised graduates
of the Sadiqi college in Tunis. Such men took part in the general
resistance to European financial control. They then dared to hope
that imperialism might, by other means, continue their own
nationalist opposition to the traditional autocracies which had first
created and then turned on them. Finally, they were bitterly to
disown imperialism when, by its alliance with old notables, it
proceeded to betray their dreams.
The second effect of guns was the disintegration of African power,

whether by decentralisation or fragmentation. Royal power could
be eroded as the increasing flow of guns, like other imported
commodities, outstripped the capacity of rulers to enforce mono-
polies on their supply. They got into the hands of those who were
able to pay, or of subordinates who then used them on their own
account. Elsewhere, the fragmentation of already fragile, stateless,
authorities followed upon the expansion of the traders' frontier
in the tropical African interior. Gun imports financed the spread
of commercial hunting and gathering, ' a type of slash-and-burn
or "swidden" commerce' which with little capital and much
destruction thrived on elephant hunting and the collection of wax,
gums and rubber.46 The characteristic European figure was the
trader Kurtz, hacking his way into his, and Africa's, heart of
darkness. Instruments of the chase for hunter bands, guns were
instruments of protection for merchant princes and, for political
bands, of power.
Decentralisation can best be seen in Morocco and in the
interlacustrine kingdoms of Uganda. Morocco provides a third
variant of failure in self-strengthening, or tanîmat as it was known
45
Ibrahim A b u - L u g h o d , "The transformation o f the E g y p t i a n elite: prelude t o t h e
'Urabl Revolt', Middle East J., 1967, a i , 3, 315-44; A. Scholch, 'Constitutional
development in nineteenth-century Egypt: a reconsideration', Middle Eastern Studies,
1974,10, 1,3-14; Roger Owen, Tbe Middle East in tie World economy 1800-1914 (London,
1981), 1 2 2 - 5 2 ; A . Scholch, ' T h e " M a n o n t h e S p o t " and t h e E n g l i s h o c c u p a t i o n o f
E g y p t in 1882', Hist. ]., 1976, 1 9 , 3, 7 7 3 - 8 5 .
46
A . G . H o p k i n s , 'Imperial business in Africa. Part 11: Interpretations', J. Afr.
Hist., 1976, 17, 2, 280.
7° 5

in North Africa, after the usage of the Ottoman Turks. Successive
sultans tried to achieve what so many of their contemporaries were
also attempting with scarcely more success, to lay hands on more
internal resources and to use them to transform not production
but the basis of power. Their objective was to construct a
coherent political class which would be protected from faction by
the denial of regional privilege, from out of the old mosaic of
exemptions and obligations which had previously been sustained
by civil war and intrigue. Morocco's failure in this project was
more spectacular than any; it was riddled with contradictions.
Tansîmat enlisted with external borrowing and contracts for
supply the very Europeans it was designed to exclude. It
scandalised the clerics but established no new learning. It antag-
onised the military tribes, the jaysh, but could not afford a new
model army. It expropriated from office notables who then
recouped their power by commercial brokerage between their
followers and foreign buyers. Loyal tribesmen became disgruntled
peasants. Searching for a bureaucracy, the sultan found an alliance
with the overmightiest of his subjects. New taxes could not keep
pace with the military drain on his treasury as he fought off rivals
for favour who protested in arms. Dissident highlanders on the
periphery could afford rifles as well as the city-dwellers of the
plain. By 1912 the French could take over from within, their
protectorate quite literally a deliverance of the sultan from the
rebels encamped about his walls.47
In Morocco the French occupation worked, with however ill
a grace, to the strengthening of the sultanate. The pre-colonial
decentralisation of power in the south-western kingdoms of
Uganda had quite the opposite colonial outcome. There the British
came to terms with subordinate power-holders rather than with
kings. In Uganda it has also been the historiographical tradition
to look to a different context of explanation. Self-strengthening
was certainly important, in face of the Egyptian advance up the
Nile. And, as elsewhere, kings began to lose control over the guns
and ideas brought in by Arab traders and Christian missionaries.
But the rearmament of these kingdoms must also be understood
47
Edmund Burke III, Prelwlt to protectorate in Morocco: precolonialprotest and resistance,
1860-1912 (Chicago and London, 1976); Ross E. Dunn, 'BO Himara's European
connexion: the commercial relations of a Moroccan warlord', J. Afr. Hist., 1980, 21,
2,23 j—j 3 ; David Seddon, Moroccan peasants: a century ojchange in the eastern Kif, fy
(London, 1981).
706

in the context of the unceasing conflict between monarchy and

aristocracy in the enjoyment of tribute, and between young kings
and their fathers' chiefs, each party often strengthened against the
other by the resources which they had mobilised in succession
war. Two kingdoms in particular were involved, Buganda and
Bunyoro, each competing with the other for land, subjects, and
access to external trade. Their kings had each created what have
been called new-model warbands by the 1880s, Buganda copying
Bunyoro's innovation. Royal pages and alien adventurers alike
were enrolled in standing regiments of fusiliers, to outgun the
nobility's militias. They lived by loot and had a dreadful reputa-
tion. In Buganda they were nicknamed the bapere, 'those who
had tarnished or gone bad'; in Bunyoro they were called
abarusura, the casually brutal thieves. They were designed to
subdue provincial chiefs but, under the stresses of British conquest,
most of their leaders turned against their kings. Their internal
factional disputes, clothed in Buganda in denominational division
between Muslims, Catholics and Protestants, determined the
politics of colonial state-building. In the late 1890s the British
hunted down the fugitive kings, Mwanga of Buganda and
Kabarega of Bunyoro, assisted by turncoat African regimental
commanders. Under the Uganda Protectorate, kingship was to
become a legal fiction, no more than an ideology to be exploited
by and against the new oligarchies over whom kings had lost
control.48
Many African kingdoms and chieftaincies have traditions of

genesis in which the founding hero is a hunter. Looking at the
nineteenth century one can see why. Hunters must be well armed,
capable of organising associations of men over large territories,
and cunning in their understanding of the hostile environment out
in the bush, the frontier wildernesses between one settlement and
another. The chase provided many articles of use and exchange,
food, clothing, and regalia; it protected agriculture and grazing
48
Michael Wright, Buganda in the heroic age (Nairobi, 1971), 27; G. N. Uzoigwe,
'Kabalega's Abarusura: the military factor in Bunyoro', Proceedings of the University of
"East Africa Social Science Conference if6g-6f (Nairobi, 1969), 308. Richard Waller, 'The
traditional economy of Buganda', M.A. (SOAS), London, 1971; Michael Twaddle,
' The Muslim revolution in Buganda \ African Affairs, 1971,71,282,54-72; D. A. Low,
'Warbands and ground-level imperialism in Uganda, 1870-1900', Hist. Studies, 1975,
16, 65, 584-97; Edward I. Steinhart, Conflict and collaboration: the kingdoms of Western
Uganda 1890—1907 (Princeton, 1977), 73—97-
7°7

from depredation; it was a sport. In the nineteenth century

hunting acquired a huge new market in the European and Indian
demand for ivory; if elephants were shot out, one could hunt men.
The demand for labour never slackened; if slaves could no longer
be legally exported overseas, African andi increasingly, local
European producers had an insatiable demand for workers on
plantations and in porterage — although it was also in porterage
that wage labour seems first to have emerged.49
These remarks are especially relevant to central Africa, from
the Congolese and Angolan coasts across to the Indian Ocean, and
the Sudan south of Khartoum. The expansion of Africa's firearmed
hunting frontiers was too short-lived a phenomenon for one to
determine whether, but for European conquest, the nineteenth-
century pioneers would have come to dominate the traditions of
political genesis recounted in future ages. Their immediate
achievements were almost entirely destructive, but that is generally
the way with processes which can, after the event, be seen as the
early stages of state-building, especially when the material base
of power is also being revolutionised.
Four stimuli to the competitive rearmament of power can be
seen on Africa's hunting frontiers, as the export staples of ivory
or rubber became locally exhausted, as the frontier of production
moved on into the interior. When a local superiority in production
was lost to competitors elsewhere, hunting communities or their
rulers tried to seize greater hold on the external trading system,
seeking to extract from transit dues what they had previously
earned from export sales. This was the motive behind Mirambo's
drive to control central Tanganyika; his name, 'corpses', was
coined by his rugaruga warbands, kinless teenage hooligans with
guns. More generally, rulers deprived of ivory duties turned to
export the asset which remained to them, subject men and women
or prisoners of war. Internal justice became more cruel, dependent
relationships more exploitative. In the 1870s and 1880s civil wars
destroyed such previously coherent polities as the Lunda and
Luba states of southern Zaire or the Mangbetu kingdom in the
north.
These were rulers' reactions. For their part, long-distance
traders compensated with two other devices, seizing power and
49
Ivor Wilks, Asante in the nineteenth century: the structure and evolution of a political order
( C a m b r i d g e , 1 9 7 5 ) . ' 9 5 ; I l i f f e , Emergence of African Capitalism, 7 1 8 .
708

controlling production. Their access to the capital which was
advanced by coastal exporters, European and Indian, allowed
them to organise larger caravans of porters and to arm them with
guns. They made their depots into stockaded strongpoints. As
export prices fell in the 1870s they enforced harder bargains with
their suppliers. African and Arab merchants in the interior were
doing for themselves what European traders on the coasts were
demanding of their governments. Trading diasporas became
networks of political control; trading became raiding. Armed
force was an investment. It reduced operating costs, at the expense
of the local inhabitants. The effects can most brutally be seen in
the southern Sudan, on the upper Nile. Raids provided porters,
guards and agents with slaves in lieu of pay. Local harvests were
comandeered to feed trading posts, cattle were seized to supply
the working capital of the ivory trade. By similar means Tippu
Tip built up his trading state to the west of Lake Tanganyika.
These were very primitive exercises in capital accumulation.
Finally, there was a potentially more constructive development.
As the power of existing rulers was fragmented by faction, as the
broken channels of tribute then ceased to generate the articles of
trade, as reserves of ivory or rubber became exhausted, so external
traders, lacking now the producer partners whom their compe-
tition had destroyed, turned to the direct control of production
themselves. They did so in two distinct ways, by involuntary mass
migration and the deliberate imposition of overlordship. The
Chokwe people of northern Angola provide the best example of
the first. By the middle of the century their colonising settlements
were obliged to become ever more mobile, as the hunting,
wax-collecting and rubber-tapping frontiers advanced. Chokwe
populations expanded as they moved, absorbing wives from other
populations who pawned their women in return for partnerships
in trade. As the Chokwe men's trading frontier enlarged, so too
did their control of women's agriculture. From this productive
base they hired out their guns in civil wars; by the 1880s they had
destroyed their Lunda hosts' patterns of power. Nyamwezi
merchant adventurers from the east probed into the same area of
what is now southern Zaire. Their methods were more direct than
the agricultural infiltration of the Chokwe. When they had
ransacked the Luba empire's ivory, their leader Msiri carved out
an outlying province, to control its copper production, instead.
709

Thus, a conquest-state was born, one of many at the time. Samory

Ture's 'Dyula revolution' in the far west of Africa, in the
interland of Sierra Leone, is the most famous example of this
transition from trade to dominion. By threading together a
military coalition along the commercial network of the Dyula
diaspora, he was able to monopolise both the export of ivory and
the import of more and more guns, rifles among them.
In this violent and discordant welter of the disintegration of
old tributary powers and the imposition of new autocracies on
previously stateless societies it is possible to discern a particular
variant of the common process called, at the time, the opening
up of Africa. No African societies had previously been entirely
closed. Ethnic identities were mutable bargaining counters to be
used in the external relations of trade, tribute or marriage as well
as being the cultural celebration of men's domestication of their
parochial botanical environments. Nor was political violence new.
What was unprecedented in nineteenth-century central Africa was
both the volume of the commodities traded and the destructive
power of guns when handled by quite small bands of men
practised in their use. All previous equations of profit and power
were upset. All branches of domestic production became more
profitable in their expanded linkages, not just hunting but
farming, herding, mining and smithing too; and power was
potentially released from political responsibility. It degenerated
into force in response to economic stress, whether drought,
depletion of the export staple, or declining export values. The
1870s and 1880s were not short of all these forms of crisis. Traders
repudiated alliances with chiefs, chiefs broke faith with their
people. Political and commercial power not only combined, it
became militarised. It was resisted. Cultivators, like traders, took
to stockades. They turned into villagers, against the insecurities
of Africa's opening out.
What was being destroyed in this process were the unstable but
workable oppositions which underwrote previous dispensations
of power, between kings, their officers, and free kinsmen; between
these citizens and the unfree; between tributary hierarchies and
market exchange; between customary or ritual constraints and the
individual agency of big men. All social categories and their
inherent rights were put in question, vulnerable to forcible
redefinition. Elaborated political structures gave place to a
710

single-minded African imperialism of unfree trade. Merchants
constructed their own authorities, armed plutocracies, one-party
states which ruled without consent. Preying on the connections
between the many branches of production, they represented none.
Samory's fruitless attempts to create a more inclusive political
order are symptomatic. He tried to use Islam as a unifying
instrument of political discourse only to find that it provoked
sharper opposition; he was obliged to pack his councils with his
own men. Other merchant adventurers also failed to make their
peace with the land. They relied less on citizen militias than on
alien mercenaries, small bands of ruffians scarcely more numerous
than the colonial troops by whom they were later defeated. Under
the shock of conquest these new autocracies disappeared almost,
but not quite, without trace. They left behind no political
communities with which colonial rulers might come to terms but
diasporas of traders and porters once again, and sometimes
colonial Africa's first migrant workers. Their agricultural popu-
lations generally resumed older identities. One reason for the
extinction of these secondary empires, as they have been called,
was that colonial states jealously represented rival means of
articulating or yoking producers to overseas accumulation. But,
more fundamentally, these African merchant states had created
not citizens who could cling to differentiated traditions of past
freedoms, only subjects differentiated by intensities of extortion.50
Their true successors were the concessionary companies which
operated within the early colonial regimes in much of this same
central zone of Africa, in French and Belgian equatorial Africa
or in Portuguese Mozambique.
Some of Africa's older kingdoms survived the European conquest,

at least in a resurrected form powerful enough to inspire and
divide the twentieth-century politics of colonial nationalisms. As
elaborated structures which bound together the holders of
competing material resources into collective ruling classes, they
50
John lliffe, A modern history of Tanganyika (Cambridge, 1979), 62; David
Birmingham, 'The forest and the savanna of Central Africa', in John E. Flint (ed.),
The Cambridge History of Africa, v: c. 1790-c. 1870 (Cambridge, 1976), ch. 7; Richard
Gray, A history of the southern Sudan igjf-rgSf (Oxford, 1961), 46-69; Yves Person,
Samori: une revolution dyula, j vols. (Dakar, 1968, 1971, 1975); Joseph C. Miller, 'Cokwe
trade and conquest in the nineteenth century', in Richard Gray and David Birmingham
(eds.), Pre-colonial African trade (Oxford, 1970), 175-201. Philip Curtin, Steven Feierman,
Leonard Thompson and Jan Vansina, African history (London, 1978), 559-43, 557~9.
389.

succeeded in coping with the mounting contradiction between the

costs of rearmament and the subversions of external capital.
Their rulers had to overcome two levels of difficulty. They
required state servants to trim the powers of the provinces, new
men with a self-interest in centralising older, more diffused, rights
to the hierarchical management and enjoyment of the goods of
their political culture. But kings could not rest there; they then
had to decide whether they could fight the temptations of their
new officials. These were minded to enforce a further redefinition
of the bargains between private wealth and communal welfare.
They could see that fortunes might be made in handling the
external commerce which funded self-strengthening. As the
global terms of trade began to turn against primary producers in
the 1860s, so conflict intensified between the public interest of
states and the private ambition of their ruling classes. Meanwhile,
European traders were becoming more impatient, the politics of
African survival more desperate. Internal and external anxiety
bred tyrannies of rule. Not all African kingdoms became tyrannical
and not all African tyrannies were the same. They faced up to
European conquest crippled by very different rivalries between
crown, officials, and community. Three examples illustrate the
various possibilities, the Muslim theocracy of the Mahdiyya, and
the kingdoms of Imerina in Madagascar and Asante in West
Africa.
Tyranny is a term to be used with caution. European observers
at the time naturally seized on any evidence they could for African
misrule. Conquerors prefer to be seen as liberators. The Mahdist
state in the Sudan was possibly the most traduced. But there is
good internal evidence for the self-destructiveness of its autocracy.
Under Khalifa 'Abdallahi, successor to the Mahdi, there was
irreconcilable conflict between the Mahdi's earliest and most
dedicated followers who were drawn from the farmers and traders
of the Nile's riverain margins, and the strong men of the
revolution, the Khalifa's own nomadic pastoral following. Faced
with the great famine of the late 1880s — and all three of these
polities were weakened by demographic decline — the Khalifa was
prepared to foster a neo-mercantilist accommodation between his
treasury's monopoly of ivory exports and private Egyptian grain
importers. His officials were not prepared to tolerate free trade
on the part of his own farmers and merchants; requisitioning was
712

all they could expect. Having put down their rebellion, he feared
to arm them against the Anglo-Egyptian reconquest in the 1890s.
Indeed, General Kitchener enlisted some of the riverain peoples
as auxiliaries against their ruler. The suppression of commercial
freedoms was no guarantee of political survival.51
Nor was salvation to be found at the opposite end of the
spectrum, by giving free rein for the monopolists of office to
become the magnates of trade. The nineteenth-century kingdom
of Imerina was torn by an unfinished conflict for control of the
crown between the noble guardians of the talismans which
protected the old confederal freedoms of its demes or clans, and
the commoner officials raised by King Andrianampoinimerina's
revolution in government, 'the princes who came up with the
pigs'. Mission Christianity became an alternative ideology of
opposition to autocracy, parallel with but scarcely sympathetic to
the old talismans. The commoner oligarchs' coup of 1869, with
their official adoption of Christianity and burning of the talismans,
restored the disrupted unity of power and cosmology, and so
ushered in a terrible period of more cohesive repression. Churches
became the most effective arms of state, mobilising labour and
taxes. Commoner oligarchs helped themselves to public funds for
private investment in the cattle, rice and slave trades, mocking
the popular Christian doctrines of self-help. They then made up
for the commercial slump by working their forced labourers
harder and by investment in peasant debt. Popular Christianity
and a revived love of the old talismans both formed an under-
ground geography of resistance. They needed only the approach
of the French invasion of 1895, which made good the paper
protectorate of 1885, to come out against the official Christianity
of the kingdom, effectively paralysing Imerina's resistance. Rebels
against both French conquest and domestic tyranny, they adopted
pseudonyms from Pilgrim's progress. One guerrilla praise-name,
'Mr Does-not-love money', summed up a half-century of moral
outrage.52
Asante's experience lay somewhere mid-way between that of
the Khalifa's Mahdiyya and Imerina. Its government's balance
between public service and private gain was overturned, as the
51
P. M. Holt, The Mabdist Slate in the Sudan rSSi-rSfS: a study of its origins, development
and overthrow ( O x f o r d , 1958), 1 7 3 - 8 4 , 214, 2 2 1 .
52
Stephen Ellis, 'Collaboration and resistance in Madagascar 1 8 9 5 - 1 8 9 9 , w i t h
special reference t o t h e k i n g d o m o f Imerina', D . P h i l , thesis, O x f o r d , 1980.
713

century advanced, by the declining profitability of war — indeed,

its escalating costs in both blood and treasure - and the growing
attractions of trade. Public funds and private war loans to the
treasury, speculations in booty and captives, had once comple-
mented each other. By the middle of the century, access to Fante
and British capital on the Gold Coast and awareness of the greater
freedoms given there to traders had made the asikafo, the big men
of Asante, restive. To lead a nation of pedlars looked quite
attractive. Previously they had invested, whether voluntarily or
not, in their public reputation, by replenishing the hierarchy of
which they were part. Now they began to invest in property and
their families' future. They began to see kings as extortioners.
Kings began to doubt their loyalty. State and ruling class broke
their mutual understandings. The conflict came to a head after the
British invasion of 1874 had exposed Asante's vulnerability and
finally discredited its generals, its conservatives. As Sir Garnet
Wolseley somewhat unnecessarily pointed out to the asantebene,
' Your Majesty... can no more prevent an army of white men
marching into your territory... than you can stop the sun from
rising every morning.'53
In the aftermath of defeat Mensa Bonsu was enthroned as the
candidate of the peace party among the councillors, the Cobdenites
of Kumasi, who favoured free trade rather than confrontation
with the British colony, and a corresponding relaxation of
Asante's own regime. But Mensa Bonsu's reign was as riven with
contradictions as any Moroccan sultan's. He pursued self-
strengthening through reform and ended by unleashing
reactionary terror. His liberalisation of the penal code, especially
the abolition of many capital crimes, angered the conservatives.
His employment of mission-educated officials unsettled some
progressives. His creation of a standing fusilier corps, perhaps to
assuage the growing clamour against conscription among Asante's
poor, antagonised the rich; it was expensive, and might be used
against them. State exactions increased, but did not cover expenses.
The asantebene enlarged his revenues with his own version of
neo-mercantilism, auctioning off trading and mining concessions
to Europeans. They seemed less of a threat to him than his own
asikafo, who were beginning to invest privately in the new rubber
boom. His increasingly beleaguered government resorted to ever
" Wilks, Asante, 508.
714

more violence. Asante collapsed into a civil war in which no
political coalition, no rival vision of the future, could long secure
ascendancy. A new king, Agyeman Prempeh, could reconstruct
the kingdom only at the cost of much decentralisation. Mean-
while, Mensa Bonsu's tyranny had already handed to the expec-
tant British the levers of Asante discontent with which to open
the way to a bloodless take-over in 1896. Disaffected provincial
chiefs and an exile community at the coast were quite ready to
help the British in reconstituting Asante to their private profit,
whatever other ideas of greater state protection for European
capital the British may have intended.54
Neither the centralising rearmament of statehood nor the
attractions of foreign capital for public servants forced these
African governments to break faith with the political under-
standings which underwrote their states. Rather, it was their
contingent and contradictory combination. Neither the revolution
of modes of domination nor the articulation of modes of
production, both terms of crushing generality, can stand alone in
historical explanation. The particularity of each local history
demands respect; there are no laws of conquest. It is arguable,
for instance, that those African peoples who remained most united
in face of European power were those whose leaders shunned both
rearmament and mercantile concessions, the Hehe of Tanganyika
or the Lozi and Ndebele of Zambezia. Their spearmen could be
mobilised effectively only by political institutions which retained
legitimacy. And it must be remembered that spears often proved
to be more effective in battle than guns until very late in the
nineteenth century, when muskets were abandoned for rifles. In
the 1820s, Shaka Zulu had good reason to scorn guns; his assegais
needed no time for reloading. The Nandi of Kenya likewise knew
how to drawfireand then charge in after a volley of muzzle-loaders;
it was not until 1895 that they learnt that this tactic had become
bloodily obsolete against riflemen.5S But it is also arguable that
the kingdom which most successfully resisted the opposing
54
Ibid., 4 7 7 - 7 2 4 ; T . C. McCaskie, 'Office, land and subjects in t h e history o f the
Manwere fekuo o f K u m a s e : an essay in the political e c o n o m y o f the Asante state', J.
Afr. Hist., 1980, 2 1 , 2, 2 0 3 ; idem, 'State and society, marriage and adultery: s o m e
considerations towards a social history o f pre-colonial A s a n t e ' , J. Afr. Hist. 1981, 2 2 ,
4, 4 8 8 ; idem, ' A c c u m u l a t i o n , wealth and b e l i e f , 3 2 - 9 .
55
J a m e s Stuart a n d D . M c K . M a l c o l m ( e d s . ) . The diary of Henry Francis Fjnn
(Pietermaritzburg, 1969), 8 1 - 2 ; A . T . M a t s o n , Nandi resistance to British rule, 1890-1906,
vol. 1 (Nairobi, 1972), 46, 143.
7M

tendencies of decentralisation and tyranny, which best maintained

a negotiable equilibrium between kings and their highly complex
administrations, and which then mounted the most effective
military resistance to conquest was Dahomey, one of the longest
involved in external trade and one of the largest importers of guns.
The paradoxes do not end there, for the peoples who often proved
most difficult to subdue, like the Xhosa on the Cape Colony's
eastern frontier, were precisely those who had no elaborate state
structure, such as Dahomey's, which Europeans might prise open
or take over. Conversely dynastic kingdoms, like those of the
Swazi or the Zulu, which had not created staffs of the sophistication
of Asante or Imerina, proved desperately vulnerable to the
defection not of classes but of princes, who hired European allies
for succession wars with the promise of damaging concessions.
And it is difficult to argue that the two nations which most
successfully pursued self-strengthening, the Ethiopians and the
Boers, achieved anything very remarkable in terms of state-
building, let alone the industrialisation of statehood.
It is conventional to maintain that Ethiopia's sovereignty would

never have survived the scramble had not her highland provinces
been hammered back into statehood after a century of civil war
and puppet emperors, the era of the princes. But the unity
reconquered by the emperor Tewodros II collapsed under its own
weight before the scramble started; his successor Yohannes IV,
emperor through the 188os, was obliged to resume the old habit
of government by civil war; and Menelik II defeated the Italians
in 1896 before embarking on any radically new and questionably
successful measures of centralisation.
The bandit noble, Kassa, rose to become the emperor Tewodros
through his dashing command of cavalry, only to become
obsessed with guns. He created a standing army he had no means
to supply. He destroyed his promising friendship with the church,
the only other possible national institution, by confiscating such
of its lands as he could lay hands on. The obduracy of his nobility
sent him mad; he even imprisoned Queen Victoria's consul. When
the British punitive expedition defeated him in 1868 and provoked
his suicide, Europeans may have considered Ethiopia easily
defeated; but Ethiopia had already gone rotten again from within.
Nearly thirty years later Menelik of Shoa as crushingly defeated
7i6

the Italians, but it is not at all clear that he did so as the emperor
of a markedly more united Ethiopia. He had resumed Tewodros's
project of self-strengthening but by entirely different methods. He
did not attempt to seize the internal privileges of church or nobles;
he expropriated the resources of outsiders instead. In their
competitions for influence Europeans were touchingly ready to
be despoiled. The Suez Canal had conferred on Ethiopia a
strategic importance as tempting as Morocco's or Egypt's; rifles
and bids for trade and influence flowed in from many quarters;
only the Italians overplayed their hand. And Menelik had some-
thing more than strategic geography to offer; his conquests of
Ethiopia's southern marchlands, following in the footsteps of his
grandfather, enabled him to control their staple exports, ivory,
gold, musk, coffee and, more discreetly, slaves. Here, in the lands
of the Oromo, who had plagued the highlands for centuries, was
Menelik's second external resource. Far from engrossing the fiefs
of his rivals he was now able to double their supply, with grants
to his supporters, ecclesiastical and lay. He did not transform
Ethiopia's political system, he expanded it. He rewarded Amhara
notables, he co-opted Oromo, both old devices, and Ethiopia's
modern history has been soured by the memory: 'Menelik gave
the land to the Amhara, and other people to the birds.' His
response to the twin disasters of rinderpest and drought was
entirely traditional, opening his granaries to the destitute and
giving out to notables such breeding stock as he could capture.
Moreover, although his huge army at Adowa was well supplied
with modern weapons, it was not a modern army. It did not
tolerate discipline, it was ravenously hungry, and on the morrow
of victory it wanted to go home. He was unable to follow up his
success except by exploiting still more strongly the continuing
European competition for influence. And when he suffered his
first stroke in 1906 his rudimentary cabinet government proved
to be no more than a new term for traditional court intrigue.56
What Ethiopia did enjoy in greater measure than any other
African kingdom or people was a national tradition of recurring
unification about the lawgiving and ritual institutions of empire
and church in face of outside threats, but the case of Afrikaner
self-strengthening clearly illustrates the limits of national senti-
56
S v e n R u b c n s o n , The survival of Ethiopian independence ( L o n d o n , 1976); J o h n
Markakis, Ethiopia, anatomy of a traditional polity ( O x f o r d , 1974), 157.
717

ment as a social construction. In 1899 most Englishmen probably

believed Afrikaners to be 'a single, solid, and compact body of
obstinate and reactionary farmers'; and it suited British imperial
policy to claim, as Consul Plowden forty years earlier had almost
but not quite said of Ethiopia, that 'progress from inside' was
therefore impossible.57 Moreover, the Boers of the (Transvaal)
South African Republic embarked on an expensive programme
of rearmament after the local and metropolitan imperialists had
clumsily shown their hand in the Jameson Raid of 1895; they
reinforced their burghers' rifles with imported machine guns and
German artillery. Unlike other African countries the Republic also
had good manufacturing and repair capacity in mine and railway
workshops. It had the social cohesion, again in contrast to
virtually all other African states, but comparable to the Nama of
South West Africa, to engage in people's guerrilla war after its
formal battlefield formations had been defeated. The Boers were
therefore vastly expensive to conquer. Britain picked up new
tropical African subjects at the bargain price of about 15 pence
each; even that sum was mostly spent on railway-building in
support of conquest, not on directly military expenditure. The
Transvaal Boers cost £1,000 a head or £250 million in all to
subdue, over 14 per cent of Britain's net national income in 1902,
a grim foretaste of the material costs of industrialised warfare.58
It must then seem as if Boer self-strengthening was supremely
successful; with 300,000 troops in the field, Britain had had to
commit more manpower than the total Boer population of the
Transvaal. But appearances deceive; by 1902 no less than 20 per
cent of the Boers under arms were fighting against their own
nation on the British side.
Afrikaner disunity was manifold. Afrikaners were no more a
unified descent group than any African tribe; the original Dutch
settlers had long since been joined by French Huguenots and
Germans. They had little commonality of culture. By the 1880s
the children of richer families were being educated in English,
even in their church schools; the twelve-year-old Jan Smuts first
caught his teacher's notice by reciting the names of all the counties
57
J. A . H o b s o n , The war in South Africa: its causes and effects ( L o n d o n , 1900), 15.
58
T r e v o r L l o y d , 'Africa and H o b s o n ' s imperialism', Past and Present, 1 9 7 2 , 5 5 , 14*;
Clive T r e b i l c o c k , ' W a r and the failure o f industrial mobilization: 1899 and 1914', in
J. M . W i n t e r ( c d . ) , War and economic development (Cambridge, 1975), 1 4 1 - 3 .
718

59
of England. Meanwhile the children of the poor, four-fifths of
the Cape Dutch, could not read even their own language in the
1870s. English was the language of money. In the Cape there were
old gentry as well as the newly noticed poor. Even before the
mineral discoveries there were great disparities of power and
wealth in the Boer republics as well. Public office went to those
who could afford it, and it yielded private dividends, both in credit
on the colonial British banks and in privileged access to state
allocations of the best sort of land, land with resident Africans
on it who could be charged rent in labour and kind. Land
speculation created landlessness more effectively amongst poor
whites than amongst blacks.
Nor were Afrikaners united at the political level. Within the
republics the politics of notables bred personal rule and factional
opposition. At the level of interstate as much as individual social
relations, economic development, as usual, divided men before it
provided the resources for political consolidation. In the 1880s
the gold of the Witwatersrand overturned South Africa's eco-
nomic geography, tipping the Anglophile gentility of the Cape
into the pocket of the ragamuffin republic. There were endless
rows between interior republics and coastal colonies over tariffs
and railway rates, with Afrikaners aligned on both sides. Even
within the South African Republic the political vision of the future
was divided. The older political generation, cultural conservatives,
looked upon gold-mining as a subversive incubus, to be exploited
only to the extent that it strengthened their control. They were
opposed by a younger generation whose skills they needed,
professional men and commercial farmers, English-educated
where they were not new immigrants from Holland. These 'pro-
gressives', secular nationalists, were prepared to co-opt inter-
national mining capital as the revenue-bearing foundation of a
modernised Anglo-Afrikaner state but on their own, not imperial,
terms. In the 1890s, the years of crisis, it became clear that the
Boers had been as sharply divided as any of Africa's peoples by the
penetration of capital and the rebuilding of states. Investment and
markets had already separated them into big men and clients. The
food market at the mines and rinderpest's destruction of their
herds now disengaged the fortunes of the big men from their
59
W. K. Hancock, Smuts, vol. i: The sanguine years, iS/o-ifrp (Cambridge, 1961),
11.
719

clients. Unfree African farm tenants, who could be made to

labour, became a better proposition than dependent white herders,
free citizens who could not. The defensive nationalism which
might have bridged this growing gap with a common ideology
was fatally compromised not only by the obvious commercial
utility of English but also by the lingering attractions of British
representative institutions. Afrikaner nationalism was necessarily
as ambiguous as Africanus Horton's or Edward Blyden's, the first
prophets of the African personality. Even import-substituting
industrialisation, to supply the mining community's consumer
needs, divided where it was intended to unite. Landless Boers
were given some employment, but their bosses were the same
political notables who had pushed them off the land.
All these divisions were opened up under the privations of a
particularly brutal war. The older political generation was soon
shouldered aside by the new, but the young men themselves
divided. Many Anglophone officials were quick to collaborate
with the British occupation, kissing the rod which scourged them.
The most determined guerrilla leaders, the bittereinders, were also
to be found among the progressives. But the ' joiners', those who
fought for the British, were disproportionately recruited from the
landless poor, the bijwoners. 'In some sense this treason', it has
been said, 'was a rebellion against class exploitation.'60
The argument hitherto has suggested that the divisions of

self-strengthening and capitalist penetration were divisions be-
tween politicians, between kings and their staff, between
monarchies and tributaries. But conflicts between power-holders
have a habit of unlocking the normal fetters on the freedom of
the powerless to express their discontents.61 Africa's contradictory
60
D o n a l d D c n o o n , ' Participation in the " B o e r war " : p e o p l e ' s war, people's n o n - war,
o r n o n - p e o p l e ' s w a r ? ' , in B . A. O g o t (ed.), War and society in Africa ( L o n d o n , 1971),
1 0 9 - 2 2 ; Stanley T r a p i d o , 'Landlord and tenant in a c o l o n i a l e c o n o m y : the Transvaal,
1 8 8 0 - 1 9 1 0 ' , / . S. Afr. Studies, 1978, 5, 1, 2 6 - 5 8 ; idem, 'Reflections o n land, office and
w e a l t h in the S o u t h African Republic, 1 8 ) 0 - 1 9 0 0 ' , in Marks and A t m o r e (eds.) Economy
and society in pre-industrial South Africa, 350-68; C. T. Gordon, The growth of the Boer
opposition to Kruger iSfo-iSf) (Cape Town, 1970); Hermann Giliomee, 'Processes in
development of the South African frontier', in H. Lamar and L. Thompson (eds.). The
frontier in history: North America and Southern Africa compared (New Haven and London,
1981), ch. 4 - p. 112 for the final quotation. I am indebted to Dr Giliomee for his
advice.
61
I have discussed this interplay between 'symmetrical' and 'asymmetrical' politics
in ' States and social processes in Africa: a historiographical survey', African Studies
Review, 1981, 24, 2—}, 139—225.
72O

experience of rearmament and opening-out raised taxes and
sharpened grievance; it also gave some of its common people and
some of its slaves new opportunities for political action. Some
traded with strangers who broke royal or chiefly controls on the
flow of goods; some villagers could afford to buy guns. Some had
access to new ideas, Islam, Christianity, cults of affliction or
anti-witchcraft movements which tried to combat the pestilence
and envy that accompanied external trade, all of them embodied
in alternative communities, with alternative protectors. The
possibilities of subverting existing inequalities of power were by
no means necessarily translated into revolutions. Those who
could actually visualise revolution, Muslim clerics or christianised
chiefs, generally used it to authorise claims on the services of their
followers which are not easily distinguishable from the
exploitations which they destroyed.
Nevertheless, there are many indications of social disorder and
uncertainties in the prerogatives of power in the cases which have
been presented here. In Tunis and in Egypt there was what can
properly be called peasant violence against landlords and traders.
On the ivory and gun frontiers of central Africa there was much
localised resistance to slavers and their allies among chiefs; it
sometimes took the form of flight and further fragmented power.
The Mahdist revolt gathered up everybody who suffered from
Egyptian taxes. The kings of Asante and the oligarchs of Imerina
alike faced popular hatred of military service or forced labour.
Menelik's programme of granary-building attempted to forestall
traditional peasant murmurings against the quartering of his
troops. But the hidden history of the Boer war suggests that
discontent was most to be expected when power was most
dislocated, at the moment of conquest.62 The bijwoners1 revenge
against their landlords was not the only revolt which the
Transvaal faced from within. Africans seized control of farms
whose owners were absent on commando service against the
imperial forces. Many Africans suffered as bitterly as the Boers
in the war; they died in equivalent numbers in Kitchener's
concentration camps. But many other Africans sold their services
to the British as scouts, town guards or transport riders, they
61
Peter Warwick, Black people and the South African War 1X99-1902 (Cambridge,
1983); William R. Nasson, 'Black society in the Cape Colony and the South African
War in 1899-1902: a social history' (Ph.D. Cambridge, 198)).

provided vast quantities of food to the army, and all at a profit

which outraged colonial conventions of African servility. It was
by no means only a white man's war.
Everywhere the conquests of Africa brought similar paradoxes
of public disaster and private profit in their train, confusions of
established orders which colonial rulers had everwhere in some
fashion to put back to rights.
CONFRONTATION AND DISORDER
There were literally hundreds of European conquests of Africa,

not one. They occurred at very different stages in European and
African relationships, at different levels of European and African
military capacity. In the first phases of the French conquest of
Algeria, in mid-century, both French and Muslims died in their
thousands, the French remembered in Berlioz's gargantuan Grand
messe des morts. By the end of the nineteenth century Europeans
were almost invulnerable against all Africans save the Ethiopians,
and small colonial wars were sentimental music hall. But Africans
died in their tens of thousands, notably the Khalifa's ansar at
Omdurman, and the Herero and Nama of German South West
Africa, three-quarters of whose total populations perished, mostly
of thirst. The British expeditionary force in Asante, in 1874, was
the first not to be decimated by malaria. Gatling machine-guns
were first tested out, and first jammed, against the Zulu; obser-
vation balloons were employed in the Sudan. The Hehe chief
Mkwawa's stone fortress of Iringa was like a house of cards before
Colonel Schele's shrapnel storm. 'There is nothing which can
come in here', the Hehe had only too truly sung, ' unless perhaps
there is something which drops from the heavens.'63
Some invasions pushed up the existing trade routes, like those
of the British in West Africa and the Germans in East Africa:
others, like the French advance in the western Sudan, cut right
across them. Some conquests were scarcely to be distinguished
from a whole series of earlier clashes, strung out over a half-century
or more, as between Britain and Asante or the French in Senegal.
But for many peoples in the interior the first European they saw
was the political officer with his police, demanding taxes. Some
63
D a n i e l R. Headrick, The tools of empire: technology and European imperialism in the
nineteenth century ( N e w Y o r k and O x f o r d , 1981); Iliffe, Modern history of Tanganyika, 112.
722

conquests were occupations from within, as Europeans took sides
in local succession wars, or recruited and drilled runaway slaves.
The British took advantage of the first sort of opportunity in
Buganda, Boers and British used the same in Zululand; while the
French advance across the western Sudan was almost a creeping
slave revolt, but scarcely an emancipation, under a foreign flag.
Most conquests were doubly from without. Alien white com-
manders were generally accompanied by a swarm of African
auxiliaries seeking to profit by a raid on neighbours, whether they
were vengeful tributaries, such as the Tswana or Shona who
helped Rhodes against the Ndebele; competitors for pasture, like
the Masai against the Nandi in Kenya; or levies provided by
locally dominant rulers who seized the chance to convert their
sphere of influence into a sub-imperialism, whether by allying
with British arms like the Ganda or, as did the Lozi, by a deft
exercise of political judo,64 guiding Western preconceptions of the
proper nature of African kingship to underwrite claims to a wider
paramountcy than the Lozi were able to enforce themselves. Some
conquests were the continuation of diplomacy by other means,
others were bleakly uncomprehending. In southern Africa and on
the west African coasts, within the missionary fringes of white
power, African leaders could appeal for protection or denounce
subjugation in French, English or Afrikaans, languages in which
they had been adept for a generation or more. In the interior,
Europeans and Africans were often obliged, metaphorically at
least, to communicate ' by gesture as if they were deaf. The sheer
number and variety of conquests is not the only reason why they
are so difficult to analyse as a whole. For they were both extended
processes and sudden, shocking events.
As processes, the European conquests generated their own
political systems, barbed alliances between formal holders of
power, white and black. As events, European victories destroyed
or discredited African political structures; they loosened the
allegiance of subordinates, they offered alternatives to submissive
labour among the poor, whether slave or free. There was a
dynamic relationship between these two qualities of conquest; one
can trace it in three phases. In the armed diplomacy of the years
preceding the first scramble, and during the 1880s too, African
rulers and chiefs used their European alliances for self-
64
I owe this concept to Dr Gwyn Prins.
723

strengthening; and in the commercial slump they battened more

on their own producers while they also challenged the European
terms of trade. When in the second scramble European officials
forcibly converted clientage into control, the subsequent decom-
position of African power opened the possibility that discontent
could become opposition, the evasion of tribute or flight by slaves.
These threats of social dissolution, finally, obliged the conquerors
to reconstruct the very alliances which they had so brusquely
overturned, and on conditions which they could not freely dictate.
For Europeans were divided in their answers to the crisis in
authority, and rival Africans looked to them for mutually in-
compatible bargains in sharing out power. All such post-conquest
reconstructions, moreover, writhed between the public constraints
of metropolitan treasuries and the private goads of merchant,
mining, or plantation capital.
This dynamic sequence of conquest compelled Europeans to
become concerned with control over ever deeper levels of African
society. Historians have tended to follow them. A scholarly
preoccupation with states and trade, royal diplomacy or the
struggles for survival by African coastal middlemen, has been
superseded by investigations into the changing relations of power
and obedience, the shifting oppositions between holders of office
and producers of commodities. The first direction of enquiry
would be more appropriate to the first scramble, the latter to the
second, if only history were ever quite so conveniently simple.
In the long decades of British paramountcy over Africa's coasts,
and during the few hectic years of the first scramble, one may
observe a dual process of political accumulation.65 European
consuls and trading factories were only one local power among
many, and not always the most formidable. They had their own
external resources, credit and naval bombardments or landing-
parties on call. But they were ignorant, they needed supplies, they
looked for allies. Consular establishments and expeditionary
forces were markets for intelligence, food, labour, and political
reputation. Their needs enlarged the local flows of power, those
daily transactions which were premised upon mutual responsibility
between contracting parties over time. Mundane relationships
became formalised. The road to empire was paved with treaties.66
61
J. M. Lonsdale, "The politics of conquest: the British in Western Kenya,
1894—1908', Hist. J. 1977, 20, 4, 841—70, for a discussion of this concept.
66
C. W. Newbury, The western Slave Coast and its rulers (Oxford, 1961), 49.
724

European needs were guaranteed by the recognition of African

statehood. Demands for abolition of the slave trade or the
mitigation of import duties were exchanged against the instru-
ments of power, annual subsidies to kings and gifts of muskets.
Initially, then, Europeans increased their own security, their
minimum necessary power of control and prediction over the
immediate political environment, by recognising African power —
including the continuing power to own slaves — and by giving
away some of their own. The process was by no means indis-
criminate. Allies had domestic rivals and enemy neighbours;
self-strengthening, it has been shown, multiplied such divisions.
Europeans had to choose between the expense of supporting their
friends and the cost of losing them. There came a time when
cutting losses, whether incurred by trader constituents or by
public administrations, seemed to be best achieved by enlarging
European control at the expense of African power. Whatever the
cause of forward political action, whether it was a local crisis of
mistrust between Europeans and Africans who were formerly
allied, or a precautionary claim staked out against European
competitors, the repudiation of existing conventions of doing
business faced Africans with a moment of choice. They could
stand by, and contemplate dying for, the public cause of self-
strengthening; they could try to make a new private accom-
modation with the opportunities of Africa's opening out. African
ruling classes divided over this choice, so did Africa's poor; how
each reacted to crisis is often taken to be a good guide to their
societies' internal tensions. But it has to be remembered that for
the most part internal divisions were closely aligned with external
contacts by the 1880s; scarcely any African people came to their
hour of trial as a closed community with a consensus, even among
their leaders, about what it was that they might be called upon
to defend, let alone among their followings and slaves. And their
freedom of decision was increasingly constrained. The great
depression had caused political accumulation to become more and
more one-sided; as Europeans took greater liberties, so Africans
found that they could take less, except along paths smoothed by
European power.
European traders became increasingly impatient of paying
tribute or road tolls to chiefs; they ceased to regard losses by
pillage as a normal business risk; they turned more readily to their
725

67
consuls for redress. But their African partners and competitors
came to feel very much the same; Senegalese traders valued their
rights to representation in the four coastal communes; Fante
commercial opinion towards the Asante oscillated between force
and conciliation in much the same rhythms as English opinion;
Yoruba traders took care to secure membership of the Lagos
Chamber of Commerce. Political allegiances wavered on all the
frontiers of contact long before any formal European acts of
annexation. To the Asantehene the British forts on the Gold Coast
must have looked infuriatingly like asylums for all sorts of
malcontents, whether criminals fleeing from his justice or sub-
ordinate chiefs who wearied of their responsibilities; the Zulu
kings felt the same about Natal. However much the Egba
demonstrated their Yoruba loyalties by expelling missionaries
after a dispute with the British colony of Lagos in 1867, they still
wanted English taught in their schools. The first thing the Niger
delta rebel Jaja did on his defeat in 1869 was to appeal for a British
protectorate against his king, George Pepple, who had himself
called for British intervention in Bonny's civil war.68 Consular
courts were looked to for judicial decision along the Mediterranean
and Atlantic coasts, and in Zanzibar the rulings of the British
consul, Sir John Kirk, were increasingly influential. The Native
Affairs secretary, Theophilus Shepstone, gradually insinuated
himself as paramount chief beyond the borders of Natal. King
Kabarega of Bunyoro failed to bring his uncle Ruyonga to heel,
protected as the latter was by Egypt in the persons of Samuel
Baker and General Gordon. This growing inability of African
rulers to bring internal disputes to a final conclusion was a factor
in their increasingly anxious tyrannies; it was also an indication
of creeping conquests from outside.
By the 1890s European conquests were increasingly wilful,
taking their cue from the French advance inland from Senegal.
Recognitions of African statehood ceased to be the condition of
European security and became instead the bargaining counters of
European sovereignty. Treaties became fraudulent prospectuses,
guaranteeing African rights only for their European protectors
to flaunt them as their own. For the Frenchman Gallieni his treaty
of Nango with Ahmad's Tukolor empire in 1880 was never
67
K l e i n , Islam and imperialism, 6 1 , 114.
68
K. Onwuka Dike, Trade and politics in the Niger delta itjo-iSSj (Oxford, 1956),
186-8.
7 26

intended to be anything but a sham; Italy's treaty of Wichale with

Menelik was similarly two-faced. Perhaps the sequence of Rhodes'
relationships with Lobengula is the most revealing, not so much
of white duplicity, although there was plenty of that, but of the
growing white appreciation that Africa could not be made more
profitable without greater control over African production. In
1891 Rhodes asked his shareholders to see the British South Africa
Company's costly police force as ' a capital outlay for the purpose
of rendering productive and interest-bearing the valuable assets'
which he had acquired for them by treaty. Two years later he
deployed this capital to do just that, to seize Matabeleland and
its supposed second Witwatersrand; his conquest of Lobengula's
kingdom turned out to be a trial run for his attempted takeover
of Kruger's republic. Conclusive political accumulation, unfet-
tered control, seemed necessary for capital accumulation, and not
in mining alone. African friendship was no longer enough.
African land was now the object. The outward forms of treaty-
making were retained; getting territory in Africa remained very
simple, as Bismarck said. 'For a few muskets one can obtain a
paper with some native crosses.'69
Why and how Africans gave such very different answers to the
choice which Europeans forced upon them in the second scramble
has long fascinated Africa's historians. African reactions to
conquest do link in often dramatic ways their subsequent colonial
histories with their more distant past; but which colonial history
and which pre-colonial past is more and more in question.
Historians first focused on the calculations of Africa's established
leaders, their capacity to rouse their people to war, their ability
to guide them in the industrious path of peace. Our growing
knowledge of pre-colonial Africa's often harshly unequal political
economies has prompted new realisations of how its social
divisions were often, but not always, sharpened under the test of
European invasion. Popular perceptions of conquest were not
only very different from their rulers'; they were frequently, and
deliberately, opposed. The first historical accounts were structured
by the assumed dichotomy between resistance to conquest and
69
A. S. Kanya-Forstner, The conquest of the western Sudan: a study in French military
imperialism (Cambridge, 1969), 71-83; Lloyd, 'Africa and Hobson's imperialism', 142;
Iliifc, Modern history of Tanganyika, 90.
7*7

collaboration with the conquerors; between the uninstructed and

the enlightened or between heroes and traitors, according to taste.
These simplicities have long been discarded and can now be
replaced. To understand the contrary human intentions one needs
first to grasp an underlying process. But the process is, again, a
dual one, the building of states and the articulation, or contra-
dictory junction, of pre-capitalist and capitalist modes of
production.
The politics of colonial state-building involved one set of
contradictions, its economics another. In political terms, Euro-
peans had so to construct their African coalitions that erstwhile
diplomatic allies could be reduced to administrative agents.
Political security ' required powerful allies with obedient follow-
ings. Administration required efficient agents who could impose
uncongenial innovations.' This is a paradox of all rule. But it was
also a transition which early colonial regimes tried to master.
Allies had to bend their backs to new demands, the delivery of
tax especially. They also had to adapt if they were to profit both
themselves and their rulers - to patronise schools, sponsor new
export crops, or send out workers. The shock of this transition,
which occurred throughout most of colonial Africa, at least in its
most accessible regions, somewhere between 1895 and 1905, could
provoke violence, sometimes the first real clash of arms between
Africans and Europeans, sometimes a rising against the terms,
now belatedly imposed, of an earlier conquest. But revolts against
this sterner European mastery only served, by their defeat, to
strengthen European control still further. In their ruins there
remained more committed African partisans of the new order,
clearer perceptions of white power, and often stronger military
establishments to protect it.70
The economics of state-building on the borderlands between
European capitalism and non-capitalist Africa created quite diff-
erent contradictions. The opposition between the politics and the
economics of the process forced tiny colonial establishments to
shoulder an increasing range of responsibilities. Europeans,
representatives of capitalist society if not capitalists themselves,
took it for granted that proprietorship in material resources
70
This sequence of conquest has been most fully worked out for German East Africa.
See, Ralph A. Austen, Northwest Tanzania under German and British rule (New Haven and
London, 1968), 3 1-61; John Iliffe, Tanganyika under German rule /foj-if/2 (Cambridge,
1969), 142-9; the quotation is from Iliffe, Modem history of Tanganyika, I Z I .
728

conferred economic rights and social power. Africans, like all

other non-capitalist peoples, knew, to the contrary, that it was
authority over the services of men and women which generated
economic power, whether within familial, tributary or slave
systems of production, or in their various combinations. Property
lay in the continual exercise of domination over social processes
of production and exchange; it was not a thing. Articulation
between these two quite different ways of perceiving and
organising the world occurred first in the market and caused few
problems for either side. Each put a higher value on the goods
supplied by the other than those which they could produce
themselves. Europeans had much more use than Africans for
ivory or palm oil; Africans had not the technical capacity to turn
out in large quantities guns, cheap cloth or domestic metalwares.
These market exchanges had generated commercial alliances
which were founded on the hidden premise of the productive
obedience of each partner's subordinates. African trader chiefs
implicitly guaranteed the labour of their wives, or their slaves, or
their slave-catching armies, as Europeans laid out for them goods
in trust against the future supply of foodstuffs, vegetable oils, or
slaves. African power-holders exacted more of such laborious
discipline as their external markets expanded, and more still as
they contracted on the eve of conquest.
But then the premise of African economic obedience began to
collapse as the diplomatic alliances of commerce were first
overwhelmed by conquest and then superseded by colonial
state-building. Loss of sovereignty could literally mean the loss
of mastery over women and men. Superiors, whether chiefs or
elders, found that their attempts to profit by the new European
demands not only for commodities but also for their subjects' or
juniors' labour could be evaded or even turned against them
within the new uncertainties of power. Slaves decamped to the
communities of their birth. Peasantries deserted their chiefs'
stockaded villages, preferring easy access to their fields. Such
'downhill migrations' can be seen everywhere. Forced to pay
colonial taxes, family heads became reluctant to pay tribute, and
increasingly they took their disputes to colonial courts. Young
men who went out to work on railway construction or down the
mines at the behest of their chief returned with a sense of private
property in their labour and in their decisions for marriage. In
729

1905 Mafingo, an old Zulu chief in Natal, summed up all these
problems with his complaint that colonial officials seemed inter-
ested only in his taxes.
They do not help us in kraal matters and the management of our wives and
families. Our sons elbow us away from the boiled mealies in the pot when we
reach for a handful to eat, saying, 'We bought these, father,' and when
remonstrated with, our wives dare to raise their eyes and glare at us. It used
not to be thus. If we chide or beat our wives and children for misconduct,
they run off to the police and the Magistrate fines us; with the result that our
families defy us and kraal-head control is ceasing to exist.71
The first colonial officials were obsessed with a search for

political security, but not to the exclusion of economic produc-
tivity. They had all experienced wars or rumours of wars. Political
order was self-evidently founded on social hierarchy. Their
African agents were worse than useless if they did not have the
authority to be obeyed. The articulation of modes of production
appeared to be undermining the means of domination, but chiefs
had to have an income, taxes had to be paid, merchants and miners
had to profit. Few officials had any faith in either the productivity
or the reliability of free peasantries. They needed political
supervision or the discipline of employment; they had to have
chiefs, landlords or foremen. If allies were to be turned into
administrative agents, their followers would have to learn again
the habit of obedience. Quite how, and with what degree of
success, the various colonial governments tried to reconstruct
authority depended much upon the nature of their competing
white constituents. Established white traders were generally
content with the old order of things, but peasant proprietors were
conveniently weaker in the market than the old noble brokers.
Chiefs and traders nevertheless often found themselves in alliance
against the intrusion of the newer clients of the state which
threatened them both, concession-hunters, mine promoters, plan-
tation syndicates, farmers, all interested in the managed supply of
African wage-labour rather than the free flow of African domestic
produce. But colonial reconstructions also had to cope with the
recent history of African attempts, from above and below,
continually to renegotiate between themselves and with whites the
shifting economic bargains of political change. Resistance and
71
David Welsh, Tie roots of segregation: Native policy in colonial Natal tSjj-rf/o (Cape
Town, 1971), 301.
73°

collaboration are best understood as the African side of the

competitive politics of state-building. Five conclusions follow.
Armed resistance was almost never 'primary', in the sense of

being an instinctive reflex action against an external intrusion.
Domestic African and external European politics were in most
regions already too interlocked for that. They continued to share
a common interest in trade, even in trade devoted to African
rearmament. British dealers kept on supplying arms to Samory,
Germans sold rifles to King Behanzin of Dahomey, while both
rulers were fighting the French. Black mineworkers in South
Africa went on buying guns with their earnings long after the
white panics of the 1870s about the growing threat from African
chiefdoms across the frontiers. Africans took advantage of
European divisions as much as their own quarrels were themselves
exploited. Virtually all Africans negotiated working accommoda-
tions with Europeans until Europeans made it impossible. Even
those who fought most fiercely, Samory, the rulers of Dahomey,
the Ndebele, the Zulu, the Herero and the Nama, did so only as
a last resort, when their European allies betrayed them. The
Mahdiyya alone was proof against all overtures, not that any bids
for its friendship, save from Menelik, were ever seriously made.
Primary resistance must also have been the response of some
stateless peoples of the interior, those very few societies which
had not become deeply involved in the advancing frontiers of
hunting and trade. But everywhere else the concept is inappro-
priate. It assumes the contiguous opposition of discrete political
systems, African and European, but the reality was their inter-
penetration, Afro-European.
Secondly, therefore, one has to abandon, once and for all, any
idea that African ' societies', or even their rulers or leaders, were
normally united in their solutions to the European problem. The
case studies in the rearmament of African statehood have shown
that for the states most fully involved in the European world there
was profound disagreement as to their national cause; stateless
societies were as chronically divided, if in different ways. Within
the trading frontiers, kingdoms were contested between what
have been described for Dahomey as ' Ethiopian' and ' Liberian'
parties. The former supported monarchies in their attempt to
maintain pre-capitalist relations of authority and exploitation;

their opponents tried to convert their official domains into private

capital. The one required European recognition, the latter Euro-
pean partnerships. The analysis would apply equally well to the
war and peace parties of Asante or the royal, usuthu, faction
and its opposition in Zululand. Foreign policies veered between
the obstructive and the accommodating according to the domestic
balance of the parties. Ideological division was perhaps still more
marked on the western fringes of Muslim Africa; Senegalese
chiefdoms, which had equivocated with Islam for centuries,
increasingly found that their tributary exactions were opposed by
armed peasantries who enjoyed clerical protection. Neither side
had any particular love for the French, but each was prepared to
bargain with them against their internal rivals.72
Just as there were no agreed policies, so also, and this is the
third conclusion, leaders had very varied capacities to carry their
people with them in whatever tactical option they had for the time
being decided upon. There were important distinctions here, not
only between kingdoms and stateless societies but also between
varieties of kingdom. It is a crude simplification, but workable
for present purposes, to visualise a continuum in the basis and
intensity of African political authority. At one end of the
spectrum, authority in stateless societies was founded upon
mastery over the ritual and material skills of domestic production
but was invested with few powers of enforcement. At the other
end, their control of administrative and military establishments
granted to the rulers of the western Sudan and the ruffians of the
hunting frontier no managerial competence in production other
than that given by the power to extort tribute in labour and kind
from subject communities of cultivators or pastoralists. The
Guinea forest kingdoms and those which survived in central
Africa lay somewhere in between. The continuum is related
to variables in both modes of production and means of
coercion.
Stateless societies were productive communities; their social
institutions were most deeply valued and most fully activated as
co-operative pathways designed to preserve life against the risks
of all but the most savage of natural disasters, those droughts
72
Patrick Manning, Slavery, colonialism and economic growth in Dahomey 1640—1960
(Cambridge, 1982), 1 4 - 1 ; ; Ivor Wilks, Political bi-polarity in nineteenth century Asante
(Edinburgh, 1971); Jeff Guy, The destruction of the Zulu kingdom: the civil war in Zululand
(London, 1979); Klein, Islam and imperialism, chs. 4 and 5.

which extended for more than a couple of seasons. Then the

desperation of individual survival dictated the repudiation of all
reciprocities, and civilised life was at an end. They had no
professional armies for Europeans to destroy, only armed citi-
zenries; they had no professional administrators or rival princes
to co-opt. In neither sense were they easy, like kingdoms, to crack
open. They could be defeated only by assaults on their capacity
to reproduce themselves, by the burning of crops or the seizure
of livestock. Alternatively, their leaders could be patronised only
if they were not then required to exercise much power. Africa's
stateless societies and petty chiefdoms fought her longest and
most bitter resistances against the colonial invasions; many had
seen it all before as zones of tribal dissidence on the margins of
predatory kingdoms, enduring the rise and surviving the fall of
successive conquering dynasties. They may not have had govern-
ments themselves, but that does not necessarily mean that they
had no perception of what conquest portended. Conversely, they
probably found it less easy to mobilise for war except to repel
specific administrative demands. Not only did stateless societies
resist longer than kingdoms but their struggles often came later
in the evolution of colonial states.
Africa's kingdoms varied enormously in the degree of separa-
tion between their state apparatus and their social communities
and institutions. This separation seems to have been most marked
in the Muslim empires of the western Sudan, but quite why this
should have been so is a contentious matter. It was partly because
in the nineteenth century they were all new, but that is itself a
reflection on the historical instability of ruling regimes in this
region. It has commonly been argued that in these open savannas,
tsetse-free and with long dry campaigning seasons, large cavalry
aristocracies were only too well equipped to cow their subjects,
and to communicate orders over long distances. Cavalry raids also
captured large numbers of slaves which could be used in
production, in administering free communities and in further
wars. All that Europeans had to do was to defeat their armies in
an afternoon and their brittle superstructures would collapse by
evening; the way was then open to negotiate understandings with
the submerged dynasties of subject peoples which were thereby
raised again to political life. There is much in this, but it probably
accords too much military prowess to horsemen and does not take
733

sufficient account of the centuries-old contradictions between
rulers and townsmen - merchant princes, artisans and clerics -
communities which supplied armies and sold their slave products,
but which also used the dry seasons for long-distance trading only
to be harassed by the warriors who rode out to war. Peasantries
had their grievances against warrior elites, but so too did African
merchant capitalists, and they had more to offer to Europeans by
way of supplies and skills in return for peace.73
Elsewhere there was much less sharp a distinction between state
and society, whether in the kingdoms of the tropical Guinea
forests, the equatorial forest of Zaire, or in the savanna of eastern
and southern Africa. In some of these kingdoms, it is true, the
founding heroes were remembered as outsiders, some fabulous,
some uncouth, and their successors were symbolically distanced
from the ties of the kingdom's component lineages. Nevertheless,
the fabric of authority generally seems to have been more
complex, more subject to the claims of representation and
consultation, altogether more civil than in the western Sudan.
Again, it is difficult to explain this contrast in a word. It must be
at least in part because the tsetse-fly prevented these from being
anything other than pedestrian states. Infantry militias were even
less easily centralised than mounted aristocracies. The proceeds
of war were correspondingly widely dispersed. Captives aug-
mented the power and, through concubinage, the followings of
the ' big men' of local communities as much as they reinforced
the wealth and majesty of kings. Lineage slavery was as common
as royal. In the forests, domestic agricultural operations required
as heavy an investment of labour as did public works. Lineages
were probably more often alliances for protection and the manage-
ment of production than actual descent groups; but while their
ideology of descent and kinship may have justified unequal obliga-
tions to elders, big men and chiefs, it did also also mark off the
freedoms of citizenship from the hazards of kinless servitude. The
result was a wider popular reservoir of self-interested loyalty
for these kingdoms to draw upon in the face of external challenge
73
Goody, Technology, tradition and the state; Robin Law, Tbehorsein West African history
(Oxford, 1980), and the review by Rosemary Harris in Africa, 1982,5a, i, 81-5; Stephen
Baicr, An economic history ofCentralNiger(Oxford, 1980), 2i-j6; Richard Roberts, 'Long
distance trade and production: Sinsani in the nineteenth century',/. Afr. Hist. 1980,
21, 2, 169-88; idem,' Production and reproduction in warrior states: Segu Bambara and
Segu Tokolor, c. 1712-1890', Int. J. Afr. Hist. Studies, 1980, 13, ), 389-419.
734

than appears to have been available to the empires of the

western Sudan.74
Nevertheless, the self-strengthenings of the nineteenth century
did little, if anything, to expand the rights of commoners in any
kingdom, except possibly to make them more grateful that they
were protected from some of the growing burdens which were
heaped upon slaves. While one can point to the relaxation of one
or two specific impositions on commoners in Asante for instance,
their general security was undermined by the growing risks of its
wars. In Muslim areas the stricter application of the sharTa
stiffened royal administration at least as much as it enhanced
citizenship with the concept of the umma, the community of the
faithful. And it is interesting that Christianity was taken up only
by officials or fugitive slaves; ordinary commoners seem to have
had little freedom for religious experimentation even in
communities in such patent disarray as in Yorubaland. It can be
asked therefore, as it has been argued for the Balkans under the
Turkish invasions of the fourteenth and fifteenth centuries, how
far some of Africa's kingdoms not only divided politically but also
fell into ruins as societies before the European advance, so that
to some extent their people acquiesced in their own defeat.75
A proper answer needs careful distinctions. There were 'crises
of monarchy', with falling revenues, rebellious conquered popu-
lations and runaway slaves, or subversively selfish high officers
of state. There were also elements of a' crisis of aristocracy', which
involved the rise of market competition and therefore of political
disobedience among free commoners. It was possible to have the
first without the second but not, it seems, the second without the
first. These distinctions are best seen in West Africa, where they
have already been most interestingly explored.76
In some coastal kingdoms there were all the elements of a crisis
of monarchy. Like the Mediterranean countries and Imerina, they
74
J o s e p h C. Miller, Kings and kinsmen: early Mbundu states in Angola ( O x f o r d , 1976);
Jan Vansina, The children of Woot: a history of the Kuba peoples ( M a d i s o n , 1978); idem,
' T o w a r d s a history o f lost corners in the w o r l d ' , Econ. Hist. Review, 1982, 2nd series,
35, 2, 1 6 5 - 7 8 ; I v o r Wilks, ' L a n d , labour, capital and the forest k i n g d o m o f A s a n t e :
a m o d e l o f early c h a n g e s ' , in J. Friedman and M. J. R o w l a n d s (eds.), The evolution of
social systems (Pittsburg, 1978), 487—534; idem, ' T h e g o l d e n stool and the elephant tail:
an essay o n wealth in A s a n t e ' , Research in Economic Anthropology, 1979, 2, 1 - 3 6 .
75
Wilks, Asante, 6 7 3 , 7 0 8 ; Ajayi, Christian missions in Nigeria; Fernand Braudel, The
Mediterranean and the Mediterranean world in the age of Philip II, v o l . 11 ( L o n d o n , 1973),
664.
76
By H o p k i n s , Economic history of West Africa, c h . 4.
735

all suffered a collapse of external revenues from the 1860s as the

terms of world trade turned against them. This clearly sharpened
differences between kings and their trading chiefs. Asante's
conquered provinces detached themselves, and some of her slaves
fled to the Gold Coast after the British expedition of 1874. The
Oyo empire had disintegrated half a century earlier and the
successor states of Yorubaland found it impossible to end their
civil wars for want of a mediatory king.77 Paradoxically however,
the most serious monarchical crises were in the western Sudan,
the area best protected against adverse commercial conditions, if
not against severe monetary inflation, by the internal, regional,
focus of its trade. This was because the Muslim empires of the
area had their own political history as well as being subject to the
fluctuations of the world economy. Samory's regime and the jihad
empire of the Tukolors were undergoing their own early crises
of state-building at the same time as the French were converting
their diplomatic influence into a network of control. Samory's
conquests were unfinished, those of the Tukolor were falling
apart. They were disputed between rival ruling houses, and
between their zealot supporters and subject peoples who did not
yet accept that they were defeated. Both aspects of uncompleted
state-building gave openings to the French. They recruited large
auxiliary forces from among the Bambara subjects of the Tukolor;
they also allied with one Tukolor faction against the other, 'so
making the collapse of the state look less tragic than it might have
been'. 78
The local history of labour relations was as full of paradox, due
to the contradictions within African slavery. Individual slave-
holders could not have created and controlled a servile work-force
on their own. They needed soldiers and some civil system of
coercion, but also an ideology which, however irksome it may
have seemed, checked them in their own oppressions while
containing the resentments of their slaves. Paternalism came in
both Islamic and lineage guises. Large-scale mineworking or
plantation slavery, the most alienating form of all, was never the
sole element of servitude; for kings and nobles were never the
only class of masters, even in the western Sudan. Traders
77
J. F. A. Ajayi and Robert S. Smith, Yoruba warfare in the nineteenth century
(Cambridge, 1964), 63-75, 125-8.
7
* B. O. Oloruntimehin, The Segu Tukulor empire (London, 1972), 520.
736

everywhere employed slaves as some of their most trusted agents;

and free smallholders often had a handful of farm slaves as well
as captive wives. The difficulty for masters was that their slaves
did not allow themselves to be treated simply as an economic
asset; they were intended to be 'extensions of their owner's will,
but they also had wills of their own'. 79 All slave-owning societies
had to make concessions to their slaves' awkward humanity by
showing some humanity themselves. African kingdoms, let alone
lineages, had few powers of surveillance; they controlled slaves
as much by accommodating them. Masters had to grant slaves
some rights lest otherwise they seize them, if not by violence then
by flight. All African slave-holders went some way to appease
their slaves with acts of manumission or processes of incorporation
into free society, by according some respect to slave families, by
permitting them some rights to land and some enjoyment of the
product of their own labour. The manner in which slaves took
advantage of the European conquest therefore depended much
upon their individual experience of the mixture of constraints and
opportunities inherent in their condition. Recent captives or
gangs of plantation workers, whose status had not begun to soften
from slave to dependant, often saw external conquest as a breach
in the walls of coercion and accordingly took flight. Farm slaves,
whose owners worked alongside them and whose families were
about them, tended to see conquest more as a means to strengthen
their claims on whatever customary rights of enfranchisement
were locally available, in order to remain where they were.
In the far western Sudan, the small states which bordered the
French colony of Senegal suffered a steady drain of fugitive slaves
over the frontier for decades, no doubt in part because their
aristocracies were distracted by constant civil war. But slave-
holders further inland, in the no less disordered region of the
Tukolor conquests, continued to send bands of freemen with their
slave retainers to the groundnut coasts of Senegambia, apparently
confident that they would return with their earnings after as many
as three seasons working as share-croppers hundreds of miles
from home.80 While political disruption therefore preceded
19
Frederick Cooper, Plantation slavery on the east coast of Africa (New Haven and
London, 1977), 255. 1 am indebted to Professor Cooper for his advice.
80
Ken Swindell, ' Serawoollies, Tillibunkas and strange farmers: the development
of migrant groundnut farming along the Gambia river, 1848-9;', J. Afr. Hist, 1980,
21, 1, 93-104.
737

conquest almost everywhere in the savanna the collapse of social

control did not take on crisis proportions until the proximity of
European armies made it safe for slave populations openly to
acquiesce in their masters' defeat. In the middle Niger region, in
both the French and British spheres of control, there then
followed classic examples of one very widespread consequence of
the articulation of modes of production. Pre-capitalist authorities
first strengthened and then lost their direct control over labour;
capitalists thereafter failed to subordinate it fully to their purposes.
In the more westerly area of the French advance, slave regimes
happened to be particularly harsh. Recent African conquests
meant that many slaves were newly captured, needing constant
reminders of their subjugation. Slave-holders had a further
incentive to work their labour harder in the large market for food
and fodder provided by French garrisons. In some polities up to
two-thirds of the population was slave. As the victorious French
destroyed slavers' armies and displaced rulers this vast servile
population began to stir. Some became dependants of the French.
They enlisted in their army, the tirailleurs se'negalais, or accepted
involuntary settlement as porters and cultivators in the so-called
villages of freedom along the army's supply routes. Very many
more, over a quarter of a million, made good their escape back
to the homes and peasant citizenships from which they had only
recently been abducted. But most stayed behind, becoming the
share-cropping dependants of their former owners. In this way
they were able to exercise more control over their lives while
retaining the protection of patronage in uncertain times.
With this wholesale erosion of their authority, African rulers
also lost the basis for independent political action. Their new
European overlords were scarcely less dismayed. They anticipated
the collapse of agriculture and trade; they feared that the only
conceivable African levers of power would fall apart in their
hands. Their civilising mission forbade the toleration of slavery,
but it was not clear what else would maintain African hierarchies.
They resolved therefore not to abolish slavery but to make it
invisible, while undermining its sources of long-term reproduc-
tion. They stopped the wars and sales which resupplied slave
labour; they decreed that children could no longer be born into
slavery; but they refused to make it easy for existing slaves to
escape their yoke by any means other than those provided by a
73«

liberal interpretation of Islamic law. In British Northern Nigeria

comparatively few slaves ran away. This was only partly due to
the fact that Lugard and his successors took more care than the
French not to discredit the local ruling, slave-holding, class in the
Sokoto caliphate. The rapid growth of export agriculture was the
more important condition for social stability. It provided
acceptable alternatives for both masters and slaves. Masters
became traders and rentiers; slaves bought land and freedom in
some areas and entered into tenancies in others. An agricultural
revolution was under way, but it was not at all the one which the
Europeans had intended. The British had hoped that former
masters would become active landlords, their former slaves a rural
proletariat. But it was not the British who made the revolution,
it was the slaves. It was a peasant settlement, not a capitalist
transition. It provided a pattern, as will be seen, for much of
colonial Africa. Peasants wanted patrons rather than employers;
their former masters similarly clung to old relations of authority,
the ones with which they were familiar. Dependants owed them
more services than employees. So European merchants had to be
content to deal not with large plantation enterprises but with
African brokers and smallholders, neither of which tolerated
interference in their local relations of marketing, credit and
production.81
In the western Sudan, then, it appears that there was no real
crisis of aristocracy, no loss of hierarchical control over
dependants, until monarchies lost their sovereignty under con-
quest. Societies fell into ruins not before, but during, the
European advance. The crisis of monarchy had come earlier
among the coastal kingdoms because their greater exposure to the
European trading frontier had stimulated both public self-
strengthening and private cupidity. But any crisis of aristocracy,
the rise of the small producer to compete with kings and chiefs
in the market, seems, like the flight of slaves, to have been a
consequence rather than a precursor of conquest. To argue
81
Richard Roberts and Martin Klein, 'The Banamba slave exodus of 1905 and the
decline of slavery in the Western Sudan', / . Afr. Hist., 1980, 21, j , 375-94; Paul
E. Lovejoy, 'Slavery in the context of ideology', pp. 11-38 in Lovejoy (ed.), The ideology
of slavery in Africa (Beverly Hills and London, 1981), 11-38; idem. Transformations in
slavery: a history of slavery in Africa (Cambridge, 1983), ch. n ; Hill, Population, prosperity
and poverty, ch. 13; for comparable processes in the decline of slavery see, for the West
African coast, Manning, Dahomey, 188-93; and, for East Africa, Frederick Cooper, From
slaves to squatters: plantation labour and agriculture in Zanzibar and coastal Kenya 1890-192/
(New Haven and London, 1980).
759

otherwise presupposes a greater degree of political freedom for
commoners than is consistent with their available strategies of
domestic survival.82 Officials became magnates in Asante and
Dahomey precisely because their office granted them the powers
of market brokerage or political tribute, and a share in road or
harbour dues. In Yorubaland the aristocratic mode of production,
with its command over slave plantations, guilds of craftsmen and
free followers alike, was only too successful in sustaining the civil
wars which produced the slaves who, in turn, produced palm oii
for export. Ibadan's military-industrial complex may have had no
king, but its coalition of landlord-brokers was unchallenged from
below. War kept their followings faithful; and the same can be
said of Ibadan's most bitter opponents, the Ekiti confederation
of north-eastern Yorubaland. British observers in Lagos thought
that virtually all Yoruba palm-oil exports were produced and
transported by slaves until the 1890s, when British power was
finally extended inland. Asante's poor, slaves and debtors, likewise,
did not give any organised voice to their discontents until driven
to despair by their conscription to fight the British. Although the
rubber boom of the closing years of the century did promote
economic individualism in Asante, this development again was
not widespread until the 1890s, when the creeping British
conquest was already far advanced. Even then, Asante and Fante
chiefs would grant to commoners the right to tap rubber only in
return for a share in the proceeds or the performance of tribute
labour. In 1898 the Fante chiefs of the Gold Coast refused to
compromise their authority by acceding to British proposals to
make their lands convertible to freeholds, but they were quite
ready to exercise their power by turning out communal labour
on public works instead. And this was in an area where one would
suppose that chiefly prerogatives had been most outflanked by
colonial jurisdiction.
It is difficult to deduce from these examples that commoners
had become a serious challenge to pre-colonial chiefs. Rather, as
citizens, they continued to profit and to be protected by their
allegiance to the big slave-owning producers who were also their
market brokers. If the experience of the Baule of the Ivory Coast
82
On this point 1 prefer the earlier interpretation of Hopkins,' Lagos Strike \ 13 8-44,
to the later one in idem, 'Economic imperialism in West Africa: Lagos, 1880-92',
Economic History Review, 2nd series, 1968, 21, 3, 588, 604, and in idem, Economic history
of West Africa, 143.
740

is any guide, free lineage members were able to evade European
demands on their chiefs for tribute labour; slaves were picked
upon instead. It was not until colonial rule forced open both roads
and markets that it became not only economically rational but
politically possible for commoners to repudiate the economic
privileges of their chiefs. Perhaps the most illuminating example
of the transition is to be seen in the origins of cocoa-planting in
the Yoruba interior around the turn of the century. It was taken
up most eagerly in what had been the more belligerent of the
feuding Yoruba states. Here, where commoners had been most
closely identified with the aristocratic pursuit of war, they also had
most to lose when military allegiances were dissolved by alien
overrule. Demobilised soldiers fell back on commercial agri-
culture, but still within the kin groups which had been forged by
self-protection in war, 'to escape the economic consequences of
the peace'.83
On the west African coast the complementary oppositions of
free smallholder and slave production greatly complicated the
capacity of kings to resist unacceptable European demands.
Commoners fought in defence of Dahomey, but behind their
backs many of their Yoruba slaves took off for home. In eastern
central Africa, by contrast, the influence of European trade was
not strong enough to sharpen such divisions, while European
demands for land and labour as well as commercial concessions
impinged upon the least as well as the greatest in African society.
It is no accident then that one of the most famous examples of
the African capacity to resist conquest was shown by the Ndebele
and Shona people of Zimbabwe. How to account for their
sustained hardihood has become a matter of illuminating
controversy. It was first understood as their response to revolu-
83
Newbury, Western Slave Coast, 126; Bolanle Awe, 'Militarism and economic
development in nineteenth century Yoruba country: the Ibadan example', J. A/r. Hist.
197J, 14, 1, 65-78; S. A. Akintoye, Revolution and power politics in Yorubaland iSjo-rfyj
(London, 1971); Wilks, Asante, 705-10; Raymond Dumett, 'The rubber trade of the
Gold Coast and Asante in the nineteenth century: African innovation and market
responsiveness',/. A/r. Hist., 1971, 12, 1, 93—8; Kwame Arhin, 'The economic and
social significance of rubber production and exchange in the Gold and Ivory Coasts,
1880-1900 ',Cabiersd" etudes A/ricaines, 1980,20, I - I , 4 9 - 6 Z ; R . E. Robinson,' European
imperialism and indigenous reactions in British West Africa 1880-1914', in H. L.
Wesseling (ed.), Expansion and reaction: essays on European expansion and reactions in Asia
and Africa (Leiden, 1978), 154-5; Timothy C. Weiskcl, Trench colonial rule and the Baule
peoples: resistance and collaboration i!if—ifii (Oxford, 1980), 214—15, 221—3; Sara
S. Berry, Cocoa, custom and socio-economic change in rural western Nigeria (Oxford, 1975),
5 3 for the concluding quotation.
741

tionary, populist, appeals by religious cult authorities who

enlarged political loyalties in opposition to secular leaders; these
had been morally discredited and materially weakened, first by
military defeat and then by enforced acquiescence in the Rhodesian
settlers' demands, but were now shamed into resuming the
struggle. This was an exciting thesis, an exploration of how, under
pressure, popular consciousness can be changed by visionary
leadership; but it has been shown to have serious problems as an
explanation. It is not that the concept of' revolution' could not
be grasped in these pre-literate societies, for lack either of a critical
tradition or of class oppositions such as might inspire a subversive
vision of society. The peoples of central Africa were well aware
of the distinctions between hierarchical kingdoms and the stateless
societies which managed to survive on their fringes; and some
territorial cults did embody, in oracular guise, submerged dyna-
sties, alternative political communities. While there is room to
doubt whether the Zimbabwean cults and spirit mediums did in
fact play such a revolutionary role, there can be no objection in
principle to their so doing, especially when popular discontent
gave them a potential following to satisfy.
There are two more substantial objections to the original thesis.
One is that it did not pay sufficient attention to the non-military
networks of co-ordinated authority which all African communities
had to have, in order to insure against ecological disaster, to
exchange specialised articles of use, and to reinforce with marriage
alliances the local authority of big men, chiefs or kings in their
control of tribute, trade and the access of their juniors to women,
land and cattle. These alliances were expressly designed to be
called upon in crisis, and a crisis is what the Zimbabwean peoples
faced, a compound of locusts, rinderpest and European claims on
what slender material resources remained. Joint action, in short,
did not require creative coordination; there had been no funda-
mental crisis of monarchy or chieftaincy to disrupt such channels
of co-operation as already existed. The second objection brings
back the question of state-building, European and African. The
' revolutionary' thesis rested on the distinction between resistance
and rebellion. Resistance, it was argued, had seen the old Ndebele
order of king, chiefs and regiments go down fighting in all their
anachronistic splendour. The subsequent rebellion could be
organised only by new men, upstarts and prophets, men not
742

imprisoned by co-optation into the new colonial order. There are

two difficulties here. Early colonial rule, even in settler Rhodesia,
had not the power to institute so sharp a break; it was always a
disputed process. And Ndebele royal successions, brought about
in this case by the death of the fugitive king Lobengula after his
war of resistance in 1893, were generally lengthy, contested affairs
which mobilised all the competing factions of the kingdom. The
rebellion of 1896 can therefore be seen as a continuation of
resistance once the political coalitions of the kingdom had been
put together anew.84 Most African rebellions are perhaps likewise
best seen as 'delayed resistances', if only because the colonial
powers so often delayed the transition between the politics of
conquest and the demands of rule.
The fourth point which needs to be made in analysing African
decisions and capabilities in the 1890s is, therefore, that even the
most shattering events of conquest were merely concentrated
episodes in the unfinished and unpredictable processes of mobi-
lising political coalitions. Decisions were reached in battle but
they were never final ones, and victories were neither won nor
consolidated by firepower alone. European demands enlarged
with time, consistent only with the contradictory requirements of
military security and minimum cost. The first colonial officials had
to extract resources of manpower and food supplies from African
allies on terms which demonstrated their own mastery but which
were not so mean as to provoke defection. Conversely, they
needed their allies' help in subduing the irreconcilables, but not
on such a scale as would turn their satellites into their protectors.
Such daily political calculations were more influential in shaping
local colonial regimes than any imperial blueprint for the con-
trolled exploitation of tropical estates. Even where, as in the
colonies of settlement and mining investment, the needs of
European clients were paramount, they could not be satisfied to
the total exclusion of African agents from the political estab-
lishment. The subsequent structures of colonial states were not
governed by their economic function; their economic function
84
Ranger, Kevolt in Southern Rhodesia for the original thesis; since criticised in Julian
Cobbing,' The absent priesthood: another look at the Rhodesian risings in 1896-1897',
J. Afr. Hist., 1977, 18, 1, 61-84; and D. N. Beach, '"Chimurenga": the Shona rising
of 1896-97', J. Afr. Hist., 1979, 20, 3, 395-420; Terence Ranger's discussion of these
and other critics is in his preface to the first paperback edition of Kevolt (London, 1979),
ix—xviii.
743

was governed by the political alliances which competing interests,

white and black, were able to stitch together in order to press their
claims on the allocation of power. Outcomes were not
predetermined; colonial officials were participants in the
bargaining, not its arbiters. African accommodation or resistance
to official impositions were thus equally plausible, and equally
risky, means of setting conditions for their inclusion in or raising
the price of their rejection from the shifting coalitions of rule.
Some early European supporters of conquest were just as liable
to suffer displacement from power as their African counterparts.
In warfare, in military supply, and in trade the events of conquest
were full of surprises.
African military resistance invited defeat and dismemberment,
as in the cases of the Shona or the Herero, the Tukolor empire
or Bunyoro. But it could as well enforce more lenient bargains
upon the Europeans both directly, as the Baule found when they
first fought off the Fiench, or the Ndebele when they almost broke
Rhodes in 1896; and indirectly, as when the Sierra Leone hut tax
revolt of 1898 warned the British off their intention to levy direct
taxes on the Gold Coast and in Nigeria. Conversely, European
victories were generally but by no means always victories for their
African allies as well. Chief Mareale of Marangu was able to
eliminate his local rivals with German arms, the Protestant
warlords of Buganda used the British against their king. Africans'
defence of their own interests could also appear to be a gratifying
steadiness on behalf of their European conquerors. The rulers of
Sokoto were beaten by the British in 1903; three years later, only
the sultan stood between the British and disaster, when the Mahdi
of Satiru wiped out a company of the West African Frontier Force.
The Mahdi was as subversive of the Fulani sultan's authority as
of the British. Their joint suppression of the revolt 'changed
relations between the British and the ruling Fulani from super-
ordination based on force to a near parity based on common
interests'.85 The chiefs of Buganda made themselves even more
indispensable as the protectors of the British when in 1897 they
helped to crush the mutiny of the British mercenary force of
Sudanese. But these were exceptional cases. The more general
experience of the African allies of conquest was to find that the
85
M. G. Smith, Government in Zaâu (Oxford, i960), 205.
744

privileges of loyalty were gradually reduced and finally spurned.

There were three stages in their rejection. First, colonial officials
sought to multiply the number of the allies which they could call
upon, in order to reduce their dependence on any one of them.
They then recruited individuals into standing armies under direct
European command, bureaucratising the use of force in order to
limit its delegation to African military contractors. By 1908, the
use of African auxiliary forces or native levies was finally banned
by imperial authorities who were determined to bring an end to
the pursuit of pacification through civil war or, as Winston
Churchill put it, disturbing without governing.86 The political
relations of conquest were finally superseded by the political
relations of control. All, allies and recalcitrants alike, had to
submit to the symbolic and material obedience of taxation. All,
even the staunchest of allies, were required to hand over their
guns. Force became a monopoly of colonial states.
The first European agencies of colonial rule fared no better in
this gradual process of removing public power from private
hands. Imperial authorities disposed of the British and German
East Africa Companies when their resources proved unequal to
maintaining national prestige against the general costs of conquest
or outright African revolt. The British also revoked the Royal
Niger Company's charter, for fear that it would not be able to
keep out the French. Rhodes' discomfiture at the hands of the
Ndebele and Shona was punished by an increase, however minor,
in imperial supervision of his company's rule. The Portuguese
officials in Mozambique similarly tried to rid themselves of
dependence on their baronage, the pra^p-holders, but with little
success; the government of Mozambique remained as honey-
combed as any African kingdom with private jurisdictions.87 The
Congo Independent State and French Equatorial Africa were
similarly company regimes, but by intention. To judge by their
wretched histories, the alternative to commercial bankruptcy was
forced labour, the death or dispersion of African workers, and the
destruction of natural assets, rubber especially. They pursued not
peace, but profit, through civil war. And even here the era of the
86
Margery Perham, Lt/gard, v o l . 11: The years of authority iffg-ifjj (London, i960)
248.
87
Lcroy Vail and L a n d c g W h i t e , Capitalism and colonialism in Mozambique: a study of
Quelimane district ( L o n d o n , 1980), chs. 2 and 3.
745

quasi-sovereign concessionary company did not long outlast the

disarmament of the scarcely more destructive warlord chief.
The main military problem in the conquests of Africa was
overwhelmingly the problem of supply. Local administrative
budgets were mostly taken up in food-buying, fuel-cutting and
porterage; draught oxen succumbed to tsetse and ammunition
wagons stuck in the mud.88 Outside Egypt and South Africa all
of Africa's first railways were built with military requirements
uppermost in mind. The military food market consolidated
colonial African alliances as firmly as the supply of auxiliary
troops. French officers in the western Sudan depended on the
network of Senegalese merchants and did all they could to
facilitate their trade, reserving their contempt for their fellow
French businessmen who profited by the expenditure of French
blood. Even when the conquerors tried to reduce the costs of food
supply it did not necessarily mean the repudiation of their allies.
The chiefs of Buganda, for instance, contemplated refusing
exorbitant British demands before they agreed to supply food free
of charge. According to the Ganda historian, Batolomayo Zimbe,
the British then ' changed their attitude towards the Baganda and
favoured the habit of discussing with us any measure they wanted
to take'. The Christian leaders well understood the importance
of supply. Their superiority as quartermasters rather than gunmen
had just won them victory over their Muslim rivals in succession
war.89 But food supply and porterage could also ruin friends and
enemies alike. The great rinderpest epi2ootic was the most
catastrophic example of indiscriminate misfortune. Pleuro-
pneumonia among slaughter stock, dysentery and pneumonia
among ill-fed and ill-clad porters, tested the loyalty of the African
allies of conquest in Kenya. The slave-holders of the western
Sudan, as already mentioned, stoked the resentments of their
slaves in driving them to produce more foodforsale to the French,
their future conquerors.
In the sphere of civil commerce, as distinct from military
supply, conquest and its enlargement of European trading
privileges did not necessarily spell the end of capital accumulation
by African merchants, or their reduction to mere commission
88
The best study of the supply problems of conquest is in Matson, Nandi resistance.
80
Kanya-Forstner, Conquest of the western Sudan, 201; Michael Twaddle,' The bakungu
chiefs of Buganda under British colonial rule 1900-1930', / . Afr. Hist., 1969, 10, 2,
313; idem, "Muslim Revolution', 67.
746

agents for foreign business houses. This was partly because

European commercial interests were so competitive with each
other, partly because African trading networks of credit and
partnership were irreplaceable. In Nigeria, the smaller British
firms were just as determined as their African partners to resist
the monopoly claims of the Royal Niger Company. German
traders in Togo, like the French traders in Dahomey, managed
to defend their established connections with African trading chiefs
against the threat of plantation concessions which would have
needed monopoly buying rights and the disruption of domestic
African labour. Even in South Africa there could be effective
alliances of white traders and African chiefs, to delay the expansion
of colonial frontiers and their harbour dues. Africans survived in
trade not only because of their role in European competitions but
also because they had the priceless capital assets of local goodwill
and political authority. When British officials decided to destroy
the independence of the Niger delta merchant prince, Nana
Olomu of Ebrohimi, in 1894, they almost ruined the British
traders who had bought their palm oil from him. Only a power
like Nana could maintain the huge and intricate network of slave
porters and canoe men which kept the puncheons of oil moving
down to the delta ports. Six years later the House Rule Ordinance
tacitly re-established the old system of servile transport workers.
To profit from colonial conquest it seemed necessary to make
one's peace with the lords of trade.90
Finally, one has to ask how far it was death rather than the
Europeans which really conquered Africa at the turn of the
century. Only the west African Guinea forest peoples seem to have
wholly escaped the dreadful mortality of human beings and their
domestic animals. A number of Africans believed that the
Europeans had called biological warfare to their aid; sorcery
might be the better term. It was thought at the time that one-third
of Ethiopia's population died of famine in the 1890s, a harrowing
experience in which civilised existence all but vanished. There was
even some cannibalism, an unheard-of degradation, when the
strong consumed the weak: 'His wife gave him indigestion,' as
the Shoans sang. Around 1890, between 10 and 50 per cent of the
«° Ncwbury, Western Slave Coast, 151 - 5; Manning, Dahomey, 174; Bcinart, Pondoland,
31-4; Cherry J. Gertzel,' Relations between African and European traders in the Niger
Delta 1880-1896',/. Afr. Hist., 1962, 3, 2, 361-6; Robinson, 'European imperialism
and indigenous reactions', 152—60.
747

people of central Kenya died of famine, dysentery and smallpox
not, certainly, because of the conquest but, equally certainly,
weakening their defences against white settlers when they arrived
little more than a decade later. Three quarters of a million may
have died of famine in what is now Tanzania during the 'nineties;
over a quarter of a million of sleeping-sickness in Uganda in the
six years after 1900, about two-thirds of the population in the
affected lake-shore areas. It has been estimated that up to half the
people died of sleeping-sickness and smallpox in the lands on
either bank of the lower river Congo. By 1912 it was thought that
the population of the entire Belgian Congo might have been
reduced by 60 per cent, from 20 million to 8 million, in the quarter
century since the Berlin Conference; but that was only a horrified
guess at the consequences of disorder among diseased populations
as they hid from the exactions of King Leopold's rubber collectors.
On equally slender evidence it can be suggested that demographic
decline was scarcely less catastrophic in northern Nigeria. Depo-
pulation had started in the mid-nineteenth century, as the Muslim
emirates harried their surroundings for slaves, disrupted culti-
vation, precipitated flight to remote areas, and thus allowed room
for the expansion of tsetse-fly bush. A study of an aggressively
commercial chiefdom in Angola ends with the presumption of its
extinction by sleeping-sickness in 1899. Disease stalked all railway
construction and mining camps. And it is not impossible that
three-quarters of all Africa's cattle died in the 1890s of the great
rinderpest epizootic.91
It is plain that the various African reactions to conquest cannot
be properly understood unless they are placed in the context of
a literal struggle for survival. The earlier discussion of the
relationships between rinderpest and the politics of conquest must
serve as a warning that only detailed local evidence can defend
such a perspective against false conclusions. There were doubtless
many peoples of Africa for whom the annual cycles of planting
and harvest or pastoral transhumance continued relatively undi-
" Ford, Trypanosomiases, 240, 592-3, 460, 489; Pankhurst, 'Ethiopian famine', 123;
Marc H. Dawson,' Disease and population decline of the Kikuyu of Kenya, 1890-1925',
in African historical demography, vol. 11 (Edinburgh, 1981), 121-38; Robert W. Harms,
River of wealth, river of sorrow: the central Zaire basin in the era of the slave and ivory trade
i)oo-igfi (New Haven and London, 1981), 231-32; John H. Harris, Dawn in darkest
Africa (London, 1912), 208; Jill R. Dias,' Black chiefs, white traders and colonial policy
near the Kwanza: Kabuku Kambilo and the Portuguese, 1873-1896',/. Afr. Hist. 1976,
17, 2, 265.
748

sturbed, people whose drudgery, particularly the labour of their

wives, was no more than usually increased by the need to pay the
dues of political subordination or ceremonial observance. But for
the many unfortunate ones, devastated by drought and pestilence,
innovation was mandatory and, with it, new forms of conflict.
Necessarily ignoring all cautions against such generalisation, one
can suggest a triangular struggle between Africa's poor and their
rulers old and new, as the premise of economic obedience not only
collapsed but was actively challenged from below. Africa's poor,
if they were dissatisfied with the ability of those above them to
guarantee their livelihood, or if they were not sufficiently
constrained, increasingly took the inauguration and maintenance
of their domestic family cycle into their own hands, in ways which
have already been outlined. Former slaves, family heads, young
men, all explored new avenues of freedom. Their African patrons
and rulers sought to recapture their services by sponsoring raids
to restock herds, by taking in famished neighbours and pawned
women, by hiring out their young men as military auxiliaries or
porters for the Europeans or, if it should come to that, by
organising their defence against European raids. The colonial
invaders tried to turn both sides of this African conflict to their
own profit, now employing individual workers, now patronising
leaders who kept control of their followings. Armed resistance
and armed collaboration alike were the means by which chiefs
defended their rights to their peoples' labour. It was one of the
many ironies of African history that the European conquest,
which first helped to dislocate African livelihoods, then offered
the best means available for their reconstruction.
Meanwhile, both beyond and within the frontiers of European

control, fresh disorders kept breaking out. In the furthest interior,
facing the western slopes of the Ruwenzori mountains, the Bashu
people did not encounter the firearms-and-ivory frontier until the
1880s. Their ritual chiefs were quite unable to control this influx
of power and, in the eyes of their people, their weakness caused
the major drought of the late 1890s. It needed inspired leadership
and the borrowing of an alien spirit cult to restore well-being. In
similar circumstances other chiefs, less imaginative perhaps,
succumbed to despair. Nandi Msuulwa, king in Ufipa, committed
suicide in 1897. His easy relations with Arab traders had come
749

to an end with the colonial partition, which disrupted the

Zanzibar commercial networks. Traders called in their loans;
Msuulwa increased his tribute exactions on his people; his people
fled. Political emasculation brought on sexual impotence. It was
fitting that it was with a gun that he chose to kill himself. In the
northern savanna, trade was similarly disrupted as rifles became
more widely available and the French seized hold of the Saharan
oases. People responded to all this chaos on the frontiers of
colonial state-building by coming together in defensive alliances
in remote corners, forests and mountains. Colonies, like kingdoms,
were plagued by marcher bandits, protectors of their people who
now look like anti-colonial peasant heroes but who were in reality
followers of an old tradition of what has been called ' marginal'
tribalism, the conscious, structured rejection of the demands of
kings.92
Almost everywhere military expenditure continued to outstrip
local colonial revenues into the early years of the twentieth
century. European governments grew increasingly impatient of
the costs of African empire, and military adventures were
notorious for disrupting production and trade. Civilian governors
who took up their posts around 1900 were all under orders to
restrain the enthusiasm of soldiers for hunting blacks. But the fate
of Africans was much the worst where, in King Leopold's Congo
Independent State, there was no metropolitan treasury to finance
the capital outlay of conquest, where, like the Arab traders before
him, the king had to accumulate his capital on the spot. The
horrors of his regime threatened to discredit all colonial empire
in the educated European mind. Economy and morality thus
combined to give a powerful impetus to the civil reconstruction
of colonial rule.
RECONSTRUCTIONS AND EVASIONS
African colonies were all different. Their new civilian governments

faced similar problems. By 1905 they were on the way to
becoming very dissimilar colonial states. Past trading history, the
91
Randall M. Packard, Chitfsbip and cosmology: an historical study ofpolitical competition
(Bloomington, 1981), chs. 6 and 7; Roy Willis, A state in the making: myth, history and
socialtransformation inpre-colonial U/ipa (Blootnington, 1981), 20J-10; Baier, CentralNiger,
62-5; Alien Isaacman, ' Social banditry in Zimbabwe (Rhodesia) and Mozambique,
1894-1907: an expression of early peasant protest', / . Southern Afr. Studies, 1977, 4, i,
1-30; for marginal tribalism, Ernest Gellner, Saints of the Atlas (London, 1969), 2-3.
75°

accidents of conquest, natural endowment, the local distribution
of kingdoms and stateless societies and their particular divisions
of labour, all set each colony apart from its neighbours. But every
government had to try to bring about the same transition from
the politics of conquest to the politics of control; to create or
expand a taxable base; and to cope with the new social disorders
which these changes evoked. Governments everywhere found a
workable solution by coming to terms with locally dominant
classes and interests, the first generally African, the second usually
European, those who could cause the most trouble if ignored and
guarantee the largest revenues if ministered to. But there was a
Limit to what governments could engineer and to what dominant
groups could achieve. Powerful interests were not always com-
patible. And those who were apparently deprived of power, the
vast majority of Africans, continued to show a remarkable capacity
to evade the requirements of the new order or to exploit its needs
for their own purposes, the most important of which, for young
men at any rate, was personal freedom. Moreover, the paradox
of conquest remained the paradox of rule, that power could not
be exercised without giving some of it away. Europeans no longer
shared out the use of armed force; they employed disciplined
soldiers who knew how to ' tramp, tramp, tramp in their military
boots', instead of barefoot levies.93 But they could not do without
the delegated authority of chiefs, labour recruiters, produce
traders or tax clerks, nor easily withdraw their implied approval
of Muslim and Christian teachers, all, the last especially, with their
own private powers of patronage. Colonial states, for all their alien
origins, soon developed characteristics common to all states. They
were shifting coalitions of interest, often with rival definitions of
authority. They were held together by governing institutions with
contradictory responsibilities. Governments had to cultivate the
goodwill of their most powerful supporters, mainly by ensuring
them a ready supply of labour. They also had to see to it that the
privileges of the powerful did not dangerously excite the
disobedience of the poor.
It is often said that, with the exceptions of Egypt and South
Africa, early colonial African trade and investment were altogether
insignificant in the global scale. That is true but liable to be
" Roland Oliver, Sir Harry Johnston and the scramble for Africa (London, 1957). 5 1 ' .
quoting Johnston to Lord Salisbury, April 1900.
751

misleading. There was no shortage of conflicting ideas on how

Africa might be turned to better account, no lack of competitors
for colonial governments' favours and, since colonial rule rested
ultimately upon their acquiescence, the actions or anticipated
reactions of Africans were generally the most pressing political
consideration. Colonial governments received divided counsels,
and with such paltry revenues they had to choose. There were two
broad possibilities before them, conservative and revolutionary.
Governments could continue to serve the same commercial
interests as had been protected by consuls, but even then they
could not carry on quite as before. Too much had been disrupted
in the politics of conquest, too much invested, and Europeans
seemed to have more power. It appeared therefore that they might
resolve the crisis of aristocracy and reconstitute the premise of
African obedience to their chiefs with administrative sanctions,
even if they did not dare to meddle in African methods of
production. Even this conservative project demanded tax, and
tribute labour on building railways or improving roads. The
largest capital investments were all made by African workers.
The alternative was to revolutionise the conditions of African
economic obedience by rewriting the property relations of
production. The strict definition of the ownership of land seemed
to be the master key to African labour. Property might variously
be vested in African chiefs, in peasant freeholders, in the colonial
states, in concessionary companies, in private European farmers.
Thus deprived of their customary rights of land usage, their
capital, Africans would be obliged to work for more than their
customary levels of family subsistence; they would cease to enjoy
the privilege of living as 'a sort of naked leisure class'94 and
become instead a proletariat, clothed in imported cottons,
industrious, respectable, the foundation of a capitalist transition.
This second, revolutionary, possibility was almost everywhere the

first to be attempted. It seemed to be the obvious way forward,
not so much because of the needs of governments' capitalist
clients, but because of the mixture of condescension and fear with
which virtually all Europeans looked upon Africans. If they were
idle, they were also dangerous. Officials feared rebellion as much
as they desired taxation. African labour which became self-
94
C. W . d e K i e w i e t , A history of South Africa, social and economic ( L o n d o n , 1941), 85.
752

disciplined by regular work would, it was thought, furnish

Europeans with the levers of social control as well as the source of
wealth. Moreover, as commodity prices rose from the mid-1890s,
European governments became increasingly concerned to
safeguard the supply of cheap foodstuffs and industrial raw
materials. Their colonial offices were dazzled, if only temporarily,
by dreams of scientifically tended tropical estates. This image of
the future contained planters and workers rather than peasants;
it foresaw units of production which would provide the new
colonial railways with freight by the wagon-load rather than in
penny packets. This, again, was a strategic rather than a specifically
capitalist vision. Nevertheless, the need for changing capitalist
strategies did undoubtedly provide a third stimulus to the
revolutionary approach. The old ' slash and burn or " swidden"
commerce' of the ivory and rubber frontiers was beginning to run
into diminishing returns; it wasted its assets, human and natural.
Africa's forest resources could not compete in cost and quality
with Malaya's new rubber plantations. Mistah Kurtz was dying;
William Lever was agitating to be his successor, anxious to invest
in palm plantations and oil mills. Lord Delamere, the hero of white
settlement in Kenya, first came to Africa in the 1890s like one of
her founder-kings, hunting ivory in the bush; he returned in the
first years of the new century to try out woolled sheep, dairy cattle
and wheat on his private land.95 On the Mediterranean coasts of
Africa a similar transition was under way. In Algeria the annual
cycle of grain-growing by white peasant farmers was giving place
to heavy investment in vineyards, which would pay only in the
long term; Egypt's cotton landlords were taking out more
mortgages to irrigate more feddans. In the south a parallel
enlargement of their liabilities was forced upon the magnates of
the Transvaal. They had begun to exhaust the easily worked
surface outcrops of the gold reefs in the 1890s. In order to exploit
the deep levels of the Witwatersrand they had literally to sink
capital in mine shafts, pumping equipment, lifting gear, and new
methods of processing the comparatively low quality ores.
Long-term investment required political confidence. Colonial
95
Robert Harms, "The end of red rubber: a reassessment',/. Afr. Hitt., 197J, 16,
1, 73-88; Charles Wilson, The history of Unilever, vol. 1 (London, 1954), 165-87;
D. K. Fieldhouse, Unilever overseas (London, 1978), ch. 9; Elspeth Huxley, White man's
country, Lord Delamere and the making of Ktnya, vol. 1: 1/70-19/4 (London, 1935), 11-59,
135-78.
753

governments looked to large-scale production for economic

survival. Big producers looked to their own political survival in
colonial states. Agents of economic growth at first, they ineluctably
became part of the political structure of empire, asfiscal,land and
labour policies became clustered about their needs. Colonial
reconstructions were therefore competitive processes, not the
blueprints of the official social engineers. And the competitors
were as much African as European. The recent politics of
conquest remained as compelling as the new politics of the
capitalist transition. The past exercise of force and the active
alliances which it had generated in African society often proved
to be less of a midwife of further innovation than its most cautious
constraint.06 For the pillars of African society needed to be
buttressed as carefully as capitalists if they were to be the
trustworthy props of colonial rule. Chiefs were unavoidably under
attack from above, in the transition from the diplomacy of
conquest to the administration of control. They were stripped of
their rights to their own public income; they had to turn out
tribute in tax and labour for the Europeans instead. But they also
stood in danger of being outflanked from below by their fellow
Africans, men who took as much advantage of the new order as
European merchants and concessionaires.
On the west coast, African rural capitalists, ambitious peasants,
showed as much confidence in the future as any mining magnate
when they mobilised the resources of their lineages to buy land,
invest labour, and plant cocoa trees which would take at least five
years to come into bearing. In South Africa too, the mineral
revolution of the last quarter of the nineteenth century and the
Anglo-Boer war at the century's turn both opened up expanding
markets in foodstuffs and transport-riding which stimulated
wealthier African peasants to purchase land and labour. These
petty capitalists were almost invariably Christians. They shared
in the most generous colonial vision of future African self-respect.
In these early years it was comparatively rare for their chiefs to
be Christian as well. Christian chiefs 'beat the drum' for the
alliance for progress between the Bible and the hoe (the plough
generally came later) in Buganda, among the Tswana, in some
parts of Yorubaland and in some of the Zulu locations of Natal.
Colonial authorities elsewhere were more often concerned that the
06
G. B. Kay, The political economy of colonialism in Ghana (Cambridge, 1972), ion.
754

new religious identity of young converts would justify and

protect a withdrawal of allegiance from their elders. The expansion
of African agricultural production was all the more threatening
when it occurred under the sponsorship of Muslim clerics, as
happened in Senegal. Here colonial Islam continued its pre-
colonial tradition of appealing to those who were most excluded
from power; it constituted an internal resistance, a resistance by
withdrawal; and a most profitable resistance too, as holy men gave
protection and direction to the labours of their otherwise
masterless young followers in the groundnut fields. It took some
considerable time for the French to accept that their most feared
opponents were in fact becoming their firmest friends. It is
probably true that African acquiescence in colonial rule was
encouraged by the improvement in the global terms of trade for
primary producers which lasted from the mid-1890s until the First
World War.97 But colonial officials were more impressed at the
time by the political threat which seemed to be posed by the
peasant response to the market.
It was the same with African labour. Seasonal or longer term
migrant labour was an African preference before it was confirmed
as official policy. Even where outside employment was first
encountered as a tributary obligation, it was also embraced as a
voluntary opportunity. Young men went out to earn their tax as
a duty to their family; they also won their own bridewealth and
independence from their fathers. Soon older men went out too,
leaving their wives to maintain their land rights and their families'
subsistence with their hoes. The intensification of women's work
in hut and field was the invisible underpinning of all colonial
Africa's economic expansion. Elders raised women's bridewealth
to match the growing value of their agricultural drudgery, so as
to get a percentage of young men's earnings. The standing conflict
between older and younger men became more bitter all over
97
Polly Hill, Tbe migrant cocoa-farmers of southern Ghana: a study in rural capitalism
(Cambridge, 1963); A. G. Hopkins, 'Innovation in a colonial context: African origins
of the Nigerian cocoa-farming industry, 1880-1910', in C. Dewey and A. G. Hopkins,
Imperial Impact (London, 1978), 8 3-96; Colin Bundy, Tbe rise and fall of tbe South African
peasantry (London, 1979); C. C. Wrigley, Crops and wealth in Uganda: a short agrarian
history (Kampala, 1959), 16-18; James Benin Webster, The African churches among tbe
Yoruba (Oxford, 1964), 113-14; J. D. Y. Peel,'Conversion and tradition in two African
societies: Ijebu and Buganda', Past and Present, 1977,77, 108-41; Etherington, Preachers,
peasants and politics; K l e i n , Islam and imperialism, c h . 1 1 ; D o n a l B . C . O ' B r i e n , Tbe
Mourides of Senegal: tbe political and economic organisation of an Islamic brotherhood ( O x f o r d ,
'97'). 5 '~57; Hopkins, Economic history of West Africa, 183 for the terms of trade.
755

Africa, mitigated only by their joint exploitation of women.

Colonial officials everywhere heard endless complaints from
elders, echoing the Zulu chief Mafingo, that outside employment
gave their juniors, especially the literate among them, a new-found
insolence. The number of workers was enormous. One is ac-
customed to think of the Transvaal goldmines as the great
consumers of migrant labour. And so they were, with around
54,000 in the mid-1890s and over 80,000 a decade later; by no
means all of these were turned out as unwilling tribute labour by
their chiefs. But the mine-owners were not the only large
employers in early colonial Africa, nor was migrant labour a new
phenomenon. The involuntary migrants of the western Sudan and
elsewhere, the slave populations, had been numbered in millions.
But probably 100,000 caravan porters, almost all of them freemen,
annually walked the trade-routes from the interior to the Tan-
zanian coast, from the 1880s until the early years of this century.
Similar numbers must have been employed elsewhere. Colonial
armies of conquest had depended on armies of porters. While
railway construction employed and, notoriously, destroyed many
African workers, one effect of the colonial transport revolution
was to discharge vast numbers of Africans from porterage into
less skilled, more menial work. Africa's petty rural capitalists soon
created an even larger demand for labour. It has been calculated
that by 1911 the Gold Coast's cocoa farms employed over 180,000
people, in a population of about 1.5 million. At that time almost
all the labourers would seem to have been members of the
farmers' families, but the first trickle of the migration from the
savanna interior to the ecologically favoured coasts of West Africa
had already begun. By 1970 it would be a flood, up to a quarter
of the savanna population.98
Africans everywhere, African women especially, were hard at
work. Europeans everywhere, government departments promin-
ent among them, were loud in their complaints that African
" Audrey Richards, Land, Labour and Diet in Northern Rhodesia (London, 1939), 104,
298 and passim; Gavin Kitching, Class and economic changes in Kenya: the making of an
African petite Bourgeoisie ifo/-if/o (New Haven and London, 1980), 20, 211, 225 and
passim; Belinda Bozzoli,' Marxism, feminism and South African studies', / . Southern Afr.
Studies, 1983, 9, 2, 139-71; Francis Wilson, Labour in the South African gold mines
1911-1969 (Cambridge, 1972), 70, table 8; Iliffe, Modern history of Tanganyika, 129;
R. Szereszewski, Structural change in the economy of Ghana, iifi-ifii (London, 1965), 57;
Samir Amin, 'Introduction', 72, in Amin (ed.), Modern migrations in Western Africa
(Oxford, 1974).
756

workers were hard to find; they were even harder to keep at work
for any length of time. The African work-force challenged the
expectations of industrious discipline in at least three disturbing
ways. Most of them saw paid labour as a means to accumulate
their own capital in land and dependants. This was as true of the
resident labourers on white farms in South Africa, Rhodesia and
Kenya as of the locally hired workers on the cocoa farms of the
Gold Coast and western Nigeria. African farmers and rural
notables, even the ex-slavemasters of Zanzibar, had no objections.
There was as yet no shortage of cultivable land; their new
dependants had permanent obligations; the payment of regular
employees would, by contrast,' waste their savings', except at the
busiest times of the agricultural year. European farmers likewise
first attracted workers away from chiefs by offering them squatter
tenancies on their land, appropriated by right of conquest.
Conflicts between white landlord and black squatter arose only
when markets expanded, by the 18 80s in South Africa but not until
the first decade of this century in Rhodesia and later still in Kenya.
In the 1890s it was thought that African squatters in the Transvaal
marketed twice as much farm produce as their Boer landlords. Not
only did this depress white farm earnings and challenge the
premise of racial inequality, it also provoked enmity between
small white farmers and the larger, especially absentee, owners.
The former increasingly wanted farmworkers, but Africans
naturally moved to farms with broader acres and easier, tenant,
conditions. The first European objection to African agricultural
workers was, then, that they mocked the rights of property by
exploiting its owners' competition for labour in situations where
the land frontier was closed in law but remained open in fact."
The other objections applied to all workers, agricultural
labourers, miners and urban employees alike. Thanks to their
continuing base in family smallholdings and to the rural social ties
which were reinforced by the tributary demands of colonial states,
workers were able to assert too much independence, whether as
groups when out at work or as individuals when they returned
home. Large employers, especially the mines of the Gold Coast
and southern Africa, were plagued by their employees right from
the start, and the unemployed found it all too easy to exist in a
marginal underworld between town and country. Gangs of
*• Hill, Migrant cocoa-farmers, 18 8 for the quotation; Trapido,' Landlord and tenant'.
757

workers deserted their recruiters when on the way to work; they

ran away from work when they got there, often in search of better
pay and less appalling conditions; they bargained as labour teams
from plantation to plantation or from mine to mine, showing a
disconcerting knowledge of working conditions and employers'
reputations elsewhere in the region; they pilfered from their
employers' property and loafed when the foreman's back was
turned. Around Johannesburg the African tradition of marginal
tribalism flourished anew among the unemployed who lived on
the criminal pickings of an extravagantly disreputable town, and
preyed on migrant workers as they journeyed home with their
earnings. Far from being individually disciplined by work,
Africans were banding together to protect themselves against the
labour market's political inequalities; they deliberately claimed
identities of which they had previously been only partly aware.
Modern tribes were often born on the way to work. But returning
workers were not necessarily obedient tribesmen. To the factional
politics of rural chiefship, which decided who should be bullied
into meeting colonial demands and who exempted by chiefly
favour, they brought the independent resources of new know-
ledge, prestige, and cash. Workers' reconstructions of their lives
were as disconcerting as peasant capitalism,' breaking down tribal
systems', as the governor of Kenya warned in 1910, and
'emancipating the peasant from the rule of his chiefs'.100
What Africans made shift to do for themselves influenced the
structures of colonial states as much as the plans which their rulers
made for them. By itself capitalist property seemed after all to be
no foundation for authority; but serviceable patterns of domin-
ance, if only they could be shored up against dissolution, already
existed in most African societies. Officials soon discovered that
it was easier to collect taxes or recruit labour by recognising local
chiefs rather than by appointing African agents over them, men
of the new age like the Christian Sierra Leoneans along the coasts
of British West Africa, the Christian Ganda chiefs who were
posted out over the surrounding areas of Uganda, or the Muslim
akidas of German East Africa and the demobilised sergeants of
100
James B. Silver, 'Class struggle and class consciousness: an historical analysis of
mineworkers in Ghana', D.Phil, thesis, Sussex, 1981, chs. 2 and 3; Charles van Onseien,
Cbibaro: African mine labour in Southern Rhodesia ifoo-tf)) (London, 1976); idem. Studies
in the social and economic history of the Witmatersrand, vol. 11: New Nineveh (Harlow, 1982),
171-201; Sir Percy Girouard, Memoranda for provincial anddistrict commissioners (Nairobi,
1910).
758

the tirailleurs in French West Africa. Such men had seemed

essential to decisive political accumulation of political power over
African society. But they had two defects as a subordinate
bureaucracy. In French and British West Africa they were not
sufficiently dependent; they exploited their citizenship, actual in
French colonies, implied in British ones, in order to voice political
opposition, in newspapers and deputations, to administrative
demands. Conversely, they seemed to be far too liable to stir up
resistance, whether to taxation or the compulsory cultivation of
export crops. Unlike local chiefs they had neither the need nor
the authority to temper the arbitrariness of their commands with
the canny reciprocities of patronage. Many Germans blamed
akidas for the most devastating colonial rebellion of all, the Maji
Maji rising of 1905-07. They were dismissed from the areas of
revolt, and more closely supervised as ' a necessary evil' elsewhere.
Europeans continued to rely on such men only in German East
Africa and in those parts of the western Sudan which had been
most disrupted by French conquest. Everywhere else they looked
for their strategic core of political support not in an African
bureaucracy but in an alliance with those whom they increasingly
perceived as the 'traditional' chiefs, the aristocracy. External
recognition removed from chiefs most of their traditional in-
securities of tenure. And African traditions became increasingly
dignified, to meet the needs of elders, chiefs and colonial officials
alike.101
What colonial officials urgently wanted was 'a class who in a
crisis can be relied on to stand by us, and whose interests are
wholly identified with ours'. And the evasive subversions of
peasant and worker reconstructions meant that, throughout
Africa, elders and chiefs did indeed share an interest with colonial
administrations in curbing the ambitions of their juniors and
subordinates. Dominated classes were exploiting whatever social
and material alternatives were opened up for them by the
articulation of the various African and capitalist modes of
production. Dominant groups, white and black, were intent on
recuperating their authority through a closer class alliance. It was
not at all a secure alliance, nor was it informed by an agreed vision
of the future. Sir Frederick Lugard, its leading British architect,
101
Iliffe, Tanganyika under German rule, 181-2; Terence Ranger, 'The invention of
tradition in colonial Africa', in Eric Hobsbawm and T. Ranger (eds.), The invention of
tradition (Cambridge, 1983), 211-62.
759

had to confess to his successor in northern Nigeria that 'we are
groping in the dark, with little knowledge of the systems of the
country'. African notables made the most of such uncertainties.
They continued to work the misunderstandings of conquest in
order to preserve for themselves a sphere of invented tradition,
in which their private authority was protected from rather than
invaded by European rule.102 Chiefs also helped to devise modern
tribal identities. The ' underdevelopment' of colonial Africa is a
shorthand term for the three-cornered conflict between the
tenacious evasions of workers and peasants and the tributary
structures of patronage within which their rulers, white and black,
sought jointly to contain them, but which capitalists sought
variously to exploit or to dissolve.
The contradictory politics of establishing colonial control and

capitalist profit had three broad outcomes. The patterns were
taking shape by 1905. Around this date all colonial administrations
were alarmed by what they could not then foresee were the final
African protests against the terms of the transition from the
politics of conquest to the era of control. The massive risings in
German East and South West Africa, the reluctant Zulu rebellion
in Natal, the Mahdist Satiru revolt and tax resistance by the Tiv
in northern Nigeria, the continued expense and uproar of the
unfinished business of imposing control over the generally
stateless peoples on colonial frontiers, all concentrated official
minds on the problem of constructing a more peaceful politics of
collaboration. And the scandals of red rubber in Leopold's Congo
persuaded other governments not to delegate any further public
responsibility to private companies. In British West Africa the
outcome was an accommodation between commerce and un-
completed conquest. In central and eastern Africa there was a
series of uneasy compromises between the politics of conquest and
a capitalist transition. Only in South Africa did a decisive political
accumulation appear to clear a path for capitalist accumulation,
and even that was not without its pitfalls.
102
Both quotations are from Lugard, in 1904 and 1908 respectively. Thefirstis quoted
in Robinson, 'European imperialism and indigenous reactions', 1)9-60; the second is
from C. W. Newbury, "The economics of conquest in Nigeria 1900-1920: Amalgama-
tion reconsidered', in £tudes africaines offtrtes a Henri Brunscbwig (Paris, 1982), zj *;
D. C. Dorward, 'Ethnography and administration:
urustra a study of Anglo-Tiv "Working
misunderstanding"',/. Afr. Hist., 1974,. 15,
>5. 5,
5 457-77.
760

The colonial authorities in British West Africa failed to
engineer a full transformation from conquest to control. Lugard
was not alone in his conviction that political security must rest
upon private property. African chiefs, former slave-masters,
clearly had to be given a self-interest in supporting British rule.
Government, he acknowledged, had 'by the very act of
introducing security for life and property, and by throwing open
fertile land for cultivation', added t o ' the difficulty of the problem
which it has to solve, namely the creation of a labouring class to
till the lands of the ruling classes'.103 But the class consciousness
of British officials gave them a false consciousness of the
perceptions of African rulers. African peasants and rural capitalists
were already striking private bargains of mutual advantage with
their chiefs, establishing household production under tenancy or
share-cropping arrangements, or by land purchase. On this
misunderstanding the whole edifice of the intended capitalist
transition collapsed in ruins. Once cautioned by the Sierra Leone
hut tax revolt, the British in the Gold Coast and in southern
Nigeria failed to press home their plans for direct taxation, for
control over the chiefs' disposal of land, and even for much of
a say in chiefs' appointments. They continued to preside over
contractual agreements between internal notables and external
traders. The old frontier of commerce was too well established
to be shouldered aside in favour of any new means of intensifying
production, whether on European plantations or African estates.
Shortly before the First World War the West African Lands
Committee conceded defeat and called it principle. It enshrined
the nebulous concept of communal land tenure under tribal
authority, in order to hide the nakedness of British weakness and
ignorance behind the fig leaf of an evolutionary paternalism.104
Modern tribes were also born as an administrative device for
categorising the unknown. Meanwhile, colonial revenues and
business turnover were expanding satisfactorily enough, thanks
as much to buoyant internal trade as to growing export
production.
Central and eastern Africa showed no such clear pattern. In the
105
Robert W. Shenton,' The development of capitalism in northern Nigeria' (Ph.D.
Toronto, 1981), 61, quoting from Lugard's political memoranda of 1906.
104
Robinson, 'European imperialism and indigenous reactions'; Anne M. Phillips,
'The makeshift settlement: colonial policy in West Africa" (Ph.D. London 1982),
ch. 4.
761

Congo and in the Portuguese colonies, governments and

companies had yet to extricate themselves from their destructive
collusions in coercion. In Kenya and the Rhodesias settler farming
had barely begun to challenge the collaborative politics of conquest
in the African 'reserves', with its encouragement of peasant
production for the market. In German East Africa the Maji Maji
revolt had been sparked off by African hatred of enforced cotton
cultivation on what were effectively state-owned fields. In its
aftermath, officials and settlers competed to define the future.
Governor Rechenberg planned to base his revenues and political
settlement on free peasant agriculture. The local settlers and their
friends in the German Reichstag had other dreams, of plantations
well supplied with labour. By 1914 the German settlers had won
the battle for reconstruction; they achieved the status to which
the white settlers in Kenya were never quite admitted. They had
become the core political elite, the one class to be relied upon in
a crisis. The Christian chiefs of Buganda had won that position
much earlier, by 1900. These leaders of 'something like a million
fairly intelligent, slightly civilized negroes of warlike tendencies,
and possessing about 10,000 to 12,000 guns' were, Sir Harry
Johnston recognised, 'the only people for a long time to come
who can deal a serious blow at British rule...'. That is why he
garrisoned his Indian troops at their capital. But his special
commission was to introduce civilian rule. That is why he
concluded his Buganda Agreement. This gave the chiefs, amongst
other things, the freehold property which Lugard was soon to
propose for the emirs of northern Nigeria. The Christian chiefs
took to the role of improving landlords with scarcely more
enthusiasm than Muslim emirs; they both preferred the more
flexible profits of power. Within a generation almost all the chiefs'
followers were peasant smallholders, and in the 1920s the British
administration was backing this ' nation of small farmers' against
what was now seen as the dead hand of landlord rights.IOS In
eastern Africa therefore, by 1905, white settlers were neither
convincing capitalists nor yet a reliable strategic elite; African
chiefs provided the essential core of political support, but were
never to become the dynamic managers of capital.
With the conclusion of the Anglo-Boer War in 1902, the British
105
Iliffe, Tanganyika under German rule; D. A. Low and R. C. Pratt, Buganda and British
overrule (Oxford, i960), 94; Wrigley, Crops and wealth, ch. 4.
762

in South Africa seemed to have a clear field for rebuilding a
society at once thriving and loyal. There appeared to be an
obvious political elite, the Randlords of the Transvaal. Only the
revival of their prosperity seemed capable of attracting the British
investment and immigration which would guarantee a unified
South Africa's adhesion to the empire. But because the state and
capital were so much stronger here than in the rest of Africa, so
too was the competitiveness of the post-war reconstruction; the
powerful had more to hope for, the weak had more to fear. The
Randlords were not at all disposed to serve as the public treasure
chest. Their first responsibility was to their overseas shareholders,
not to the local society growing up around the mines. British
officials deplored their reluctance to be incorporated as a
governing elite within the enlarged municipality of Johannesburg,
their selfish desire 'to be exempted from the labour of local
government, the combined business of all businesses, and the
common interest of all interests'.106 The alliance between state and
capital here wa^ no easier than the alliance between colonial states
and African aris "ocracies elsewhere. The mass of Africans were
still more unwilling to play their part in the scheme of recon-
struction. They had seized the chance to expand their peasant
economies during the Anglo-Boer war. They now refused to be
pushed back into the role of mine labour on reduced wages. The
new British South African state needed the mines' revenues. The
mines needed the labour. It seemed cheaper, both materially and
politically, to import the workers from China rather than to
recapture the local African peasantries. Neither calculation proved
to be correct. The recruitment and transport of Chinese coolies
cost twice as much as African migrant workers. And the white
mineworkers, Boers and British alike, were frightened that this
collaboration between state and capital would depress the price
of all labour, their own included. Boer opinion was further
inflamed by the aggressive efficiencies of British administration,
perhaps especially by its introduction of compulsory education for
all whites. To the Afrikaners, their past had been destroyed in the
war; their future had been threatened too, by the deaths of their
women and children in the concentration camps. Now that future
106
'Report of the General Purposes Committee of the Town Council upon the
Memorandum of the Chamber of Mines', October 1901, in Lionel Curtis, With Milner
in South Africa (Oxford, 1951), 190.
763

seemed likely to be placed in chains. The rural Afrikaner poor
were fleeing from the devastations of rinderpest, war, drought and
commercialising landlords. They trekked into the towns, into the
arms of patrons who were already there, English employers,
English trade unionists, English churches and now, it appeared,
English-language schools run by the state. Boer notables,
landlords, lawyers and preachers, would not let their following
disappear so easily. The British imagined that they had gone to
war against Afrikaner nationalism, but it emerged instead as
another competitor in the post-war reconstruction. And the threat
of Chinese labour ensured that it would attract the votes of the
English workers whom British officials had thought would be
solid for the empire. Nowhere was colonial reconstruction quite
so vigorous; nowhere was it quite so confounded.107
The future of colonial Africa was everywhere undecided, and in

1905 much of Africa was not yet colonial. Ethiopia and Morocco,
even the Egba in Yorubaland, remained independent. Colonial
boundaries were still frontiers of opportunity for smugglers even
where they were no longer being fought for. Nevertheless, the
civil reconstructions of government which were in progress
signalled the end of the beginning of colonial rule. The
inauguration of the French West African Federation in 1905
marked the most comprehensive attempt to suppress the political
and fiscal anarchy of military conquest. The characteristic division
among rulers was no longer between soldiers and civilians but
between political officers and legal departments. The violence of
conquest was giving way to the rule of law. However, since
African workers had evaded the constraints of property, legal
codes themselves bristled with coercions. Labour service had to
be forced out where the conditions for free labour could not be
imposed. African resistance to tributary labour service or rising
rates of taxation periodically demanded a punishment of which
the violence shamed the new self-image of colonialism. Natal's
ferocious repression of the Zulu resistance to hut tax would have
107 pctcr Richardson, Chinese mine labour in the Transvaal (London, 1982); Hermann
Giliomee, 'Historical introduction' to A. du Toit and H. Giliomee (cds.), Afrikaner
political thought, 11: tiyo-ifio (Berkeley, forthcoming). My conclusion that the South
African outcome was not as the British intended is entirely conventional. I would not
however dispute that the unintended outcome was in many ways satisfactory to Britain.
For this latter view see chapter 8 by Shu la Marks; also, Marks and Trapido,' Lord Milner
and the south African State*.
764

passed without comment ten years ealier. But in 1907 the colony's
' disgusting butchery of natives' earned it the title of' the hooligan
of the British Empire' from Churchill, Under-Secretary at the
Colonial Office.108 On his visit to British East Africa later that
year, he found occasion to lecture an exasperated district official
on the necessary majesty of the law. It might be infuriatingly
pedantic in overruling the official's judgement on a technicality;
but in so doing the law, Churchill believed, enhanced rather than
reduced his personal standing in the eyes of his tribesmen. It
showed that even the district commissioner had to be obedient
10 the immense and mysterious source of authority by which he
was at the same time supported.109 The peaceful mastery of
colonial governments over their subjects now seemed to depend
upon the self-restraint of their rulers, except — and the contra-
diction was fundamental - when they required their subjects'
labour.110
It is also possible to see now, as it was not then, the first signs
of the beginning of the end of colonial rule. In Mediterranean
Africa, colonial powers were already beginning to be submerged
by what has been called ' the retaliation of Eastern fertility', as
native populations expanded and agricultural yields declined.111
In Africa south of the Sahara populations were still reeling, but
young men were on the move. Christianity was claiming its
thousands of youthful enthusiasts, Islam its tens of thousands. The
first modern generation of African nationalist intellectuals, men
like Africanus Horton and Edward Blyden in West Africa, Tiyo
Soga and Tengo Jabavu in the south, had passed or was passing
from the scene. The 'nation' which they had most nearly
represented, the African Christian and Muslim officials of the
early colonial regimes, was also being cast aside in the reconstitu-
tion of African aristocracy. A younger generation now going to
school, the second political generation of colonial Africa, was to
place its faith in the rule of law. A third generation, the men who
would challenge the rule of colonial law and become the elder
loi
R. H y a m , Elgin and Churchill at the Colonial Office, ifoj-ifot: (be watershed of the
Empire—Commonwealth ( L o n d o n , 1968), 2 ) 1 .
109
W i n s t o n Churchill, My African journey ( L o n d o n , 1908), 1 6 - 7 .
110
K a y , Political economy of colonialism in Ghana, 9 ; J o h n Lonsdale and Bruce Berman,
'Coping with the contradictions: the development of the colonial state in Kenya,
1895-1914',/. Afr. Hist., 1979, 20, 4, 487-505.
1
'' Jacques Bercjue, French North Africa: the Maghrib between two world wars (London,
1967), 88.
765

statesmen of post-colonial Africa, was being born. Nnamdi

Azikiwe, Hastings Banda, Habib Bourguiba, Felix Houphouet-
Boigny and Leopold Senghor were all alive in 1905 or the year
after. Albert Luthuli, son and grandson of Zulu Christians and
the epitome of the second political generation, became a mission
muleteer in 1908. The third generation was not far behind. In 1909
Kamau Ngengi, a Kikuyu teenager later to be known as Jomo
Kenyatta, presented himself at his local Church of Scotland
mission to become apprenticed as a carpenter.
766

BIBLIOGRAPHICAL ESSAYS
I. AFRICA ON THE EVE OF PARTITION
The bibliographical material relevant to an introductory survey

is potentially far more extensive than that for any of the
succeeding chapters. Indeed, a full treatment would probably
include a major proportion of the complete bibliographical
section of volume 5 of this History. What follows is, in contrast,
a deliberately selective indication of the items most essential to
an understanding of the total African scene around 1870, and
those which have particularly influenced the shape and scope of
chapter 1.
For the societies living around the shores of the Sahara, the
notion of 'desert-side' economies comes from the works of
Curtin, Baier and Lovejoy. Curtin (1975) is a study of Senegambia
during the time of the slave trade, and therefore deals more with
the Atlantic than the Saharan trade, but many of the ideas that
emerge from this book have proved extremely useful for the study
of the economic and social history of western Africa as a whole.
The effects of cycles of drought and recovery on the economies
of the central Sudan, and the social ramifications of these
economies, appears in the study of desert-savanna trade by Paul
E. Lovejoy and Stephen Baier (1975). Another important analysis
of a desert-side economy appears in the earlier chapters of Baier
(1980). Studies of the social impact of such an economy occur in
the chapters by Baier and Lovejoy, and Roberta Ann Dunbar, in
Miers and Kopytoff(i977). The theme of desert-side economies
is also dealt with in D. D. Cordell (1977) and R. Roberts (1980).
Baier (1980) is also an excellent introduction to the subject of
trans-Saharan trade in general. The historiography of this trade
is now very large, and has exposed not only its intricacies and
ramifications, but also the scale of its operations towards the end
of the nineteenth century. One of the earliest studies of the
trans-Saharan trade was Bovill (1933, revised 1958 and 1968).
Boahen (1964) examines British interests in the trade up to the
767

middle of the nineteenth century, and the campaign to end the

trade in slaves connected with it. The great work by J.-L. Miege
(1961—3) is concerned with the western routes of the trade.
Boahen's conclusion that the trans-Saharan trade was in decline
in the nineteenth century was questioned by C. W. Newbury
(1966), which in turn was revised by M. Johnson (1976).
Johnson's series of penetrating and marvellously lucid articles
(1970, 1973a, 1973b, 1976) provide many insights into the
complex relationships between West African economies and those
of Mediterranean Africa and Europe. The chapter by L. Brenner
in McCall and Bennett (1971), which concentrates mainly on the
mid-century, was a timely survey of the impact of the trans-Saharan
trade on sub-Saharan Africa as a whole.
In relation to North Africa, Brett (1982) advances the discussion
on problems and perspectives (problems echoed elsewhere in
Africa, notably South Africa) that has been engaging the attention
of historians during two decades or more since the achievement
of independence by the Maghribi countries. Earlier stages of this
discussion have been Douglas Johnson (1964) and Valensi (1969),
also Brett (1972). For Morocco, volumes 11 and in of Miege deal
with the nineteenth century. Trout (1969) is useful on Morocco's
relations with the Saharan and Sudanic lands to the south. Ernest
Gellner (1969) and Robert Montagne (1973) are excellent books
on crucially important elements in the Moroccan situation.
For Algeria during the 1870s Ageron (1968 and 1979) are very
useful, and Julien (1964) Vol. 1 is still of value. The essential
introductory book for the 1870s in Tunisia is Ganiage (1968). For
the British side of the situation there is Marsden (1971). For the
period just before the 1870s, Valensi (1977) offers a brilliant study
of rural conditions in the first half of the nineteenth century. Both
Van Krieken (1976) and Arnold Green (1978) lead one to the
conclusion that Tunisia was capable of a positive response to the
processes of modernisation. The most original part of Leone
(1957—60) is devoted to Libya. One still turns to the remarkable
book by Evans-Pritchard (1949) for a study of the Sanusi
brotherhood.
A comprehensive history of modern Egypt in any language
remains to be written. Extremely useful are Holt (1966) and
Vatikiotis (1969). Little (1958) is still of value. For the period of
Khedive Isma'il there is the huge and unfinished study by Douin
768

I AFRICA ON THE EVE OF PARTITION
(1936-41). Two works by Baer (1962 and 1969) are essential

reading. A work of great originality on some of the main themes
of modern Egyptian history, which can be traced back to the
nineteenth century, is Berque (1967). Landes (1958) was a seminal
work, not only for Egypt, but for the whole subject of the
European scramble for Africa.
The standard general history of the Sudan is Holt (1961). Hill
(1959) is still the best introduction to the history of Egyptian rule
in the Sudan. The consequences of the extension of that rule over
the Equatorial Nile and the Bahr al-Ghazal are carefully examined
by Gray (1961). Holt (1958) takes up the story where Hill leaves
off. A Sudanese point of view is Shibeika (1947). For the expansion
of Sudanese trade and Islam into northern Equatorial Africa, see
Yusuf Fadl Hasan (1971), especially the essay by Robert O. Collins
and Dennis D. Cordell (1979). An interesting article which deals
with yet another of the 'frontier' situations that has been
discussed above is Alessandro Triulzi (1975).
For the main themes of Ethiopian history during the 1870s,
Trimingham (1952) remains one of the best secondary sources for
information on the economic history of the country, along with
Pankhurst (1968). Crummey (1972) is a study of religious and
diplomatic developments just prior to the decade of the 1870s.
Hoben (1973) is an important study of land tenure. H. S. Lewis
(1965) deals with Jimma, one of the south-eastern border states
incorporated by Menelik. The most outstanding study of Somalia
is the work of the great Italian scholar, Cerulli (1957—9).
I. M. Lewis (1965) includes a controversial examination of the
ethnic factors which have determined much of Somali history.
In West Africa the period just prior to partition has been well
served by its historians. Many of the general works contain great
resources of information, analysis and comment. For example,
Hargreaves (1966, 1974), Fage (1969, 1978), and Curtin's own
contribution to Curtin, Feierman, Thompson and Vansina (1978);
Isichei (1977); Ajayi and Crowder (1974). On the economic side,
Hopkins (1973) is of unique importance, while the work of
Newbury (1961, 1966, 1969, 1971, 1972) sheds much light upon
the economic and social changes of the nineteenth century. The
nineteenth-century West African social and political economies
have been studied by Marxist and neo-Marxist scholars, many
influenced by French Marxist social anthropologists such as
769

Meillassoux, Terray and Godelier. Law (1978) is a witty and

critical analysis of some of the large and expanding literature on
African modes of production and class differentiation. Internal
slavery is one of the major features of the period, and Cooper
(1979) provides an excellent assessment of the literature and of
the conceptual problems involved. Further views are developed
by Martin Klein and Paul Lovejoy (1979) and by Law (1978).
The literature on specific economies or political societies in
West Africa is large and varied: only a few examples can be cited
here. In a class by themselves for sustained research and command
of detail are the studies of the southern Mande empire of Samory
Ture by Person (1968-75) and of nineteenth-century Asante by
Wilks (1975). Other outstanding examples are Adeleye (1974),
Isichei (1973), Akintoye (1971), Oloruntimehin (1972) and
Latham (1973).
In contrast with that of West Africa, the historiography of West
Central Africa on the eve of partition is still lacking in any good
general work dealing with the region as a whole, and therefore
presents a fragmented appearance. For Cameroun, Marion
Johnson (1973) is most informative on the economic history of
the area during the period just prior to partition. For Gabon there
are the studies of Patterson (1975a and 1975 b), the latter raising
questions of much wider applicability. Avaro (1981) and Martin
(1972) are both excellent. Coquery-Vidrovitch (1969) is still very
useful on the period prior to the scramble. Vansina (1966) is the
indispensable guide to the pre-colonial history of the savanna zone
of Central Africa, including some of the areas that stretch
northwards into the Zaire basin; his study of the Teke kingdom
(1973) is also highly relevant to this chapter. Harms (1981) is a
carefully researched monograph which recounts the rise and fall
of the Bobangi trading communities. Harms also contributed an
essay to Miller (1980), which includes several other studies
relevant to this area and period. There are some informative
chapters in Gifford and Louis (1967), and also in Gann and
Duignan (1969). Again many of the themes concerning trade,
economic change and political instability in eastern and western
Bantu Africa on the eve of partition, are studied most authorita-
tively in the essays edited by Gray and Birmingham (1970).
The most informative introduction to East African history
during the pre-colonial finale is contained in the relevant chapters
77°

of Oliver and Mathew (1963). This can be supplemented by Ogot

(1974), especially the chapter by Alpers. A single-country history
which has set new standards in African historiography is Iliffe
(1979). For the beginnings of European missionary endeavour
there is Oliver (1952). Although it ends with the death of Sultan
Sayyid Sa'Id, Nicholls (1971) and Cooper (1977) are indispensable
for an understanding of the history of the East African coast. For
northern Mozambique there is the excellent study by Alpers
(1975) which examines the activities of the Yao trading
communities.
Some of the works dealing with the huge area of Bantu Africa
covered by the outreach of Portuguese trading activities have
already been cited: Vansina (1966); Gray and Birmingham (1970);
and Miller (1980). To these must be added David Birmingham
and Phyllis Martin (1973) and Birmingham's masterly contribu-
tions to earlier volumes of the Cambridge history of Africa,
republished in a separate volume (1981). Oliver (1957) enlightens
areas of west and east, as well as central African history, as do
the papers edited by Ranger (1968). There is the great monograph
on the colonial wars in Angola by Pelissier (1977) and, on
southern Angola, the important study by Clarence-Smith (1979).
For Zambia there are the two works by A. Roberts (1973 and
1977). For Malawi, there is the collection of essays edited by
Bridglal Pachai (1972). The history of the lower Zambezi valley
in Mozambique during this period is dominated by Isaacman
(1972) and Newitt (1973). Another important work is Leroy
Vail and Landeg White (1980).
The major general history of Madagascar is still that by
Deschamps (1972). There is also P. M. Mutibwa (1973).
For Zimbabwe the best ethno-historical background is to be
found in Beach (1980). Stokes and Brown (1966) is still useful as
an introduction, while Ranger (1967) deals with the period
leading up to the first confrontation with the white colonists.
The essential introductory work to this period in South Africa
is the first volume of Wilson and Thompson (1969). Also very
useful is the collection of essays on southern African societies
edited by Thompson (1969). De Kiewiet (1941) remains full of
vigour and insights. Many of the main themes pursued in this
chapter are examined by the contributors to Marks and Atmore
(1980).

2. THE EUROPEAN PARTITION OF AFRICA
'If an historian conscientiously tries to encompass all the elements

... relating to [the partition] — the economic, social, strategic,
intellectual, cultural, religious, personal, domestic—political,
Afrocentric, peripheral and the rest - he is getting close to
attempting a sort of total history of the period' - P. M. Kennedy,
Historical] J[ourna/], 1977, 20, 762. Nevertheless, however multi-
farious these 'elements' and the drives and motives which they
generated, the partition in fact consisted of a series of acts of state
— either unilateral annexations (or assertions of control), or
diplomatic bargains. Decisions to assert control were often
(though by no means always) prompted by situations in Africa.
But these decisions were necessarily taken within the constraints
of great-power relationships in Europe and throughout the
world; while diplomatic bargains reflected, more or less directly,
the fluctuating pattern of great-power alignments and antag-
onisms. Above all, until the closing years of the scramble no
leading policy-maker would jeopardise metropolitan security,
even remotely, for the sake of empire in tropical Africa. Between
January 1887 and December 1889 the British cabinet considered
only five papers relating to the partition of tropical Africa; it
considered twenty on the perceived threat of a French invasion
and on the necessary naval and military precautions. Indeed,
throughout the 1880s Africa (other than Egypt) remained a very
minor preoccupation for the cabinet. Between 1880 and 1889 it
considered only fifteen papers on tropical Africa (and only about
thirty on southern Africa), compared with nearly 300 on the Near
and Middle East (including Egypt) and on Central Asia.
Given these strategic and diplomatic constraints, which also
affected, mutatis mutandis, powers other than Britain, most serious
studies of the partition - as opposed to studies in the theory of
imperialism - begin, though they do not necessarily end, by
exploring the official records of foreign and colonial ministries,
and the 'semi-official' papers of ministers and senior officials.
Most of them also exploit the records of unofficial commercial or
other organisations wherever these are relevant. For some of the
partitioning powers, notably Britain, France and Germany, there
are guides to the diplomatic and colonial archives; but the wide
chronological and geographical range of these, and their omission
772

2 THE EUROPEAN PARTITION OF AFRICA
of unofficial material, make them rather inconvenient instruments

of initial orientation on 'sources for the partition'. A more
practical approach is to consult the ' lists of sources' in competent
secondary works which have utilised the relevant material. Simply
as samples, the following may be noted. For British sources:
J. E. Flint (i960); R. Anstey (1962); G. N. Sanderson (1965);
J. S. Galbraith (1972); G. N. Uzoigwe (1974); A. N. Porter
(1980). Sources for both British and French activities in West
Africa are indicated by J. D. Hargreaves (1963). For sources on
various aspects of French policy, see: J. Ganiage (1959); H.
Brunschwig (1963); C. M. Andrew (1968); A. S. Kanya-Forstner
(1969); P. Grupp (1972); M. Michel (1972). For German policy,
see (e.g.) F.-F. Miiller (1959); and above all H.-U. Wehler (1969).
For the Congo State, see R. O. Collins (1968); and especially the
very rich references in the numerous contributions by J. Stengers
- e.g., that in P. Gifford and W. R. Louis, France and Britain in
Africa (1971).
These works (and many others) also contain rich bibliographies,
often critically annotated, of published material (but note that the
third, soft-back, edition (1972) of H.-U. Wehler (1969) retains the
Quellenver\eichnis but lacks the bibliography). There are useful
bibliographies for South Africa during the period of partition in
J. S. Marais (1961) with material in Afrikaans, and in D. M.
Schreuder (1980). The numerous 'Bulletins Historiques' on
imperialism, colonialism and Afrique noire contributed by H.
Brunschwig to R[#>#e] H{istorique] (especially 1957ft".) are very
useful guides; as are the relevant' review articles' that appear from
time to time in HJ, J[ournal of] A[frican) H[istory] and J[ournal of]
I[mperial and] Commonwealth] H[istory].
Attempts at overall explanation or interpretation of the partition
are often difficult to distinguish from works on the 'theory of
imperialism'. Recent works attempting an overall explanation,
and often critically surveying the rival'general theories', include:
L. H. Gann and P. Duignan (1969); G. N. Sanderson, JICH,
1974 and in E. F. Penrose, ed. (1975); W. Baumgart (1975,
English trans. 1982); and W. J. Mommsen (1977). Two powerful
explanatory models, applicable primarily to Britain and Germany
respectively, have been developed in the past two decades.
R. Robinson and J. Gallagher, in Africa and the Victorians (1961),
and elsewhere, have introduced the concepts of'informal empire',
773

'strategic imperialism' and 'the local crisis'. Their work has

revived the serious study of imperial history in the English-
speaking countries; and has generated a rich literature of comment
and criticism, some of which may be sampled in W. R. Louis, ed.
(1976). H.-U. Wehler has developed, in his massive Bismarck und
der Imperialismus (1969), the model of 'manipulated social
imperialism'. He has summarised some of his conclusions in P[ast
and] P[resettt], 1970; and in R. Owen and B. Sutcliffe, ed., Studies
in the theory of imperialism (1972). Criticism has concentrated on
Wehler's methodology (cf. G. Eley, Social history, 1976); and - as
befits so uncompromisingly Eurocentric a theory - on his analysis
of the Bismarckian—Wilhelmian economy and polity. More
directly relevant to the partition is the brief comment by P. M.
Kennedy, PP, 1972.
The battle between 'empiricist' historians and (neo-)Marxist
theoreticians of imperialism continues briskly; but usually with
more noise than casualties, thanks to the fundamental miscon-
ceptions, on both sides, indicated by E. T. Stokes, HJ, 1969;
B. Porter, JICH, 1980; and N. Etherington, A\frican] A[ffairs],
1982. However, Lenin - or at least 'Lenin misapplied' - seems
to be tottering: R. C. C. Law, JAH, 1983, writes him off as a
'rather battered Aunt Sally' and invites attention to 'Bukharin
and the Austro-Marxists'. Owen and Sutcliffe, in Studies (1972),
publish a series of seminar confrontations between empiricists and
mainly Marxist theoreticians; but the level of communication
achieved does not seem to have been as high as that of the papers
presented. However, the editors have appended an invaluable
bibliography of over 250 works - of all schools of thought -
concerned with or relevant to the theory of imperialism; and this
material has not been merely listed, but analysed and classified.
In spite of the constraints of diplomacy and strategy, and the
possible pressures of socio-economic development in the metro-
pole, most historians attach importance, for some regions of
Africa at least, to local Afro-European relations (including that
of collaboration) before and during the partition. Indeed, R.
Robinson, in Owen and Sutcliffe, Studies (1972), presents col-
laboration as a general, even crucial, 'non-European foundation
of European imperialism'. In this field a complete bibliography
would evidently include the internal history of all African regions
accessible to Europeans before the partition: it is impossible to
774

give more than samples of the very rich literature. For West
Africa, see: H. Brunschwig (1963); K. O. Dike (1956); J. D.
Hargreaves (1963); G. I. Jones (1963); and above all A. G.
Hopkins (1973). For the maritime Congo, R. Anstey (1962). For
East Africa, R. Co up land (1938 and 1939); R. Oliver and
G. Mathew, ed. (1963); C. S. Nicholls (1971); and J. S. Galbraith
(1972). For Ethiopia, S. Rubenson (1976) seems exhaustive for
external contacts from 1800 to about 1885. For the internal
development of the Egyptian crisis from 1878 to 1882, A. Scholch
(? 1972, English trans. 1981) completely supersedes the previous
historiography. On the remoter financial background of this
crisis, consult D. S. Landes (1958). The relevant literature on
southern Africa is enormous. W. M. Macmillan (1929, reprinted
1963) is a useful starting-point. For further guidance, see the very
copious references in A. Atmore and S. Marks, JICH, 1974;
reprinted in Penrose (1975).
African resistance was less the polar opposite of collaboration
than the other end of a broad spectrum of strategies adopted by
Africans for coping with Europeans. Many African groups and
leaders moved across this spectrum - sometimes successively in
opposite directions — as the partition developed. The very rich
literature is best approached through the regional chapters. There
are interesting and sometimes controversial general treatments by
T. O. Ranger, JAH, 1968, and in his contribution to Gann and
Duignan, ed. (1969); and by A. S. Kanya-Forstner, Canadian
Historical Papers, 1969. For aspects of resistance in East Africa see
(e.g.): F.-F. Muller (1959); and J. M. Lonsdale, HJ, 1977. For
West Africa, M. Crowder, ed. (1971); B. O. Oloruntimehin
(1972); and Y. Person's monumental Samori: une revolution Dyula,
3 vols. (1968, 1970, 1975).
Much work has been done on the penetration and disruption
of African by metropolitan economies, especially in South and
West Africa. For South Africa see, both as an introduction to the
literature and as a powerful statement in its own right, Atmore
and Marks, JICH, 1974 and in Penrose (1975); for West Africa,
the cogent and carefully documented exposition of A. G. Hopkins
(1973). How far did interactions, and ultimately conflicts, between
metropolitan and African economies afford motives for territorial
annexation ? How far, and in what circumstances, did metropolitan
policy-makers act from these motives? These problems have been
775

investigated (sometimes for both British and foreign imperialism

and in a global rather than a merely African context) by D. C. M.
Platt (1968) and PP, 1968; by D. K. Fieldhouse (1973); and by
W. G. Hynes (1979). Their answers are complex and defy easy
generalisation, except perhaps in the banal phrase' it all depended'.
Policy-makers were not always convinced of the real value of the
economic interests involved; and even when they were so
convinced, they often could not afford to give economic con-
siderations absolute priority over other (strategic, diplomatic,
domestic—political) factors. Attempts to demonstrate the primacy
of the economic factor often experience difficulty at the point of
articulation between the ' needs of the economy' and the action
of the policy-makers. For instance, Atmore and Marks (JICH,
1974) strive to show that the interventions of the 'imperial factor'
in nineteenth-century South Africa were usually prompted by the
'development and demands of the British economy'. But their
references to relevant statements, minutes or directives by the
metropolitan policy-makers are rather scanty.
Finally, how far were political decisions influenced by over-
weening 'Hobsonian' pressures from the capitalists? S. Marks,
discussing (JAH, 1982) the economic and other factors leading
to the South African War, emphasises rather the economically
determined 'structural constraints within which political action
takes place'; the Hobsonian capitalists are indeed still around, but
apparently in a secondary role. Detailed analyses of the relationship
of capitalists to imperialism have in fact produced complex, even
paradoxical results. In both Britain and France many of the most
powerful capitalist groups (including la haute banque in France)
were either very belated converts to imperialism or were from first
to last quite uninterested in colonial expansion: Hynes (1979);
C. M. Andrew and A. S. Kanya-Forstner, HJ, 1976; cf. L.
Abrams and D. J. Miller, HJ, 1976. Even when the capitalists were
imperialists, their pressures upon policy-makers were by no means
always successful. It all depended on 'the total situation'. If the
policy-makers saw no problems - well, why not annex? But very
often they did see problems — domestic—political, financial,
diplomatic, strategic; and these, especially the two latter, were
frequently insuperable obstacles: Hynes (1979); B. M. Ratcliffe,
JICH, 1979; B. Porter, JICH, 1980.
Especially from the late 1880s the partition became geared, ever
776

more closely but often in complex and subtle ways, to the

fluctuating alignments and antagonisms of the Powers in Europe.
G. N. Sanderson (1965) emphasises this aspect; cf. idem, HJ, 1964.
Other useful studies from this point of view are :J. D. Hargreaves,
Cambridge] Historical] J[ournal\, 1953; J. A. S. Grenville
(1964); C. J. Lowe, E[nglisb) H\istorical\ R[evieu>), 1966; R. L.
Hess in P. Gifford and W. R. Louis, ed., Britain and Germany
in Africa (1967); and, with a far wider scope than merely
diplomatic history, P. M. Kennedy (1980). For the naval balance
in the 1880s and 1890s, A. J. Marder (1940) is authoritative; there
is unfortunately nothing comparable for the 1870s.
The abortive prelude to the coastal partition in the late 1830s
and early 1840s was one aspect of a general challenge to British
Mediterranean and oceanic hegemony by France, the only other
survivor as a great power from the multilateral imperialist
conflicts of 'early modern' times. Useful introductions to this
episode are: B. Schnapper(i96i)andinRH, 1961;?!. Brunschwig
(1963); and R. M. Waller (Ph.D. thesis, 1979). G. S. Graham
(1967) is indispensable for the background of ocean strategy.
Khedive Isma'Il was the pioneer of extensive territorial empire
in Africa; and a pioneer of the ' civilising mission' as a matter of
public rather than merely private concern. His activities have been
strangely neglected, both by historians of the partition and by
theorists of imperialism. True, effective access to Egyptian
archives is not easy even for the linguistically competent; but it
is not impossible, as Hill, Scholch and others have demonstrated.
Meanwhile, M. F. Shukry (1953) and R. L. Hill (1959) are con-
venient introductions and useful guides to sources, including
those in European languages. G. Douin (1936-41) is incomplete,
terminating in 1876; but it is an unexploited mine of information.
Work on the mid-Victorian change (or 'discontinuity') in
British attitudes to empire (though not necessarily of course to
formal annexation) has until recently concentrated on the critical
examination of 'The imperialism of Free Trade': e.g., D. C. M.
Platt, Ec[onomic] H[istory] R[evieu>], 1968 and 1973. Work is only
beginning on the changing attitudes of policy-makers to the value
and significance of empire. F. Harcourt, HJ, 1980, breaks new
ground by tracing these changes as far back as 1866—68. Much
more work is needed on the 1870s. C. C. Eldridge (1973) has
useful material; W. D. Mclntyre (1967) adduces valuable com-
777

parative evidence from south-east Asia and Australasia, but is

suggestive rather than explicit. Carnarvon has been investigated
by C. F. Goodfellow (1966), but in too narrowly South African
a context; he urgently needs further and fuller attention.
For the more abrupt policy revolution in France in 1879-81 see,
for West Africa, C. W. Newbury and A. S. Kanya-Forstner,
JAH, 1969. Its military and political consequences in the Niger
basin are admirably set out in A. S. Kanya-Forstner (1969). Its
consequences in Madagascar appear, rather sketchily but quite
clearly, in J. S. Swinburne (M.Phil, thesis, 1969). Its motives are
further discussed by A. S. Kanya-Forstner in Owen and Sutcliffe,
Studies (1972). Some of the relevant metropolitan background is
filled in by Y. Conjo, RH, 1972. There are sidelights in J. P. T.
Bury, EHR, 1967; and in C. R. Ageron, R[evue] F[ranfaise
d']H[istoire d']O[utre]m[er], 1972. Here again, more work is
needed: most obviously, perhaps, on the detailed relationship
between the onset of major French expansion in Africa and the
vastly improved international position of France after the
Congress of Berlin (1878).
The coastal origins of the West African scramble, and its
subsequent development and extension down to 1889, have been
lucidly set out by J. D. Hargreaves (1963 and 1974). The
accelerating effect of British and French tariff policies on the west
coast has been demonstrated by C. W. Newbury: EcHR, 1968;
and in his contribution to Gifford and Louis, France and Britain
('970-
For the French occupation of Tunisia, J. Ganiage (1959) is
authoritative; cf. his contribution to Gifford and Louis (1971). On
the British occupation of Egypt, A. Scholch, H], 1976, is
important but not entirely convincing: in particular, he under-rates
the importance of Anglo-French rivalry in summer 1882. On this,
Robinson and Gallagher (1961) may still be consulted with profit.
But see also J. W. Parsons (Ph.D. thesis, 1977); and above all
A. Ramm in Gifford and Louis (1971).
On the interlocking Congo disputes of the early 1880s, the
abortive Anglo-Portuguese treaty of 1884, the Berlin West
African Conference and the origins of the Congo State, see:
A. Roeykens (1958); R. Anstey (1962); H. Brunschwig (1963),
and Cahiersdy etudesafricaines, 196 5; C. Coquery-Vidrovitch(i969).
J. Stengers has made very important contributions in L.e Flambeau
778

(Brussels), 1963; in Gifford and Louis, France and Britain (1971);

and in RFHOM, 1976. On the Berlin Conference itself, S. E.
Crowe (1942) is still useful. There is a more recent discussion (with
a very full bibliography) by W. R. Louis in Gifford and Louis
Later developments of King Leopold's imperialism have been

studied by P. Ceulemans (1959); L. Ranieri (1959); G. N.
Sanderson (1965); R. O. Collins (1968); J. Stengers (e.g.) in
Owen and Sutcliffe, Theories (1972); and I. R. Smith (1972).
Pending the expected major work on Leopold from the pen of
Professor Stengers, N. Ascherson (1963) is well-informed and
perceptive.
A. J. P. Taylor (1938) is no longer satisfactory as an analysis
of Bismarck's 'bid for colonies'; W. O. Aydelotte (Philadelphia,
1937), and CHJ, 1937, is however still useful for the Angra
Pequena affair. The study of Bismarck's colonialism is at present
dominated by the works of H.-U. Wehler. Important contribu-
tions from various points of view (including the official Marxism
of the DDR) have however been made by: F.-F. Miiller (1959);
H. A. Turner in Gifford and Louis, Britain and Germany (1967);
H. Washausen (1968); H. Pogge v. Strandmann, PP, 1969;
H. Stoecker (1977); and P. M. Kennedy (1980), and PP, 1972.
H. P. Merritt, HJ, 1978, usefully investigates the motives for the
most unexpected of Bismarck's colonial' bids', that in East Africa.
H.-U. Wehler applies his 'social imperialism' model to post-
Bismarckian colonialism in works dealing primarily with the
internal history of Germany. The renewal of German colonialist
pressure in the early 1890s, and the consequent development of
a very sharp Anglo-German antagonism in Africa, have been
explored by: T.A.Bayer (1955); G.N.Sanderson (1965); J.
Butler in Gifford and Louis, Britain and Germany (1967); and have
been evaluated by P. M. Kennedy (1980).
G. N. Sanderson (1965) attempts a comprehensive analysis of
international competition in the upper Nile basin from its origins
in the later 1880s to the Anglo-French settlement of March 1899.
He has presented his conclusions on Anglo-French rivalry in
Gifford and Louis, France and Britain (1971). Other important
contributions on the Upper Nile problem have been made by:
J. Stengers, Revue beige de philologie et d'histoire, 1958 and i960 —
on the Monteil mission; R. O. Collins in Gifford and Louis,
779

Britain and Germany (1967) — on the 'Nilotic' aspects of the

Anglo-German Agreement of July 1890; C. M. Andrew (1968) -
on Delcasse and the Fashoda crisis; M. Michel (1972) — on the
Marchand mission; C. M. Andrew and A. S. Kanya-Forstner,
JICH, 1974 and in Penrose (1975) - on 'Gabriel Hanotaux, the
Colonial Party and the Fashoda strategy'.
Until the completion of Professor Hargreaves' major work, the
final stages of the Anglo-French scramble in West Africa lack any
comparably thorough analysis. There are however valuable
contributions on various aspects by: M. Perham (1956); J. E.
Flint (i960); J. A. S. Grenville (1964); A. S. Kanya-Forstner
(1969), HJ, 1969, and his contribution to Gifford and Louis, France
and Britain (1971); J. Ganiage, RH, 1975; C. Hirshfield (1979).
Work on the French colonialist movement, pioneered by
H. Brunschwig (i960), has been intensively pursued by C. M.
Andrew, A. S. Kanya-Forstner and others. See: C. M. Andrew
and A. S. Kanya-Forstner, JICH, 1974 and in Penrose (1975); the
same, HJ, 1971, 1974 and 1976; C. M. Andrew, P. Grupp and
A. S. Kanya-Forstner, RFHOM, 1975; C. M. Andrew, Transac-
tions of the Royal Historical Society, 1976; S. M. Persell, HJ, 1974;
L. Abrams and D. J. Miller, HJ, 1976. There is a major study of
Gabriel Hanotaux by P. Grupp (1972); and of Eugene Etienne
by H. Sieberg (1968).
There is nothing remotely comparable for Italian colonialism.
Indeed, there appears to be no serious work of synthesis, or even
of orientation, on Italian imperialism during the period of
partition. For frustrated Italian aspirations in Tunisia, see
J. Ganiage (1959). For Somalia, see R. L. Hess (1966). For the
Ethiopian adventure, the most useful books apear to be: C. Conti
Rossini (1935); C. Giglio (1958-72); and above all R. Rubenson
(1964 and 1976). For Italian aspirations in the Nile valley,
M. P. Hornik (? 1939) is still useful; but see, faute de mieux,
G. N. Sanderson (1965) and HJ, 1964.
Considerable attention has been devoted to the roles of
Ethiopia during the partition: resistance to Italy, Shoan territorial
expansion, relations with the Mahdist state and intervention in the
contest for the Upper Nile. See: the two works by S. Rubenson
cited above; R. A. Caulk (Ph.D. thesis, 1966), and Transafrican
Journal of History, 1971; Z. Gabre Sellassie (1975); H. G. Marcus
(1975). For Ethiopia and the Nile valley, see G. N. Sanderson
780

(A<)6)),JAH, 1964 and Sudan Notes and Records, 1969. For Russian
activities in Ethiopia, C. Zaghi (1973) supersedes the rather
sketchy treatment by C. Jesman (1958).
Serious work has only just begun on the revival of Portuguese
imperialism in the late nineteenth century. See M. D. D. Newitt
(1981); W. G. Clarence-Smith, Journal of Southern African Studies,
1979. P. R. Warhurst (1962) and E. Axelson (1967) examine the
fate of Portugal in the diplomacy of partition.
The historiography of the South African scramble, and of the
origins of the South African war, still lacks a fully adequate
biography of either Milner or Rhodes; however, for Rhodes see
J. E. Flint (1976). The early study of the 'imperial factor' by
C. W. de Kiewiet (1937) has been supplemented by analyses
mainly of the political and diplomatic factors by: J. S. Marais
(1961); G. H. L. Le May (1965); L. M. Thompson, in M. Wilson
and L. M. Thompson, ed., 11 (1971); D. M. Schreuder (1969 and
1980); A. N. Porter (1980) and in JICH, 1972. However, much
recent work has concentrated on the political implications of
economic development, often insisting upon the crucial impor-
tance of economic interests and pressures and indeed of the
developmental needs of the British metropolitan economy.
Examples are: G. Blainey, EcHR, 1965; A. H. Jeeves, Canadian
historical papers, 1973; D. Denoon, HJ, 1980; J. J. Van-Helten,
JAH, 1982; and - challenging the primacy of economic factors
— R. V. Kubicek (1979). There are useful surveys and critiques of
recent writing in: A. Atmore and S. Marks, JICH, 1974 and in
Penrose (1975); N. Etherington, A A, 1982; and S. Marks, JAH,
1982.
Work is still rather patchy and impressionistic on the relation-
ship between the scramble, imperialist ideology and changes in
the 'climate of opinion'. B. Semmel (i960) and M. D. Bidiss, HJ,
1972, discuss the development and influence of' scientific' Social-
Darwinist political ideas; but the most useful exposition of the
overtly imperialist and racialist content of Social Darwinism is
perhaps still that of W. L. Langer (2nd edn, 1951). S. Koss, HJ,
1975, reveals the existence of an unexpected (Wesleyan) imperialist
influence upon the English petty bourgeoisie. There seems to have
been no serious exploration of the influence, especially upon
impressionable adolescents, of such authors as G. A. Henty and
H. Rider Haggard in Britain or Karl May in Germany; or of the
781

Boy's Own Paper, which commenced publication in 1879.

A. N. Porter (1980) traces the efforts of imperialist politicians to
manipulate British' public opinion' on the South African question.
B. Porter (1968) demonstrates the ineffectiveness of Liberal-radical
'critics of empire' in stemming the tide of British imperialist
sentiment in the later 1890s; but R. Price (1972) effectively
deflates, so far as Britain is concerned, the myth that imperialism
was the specific creed of the 'nationalist masses'. For Germany,
H. C. Schroder (1968) examines the anti-imperialist opposition of
the SPD, and traces the decline of this opposition after c. 1900.
Nationalistic xenophobia in Europe could be both a cause and
an effect of imperial rivalry. H. Brunschwig, in Gifford and Louis,
France and Britain (1971), traces the strengthening of French
colonialist sentiment by the popularisation, after c. 1880, of the
traditional combativeness to Britain overseas of the Marine and
the Colonies. P. M. Kennedy (1980) analyses perceptively the
contribution of Anglo-German friction in Africa to the general
' antagonism' between these two powers.
3. NORTH AFRICA, 1870-1905
Historical writings concerning North Africa are mainly the work

of French historians, as might be expected. Recently Tunisians
and Algerians, also writing in French, have been taking up the
subject. But most of these studies deal with the origins and growth
of the nationalist movement, in a sort of hagiography. As for
works in English, they mostly confine themselves to repeating or
adding to published French works.
Algeria. For the 1871-1914 period, there is a large body of
work, much of it polemical. Andre Julien made his name with
his Histoire de I'Afrique du Nord, published in 1931. The third part
of this (roughly 1830-1925), which had not previously been
published, was partially incorporated in VAlge'rie contemporaine, a
work hostile to the armed forces and the colonists. In their theses
Xavier Yacono and Andre Nouschi concentrated on the diffi-
culties experienced by the colonising forces, and on the standard
of living of the native population. Robert Ageron is more
interested in political questions, in Les Algeriens musulmans et la
France and, in particular, in I'Histoire de I'Alge'rie contemporaine,
volume 11 of which is his work.
782

4/5 WESTERN AFRICA, 1870-1905
Historical studies on the two protectorates of Tunisia and
Morocco are virtually as numerous, but generally calmer. Pierre
Guillen made his name by a fine thesis, UAllemagne et le Maroc,
but has had no disciples in this field, at least for the period prior
to 1900, for the large volume of work in progress deals with inter-
national rivalries or with the early years of the protectorate. On
the other hand, the entire history of the regency of Tunis from the
beginning of the eighteenth century to the middle of the twentieth
has now been covered. The value and the quality of these
productions naturally varies, but in them can be found the details
of the Tunisian archives of Dar el Bey in both Arabic and French.
This movement was initiated by two theses, that of Jean
Ganiage on the Origines du protectorate and that of Andre Martel
on the Tunisian South. Since then, the former has interested
himself in international relations between 1871 and 1914, par-
ticularly problems relating to the Mediterranean, and also in his
other specialisation, historical demography. The latter, now
teaching at the University of Montpellier, has made his name in
military history. At Tunis, in addition to the many publications
relative to the twentieth century, we should mention the studies
of Mohammed Cherif on the eighteenth and nineteenth centuries
(the problem of the state) and those of Khelifa Chateur on the
regency after 1815.
4 AND 5. WESTERN AFRICA, 1870-1905

The richest sources for the study of West African history in this
period are to be found in the archives of European governments,
missionary societies and commercial concerns which were intensi-
fying their activities in the region. Certainly, the bias of such
sources may incline the unwary student towards an imperial
perspective; but those capable of formulating their own questions
and applying the principles of historical criticism will find
evidence of a richness unparalleled in earlier periods. Formal
despatches by governors may indeed be unrewarding, but their
apparently obscure enclosures, for example, frequently contain
vital clues.
The most satisfactory guides to such sources in European
countries are usually in the series oi Guides to the sources of the history
of Africa south of the Sahara, published by the International Council
783

on Archives; the two volumes for France (Zug, 1971, 1976) are
particularly important for this period. Unfortunately no such
volume has appeared for the UK, but N. Matthews and M. D.
Wainwright, Guide to manuscripts and documents in the British Isles
relating to Africa (London, 1971), used in conjunction with Public
Record Office Handbooks, nos. j and 13, dealing with the records
of the Colonial and Foreign Offices respectively, may partly fill
the gap. Information may also be found in the series of Guides to
materials for West African history in European archives published by
the Athlone Press; but where a volume has been published by the
International Council on Archives, it is to be preferred. For
business records, see also A. G. Hopkins (1976). Guides to
archival sources in Africa are less readily available; regrettably,
few independent governments have been able to give high
priority to the staffing and development of public record offices,
and researchers may encounter difficulties in using these rich
resources. Senegal is one exception; researchers into the history
of any of the territories of French West Africa (AOF) should
consult the fascicules oi the Repertoire des archives edited by C. Faure
and J. Charpy (Rufisque, 1955,ff).
Besides the documentary evidence available in European
languages, there are exceptional opportunities - and exceptional
dangers — for the use of African oral traditions. During this period
many of these were recorded, and sometimes published, both by
historically minded administrators like Richmond Palmer and
Maurice Delafosse and by African scholars like the Reverend
Samuel Johnson. No general survey of such materials - and of
the formidable critical problems involved - has been attempted.
In recent years academics have in some areas made more systematic
attempts to record and interpret oral sources (cf. D. Henige, Oral
historiography, 1982).
Some early administrators interested themselves, for specific
purposes, in the recent history of Muslim communities; mention
may be made of the studies carried out by Paul Marty for the
various colonies of AOF. A vastly more important source is of
course the considerable body of theological, literary and adminis-
trative writing, in Arabic or in African languages, produced
within Islamic states or communities; problems relating such
material to other forms of evidence are discussed in the first
chapter of David Robinson's forthcoming study of al-Hajj 'Umar.
784

4/5 WESTERN AFRICA, 1 8 7 0 - 1 9 0 5
The historical potential of contemporary printed sources was
by no means exhausted by earlier writers of colonialist history.
The books published, even with tendentious purposes, by
travellers, administrators, soldiers, traders and missionaries such
as Binger, Flegel, Gallieni, Kingsley, Peroz, Ramseyer and Kuhne
can still prove revealing, about the peoples among whom they
worked as well as about their own purposes. And there is a
substantial body of printed testimony by Africans. Authors like
Blyden, Horton and Otonba Payne are by now well known; but
there is still much to be learned from the columns of contemporary
African newspapers.
Secondary studies
During the 1950s and 1960s research on this period tended to

concentrate on studies of African responses to the encroachment
of European power and to the early structures of colonial rule.
Often the emphasis was directly on examples of armed resistance;
there are some good essays of this type in Crowder (1971) and
in Rotberg and Mazrui (1970). Many important studies, solidly
based in archives of the colonial administration but supplementing
them in varying degrees by use of oral traditions and reports from
the African press, originated in the University of Ibadan: examples
include Anene (1966); Aderibigbe (1959); Ikime (1969); Afigbo
(1972); Akintoye (1971); Atanda (1973); and Asiwaju (1976).
Adeleye (1971); Balogun (1971); Abubakar (1977); and Fika
(1978) treat similar themes with the use of Arabic material from
northern Nigeria.
For Ghana, comparable themes are treated in Agbodeka (1971),
Braimah and Goody (1967), Staniland (1975) and Amenumey
(1966). Here however the major research effort has concerned the
internal development of the Asante state; see Wilks (1975), Arhin
(1967) and Lewin (1978). Arhin (1979) treats its economic
relations. Tordoff (1965), though not of this school, is still useful.
In Sierra Leone Ijagbemi (1968) has worked on the Temne,
Abraham (1978) on the Mende, each with somewhat different
emphases. Fyfe (1962) remains the basic work on the Sierra Leone
Colony, but may be supplemented by Porter (1963), Spitzer (1974)
and Harrell-Bond et al. (1978). Mahoney (1963) is the best work
on the Bathurst community.
785

The nineteenth-century history of colonial Senegal is being

studied by Roger Pasquier, who has published some useful articles
(i960, 1962); its political history is dealt with by Idowu (1966)
and later by G. W. Johnson (1971). Relationships between
Europeans and Africans in Senegambia are treated by Ganier
(1965), D. Robinson (1975), Charles (1977), Bathily (1975), Klein
(1968) and Quinn (1972). Earlier studies of the Tukolor empire
created by al-Hajj 'Umar, such as Saint-Martin (1967) and
Oloruntimehin (1972), will be largely superseded by the forth-
coming book of David Robinson. Stewart (1973) and Desire-
Vuillemin (1962) deal with Mauritania.
The historiography of the western Sudan is dominated by the
late Yves Person, and in particular by his vast and remarkably
wide-ranging study of Samory (1968-75); Amselle (1977) is a
valuable monograph relating to the same region. Diallo (1972) and
McGowan (1981) deal with Futa Jalon. French relations with the
Baule are well treated by Weiskel (1980); with Dahomey by Ross
(1967) and Garcia (1976).
There has been a great deal of writing on theoretical questions
concerning imperialism, and specifically many interesting studies
of the French colonial party; but these are reviewed here only in
so far as the authors have attempted to study the initiatives and
reactions of Africans during the period of partition. The important
work of Robinson and Gallagher (1961, 1981) attempts this at a
general level; there is closer focus on West Africa in Crowder
(1968), and in some of the contributions to Gann and Duignan
(1969-75); and to Gifford and Louis (1967, 1971, 1984). More
specialised studies which exemplify this approach include Flint
(i960); Atger (1962); Kanya-Forstner (1969); Obichere (1971);
Uzoigwe (1974); and Hargreaves (1963 and 1974); also some of
the essays in Stoecker (1960-8). Newbury's documentary collec-
tion (1971) is a valuable resource. Studies of African warfare
include Crowder (1971), Robert Smith (1976) and Smaldone
(*977)-
The major contribution to the economic history of the period
made by Hopkins (1973) has still to be fully assimilated; some
distinguished historians remain resistant to his work (cf. Hopkins,
1968, 1972; Ajayi and Austen, 1972). Baier (1980) is an important
study of the effects of colonial control on the 'desert-side
economy'. Important studies of Dahomey are Manning (1982)
786

4/5 WESTERN AFRICA, 1 8 7 0 - I 9 0 5
and Coquery-Vidrovich in Meillassoux (1971). On the growth of
the cocoa industry Berry (1975) has applied to western Nigeria,
with considerable historical depth, the approach which Hill (1956,
1963) pioneered on the Gold Coast. A good study of the relatively
short-lived rubber trade has been made by Dummett (1971). Jeng
(1978) and Swindell (1980) deal with aspects of Senegambian
groundnut production. Political aspects of cotton-growing are
treated by Nworah (1971) and M. Johnson (1974). Recent studies
of European mercantile organisation are surveyed by Hopkins
(1976).
The most comprehensively successful attempt to move from
economic history to a study of the social consequences of
colonialism has been made by the French Marxist, Suret-Canale
(1964). Hausen (1971) and Wirz (1973) are important for Kamerun.
Isichei on the Igbo (1976) has good material on social changes.
The evolution during this period of African systems of' domestic
slavery' is emerging as a topic of great importance; see Miers and
Kopytoff (1977), and for specific examples Bouche (1968), Renault
(1971) and Grace (1975). Hopkins (1966) is also relevant. Western
education in French West Africa is fully treated by Bouche (1975);
there is nothing comparable for British Africa in this period.
The fortunes of Islam under colonial rule have been most
thoroughly studied with reference to Senegal; see Monteil (1966)
and O'Brien (1971), though the latter's emphasis is on later
periods. Gbadamosi (1978) is important for Nigeria. Important
work on the influence of Christian missions has also been done
in Nigeria, beginning with Ajayi (1965) and Ayandele (1966). In
the latter's biography of James Johnson (1970) this led on to study
of the ideas and political activities of members of the educated
African elite, which were very popular in the 1960s. Perhaps the
mostdurableworkofthisperiodisKimble(i963).G. W. Johnson
(1971), Curtin (1972), Spitzer (1974) and Cole (1975) set such men
in broader social contexts. Pallinder-Law (1973) is interesting on
political experiments in Abeokuta. Higher priority in the study
of African political history may now be given to localised study
of political and social conflicts in rural areas, and of the connections
between these and the activities of the educated elite, such as
Simensen (1975).
787

6. WESTERN EQUATORIAL AFRICA
French Congo and Gabon, 1886—190/

Most of the archival documents are located in Paris, Section
Outre-Mer (S.O.M.) of the Archives Nationales. The Brazza
material, inherited from the family's private papers, includes a
series of fundamental dossiers on the Mission de l'Ouest-africain
(M.O.A.), the Commissariat General (1886—97), and the mission
of 1905. The concessions material comprises about a hundred files
on the companies. All this is complemented by the Archives de
la Marine (Serie BB.4) and by the material deposited in the S.O.M.
at Aix-en-Provence (the political reports of the different colonies,
and the reports of the Inspecteurs Generaux).
Studies of the region and the period are both recent and
penetrating; cf. in particular Gilles Sautter, De /'At/antique aufleuve
Congo: une geographie du sous-peuplement, Kepublique Gabonaise,
Kepublique du Congo, Paris, 1966, which includes excellent historical
accounts. On the period in general, these are the works of
C. Coquery-Vidrovitch, Bravga et la prise de possession du Congo: la
mission de rOuest-africain, I88J-J (1969) (documents and commen-
taries), and Le Congo (AEF) au temps des grandes compagnies
concessionaires, 1898-19)0 (1972). In these works will be found the
bibliographical references to numerous accounts of the period. To
these may be added a series of good monographs such as: Marc
Michel, La mission Marchand, 189/—1899 (1972); Georges Mazenot,
La Likouala-Mossaka, 1878-1920 (1971); A. de Mazieres, Liotard
et le haut-Oubangui, these de 3e cycle (Universite Paris I, Ecole des
Hautes Etudes, 1975). Finally, on the Bateke people, the
remarkable ethno-historical study of J. Vansina: The Tio kingdom
of the Middle Congo, 1880-1892 (1973).
King Leopold's Congo, 1886-1908

King Leopold will not be remembered gratefully by students of
history. On the eve of the annexation of the Congo State by
Belgium, he was heard to say: 'They can have my Congo, but
they have no right to know what I have done there.' This was
no mere jest: it led to the systematic destruction of the state
archives. Only some fragments have survived (listed in Van
788

6 WESTERN EQUATORIAL AFRICA
Grieken-Taverniers, 1981); they include the papers relating to

foreign affairs (Van Grieken-Taverniers, 1955). Ironically enough.
Leopold was mostly defeated by himself: he took no precautions
to have his own private papers destroyed, and these are now
available in the Royal Archives (Vandewoude, 1965). Other
revealing insights are provided by the papers of high officials or
advisers (many of which are preserved in the Belgian State
Archives or at the 'Musee Royal de l'Afrique Centrale' at
Tervuren), their correspondence (e.g. LeopoldII et Beernaert, 1920)
or their memoirs (for Banning, see Banning, 1927, and Stengers,
i95 5)-
Some older biographies of the king are still useful (de
Lichtervelde, 1926; Daye, 1934). Recent research is used in the
latest one, by Barbara Emerson (1979), as well as in the general
works on Leopold's Congo (Slade, 1962; Gann and Duignan,
1979). The king's special kind of imperialism was analysed by
Stengers (1972).
The central administration of the State is described in the first
part of Lutumba's strangely named Histoire du Zaire (1972).
Disentangling the finances of the state and the financial affairs of
the king, which are extraordinarily mixed up, requires a kind of
detective work, at which Stengers has made several attempts
(1957, 1980). The most general survey of foreign policy is that
of van Zuylen (1959). Foreign archives have been used to study
the attitudes of Britain (Cookey, 1968), Germany (Willequet,
1962) and Italy (Ranieiri, 1959). The push towards the Nile can
be followed in the works of Stengers (1969),- Sanderson (1965) and
Collins (1962 and 1968).
A vast amount of material on the European side of the work
of the great companies is assembled in Waltz (1917). Useful special
studies include Cornet (1947) and Katzenellenbogen (1973).
For the development of the Congo campaign in Britain, apart
from Cookey's book, one may refer to the most recent biography
of Morel by C. Cline (1980), and to Louis and Stengers (1968).
In Belgium the indictment by Cattier (1906) remains a classic
which is still worth reading.
The steps towards the annexation to Belgium have been studied
by Stenmans (1949) and Stengers (1963).
789

Stanley left us the expression 'Congo Free State' as a translation
of Etat independent du Congo, an early example of' newspeak' that
still haunts us today and is the worst feature of his The Congo and
the founding of its Free State (1885). This is the first of a series of
early histories culminating in the works of Wauters (1911),
Masoin (1912/1913) and Morel (Louis and Stengers, 1968). After
its demise, interest picked up only after 1945 when J. Cornet was
the quasi-official historian of Belgian colonialism, while R. Slade
ushered in the era of decolonisation in 1962. Her work, outdated
as it is, was not superseded by L. H. Gann and P. Duignan (1979).
Today there exists neither a satisfactory introduction nor a solid
interpretation of this period in Zaire's history.
Scholars are so impressed by the continuities between this
regime and its successor that they tend to include the Congo State
period in studies devoted to 'the colonial period'. Jewsiewicki
(1972) views 1891 and 1920 as major turning points rather than
1879, 1885 or 1908, and most others follow him. As a result little
research is currently devoted specifically to this period. The vogue
for diplomatic history has passed as the most relevant archives
have been combed, while the study of institutions (the military,
the judiciary, the administrations, the missions and the major
companies) are spurned, even though they need adequate study.
Research today concentrates on regional history (Cornet, Martens,
Mumbanza, Ndua Solol, Bustin) or on specific topics such as
education (Yates) or a mission (Storme), or on some train of
striking events (Bimanyuj. The pace of research remains slow,
and major themes such as the conquest itself, economic exploita-
tion including atrocities, financial history, the evolution of
societies and cultures in the country all still await their
historians.
Yet this is not for a lack of sources! Abundant public archives
in Belgium are accessible and were described by M. Van Grieken-
Tavernier's 'archives' in Livre blanc (Academie des Sciences
d'Outre-mer, Brussels, 1962). Only the Belgian Royal Archive is
not open to one and all. In Zaire the archives have been described
by B. Jewsiewicki in Etude analytique des archives administratives
Zai'roises (Lubumbashi, 1973 (stencil)) and are summarized in
Vellut (1974: 107-5 0- Missions and companies usually have kept
some papers, and those of the major churches are well kept.
Private papers are harder to find as they are dispersed, even though
790

7/8 SOUTHERN AND CENTRAL A F R I C A , 1867-I9IO
the Musee Royal de l'Afrique Centrale has acquired important
lots and attempts to centralise more.
Moreover the amount of publication by contemporaries is quite
extraordinary, exceeding by far what was usual in other colonies.
The bibliographies of Wauters (1895), Simar (1912) and Vellut
(1974) are precious in this regard even though they do not list
all of them, especially after 1895. The Congo State attracted agents
from many nationalities as well as a throng of adventurous
travellers. Their reminiscences and accounts are especially valuable
when the authors belonged to less exalted echelons or wrote for
a simple public in Europe. The less ambitious the account, the
better. It then tended to be much more concrete. Such sources
can be found in practically any European language — even
Hungarian, but apparently not in Finnish, although some Finns
served in the Congo. Jespersen (Hulstaert) and Armani are listed
as examples in the bibliography. Both were agents for the state.
The former wrote in Danish and documents the administrative
occupation of a portion of the equatorial province where he is well
remembered in tradition. The second one, an Italian, still remains
our main source about southern Manyema after the Arabs,
especially from 1904 to 1906. The full wealth of these sources is
still unknown to most researchers. Nor have the myriad local
ethnographies of the time been studied as sources for social
history, although they are our main founts of information towards
this end (de Calonne Beaufaict).
Oral reminiscences were first gathered in the 1930s, especially
in the equatorial region, and can still be gathered today for the
closing years of the period, although almost all such accounts are
by now hearsay related by children or grandchildren of the
participants. Yet in conjunction with written sources, whether
published or not, these are essential elements for an all-round
understanding of the period (Hulstaert). Recent research (Mum-
banza, 1980) has shown that such data can still be gathered and
considerably enrich other records.
7 AND 8. SOUTHERN AND CENTRAL AFRICA, 1867-19IO

South of the 'Limpopo
South Africa is an intensely divided society, and its historiography
inevitably reflects its divisions. The period between 1870 and 1910
79 *

in particular is covered by a large, growing and controversial

literature, and it is impossible in an essay of this scope to do more
than allude to some of the older preoccupations, and outline very
briefly the more recent interpretations and areas of debate. The
published and unpublished primary source material for this period
is almost embarrassingly abundant. Enterprising researchers have
found evidence in archives and libraries as far afield as Poland and
the American south, in languages as different as Hindi, Yiddish
and Zulu. The more mundane sources include a variety of
government publications, statistical material, departmental re-
ports, commission and select committee evidence and reports,
parliamentary debates and legislation, the archives of the Transvaal
Chamber of Mines and individual mining houses, land registries
and business records, mission archives and collections of private
papers, political ephemera and trade union collections, and
colonial and African newspapers. Many major collections of
private papers (e.g. Smuts, Milner, Merriman, Fitzpatrick,
Phillips) have been published for this period. In addition, several
South African universities and the South African Institute of Race
Relations are now establishing oral archives. For a listing of the
official sources, see the bibliography in the Oxford history of South
Africa (ed. Wilson and Thompson), vol. n. There are a variety
of bibliographic guides to published works and theses available.
Much of the recent debate in South African historiography has
centred on the mineral revolution, and it is in their interpretation
of its meaning and effects that recent historians have differed most
from that of their predecessors, such as Van der Horst or Hobart
Houghton. Articles by Legassick and Trapido in the early 1970s
established the significance of South Africa's industrial revolution
in this period. During it the modern South African state was
created and the preconditions for capital accumulation were
established and subsequent class structures shaped. Whereas much
of the earlier, liberal literature regarded contemporary racist
practices in the Republic as an atavistic hang-over from the
seventeenth- and eighteenth-century 'frontier tradition', recent
work has stressed the way in which racist practices accompanied
the capitalist development of South Africa in these years and were
intrinsically related to it.
The interaction of pre-industrial and industrial societies has
been dealt with in a number of doctoral theses many of which are
79 2

7/8 SOUTHERN A N D CENTRAL A F R I C A , 1867-I9IO
now beginning to appear in print. Although De Kiewiet's
pioneering work, The imperial factor in South Africa (1937) remains
indispensable for the 1870s and 1880s, the new research explores
in far greater detail the impact of the mineral discoveries and the
scramble for southern Africa on African societies. Studies by
Beinart (Mpondo), Bonner (Swazi), Cobbing (Ndebele), Delius
(Pedi), Guy (Zulu), Harries (Tsonga), Kimble and Burman
(Sotho), Parsons (Ngwato), Shillington (southern Tswana), are all
concerned both with the nature of pre-industrial African societies
and their social, economic and political articulation with colon-
ialism and capitalism. The essays in Palmer and Parsons have a
similar focus and deal with both central and southern Africa.
There is no single work on the wars of conquest in South Africa
in the 1870s and 1880s, although they are touched on in all the
above, and Saunders and Atmore (in Rotberg and Mazrui) deal
more specifically with the conquest of the Xhosa and Sotho
respectively. The rise and fall of the African peasantry more
generally is the subject of work by Bundy, while white—black
agrarian social relations are dealt with in a series of major articles
on the Transvaal by Trapido, in work by Delius and in theses by
Keegan on the Orange Free State and Morris on the capitalisation
of South African agriculture more generally.
Most of this work also throws light upon the origins of migrant
labour and the nexus of contemporary laws controlling African
labour. A pioneering article by Wolpe (1972), suggesting the
origins of the system in capital's need for a cheap labour supply
whose welfare costs could be borne in the reserves, has been
modified by these detailed case studies, which show the system
emerging out of the complex struggles between the mine mag-
nates, colonial authorities, the traditional ruling class and African
homesteads over the labour power of young men. The essays in
Marks and Atmore and in Marks and Rathbone contain a useful
introduction to these debates. The significance of the early days
on the diamond fields for the evolution of migrant labour, the
compound system and institutionalised racism is well argued in
R. V. Turrell's Ph.D. thesis on Kimberley, while the early gold-
mining industry has received even more attention. Starting with
the publication of F. R. Johnstone's Class, Race and Gold, the
significance of the gold mines has if anything overshadowed the
literature on the earlier period. Further monographs include
793

N. Levy, P. Richardson, J. J. Van-Helten, R. V. Kubicek and

D. Yudelman. The social history of the period has been brilliantly
handled by C. Van Onselen in his two-volume collection of essays
on the social and economic history of the Witwatersrand. Very
different in approach but of value is the two-volume composite
Afrikaners in die Goudstad (E. L. P. Stals et a/.), produced by
Afrikaner historians.
White politics and the' native policy' developed by the colonies
and republics have been dealt with by Trapido, McCracken and
Davenport for the Cape; Welsh, Marks and Guy for Natal and
Zululand; Uys, Trapido, Gordon and Delius for the South
African Republic and Keegan for the Orange Free State. There
is also a considerable literature in Afrikaans, some of it published
in the Archives yearbook for South African history. A useful intro-
duction to the politics of the period is through the many bio-
graphies of imperial and colonial actors. The literature on Rhodes
is particularly voluminous, but Flint provides the most useful
starting point. Lewsen's Merriman and older biographies of
Schreiner and De Villiers are valuable for Cape liberalism. There
are biographies of leading Boer figures such as Burghers, Kruger,
Stein and Botha in Afrikaans.
Although the ' scramble' for southern Africa is touched on in
a number of general works on the partition of Africa, or in
detailed case-studies for parts of the region, the only recent
synthesis on the 'scramble' in South Africa and Rhodesia, seen
largely in terms of the politics of partition, is Schreuder's The
scramble for Southern Africa, 1877-189J. This also contains a
valuable bibliographical essay on the period covered. Unfortu-
nately this work ends effectively before fully considering the
South African war of 1899—1902. This war and the events which
preceded it have been a matter of controversy ever since the
publication of J. A. Hobson's The South African war (1900).
Recent interpretations have moved away from the rather
personalised accounts of causation in much of the early literature.
Some of the literature of the war is reviewed critically in Marks
and Trapido (1979) and Marks (1981). Much of the debate has
centred on the relationship of imperial economic and strategic
interests, and the needs of mining industry. Important contribu-
tions have been made by Jeeves, Mendelsohn, Spies, Duminy and
Denoon. The most readable recent narrative history of the war
794

7/8 SOUTHERN AND CENTRAL AFRICA, 1867-I9IO
is Pakenham's The Boer war, although the collection edited by
Warwick is in general a more scholarly introduction to the
substantive issues. The participation of blacks in this ostensibly
'white man's war' is dealt with by Warwick and Nasson,
Comaroff and Willan.
The reconstruction period has also been subjected to scrutiny
since Denoon's A grand illusion. Again, it has been the relationship
of the Milner administration to the mine magnates and more
generally of the relationship of the war and reconstruction to the
capitalist development of South Africa which has been to the fore
in the debate (see Denoon, Mawby, Marks and Trapido, Legassick,
Johnstone, Yudelman). For the unification of South Africa,
Thompson remains the standard work, though the revisionist
historiography stresses the economic underpinning of the con-
stitutional changes and questions the notion that unification
marked an imperial defeat. Chanock's Unconsummated union has a
stimulating introductory section on the nature of the Union that
was formed in 1910, and British imperial interests in the region.
The white working class has received rather more attention
than the black in this period, but both are the subject of a growing
literature (see the introductory chapters of Davies, Yudelman,
Burke and Richardson, and the work by Katz). Bozzoli has also
dealt with the ideology of English-speaking industrialists in this
period. For black workers see the collections produced by Bozzoli
and Webster out of the history workshops held by the University
of Witwatersrand, and by Marks and Rathbone and by Warwick.
To understand the formation of the African elite it is necessary
to turn in the first instance to the work of Christian missions. The
impact of Christianity in Natal—Zululand is dealt with directly by
Etherington for the early part of this period and by Unterhalter
for Nqutu district, Zululand until 1906; Hunt Davis, Saunders,
Delius, Parsons, Dachs, Sundkler all make valuable contributions
to our understanding of its very varied manifestations. As a class,
the African petty bourgeoisie has been most sympathetically
handled by B. Willan in his ' The life and times of Sol T. Plaatje',
a work which stands with Shepperson and Price in the writing
of African biography.
795

North of the Limpopo

This is a period in which primary written material becomes much
richer also in the case of the territories north of the Limpopo.
From the 1870s published travel literature, and the letters and
diaries of missionaries and traders, increase in number and value,
while from the 1880s and 1890s, these can be supplemented by
Foreign Office, British South Africa Company and local admin-
istrative records. A number of settler accounts of early Rhodesia
have recently been republished.
For the region as a whole, the two-volume History of Central
Africa, edited by D. Birmingham and P. Martin, contains chapters
by acknowledged experts on the region and synthesises much of
the recent research. It still needs to be supplemented by earlier
outline histories and the essays in composite volumes, such as
Stokes and Brown, and Pachai (ed.), Early history of Malawi.
Although much of his work is now dated, the first archivally based
histories by Gann on the Rhodesias still contain useful material;
his The birth of a plural society remains an outstanding account of
the period in Zambia. Hanna and Oliver provide a solid basis for
an understanding of the establishment of British rule in central
Africa. Barnes (1954) was one of the first to look at the history
of an African people and the impact of colonial rule, using both
documentary and oral sources. In the 1960s his lead was followed,
especially in Zambia in a series of regional ethnic histories based
on oral and documentary evidence.
At the heart of the Zambian and Malawian experience in this
period was the extension of the slave and ivory trade, the advent
of missionaries and traders from the east coast, the penetration
of the region from the south, its annexation by Crown and
Company and the imposition of alien rule. The relevant chapters
in Roberts's succinct outline History of Zambia are the most useful
starting point. There is no equivalent for Malawi, though the
introductory pages of McCracken's Politics and Christianity in
Malawi contain a useful synthesis of late-nineteenth-century
Malawian history before it deals in detail with the effects on
Malawi of the Scottish missionaries. In general, historians have
perhaps dealt with the impact of missionaries on Malawi more
exhaustively than any other single topic (Ross, Stuart, the
Lindens, Tangri). The chapter by Vail in Birmingham and Martin,
796

7/8 SOUTHERN AND CENTRAL AFRICA, 1867-19IO
and his articles, deal graphically with the impact of colonial and
company rule on African peoples, and trace the repercussions of
early land and labour policies on later developments in Malawi
and eastern Zambia. This work is valuable also in questioning the
focus on the 'tribe' as the natural unit of analysis. Vaughan's
recent doctoral thesis examines social and economic developments
at a regional level, both for the immediate pre-colonial, slave-
trading period and subsequently. There is a considerable amount
of regional monograph material, dealing with individual ethnic
groups: thus, the Lozi (Caplan, Mainga, Prins, Van Horn,
Clarence-Smith), the Bemba (Roberts), the Ngoni (Barnes, Read,
Rau), the Luvale (White, Cunnison), the Tonga (Colson, Mat-
thews) and the Ngonde (Kalinga) have all received considerable
attention. All these works deal with either part or the whole of
this period as a section of wider studies, straddling an earlier or
a later period (or both).
The much greater white presence in Southern Rhodesia has
given its recent historiography an affinity with that of South
Africa, and there are rather fewer detailed studies of African
peoples built on oral sources. Nevertheless, the researches of
Cobbing, Beach and Bhebe provide a solid foundation for the
study of the Ndebele and Shona polities on the eve of the colonial
occupation, and for their early responses to colonialism. The
actual nature of the British South Africa Company and the early
years of its rule are handled in masterly fashion by Galbraith in
Crown and charter. Ranger's Kevoltin Southern Rhodesia also contains
an invaluable account of early colonial rule, although his stimu-
lating analysis of the actual Shona—Ndebele uprising is contested
and should be read in conjunction with the articles by Cobbing
and Beach in the Journal of African History. Phimister's economic
history of Zimbabwe (in Birmingham and Martin, and a forth-
coming general synthesis which looks in considerable detail at
this period) moves beyond his early work on gold-mining and on
Rhodes, Rhodesia and the Rand. The African peasant response
to colonialism and land policies have been thoroughly studied by
Palmer and should be supplemented with the introductory
chapters of Machingaidze on white farming. African mine labour
and conditions in the mining industry form the subject of Van
Onselen's Chibaro, which marked a watershed more generally in
African labour history.
797

South West Africa

For South West Africa, as is the case north of the Limpopo, this
is a period which sees great increase in the amount of primary
material available both in print and in German colonial archives
in Potsdam, East Germany, as well as in German and Finnish
mission headquarters. The region remains neglected historio-
graphically, although there are signs of an awakening interest in
the past as it approaches independence from South Africa.
The most exhaustively covered aspect of South West African
history is probably its annexation by Germany. The useful
bibliographical essay on Germany's participation in the' scramble'
in Gifford and Louis (1967) should be supplemented with more
recent articles by Hans-Ulrich Wehler, Paul Kennedy and Pogge
von Strandmann. Recent interpretations are incorporated in
Schreuder (1980). Vedder's South West Africa in early times deals
with the first half of this period, but needs to be used critically,
as do the accounts by Goldblatt and Esterhuyse, which do not
go beyond rather poor narrative. In some sense, they are all
concerned to provide a justification for subsequent German
colonial rule in their account of the internecine warfare of the
region. None of them deal with the socio-economic reasons for
the political upheavals so evident in the late nineteenth century,
or the changes African societies were undergoing. Bley and
Drechsler are the only two detailed monographs based on
unpublished source material in East Germany to be published in
English, and they move well beyond any of the above in their
analysis of the nature and impact of German colonial rule and the
Nama—Herero uprising. The diary of Hendrik Witbooi, published by
the Van Riebeeck Society in English and Afrikaans, constitutes
a unique record of an African's response to colonial rule.
None of these works deals with the Ovambo, who only came
under formal colonial rule in 1915. Apart from Stals (1968), an
M.A. thesis and articles by Moorsom and an important joint
article by Clarence-Smith and Moorsom on class formation and
early labour migration, the history of the Ovambo in this period
has been virtually unexplored.
798

9 PORTUGUESE COLONIES AND MADAGASCAR
9 PORTUGUESE COLONIES AND MADAGASCAR, 1870-1905
Angola and Mozambique

The historiography of Portuguese-speaking Africa is generally
poor in comparison to that of other parts of the continent, but
this particular period is better documented than any other.
Partition and colonial expeditions produced a flood of reports,
articles and books, by officials, officers, missionaries, explorers,
company representatives, journalists, settlers and Creoles. Large
amounts of this material were published, usually but by no means
always in Portuguese. For Angola, the books by Capelo and Ivens
(1886), Carvalho (1890-4), Pimentel (1903), Nevinson (1906) and
Couceiro (1910) are probably the most useful. For Mozambique,
Andrade (1907), Enes (1913), Ornellas (1934-6), Botelho (1936)
and Coutinho (1941) can be consulted. However, these are only
the tip of an iceberg of published primary material, which includes
a mass of polemical pamphlets and a vigorous local press in the
colonies themselves.
Historians have not yet caught up with this immense wealth
of sources, and have been handicapped by research conditions.
The long years of dictatorship and colonial wars hampered
foreign scholars and stultified the academic community within the
Portuguese empire. Research in Portugal itself has opened up
dramatically since 1974, but conditions in Angola, wracked by
civil war and foreign invasions, have become even worse than
before. The struggling school of local historians in Mozambique
has begun valuable research but has been hampered by extremely
difficult political conditions. The wanton murder of Ruth First
in Maputo has been a cruel blow to Mozambique historians. The
collection of oral testimony has suffered most from these
conditions, but the vast body of official archives in Lisbon,
Luanda and Maputo, poorly catalogued and organised, has only
begun to be systematically exploited. The large and well cata-
logued Spiritan mission archives for Angola have been used rather
more, but the Jesuit and Franciscan archives for Mozambique
remain untouched. Scholars have relied disproportionately on
British Foreign Office documents and Protestant mission sources.
The most recent general survey of the Portuguese empire is
provided by Newitt (1981), which replaces the older syntheses by
799

Duffy (1959, 1962) and Hammond (1966). The weak point in

Newitt is the examination of the origins of colonialism in
Portugal, but this is well covered by Pirio (1982), Alexandre
(1979) and Clarence-Smith (forthcoming). Capela's books (1973,
1975 and 1977) are written rather fast, but also contain much
valuable material on this question. Almeida (1978-9) provides a
year by year chronology and large bibliography, while Duffy
(1967) surveys the labour question.
Among general histories of Angola, Wheeler and Pelissier
(1971) and Henderson (1979) are both rather weak, although
Wheeler is useful on the reactions of the Creole elite to colonial
rule, and Henderson is good on Protestant missions. Bender
(1978) purports to be a general history, but is in effect a set of
essays on white settlement. His essay on convict settlement is the
most useful for this period. Pelissier (1977) is a monumental
compilation on the campaigns of colonial conquests, which
provides a great deal of information on a wealth of other topics.
Samuels (1970) covers education and missions, and Brasio
(1966-71) reproduces a large number of documents on the
Spiritan missions. The Protestant churches are dealt with by
Soremekun (1971) and Okuma (1964). The only regional mono-
graph is Clarence-Smith (1979), but there are important articles
by Miller (1970 and 1973), Dias (1976), Birmingham (1978),
Wheeler and Christensen (1973) and Soremekun (1977).
The historiography of Mozambique has a much more regional
flavour than that of Angola, and the only general history,
Henriksen (1978), is decidedly weak for this period. The Zambezi
valley and adjoining parts of central Mozambique have attracted
the most attention. The excellent book by Vail and White (1980)
is strongly recommended, with its subtle use of official records,
private company documents and oral testimony to reconstruct the
history of the plantations of the Zambezi delta. Newitt (1973)
covers the Zambezi valley from a wider and more political
perspective, and is generally more reliable than Isaacman (1972
and 1976). Papagno (1972) is especially useful for an understanding
of the Lisbon politics behind the granting of land concessions in
the Zambezi valley. The chapter on the Mozambique Company
by Neil-Tomlinson (1978) illuminates other aspects of economic
history, while Bhila (1982) investigates the history of one small
area, Manyika, in depth. Northern Mozambique is the subject of
800

Hafkin's thesis (1975), the last chapter of Alper's book (1975) and
an article by Neil-Tomlinson (1977), but the focus of most of this
work is on the coast. For southern Mozambique, Liesegang (1967)
tells the story of the Gaza empire, while Katzenellenbogen (1982)
charts the complex colonial rivalries over railways, ports and the
supply of labour. The African origins of the migrant labour
system are the subject of afiercedebate between Harris (1959) and
Rita-Ferreira (i960 and 1963).
Madagascar and France

The archives of the Malagasy government in Tananarive have
been partly classified and are open for consultation. In France the
main deposits are in the Section d'Outre-mer of the Archives
Nationales (formerly Archives des Colonies) and, for the years
prior to 1897, the Archives des Affaires Etrangeres (Fonds
Memoires et Documents, Archives, volumes 67, 68 and 141—46).
In Britain, there are the Public Record Office and the Archives
of the London Missionary Society (now deposited at the School
of Oriental and African Studies).
For contemporary works on the colonial period, the Histoire
des Rots by Pere Malzac is a resume of Pere Callet's Tantara ny
Andriana, with a supplement covering contemporary events, as
seen by a Catholic priest. Passfield Oliver and Sibree give vivid
descriptions of the island, and present British Protestant points
of view. The Grandidiers' enormous Ethnographie de Madagascar
and Histoire politique et coloniale are based on notes gathered by
Alfred Grandidier during his explorations (1865-70), together
with additional information obtained later. Catat's exploration
(1889—90) describes the condition of the plateau and northern and
eastern coasts with numerous photographs. Gustave Julien's
work on the Institutions po/itiques et sociales is of major importance,
particularly in respect of traditional law and the reforms. Mart-
ineau's study: Madagascar en 1X94, ls a n outstanding description
of the state of the island, as seen by a Frenchman preparing public
opinion for the conquest.
As regards the conquest itself, the works of the ministers,
Gabriel Hanotaux and Andre Lebon, are valuable. The report of
General Duchesne and the book by Anthouard and Ranchot give
details of the expedition. For Gallieni himself, the essential facts
801

are contained in his reports and letters. Considerable extracts from

these, with an introduction and notes, can be found in Gallieni
pacificateur by H. Deschamps and P. Chauvet. Carol's book, while
not doing justice to Gallieni, is a splendid evocation of Tananarive
at the time. On the pacification of the south, the documents to
be consulted are the fascinating letters from Lyautey, Augagneur's
indictment, and the clarification by H. Deschamps (Les Antaisaka,
pp. 178—82). The principal journals to be consulted are: the
Antananarivo Annual for the Merina period, and Notes, reconnai-
ssances and explorations and the Bulletin de F Acade'mie Malgache for
the Gallieni period. An impressive list of review articles and
journals can be found in Grandidier's Bibliographie de Madagascar,
vol. 1, part 2; the first part is devoted to books.
Among the numerous later works are two general histories:
H. Deschamps' Histoire de Madagascar and that of Ralaimihoatra.
On the Merina period, there are the studies of Chapus and the
Pere Boudou, the former Protestant and the latter Jesuit, which
are both excellent and reasonably objective; the British point of
view given by Wastell; the sympathetic biography of Rainilaia-
rivony by Chapus and Mondain; the humorous and moving
account of Rainandriamampandry by E. F. Gautier, and the
useful and conscientious theses of Mutibwa and Randrinarisoa.
The Histoire militaire de Madagascar provides the important official
account of the military operations during the conquest and the
pacification campaigns. Concerning Gallieni himself, there are
Lyautey's letters, and various accounts by his faithful friends, such
as Guillaume Grandidier, Gheusi and Marius Ary Leblond.
IO. EAST AFRICA, 1870-1905
Cornerstones for the modern historiography of East Africa were

laid in 1963 and 1965 with the publication of the first two volumes
of the Oxford History of East Africa. Between 1965 and 1979, the
years of research and publication of primary concern here, the
initiative in many ways shifted to African universities. The results
of work in that period have been published in Zamani: a survey
of East African history, edited by B. A. Ogot (with John Kieran
in the first edition), and numerous country-oriented volumes.
Tanzania before ipoo, edited by A. D. Roberts, was followed by
'The peoples of Uganda in the 19th Century', a special number
802

IO EAST AFRICA, 1870-1905
of Tarikh, and by Kenya before 1900, edited by B. A. Ogot.

National historical associations have promoted serial publications
such as Hadith and the Kenya Historical Review, the pamphlet series
of the Historical Association of Tanzania, and so on. A history
of Tanzania, edited by I. N. Kimambo and A. J. Temu, like the
rest of the publications of this genre, was aimed at enriching the
historical literature available to secondary and undergraduate
students. Such primers, while they frequently contained the most
advanced findings available at the time of publication, carried a
minimum of bibliographical and other scholarly trappings.
Very full bibliographies were prepared for the Oxford History
of East Africa and a comprehensive listing of sources has been
provided by John Iliffe in his Modern history of Tanganyika. Valuable
supplementary bibliographies include A. D. Roberts, 'A biblio-
graphy of primary sources for Tanzania, 1799—1899', Tanzania
Notes and Records, 73 (1974); August Nimtz, 'Islam in Tanzania:
an annotated bibliography', ibid., 72 (1973); and J. B. Webster,
et a/., A bibliography on Kenya, Syracuse, 1967.
Primary sources for the history of East Africa have been
discovered, rediscovered and retrieved since 1965 by a large
number of researchers, only a fraction of whom can be acknow-
ledged in this space. At the same time, gaps in the surviving
archives, lack of facility in German, and inaccessibility of official
records owing to political upheaval have inhibited studies of the
transition from the late Swahili to the early colonial era.
Archives bearing upon commercial relations in East Africa are
spread far and wide, being lodged in Salem (Massachusetts),
Marseilles (France), Hamburg (Germany), Great Britain and
India. The British role in Zanzibar took on a more explicit
political character in the 1870s and the consular correspondence
and court records for the 1870s and 1880s provide valuable
materials concerning the affairs of Indians who were British
subjects. The Zanzibar Archives were effectively closed from the
time of the revolution in 1964 until 1974. The staff of the
Department of History, University of Dar es Salaam, undertook
a team research effort in Zanzibar in 1975 and 1976, the results
of which were still unpublished in 1983. Examples of important
work done without the benefit of archival research in Zanzibar
itself are Abdul Sheriff's yet unpublished but rightly influential
thesis, 'The rise of a commercial empire: an aspect of the
803

economic history of Zanzibar, 1770—1873' (London, 1971), and

Frederick Cooper's two volumes, Plantation slavery on the east coast
of Africa and From slave to squatter.
The official archives for the German period in mainland
Tanzania have been comprehensively described in the Guide to the
German records, z vols., Dar es Salaam and Marburg, 1973. These
materials are microfilmed. German records suffered much from
burial and destruction during the First World War and those that
survived were kept only selectively by the British. The Germans
published extensively in such periodicals as the Deutsches
Kolonialblatt, so that the record is more intact than the archives
alone suggest. Missionary archives also compensate for the
patchiness of official records. The Roman Catholic Holy Ghost
Fathers' archives (in Paris) and the White Fathers' archives (in
Rome) indeed provide records of continuous observation from
the 1870s through 1905. The London Missionary Society Archives
(at the School of Oriental and African Studies, London) contain
materials of greater value for the period before 1895 than
afterward. Late-coming missionary groups, including the German
Benedictines and Protestants of the Bethel, Leipzig, Berlin and
Moravian societies took up spheres of influence in many parts of
the country beginning in the late 1880s and thus began to
chronicle local affairs in a detailed way. Missionary sources have
been generally described in Roland Oliver's The missionary factor
in East Africa. More detail on sources for the German Protestants
is given in M. Wright, German missions in Tanganyika, ifyi-1941.
A work which is particularly impressive in using newly discovered
German official records, neglected missionary sources, oral data
and a command of geography to provide a regional history is
Lome E. Larson's 'A history of the Mahenge (Ulanga) District,
c. 1860-1957' (Ph.D. thesis, Dar es Salaam, 1976).
The Maji Maji Research Project of the University of Dar es
Salaam generated a quantity of oral data, some of which was made
available in mimeographed form, edited by John Iliffe, in 1968.
No other concerted effort to gather oral data for the period before
1905 was made between 1965 and 1979. It may be noted, however,
that the Germans recorded many accounts by Africans, examples
of which are found in the works of Carl Velten and Elize
Kootz-Kretschmer, Die Safwa (3 vols.).
The official record for Kenya in the later nineteenth century is
804

IO EAST AFRICA, 187O-I9O5
comparatively spotty and almost non-existent for such areas as the
arid, pastoral north. The Kenya coast shares with other parts of
the Muslim littoral a depth of indigenous writings, legal docu-
ments, travellers' observations and missionary chronicles. Con-
siderable gathering of oral data attended the use of these
documentary sources, as is evident in the bibliographies attached
to the work of Spear, Brantley, Ylvisaker and others listed in the
bibliography. Unpublished records of the Imperial British East
Africa Company and its agents have been described in the Oxford
History of East Africa. Supplementary papers are to be found in
the Macmillan Library in Nairobi and in the Colonial Records
Project collections at Rhodes House in Oxford. The National
Archives of Kenya do not contain extensive materials for the
period before 1905, owing to the general absence of systematic
district administration up to that time and the Secretariat fire of
1939. The Public Record Office must thus be regarded as the
primary repository for a continuous series of official records on
early Kenya. This observation takes into consideration the fact
that missionaries occupied the central and western parts of the
country only after the turn of the century, on the heels rather than
in advance of the colonial government. The various sources for
the early Kenyan period are listed in G. H. Mungeam, British rule
in Kenya, 189j-1912 (1966).
Led by B. A. Ogot, Kenyan historiography based itself exten-
sively in oral data. The cultural complexity of Kenya stimulated
localised and ethnic historical research, some critical problems in
which have been discussed by John Lonsdale in a review article,
'When did the Gusii (or any other group) become a "Tribe"?',
Kenya Historical Review (1970), 5, 1. Retrieval of history through
oral data has extended to the least documented people of East
Africa in such work as that of John Lamphear on the Jie. For
the southern pastoralists, the Masai and related or interacting
peoples, the new historiography is especially strong for the 1880s
and 1890s.
In Uganda, Makerere University had been a centre for docu-
mentation and research for all East Africa up to 1965. Thereafter
as a more nationally oriented university, its library is notable for
its holdings of private papers and historical manuscripts by
Baganda authors belonging to the vocal, literate, partisan elites
who were so important in the extension of British colonial rule.
80;

Translations of works by Apolo Kagwa have been published by

M. S. Kiwanuka. John Rowe prepared translations and made
many available through the Co-operative Africana Microfilm
Project. The Secretariat archives in Entebbe begin in 1894 but do
not assume great volume before the turn of the century. The
extension of direct administration occurred mainly after 1902 in
Uganda. Once again, personal papers and writings of colonial
agents fill the gaps, as do the archives of the White Fathers and
the Church Missionary Society (in London), organisations that
formed parties intimately engaged in the affairs of Buganda well
before the coming of formal colonialism.
In the early 1970s, the Department of History at Makerere
undertook a crash programme of oral data collection in the
northern and Kigezi parts of Uganda which had been neglected
in earlier research. Critical reflection upon the methods entailed
in this research and the work itself had to be suspended when
conditions deteriorated during the regime of Idi Amin. An
excellent study of the transition to colonialism within Uganda
with regional focus was published by John Tosh. For the
kingdoms, political history and inter-state relations continued to
be a preoccupation as it had been before 1965. The social
formation of a pre-colonial and early colonial state has been
discussed in two works on Ankole. S. Karugire greatly advanced
the discussion of caste in his Kingdom ofNkore, while M.Doornbos
showed that the kingdom did not collapse and indeed was
reconstructed in ways that favoured the monarchy and sealed a
neo-traditional order. Plainly, the more that is known of the early
colonial reconstruction, the more critical and effective will be the
use of evidence drawn from elite and non-elite informants.
The source materials for East African history by 1979
approached a point of closure, especially as the recollections of
elderly people no longer reached back to 1900 and before, and
profound events and crises in many places had dislocated old
communities and livelihoods. Yet the materials for the history of
East Africa from 1870 to 1905, albeit scattered, are nevertheless
massive, and can support increasingly sophisticated enquiries as
the balance of external and internal factors is weighed.
Between 1965 and 1979 the writing of internal history was
essentially nationalist in the sense of confining itself within the
boundaries of the states as they were at the time of independence.
806

I I NILE BASIN AND THE EASTERN HORN
Within these confines, however, much work reflects the larger

processes of history within East Africa and has comparative
significance. National histories such as Iliffe's Modern history of
Tanganyika exemplify how the transition to colonialism can be
treated in holistic, if geographically limited terms. The discussion
of the ecological aspect reflects the most advanced thinking on
this important new theme. While little is made of the common
history of the mainland and the Zanzibar portions of the present
United Republic of Tanzania, Iliffe's work is significant because
the mainland formed the heart of the Swahili commercial system
with all of its differential incorporation of localities and regions.
Also notable among the pioneering works published in the 1970s
was John Lonsdale's article on the conquest of western Kenya,
which treats local conditions and sequences of military, political
and commercial interaction in a meticulous way. The history of
the Lake Victoria basin provides an especially attractive oppor-
tunity for future research, for it has a regional cohesion, national
and international ramifications and ample source materials.
II. THE NILE BASIN AND THE EASTERN HORN,
c. 1875—c. 1907
Egypt
Some idea of the nature and scope of official Egyptian archives
for this period can be gained from H. A. B. Rivlin (1970); and
from the Quellenverêichnis of A. Scholch (?i972; English trans.
1981). There are essays on other relevant Egyptian sources in
P. M. Holt, ed., 'Political and social change in modern Egypt (1968).
For British official reaction to the Egyptian crisis, and British
military intervention and occupation, see: the Gladstone-
Granville correspondence edited by A. Ramm, 2 vols. (1962); the
diplomatic correspondence at the Pjublic] R[ecord] O[ffice],
F[oreign] O[ffice] 78 and FO 141; the Cromer Papers (PRO, FO
633); the Granville Papers (PRO 30/29); the Rosebery Papers
(National Library of Scotland, Edinburgh); and the Salisbury
Papers (Christ Church, Oxford). Cromer's Annual Reports are
useful if seen as the pleadings of an advocate rather than the
verdicts of a judge.
There are useful bibliographies, with copious references to
807

Egyptian historical and other writing in Arabic, in: Scholch (as

above); A. Hourani (1962, reprinted 1970); and P. J. Vatikiotis
(1969, revised edn 1980) - for the Arabic references see Vatikiotis'
footnotes. R. O. Collins, in R. W. Winks, ed. (1966), is also
useful. E. R. J. Owen (1969) has a very rich bibliography of
modern Egyptian economic history. For material on, or relevant
to, the history of landownership, see G. Baer (1962).
The historiography for this period is copious, but rather little
of it is fully satisfactory. At least four 'schools' of Egyptian
historiography exist (or have existed), each seeking to propagate
in English, French or Arabic its own version of modern Egyptian
history. The British 'proconsular' school was dedicated to the
creation of a 'correctly informed' public opinion in Britain. The
French school was, more or less consciously, a literary riposte to
the British occupation. The Egyptian nationalist school, in its
early stages often under strong French influence, has treated
history as a weapon in the national struggle. It has introduced into
its historiography its own political presuppositions and preoccu-
pations, sometimes to the point of producing (as Scholch has
bluntly remarked) 'wishful thinking' rather than history. The
' dynastic' school, under the patronage of Fu'ad I, was granted
privileged access to Egyptian official archives; at its best (e.g.
G. Douin, M. F. Shukry) it produced works of permanent value,
at any rate as repertories of information.
In spite of their divergent political outlooks and objectives, all
four 'schools' had much in common. The 'early modern' history
of Egypt in the Ottoman—Mamluk period was ignored or
regarded as irrelevant to later developments. There was a heavy
emphasis on narrowly political history, on Anglo-Egyptian rela-
tions, and - among the nationalists - on ideological and political
manifestations (or alleged manifestations) of Egyptian national-
ism. There was very little interest in social and economic history,
or in problems of structural change. In some of these fields a little
competent detailed work was done, but not usually by historians.
Economists and agronomists wrote on some aspects of economic
development; 'orientalists' on learned and popular Islam; des-
criptive sociologists on (especially rural) society; an educationalist
on the history of education. Historians of all four schools were
virtually unanimous in ignoring this work. Their typical product
was the ' survey' of a fairly lengthy period; but for these ' surveys'
808

II NILE BASIN AND THE EASTERN HORN
the supporting detailed monographs were almost entirely lacking.

It is very hard to find much genuinely critical historiography
before about 1950.
There is no 'standard' history of Egypt during the Isma'U-
'Urabl-Cromer era. P. J. Vatikiotis (1969, revised edn 1980) is a
'modern history' from Muhammad 'All onwards; but it is
particularly valuable for intellectual, literary and journalistic
developments in the late nineteenth and early twentieth centuries.
P. M. Holt (1966), with an even broader chronological and
geographical span, sees nineteenth-century Egypt against the
background of the 'early modern' Ottoman Near East. J. Berque
(1967, English trans. 1972) is less a 'history' than an extended
meditation on Egyptian history since about 1880. Its learning
(especially in social history) and its sympathetic concern for
contemporary Egypt are alike profound. The essays collected by
P. M. Holt in Political and social change (1968) are excellent examples
of the mid-century historiographical break-through; those by
J. N. D. Anderson, G. Baer, A. Goldschmidt and P. O'Brien are
directly relevant to the period 1875-1907.
Two major aspects of the Egyptian economy have been
thoroughly investigated: the history of landownership since 1800
by G. Baer (1962); and the role of cotton in the Egyptian
economy by E. R. J. Owen (1969). The study of the recent
economy by R. Mabro (1974) has a brief but useful historical
introduction, and some of its statistical tables go back to the 1880s.
Rural Egypt and the village community have been studied by
Hamid 'Ammar (19 5 4); by J. Berque (1957); and by G. Baer, who
has collected some of his important essays in both rural and urban
social history in his Studies in the social history of modern Egypt (1969).
For popular Islam, see J. W. McPherson (1941) and M. Gilsenan
(1973). For Egyptian Islam generally in this period, H. A. R. Gibb
(1947) is still useful; so is C. C. Adams (1933) on Muhammad
'Abduh. N. R. Keddie (1968) grapples with the elusive Jamal
al-Dln al-Afghani; E. Kedourie (1966) examines the links between
Afghani and 'Abduh. A. Hourani (1962) sensitively analyses
Egyptian religious, political and social thought. N. Safran (1961)
is interesting on the emergence of Egyptian nationalism as a
response to political, economic and social change. J. M. Ahmed
(i960) is a concise and lucid introduction to early nationalist
thought. On education, J. Heyworth-Dunne ([1938], reprinted
809

1968), writes with a broader scope than his title might suggest.
The history of the enormous and pervasive Egyptian bureaucracy
has been strangely neglected. M. Berger (1957) hardly attempts
historical analysis. F. R. Hunter, M[iddle] E[astern] S[tudies], 1983,
is not only useful in itself but a welcome sign of interest.
G. Douin, vols. 1 and 11 (1933, 1934) has much useful inform-
ation on Isma'Il's reign down to 1873, where he stops so far as
Egypt itself is concerned. The origins of Isma'Il's financial
problems may be studied in D. S. Landes (1958), and the beha-
viour of his European creditors in J. Bouvier (1968). Of the works
of the British 'proconsular' school on the crisis of 1876—83 and
on Cromer's regime, the most useful is probably that of Cromer
himself (1908). To an alert and well-informed reader, much that
Cromer writes is very revealing. Contemporary British opposition
to 'proconsular' orthodoxy is represented by W. S. Blunt (1907).
Some British historiography since about 1950 has seen the
occupation less as an administrative triumph than as a succession
of missed opportunities and imaginative failures: cf. J. Marlowe
(195 4 and 1970); P. Mansfield (1971). Although these authors still
see Egypt primarily in relation to Britain and rarely use Egyptian
sources, they evidently present a more balanced picture than
their 'proconsular' predecessors. From the copious writings
of the Egyptian nationalist school, the bibliography includes
two examples by 'Abd al-Rahman al-Rafi'I, its most prolific
author.
A. Scholch (? 1972, English trans. 1981) has by his massive
archival and prosopographical research rendered obsolete most
previous work on the origins and development of the 'UrabI
revolution. In particular, he has cast grave doubt upon the
interpretations of J. M. Landau (1953 and 1954); and of Ibrahim
Abu-Lughod, Middle East Journal, 1967. In MES, 1974, he calls
for a reconsideration of nineteenth-century Egyptian ' constitu-
tional development' in the light of his findings. Scholch has also
demonstrated (Historical Journal, 1976) the existence in 1882 of
strong pressure from the British ' men on the spot' for a military
solution to the Egyptian crisis. But he has not demonstrated that
this pressure was crucial; and he virtually ignores the increasing
fear in London that British influence in Egypt might be destroyed
by an alignment between France and the 'Urabists. On this and
other diplomatic aspects, see: R. Robinson and J. Gallagher
810

(1961); and A. Ramm in P. Gifford and W. R. Louis, ed., France

and Britain in Africa (1971).
Work on Cromer's Egypt, apart from that of the proconsular
and nationalist schools and the essentially derivative writings of,
for example, Mansfield and Marlowe, is still very scanty. Afaf Lutfi
al-Sayyid (1968) is useful on the re-emergence of Egyptian
nationalism, at first under the patronage of Khedive 'Abbas Hilml.
But she could with advantage have told us more about Egypt and
less about Cromer, on whom she has nothing very original to say.
A. Goldschmidt, in P. M. Holt, ed. (1968) and Walid Kazziha,
MES, 1977, throw light on the origins of organised political
parties in Egypt. Cromer's economic policies have been analysed
by E. R. J. Owen in his book (1969), and in MES, 1966. R.
L. Tignor (1966) is informative on some of Cromer's adminis-
trative policies but hardly adequate as an analysis of
' modernisation'.
Sudan
No records of the Turco-Egyptian administration have survived
in the Sudan: those that have survived in Cairo have yet to be
effectively exploited for the decade 1875—84. Much of our
information for this period is derived from books by, or based
on material left by, Europeans serving as khedivial officials:
Romolo Gessi, Charles Gordon, Rudolf Slatin, Eduard Schnitzer
('Emin Pasha'). But these Europeans served mainly in the south
or the far west (Darfur); and the Turco-Egyptian officials who
governed the 'metropolitan' northern and central provinces did
not write books or attract biographers. Paradoxically, in the final
decade of the Turkiyya there is more and better available source
material for the southern than for the northern Sudan. R. C. Slatin
(1896) is of course a 'primary source' for the west during his
governorships in Darfur (1879—84). The memoirs of Yusuf
Mikha'Tl, which have been edited by Salih Muhammad Nur
(Ph.D. London, 1963), are copious on politics in Kordofan on the
eve of the Mahdiyya. But for the central and northern provinces
we have only one or two sidelights and a snapshot. There are
sidelights in the memoirs of Babikr Bedri, vol. 1 (1959, English
trans. 1969). The snapshot is the Report (1883) by Lieut.-Col.
J. D. H. Stewart on administrative and other conditions in the
811

Sudan (British Parliamentary Paper [C. 3670]). However, Na'iim
Shuqayr [1903] drew heavily upon local Sudanese knowledge and
on captured Mahdist documents for his account of the late
Turkiyya and the early career of the Mahdi. His material on the
Mahdi has been summarised in German by E. L. Dietrich, Der
Islam, 1925. There is no translation into any other language.
The central archives of the Mahdist state are well preserved and
admirably arranged in the Central Record Office (Ddral-Wathaiq)
at Khartoum. Their contents and scope are described by
P. M. Holt in The Mahdist State in the Sudan (2nd edn 1970) and
in S[udan] N[otesand] R[ecords], 1955. Muhammad Ibrahim Ahmad
Abu Sallm has published in Arabic a guide to the records of the
Mahdi (1969); and, among much other Mahdist archival material,
a selection of Mahdist documents (1969). For full details of these
archival publications and of recent Sudanese historiography on
the Mahdiyya, see: P. M. Holt (1970); and P. M. Holt and
M.W.Daly (1979). P. M. Holt has published, in English, a
calendar of the Khalifa's correspondence with his general,
Mahmud Ahmad (1956). H. Shaked has published (1978) a
summary—translation of an official hagiography of the Mahdi.
Babikr Bedri, Na'fim Shuqayr and Yusuf Mikha'Il (see above) are
all valuable sources for the Mahdist revolution and state. Babikr's
memoirs are particularly interesting as the experiences of a
rank-and-file Mahdist who was not a member of the new
' Establishment'.
P. M. Holt has pointed out, in St Antony's Papers, no. 4(1958),
that some of the English-language sources on the Mahdiyya —
J. Ohrwalder (1892) and R. C. Slatin (1896) - were 'written u p '
as anti-Mahdist propaganda by F. R. Wingate from the rough
drafts of the authors. (The memoirs of Fr P. Rossignoli, published
in English translation by F. Rehfisch, SNR, 1967, escaped this
processing.) Wingate's own contribution (1891, reprinted with an
introduction by P. M. Holt, 1968) contains comparatively little
propaganda and looks more like a compendium of information
for use by a conquering and occupying army. In 1889—90 hopes
of early reconquest were very high in both Cairo and London.
Sources for 'British policy towards' the Mahdist state are
indicated in Mekki Shibeika (1952) and G. N. Sanderson (1965).
The ' staff-work' for the earliest civil administration under the
Condominium was the responsibility of the Cairo HQ of the
812

Egyptian Army intelligence branch. Early administrative papers

are therefore to be found in the CAIRINT class at the Central
Record Office in Khartoum. But they are also to be found in the
INTEL class - the records of the Assistant Director of Intelligence
stationed in Khartoum. Whatever the original criterion of
distinction (possibly the office that originated the correspondence),
it was evidently not subject-matter; down to about 1905 almost
anything 'administrative' can be found in either class. But
thereafter, as the Sudan administration developed and the link
with the Cairo Intelligence HQ became more tenuous, CAIRINT
gradually ceased to carry general administrative material. Except
for the Financial Secretary's office, the organisation of government
'departments' was still rudimentary in 1907; very little 'depart-
mental ' correspondence (other than financial) has survived from
this period.
However, for early Condominium administration the 'Sudan
Archive' at the University of Durham is at least as important as
the records at Khartoum. This archive holds both the Wingate
and the Slatin Papers, as well as other important deposits: see
G. Warburg, The Sudan under Wingate (1971) for further details.
The Wingate Papers are particularly important in view of Win-
gate's practice of transacting official business through quasi-
personal correspondence.
There are good general bibliographies of the Sudan: R. L. Hill
(1939); Abdel Rahman El Nasri (1962). But they are now rather
out-dated; as is G. N. Sanderson's survey of the historiography
of the modern Sudan, J\ournal of] A\frican\ H[istory], 1963.
M. W. Daly has compiled an annotated select general bibli-
ography, Sudan, World Bibliographical Series, vol. 40 (1983). The
entries are admirably arranged and cross-referenced. For English-
language material on the modern history of the Sudan, the
selection is judicious and misses little of importance. J. A. Dagher
(Yusuf As'ad Daghir) has compiled a bibliography of Arabic
sources for the Sudan (1968). R. L. Hill's Biographical dictionary
of the Anglo-Egyptian Sudan (1951, 2nd edn, Biographical dictionary of
the Sudan, 1967) is indispensable. J. O. Voll has compiled a
Historical dictionary of the Sudan (1978).
Only one 'general history' of the modern Sudan merits
mention. P. M. Holt and M. W. Daly (1979), the revised and
extended third edition of P. M. Holt's Modern history of the Sudan
813

(1961), is learned, lucid and concise. It has an excellent select

bibliography, which includes material in Arabic. For the general
development of the nineteenth-century Sudan, see P. M. Holt's
essay on ' Modernization and Reaction', in his Studies in the history
of the Near East (1973).
Sudan Notes and Records (1918 — in progress) is very rich in
regional and ethnic historical studies for the period from the
Turkiyya to the early Condominium. G. N. Sanderson, )A.H,
1963, is a convenient guide to much of this material. An author
and subject index to SNR, vols. 1 to 55, 1918—74, has been
published (1983). There is much similar material outside the pages
of SNR: W. Hofmayr (1925) on the Shilluk; A. Paul (1954,
reprinted 1971) on the Beja; S. Santandrea (1964) on the western
Bahr al-Ghazal, also (1977) on the town of Wau; E. E. Evans-
Pritchard (1971), on the Azande; R. S. O'Fahey (1980) on Darfur;
K. Hodnebo (Cand. Fil. thesis, Bergen, 1981) on the ecological
history of Equatoria; L. E. M. Kapteijns (D.Litt. thesis, Amster-
dam, 1982) on Dar Masallt; D. H. Johnson, JAH, 1982, on the
Dinka and the Nuer.
For the Islamic background of this period see: P. M. Holt, The
Mahdist State (1970); P. M. Holt's essay on 'Holy families' in his
Studies (1973); J. S. Trimingham (1949, reprinted 1965); and
J. O. Voll (Ph.D. thesis, Harvard, 1969).
The thin historiography of the northern Sudan during the late
Turkiyya reflects the paucity of the sources. The importance of
Na'um Shuqayr [1903] has already been indicated. G. Douin, vol.
Ill: UEmpire africain (1936-41) peters out c. 1876. R. L. Hill
(1959) has some interesting information and insights, but hardly
presents a continuous history for this period. For Gordon's first
governor-generalship, 1877—9 (during which Gordon was how-
ever heavily preoccupied with the south-western Sudan) see B. M.
Allen (1931). There is an excellent historical geography of Turco-
Egyptian Khartoum by R. C. Stevenson, SNR, 1966. The
southern Sudan is better served for this period. For the south as
a whole, see M. F. Shukry (1937) and J. R. Gray (1961); On Islam
in the south, see Gray's essay in M. Brett, ed. (1973)- For Gordon
as 'general-governor' of Equatoria, 1874-6, see M. F. Shukry
('95 3) ~ a n edition of Gordon's correspondence with Khedive
Isma'Il. G. Birkbeck Hill published (1881) material illustrating
Gordon's policies and activities in the southern Sudan over the
814

whole period 1874-9. A. Thuriaux-Hennebert (1964) has written

usefully on the Egyptian administration of the Azande. For the
Bahr al-Ghazal see: on or relating to al-Zubayr Rahma Mansur,
H.C.Jackson (1913, reprinted 1970) and W. K. R. Hallam
(1977); on Romolo Gessi, Gessi's own memoirs (1891, English
trans. 1892) and the two works by C. Zaghi (1939 and 1947). On
Equatoria under ' Emin Pasha' and the final collapse of Egyptian
rule in the south, see: the selection of Emin's letters and journals
edited by G. Schweinfurth et al. (1888); G. Schweitzer (1898,
English trans. 1898); A. J. Mounteney-Jephson (1890, reprinted
1969); R. O. Collins (1962); and, with a first-rateQuellenverêichnis
and bibliography, I. R. Smith (1972). There is an extensive
literature on Emin. His Tagebiicher have been published: F.
Stuhlmann, 5 vols. (1917—27).
For the Mahdist revolution and state, the second edition of
P. M. Holt's Mahdist State (1970) is fundamental and indispensa-
ble. Holt has also written, SNR, 1959, a brief but valuable
assessment of the significance of the Mahdiyya in Sudanese
history. On Gordon's mission and death, there is an enormous
hagiographical and, more recently, 'psychological' literature,
almost all of it useless as a contribution to Sudan historiography.
The only satisfactory study is still that in B. M. Allen (1931).
Other useful work on the Mahdiyya in English includes:
H. C. Jackson (1926) on Osman Digna ('Uthman Diqna);
A.B.Theobald (1951) and SNR, 1950; L. E. M. Kapteijns,
African perspectives (Leiden), 1977; and Ahmad 'Uthman Ibrahim,
SNR, 1979. There is now an increasing body of monographic
work on the Mahdiyya by Sudanese scholars writing in Arabic.
Examples are: a study of the theological debate between the
Mahdi and the orthodox 'alamo" by 'Abdallah 'All Ibrahim (1968),
and two regional studies — on Darfur by Musa al-Mubarak al-Hasan
[1971], and on Kordofan by 'Awad 'Abd al-Hadi al-'Ata
Apart from the material on Emin, historiography is scanty for

the southern Sudan during the Mahdiyya. For Equatoria,
R. O. Collins (1962) continues the story down to 1898. But on
the Bahr al-Ghazal he is almost completely silent; and the material
available in S. Santandrea (1964) and A. Thuriaux-Hennebert
(1964) is very fragmentary. And there is very little on the great
Nilotic peoples - Dinka, Nuer and Shilluk - who between them
815

probably amounted to more than half of the southern Sudanese

population.
Relations between the Mahdist State and Ethiopia have received
considerable attention. See: G. N. Sanderson, JAH, 1962 and
1964, and SNR, 1969; P. M. Holt, Mahdist State (2nd edn 1970);
R. A. Caulk, Transafrican Journal of History, 1971; Muhammad
Sa'Id al-Qaddal, in Arabic (1973); A. Triulzi, JAH, 1975.
The diplomacy of European competition in the Upper Nile
basin, the diplomatic background of the 'reconquest' of 1896-8,
and the development of Kitchener's campaign may be followed
in: G. N. Sanderson (1965) and SNR, 1959; A.B.Theobald
(1951); and P. Magnus (1958). The battle of Karari (Omdurman)
has been analysed in great detail from the Sudanese side by 'Ismat
Hasan Zulfu (1973, English trans. 1980). This work contains
much of value on the military organisation of the later Mahdist
state — a topic not treated in detail by Holt.
The making of the Anglo-Egyptian Condominium Agreement
of January 1899 has been studied, from various points of view,
by: Mekki Abbas (1952); Mekki Shibeika (1952); G. N. Sander-
son (1965); and Muddathir 'Abd al-Rahlm (1969). There is no
adequate treatment of Kitchener's brief governor-generalship;
there is relevant information in P. Magnus (1958) and G. R.
Warburg (1971). However, Warburg (op. cit.) is primarily con-
cerned with the structure and policies of Wingate's administration,
of which he presents a comprehensive and meticulously docu-
mented analysis, at all events for the northern Sudan; he is more
sketchy and less reliable on the south. He has studied the relations
of this administration with Cromer and his successors in Asian
and African studies (Annual of the Israel Oriental Society), 1969; and
its (Islamic) religious policy in ibid., 1971. His brief but meaty
paper in MES, 1970, is especially useful on Egyptian financial
subventions to the Sudan; and on Wingate's efforts to reduce
Egyptian influence and pave the way for an exclusively British
regime. G. N. Sanderson, in his Introduction to the English
translation of Babikr Bedri's Memoirs, vol. 11 (1980) has examined
the hastily improvised military administration of 1898-9 and
traced some of its lines of evolution under Wingate. The role of
Rudolf Slatin in the administration is the principal theme of the
biographical essay by R. L. Hill (1965). For 'All Dinar's seizure
of power in Darfur and his early relations with the Sudan
816

government, see A. B. Theobald (1965). R. L. Hill, in Sudan
transport (1965), traces the origins and development of the railway
system.
Early Sudanese resistance to British rule in the northern Sudan
has been analysed by A. S. Cudsi (M.Sc. Econ. thesis, Khartoum,
1969); by S. A. R. Bukhari (M.Phil, thesis, London, 1972), in
the context of the administration's overall problem of military
security; and by Hasan Ahmad Ibrahim, International Journal of
African Historical Studies, 1979. For early administration in the
south, and resistance to it, D. C. E. ff.Comyn (1911) is a much
more perceptive and useful testimony than its title might suggest.
There are useful sidelights in F. X. Geyer (1914). There are
accounts and analyses by: R. O. Collins and R. Herzog, JAH,
1961; S. Santandrea (1967); A. S. Cudsi (1969); R.O.Collins
(1971); and G. N. Sanderson, in Etudes... HenriBrunschwig (1982).
F. X. Geyer (1914) describes his work in the re-establishment
of the Catholic mission in the southern Sudan. The re-
establishment of the missions, especially in relation to the
administrative objectives and policies of Cromer, Kitchener and
Wingate, has been studied by: R. L. Hill, MES, 1965; R. O.
Collins (1971); and L. M. Passmore Sanderson and G. N. Sander-
son (1981).
J. Ward (1905) is illustrated with over 500 early photographs
of the Sudan and the Sudanese; and it incorporates, often verbatim,
some interesting reports of early patrols and reconnaissances.
F. X. Geyer (1914) has nearly 400 photographs.
Ethiopia
The historiography of Ethiopia for the reigns of Yohannes IV and
Menelik II suffers from the loss or destruction of almost all official
Ethiopian records, and the inaccessibility of any scraps that
remain. But it has also suffered, until quite recently, from the
inability or reluctance of European historians of modern Ethiopia
to master a difficult language and to get to grips with the complex
and subtle structure of Ethiopian society and institutions.
Nevertheless, there is now a growing body of reliable and usable
work, comparable in volume with the usable work on Egypt in
this period - though hardly with that on the Sudan.
The archives of the major European powers, and the Ottoman,
8,7

Egyptian and Mahdist archives, contain not only material relating

to Ethiopia but the correspondence of Ethiopian emperors and
kings with foreign rulers. This correspondence, which not
infrequently throws light on internal Ethiopian affairs, has been
traced for the nineteenth century by S. Rubenson's exhaustive
searches of almost all the relevant archives, including mission
archives and other unofficial collections. He reports on his
findings in detail in The survival of Ethiopian independence (1976).
Rubenson also reports on Ethiopian chronicle material both
published and in manuscript: for the Yohannes-Menelik era,
published material is evidently meagre. The official Res Gestae of
Menelik are indeed available in French translation with a useful
editorial apparatus: Guebre Sellassie, ed. M. de Coppet, 2 volumes
(1930, 1931). But whatever its value as a specimen of official
Ethiopian historiography, this work is too stylised to be of much
immediate assistance to a political historian, and it is discreet
almost to the point of total silence on Menelik's relations with
European powers. Many of the Europeans who resided or
travelled in Ethiopia in this period wrote about their experiences.
Their accounts, obviously of varying evidential value, are con-
veniently listed in Z. Gabre-Sellassie (1975) and H. G. Marcus
(1975). Alfred Ug, the most influential and long-serving of
Menelik's European advisers, has been the subject of a biography
by C. Keller (1918) and a study by W. Loepfe (1974).
A historical dictionary of Ethiopia, with a rich bibliography,
has been compiled by C. Prouty and E. Rosenfeld (1981); and a
bibliography of the modern history of Ethiopia by H. G. Marcus
(1972). There are useful introductory 'surveys' of Ethiopia by
D. J. Mathew (1947) and by E. Ullendorff(i96o). J. S. Triming-
ham (1952, reprinted 1965) is not only indispensable for Ethiopian
Islam but very useful for general orientation on Ethiopian history
and ethnography. There is unfortunately nothing comparable to
Trimingham for Ethiopian Christianity and the Ethiopian Church
in the nineteenth century. The most recent 'general history' of
Ethiopia in English is that by R. Greenfield (1965, 2nd edn 1967).
On the Yohannes-Menelik period he is brief but often perceptive;
he prints, in English translation, Menelik's famous 'Circular to
the Powers' of April 1891. The two-volume general history by
A. Bartnicki and J. Mantel-Niecko (German trans, from the
Polish, 1978) has been condemned as totally worthless by E.
818

Haberland, JAH, 1981. Nevertheless, for this period it contains

useful information competently presented; nor is the commitment
to 'dogma' (of which Haberland complains) very obtrusive. It
also has a useful bibliography.
The history of the ' non-Amhara' peoples of Ethiopia tends to
be neglected in the historiography of this period, and has to be
sought mainly in ethnographic works. Since 1959 the Frobenius-
Institut at the University of Frankfurt-am-Main has published a
series of major studies on the peoples of southern Ethiopia.
Perhaps the most useful to the modern historian is that on the
southern Galla (Oromo) by E. Haberland (1963). G. W. B.
Huntingford (1955) has written more generally on the Oromo, and
on the kingdom of Kaffa; and there is a very useful study of Jimma
Abba Jifar by H. S. Lewis (1965). For the Muslim and Muslim-
influenced peoples of Ethiopia, including the Muslim Oromo and
such important groups as the Danikil ('Afar), the Muslim groups
of Eritrea and the Sidama peoples, J. S. Trimingham (1952)
provides at least an introduction: unfortunately, Trimingham's
work lacks a bibliography.
At present, the most useful general guides to the Yohannes—
Menelik period appear to be Z. Gabre-Sellassie (1975) on
Yohannes, and H. G. Marcus (1975) on Menelik; though neither
author fully succeeds in integrating his hero's career into the
complex political, religious, economic and social structure of
Ethiopia. C. P. Rosenfeld's Chronology ojMenilek 7/(1976) is in fact
a very copious repertory of information from 1844 to 1913,
admirably arranged and as far as possible precisely dated (where
chronological precision is impossible, the author says so). For
such themes as the chronology of Menelik's campaigns of
conquest, the scale and tempo of his arms supplies, the payment
of tribute to and by him, Rosenfeld is not merely convenient but
virtually indispensable.
R. Pankhurst has written extensively on Ethiopian economic
history since 1800: (1968); ]\ournal of\ Ethiopian] S[tudies], 1964;
Journal of Semitic Studies, 1964. R. A. Caulk has worked on the
urban and economic history of Harar: JES, 1971; JAH, 1977.
R. Pankhurst has written on the foundation by Menelik of Addis
Ababa as a new 'fixed capital', JAH, 1961; and P. Garretson on
the early history and development of Addis Ababa (Ph.D. thesis,
London, 1974). R. Pankhurst has also written on State and land in
819

Ethiopian History (1966). D. Crummey has analysed, in two

important papers {Past and Present, 1980; JA.H, 1983), the rela-
tionship between hereditary ' nobility', hereditary landownership
and political power. H. Erlich, JAH, 1974, examines the ascent
to the highest ranks of the nobility of a 'king's man' of lowly
social origin. R. A. Caulk has analysed, for nineteenth-century
Ethiopia, both the political significance of firearms, J^4H, 1972;
and the inter-action of'religion and state', JES, 1972. Much of
this recent analytical work has yet to be assimilated into the
general historiography.
Ethiopia's role in the partition of Africa, both as a successful
' resister' and as an active participant, has attracted much scholarly
attention. For work of this kind, some Ethiopian sources are
available, though not of course in Ethiopian archives; and the
unsolved problems of the internal structure and dynamics of
Ethiopia are a less crippling handicap. There are moreover useful
European sources: not only diplomatic correspondence in the
relevant national archives (on which S. Rubenson (1976) should
be consulted); but also valuable published testimonies. G.
H. Portal 'wrote up' his mission to Yohannes twice ([1888] and
1892). For the Rodd mission of 1897 there is A. E. W. Gleichen
(1898) and, better, the official British Precis of information (1897).
C. Zaghi has edited (1956) Salimbene's diary of his mission to
Menelik; and C. Michel (1900) is frank and informative on the
French missions that attempted to reach the White Nile with
Menelik's 'assistance' - though he tends to blame Lagarde, rather
than Menelik, for their failure.
For Ethiopian resistance, see S. Rubenson (i960); and above
all his Survival of Ethiopian independence (1976). Important contri-
butions have also been made by R. A. Caulk (Ph.D. thesis,
London, 1966), and Transafrican Journal of History, 1971. However,
after 1889 Rubenson concentrates rather narrowly on the local
consequences of the Treaty of Wichale, on which see also
S. Rubenson (1964) and JAH, 1964; he has little to say on the
fluctuating pattern of European alignments and antagonisms
within which Menelik had to make his diplomatic moves. On
these themes see G. N. Sanderson (1965). Rubenson virtually
ends his story in March 1896; but Menelik himself did not believe
that he had completely solved Ethiopia's European problem at
Adowa. For Menelik's interventions in the politics and diplomacy
820

of the Nile Valley between 1896 and 1898, see: H. G. Marcus

(1975), and JAH, 1963 and 1966; G. N. Sanderson (1965) and
JAH, 1964. One of the most important of these initiatives was
Menelik's successful attempt to reach an entente with Khalifa
' Abdallahi: on this, and on Mahdist-Ethiopian relations generally,
see the works cited above, p. 816. On Russian activities in
Ethiopia, C. Zaghi (1975) supersedes the sketchy, though some-
times perceptive, treatment by C. Jes'man (1958).
Little seems to have been written on the internal politics and
conditions of Ethiopia between 1902 and 1908. H. G. Marcus
(1975) describes in interesting detail the 'domestic' organisation
of Menelik's imperial household. H. Darley (1926, reprinted 1972)
has some vivid, not to say lurid, side-lights on conditions in
south-western Ethiopia: he frequently alludes to the deterioration
that followed Menelik's incapacity. Otherwise, historians have
concentrated on the complex diplomatic and financial struggle for
control of the then unfinished Ethiopian railway; and on the
tortuous European diplomacy which ultimately led, in the Tri-
partite Treaty of 1906, to the frustration of renewed Italian
territorial aspirations and the stabilisation of Ethiopia's inter-
national position. Marcus (1975) is informative on both topics;
and has written specifically on the Tripartite Treaty, JES, 1964.
For the railway question, see T. L. Gilmour (1906); W. Loepfe
(1974); and K. V. Ram, Journal of Imperial and Commonwealth
History, 1981.
Somalia
Archival material for much of this period of Somali history will
be found in the records of the British India, Foreign and Colonial
Offices; and in those of the Archivio Storico delfex-Ministero
delF Africa ltaliana. In this period the British government published
two 'Blue Books' on relations with, and operations against,
Muhammad 'Abdallah Hasan (Cd. 1597, 1903; Cd. 5000, 1910);
and a two-volume Official history of the operations in Somaliland,
1901-4 (War Office, 1907).
Historiography is very scanty. There are useful bibliographies
in 1. M. Lewis (1961, reprinted 1967); in R. L. Hess (1966); and
in S. S. Samatar (1982). That in Samatar is the most compre-
hensive. But it omits a few relevant titles to be found in Lewis
and Hess; and it is considerably inflated by material which, while
821

doubtless relevant to the author's field of study, is hardly relevant

to Somali history.
This period of Somali history is incomprehensible without
some knowledge of: Somali socio-political structure and institu-
tions; the nineteenth-century development of Somali Islam; and
the role of the Somali language as a political weapon. These three
closely linked topics are admirably analysed, with varying
emphases, by: I. M. Lewis (1961, reprinted 1967)-and cf. Lewis'
essay in I. M. Lewis, ed. (1966); B. W. Andrzejewski and I.
M. Lewis (1964); and S. S. Samatar (1982).
The best and most balanced treatment of Somali history in this
period appears to be that by Samatar (1982) although it is no more
than a sketch in a work whose primary focus is socio-linguistic
rather than historical. Among other virtues, Samatar relates
Muhammad 'Abdallah Hasan's movement clearly and cogently to
the clan and lineage conflicts of' normal' Somali politics; and he
takes the Sayyid's Somali opponents seriously. I. M. Lewis (1965,
revised edn 1980) is useful on Egyptian and early British admin-
istration; helpful (though not always completely accurate) on
the Anglo-French 'scramble' and partition of 1884-8; and
admirable on the social and religious background. But he is almost
silent on the internal development of the Sayyid's movement
between 1899 and 1908; and on the British attempts to destroy
it by military action between 1901 and 1904.
Five substantial pieces of writing have been devoted more or
less exclusively to the career of Muhammad 'Abdallah Hasan.
That by Sahykh Jama' 'Umar 'Tsa (1976) is in Somali. That by
D. Jardine (1923, reprinted 1969) is over-loaded with military
detail; but when Jardine escapes from this preoccupation he can
often be perceptive, in spite of some routine ' official' denigration
of the Sayyid. F. S. Caroselli (1931) owes much of his factual
material to Jardine, but is critical of British policy. An Egyptian
interpretation by 'Abd al-Sabiir Marzuq (Cairo, 1964) sees the
Sayyid's movement as part of an anti-British jihad common to
Egypt, the Sudan and Somalia. Oddly, he calls Muhammad
'Abdallah a 'Mullah'. R. L. Hess, in N. R. Bennet, ed. (1968), is a
competent (but very one-sided) synthesis of the published work,
with some original material on the Sayyid's relations with the
Italians.
I am indebted to an unpublished M.A. dissertation (University
822

of London, 1975) by Dr L. E. M. Kapteijns for information on

the treatment of Muhammad 'Abdallah in Shaykh Jama' 'Umar
'Isa's Arabic-language History of the Somali in medieval and modern
times (1965). Shaykh Jama' evidently went to great pains to collect
Somali oral traditions (or at any rate those favourable to the
Sayyid); and he bases his account firmly on these, but not without
occasional anachronistic inputs. Shaykh Jama' relates (as do
Samatar and most European writers) a number of anecdotes about
the Sayyid's early life, and especially about his alleged clashes with
British officials; but it is very difficult to assess the evidential value
of these stories. The best guide to the Sayyid's ideas is obviously
his own utterances. His Somali-language Diwdn (Collected
Verse) has been published (but apparently with some bowdleri-
sation and expurgation) by the indefatigable Jama' 'Umar 'Tsa
(Mogadishu, 1974). A scholarly edition in an European language
would be an invaluable instrument de travail.
For the Egyptian period in northern Somalia, G. Douin, vol.
in: U'Empire africain (Cairo, 1936—41), describes the establish-
ment of Egyptian control in considerable detail, but is very thin on
administration. For the establishment of the British protectorates,
and British administration down to 1905, see A. M. Brockett
(D.Phil, thesis, 1969). The establishment of the Italian pro-
tectorates is traced by R. L. Hess (1966); Hess also illustrates the
impact of the Sayyid's movement and propaganda upon Italian
Somaliland, ]AH, 1964. B. G. Martin (1976) attempts, not very
convincingly, to explain Muhammad 'Abdallah's quarrel with the
Qadiriyya as a dispute about the permissible limits of tawassul (the
invocation of Muslim saints as mediators between God and man).
Samatar (1982) helps to elucidate the political and ecological
constraints and opportunities that determined the strategy of the
Sayyid between 1899 and 1910. The British counter-strategy, and
the fetters placed upon it by London, can be discerned amidst the
dense 'bush' of military detail in the pages of Jardine (1923).
There does not appear to be anything substantial on the Sayyid's
sojourn at Ulig.
823

BIBLIOGRAPHY
GENERAL
The following are the principal African and regional histories, and other
important works covering the whole continent.
Many other works are cited in several chapters: full details are given under
the earliest chapter, 'see above' and 'see below' relate to references in the same
chapter.
Ajayi, J. F. A. and Crowder, M. (eds.), History of West Africa, n. London, 1974.

Curtin, P., Feierman, S., Thompson, L. and Vansina, J., African History.
Boston, Toronto, London, 1978.
Fage, J. D., A history of West Africa: an introductory survey (4th edn of An
introduction to the history of West Africa). Cambridge, 1969.
A history of Africa. London, 1978.
Gann, L. H. and Duignan, P. (eds.), Colonialism in Africa 1870-1960 (Cam-
bridge, 1969—75): 1, The history and politics of colonialism, 1870-1914
(1969); 11, The history and politics of colonialism, 1914-1960 (1970-
reprinted 1982); in, Profiles of change: African society and colonial rule,
ed. V. Turner (1971); iv, The economics of colonialism (1975); v, A
bibliographical guide to colonialism in Africa (1973).
Gifford, P. and Louis, W. R. (eds.), Britain and Germany in Africa. New Haven,
1967.
(eds.), France and Britain in Africa. New Haven, 1971.
Julien, Ch.-A., Histoire de I'Afrique du Nord. Parie, 1921. (English edn ed.
C. C. Stewart. London, 1970).
Oliver, R. and Mathew, G. (eds.), History of East Africa, 1. Oxford, 1963.
Robinson, R. and Gallagher, J. with Denny, A., Africa and the Victorians: the
official mind of Imperialism. London, 1961 (2nd edn, 1981).
Wilson, M. and Thompson, L. M. (eds.), The Oxford history of South Africa, 2
vols. Oxford, 1969.
ARCHIVES
The following are the principal references to archival sources in Europe.

Archival and bibliographical sources for particular areas are included under
the relevant chapter.
Archives de la Marine (Serie BB, 4), Paris.

Archives des Affaires Etrangeres (Fonds Memoires et Documents), Archives
67, 68, 141—6, Paris.
Archives Nationales, Section Outre-mer (S.O.M.), Paris and Aix-en-Provence.
824

Giglio, C , Inventario delle Fonti manuscritte relative alia Storia dell' Africa del Nord
esistenti in Italia, i vols. Leiden, 1971, 1972.
Giglio, C. and Lodolini, E., Guida delle Fonti per la Storia dell'Africa a Sud del
Sahara esistenti in Italia, 2 vols. Zug, Switzerland, 1973, 1974.
Guide to materials for West African history in European archives (Athlone Press
1962- ) : 1, Belgium and Holland (1962); 11, Portugal (1965); in, Italy
(1965); iv, France (1968); v, United Kingdom (1973).
Guides to the sources of the history of Africa south of the Sahara (International Council
of Archives 1970-76): 1, Federal Republic of Germany (1970); 11, Spain
(1971); in, iv, France (1971, 1976); v, vi, Italy (1973, 1974); VII, Holy
See (still unpublished?); v m , Denmark, Norway and Sweden (1971); ix,
Netherlands (1978).
Matthews, N. and Wainwright, M. D. (compilers), ed. J. D. Pearson, A guide
to manuscripts and documents in the British Isles relating to Africa. London,
1971.
Public Record Office (P.R.O.), List of Cabinet papers, 1880-1914. London,
1964.
Public Record Office Handbooks: no. 3, Colonial Office; no. 13, Foreign
Office.
I. AFRICA ON THE EVE OF PARTITION
Adeleye, R. A., Power and diplomacy in northern Nigeria, 1804-1906. London,

1971.
Agbodeka, F., African politics and British policy in the Gold Coast, 1868-1900.
London, 1971.
Ageron, C. R., Les Algeriens musulmans, et la France 1871-1919. Paris, 1968.
Ajayi, J. F. A., Christian missions in Nigeria, 1841-91. London, 1965.
Ajayi, J. F. A. and Crowder, M. (eds.), History of West Africa {see under
General).
Akintoye, S. A., Revolution and power politics in Yorubaland 1840-189). London,
1971.
Alagoa, E. J., 'Nineteenth century revolutions in the Eastern States and
Calabar', 1971,/. Hist. Soc. Nig., 5, 4, 565-74.
Alpers, E. A., Ivory and slaves in East Central Africa. London, 1975.
Arhin, K., West African leaders in Ghana in the nineteenth and twentieth centuries.
London, 1979.
Ayandele, E. A., The missionary impact on modern Nigeria, 1842—1914. London,
1966.
The educated elite in the Nigerian society. Ibadan, 1974.
Baer, G., A history of landownership in modern Egypt, 1800-19 to. London, 1962.
Studies in the social history of modern Egypt. Chicago, 1969.
Baier, S., An economic history of central Niger. Oxford, 1980.
'Trans-Saharan trade and the Sahel: Damergu, 1870—1930', / . Aft. Hist.,
1977, 18, 1, 37-6o.
Beach, D. N., The Shona and Zimbabwe. London, 1980.
Bennett, N. R., Studies in East African history. Boston, 1963.
Mirambo of Tanzania 1840^-1884. New York, 1971.
825

BIBLIOGRAPHY
Berque, J., French North Africa. London, 1967.
L.'inte'rieur du Maghreb. Paris, 1979.
Biobaku, S. O., The Egba and their neighbour*, 1842-1872. Oxford, 1957.
Birmingham, D., 'The copper barons of Cazengo', J. Afr. Hist., 1978, 19, 4,
525-38.
Birmingham, D. and Martin, P. M., History of Central Africa. London,
1983.
Boahen, A. A., Britain, the Sahara and the western Sudan, 1788-1861. Oxford,
1964.
Bouroue-Avaro, J. a., Unpeuplegabonais a I'aube de la colonisation: le has Ogowe
auXlX' siecle. Paris, 1981.
Bovill, E. W., The Golden Trade of the Moors 2nd edn, revised R. Hallett.
London, 1968.
Brenner, L., ' T h e North African trading community in the nineteenth century
Central Sudan', in D . E. McCall, and N . R. Bennett (eds.), Aspects of West
African Islam. Boston, 1971.
Brett, M., 'Problems in the interpretation of the history of the Maghreb in the
light of some recent publications', J. Afr. Hist., 1972, 13, 3, 489-506.
' Modernisation in 19th century North Africa', The Maghreb Review, 1982,7,
1-2, Jan-April, 16-22.
Brooks, G. E., 'Peanuts and colonialism: consequences of the commercialis-
ation of peanuts in West Africa, 1830-70',/. Afr. Hist., 1925,16, 1, 29-54.
Brown, B., 'Muslim influence on trade and politics in the Lake Tanganyika
region', Int. J. Afr. Hist. Studies, 1971, 4.
Brown, L. C , The Tunisia of Ahmad Bey 18)7-18JJ. Princeton, 1974.
Burke, i n , E., Prelude to Protectorate in Morocco: pre-colonialprotest and resistance,
i860—if 12. Chicago and London, 1976.
Caulk, R. A., 'Harar town and its neighbours in the nineteenth century', / .
Afr. Hist., 1977, 18, 3, 369-86.
Cerulli, E., Somalia: scritti vari editi ed inediti, 3 vols. Rome, 1957, 1959, 1964.
Clarence-Smith, W. G., Slaves, peasants and capitalists in Southern Angola,
1840-1926. Cambridge, 1979.
Cookey, S. J. S., King Ja Ja of the Niger Delta. Nok, 1974.
Cooper, F., Plantation slavery on the east coast of Africa. New Haven, 7977.
'The problem of slavery in African studies',/. Afr. Hist., 1979,20,1, 102-25.
Coquery-Vidrovitch, Catherine, Braâ et la prise de possession du Congo. Paris,
1969.
Cordell, D. D., 'Eastern Libya, Wadai and the Sanusiya',/. Afr. Hist., 1977,
18, 1, 21-36.
' Blood partnership in theory and practice: the expansion of Muslim power
in Dar al-Kuti, / . Afr. Hist., 1979, 20, 3, 379-94.
Crummey, D., Priests and politicians. Oxford, 1972.
Curtin, P., Economic change in precolonial Africa: Senegambia in the era of the slave
trade. Madison, 1975.
Curtin, P. et a/., African history (see under General).
Dampierre, E. de, Un ancien royaume Bandia du Haut-Oubangui. Paris, 1967.
Darkwah, R. H. K., Shewa, Menelik and the Ethiopian empire, i8ij-i88f.
London, 1975.
826

De Kiewiet, C. W., The imperial factor in South Africa. Cambridge, 1937.
Deschamps, H., Histoire de Madagascar. Paris, 1961; 4th edn 1972.
Dias, J. R., 'Black chiefs, white traders and colonial policy near the Kwanza:
Kabuku Kambilo and the Portuguese, 1873-1896', J. Afr. Hist., 1976,
17, 2, 245-65.
'Famine and disease in the history of Angola, r.1830-1930', J. Afr. Hist.,
1981, 22, 3, 349-78.
Dike, K. O., Trade and Politics in the Niger Delta 1830-188;. Oxford, 1956.
Douin, G., Histoire du rigone du Khedive Ismail, 4 parts. Cairo, 1936-41 (3 vols.
bound as 6, Rome, 1933-4). part in: L.'empire africain.
Dumett, R. E., 'The rubber trade of the Gold Coast and Asante in the
nineteenth century', J. Afr. Hist., 1971, 12, 1, 79—101.
' John Sarbah the elder, and African mercantile entrepreneurship in the Gold
Coast in the late nineteenth century', J. Afr. Hist., 1973, 14, 4, 653-79.
Dumett, R. E. and Schwartz, B. K.., West African culture dynamics: archaeological
and historical perspectives. Chicago, 1979-
Evans-Pritchard, E. E., The Sanusi of Cyrenaica. Oxford, 1949.
Fage, J. D., History of West Africa (see under General).
History of Africa (see under General).
Ganiage, J., lies origines du protectorat franfais en Tunisie, 1861-1881. Tunis, 1959
(2nd edn 1968).
Gbadamosi.T. G. O., The growth of Islam among the Yoruba, 1841-1908. London,
1979.
Gellner, E., Saints of the Atlas. London, 1969.
Gifford, P. and Louis, W. R. (eds.), Britain and Germany (see under General).
Trance and Britain (see under General).
Gray, J. R., A history of the southern Sudan 18)9—1889. London, 1961.
Gray, J. R. and Birmingham, D., Pre-colonial African trade: essays on trade in
central and eastern Africa before 1900. London, 1970.
Green, Arnold, The Tunisian ulama 1873-19/;: social structure and response to
ideological currents. Leiden, 1978.
Guy, J., The destruction of the Zulu kingdom. London, 1979.
Hargreaves, J. D., Prelude to the partition of West Africa. London, 1966.
West Africa partitioned, 1: The loaded pause 188J-89. London, 1974; vol. n:
The Elephants and the Grass (London, 1984).
Harms, R. W., River of wealth, river of sorrow. New Haven and London, 1981.
Hartwig, G. W., 'The Victoria Nyanza as a trade route in the nineteenth
century', J. Afr. Hist., 1970, n , 4, 535-52.
The art of survival in East Africa: the Kerebe and long-distance trade i8}o-i89j.
New York: 1976.
Hill, R. L., Egypt in the Sudan, 1820-1881. London, 1959.
Hoben, A., Land tenure among the Amhara of Ethiopia. Chicago and London,
»973-
Hogendorn, J. S., 'The economics of slave use on "Plantations" in the Zaria
emirate of the Sokoto caliphate, Int. J. Afr. Hist. Studies, 1977,10,369-83.
' Slave acquisition and delivery in pre-colonial Hausaland', in R. Dumett and
B. K. Swartz (eds.), West African culture dynamics: archaeologicaland historical
perspectives. Chicago, 1979.
827

BIBLIOGRAPHY
Holt, P. M., Egypt and the Fertile Crescent, //16-1922. London, 1966.
The Mahdist State in the Sudan 1881-94. Oxford, 1958 (2nd edn, revised and
enlarged, 1970).
A modern history of the Sudan. London, 1961, 196}. With M. W. Daly, 3rd edn,
History of the Sudan from the coming of Islam to the present day. London, 1979.
Hopkins, A. G., An economic history of West Africa. London, 1973.
Ijagbemi, E. A.,' The Freetown colony and the development of legitimate trade
in the adjoining territories', /. Hist. Soc. Nigeria, 1970, 5, 2, 243-56.
IlifFe, J., A modern history of Tanganyika. Cambridge, 1979.
Isaacman, A., Mozambique: the Africanisation of a European institution: the Zambesi
Pra%os, iyjo—1902. Madison, 1972.
The tradition of resistance in Mozambique: anti-colonial activity in the Zambesi valley,
18/0-/92/. London, 1976.
Isichei, E., The Ibo people and the Europeans: the genesis of a relationship, to 1906.
London, 1973.
History of West Africa since 1800. London, 1977.
Johnson, Douglas, 'Algeria, some problems of modern history',/. Afr. Hist.,
1964, 5, 2, 221-42.
Johnson, M., 'The cowrie currencies of West Africa', part 1, / . Afr. Hist.,
1970, i i , 1, 17-49; P art *» J- Afr- Hit/> l97°< "» 3. 331—53-
'By ship or by camel: the struggle for the Cameroons ivory trade in the
nineteenth century', / . Afr. Hist., 1973(3), 19, 4, 539-49.
'Cloth on the banks of the Niger',/. Hist.Soc. Nigeria, 1973(^,6,4,353—64.
'Calico caravans', /. Afr. Hist., 1976, 17, 1, 95-117.
Jones, G. I., The trading states of the Oil Rivers. London, 1963.
Julien, Ch.-A., Histoire de I'Alge'rie contemporaine, vol. 1. Paris, 1964.
Le Maroc face aux impe'rialismes, 1411-19/6. Paris, 1978.
Kalck, P., Histoire de la Re'publique Centrafricaine. (Paris, 1974).
Kanya-Forstner, A. S., The conquest of the western Sudan: a study in Trench
imperialism. Cambridge, 1969.
Klein, M., Islam and imperialism in Senegal. Edinburgh, 1968.
Klein, M. and Lovejoy, P. E., 'Slavery in West Africa' in H. Gemery and
J. S. Hogendorn (eds.), The uncommon market: essays in the economic history
of the Atlantic slave trade. New York, 1979.
Knapp, W., North West Africa: a political and economic survey. Oxford, 1977.
Kopytoff, J. H., A preface to modern Nigeria: the 'Sierra Leonians' in Yoruba,
/8J0-/890. Madison, 1965.
Landes, D. S., Bankers and Pasbas. London, 1958.
Latham, A. J. H., 'Currency, credit and capitalism on the Cross River in the
pre-colonial era',/. Afr. Hist., 1971, 12, 4, 599-605.
'Witchcraft accusations and economic tension in pre-colonial Old Calabar',
/. Afr. Hist., 1972, 13, 2, 249-60.
Old Calabar 1600-1891. Oxford, 1973.
Law, R. C. C, 'Royal monopoly and private enterprise in the Atlantic trade:
the case of Dahomey',/. Afr. Hist., 1977, 18, 4, 555-77.
'In search of a Marxist perspective on pre-colonial tropical Africa', /. Afr.
Hist., 1978, 19, 3, 44»-5*-
828

' Slaves, trade and taxes: the material basis of political power in pre-colonial
West Africa', Research in Economic Anthropology, 1978, 1, 37-52.
Leone, E., La coloniâîone dell'Africa del Nord, 2 vols. Padua, 1957, i960.
Lewin, T. J., Asante before the British: the Prempeanyears, 1871-1900. Kansas,
1978.
Lewis, H. S., A Galla monarchy. Madison, 1965.
Lewis, I. M., The modern history of Somaliland. London, 1965 (2nd, revised, edn,
A modern history of Somalia, Nation and State in the Horn of Africa, London,
1980).
Little, T., Egypt. London, 1958.
Lovejoy, P. E., 'Long-distant trade and Islam: the case of the nineteenth
century Hausa Kola-trade',/. Hist. Soc. Nigeria, 1971, 5, 4, 537-47.
'The Kambarin Beriberi: the formation of a specialised group of Hausa
Kola-traders in the nineteenth century',/. Afr. Hist., 1973, 14, 4, 633-51.
'Interregional monetary flows in the precolonial trade of Nigeria', J. Afr.
Hist., 1974, 15,4, 563-85.
'Plantations in the economy of the Sokoto caliphate', / . Afr. Hist., 1978,
19, 341-68.
Lovejoy, P. E. and Baier, S., 'The desert-side economy of the Central Sudan',
Int. J. Afr. Hist. Studies, 1975, 7, 4, 551-81.
McCall, D. F. and Bennett, N. R. (eds.), Aspects of West Africa Islam. Boston,
1971.
Marks, S. and Atmore, A. (eds.), Economy and society inpre-industrial South Africa.
London, 1980.
Marsden, A., British diplomacy and Tunis (1871—1902). New York, 1971.
Martin, P. M., The external trade of the Loango coast 1176-1870. Oxford, 1972.
Mason, M., ' Captive and client labour and the economy of the Bida emirate
1857-1901', J. Afr. Hist., 1973, 14, 3, 45 3-7I-
Matthews, T. I.,' Portuguese, Chikunda and the peoples of the Gwembe valley:
the impact of the " lower Zambezi complex " on southern Zambia', J. Afr.
Hist., 1981, 22, 1, 23-41.
Meillassoux, C. (ed.), The development of indigenous trade and markets in West Africa.
London, 1971.
(ed.), L'esclavage en Afrique precoloniale. Paris, 1975).
Miege, J-L., Le Maroc et /'Europe, 18)0—1894, 3 vols. Paris, 1961-3.
Miers, S. and Kopytoff, I. (eds.), Slavery in Africa: historical and anthropological
perspectives. Madison, 1977.
Miller, J. C. (ed.), The African past speaks. Folkestone, Eng. and Hamden,
Conn., 1980.
Montaigne, Robert, The Berbers, their social andpolitical organisation. (Translation
by David Sedden, London, 1973 of La vie sociale et la vie politique des
Berberes, Paris, 1931.)
Mutibwa, P. M., The Malagasy and the Europeans, 1861-189). Ibadan, 1973.
al-Naqar, 'U., The pilgrimage tradition in West Africa. Khartoum, 1972.
Newbury, C. W., The western slave coast and its rulers. Oxford, 1961.
'North African and western Sudan trade in the nineteenth century: a
re-evaluation', J. Afr. Hist., 1966, 7, 2, 233-46.
829

BIBLIOGRAPHY
'Trade and authority in West Africa' in Gann, L. H. and Duignan, P. (eds.),
Colonialism in Africa (see under General).
'Prices and profitability in early nineteenth century West Africa trade' in
C. Meillassoux (ed.), The development of indigenous trade and markets in West
Africa. London, 193 1.
'Credit in early nineteenth century West African trade',/. Afr. Hist., 1972,
13, 1, 81-95.
Newitt, M. D. D.,' Angoche, the slave trade and the Portuguese, c. 1844-1910',
/ . Afr. Hist., 1972, 13, 4, 659-72.
Portuguese settlement on the Zambesi. London, 1973.
Nicholls, C. S., The Swahili Coast: politics, diplomacy and trade on the East African
littoral, 1798-1816. London, 1971.
Northrup, D., 'The compatibility of the slave and palm oil trades in the Bight
of Biafra', / . Afr. Hist., 1976, 17, 3, 353-64.
Trade without rulers: pre-colonial economic development in south-eastern Nigeria.
Oxford, 1978.
Oded, A., Islam in Uganda. New York, 1974.
Ogot, B. A. (ed.), Zamani: a survey of East African history. 2nd edn, Nairobi,
1974-
(ed.), Kenya before 1900. Nairobi, 1976.
Oliver, R., The Missionary Factor in East Africa. London, 1952; 2nd edn 1965.
Sir Henry Johnston and the scramble for Africa. London, 1957.
Oliver, R. and Mathew, G. (eds.), History of East Africa (see under General).
Oloruntimehin, B. O., The Segu Tukulor empire. Ibadan, 1972.
Pachai, B. (ed.), The early history of Malawi. London, 1972.
Page, M. E., 'The Manyema hordes of Tippu Tip', Int. J. Afr. Hist. Studies,
1974, 1, 69-84.
Pallinder-Law, A., 'Aborted modernisation in West Africa?-the case of
Abeokuta', J. Afr. Hist., 1974, 15, 1, 65-82.
Palmer, R. and Parsons, N. (eds.), The roots of rural poverty in central and southern
Africa. London, 1977.
Pankhurst, R. K. P., Economic history of Ethiopia 1800—19);. Addis Ababa, 1968.
Patterson, K. D., The northern Gabon coast to 187;. Oxford, 1975(8).
'The vanishing Mpongwe: European contact and demographic change in
the Gabon river',/. Afr. Hist., 197j(b), 16, 2, 212-38.
Peel, J. D. Y., 'Kings, titles, quarters: a conjectural history of Ilasha', History
in Africa, 1979, 6.
Pelissier, R., Lesguerresgrises: resistance et revoltes en Angola, 184J-1941. Orgeval,
•977-
Person, Y., Samori: une Revolution Dyula, 3 vols. Dakar, 1968, 1970, 1975.
Peterson, J., Province of freedom: a history of Sierra Leone, 1787—1870. London,
1869.
Phimister, I. R.,' Precolonial goldmining in southern Zambezia', African Social
Research, 1976, 21, 3, x—30.
Porter, A., Creoledom: a study of the development of Freetown society. London, 1963.
Quinn, C. A., Mandingo kingdoms of the Senegambia: traditionalism, Islam, and
European expansion. Evanston/London, 1972.
830

Ranger, T. O., Revolt in Southern Rhodesia 1896-7. London, 1967.

(ed.), Aspects of Central African history. London, 1968.
Ranger, T. O. and Kimambo, I. N. (eds.), The historical study of African religion.
London, 1972.
Reynolds, E., Trade and economic change on the Gold Coast, 1807-1874. London,
1974.
Roberts, A. D., Tanzania before 1900. Nairobi, 1968.
A history of the Bemba: political growth and change in north-eastern Zambia before
1900. London, 1973.
A history of Zambia. London, 1976.
Roberts, R., 'Long distance trade and production: Sinsani in the nineteenth
c e n t u r y ' , / . Afr. Hist., 1980, 21, z, 169-88.
Rubenson, S., The survival of Ethiopian independence. London, 1976.
Shibeika, Makki, Ta'rikh muluk al-Suddn. Khartoum, 1947.
Slade, R., King Leopold's Congo. London, 1962.
Smith, R. S., 'Peace and palaver: international relations in colonial West
Africa', J. Afr. Hist., 1973, 14, 4, 599-621.
Stokes, E. and Brown, R. (eds.), The Zambeîan past: studies in Central African
history. Manchester, 1966.
Swindell, K . , ' Serawoollies, Tillibunkas and strange farmers: the development
of migrant ground-nut farmers along the Gambia river', / . Afr. Hist.,
1980, 21, 1, 93-104.
Syfert, D. N., 'The Liberian coasting trade, 1822-1900', / . Afr. Hist., 1977,
18, 2, 217-35.
Thompson, L. M. (ed.), African societies in southern Africa. London, 1969.
Trimingham, J. S., Islam in Ethiopia. London, 1952; reprinted 1965.
Triulzi, A . , ' Trade, Islam and the Mahdia in north-western Wallaga, Ethiopia',
J. Afr. Hist., 1975, 16, 1, 55—71-
Trout, F. E., Morocco's Saharan frontiers. Geneva, 1969.
Vail, L. and White, L., Capitalism and colonialism in Mozambique: a study of
Quelimane district. London, 1980.
Valensi, L., Le Maghreb avant la prise d'Alger 1790-18jo. Paris, 1969.
On the eve of colonialism: North Africa before the French conquest. London, 1977.
Van Krieken, G., Khe'redine et la Tunisie I8;O-I88I. Leiden, 1976.
Vansina, J., Kingdoms of the savanna. Madison, 1966.
The Tio kingdom of the Middle Congo 1880-1892. London, 1973.
The children of Woot. Madison, 1978.
Vatikiotis, P. J., The modern history of Egypt, London, 1969 (revised edn, The
History of Egypt from Muhammad AH to Sadat. London, 1980).
Vellut, J-L., 'Notes sur le Lunda et la frontiere luso-africaine, 1700—1900',
Etudes d'histoire africaine, 1972, 3, 61-166.
Wilks, I., Asante in the nineteenth century. Cambridge, 1975.
Wilson, A., ' L o n g distance trade and the Luba Lomani empire', J. Afr. Hist.,
1972, 13,4, 575-89.
Wilson, M. and Thompson, L. M. (eds.), History of South Africa {see under
General).
Ylvasaker, M., Lamu in the nineteenth century: land, trade and politics. Boston, 1979.
831

BIBLIOGRAPHY
Yoder, J. C , 'Fly and elephant parties: political polarisation in Dahomey
1840-1870',/. Afr. Hist., 1974, 15, 3, 4I7-3 2 -
Yusuf Fadl Hasan (ed.), Sudan in Africa. Khartoum, 1971.
2. THE EUROPEAN PARTITION OF AFRICA
Abrams, L. and Miller, D. J., 'Who were the French Colonialists? A

reassessment of the Parti Colonial', Hist. / . , 1976, 19, 3, 685-725.
Ageron, C. R.,' Gambetta et la reprise de l'expansion coloniale', Revuefranfaise
d'bistoire d'outremer, 1972, 59, 165—204.
Andrew, C. M., Tbeophile Delcassi and the making of the Entente Cordiale. London,
1968.
'The French colonialist movement during the Third Republic', Transactions
of the Royal Historical Society, 1976, 5th series, 26, 143-66.
Andrew, C. M., Grupp, P. and Kanya-Forstner, A. S.,' Le mouvement colonial
francais et ses principals personnalites, 1890-1914', Revue franfaise
d'bistoire d'outremer, 1975, 62, 640—73.
Andrew, C. M. and Kanya-Forstner, A. S.,' The French " Colonial Party "; its
composition, aims and influence, 1885—1914', Hist. / . , 1971, 14, 1,
99—128.
'The Groupe Colonial in the French Chamber of Deputies, 1892—1932', Hist.
/•> »974. 17. 4, 837-66.
'Gabriel Hanotaux, the Colonial Party and the Fashoda strategy', J. Imp.
C'wealtb Hist., 1974, 3, 1, 55-104. Also in E. F. Penrose (ed.), European
imperialism... {see below).
' French business and the French colonialists', Hist. / . , 1976,19,4,981-1000.
Anstey, R., Britain and the Congo in the nineteenth century. Oxford, 1962.
Ascherson, N., The King incorporated: Leopold the Second in the age of trusts.
London, 1963.
Atger, P., La France en CSte d'lvoire de 184J a /Spj. Dakar, 1962.
Atmore, A. and Marks, S., 'The imperial factor in South Africa in the
nineteenth century: towards a reassessment',/. Imp. Cwealth Hist., 1974,
3, 1, 105-39. Also in E. F. Penrose (ed.), European Imperialism... {see
below).
Axelson, E., Portugal and the scramble for Africa. Johannesburg, 1967.
Aydelotte, W. O., Bismarck and British colonial policy: the problem of South West
Africa. Philadelphia, 1937; reprinted Wcstport, Conn., 1970.
' The first German colony and its diplomatic consequences', Cambridge Hist.
]-, «937. 5. 3. *9'-3>3-
Baumgart, W., Der Imperialisms. Idee und Wirklichkeit der englischen und
fran^psiscben Kolonialexpansion, 1880-1914. Wiesbaden, 1975.
English translation: Imperialism: The idea and reality of British and French
colonial expansion, 1880—if 14. Oxford, 1982.
Bayer, T. A., England und der neue Kurs, 1890-189}. Tubingen, 1955.
Berge, F., Le Sous-Secretariat etlesSous-Secretaires d'Etataux Colonies. Paris, 1962.
Bidiss, M. D . , 'Racial ideas and the politics of prejudice, 1850-1914', Hist. ].,
'97*, 15. 3. 570-82-
832

Blainey, G., 'Lost causes of the Jameson Raid', Econ. Hist. Review, 1965, 2nd
series, 18, 350-66.
Brunschwig, H., 'Bulletins historiques' and review articles on imperialism and
colonisation in Africa, Revue Historique: 1957 (no. 441); 1961 (460); 1962
(461); 1965 (475); 1966 (480); 1968 (485); 1968 (486); 1972 (501); 197}
(5°5); '975 O 0 8 ); >976 (519); '977 ( S M ) ; 19 80 (5 55); «9 82 (54«)-
Mytbes et realites de I'imperialism! colonial franfais. Paris, i960. English trans:
Frerub colonisalism: myths and realities. London, 1966.
L'averement de I'Afrique noire. Paris, 1963.
'Les origines du partage de l'Afrique Occidentale', / . Afr. Hist., 1964, 5,
1, 121-5.
' La negotiation du traite Makoko', Cabiersd'Etudes Africaines, 1965, 5, 5-56.
Also in: J. Glenisson, J. Vansina and A. Walker (ed.), Braâ explorateur,
les traites Makoko, 1880-1882. Paris and The Hague, 1972.
'Politique et economie dans l'Empire francais d'Afrique noire, 1870-1914',
J. Afr. Hist., 1970, 11, 3, 401-17.
Le partage de l'Afrique noire. Paris, 1971.
' Anglophobia and French African policy', in P. Gifford and W. R. Louis,
Frame and Britain... (1971), {see under General).
'Conclusion' [to part I, 'Imperial rivalry'], in P. Gifford and W. R. Louis,
France and Britain... (see under General).
'Scramble et "course au clocher"', / . Afr. Hist., 1971, 12, 1, 139-41.
Bury, J. P. T., 'Gambetta and overseas problems', Eng. Hist. Review, 1967, 82,
323, 277-95.
Butler, J.,' The German factor in Anglo-Transvaal relations', in P. Gifford and
W. R. Louis, Britain and Germany.. .(1967) {see under General).
Cain, P. J. and Hopkins, A. G., 'The political economy of British expansion
overseas", Econ. Hist. Review, 1980, 2nd series, 33, 4, 463-90.
Caplan, G. L.,'Barotseland's scramble for protection',/. Afr. Hist., 1969,10,
Caulk, R. A., ' The origins and development of the foreign policy of Menelik
II, 1865-1896'. Ph.D. thesis, London, 1966.
'Yohannes IV, the Mahdists and the colonial partition of north-east Africa',
Transafrican J. Hist., 1971, 1, 2, 23-42.
Cecil, (Lady) G., Life of Robert, Marquis of Salisbury, vols. in and iv. London,
'93>. «93 2 -
Ceulemans, P., La question Arabe et le Congo, 188$-1892. Brussels, 1959.
Chamberlain, M. E.,' Clement Hill's memoranda and the British interest in East
Africa', Eng. Hist. Review, 1972, 87, 344, 533—47.
Clarence-Smith, W. G., 'The myth of uneconomic imperialism; The Portu-
guese in Africa, 1836—1926', J. Southern Afr. Studies, 1979, 5, 165—
Collins, R. O., 'Origins of the Nile struggle: Anglo-German negotiations and
the Mackinnon Agreement of 1890' in P. Gifford, and W. R. Louis,
Britain and Germany... (1967), {see under General).
King Leopold, England and the Upper Nile, 1890—1909. New Haven and
London, 1968.
Conjo, Y., 'Le "plan Freycinet", 1878-1882: un aspect de la "grande
833

BIBLIOGRAPHY
depression" economique en France', Revue historique, 1972, 248, 503,
49-86.
Conti Rossini, C , Italia ed Etiopia dal Trattato d'Uccialli alia Battaglia di Adua.
Rome, 1935.
Coquery-Vidrovitch, C , 'Le blocus de Whydah (1876-1877) et la rivalite
franco-anglaise au Dahomey', Cahiersd'etudesafricaines, 1962, 2, 373—419.
Bra?ga... (see under ch. 1).
Coupland, R., East Africa and its invaders. Oxford, 1938.
The exploitation of East Africa, i8;6-i8fo. London, 1939; reprinted, with
Introduction by J. Simmons, London, 1968.
Crowder, M. (ed.), West African resistance: the military response to colonial
occupation. London, 1971.
Crowe, S. E., The Berlin West African Conference. London, 1942.
Dachs, A. J., 'Missionary imperialism - the case of Bechuanaland', J. Afr.
Mist., 1972, 13, 4, 647-58.
Darcy, J., France et Angleterre: cent anne'es de rivalite coloniale. Paris, 1903.
Davenport, T. R. H., The Afrikaner Bond: the history of a South African political
party, 1880—if 11. Cape Town, 1966.
De Kiewiet, C. W., Imperial factor (see under ch. 1).
Denoon, D., 'Capital and capitalists in the Transvaal in the 1890s and 1900s',
Hist. J., 1980, 23, 1, 111-32.
Deschamps, H., Histoire de Madagascar (see under ch. 1).
Dike, K. O., Trade and politics (see under ch. 1).
Douin, G., Regne du Khedive Ismail (see under ch. 1).
Eldridge, C. C , England's Mission: the imperial idea in the age of Gladstone and
Disraeli 1868-1880. London, 1973.
Eley, G., 'Defining social imperialism: use and abuse of an idea', Social Hist.,
1976, 13, 265—90.
Etherington, N., ' Frederic Elton and the South African factor in the making
of Britain's East African empire',7. Imp. Cwealth. Hist., 1981,9, 3, 256-74.
'Theories of Imperialism in Southern Africa re-visited', African Affairs,
1982, 81, 324, 385-407.
Faber, R., The vision and the need: late Victorian imperial Aims. London, 1956.
Feis, H., Europe: The world's banker, 1870—1914. New Haven, 1930; reprinted
New York, 1965.
Fieldhouse, D. K., ' " I m p e r i a l i s m " - a n historiographical revision', Econ.
Hist. Review, 1961-2, 2nd series 14, 2, 187-209.
The theory of capitalist imperialism. London, 1967.
Economics and empire. London, 1973.
Flint, J. E., Sir George Goldie and the making of Nigeria. London, i960.
' The Wider Background to Partition and Colonial Occupation', in R. Oliver
and G. Mathew, History of East Africa (1963), (see under General).
Cecil Rhodes. London, 1976.
Gabre-Sellassic, Z., Yobannes IV of Ethiopia. Oxford, 1975.
Galbraith, J. S., Mackinnon and East Africa, 1878-yj. Cambridge, 1972.
'Cecil Rhodes and his "cosmic dreams": a reassessment', / . Imp. Cwealth.
Hist., 1973, 1, 2, 173-89.
834

2 T H E E U R O P E A N P A R T I T I O N OF A F R I C A
Gallagher, J. and Robinson, R., ' T h e imperialism of Free Trade', Econ. Hist.
Review, 1953, 2nd series, 6, 1, 1—15.
Ganiage, J., Origines du protectorat {see under ch. 1).
U expansion colonial de la France sous la Troiseme Ripubliquc. Paris, 1968.
'France, England and the Tunisian affair' in P. Gifford and W. R. Louis,
France and England... {see under General).
' U n episode du partage de l'Afriquc: les affaires du bas Niger (1894-1898),
Revue Historique, 1975, 254, 515, 149-88.
Gann, L. H. and Duignan, P. (eds.), Colonialism in Africa, 1 {see under General).
' Reflections on imperialism and the scramble for Africa', in L. H. Gann and
P. Duignan, Colonialism in Africa, 1 {see under General).
Gertzel, C. J., 'Relations between African and European Traders in the Niger
D e l t a ' , / . Afr. Hist., 1962, 3, 2, 361-66.
Gifford, P. and Louis, W. R. (eds.), Britain and Germany {see under General).
(eds.), France and Britain {see under General).
Giglio, C , UItalia in Africa: Etiopia-Mar Rosso, 5 vols. Rome, 1958-72.
lnventario... {see under Archives).
Giglio C. and Lodolini, E., Guida delle Fonti... {see under Archives).
Girardet, R., Uide'e coloniale en France, 1871—1962. Paris, 1972.
Goodfellow, C. F., Great Britain and South African Confederation. Cape Town,
1966.
Graham, G. S., Great Britain in the Indian Ocean, 1810-18jo. Oxford, 1967.
Grenville, J. A. S., Lord Salisbury and foreign policy. London, 1964.
Grupp, P., Theorie der Kolonialexpansion und Methoden der imperialistschen Aussen-
politik bei Gabriel Hanotaux. Berne and Frankfurt/Main, 1972.
Guillen, P . , ' Milieux d'affaires et imperialisme colonial', Relations Internationales,
•974, 1, 57-69-
Hammond, R. J., Portugal and Africa, i8ij-ipio, a study in uneconomic imperialism.
Stanford, Calif., 1966.
Hanna, A. J., The beginnings of Nyasaland and North-Eastern Rhodesia. Oxford,
1956.
Harcourt, F., 'Disraeli's imperialism, 1866-1868: a question of timing', Hist.
J., 1980, 23, 1, 87-109.
Hargreaves, J. D., 'Entente manquie: Anglo-French relations 1895-1896',
Cambridge Hist. ]., 1953, 11, 1, 65-92.
'Towards a history of the partition of Africa', J. Afr. Hist., i960, 1, 1,
97-109.
Prelude... {see under ch. 1).
' The Tokolor empire of Segou and its relations with the French', Boston
University Papers on Africa, 1966, 2, 125-45.
'British and French imperialism in West Africa', in P. Gifford and W.
R. Louis, France and Britain... {see under General).
West Africa partitioned, vol. 1 {see under ch. 1).
Headrick, D. R., The tools of empire: technology and European imperialism in the
nineteenth century. New York and Oxford, 1981.
Helly, D., ' " I n f o r m e d " opinion on tropical Africa in Great Britain,
1860-1890', African Affairs, 1969, 68, 272, 195-217.
835

BIBLIOGRAPHY
Hertslet, E., The Map of Africa by treaty. 3rd edn, revised R. W. Bryant and
H. L. Sherwood, 3 vols., London, 1909; reprinted London, 1967.
Hess, R. L., Italian colonialism in Somalia. Chicago and London, 1966.
'Germany and the Anglo-Italian colonial entente' in P. Gifford and
W. R. Louis, Britain and Germany... (see under General).
Hieke, E., G. L. Gaiser: Hamburg-Westafrika. Hamburg, 1949.
Hill, R. L., Egypt in the Sudan (see under ch. 1).
Hirshfield, C , The diplomacy of partition: France, Britain and the creation of Nigeria,
1890-1898. The Hague, 1979.
Hobson, J. A., Imperialism - a study. London, 1902.
Hopkins, A. G., 'Economic imperialism in West Africa: the case of Lagos,
1880-92', Econ. Hist. Review, 1968, 2nd series, 21, 580-606; see also
"'Economic Imperialism in West Africa": a rejoinder', ibid., 25, 1972,
307-12.
An economic history (see under ch. 1).
Hornik, M. P., Der Kampf der Grossmdchte um den Oberlau des Nil. Vienna, n.d.
[?i939].
Hyam, R., 'The partition of Africa', Hist. J., 1964, 7, 1, 154-69.
Britain's imperial century, iSrj-1914. London, 1976.
Hynes, W. G., 'British mercantile attitudes towards imperial expansion', Hist.
J., 1976, 19, 4, 969-79.
The economics of empire. London, 1979.
Jeeves, A. H., 'The Rand capitalists and the coming of the South African war,
1896—1899', Canadian Historical Papers, 1973, 61—83.
Jesman, C , The Russians in Ethiopia. London, 1958.
Jones, G. I., Trading states (see under ch. 1).
Kanya-Forstner, A. S., Conquest of Western Sudan (see under ch. 1).
' French African policy and the Anglo-French Agreement of 5 August 1890',
Hist. J., 1969, 12, 4, 628—50.
'Myths and realities of African resistance', in Canadian Historical Associa-
tion, Historical Papers 1969, 185-98.
'Military expansion in the Western Sudan - French and British style', in
P. Gifford and W. R. Louis, France and Britain... (see under General).
'French expansion in Africa - the mythical theory', in R.Owen and
B. Sutcliffe, Studies... (see below).
Kennedy, P. M., 'German colonial expansion: Has the "manipulated social
imperialism" been ante-dated?', Past and Present, 1972, 54, 134-41.
'The theory and practice of imperialism', Hist. ]., 1977, 20, 3, 761-69.
The rise of the Anglo—German antagonism. London, 1980.
Koss, S., ' Wesleyanism and empire', Hist. ]., 1975, 18, 1, 105-18.
Kubicek, R. V., Economic imperialism in theory and practice: the case of South
African gold-miningfinance,1886—1914. Durham, N.C., 1979.
Landes, D . S., Bankers and pashas (see under ch. 1).
' Some thoughts on the nature of economic imperialism',]. Econ. Hist., 1961,
21, 496—512.
Langer, W. L., The diplomacy of imperialism. 2nd edn, New York, 1951.
Laurence, P., The life of John Xavier Merriman. London, 1950.
836

Law, R. C. C , 'Imperialism and partition', J. Afr. Hist., 1983, 24, 1, 101-4.
Le May, G. H. L., British supremacy in South Africa, 1899-1907. Oxford, 1965.
Lenin, V. I., Imperialism, the highest stage of capitalism, (first published Petrograd,
19I7)-
Lloyd, T., 'Africa and Hobson's imperialism', fast and Present, 1972, 55,
150-53-
Lockhart, J. G. and Woodhouse, C. M., Rhodes. London, 1963.
Lonsdale, J. M., 'The politics of conquest: The British in Western Kenya,
1894-1908', Hist. J., 1977, 20, 4, 841-70.
Louis, W. R., 'Sir Percy Anderson's grand African strategy, 1883-1893', Eng.
Hist. Review, 1966, 8i, 319, 292-314.
'Great Britain and German expansion in Africa, 1884-1919', in P. Gifford
and W. R. Louis, Britain and Germany... (see under General).
"The Berlin Congo Conference' in P. Gifford and W. R. Louis, France
and Britain... (see under General),
(ed.), Imperialism: the Robinson and Gallagher controversy. New York and
London, 1976.
Lowe, C. J., 'Anglo-Italian differences over East Africa, 1892-1895, and their
effects upon the Mediterranean entente', Eng. Hist. Review, 1966, 81, 319,
315-36.
The reluctant imperialists, 2 vols. London, 1967.
McGowan, W., 'Fula resistance to French expansion into Futa Jallon,
1889-1896', / . Afr. Hist., 1981, 22, 2, 245-61.
Mclntyre, W. D., 'British policy in West Africa: the Ashanti expedition of
1873-4', Hist. ]., 1962, 5, 1, 19-46.
The imperial frontier in the tropics. London, 1967.
Macmillan, W.M., Ba»/«, Bow<w»</Br»'/0». London, 1929; reprinted Oxford, 1963.
Marais, J. S., The fall of Kruger's Republic. Oxford, 1961.
Marcus, H. G., The life and times of Menelik II. Oxford, 1975.
Marder, A. J., The anatomy of British sea-power: a history of British naval policy in
the pre-Dreadnought era, 1880—1901. New York, 1940.
Marks, S., 'Scrambling for South Africa', 7. Afr. Hist., 1982, 23, 1, 97-113.
Marseille, J.,' Les relations commerciales entre la France et son empire colonial
de 1880 a 1913', Relations international, 1976, 6.
Merritt, H. P., 'Bismarck and the first partition of East Africa', Eng. Hist.
Review, 1976, 91, 360, 585-97.
' Bismarck and the German interest in East Africa, 1884-1885', Hist. J., 1978,
21, 1, 97-116.
Michel, M., 'La mission Marchand, 189J-1899. Paris and The Hague, 1972.
Mommsen, W. J., Imperialismustheorien: Ein Oberblick iiber die neueren Imperialis-
musinterpretationen. Gottingen, 1977.
Miiller, F. F., Deutscbland-Zarnybar-Ostafrika. Berlin (DDR), 1959.
Newbury, C. W., 'The development of French policy on the lower and upper
Niger, 1880-1898',/. Mod. Hist., 1959, 31, 1, 18-23.
Western Slave Coast (see under ch. 1).
' Victorians, republicans and the partition of West Africa',/. Afr. Hist., 1962,
3. 3. 493-5O"-
837

BIBLIOGRAPHY
' The protectionist revival in French colonial trade: the case of Senegal',
Econ. Hist. Review, 1968, 2nd series, 21, 351-48.
British policy towards West Africa: select documents 1871-1914. Oxford, 1971.
' The tariff factor in Anglo-French West African partition' in P. Gifford and
W. R. Louis, France and Britain... (see under General).
Newbury, C. W. and Kanya-Forstner, A. S., 'French policy and the origins of
the scramble for West Africa', J. Afr. Hist., 1969, 10, 2, 253-76.
Newitt, M. D . D., Portugal in Africa, the last hundredyears. London, 1981.
Nicholls, C. S., Swahi/i Coast (see under ch. 1).
Nwoye, R. E., The public image of Pierre Savorgnan de Braâ and the establishment
of French imperialism in the Congo. Aberdeen, 1981.
Obichere, B. L., West African states and European expansion: the Dahomey—Niger
hinterland, 188j—1898. New Haven and London, 1971.
' The African factor in the establishment of French authority in West Africa,
1880-1900', in P. Gifford and W. R. Louis, France and Britain... (see under
General).
Oliver, R., The missionary factor (see under ch. 1).
Sir Harry Johnston (see under ch. 1).
Oliver, R. and Mathew, G., History of East Africa, 1 (see under General).
Oloruntimehin, B. O., Tukulor Empire (see under ch. 1).
Owen, R. and Sutcliffe, B. (eds.), Studies in the theory of imperialism. London,
1972.
Parsons, J. W., 'France and the Egyptian question, 1875-1894'. Ph.D.
Cambridge, 1977.
Penrose, E. F. (ed.), European imperialism and thepartition of Africa. London, 197 5.
Perham, M., Lugard, 1: Tie years of adventure 1886-1898; n: The years of authority
1898—194;. London, 1956, i960.
Persell, S. M., 'Joseph Chailley-Bert and the importance of the Union Coloniale
franfaise', Hist. J., 1974, 17, 1, 176-84.
Person, Y., Samori (see under ch. 1).
'French military imperialism',/. Afr. Hist., 1972, 13, 3, 507-10.
Platt, D. C. M., Finance, trade and politics in British foreign policy, 180-1914.
Oxford, 1968.
' Economic factors in British policy during the " New Imperialism "', Past
and Present, 1968, 39, 121-38.
'The imperialism of free trade: some reservations', Econ. Hist. Review, 1968,
2nd series, 21, 296-306.
'Further objections to an "imperialism of free trade"', 1830-1860', Econ.
Hist. Review, 1973, 2nd series, 26, 77-91.
Pogge von Strandmann, H.,' Domestic origins of Germany's colonial expansion
under Bismarck', Past and Present, 1969, 42, 140-59.
' The German role in Africa and German imperialism', African Affairs, 1970,
69, 277, 381-89.
Porter, A. N., 'Lord Salisbury, Mr Chamberlain and South Africa, 1895-9',
J. Imp. C'wealtb Hist., 1972, 1, 1, 3-26.
The origins of the South African War: Joseph Chamberlain and the diplomacy of
imperialism, 189/—1899. Manchester, 1980.
838

Porter, B., Critics of empire. London, 1968.

The lion's share: a short history of British imperialism, iSjo-1970. London, 1975.
'Imperialism and the scramble', J. Imp. Cwealth Hist., 1980, 9, 1, 76-81.
Price, R., An imperial war and the British working class. London, 1972.
Public Record Office, List of Cabinet papers (see under Archives).
Rabut, E., 'Le dessein colonial des milieux d'affaires implantes en Afrique
noire', in fitudes africaines offertes a Henri Brunschwig. Paris, 1982.
Ramm, A. (ed.), The political correspondence of Mr Gladstone and Lord Granville,
1876-1886, 2 vols. Oxford, 1962.
'Great Britain and France in Egypt, 1876-1882', in P. Gifford and W.
R. Louis, Trance and Britain... (see under General).
Ranger, T. O., 'Connexions between "primary resistance" movements and
modern mass nationalism in East and Central Africa', / . Afr. Hist., 1968,
9» 3. 437-5 3; 4. 631-41.
' African reaction and resistance to the imposition of colonial rule in East
and Central Africa', in L. H. Gann and P. Duignan (eds.), Colonialism in
Africa, I (see under General).
Ranieri, L., LesrelationsentreVEtat IndependantduCongoetrItalic. Brussels, 1959.
Ratcliffe, B. M., 'Commerce and empire: Manchester merchants and West
Africa, 1873-1895', / . Imp. C'wealth Hist., 1979, 7, 3, 293-320.
Reade, W. Winwood, The martyrdom of man. London, 1872.
Roberts, T. W., 'Railway imperialism and French advances towards Lake
Chad, 1890-1900', Ph.D. Cambridge, 1973.
Robinson, R., 'Non-European foundations of European imperialism: sketch
for a theory of collaboration', in R. Owen and B. Sutcliffe, Studies (see
above)'.
Robinson, R. and Gallagher, J., Africa and the Victorians (see under General).
' The partition of Africa', in The new Cambridge modern history, xi. Cambridge,
1962.
Roeykens, A., Leopold 11 et T Afrique, I8JJ-I88O. Brussels, 1958.
Rubenson, S., Wichale XVII: the attempt to establish a protectorate over Ethiopia.
Addis Ababa, 1964.
Survival... (see under ch. 1).
Sanderson, G. N . , ' The European Powers and the Sudan in the later nineteenth
century', Sudan Notes and Records, 1959, 40, 79-100.
'The Anglo-German agreement of 1890 and the Upper Nile', Eng. Hist.
Review, 1963, 78, 306, 49—72.
'The foreign policy of the Negus Menelik, 1896-1898',/. Afr. Hist., 1964,
5, 1, 87-97.
'England, Italy, the Nile Valley and the European balance, 1890-91', Hist.
J; I9 6 4, 7. »> 94-H9-
England, Europe and the Upper Nile, 1882-1899. Edinburgh, 1965.
'Conflict and co-operation between Ethiopia and the Mahdist State,
1884-98', Sudan Notes and Records, 1969, 50, 15-40.
' The origins and significance of the Anglo-French confrontation at Fashoda,
1898' in P. Gifford and W. R. Louis, France and Britain... (see under
General).
839

BIBLIOGRAPHY
'The European partition of Africa: coincidence or conjuncture?', J. Imp.
Owealth. His/., 1974, 3, 1, 1-54; a 'so in Penrose, E. F. (ed.), European
imperialism... (see above).
Schmidt, M. E., 'Prelude to intervention: Madagascar and the failure of
Anglo-French diplomacy, 1890-95', Hist. J., 1972, 15, 4, 715-30.
Schnapper, B., ' La politique et le commerce francais dans le Golfe de Guinee
de 1838 a 1871. Paris, 1961.
' La Politique des " points d 'appui" et la fondation des comptoirs fortifies
dans le golfe de Guinee (1837-1843)', Revue historique, 1961, 225, 457,
99-120.
Scholch, A., Agypten den Agyptem!. Zurich and Freiburg i. Br., n.d. [71972];
English trans., Egypt for the Egyptians! London, 1981.
'The "Men on the spot" and the English occupation of Egypt in 1882',
Hist. J., 1976, 19, 3, 773-85-
Schreuder, D. M., Gladstone and Kruger, Liberal government and colonial 'Home
Rule', 1880-188j. London, 1969.
The scramble for Southern Africa, 18jy-i89 j . Cambridge, 1980.
Schroder, H. O , SotîalismusundImperialisms: die Auseinanderset\ungderdeutschen
Soîaldemokratie mit dem Imperialismusproblem und der' Weltpolitik' vor 1914.
Hanover, 1968.
Semmel, B., Imperialism and social reform: English social thought, 189/-1P14.
London,i960.
Shepperson, G., 'Africa, the Victorians and imperialism', Revue Beige de
philologie et d'histoire, 1962, 40, 4, 1228-38.
Shukry, M. F. (Muhammad Fu'ad Shukrl), Equatoria under Egyptian rule. Cairo,
'953-
Sieberg, H., Eugene Etienne und die franypsische Kolonialpolitik. Cologne, 1968.
Smith, I. R., The Emin Pasha relief expedition, 1886—90. Oxford, 1972.
Stengers, J., ' Aux origines de Fachoda: I'expedition Monteil', Revue Beige de
philologie et d'histoire, 1958, 36, 436-50; i960, 38, 366-404 and 1040-1065.
' L'imperialisme colonial de la fin du XIXe siecle: mythe ou realite?', / . Afr.
Hist., 1962, 3, 3, 469-91.
'Leopold II et la fixation des frontieres du Congo', L* Flambeau, Brussels,
1963 (Mar.-Apr.), 153-97-
'British and German imperial rivalry: a conclusion', in P. Gifford and
W. R. Louis, Britain and Germany... (see under General).
'Une facette de la question du Haut-Nil: le mirage soudanais',_/. Afr. Hist.,
1969, 10, 4, 599—622.
'King Leopold and Anglo-French rivalry, 1882-1884', in P. GifFord and
W. R. Louts, France and Britain... (see under General).
' King Leopold's imperialism', in R. Owen and B. Sutcliffe, Studies (see above).
'Leopold II et Brazza en 1882. Documents inedits', Revuefranfaise d'histoire
d'outremer, 1976, 63, 105-36.
' Le Katanga et le mirage de Tor', in Etudes africaines offertes a Henri Brunschwig,
Paris, 1982.
Stoecker, H., Drang nach Afrika. Die koloniale Expansionspolitik und Herrscbaft
des deutschen Imperialismus in Africa... Berlin (D.D.R.), 1977.
840

3 NORTH AFRICA
Stokes, E. T., 'Milnerism', Hist. J., 1962, 5, 1, 47~6°-
' Late Nineteenth-Century colonial expansion and the attack on the theory
of economic imperialism: a case of mistaken identity', Hist. ]., 1969, 12,
2, 285-501.
'Uneconomic imperialism', Hist. ]., 1975, 18, 2, 409-16.
Swinburne, J. S., 'The influence of Madagascar on international relations,
1878-1904'. M.Phil. London, 1969.
Taylor, A. J. P., Germany's first bid for colonies, 1884-188}. London, 1958 and
Hamden, Conn, 1967.
Thompson, L. M., 'Great Britain and the Afrikaner republics, 1870-1899', in
M. Wilson and L. M. Thompson, History of South Africa, 11 (see under
General).
Thornton, A. P., Doctrines of imperialism. New York, 1965.
Turner, H. A., 'Bismarck's imperialist venture: anti-British in origin?' in
P. Gifford and W. R. Louis, Britain and Germany... (see under General).
Uzoigwe, G. N., Britain and the conquest of Africa: the age of Salisbury: Ann Arbor,
Mich., 1974.
Van der Poel, J., The Jameson Raid. Cape Town, 1951.
Van-Helten, J. J.,' Empire and high finance: South Africa and the international
gold standard, 1890-1914', J. Afr. Hist., 1982, 23, 4, 529-48.
Vignes, K..,' Etude sur la rivalite d'influence entre les puissances europeennes
en Afrique equatoriale et occidentale depuis l'acte general de Berlin,
jusqu'au seuil du XXc siccle', Revue francaise d'bistoire d'outremer, 1961, 48,
S—95-
Waller, R. M., ' Anglo—French diplomacy overseas, 1835—45, with special
reference to West Africa and the Indian Ocean'. Ph.D. London, 1979.
Warhurst, P. R., Anglo-Portuguese relations in south-central Africa, 1890-1900.
London, 1962.
Washausen, H., Kolonialpolitik des deutschen Reiches 1880 bis 1898. Hamburg,
1968.
Wehler, H-U., Bismarck und der Imperialisms. Cologne and Berlin, 1969.
'Bismarck's imperialism, 1862-1890', Past and Present, 1970, 48, 119-55.
'Industrial growth and early German imperialism' in R. Owen and B.
Sutcliffe, Studies... (see above).
Wesseling, H. L. (ed.), Expansion and reaction. Leiden, 1979.
'The Netherlands and the partition of Africa', / . Afr. Hist., 1981, 22, 4,
495-509.
Zaghi, C, / Russi in Etiopia, 2 vols. Naples, 1975.
}. NORTH AFRICA
Adam, A., Casablanca: Essai sur les transformations de la societe marocaine, 2 vols.
Paris, 1968.
Ageron, R.,' La France a-t-elle eu une politique kabyle?' Revue historique, i960,
311-52.
'Jules Ferry et la question algerienne en 1892', Revue a" hist. mod. et contemp.,
1961, 127-46.
841

BIBLIOGRAPHY
Les Algeriens musulmans (see under ch. i).
Histoire de r Algerie contemporaine, n (i 871—1954). Paris, 1979-
Albertini, E., Marcais, G. and Yver, G., UAfrique du Norddans I'histoire. Paris,
i9}9.
Ansky, M., Les Juifs d'Algerie (iS71-194)). Paris, 1950.
Aubin [Descos], E., Le Maroc d'aujourd'bui. Paris, 1905.
Aumont-Thieville, P., Du regime de I'indigenat en Algerie. Paris, 1906.
Bardin, P.,'Les debuts difficiles du protectorat tunisien (mai 1881-avril 1882)',
Revue d'bist. diplom., 1971, 1—48.
Benhabiles, G, L' Algerie francaise vue par un indigene. Algiers, 1914.
Bernard, A., L'Algerie. Paris, 1930. (Histoire des colonies francaises, vol. 11.)
Betts, R., Assimilation and association in French colonial theory (1890-1911). New
York, 1961.
Brignon, J. and Amine, A., Histoire de Maroc. Paris, 1947.
Brunschwig, H., Mythes et realties (see under ch. 2).
Cambon, H., Histoire de la Regence de Tunis. Paris, 1948.
Histoire du Maroc. Paris, 1952.
Chalon, P., Les richesses minerales de VAlgerie et de la Tunisie Paris, 1907.
Chateur, K., Insurrection et repression dans la Tunisie du XIX" siecle. La mehalla
de Zarrouk. Tunis, 1978.
Chouraqui, A., Les Juifs d'Afrique du Nord. Paris, 1959.
Coppolani, X. and Depont, O., Les confreries religieuses musulmanes. Algiers, 1897.
Demontes, V., L' Algerie e'conomique, 5 vols. Algiers, 1922-30.
L'Algerie agricole. Paris, 1930.
L''Algerie industrielte et commercante. Paris, 1930.
Deschamps, H., Mithodes et doctrines colontales de la France. Paris, 1953.
Despois, J., L'Afrique du Nord. Paris, 1949.
Deverdun, G., Marrakecb, des origines a 1912. Rabat, 1959.
Doutte, E., L'Islam algerien en 1900. Algiers, 1900.
Doutte, E. and Gautier, E., Enquete sur la dispersion de la langue berbere en Algerie.
Algiers, 1913.
Dupuy, H., Les impSts arabes en Algerie. Montpellier, 1910.
Ficaya, P., Le peuplement italien en Tunisie. Paris, 1931.
Franc, J., Le cbefd'oeuvre colonial de la France, la colonisation de la Mitidja. Algiers,
1928.
Ganiage, J., 'Une affaire tunisienne, 1'affaire de FEnfida', Revuea/ricaine, 1955,
341-78.
Une entreprise italienne de Tunisie au milieu du XIXe siecle. La thonaire de Sidi
Daoud. Tunis, i960.
La population europe'enne de Tunis au milieu du XIXe siecle. Tunis, i960.
'La population de la Tunisie vers i860. Essai devaluation d'apres les
registres fiscaux', Etudes magbrebines, 1964, 115-98.
Origines du protectorat (see under ch. 1).
L'expansion coloniale (see under ch. 2).
'The Tunisian affair' (see under General).
L'expansion coloniale et les rivalite's internationales (1871-1914), 2 vols. Paris,
•974-5-
842

3 NORTH AFRICA
Ganiage, J. and Deschamps, H., L'Afrique au XXe siecle {1900-1966). Paris,
1966.
Girault, A., Principes de colonisation et de legislation toloniale, 5 vols. Paris, 1927-9.
Giaccardi, A., La conquista di Tunisi. Milan, 1940.
Gsell, S., Marcais, G. & Yver, G., Histoire de I'Algerie. Paris, 1927.
Guillen, P.,' L'implantation de Schneider au Maroc'. Revue d'hist. diplom., 1965,
113-68.
L'Allemagne et le Maroc de 1870 a if or. Paris, 1967.
Les emprunts marocains de 1902 et 1904. Paris, 1969.
' The entente of 1904 as a colonial settlement', in P. Gifford and W. R. Louis,
France and Britain... {see under General).
Hajoui, M., Histoire diplomatique du Maroc {1900-1912). Paris, 1937.
Hardy, G., 'Le Maroc (Histoire des colonies franchises, vol. 11). Paris, 1931.
La Tunisie (Histoire des colonies francaises, vol. ti). Paris, 1931.
Harris, W., Morocco that was. London, 1921.
Isnard, H., La vigne en Algerie. Etude geographique, 2 vols. Gap, 1951—4.
Julien, Ch.-A., Afrique du Nord {see under General).
Kraiem, M., La Tunisie precoloniale. Tunis, 1973-
[Lagrange] Etude sur la colonisation officielle en Tunisie. Tunis, 1919.
Lahbabi, M., Legouvernement marocain a I'aube du XX' siecle. Rabat, 1957.
Lanessan, J. L. de, La Tunisie, 2nd edn. Paris, 1917.
Langer, W., 'The European powers and the French occupation of Tunis',
Amer. Hist. Review, 1925, 55-78 and 251-65.
Leone, E., Co/oniâîone {see under ch. 1).
Leroy-Beaulieu, P., U Algerie et la Tunisie. 2nd edn, Paris, 1897.
Le Tourneau, R., Fes avant le protectorat. Casablanca, 1949.
Loth, G., Le peuplement italien en Tunisie et en Algerie. Paris, 1905.
Mabrouk, M., Administration et personnel administratif de la Tunisie beylicale. Tunis,
1971.
Mahjoubi, A., L'e'tablissement du protectoratfranfais en Tunisie {1SS1—1886). Tunis,
1977-
Marsden, A., Diplomacy and Tunis {see under ch. 1).
Martel, A., Les confins sahariens de la Tunisie {1881—1911). Paris, 1965.
Martin, C , Les Israelites alge'riens. Paris, 1936.
Histoire de /'Algerie francaise {1830—1962). Paris, 1963.
Masmoudi, A. and Smida, M., Histoire generate de la Tunisie, iv: Les temps
modernes. Tunis, 1967.
Miege, J. L., Maroc et Europe {see under ch. 1).
Miege, J. L. and Hugues,E., LesEurope'ensd CasablancaauXIXesiecle. Paris, 1954.
Montagne, R., lês Berberes et le Makh\en dans le sud du Maroc. Paris, 1930.
Murphy, A., The ideology of French imperialism. Washington, 1948.
Nouschi, A., Enquite sur le niveau de vie des populations rurales constantinoises de la
conquetejusqtl'en 1919. Paris, 1961.
O'Donnell, D., Lavigerie in Tunisia. The interplay of imperialist and missionary.
Athens, Ga., 1979.
P.H.X. [d'Estournelles de Constant], La politique franfaise en Tunisie. Le
protectorat et ses origines {18)4-1891). Paris, 1891.
843

BIBLIOGRAPHY
Peyerimhof, F. de, Enquete sur le resultat de la colonisation officielle en Algerie de

i8pi a 189/, i vols. Algiers, 1906.
Piquet, V., Les riformes en Algerie et le statut des indigenes. Paris, 1919.
UAlgerie franfaise. Un siecle de colonisation (18)0-19)0). Paris, 1930.
Poncet, J., La colonisation et I'agriculture europeennes de Tunisie depuis 1SS1. Paris,
1962.
Raymond, A.,' Les liberaux anglais et la question tunisienne', Cahiers de Tunisie,
'95 5, 4 « - 6 j -
Renouvin, P., La politique exterieure de Deleassi. Paris, 1962.
Rinn, L., Marabouts et khouans. Etude sur I'lslam en Algerie. Algiers, 1884.
Roberts, S. H., A history of French colonial policy (/870-190;), 2nd edn. London,
1963.
Rosenbaum, J., Frankreich in Tunisien: Die Anfdnge des Protektorats (1881-1884).
Freiburg im Breisgau, 1971.
Serra, E., La questione tunisina da Crispi a Kudini. Milan, 1967.
Smida, M., Khe'reddine, ministre reformateur (187)-1877). Tunis, 1971.
Taylor, A., 'British policy and the Morocco question 1886-1902', Eng. Hist.
Review, 1951, 34*-"73-
Tchaidje, A., La question tunisienne et la politique ottomane (1881-191)). Erzerum,
1963.
Terrasse, H., Histoire du Maroc. 2 vols. Casablanca, 1950.
Tinthoin, R., Colonisation et evolution des genres de vie dans la region ouest d'Oran,
de 18)0 a t88j. Ozan, 1947.
Valensi, L., Fellahs tunisiens. Veconomic rurale et la vie des campagnes aux XVIIIe
et XIXe siecles. Paris, 1977.
Van Krieken, G., Khe're'dine (see under ch. 1).
Vignes, K., Le gouverneur general Tirman et le systeme des rattacbements. Paris
1958.
Weisgerber, F., Au seuil du Maroc moderne. Casablanca, 1948.
Windelband, W., Die auswdrtige Politik der europdische Grossmdchte in der Neut(eit.
Essen, 1936.
Yacono, X., Les bureaux arabes et revolution des genres de vie indigenes dans I'ouest
du Tell alge'rien. Paris, 1953.
La colonisation des plaints du Chelif, 2 vols. Algiers, 1955—6.
4 A N D 5. W E S T E R N A F R I C A , 1870-1886, 1886-I905
Abraham, A., Mende government and politics under colonial rule. Freetown, 1978.
Topics in Sierra Leone history; a counter-colonial interpretation. Freetown,
1976.
Abrams, L. and Miller, D . J., 'The French colonialists?' (see under ch. 2).
Abubakar, Sa'ad, The Lamibe of Fombina; a political history of Adamaaa,
1809-190). Zaria, 1977.
Adamu, Mahdi, The Hausa factor in West African history."L^tva.,1978.
Adeleye, R. A., 'Rabih b. Fadlallah and the diplomacy of European imperial
invasion in the central Sudan, 1893—1902', J. Hist. Soc. Nigeria, 1970, 5,
}. 399~4«8.
Power and diplomacy (see under ch. 1).
844

4 / 5 WESTERN AFRICA, 1 8 7 0 - 1 9 0 5
'The Sokoto caliphate in the nineteenth century', in J. F. A. Ajayi, and
M. Crowder, History, vol. 11 (see under General).
Aderibigbe, A. A. B., 'Expansion of the Lagos Protectorate, 1863-1900'.
Ph.D. thesis, London, 1959.
'The Ijebu expedition, 1892', in Historians i»-tropical Africa. Salisbury
(Harare), 1962.
Afigbo, A. E., The warrant chiefs: Indirect rule in south-eastern Nigeria, 1891-1929.
London, 1972.
Agbodeka, F., African politics (see under ch. 1).
Ajayi, J. F. A., Christian missions (see under ch. 1).
Ajayi, J. F. A. and Austen, R. A.,' Hopkins on Economic Imperialism in West
Africa', Econ. Hist. Review, 1972, 2nd series, 25, 303-6.
Akindele, A. and Aguessy, C., Contribution a I'etude de I'histoire de I'ancien rqyaume
de Porto Novo (Memoires de FI.F.A.N.), Dakar, 1953.
Akintoye, S. A., Revolution and power politics (see under ch. 1).
Amenumey, D. E. K., 'The Ewe people and the coming of European rule,
1850-1914'. Ph.D. thesis, London, 1966.
Amselle, J.-L., Les Ne'gociants de la Savane. Histoire et organisation sociale des
Kooroko (Mali). Paris, 1977.
Andrew, C. M., 'The French colonialist movement during the Third Republic;
the unofficial mind of imperialism', Transactions Royal Hist. Soc, 5 th series
26, 1976.
Andrew C. M. and Kanya-Forstner, A. S., France overseas: the Great War and
the climax of French imperial expansion. London, 1981.
Anene, J. C , Southern Nigeria in transition, 188J-1906. Cambridge, 1966.
Arhin, K., 'The structure of Greater Ashanti', J. Afr. Hist., 1967, 8.
(ed.), The papers of George Ekem Ferguson. Leiden and Cambridge, 1974.
West African traders (see under ch. 1).
Asiwaju, A. I., Western Yorubalandunder European rule, 1889-194 j . A comparative
analysis of French and British colonialism. London, 1976.
Atanda, J. A., The new Oyo empire: indirect rule and change in western Nigeria,
1894-19)4. London, 1973.
Atger, P., La France en Cote d'lvoire (see under ch. 2).
Ayandele, E. A., The missionary impact (see under ch. 1).
'The ideological ferment in Ijebuland, 1892—1943 ', African Notes (University
of Ibadan), 1970, 5, 3.
Holy Johnson: pioneer of African nationalism, 18)6-1917. London, 1970.
Baier, S., Central Niger (see under ch. 1).
Ballard, J. A.,' The Porto Novo incidents of 1923 ; politics in the colonial era',
Odu, July 1965, 2, 1.
Balogun, S. A., 'Gwandu emirates in the nineteenth century with special
reference to political relations'. Ph.D. thesis, Ibadan, 1971.
Bathily A., 'Imperial and colonial expansion in Senegal in the nineteenth
century: with particular reference to the economic, social and political
development of the kingdom of Gajaaga (Galam)'. Ph.D. thesis,
Birmingham, 1975.
Berry, S. S., Cocoa, custom and socio-economic change in rural western Nigeria. Oxford,
'975-
845

BIBLIOGRAPHY
Blyden, E. W., 'Ethiopia stretching out her hands unto God; or, Africa's
service to the world'. Discourse delivered before the American Coloni-
zation Society, May 1880, in Christianity, Islam and the Negro race. 1887;
reprinted Edinburgh, 1967.
Bouche, D., Les villages de liberte en Afrique noire franfaise, 1887-1910. Paris and
The Hague, 1968.
L'enseignement dans les territoires franfais de I'Afrique occidentale de 1817 a 1920.
Mission civilisatrice ou formation d'une elite?, 2 vols. Lille, 1975.
Braimah, J. A. and Goody, J. R., Salaga: the struggle for power. London, 1967.
Brenner, L., The Shehus of Kukawa: a history of the al-Kanemi dynasty of Bornu.
Oxford, 1973.
Brunschwig, H., L'avenement {see under ch. 2).
'French expansion and local reactions in Black Africa in the time of
imperialism (1880—1914)', in H. L. Wesseling (ed.), Expansion and reaction
(Leiden, 1979).
Charles, E. A., PrecolonialSenegal: The Jolof Kingdom, 1800 to 1890. Boston, 1977.
Cole, P. D . , Modern and traditional elites in the politics of Lagos. Cambridge, 1975-
Coquery-Vidrovitch, C , 'Le blocus de Whydah' (see under ch. 2).
' De la traite des esclaves a ('exportation de l'huile de palme et des palmistes
au Dahomey', in Meillassoux, C. (ed.), Indigenous trade (see under ch. 1).
Cornevin, R., 'The Germans in Africa before 1918', in L. H. Gann and
P. Duignan, Colonialism in Africa (see under General).
Crowder, M., West Africa under colonial rule. London, 1968.
(ed.), West African resistance, (see under ch. 2).
Curtin, P. D. (ed.), Africa and the West: intellectual responses to European culture.
Madison, 1972.
Debrunner, H. W., A church between colonial powers: a study of the Church in Togo.
London, 1965.
De Cardi, N., 'A short description of the natives of the Niger Coast
Protectorate, with some account of their customs, religion, trade. App.
1 to Mary H. Kingsley, West African studies, 1st edn, London, 1899.
Desire-Vuillemin, G.; Contribution a I'histoire de la Mauritanie. Dakar, 1962.
DiaJlo, T., Les institutionspolitiquesau Fouta Djallon auXIXe siecle. Dakar, 1972.
Dumett, R. E., 'The rubber trade' (see under ch. 1).
Ekundare, R. O;, An economic history of Nigeria, i860—1960. London, 1973.
F a u r e , C. a n d C h a r p y , J., Repertoire des archives. Rufisque, 1955.
Fika, A. M., The Kano civil war and British over-rule, 1882-1940. Ibadan, 1978.
Fisher, H., 'The early life and pilgrimage of al-Hajj Muhammad al-Amin the
Soninke (d. 1887)',/. Afr. Hist., 1970, 11, i, 51-69.
Flint, J. E., Sir George Goldie (see under ch. 2).
Forde, C. D. and Kaberry, P. M. (ed.), West African kingdoms in the nineteenth
century. London, 1967.
Fyfe, C. H., A history of Sierra Leone. Oxford, 1962.
Ganier, G., 'Lat Dyor et le chemin de fer de l'arachide (1876-86)', Bulletin de
1'I.F.A.N., 1965, series 8, 27, 223-81.
Gann, L. H. and Duignan, P. (eds.), Colonialism in Africa (see under General).
Garcia, L. M., 'Archives et tradition orale. A propos d'une enquete sur la
846

4/5 WESTERN A F R I C A , X87O-I9O5
politique du royaume de Danhome a la fin du 19c siecle', Cahiers a"etudes
africaines, 1976, 16(61-2), 189-206.
Gbadamosi, G. O., Islam among the Yoruba (see under ch. 1).
Gertzel, C , 'Commercial organisation on the Niger coast, 1852-1891',
Historians in Tropical Africa (Leverhulme Foundation). Salisbury (Harare),
1962, 289-504.
Gifford, P. and Louis, W. R. (eds.), Britain and Germany (see under General).
France and Britain (see under General).
Grace, J. J., Domestic slavery in West Africa, with particular reference to the Sierra
Leone Protectorate 1896-1927. London, 1975.
Gray, J. M., A history of the Gambia. Cambridge, 1940.
Gueye, L., Itine'raire africain. Paris, 1966.
Guide to materials for West African history (see under Archives).
Guide to sources — Africa south of the Sahara (see under Archives).
Haliburton, G. M., The prophet Harris: a study of an African prophet and his mass
movement in the Ivory Coast and the Gold Coast, 1913-191J. London, 1971.
Hargreaves, J. D., 'The evolution of the Native Affairs Department', Sierra
Leone Studies, n.s.3, 1954.
A life of Sir Samuel Lewis. London, 1958.
' The establishment of the Sierra Leone Protectorate and the insurrection of
1898', Cambridge Hist. / . , 1956, 12.
Prelude (see under ch. 1).
'The Tokolor Empire' (see under ch. 2).
West African Partitioned, vol. 1 (see under ch. 1).
Harrell-Bond, B., Howard, A. M. and Skinner, D. E., Community leadership and
the transformation of Freetown (1801-1976). The Hague, 1978.
Hausen, K., Deutsche Kolonialherrschaft in Afrika. Wirtschaftinteressen und Kolo-
nialverwaltung in Kamerun vor 1914. Zurich, 1971.
Henige, D., Oral historiography. London, 1982.
Hill, P., The Gold Coast cocoa farmer: a preliminary survey. London, 1956.
Migrant cocoa farmers of southern Ghana. Cambridge, 1963.
Hopkins, A. G., 'The Lagos strike of 1897: an exploration in Nigerian labour
history', Past and Present, 1966, 35, 133-55.
' Economic imperialism' (see under ch. 2).
Economic history (see under ch. l ) .
'Imperial business in Africa. Part 1: Sources; Part 2: Interpretations',/. Afr.
Hist., 1976, 17, 29-48, 267-90.
Idowu, H. O., 'The Conseil General in Senegal, 1879-1920', Ph.D. thesis,
Ibadan, 1966.
Igbafe, P. A., 'Western Ibo society and its resistance to British rule: the
Ekumeku movement, 1898-1911',/. Afr. Hist., 1971, 12, 441-59.
Ijagbemi, E. A., 'The history of the Temne in the nineteenth century'. Ph.D.
thesis, Edinburgh, 1968.
'The Yoni expedition of 1887: a study of British imperial expansion in Sierre
L e o n e ' , / . Hist. Soc. Nigeria, 1974, 7, 2, 241-54.
Ikime, O., Niger Delta rivalry: Itsekiri-Urhobo relations and the European presence,
1884—19j6. London, 1969.
847

BIBLIOGRAPHY
Isichei, E., A history of the Igbo people. London, 1976.
Jeng, A. A. O., 'An economic history of the Gambia groundnut industry,
1830—1924: the evolution of an export economy'. Ph.D. thesis, Birming-
ham, 1978.
Johnson, G. W., The emergence of black politics in Senegal: the struggle for power
in the four communes, if 00—if 20. Stanford, 1971.
Johnson, M., 'Cotton imperialism in West Africa', African Affairs, 1974, 73,
178-87.
Kanya-Forstner, A. S., Conquest (see under ch. 1).
Kimble, D., A political history of Ghana, 18)0-1928. Oxford, 1963.
Klein, M. A., Islam and imperialism (see under ch. 1).
Last, D . M., The Sokoto caliphate. London, 1967.
Lewin, T. J., Asante before the British (see under ch. 1).
Low, D . A., Lion rampant: essays in the study of"British imperialism. London, 1973.
McGowan, W., 'Fula resistance' (see under ch. 2).
Mahoney, F., 'Government and opinion in the Gambia, 1816-1901', Ph.D.
thesis, London, 1963.
Manning, P., Slavery, colonialism and economic growth in Dahomey. Cambridge, 1982.
Markov, W. and Sebald, P., 'The treaty between Germany and the Sultan of
G w a n d u ' , / . Hist. Soc. Nigeria, 1967, 4, 1, 141-53.
Matthews, N. and Wainwright, M. D., Guide to manuscripts (see under Archives).
Meillassoux, C. (ed.), Indigenous trade and markets (see under ch. 1).
Meniaud, J., Lespionniers du soudan: avant, avec et apres Archinard, 2 vols. Paris,
1931.
Metcalfe, G. E., Great Britain and Ghana: Documents of Ghana history, 1867-19)7.
London, 1964.
Michel, M.,' Les plantations allemandes du mont Cameroun, 18 8 5 -1914', Revue
franfaise d'bistoire d'outre-mer, 1976, 58.
Miers, S. and Kopytoff, I. (ed.), Slavery in Africa (see under ch. 1).
Mille, P., 'The black vote in Senegal',/. Royal Afr. Soc, 1901, 1.
Monteil, V., U'Islam noir. Paris, 1964.
Esquisses Senegalaises. I.F.A.N., Initiations et etudes africaines, 21. Dakar,
1966.
Muffett, D . J. M., Concerning brave captains. London, 1964.
Newbury, C. W., British policy (see under ch. 2).
Nworah, K. D . , ' The West African operations of the British Cotton Growing
Association, 1904-1914', Afr. Hist. Studies, 1971, 4, 315-30.
Nyambarza, D., 'Le marabout El Hadj Mamadou Lamine d'apres les archives
francaises', Cabiers d'etudes africaines, 1969, 9 (33), 124-45.
Obichere, B. L., West African states (see under ch. 2).
O'Brien, D . B., The Mourides of Senegal. The political and economic organisation of
an Islamic brotherhood. Oxford, 1971.
Oloruntimehin, B. O., The Segu Tukolor empire (see under ch. 1).
Omu, F. I. A., Press and politics in Nigeria, 1880—19)7. London, 1978.
Pallinder-Law, A., 'Government in Abeokuta, 1830-1914, with special refer-
ence to the Egba United Government, 1898-1914', Ph.D. thesis,
Goteborg, 1973-
848

4/5 WESTERN AFRICA, 1870-1905
Pasquier, R., 'Villes au Senegal au XIXe siecle', Revue franfaise d'bistoire
d'outre-mer, i960, nos. 168-9, 387-426.
'Les debuts de la presse au Senegal', Cabiers d''etudes africaines, 1962, 2, 7,
477-90.
Pedler, F. J., Tbe lion and tbe unicorn in Africa: a history of tbe origins of tbe United
Africa Company, ijSy-ifji. London, 1974.
Penrose, E. F. (ed.), European imperialism {see under ch. 2).
Perham, M., Lugard (see under ch. 2).
Person, Y., Samori (see under ch. 1).
Public Record Office Handbooks (see under Archives).
Porter, A. T., Creoledom (see under ch. 1).
Quinn, Charlotte, 'A 19th century Fulbe state', J. Afr. Hist., 1971, 12.
Mandingo kingdoms (see under ch. 1).
Renault, F.,' L'abolition de 1'esclavage au Senegal: l'attitude de Padministration
francaise, 1848-1905', Revue franfaise d'bistoire d'outre-mer, 1971, 58, 5-81.
Roberts, T. W., 'Railway imperialism* (see under ch. 2).
Robinson, D., Chiefs and clerics: the history of Abdul Bokar Kan and Futa Toro,
iSjj-igfi. Oxford, 1975.
Tbe Holy War of Vmar Tal in tbe Senegal and Niger River valleys (forthcoming).
valleys (forthcoming).
Robinson, R., 'Non-European foundations' (see under ch. 2).
Roche, C , 'Portraits de chefs casamancais du XIXe siecle', Revue franfaise
d'histoire d'outre-mer, 1971, 58, 451—67.
Ross, D . A., 'The autonomous kingdom of Dahomey, 1818-94'. Ph.D. thesis,
London,1967.
Ross, D., 'Dahomey', in Crowder, M. (ed.), West African resistance (see under
ch. 2).
Rotberg, R. and Mazrui, A., Protest and power in black Africa. New York, 1970.
Saint-Martin, Y., L.'empire Toucouleur et la France; un demi-siecle de relations
diplomatiques. Universite de Dakar: Faculte des Lett res et Sciences
Humaines, 1967, 11.
Salifou, A., Le Damagaram ou Sultanat de Zinder au ipe siecle. Niamey, 1971.
Sanneh, L., Tbe Jakbanke: tbe history of an Islamic clerical people of the Senegambia.
London, 1979.
Senkomago, N. S., 'The kingdom of Porto Novo with special reference to its
external relations, 1862—1908'. Ph.D. thesis, Aberdeen, 1976.
Short, A. W., 'Continuity and change in West African foreign relations.
Problems of some stranger communities in Ghana in the nineteenth and
twentieth centuries'. Ph.D. thesis, Aberdeen, 1978.
Simensen, Jarle, 'Commoners, chiefs and colonial government; British policy
and local politics in Akim Abuakwa, Ghana under colonial rule', Ph.D.
thesis, Trondheim, 1975.
Smaldone, J., Warfare in the Sokoto caliphate: historical and sociological perspectives.
Cambridge, 1977.
Smith, R. S.,' Nigeria-Ijcbu', in Crowder, M. (cd.), West African resistance (see
under ch. 2).
849

BIBLIOGRAPHY
Smith, R., Warfare and diplomacy inpre-colonial West Africa. London, 1976.
Smith, W. D., The German colonial empire. Chapel Hill, 1978.
Spitzer, L., The Creoles of Sierra Leone: responses to colonialism, 1870-194}.
Madison, 1974.
Staniland, M., The lions of Dagbon: political change in Northern Ghana. Cambridge,
"975-
Stewart, C. C , Islam and social order in Mauritania. Oxford, 1973.
Stoecker, H. (ed.), Kamerun unter deutscher Kolonialberrschaft. Studien, Institut fur
allgemeine Geschichte, Humboldt-Universitat. E. Berlin: vol. 1, i960;
vol. n, 1968.
Suret-Canale, J., 'La Guinee dans le systeme colonial', Presence africaine,
1959-60, 29.
Afrique noire occidental et centrale, vol. 11: L'ere colonials. Paris, 1964 (English
edn, French colonialism in tropical Africa, 1900—194}. London, 1971).
Swindell, K., ' Serawoollies, Tillibunkas' (see under ch. 1).
Tamuno, T. N., The evolution of the Nigerian state: the southern phase, 1898-1914.
London, 1972.
Tasie, G. O. M., Christian missionary enterprise. Leiden, 1978.
Thomas, R. G., 'Forced labour in British West Africa: the case of the Northern
Territories of the Gold Coast, 1906-1927', / . Afr. Hist., 1973,14, 79-103.
Tordoff, W., Ashanti under the Prempehs 1888-19);. London, 1965.
Uzoigwe, G. N., Conquest of Africa (see under ch. 2).
Vietor, J. K., Die Geschichte und kulturelle Enhvicklung unserer Schut^gebieten.
Berlin, 1913.
Vincent, D. B. [Mojola Agbebi], Africa and the Gospel. Sierra Leone and Lagos,
n.d.
Webster, J. B., The African churches among the Yuruba, 188}-1922. Oxford, 1964.
Wieskel, T. C , French colonial rule and the baule peoples: resistance and collaboration,
1889-1911. Oxford, 1980.
Wilks, I., Asante (see under ch. 1).
Wirz, A., Vom SkJavenhandel %um Kolonialhandel; Wirtschaftsraume und Wirtsch-
aftsformen in Kamerun vor 1914- Zurich, 1973.
6. WESTERN EQUATORIAL AFRICA
A. French Congo and Gabon

Archives de la Marine (see under Archives).
Archives Nationales, S.O.M. (see under Archives).
Augouard, Mgr, 28anne'esauCongo: lettresdeMgr Augouard, 1 vol. Poitiers, 1905.
Avaro, J. a B., (see Bouroue-Avaro, J. a).
Balandier, G., Sociologie actuelle de /''Afrique noire [Fang et Bakongo en Afrique
equatoriale]. Paris, 196;.
Bouroue-Avaro, J. a, Unpeuplegabonais (see under ch. 1).
Bruel, G., La France equatoriale africaine: le pays, les habitants, la colonisation, les
pouvoirs puhliques. Paris, 1935.
850

Brunschwig, H., Noirs et Blancs dans I'Afrique noirefrancaise, ou comment le colonise
dcvient colonisateur (1870-1914). Paris, 1983.
Burton, R. F., Two trips to Gorilla Land and the cataracts of the Congo, 2 vols.
London, 1876.
Chavannes, Ch. de, Les origines de i'AEF: le Congo franfais. Ma collaboration avec
Bra^Z", 18i6-i894. Nos relations jusqu'a sa mort, 190 j . Paris, 1937.
Congo. Notice a I''usage des emigrants. Ministere des Colonies, Melun, 1895.
Coquery-Vidrovitch, Catherine, Braaga... (see under ch. 1).
Le Congo au temps des grandes compagnies concessionnaires 1898-19)0. Paris/The
Hague, 1972.
Deschamps, H., Traditions orales et archives au Gabon. Contribution a Tethno-histoire.
Paris, 1962.
'Quinze ans de Gabon (les debuts de l'etablissement fran9ais, 1839—1853)*,
Revuefrancaise d'histoire d'outre-mer, 180-181 (1963), 2 8 3 - 3 4 5 ; 186 (1965),
92—126.
Fox-Bourne, H. R., Civilisation in Congoland: a story of international wrong-doing.
London, 1903.
Gaulme, F., Lepays de Cama, un ancien Etat cotierdu Gabon etses origines. Paris, 1981.
Leroy-Beaulieu, P., Les grandes compagnies de colonisation. Paris, 1895.
Louis, W. R. and Stengers, J., E. D. Morel's history of the Congo Reform
Movement. London/Oxford, 1968.
Mangongo-Nzambi, A., 'La delimitation des frontieres du Gabon', Cahiers
a"etudes afrieaines, 1969, 9, 33, 5-55.
Martin, P., Loango coast (see under ch. 1).
Mazenot, G., La Likouala—Mossaka, 1878—1920. Paris/The Hague, 1971.
Mazieres, A. de,' Liotard et le haut Oubangui'. These de 3eme cycle, Universite
Paris I, 1975.
Mbokolo, E., Noirs et blancs en Afrique equatoriale: les socie'tes cotieres et la
penetration franfaise (1820-1874). Paris, 1981.
Metegue N'nah, N., L''implantation coloniale au Gabon. Resistance d'un peuple
(18)9-1920). Paris, 1981.
Meye, F., 'Note d'histoire: Emane Tole de Ndjole, vers 1900', Re'alite's
gabonaises, 1965, 26, 46—7.
Michel, M., La mission Marchand (see under ch. 1).
Patterson, K. D., The Mpangue and the Orungi of the Gabon coast. Oxford, 1975.
Rey, P.-P., Colonialisme, nio-colonialisme et transition au capitalisme. Exempie de la
COMILOG au Congo-Brazzaville. Paris, 1971.
Robineau, C , L'e'volution economique et sociale en Afrique. L'exemple de Souanke,
Re'publique du Congo-Brazj(aville. Paris, 1964.
Saintoyant, J., L'affaire du Congo 190/ (ed. Ch.-A. Julien). Paris, i960.
Sautter, G., De I'Atlantique au fleuve Congo. Une geographie du sous-peuplement:
Re'publique du Congo, Ripubliqut gabonaise. Paris, 1966.
Sillans, R. and Raponda-Walker, A., Rites et croyances des peuples du Gabon. Paris,
1962.
Suret-Canale, J., L'e're coloniale (see under chs. 4/5)-
Thompson, V. and Adloff, R., The emerging states of French Equatorial Africa.
Stanford and London, 1960.
851

BIBLIOGRAPHY
Vansina, J., The Tio kingdom (see under ch. i).
Veistroffer, A., Vingt ans dans la brousse africaine. Souvenirs d'un ancien membrt de
la Mission Savorgnan dans I'Ouest africain (1883-1903). Lille, 1931.
Zorzi, E., Al Congo con Braâ. Viaggio di due exploratori Italian: nel carteggio e
nel 'Giornaliy inediti di Altilio Pecile (1883-1886). Rome, 1940.
B. King Leopold's Congo

Abel, A., Les musulmans noirs du Maniema. Brussels, i960.
Anstey, R., ' T h e Congo rubber atrocities - a case study', Afr. Hist. Studies,
197'. 4. 1. 59-76.
Armani, L., Diciotto mesi al Congo. Milan, 1907.
Banning, E., Memoires politiques et diplomatiques. Brussels, 1927. {See also below
under Stengers, J., ed.)
Belgique coloniale, Brussels, 1895-1904. Then becomes: Be/gique maritime et
colon/ale, 1904-1960.
Bentley, W . H., Pioneering on the Congo, 2 vols. London, 1900.
Bimanyu, D . K., The Wangwana of eastern Zaire. Ph.D. dissertation, London,
1976.
Biographie Coloniale beige, 5 vols. Brussels, 1948-195 8. Biographie Beige d'Outre-
Mer, 2 vols. Brussels, 1968, 1973. (The second is the continuation of the
first.)
Bontinck, F. and Janssen, K., L'autobiographic de Homed ben Mohammed el Murjebi
Tippo Tib (ca.1840-190}). Brussels, 1973.
Bustin, E., Congo-Kinshasa: Guide bibliographique. Cahiers du CEDAF, 1971, no. 3.
Lunda under Belgian rule: the politics of ethnicity. Cambridge, Mass., 1975.
Caltier, F., Etude sur la situation de VEtat Independant du Congo. Brussels, 1906.
Ceulemans, P., La question arabe et le Congo (1883-1892). Brussels, 1959.
Cline, C. A., E. D. Morel 1871-1924: the strategies of protest. Belfast, 1980.
Collins, R. O., The Southern Sudan 1883-1898: a struggle for control. New Haven,
1962.
King Leopold, England and Upper Nile (see under ch. 2).
Cookey, S. J. S., Britain and the Congo Question, i88j-rpij. London, 1968.
Cornet, R. J., Katanga: le Katanga avant les Beiges. 3rd edn, Brussels, 1946.
La bataille du rail: la construction du chemin de fer de Matadi au Stanley Pool.
Brussels, 1947; 4th edn, 1958.
Maniema: la lutte contre les Arabes esclavagistes au Congo. Brussels, 1952.
Daye, P., Leopold II. Paris, 1934.
De Calonne Beaufaict, A., La penetration de la civilisation au Congo beige. Brussels,
1912.
de Lichtervelde, L., Leopold II. Brussels, 1926.
Dereine, A., ' L e soulcvement des Babua (1900-1901)', Africa-Tervuren, 1964,
10, 2, 29—48.
Emerson, B., Leopold II of the Belgians, King of colonialism. London, 1979.
Flament, F. et al.. La Force Publique de sa naissance a 1914. Brussels, 1952.
Gann, L. H. and Duignan, P., The rulers of Belgian Africa 1884-1914. Princeton,
1979.
852

Harms, R., 'The end of red rubber: a reassessment', J. Afr. Hist., 1975, 16,
73-88.
Hulstaert, G.,' Le voyage au Congo d'un officier danois. Notes et commentaires
sur les sejours a l'Equateur de Knud Jespersen (1898-1908)', Enquetes et
documents d'histoire africaine, 1980,4, 1-100.
Janssens, E. and Cateaux, A., Les Beiges au Congo, 3 vols. Antwerp, 1908/12.
Jewsiewicki, B., 'Notes sur l'histoire socio-economique du Congo
(1880-1960)', Etudes d'bistoire africaine, 1972, 3, 209-41.
Etude analytique des archives âiroises. Lubumbashi, 1973 (stencil).
Katzenellenbogen, S. E., Railways and the copper mines of Katanga. Oxford, 1973.
Lambert-Culot, M. C , 'Les premieres annees en Afrique du Comite Special
du Katanga', Etudes d'histoire africaine, 1972, 3, 275-311.
Lejeune Choquet, A., Histoire militaire du Congo. Brussels, 1906.
Leopold II et Beemaert d'apres leur correspondance inedite de 1884 a 1894, ed. E. Van
der Smissen, 2 vols. Brussels, 1920.
Louis, W. R. and Stengers, J., E. D. Morel's history (see above under ch. 6A).
Lutumba-Lu-Vilu Na Wundu, Histoire du Zaire. L'administration centrale du
Ministire beige des Colonies, 1908-1940. Kinshasa, 1972.
Martens, D. S., A history of European penetration and African reaction in the Kasai
region of Zaire, 1880-1908. Ph.D. dissertation, Simon Fraser University,
1980.
Masoin, F., Histoire de I'Etat Independant du Congo, 2 vols. Namur, 1912-13.
Meyers, J., Leprix d'un empire. 3rd edn, Brussels, 1969.
Mille, P., Le Congo Leopoldien. Paris, 1905.
Mouvement des missions catholiques au Congo. Brussels, 1889-1914.
Mouvement geograpbique. Brussels, 1884—1924.
Mumbanza mwa Bamwele, 'Les etablissements d'enseignement public a
l'epoque de l'E.I.C. La colonie scolaire de Nouvelle-Anvers (1892-1913)',
Etudes d'bistoire africaine, 1976, 8, 87-129.
Histoire despeuples riverains de I'entre Zaire-Ubangi: evolution sociale et e'conomique
(ca.ijoo-ipjo). Ph.D. dissertation, National University of Zaire, 1980 (2
vols.).
Ndua Solol, 'Mwant Yav Mushid (c. 1856-1907)', Etudes d'histoire africaine,
«973. 5. 2 5~5°-
Peemans, J. P., Le role de I'Etat dans la formation du capital au Congo pendant la
pe'riode coloniale (188j—1960). Louvain, 1973.
Ranieri, L., Les relations entre I'Etat Independant du Congo etl'Italie. Brussels, 1959.
Sanderson, G. N., England, Europe and the Upper Nile (see under ch. 2).
Shaloff, S., Reform in Leopold's Congo. Richmond, 1970.
Simar, T., 'Bibliographic congolaise de 1895 a 1910', Revue congolaise, 1912,
252-83. 354-8i.
Slade, R., English-speaking missions in the Congo Independent State. Brussels, 1959.
King Leopold's Congo (see under ch. 1).
Stanley, H. M., Through the Dark Continent. London, 1879.
The Congo and the founding of its Free State, 2 v o l s . L o n d o n , 1885.
S t e n g e r s , J . ( e d . ) , Textes ine'dits d'Emile Banning. B r u s s e l s , 1955.
Combien le Congo a-t-il coute a la Belgique? Brussels, 1957.
853

BIBLIOGRAPHY
Belgique et Congo. L'elaboration de la Charte Coloniale. Brussels, 1963.

'The Congo Free State and the Belgian Congo before 1914' in L. H. Gann
and P. Duignan, Colonialism in Africa, 1 (see under General).
'La question du Haut-Nil' (see under ch. 2).
' King Leopold's imperialism' (see under ch. 2).
'La dette publique de PEtat Independant du Congo 1879-1908', in La delte
publique au XVIIle et XIXe siecles. Colloque international. Brussels, 1980.
Stenmans, A., La reprise du Congo par la Belgique. Brussels, 1949.
Storme, M., La mutinerie militaire au Kasai en 189 j . Introduction. Brussels, 1970.
Thuriaux-Hennebert, A., Les Zande dans I'histoire du Bahr-el-Ghaâl et de
I'Equatoria. Brussels, 1964.
Vandewoude, E., Inventaire des archives relatives au developpement exterieur de la
Belgique sous le regne de Leopold II. Brussels, 1965.
Van Grieken-Taverniers, M. Inventaire des archives des Affaires etrangeres de /' Etat
Independant du Congo et du Ministere des Colonies, 1881-1914. Brussels, 1955.
Guide des archives africaines du Ministere des Affaires africaines, 1881-1962.
Brussels, 1981.
van Zuylen, P., L'Echiquier congolais ou le secret du Roi. Brussels, 1959.
Vellut, J.-L., Guide de I'etudiant en histoire du Zaire. Kinshasa, 1974.
Les bassins miniers de I'ancien Congo Beige. Essai d'histoire economique et sociale
(1900-1966). Cahiers du CEDAF, Nov. 1981, no. 7.
Verbeken, A., La revolte des Batetela en 189;. Textes ine'dits. Brussels, 1958.
Waltz, H., Das Konêssionsn/esen im belgischen Kongo, 2 vols. Jeaa, 1917.
Wauters, A. J., Histoirepolitique du Congo beige. Brussels, 1911.
Wauters, A. J. and Buyl, A., Bibliographic du Congo, 1880-189/. Brussels, 1895.
Willequet, J., Le Congo beige et la Weltpolitik 1894—1914. Brussels, 1962.
Yates, B., The Missions and educational development in Belgian Africa: 1876—1908.
Ph.D. dissertation, Columbia University, New York, 1968.
Young, C , Politics in the Congo: decolonisation and independence. Princeton, 1965.
7 A N D 8. S O U T H E R N A F R I C A , 1867-1886
S O U T H E R N AND CENTRAL AFRICA, 1886-19IO
Amery, L. C. M. S. (ed.), The Times history of the war in South Africa 1899-1902,
7 vols. London, 1900-1909.
Arndt, E. H. D., Banking and currency development in South Africa 1812-1927.
Cape Town, 1928.
Arrighi, G., The political economy of Rhodesia. The Hague, 1967.
'Labour supplies in historical perspective: a study of the proletarianisation
of the African peasantry in Rhodesia',/. Development Studies, 1970, 6, 3.
Atmore, A. and Marks, S., 'The Imperial Factor' (see under ch. 2).
Axelson, E., Portugal and the scramble (see under ch. 2).
Aydelotte, W. O., Bismarck (see under ch. 2).
Barnes, J. A., Politics in a changing society - the Fort Jameson Ngoni, Manchester,
«954-
Beach, D. N., 'The rising in south-western Mashonaland, 1896-7'. Ph.D.
London, 1971.
854

7/8 SOUTHERN AND CENTRAL AFRICA
'Ndebele raiders and Shona power', / . Afr. Hist., 1974, 15, 4.
'" Chimurenga": The Shona risings of 1896-7', / . Afr. Hist., 1979, 20, 3.
Beinart, W.,' European traders and the Mpondo paramountcy, 1878-1886', J.
Afr. Hist., 1979, 20.
'Conflict in Qumbu: rural consciousness, ethnicity and violence in the
colonial Transkei, 1880-1913 ', J. Southern Afr. Studies, 1981, 8, 1.
The political economy of Pondoland. Cambridge, 1982.
Benyon, J. A.,' Basutoland and the High Commission, with particular reference
to the years 1868-84. The changing nature of the imperial government's
"special responsibility for the territory"*. D.Phil. Oxford, 1968.
Benyon, J., 'The process of political incorporation', in W. D. Hammond
Tooke (ed.), The Bantu-speaking peoples of Southern Africa. London, 1974.
Benyon, J. A., Proconsul and paramountcy r the High Commission, British supremacy
and the sub-continent 1806—1910. Pietermaritzburg, 1980.
Bhebe, N. M. B., 'Christian missions in Matabeleland 1859-1923'. Ph.D.
London, 1972.
Birmingham, D . and Martin, P. M. (eds.), Central Africa (see under ch. 1).
Bitensky, M. F., 'The South African League'. M.A., Univ. of Witwatersvand,
1951.
Blainey, G., 'Lost causes...' (see under ch. 2).
Bley, H., South West Africa under German rule 1894-1914, trans. H. Ridley.
London, 1971.
Bonner, P., Kings, commoners and concessionaires: the evolution and dissolution of the
19 th century SwaqJ state. Cambridge, 1983.
Bozzoli, B., The political nature of a ruling class: capital and ideology in South Africa
1890-19)}. London, 1981.
Bradlow, E., ' The Cape government's rule of Basutoland', Archives Year Book
for South African History, 1968, 3 1 , 2 .
Brock, S. M., 'James Stewart and Lovedale: a reappraisal of missionary
attitudes and African response in the eastern Cape, South Africa,
1870-1905'. Ph.D. Edinburgh, 1974.
Brookes, E. H., The history of Native policy in South Africa from 18)0 to the present
day. Cape Town, 1924.
Brookes, E. H. and Webb, C. de B., A history of Natal. Cape Town, 1967.
Brown, R., The Ndebele succession crisis 1868-77. Central African Historical
Association, Local series 5. Salisbury, 1962.
Bundy, C , 'The emergence and decline of a South African peasantry', African
Affairs, Oct. 1972, 71, 285.
The rise and fall of the South African peasantry. London, 1979.
Burke, G. and Richardson, P., 'The migration of miners' phthisis between
Cornwall and the Transvaal, 1876-1918', / . Southern Afr. Studies, 1978,
4, *••
Burman, S. B. (ed.), Thejustice of the Queen's government: the Cape's administration

of Basutoland 1871-84. Leiden, 1976.
Chiefdom politics and alien law: Basutoland under Cape rule, 1871-1884. London,
1981.
Butler, J., 'Sir Alfred Milner on British policy in South Africa in 1897', in
855

BIBLIOGRAPHY
J. Butler (ed.), Boston University Papers on African History, vol. i. Boston,
1964.
The Liberal Party and the Jameson Raid. Oxford, 1968.
'Cecil Rhodes', Int. J. Afr. Hist. Studies, 1977, 2, 259-81.
Cairns, H. A. C , Prelude to imperialism: British reactions to Central African society,
1S40— 1890. London, 1965.
Campbell, W. B., 'The South African frontier, 1865-1885', Archives Yearbook
for South African History, 1959.
Caplan, G. L., ' Barotscland's scramble for protection' {see under ch. 2).
The elites of Barotseland, 1878-1969: a political history of Zambia's western
province. London, 1970.
Cartwright, A. P., Gold paved the way: the story of the Goldfields group of companies.
London, 1967.
Golden age: the story of the industrialisation of South Africa and the part played
in it by the Corner House Group of Companies, 1910-67. Cape Town, 1968.
The first South African: the life and times of Sir Percy Fit^patrick. Cape Town,
1971.
Cell, J., The highest stage of white supremacy: the origins of segregation in South Africa
and the American South. Cambridge, 1982.
Chamberlain, M. E., The rise andfall of Germany's colonial empire. London, 1930.
Chanock, M., Unconsummated union: Britain, Rhodesia and South Africa,
1900-194). Manchester, 1977.
Clarence-Smith, W. G., 'Slaves, commoners and landlords in Bulozi,
< \ I 8 7 5 - I 9 O 6 ' , J. Afr. Hist., 1979, 20, 2.
Clarence-Smith, W. G. and Moorsom, R., ' Underdevelopmcnt and class
formation in Ovamboland, 1845-1915',_/. Afr. Hist., 1975,16, 3, 365-82.
Clay, G., Your friend, Ltwanika: Litunga of Barotseland, 1842-1916. London,
1968.
Cobbing, J. R. D . , ' Lobengula, Jameson and the occupation of Mashonaland',
Rbodesian History, 1973, 4, 39—56.
'The Ndebele under the Khumalos', Ph.D. Lancaster, 1976.
' The absent priesthood: another look at the Rhodesian risings of 1896—7',
J. Afr. Hist., 1977, 18, 1.
Coillard, F., On the threshold of Central Africa. London, 1902.
Colson, E., The Plateau Tonga. Manchester, 1958; reprinted 1970.
Colvin, I., The life of Jameson, 2 vols. London, 1922-3.
Comaroff, J. L. (ed.), The Boer War diary of Sol T. Plaatje, an African at Mafeking.
London, 1973.
Consolidated Goldfields, The goldfields, 1887-19)7. Johannesburg, 1937.
Cronwright-Schreiner, S. C. (ed.), The life and letters of Olive Schreiner,
1876-1920. London, 1924.
Cunnison, I. (tr. and ed.), Historical traditions of the eastern Lunda, Rhodes-
Livingstone Communication no. 23. Lusaka, 1962.
Curtis, L., With Milner in South Africa. London, 1951.
Dachs, A. J., 'Missionary imperialism' {see under ch. 2).
'Rhodes's grasp for Bechuanaland, 1889-1896', Rbodesian History, 1971, 2.
'Missionary imperialism - the case of Bechuanaland', / . Afr. Hist., 1972,
i3»4-
856

7/8 SOUTHERN AND CENTRAL AFRICA
(ed.), Christianity south of the Zambezi. Salisbury, 197}.
(ed.), The papers of John Mackenzie. Johannesburg, 1975.
Davenport, T. R. H., The Afrikaner Bond {see under ch. 2).
South Africa: a modern history. London, 1977.
Davenport, T. R. H. and Hunt, K., The right to the land. Cape Town, 1974.
Davey, A. M., The British pro-Boers, 1877-1902. Cape Town, 1978.
Davies, R. H., 'Mining capital, the state and unskilled white workers in South
Africa, 1901-1913 ', J. Southern Afr. Studies, 1976, 3, i.
Capital, State and white labour in South Africa 1900-1960. Brighton, Sussex, 1979.
De Kiewiet, C. W., The imperial factor (see under ch. 1).
A history of South Africa: social and economic. Oxford, 1937.
Delius, P. N., 'The Pedi polity under Sekwati and Sekhukhune', Ph.D.
London, 1980 (published as The land belongs to us, q.v.).
Delius, P.,' Abel Erasmus: power and profit in the eastern Transvaal', research
paper, Institute of Commonwealth Studies. Oxford, 1981.
' The land belongs to us': the Pedi polity, the Boers and the British in the nineteenth
century Transvaal. Johannesburg and London, 1983.
Delius, P. and Trapido, S., ' Inboekselings and Oorlams: the creation and
transformation of a servile class', / . Southern Afr. Studies, 1982, 8, 2.
Denoon, D. J. N., 'The Transvaal labour crisis, 1901—6', J. Afr. Hist., 1967,
7. 3-
'"Capitalist influence" and the Transvaal Government during the Crown
Colony period, 1900-1906', Hist. ]., 1968, 11, 2.
'Participation in the "Boer War": People's War, People's non-war or
non-People's War?' in B. A. Ogot (ed.), War and society in Africa, London,
1972.
A grand illusion: the failure of imperial policy in the Transvaal colony during the
period of reconstruction, 1900—190;. London, 1973.
' Capital and capitalists' (see under ch. 2).
Drechsler, H., Let us diefighting:the struggle of the Herero and Nama against German
imperialism, 1884—191;. London, 1980.
Duminy, A. H.,' Sir Alfred Milner and the outbreak of the Anglo-Boer War'.
Ph.D. Natal, 1976.
Duminy, A. H. and Ballard, C. (eds.), The Anglo-Zulu war: new perspectives.
Pietermaritzburg, 1981.
Duminy, A. H. and Guest, W. R., Fit^patrick, South African politician: selected
papers 1888-1906. Johannesburg, 1976.
Du Plessis, J. A., A history of Christian missions in South Africa. London,
1911.
Engelbrecht, S. P., Thomas Francois Burgers: a biography. Pretoria, 1946.
Engelenburg, F. W., General Louis Botha. London, 1929.
Esterhuyse, J. H., South West Africa, 1880-94: the establishment of German
authority. Cape Town, 1968.
Etherington, N. A.,' Mission Station melting pots as a factor in th rise of South
African nationalism', Int. J. Afr. Hist. Studies, 1976, 9, 4.
Preachers, peasants and politics in South-east Africa, i8))-8o. London, 1978.
' Labour supply and the genesis of the South African Confederation in the
1870s',/. Afr. Hist., 1979, 20, 2.
857

BIBLIOGRAPHY
Eybers, G. W. (ed.), Select constitutional documents illustrating South African history
i/pj-rpro. London, 1918.
Fitzpatrick, J. P., The Transvaal from within. London, 1899.
Flint, J., Cecil Rhodes. London, 1976.
Frankel, S. H., Capital investment in Africa. London, 1938.
Investment and the return to capital in the South African gold mining industry.
Cambridge, 1967.
Fraser, M. and Jeeves, A. (eds.), All that glittered: selected correspondence of Lionel
Phillips 1890—1924. Cape Town, 1977.
Gaitskell, D. L-, 'Female mission initiatives: black and white women in three
Witwaterstand churches, 1905-1939', Ph.D. London, 1981.
Galbraith, J. S., 'Engine without a governor: the early years of the British
South Africa Company', Rhodesian History, 1970, 1.
Crown and Charter: the earlyyears of the British South Africa Company. Berkeley,
Calif., 1974.
Gandhi, M. K., Satyagraha in South Africa. Ahmedabad, 1928.
Gann, L. H., The birth of a plural society under the British South Africa Company:
Northern Rhodesia 1894-1914. Oxford, 1959.
A history of Northern Rhodesia, early days to 19 fj. London, 1964.
A history of Southern Rhodesia: early days to 19)4. London, 1965.
Garson, N. G., 'The Swaziland question and a road to the sea, 1887-95',
Archives Year Book for South African History, vol. 11, 195 7.
Garvin, J. L., The life of Joseph Chamberlain, 3 vols. London, 1932—4.
Garvey, B., 'The development of the White Fathers' mission among the
Bemba-speaking peoples, 1891—1964', Ph.D. London, 1974.
Gelfand, M., Lakeside pioneers. Oxford, 1964.
Gifford, P. and Louis, W. R. (eds.), Britain and Germany {see under General).
Ginwalla, F., 'Class consciousness and control: Indian South Africans,
1860-1946', D.Phil. Oxford, 1974.
Glass, S., The Matabele war. London, 1968.
Goldblatt, I., History ofSouth West Africa from the beginning of the nineteenth century.
Cape Town, 1971.
Gollin, A. M., Proconsul in politics. London, 1964.
Goodfellow, C. F., Great Britain and confederation (see under ch. 2).
Gordon, C. T., The growth of Boer opposition to Kruger. Cape Town, 1970.
Guy, J. J., The destruction of the Zulu kingdom (see under ch. 1).
Hall, K. O., Imperial proconsul, Sir Hercules Robinson and South Africa 1881-1889.
Kingston, Ont., 1980.
Hancock, W. K., Jan Christiaan Smuts, vol. 1: The sanguine years, 1870—1919,
Cambridge, 1962; vol. 11: Thefieldsofforce, 1919-19/0, Cambridge, 1968.
Hancock, W. K. and Van der Poel, J. (eds.), Selections from the Smuts Papers,
4 vols. Cambridge, 1966.
Hanna, A. J., Beginnings of Nyasaland (see under ch. 2).
Harries, P., 'Kinship, ideology, and the nature of pre-colonial labour
migration - labour migration from the Delagoa Bay hinterland to South
Africa up to 1895', in S. Marks and R. Rathbone, Industrialisation and social
change (see below).
858

7 / 8 SOUTHERN AND CENTRAL AFRICA
Headlam, C , The Milner papers, z vols. London, 1931-}.
Hemson, D., 'Class consciousness and migrant workers: dockworkers of
Durban'. Ph.D. Warwick, 1980.
Henderson, W. O., Studies in German colonial history. London, 1962.
Hermitte, E., 'An economic history of Barotseland, 1800-1940'. Ph.D.
Northwestern, 1974.
Hobson, J. A., The war in South Africa, its causes and effects. London, 1900.
Imperialism. London, 1902.
Hofmeyr, J. H. with Reitz, F. W., The life of Jan Hendrik Hofmeyr {Onê Jan).
Cape Town, 1913.
Horell, M., African education: some origins, and development until 19 jj. Johan-
nesburg, 1963.
Houghton, D. Hobart, The South African economy. Cape Town, 1964.
Houghton, D. Hobart and Dagut, J., Source material on the South African economy,
1860-1970, vols. 1 and 2. Cape Town, 1972-
Hunt, D. R., 'An account of the Bapedi', Bantu Studies, 1931, 5, 4.
Hunt Davis, R., 'School vs. blanket and settler: Elijah Makiwane and the
leadership of the Cape School community', African Affairs, 1979, 78, 310.
Huttonback, R. A., Gandhi in South Africa, British imperialism and the Indian
question', 1860—1914. Cornell, 1971.
Hyam, R.,"Elginand Churchill at the Colonial Office, 190 j-8. London, 1968.
Innes, D., Anglo-American and the rise of modern South Africa (London, 1984).
Innes, J. Rose, Sir James Rose Innes: selected correspondence 1884-1902 (ed.
M. Wright), Van Riebeeck Society, 2nd series, 3. Cape Town, 1972.
Jabavu, D. D. T., The life of John Tengo Jabavu. Lovedale, 1922.
Jeeves, A. H., 'The Rand capitalists and Transvaal politics, 1892-9', Ph.D.
Queen's University, Ontario, 1971.
' The Rand capitalists' (see under ch. 2).
' The control of migratory labour on the South African gold mines in the
era of Kruger and Milner', J. Southern Afr. Studies, 1975, 2, 1.
Johnston, Sir H. H., British Central Africa. London,. 1897.
The story of my life. London, 1923.
Johnstone, F. A., Class, race and gold: a study of class relations and racial
discrimination in South Africa. London, 1976.
Kalinga, O. J., 'The Ngonde of Northern Malawi, f.1600-1895'. Ph.D.
London, 1974.
Kaplan, G. L., The elites of Barotseland, 1878-1969. A political history of Zambia's
Western Province. London, 1970.
Katz, E. N., 'White workers' grievances and the industrial colour bar,
1902-13', S. Afr. J. Econ., 1974, 42, 2.
A trade union aristocracy. Johannesburg, 1977.
Kaye, H., The tycoon and the president: Alois Hugo Nellmapius 1847-9). London,
1978.
Keegan, T., 'The restructuring of agrarian class relations in a colonial
economy: the Orange River Colony, 1902-1910', J. Southern Afr. Studies,
•979. 5. 2-
' The transformation of agrarian society and economy in industralising South
859

BIBLIOGRAPHY
Africa: the Orange Free State grain belt in the early twentieth century',
Ph.D. London, 1981.
Kennedy, P., 'German colonial expansion' (see under ch. 2).
Keto.C. T.,' Black Americans and South Africa, 1890-1910', Current Bib. Afr.
Affairs, 1972, 5.
Kimble, J., 'Labour migration in Basutoland 1870-1885', in S. Marks and
R. Rathbone, Industrialisation and social change (see below).
Krishnamurthy, B. S., 'Land and labour in Nyasaland 1891-1914', Ph.D.
London, 1964.
Kriiger, D. W., Paul Kruger, 2 vols. (Afrikaans). Johannesburg, 1961-3.
Kubicek, R. V. Economic imperialism (see under ch. 2).
Lacey, M., Working/or Boroko: the origins of a coercive labour system in South Africa.
Johannesburg, 1981.
Legassick, M., 'The making of South African "Native Policy", 1903-192}:
the origins of "segregation"'. Unpublished paper, Institute of Com-
monwealth Studies, London, 1972.
'British hegemony and the origins of segregation in South Africa'. Unpub-
lished paper, Institute of Commonwealth Studies, London, 1973.
Le May, G. H. L., British supremacy (see under ch. 2).
Letcher, O., The gold mines of South Africa. London, 1936.
Levy, N., The foundations of the South African cheap labour system. London, 1982.
Lewis, C. and Edwards, G. E., Historical records of the Church of the Province of
South Africa. London, 1934.
Lewsen, P., 'The first class in responsible government in the Cape Colony',
Archives Year Book for South African History, 1942, 5, 1.
(ed.), Selections from the correspondence of J. X. Merriman, 4 vols. Cape Town,
i960, 1963, 1966, 1969.
John X. Merriman: paradoxical South African statesman. Yale, 1952.
Leyds, W. J., The first annexation of the Transvaal. London, 1906.
Linden, I. with Linden, J., Catholics, peasants and Cewa resistance in Nyasaland,
iSSf—ipjp. London, 1974.
Livingstone, W. P., Laws of Livingstonia. London, 1921.
Lockhart, J. G. and Woodhouse, C. M., Rhodes. London, 1963.
Lovell, R. I., The struggle for South Africa, 1S7S—99: a study in economic
imperialism. New York, 1934.
McCracken, J. L., The Cape Parliament 18J4-1910. Oxford, 1967.
Politics and Christianity in Malawi, rfyj-ryjo. Cambridge, 1977.
Macdonald, R., 'A history of African education in Nyasaland 1875—1945'.
Ph.D. Edinburgh, 1969.
(ed.), From Nyasaland to Malawi. Nairobi, 1975.
Machingaidze, V. E. M.,' The development of the settler capitalist agriculture
in Southern Rhodesia with particular reference to the role of the State,
1908—1939'. Ph.D. London, 1980.
Mackenzie, J. M.,' African labour in the Chartered Company period', Kbodesian
History, 1970, 1, 43-58-
Mackenzie, W. D., John Mackenzie: South African missionary and statesman.
London, 1902.
860

Macmillan, H. W., 'The origins and development of the African Lakes
Company, 1878-1908'. Ph.D. Edinburgh, 1970.
Macquarrie, J. W. (ed.), The reminiscences of Sir Walter Stanford, 2 vols. Cape
Town, 1958, 1962.
Magubane, B., The political economy of race and class in South Africa. New York,
'979-
Mainga Mutumba, Buloî under the Luyana kings: political evolution and state
formation in pre-colonial Zambia. London, 1973.
Malan, F. S., Die Konvensie-dagboek van by Edelagbare Franfois Stepbanus Malan (ed.
J. K. Keller). Cape Town, 1925.
Malherbe, E. J., Education in South Africa, 16)2-1922. Cape Town, 1925.
Mansergh, N., South Africa, 1906—1961: the price of magnanimity. London, 1962.
Marais, J. S., The Cape coloured people 16)2-1937. London, 1939.
Kruger's republic (see under ch. 2).
Marks, S., Reluctant rebellion: an assessment of the 1906-8 disturbances in Natal.
Oxford, 1970.
'Scrambling for southern Africa' (see under ch. 2).
Marks, S. and Atmore, A. (eds.), Economy and society (see under ch. 1).
Marks, S. and Rathbone, R., Industrialisation and social change in South Africa:
African class formation, culture and consciousness, 1870—19)0. London,
1982.
Marks, S. and Trapido, S., 'Lord Milner and the South African state', History
Workshop, Nov. 1979, 8.
Marlowe, J., Milner: apostle of Empire. London, 1976.
Matsebula, J. S. M., A history of Swaziland. Cape Town, 1972.
Matthews, J. W., lncwadi Yami. London, 1887.
Matthews, T. I.,' The historical traditions of the peoples of the Gwembe valley,
middle Zambezi', 2 vols. Ph.D. London, 1976.
Mawby, A. A., 'The political behaviour of the British population of the
Transvaal, 1902-1907', Ph.D. Univ. of Witwatersrand, 1969.
'Capital, government and politics in the Transvaal, 1900-6: a revision and
a reversion', Hist. J., 1974, 18, 2.
Maylam, P. R., 'The making of the Kimberley—Bulawayo railway: a study in
the operations of the British South Africa Company', Khod. Hist., 1977,
8.
Rhodes, the Tswana and British colonialism: collaboration and conflict in the
Bechuanalandprotectorate iSS)-iS99. Westport and London, 1980.
Mebelo, H. S., Reaction to colonialism: a prelude to the politics of independence in
northern Zambia, 189^-19)9. Manchester, 1971.
Meek, Y. S. et al.. Documents of indentured labour, Natal 18)1-1917. Durban,
1980.
Mendelsohn, R., 'Blainey and the Jameson Raid: the debate renewed', J.
Southern Afr. Studies, April 1980, 6, 2.
Michell, L., The life of the Rt Hon. Cecil J. Rhodes, 2 vols. London, 1910-12.
Molteno, P. A., Selections from the correspondence of P. A. Molteno, 1892-1914 (ed.
V. Solomon). Cape Town, 1981.
Moorsom, R.,' Colonialism and proletarianisation: an exploratory investigation
861

BIBLIOGRAPHY
of the formation of the working class in Namibia under German and South
African colonial rule to 1945'. M.A. Sussex, 1973.
Morris, D. R., The washing of the spears. London, 1966.
Morris, M., 'The development of capitalism in South African agriculture: class
struggle in the countryside', Economy and Society, 1976, 5, 3.
Morrow, L. F., 'State formation, factional rivalries and external trade in the
Ndebele kingdom, 1823-1884; problems and methods of maintaining
royal authority in a heterogeneous conquest state', Ph.D. Duke University,
1975-
Mouton, J. A . , ' General Piet Joubert in die Transvaalse Geskiedenis', Archives
Yearbook for South African History, 1957, 1.
Nasson, W. R., 'Black society in the Cape Colony and the South African war
of 1899—1902: a social history'. Ph.D. Cambridge, 1983.
Newbury, C , 'Out of the pit: the capital accumulation of Cecil Rhodes', J.
Imp. C'wealth Hist., 1981, 10, 1.
Ngcagco, L. D., 'Aspects of the history of the Ngwaketse to 1910'. Ph.D.
Dalhousie University, 1975.
Oliver, R., Sir Harry Johnston {see under ch. 1).
Pachai, B. (ed.), Malawi past and present. Limbe, 1967.
(ed.), Early history (see under ch. 1).
(ed.), Land and politics in Malawi C.187J-1971. 1978.
Palley, C , The constitutional history and law of Southern Rhodesia 1888—19 JJ.
Oxford, 1966.
Palmer, R., Land and racial domination in Rhodesia. London, 1977.
Pakenham, T., The Boer War. London, 1979.
Palmer, R. and Parsons, N. (eds.), Roots of rural poverty {see under ch. 1).
Parsons, Q. N., 'Khama III, the Bamangwato and the British, with special
reference to 1895-1923'. Ph.D. Edinburgh, 1973.
Perrings, C , 'The production process, industrial labour strategies and worker
responses in the South African gold mining industry' (review article), J.
Afr. Hist., 1977, 18, 1.
Phillips, L., Some reminiscences. London, n.d. [1925?].
Phimister, I. R., 'Rhodes, Rhodesia and the Rand', / . Southern Afr. Studies,
•974, 1. i-
Pillay, B., British Indians in the Transvaal, 188;-1906. London, 1976.
Pogge von Strandmann, H., 'Germany's colonial expansion' {see under ch. 2).
Porter, A. N., The Origins of the South African war {see under ch. 2).
Preston, A. (ed.), Sir Garnet Wolseley's Natal diaries. Cape Town, 1971.
(ed.), The South African journal of Sir Garnet Wolseley. Cape Town, 1973.
Prins, G., The hidden hippopotamus: reappraisal in African history. The early colonial
history of western Zambia. Cambridge, 1980.
Purkis, A. J., 'The politics, capital and labour of railway building in the Cape
Colony, 1870-1885'. D.Phil. Oxford, 1978.
P y t a h , G . B . , Imperial policy and South Africa, 1902—10. L o n d o n , 1 9 5 5 .
Ranger, T. O., Revolt (see under ch. 1).
(ed.), Central African history (see under ch. 1).
'Primary resistance' (see under ch. 2).
862

'African reaction and resistance' (see under ch. 2).
The African voice in Southern Rhodesia, 1898-19)0. London, 1970.
The agricultural history of Zambia. Lusaka, 1971.
'Growing from the roots. Reflections on peasant research in Central and
southern Africa',/. Southern Afr. Studies, 1979, 5, 1.
Ranger, T. O. and Kimambo, I. (eds.), African religion (see under ch. 1).
Ranger, T. O. and Weller, J., Themes in the Christian history of Central Africa.
London, 1975.
Rau, E. W., 'Mpezeni's Ngoni of eastern Zambia 1870-1920', Ph.D. UCLA,
1974.
Read, M., The Ngoni of Nyasaland. Oxford, 1956.
Reitz, D . , Commando, a Boer journal of the Boer War. L o n d o n , 1931.
Rennie, J. K., 'Christianity, colonialism and the origins of nationalism among
the Ndau of Southern Rhodesia, 1890-1935', Ph.D. Northwestern,
«973-
' White farmers, black tenants and landlord legislation: Southern Rhodesia
1890-1930', J. Southern Afr. Studies, Oct. 1978, 5, 1.
Richardson, P., Chinese mine labour in the Transvaal. London, 1982.
Roberts, A. D., 'Firearms in North-Eastern Zambia before 1900', Transafrican
J. Hist., 1970, 2.
Bemba (see under ch. 1).
Zambia (see under ch. 1).
Roberts, B., The diamond magnates. London, 1972.
Ross, A. C , 'Origins and development of the Church of Scotland mission at
Blantyre, Nyasaland 1875-1926', Ph.D. Edinburgh, 1968.
Rotberg, R. I., Christian missionaries and the creation of Northern Rhodesia
1880-1924. Princeton, 1965.
The rise of nationalism in Central Africa: the making of Malawi and Zimbabwe,
187)-1964. Cambridge, Mass., 1966.
Rotberg, R. I. and Mazkui, A. A., Protest and power in black Africa. London,
1970.
Rouillard, N . (cd.), Matabele Thompson: his autobiography and story of Rhodesia.
Johannesburg, 1957.
Saker, J. and Aldridge, J., 'The origins of the Langeberg rebellion', J. Afr.
Hist., 1971, 12, 2.
Sanderson, F. E., 'The development of labour migration from Nyasaland,
1891-1914',/. Afr. Hist., 1961, a.
Sauer, H., Ex Africa. London, 1937.
Saundcrs, C. C , 'Tile and the Thembu Church', / . Afr. Hist., 1970, n , 4.
'The annexation of the Transkeian territories (1872-95)', D.Phil. Oxford,
1972.
'The Transkeian rebellion of 1880-1: a case study of Transkeian resistance
to white control', S. Afr. Hist. J., Nov. 1976.
'The 100 years war. Some reflections on African resistance on the
Cape-Xhosa frontier', in D. Chanaiwa (ed.) Profiles of self-determination.
(Northridge, 1979).
863

BIBLIOGRAPHY
Saunders, C. C. and Derricourt, R. (eds.), Beyond the Cape frontier: studies in the
history of the Transkei and Ciskei. London, 1974.
Schapera, I., Tribal innovators. Tswana chiefs and social change, 1791-/940. London
and New York, 1970.
Schreuder, D. M., Gladstone and Kruger {see under ch. 2).
' The cultural factor in Victorian imperialism - a case study in the " civilizing
mission" on the Cape frontier, 1870-84',/. Imp. C'wealth Hist., 1976, 4,
3. 2 »3-Ji7-
Scramble for southern Africa {see under ch. 2).
Shepperson, G. and Price, T., Independent African. Edinburgh, 1958.
Shillington, K., ' Land loss, labour and dependence: the impact of colonialism
on the southern Tswana, 1870-1900'. Ph.D. London, 1981.
Sillery, A., Founding a protectorate: history of Bechuanaland, iSSj-iSff. The
Hague, 1965.
John Mackenzie of Bechuanaland': a study in humanitarian imperialism, /Sjj-/Sff.
Cape Town, 1971.
Simons, H. J. and R. E., Class and colour in South Africa iijo-1910. Harmon-
dsworth, 1969.
Slater, H.,' Land, labour and capital in Natal: the Natal Land and Colonisation
Company, 1860-1948, J. Afr. Hist., 1975, 16.
Spies, S. B., The origin of the Anglo-Boer war. London, 1972.
Methods of barbarism ? Roberts and Kitchener and civilians in the Boer Republics,
January /poo—May /902. Cape Town, 1977.
Stals, E. L. P., 'Die aanraking tussen blankes en Ovambo in Suidwes-Afrika,
1850-1915', Archives Year Book for South African History, 1968, 3 1 ,
11.
Stals, E. L. P., Fourie, J. J. tt al., Afrikaners in die Goudstad, /SSi—1924, vol.
1. Pretoria, 1978.
Stewart, J., Lovedale: past and present. Lovedale, S.A., 1887.
Stigger, P., 'Volunteers and the profit motive in the Anglo—Ndebele war of
1893', Rhod. Hist., 1971, 2, 11-23.
The lând Commission of 1894 and 'be land. Salisbury, 1980.
Stokes, E. T., 'Milnerism', Hist. J., 1962, 6, 1.
Stokes, E. T. and Brown, R. (eds.), The Zambesian past {see under ch. 1).
Stuart, R. G., 'Christianity and the Chewa: the Anglican case, 1885-1950',
Ph.D. London, 1974.
Sundkler, B. G. M., Bantu prophets in South Africa. 2nd edn, London, 1961.
Sutton, I. B., 'The 1878 rebellion in Griqualand West and adjacent territories'.
Ph.D. London, 1975.
Swanson, M. W., 'Urban origins of separate development', Race, 1968, 10.
'" The Durban System ": roots of urban apartheid in colonial Natal', African
Studies, 1976, 35.
'The sanitation syndrome: bubonic plague and urban Native policy in the
Cape Colony 1900-1909', J. Afr. Hist., 1977, 18, 3.
Tangri, R. K., 'The development of Modern African politics and the
emergence of a nationalist movement in colonial Malawi, 1871-195 8'.
Ph.D. Edinburgh, 1970.
864

Tayal, M. J., 'Indian indentured labour in Natal, 1890-1911', Indian Econ. and
Social Hist. Review, 1978, 14, 4.
'Gandhi: the South African experience'. D.Phil. Oxford, 1980.
Taylor, A. J. P., Germany"sfirstbid {see under ch. 2).
Thompson, L. M., The unification of South Africa, 1902-10. Oxford, i960.
Tindall, B. A. (ed.), James Rose Innes: an autobiography. Cape Town, 1949.
Tlou, T., ' A political history of northwestern Botswana to 1906*. Ph.D. Univ.
of Wisconsin, 1972.
Townshend, M. E., The rise andfall of Germany" s colonial empire, 1886—1918. New
York, 1966.
Trapido, S.,' African divisional politics in the Cape Colony, 1884-1910',]. Afr.
Hist., 1968, 9, 1.
' White conflict and non-white participation in the politics of the Cape of
Good Hope, 185 3-1910'. Ph.D. London, 1970.
' The origin and development of the African political organisation'. Collected
seminar papers on the societies of southern Africa in the 19th and zoth
centuries, vol. 1. Institute of Commonwealth Studies, London, 1970.
' South Africa in a comparative study of industrialization', / . Devel. Studies,
1971, 7, 309—20.
'Landlord and tenant in a colonial economy: the Transvaal 1880—1910', / .
Southern Afr. Studies, 1978, 5, 1.
Trollope, A., South Africa, 2 vols. London, 1878.
Turrell, R., "The 1875 Black Flag revolt on the Kimberley diamond fields',
J. Southern Afr. Studies, 1981, 7, 2.
Turrell, R. V., 'Capital, class and monopoly: the Kimberley diamond fields
1871-1889', Ph.D. London, 1982.
Turrell, R., 'Kimberley: labour and compounds, 1871-1888', in S. Marks and
R. Rathbone, Industralisation andsocialchange in South Africa. London, 1982.
Turrell, R. V.,'Rhodes, De Beers and monopoly',}. Imp. C wealth Hist., 1982,
10, 3.
Unterhalter, E., 'Religion, ideology and social change in the Nquthu district
of Zululand, 1879-1910'. Ph.D. thesis, London, 1981.
Uys, C. J., In the era of Shepstone: being a study of British expansion in South Africa,
1842-1877. Lovedale, 1933.
Vail, L.,' Ecology and history: the example of eastern Zambia', J. Southern Afr.
Studies, 1977, 3, 2.
' The state and the creation of colonial Malawi's agricultural economy', in
R. Rotberg (ed.), Imperialism, colonialism and hunger: East and Central Africa,
Lexington, Mass./Toronto, 1983.
Van der Horst, S., Native labour in South Africa. 1942; reprinted London, 1972.
Van der Poel, J., Railways and customs policies in South Africa. London, 1933.
The Jameson Raid. London, 1951.
Van der Ross, R., 'The founding of the A.P.O. in 1903 - t h e role of Dr
Abdurrahman', Munger Africana Notes, no. 28. Pasadena, 1975.
Van-Helten, J. J., 'German capital, the Netherlands Railway Company and the
political Economy of the Transvaal 1886-1900', J. Afr. Hist., 1978,
19.3
86;

BIBLIOGRAPHY
' British and European economic investment in the Transvaal: with specific
reference to the Witwatersrand gold fields and district 1886-1910'. Ph.D.
London, 1981.
Van-Helten, J. J. and Williams, K., '"The crying need of South Africa": the
emigration of single British women to South Africa, 1901-10', J. Southern
y4.fr. Studies, 1983, 10, 1.
Van Horn, L., 'The agricultural history of Barotseland 1890—1964', in Palmer,
R. and Parsons, N., The roots of rural poverty {see above).
Van Jaarsveld, F. A., The awakening of Afrikaner nationalism, 1868—1881. Cape
Town, 1961.
Van Onselen, C , 'Reactions to Rinderpest in South Africa 1896-7', / . Afr.
Hist., 1972, 13, j .
'Worker consciousness in black miners: Southern Rhodesia 1900—1930',/.
Afr. Hist., 1973, 14, 2.
' Black workers in Central African industry: a critical essay on the histori-
ography and sociology of Rhodesia', / . Southern Afr. Studies, 1975, 1, 2.
Chibaro. African mint labour in Southern Rhodesia 1900—19)0. London, 1976.
Studies in the social and economic history of the Witwatersrand, 1886-1914, vol.
1: New Babylon; vol. 11: New Nineveh. Johannesburg and London, 1982.
Van Velsen, J., 'The missionary factor among the Lakeside Tonga of
Nyasaland', Rhodes-Livingstone J., 1959, 26.
Vaughan, M. A., 'Social and economic change in southern Malawi: a study
of rural communities in the Shire highlands and upper Shire valley from
the mid 19th century to 1915'. Ph.D. London, 1981.
Vedder, H., South West Africa in early times (trans. C. G. Hall). London, 1928.
'Vindex', Cecil Rhodes: his political life and speeches. London, 1900.
Walker, E. A., Lard de Villiers and his times. London, 1925.
W. P. Schreiner: a South African. London, 1937.
Wallis, J. P. R., Fit%: the story of Sir Percy Fit^patrick. London, 1954.
Walshe, P., 'The origins of African political consciousness in South Africa',
J. Modern Afr. Studies, 1969, 7, 4.
Walton, E. H., The inner history of the National Convention of South Africa. Cape
Town, 1912.
Warhurst, P. R., Anglo-Portuguese relations {see under ch. 2).
'The Tete agreement', Rhod. Hist., 1970, 1.
Warwick, P. (ed.), The South African war: the Anglo-Boer war. London, 1980.
Black people and the South African war, 1899-1902. Cambridge, 1983.
Webster, E. (ed.), Essays in South African labour history. Johannesburg, 1978.
Wehler, H.-U., 'Bismarck's imperialism', {see under ch. 2).
Weller, J. C , The priest from the lakeside: the story of Leonard Kamungu of Malawi
and Zambia, 1877-191). Blantyre, 1971.
Wells, J., Stewart of Lovedale. London, 1908.
Welsh, D., The roots of segregation: Native policy of Natal {184J-1910). Cape
Town, 1971.
White, C. M. N., 'A preliminary survey of Luvale rural economy' (29, 1959);
'An outline of Luvale social and political organisation' (30, i960);
' Elements in Luvale beliefs and ritual' (32, 1961); 'Tradition and change
866

in Luvale marriage' (34, 1962), Rhodes-Livingstone papers. Manchester,
1959-62.
Wilde, R. H., 'Joseph Chamberlain and the South African Republic 1895-99',
Archives Year Book for South African History, 1956.
Willan, B. P. (ed.), Diary of the siege of Mafeking, October 1899 to May 1900, by
Edward Ross. Cape Town, 1980.
Sol Plaatje: a biography. London and Johannesburg, 1984 (forthcoming).
Williams, B., Cecil Rhodes. London, 1921.
Williams, G. F., The diamond mines of South Africa: some account of their rise and
development. New York, 1902.
Wills, A. J., Introduction to the history of central Africa. London, 1964.
Wilson, F. and Perrot, D., Outlook on a century: South Africa 1870-1970.
Lovedale/Johannesburg, 197}.
Wilson, M. and Thompson, L. (eds.). History of South Africa {see under
General).
Wolpe, H., 'Capitalism and cheap labour-power: from segregation to apar-
theid', Economy and Society, 1972, 1, 4.
Worsfold, W. B., South Africa. London, 1897.
Wright, M. and Lary, P.,' Swahili settlements in northern Zambia and Malawi',
Afr. Hist. Studies, 1971, 4, 3.
Yudelman, D., 'Lord Rothschild, Afrikaner scabs and the 1907 white miners'
strike: a state-capital daguerrotypc', African Affairs, April 1982, 81.
The emergence of modern South Africa. State, capital and the incorporation of
organised labor on the South African gold fields, 1902-19)9. (Westport,
Conn./London, 1983).
9. PORTUGUESE COLONIES AND MADAGASCAR
A. Angola and Mozambique

Alexandre, V., Origem do colonialismo portugues moderno, 1822-1891, Portugal no
seculo xix, ed. M. Halpern Pereira, vol. m. Lisbon, 1979.
Almeida, P. Ramos de, Historia do colonialismo portugues em Africa, 3 vols.
Lisbon, 1978-9.
Alpers, E., Ivory and slaves (see under ch. i).
Andrade, A. Freire de, Relatdrios sobre Mozambique, 4 vols. Lourenco Marques,
1907.
Axelson, E., Portugal and the scramble (see under ch. 2).
Bender, Gerald, Angola under the Portuguese. London, 1978.
Bhila, H. H. K., Trade and politics in a Shona kingdom; the Manyika and their
African and Portuguese neighbours, 1J7J-1902. London, 1982.
Birmingham, D . , 'Coffee barons' (see under ch. 1).
Botelho, J. Teixeira, Historia militar e politico dos Portugueses em Mozambique de
I8JJ aos nossos dias. 2nd edn, Lisbon, 1936.
Brasio, A., Angola, Spiritana Monumenta Historica, Series Africana, 5 vols.
Pittsburg and Louvain, 1966-1971.
Capela, J., 0 vinho para 0preto. Oporto, 1973.
867

BIBLIOGRAPHY
A burguesia mercantil do Porto e as colonias, 18)4-1900. Oporto, 1975.
0 imposto de palhota e a introdufao do modo de produfdo capitalist/! nas colonias.
Oporto, 1977.
Capelo, H. Brito, and Ivens, R., De Angola a contra-costa. Lisbon, 1886.
Carvalho, H. Dias de, Expedifio portuguesa ao Muatianvua, 7 vols. Lisbon,
1890-1894.
Clarence-Smith, W. G., Slaves, peasants and capitalists {see under ch. 1).
The third Portuguese empire, the political economy of overseas expansion, 1821-197 j .
Manchester (forthcoming).
Contreiras, M. Martins, A provincia de Angola. Lisbon, 1894.
Couceiro, H. de Paiva, Angola, dos annos de governo. Lisbon, 1910.
Coutinho, J. Azevedo, Memdrias de urn velho marinheiro e soldado de Africa. Lisbon,
1941.
Dias, J., 'Black chiefs' (see under ch. 1).
Drechsler, H., 'Germany and Southern Angola, 1898-1903', Presence africaine,
1962, 42-5, 51-69.
Duffy, James, Portuguese Africa. Cambridge, Mass., 1959.
Portugal in Africa. Harmondsworth, 1962.
A question of slavery; labour policies in Portuguese Africa and the British
protest, 1810-1920. Oxford, 1967.
Enes, A., Mozambique. 2nd edn, Lisbon, 1913.
Great Britain, Admiralty, A manual of Portuguese East Africa. London,
1920.
Hafkin, N . , ' Trade, society and politics in northern Mozambique, c. 175 3—1915 *.
Ph.D. thesis, Boston, 1973.
Hammond, R. J., Portugal and Africa (see under ch. 2).
Harris, M.,' Labour emigration among the Mozambique Thonga: cultural and
political factors', Africa, 1959, 29, 50-65.
Henderson, L., Angola, five centuries of conflict. Ithaca, 1979.
Henriksen, T., Mozambique: a history. London, 1978.
Isaacman, A., Zambezi pra%ps (see under ch. 1).
The tradition of resistance (see under ch. 1).
Katzenellcnbogen, S. E., Railways and copper mines (see under ch. 6B).
South Africa and Southern Mozambique: labour, railways and trade in the making
of a relationship. Manchester, 1982.
Liesegang, G., Beitrdge %ur Geschichte des Reicbes der Ga%a Nguni im sudlichen
Mozambique, 1820—189}. Cologne, 1967.
Marques, A. de O., Historio de Portugal, vol. 11. Lisbon, 1973. The English
edition (New York, 1972) is less detailed.
Marvaud, A., L* Portugal et ses colonies. Paris, 1912.
Miller, J. C , 'Cokwe trade and conquest in the nineteenth century', in
J. R. Gray and D. Birmingham (eds.), Pre-colonial African trade (see under
ch. 1).
'Slaves, slavers and social change in nineteenth century Kasanje', in
F. Heimer (ed.), Social change in Angola. Munich, 1973.
Neil-Tomlinson, Barry, 'The Nyassa chartered company, 1891-1929', / . Afr.
Hist., 1977, 18, 1, 109—28.
'The growth of a colonial economy and the development of African labour:
868

9 PORTUGUESE COLONIES A N D MADAGASCAR
Manica and Sofala and the Mozambique chartered company, 1892-1942',
in Centre of African Studies, Edinburgh, Mozambique. Edinburgh, 1979.
Nevinson, H., A modern slavery. New York, 1906.
Newitt, M. D. D., ' Angoche' (see under ch. 1).
Portuguese on the Zambezi (see under ch. 1).
Portugal in Africa (see under ch. 2).
Okuma, Thomas, 'The social response of Christianity in Angola: selected
issues'. Ph.D. thesis, Boston, 1964.
Ornelas, Aires de, Colectanea das suas principais obras milltares e coloniais (ed.
E. Lupi), 3 vols. Lisbon, 1934—6.
Papagno, G., Colonialismo e feudalesimo, la questione dei praôs da coroa nel
Moâmbico aliafinedel secolo XIX. Turin, 1972. (There is also a Portuguese
translation, Lisbon, 1980).
Pelissier, R., 'Campagnes militaires au Sud-Angola, 1885-1915', Cahiers
d'etudes africaines, 1969, 9, 33, 54-123.
Les guerres grises (see under ch. 1).
Pimentel, Fernando, Investigacdo commercial na provincia de Angola em 1902-190}.
Oporto, 1903.
Pirio, G., 'Commerce, industry and empire: the making of modern Portuguese
colonialism in Angola and Mozambique, 1890-1914', Ph.D. thesis, Los
Angeles, 1982.
Rita-Ferreira, A., 'Labour emigration among the Mocambique Thonga:
comments on a study by Marvin Harris', Africa, i960, 30.
O movimento migratorio de trabalbadores entre Mozambique e a Africa do Sul.
Lisbon, 1963.
Samuels, M., Education in Angola, 1878-1914. New York, 1970.
Santos, E. dos, A questao da Lunda, 188j—1894. Lisbon, 1966.
Soremekun, F.,' Religion and politics in Angola: the American Board missions
and the Portuguese government, 1880-1922', Cabiers d'etudes africaines,
>97i. 9. 43. 341-77-
' Trade and dependency in Central Angola: the Ovimbunda in the nineteenth
century', in R. Palmer and N. Parsons (eds.), Roots of rural poverty (see under
ch. 1).
Vail, L. and White, L., Capitalism and colonialism (see under ch. 1).
Warhurst, P. R., Anglo-Portuguese relations (see under ch. 2).
Wheeler, D., 'Gungunhana', in N. Bennett (ed.), Leadership in eastern Africa.
Boston, 1968.
'Gungunyane the negotiator: a study in African diplomacy', J. Afr. Hist.,
1968, 9, 4, 585-602.
Wheeler, D. and Christensen, D., 'To rise with one mind: the Bailundu revolt
of 1902', in F. Heimer (ed.), Social change in Angola. Munich, 1973.
Wheeler, D. and Pelissier, R., Angola. London, 1971.
B. Madagascar and Trance

Antananarivo Annual. Tananarive, 1875-1900.
D'Anthouard and Ranchot, L'expedition de Madagascar, journaux de route. Paris,
1930.
869

BIBLIOGRAPHY
Archives des Affaires Etrangeres (see under Archives).

Atlas de Madagascar. Tananarive, 1968.
Augagneur, V., Erreurs et brutalites coloniales. Paris, 1927.
Boudou, A., Les Je'suites a Madagascar au XIX' siecle, 2 vols. Paris, 1942.
Bulletin de I'Academic Malgache. Tananarive, 1902—.
Bulletin du Comite de Madagascar. Paris, 1895—1899.
Carol, J., Chez ^" Hova, au pays rouge. Paris, 1898.
Catat, L., Voyage a Madagascar, 1889—90. Paris, 1895.
Chapus, G. S., Quatre vingts anne'es d'influence europeenne en lmerina. Tananarive,
1925.
Chapus, G. S. and Mondain, G., Un homme d'etat malgache, Kainilaiarivony. Paris,
•953-
Deschamps, H., Les Antaisaka. Tananarive, 1938.
Les migrations inte'rieures a Madagascar. Paris, 1959.
Histoire de Madagascar (see under General).
Deschamps, H. and Chauvet, P., Gallieni pacificateur. Paris, 1949.
Duchesne, General, Rapport sur I'expedition de Madagascar. Paris, 1896.
Gallieni, J. S., Rapport dy ensemble sur la pacification, I'organisation et la colonisation
de Madagascar, 1896—9. Paris, 1899.
La pacification de Madagascar. Paris, 1900.
Madagascar de 1896 a 190), 1 vols. Tananarive, 1905.
Neuf ans a Madagascar. Paris, 1908.
Lettres de Madagascar. Paris, 1928.
Gautier, E. F., Trois Heros. Paris, 193 1.
Gheusi, Gallieni et Madagascar. Paris, 1931.
Grandidier, A. and G., Ethnographie de Madagascar, 5 vols. Paris, 1908—28.
Grandidier, G., Bibliographic de Madagascar, 3 vols. Paris and Tananarive,
19°5-5 7-
Grandidier, G. and Decary, R., Histoire politique et coloniale, vol. v. Paris; part
1, 1942; part 2, 1956; part 3, 1958.
Grandidier, G., Gallieni. Paris, 1931.
Hanotaux, G., L'affaire de Madagascar. Paris, 1896.
Histoire militaire de Madagascar. Paris, Exposition coloniale de 1931.
Julien, G., Institutionspolitiques et sociales de Madagascar, 2 vols. Paris, 1908.
Leblond, M. A., Gallieni park, 2 vols. Paris, 1920.
Lebon, A., La pacification de Madagascar. Paris, 1928.
Lyautey, H., Lettres du Tonkin et de Madagascar, 2 vols. Paris, 1930.
Lettres du sud de Madagascar if 00—ip02. Paris, 1935.
Malzac, V., Histoire du rqyaume Hova depuis ses originesjusqu'a safin.Tananarive,
1912.
Martineau, A., Madagascar en 1894. Paris, 1894.
Mutibwa, P. M., The Malagasy and the Europeans (see under ch. 1).
Oliver, P., The true story of the French dispute in Madagascar. London, 1885.
Madagascar, an historical and descriptive account of the island. London, 1886.
Ralaimihoatra, E., Histoire de Madagascar. Tananarive, 1966.
Randrianarisoa, P., La diplomatie malgache face a la politique des grandes puissances,
1882-189). Paris, 1970.
870

IO EAST AFRICA, 1870-1905
Sibree, J., Fifty years in Madagascar. London, 1924.
Wastell, R. I. P., 'British imperial policy in relation to Madagascar', Ph.D.
London, 1944.
IO. EAST AFRICA, 187O-19O5
Alpers, E. A., The East African slave trade. Nairobi, 1967.

Bennett, N. R., Mirambo (see under ch. 1).
A history of the Arab state of Zanzibar. London, 1978.
Berntsen, J. L.,' The Maasai and their neighbors: variables of interaction', Afr.
Econ. Hist., 1976, 2, 1—11.
Maasai and lloikop: ritual experts and their followers. Occasional Paper 9.
Madison, 1977.
Brown, B., 'Muslim Influence...' (see under ch. 1).
Brown, W. T.,' Politics of business: relations between Zanzibar and Bagamoyo
in the late nineteenth century*, Afr. Wist. Studies, 1971, 4, 3, 631—44.
Cameron, V. L., Across Africa. London, 1877.
Cooper, F., Plantation slavery (see under ch. 1).
From slave to squatter: plantation labour and agriculture in Zanzibar and coastal
Kenya, 1890 to 192J. New Haven/London, [f.1980].
Cummings, R.,' A note on the history of caravan porters in East Africa', Kenya
Hist. Review, 197}, 1, 2, 109-38.
Doornbos, M. R.,' Images and realities of stratification in pre-Colonial Nkore',
Canadian J. Afr. Studies, 1973. 7. )» 477~95-
Esmail, A., 'Towards a history of Islam in East Africa', Kenya Hist. Review,
1975, 3. ' . >47-58-
Fallers, L. A. (ed.), The King's men; leadership and status in Buganda. London, 1964.
Feierman, S., The Sbambaa kingdom. Madison, 1974.
FitzGerald, W. W. A., Travels in the coastlands of British East Africa and the islands
of Zanzibar and Pemba. London, 1898.
Ford, J., The role of the trypanosomiases in African ecology. Oxford, 1971.
Frontera, A., 'The Taveta economy in the pre-colonial period', Kenya Hist.
Review, 1977, 5, 1, 107-14.
Good, C. M., 'Salt, trade, and disease: aspects of development in Africa's
northern Great Lakes region', Int.]. Afr. Hist. Studies, 1972, j , 4, 543-86.
Gray, R. and Birmingham D. (eds.), Pre-colonial African trade (see under ch. 1).
Harlow, V. and Chilver, E. M. (eds.), History of East Africa, vol. 11. Oxford,
1965.
Harries, L. (ed.), Swahili prose texts. Oxford, 1965.
Hartwig, G. W., The art of survival (see under ch. 1).
Iliffe, J., 'The Organization of the Maji-Maji rebellion', / . Afr. Hist., 1967,
8. }. 495-51 i-
(ed.), Maji Maji research project collected papers. Dar es Salaam, 1968 (mimeo).
History of Tanganyika (see under ch. 1).
Jackson, K. A., 'The family entity and famine among the nineteenth century
Akamba of Kenya: social response to environmental stress', J. of Family
Hist., 1976, 1, 2, 193-216.
871

BIBLIOGRAPHY
'Ngotho (the ivory armlet): an emblem of upper-tier status among the 19th
century Akamba of Kenya, <\i83o—1880', Kenya Hist. Review, 1977, 5, 1,
55-69.
Kamuhangire, E. R., 'The pre-colonial economic and social history of East
Africa, with special reference to south-western Uganda salt lakes region',
Hadith, Nairobi, 1976, 5, 67-91.
Karugire, S. R., 'Relations between the Bairu and Bahima', Tarikh, 1970, 3,
22-33.
A history of the kingdom of Nkore in Western Uganda to 1896. London, 1971.
Kieran, J. A., 'Abushiri and the Germans', Hadith, Nairobi, 1970,2, 157-201.
'Christian villages in North-eastern Tanzania', Transafrican ]. Hist., 1971,
1, 24-38-
Kimambo, I. N. and Temu, A. J., A history of Tanzania. Nairobi, 1969.
Kjekshus, H., Ecology control and economic development in East African history.
London, 1977.
Lamphear, J., The traditional history of the Jie of Uganda. Oxford, 1976.
Langworthy, H. W., 'Swahili influence in the area between Lake Malawi and
the Luangwa river', Afr. Hist. Studies, 1971, 4, 3, 575-602.
Linden, I., Church and revolution in Rwanda. Manchester, 1977.
Lonsdale, J. M.,' When did the Gusii (or any other group) become a " tribe " ?',
Kenya Hist. Review, 1970, 5, 1, 123-}}.
' The politics of conquest' {see under ch. 2).
Low, D . A., Buganda in modern history. Berkeley, 1971.
Mangat, J. S., A history of'the Asians in East Africa c. 1886to 194}\ London, 1969.
'The India Office and East Africa, 1889-1903', Hadith, Nairobi, 1970, 2,
202-11.
Mapunda, O. B. and Mpangara, G. P., The Maji Maji war in Ungoni. Dar es
Salaam, 1969.
Martin, B. G., 'Muslim politics, and resistance to colonial rule', J. Afr. Hist.,
1969, 10, 3, 471-86.
Matson, A. T., Nandi resistance to British rule, 1890—1906. Nairobi, 1972.
Maxon, R. M., 'Gusii resistance to British rule and its suppression, 1908',
Transafrican J. Hist., 1972, 2, i, 65-82.
Mungeam, G. H., British rule in Kenya 189^-1912. Oxford, 1966.
Munro, J. F., Colonial rule and the Kamha. Oxford, 1975.
Muriuki, G., 'Kikuyu reaction to traders and British administration,
1850—1904', Hadith, Nairobi, 1968, 1, 101—118.
Nimtz, A. H. Jr., The role of the Muslim Sufi order in political change: an overview
and microanalysis from Tanzania. Ph.D. Indiana, 1973.
Ochieng, W. R., A pre-colonial history of the Gusii of Western Kenya. Nairobi, 1974.
Ogot, B. A. (ed.). War and society {see under ch. 7/8).
(ed.), Zamani {see under ch. 1).
Oliver, R., The missionary factor {see under ch. 1).
Oliver, R. and Mathew, G. (eds.), History of East Africa {see under General).
Page, M. E., "The Manyema hordes' {see under ch. 1).
Pirouet, M. L., Black Evangelists. The spread of Christianity in Uganda, 1891-1914.
London, 1978.
872

1O EAST A F R I C A , 1870—1905
Ranger, T. O. and Kimambo, I. N. (eds.), African religion {see under ch. 1).
Ranger, T. O., Dance and society in eastern Africa. London, 1975.
Redmond, P., 'Maji Maji in Ungoni: A reappraisal of existing historiography',
Int. J. Afr. Hist. Studies, 1975, 8.
Rennie, J. K., 'The precolonial kingdom of Rwanda: a reinterpretation*,
Transafrican J. Hist., 1972, 2, 11-54.
Roberts, A. D . , ' The history of Abdullah Ibn Suliman', African Social Research,
1967, 4, 241-70.
Roscoe, J., The Bakitara. Cambridge, 1923 (reprinted 1968).
Rowe, J. A., 'The purge of Christians at Mwanga's court', / . Afr. Hist., 1964,
J. 5 5-7«-
'Lugard at Kampala', History papers, Makerere, 1969.
'The Baganda revolutionaries', Tarikh, 1970, 3, 34—46.
Salim, A. I., The Swabili-speakingpeoples of Kenya's coast, jgfj-rpij. Nairobi, 1973.
'" Native or non-native? " The problem of identity and the social stratification
of the Arab—Swahili of Kenya', Haditb, Nairobi, 1976, 6, 65-85.
Shorter, A.,' Nyungu ya Mawe and the " Empire of the Ruga Rugas "', / . Afr.
Hist., 1968, 10, 2, 235-59.
Chiefsbip in western Tanzania. London, 1972.
Simpson, D., Dark companions. New York, 1976.
Smith, I. R., Emin Pasha relief {see under ch. 2).
Soff, H., 'Sleeping sickness in the Lake Victoria region of British East Africa,
1900-1915', Int. J. Afr. Hist. Studies, 1969, 2, 2.
Steinhart, E. I., Conflict and collaboration: kingdoms of western Uganda, 1890-1907.
Princeton, 1977.
Strobel, M. A., 'From Lelemama to lobbying', in N. Hafkin and E. Bay (eds.),
Women in Africa. Stanford, 1976.
Swann, A. J., Fighting the slave hunters in central Africa. London, 1969.
Temu, A. J., 'The role of the Bombay Africans on the Mombasa coast,
1874-1904', Hadith, Nairobi, 1971, 3, 53-81.
Tosh, J., 'The Langi of northern Uganda', Tarikh, 1970, 3, 59-68.
Tuma, T., 'African chiefs and church work in Busoga Province of Uganda,
1900-1940, part I', Kenya Hist. Review, 1976, 4, 2, 283-95.
'Some reflections on missionary extension to Busoga, 1891-1894', Makerere
Hist. ]., 1976, 2, 1, 27-37.
Twaddle, M., 'The Bakungu chiefs of Buganda under British rule, 1900-1930',
J. Afr. Hist., 1969, 10, 2, 309-22.
'The Muslim revolution in Buganda', African Affairs, 1972, 71, 54-72.
Uzoigwe, G. N., 'Kabalega and the making of a new Kitara', Tarikh, 1970,
3. ?-*«•
'Kabalega's Abarusura; the military factor in Bunyoro-Kitara', in Revolution
and revolt in Bunyoro-Kitara. Kampala, 1970.
West, H. W., Land policy in Buganda. Cambridge, 1972, ch. 2.
Whiteley, W. H. (trans.), Maishaja Hamedbin Muhammed el Murjebi Yanni Tippu
Tip kwa manenojake mwenyewe. Nairobi, 1971.
Wright, M., 'Local roots of policy in German East Africa',/. Afr. Hist., 1968,
" 10, 4, 621-30.
873

BIBLIOGRAPHY
Wright, M., German missions in Tanganyika 1891-1941. London, 1971.

Wright, M. and Lary, P., 'Swahili settlements {see under ch. 7/8).
Wright, M. A., Buganda in the heroic age. Nairobi, 1971.
II. THE NILE BASIN AND THE EASTERN HORN
A. Egypt
'Abd al-Rahman al-Rafi'I, Al-thawra al-'uribiya wa'l-ihtilalal-injiliî'{The 'Urdbi
revolution and the English occupation). Cairo, 1949.
Mustafa Kami!. Cairo, 1950.
Adams, C. C , Islam and modernism in Egypt. London, 1953.
Ahmed, J. M. (Jamal Muhammad Ahmad), The intellectual origins of Egyptian
Nationalism. London, i960; reprinted, 1968.
Ammar, H. (Hamid 'Ammar), Growing up in an Egyptian village. London, 1954.
Anderson, J. N. D., 'Law reform in Egypt: 1850-1950' in P. M. Holt (ed.).
Political and social change {see below).
Atkins, R. A., 'The [British] Conservatives and Egypt, 1875-1880', / . Imp.
Cwealth Hist., 1974, 2, 2, 190-205.
Baer, G., 'The village shaykh in modern Egypt', Studies in Islamic history and
civilisation, vol. ix. Jerusalem, 1961.
Landownersbip in modern Egypt {see under ch. 1).
'Social change in Egypt: 1800-1914', in P. M. Holt (ed.), Political and social
change {see below).
Social history {see under ch. 1).
Baring, E. (Earl of Cromer), Modern Egypt, 2 vols. London, 1908.
Berger, M., Bureaucracy and society in modern Egypt. Princeton, 1957.
Berque, J., Histoire sociale d'un village egyptien au XX siecle. Paris and The Hague,
•957-
L'Egypte: impirialisme et revolution. Paris, 1967. English translation: Egypt,
imperialism and revolution. London, 1972.
Blunt, W. S., Secret history of the English occupation of Egypt. London, 1907.
Bouvier, J., 'Les intents financiers et la question d'Egypte 1875-1876', in
J. Bouvier, Histoire e'conomique et histoire sociale: rechercbes sur le capitalisme
contemporain. Geneva, 1968.
Chamberlain, M. E., 'The Alexandrian massacre of 11 June 1882 and the
British occupation of Egypt', Middle Eastern Studies, 1977, 13, 1, 14-39-
Chesnel, E., Plaies d'Egypte: les Anglais dans la valle'e du Nil. Paris, 1888.
Collins, R. O . , ' Egypt and the Sudan', in R. W. Winks (ed.), The historiography
of the British Empire-Commonwealth. Durham, N.C., 1966.
Colvin, A., The making of modern Egypt. London, 1906.
Cromer, Earl of, see Baring, E.
Crouchley, A. E., The economic development of modern Egypt. London, 1938.
Douin, G., Kegne du Khedive Ismail {see under ch. 1).
Elgood, P. G., The transit of Egypt. London, 1928.
GaJbraith, J. S., "The trial of Arabi Pasha',/. Imp. Cwealth Hist., 1979, 7, 3,
274-92.
874

Gibb, H. A. R., Modern trends in Islam. Chicago, 1947.

Gilsenan, M., Saint and Sufi in modern Egypt. Oxford, 1973.
Goldschmidt, A., "The Egyptian Nationalist Party: 1892-1919', in P. M. Holt
(ed.), Political and social change {see below).
Haddad, G. M., 'The Arabi revolt - comparisons and comments', The Muslim
World, 1964, 54.
Hey worth-Dunne, J., Introduction to the history of education in modern Egypt.
London, [1938] (reprinted, 1968).
Holt, P. M., The Fertile Crescent (see under ch. 1).
(ed.), Political and social change in modern Egypt. London, 1968.
Hourani, A., Arabic thought in the Liberal Age. London, 1962; reprinted, 1970.
Hunter, F. R., 'Egypt's high officials in transition..., 1849-1879', Middle
Eastern Studies, 1983, 19, 3, 277—300.
Ibrahim Abu-Lughod, ' The transformation of the Egyptian elite: prelude to
the 'Urabi revolt', Middle East J., 1967, 12, 325-44.
Issawi,C.,'Egypt since 1800: a study in lop-sided development',/. Econ. Hist.,
1961, 21, 1, 1-25.
Keddie, N. R., An Islamic response to imperialism. Berkeley and Los Angeles,
1968.
Kedourie, E., Afghani and'Abduh. London, 1966.
Kerr, M. H., Islamic reform. Berkeley and Los Angeles, 1966.
Landes, D. S., Bankers and Pashas, (see under ch. 1).
Landau, J. M., Parliaments and parties in Egypt. Tel Aviv, 195 3, and New York,
1954.
' Prolegomena to a study of secret societies in modern Egypt', Middle Eastern
Studies, 1965, 1, 2, 135—86.
McPherson, J. W., The Moulids of Egypt. Cairo, 1941.
Mabro, R., The Egyptian economy, if J2—1972. Oxford, 1974.
Malct, E., Egypt 1879-188). London, 1909.
Mansfield, P., The British in Egypt. London, 1971.
Marlowe, J., Anglo-Egyptian relations, 1800-ipjj. London, 1954; 2nd edn (to
1956), London, 1965.
Cromer in Egypt. New York, 1970.
Milner, A. (Viscount Milner), England in Egypt. London, 1892.
Mommsen, W., Imperialisms in Agypten. Munich and Vienna, 1961.
O'Brien, P., 'The long-term growth of agricultural production in Egypt:
1821-1962', in P. M. Holt (ed.), Political and social change (see above).
Owen, E. R. J., 'Lord Cromer and the development of Egyptian industry',
Middle Eastern Studies, 1966, 2, 4, 282-301.
Cotton and the Egyptian economy. Oxford, 1969.
Parsons, J. W., 'France and the Egyptian question' (see under ch. 2).
Polk, W. R. and Chambers, R. L., Beginnings of modernisation in the Middle East.
Chicago, 1968.
Ramm, A. (ed.), The Political Correspondence of Mr Gladstone and Lord Granville,
1876-1886, 2 vols. Oxford, 1962.
'Great Britain and France in Egypt, 1876-1882', in P. Gifford and W.
R. Louis (eds.), France and Britain (see under General).
875

BIBLIOGRAPHY
Rivlin, H. A. B., The Ddr al-Wathaiq in 'Abdin Palace at Cairo as a source for
the study of the modernisation of Egypt in the nineteenth century. Leiden, 1970.
Robinson, R. and Gallagher, J., with Denny, A., Africa and the Victorians (see
under General).
Safran, N., Egypt in search of political community. Cambridge, Mass., 1961.
Al-Sayyid, Afaf Lutfi, Egypt and Cromer. London, 1968.
Scholch, A., Agypten den Agyptern! (see under ch. 2).
' Constitutional development in nineteenth-century Egypt - a reconsidera-
tion', Middle Eastern Studies, 1974, 10, 1, 3-14.
'The "Men on the Spot'" (see under ch. 2).
Steppat, F., 'Nationalisms und Islam bei Mustafa KamiF, Welt des Islams,
1956, 4, 4, 241-341.
Tignor, R. L., 'Some materials for a history of the 'ArabI revolution: a
bibliographical survey', Middle East J., 1962, 16, 239-48.
Modernisation and British colonial rule in Egypt, 1882-1914. Princeton, 1966.
Vatikiotis, P. J., The history oj Egypt: from Muhammad AH to Sadat. London,
1980 (revised edn of The modern history of Egypt. London, 1969).
Walid Kazziha, 'The Jaridah-Ummah group and Egyptian polities', Middle
Eastern Studies, 1977, 13,4, 373-85.
Warburg, G. R., 'The Sudan, Egypt and Britain, 1899-1916', Middle Eastern
Studies, 1970, 6, 2, 163-78.
'British rule in the Nile Valley, 1882-1956 and Robinson's theory of
collaboration', Asian and African Studies (J. Israel Oriental Society), 1981,
»S. 5. 287-322.
B. Sudan
'Abdallah 'Ali Ibrahim, Al-Sir' bayn al-Mabdi wdl-ulama' (The Disputation
between the Mahdi and the 'Ulama'). Khartoum, 1968.
'Abd al-Rahman al-Rafi'I, Mist wal-Sudan fi avail '•ahd al-ihtildl (Egypt and the
Sudan at the beginning of the occupation era). Cairo, 1942.
Abdel Rahman el Nasri ('Abd al-Rahman al-Nasri), A bibliography of the Sudan,
19)8-19)8. London, 1962.
Ahmad 'Uthman Ibrahim,' Some aspects of the ideology of the Mahdiya', Sudan
Notes and Records, 1979, 60, 28-37.
Allen, B. M., Gordon and the Sudan. L o n d o n , 1931.
'Awad 'Abd al-Hadl al-'Ata, Ta'rikh Kurdufan al-siyasiftl-Mabdiyya, 1881-1899
(Political history of Kordofan during the Mabdiyya). Khartoum, 1973.
Babikr Badrl, Ta'rikh hayati, 3 vols. [Khartoum], 1959—61.
The memoirs of Babikr Bedri; vol. 1, trans. Yusuf Bedri and George Scott
(introd. P. M. Holt), London, 1969; vol. n, trans, and ed. Yusuf Bedri
and Peter Hogg (introd. G. N. Sanderson), London, 1980.
Baring, E. (Earl of Cromer), Modern Egypt (see above under ch. 11 A).
Biobaku, S. and Muhammad al-Hajj, 'The Sudanese Mahdiyya and the
Niger-Chad Region', in Lewis, I. M. (ed.), Islam in tropical Africa.
London, 1966.
Brett, M. (ed.), Northern Africa: Islam and modernisation. London, 1973.
876

Bukhari, S. A. R., 'Military aspects of internal security in the Anglo-Egyptian

Sudan, 1898-192$'. M.Phil, thesis, London, 1972.
Caulk, R. A., 'Johannes I V {see under ch. 2).
Churchill, W. S., The River War. London, 1899.
Collins, R. O., The southern Sudan {see under ch. 6B).
Leopold, England and Upper Nile {see under ch. 2).
Land beyond the rivers. The southern Sudan 1898-1918. New Haven and London,
1971.
Collins, R. O. and Herzog, R., ' Early British administration in the southern
Sudan',/. Afr. Hist., 1961, 2, 1, 119-55.
Comyn, D. C. E. ff., Service and sport in the Sudan. London, 1911.
Cromer, Earl of, see Baring, E.
Cudsi, A. S., 'Sudanese resistance to British rule 1900-1920'. M.Sc. (Econ.)
thesis, Khartoum, 1969.
Dagher, J. A. (Yiisuf As'ad Daghir), Al-Usulat-'Arabtya If'l-dirdsat al-Sudaniya
{Sudanese bibliography, Arabic Sources, 187J-1967). N.p. [Beirut], 1968.
Daly, M. W., 'The development of the Governor-Generalship of the Sudan,
1899-1934', J. Afr. Hist., 198}, 24, 1, 77-96.
Sudan (Select annotated bibliography, World Bibliographical Series, vol. 40).
Oxford/Santa Barbara, Calif., 1983.
Dietrich, E. L., 'Der Mahdi Mohammed Ahmed vom Sudan nach arabischen
Quellen', Der Islam, 1925, 14, 199-288. (Partial translation of Naum
Schoucair, q.v.)
Douin, G., Khedive Ismail, in: Uempire africain {see under ch. 1).
Evans-Pritchard, E. E., The Aânde: history and political institutions. Oxford,
1971.
Gessi, R., Sette anni nel Sudan Egiîano. Milan, 1891; 2nd edn, 1930.
English trans., Seven years in the Soudan. London, 1892.
Geyer, F. X., Durch Sand, Sump/ und Wald: Missionsreisen in Zentral-Afrika.
Freiburg im Breisgau, 1914.
Ghada Hashem Talhami, Suakin and Massava under Egyptian rule, I86J-I88J.
Washington, D.C., 1979.
Gray, J. R., Southern Sudan {see under ch. 1).
' Islam in the southern Sudan' in Brett, (ed.), Northern Africa {see above).
Great Britain, Parliamentary Paper, Report on the Soudan by Lieutenant-Colonel
Stewart. Egypt, no. 11 (1883) [C.3670].
Hake, A. Egmont (ed.), The journals of Major-Gen. C. G. Gordon, C.B., at
Khartoum. L o n d o n , 1885.
Hallam, W. K. R., The life and times of Rabih Fadl Allah. Ifracombe, 1977.
Hassan Ahmed Ibrahim, ' Mahdist risings against the Condominium Govern-
ment in the Sudan, 1900-1927', Int. J. Afr. Hist. Studies, 1979, 12, 3,
440-71.
' The policy of the Condominium government towards the Mahdist political
prisoners, 1898-1932', Sudan Notes and Records, 1974, 55, 33-45.
Hil\,G.Birkbcck{ed.),Colonel Gordon in central Africa, 1874-/879. London, 1881.
Hill, R. L., A bibliography of the Anglo-Egyptian Sudan, from the earliest times to
19)7. London, 1939.
877

BIBLIOGRAPHY
' T h e G o r d o n literature', Durham University J., 1955, n e w series, 47, 3,

97-101.
Egypt in the Sudan {see under ch. 1).
' The period of Egyptian occupation [of the Sudan], 18 20-18 81', Sudan Notes
and Records, 1959, 40, 100—6.
Slatin Pasha. London, 1965.
Sudan Transport. London, 1965.
' Government and Christian missions in the Anglo-Egyptian Sudan', Middle
Eastern Studies, 1965, 1, 2, i i } - ) 4 .
A biographical dictionary of the Sudan, 2nd edn. London, 1967. (1st edn: A
biographical dictionary of the Anglo-Egyptian Sudan. Oxford, 1951.)
Hodnebo, K.,' Cattle and flies, a study of cattle keeping in Equatoria Province,
the southern Sudan, 1850-1950'. Cand. Fil. thesis, Bergen, 1981.
Hofmayr, W., Die Schilluk: Geschichte, Religion und Leben eines Nilotenstammes.
M o d l i n g (Vienna), 1925.
Holt, P. M . , ' T h e archives o f the Mahdia', Sudan Notes and Records, 1955, 36,
1, 71—80.
A calendar of the correspondence of the Khalifa 'Abdallahi and Mahmoud Ahmed,
A.H. i)i j 11897-8 A.D. Khartoum, 1950 [sic pro 1956].
' The source-materials of the Sudanese Mahdia' in St Antony's Papers, no. 4:
Middle Eastern Affairs, 1, 108-18. London, 1958.
' The place in history of the Sudanese Mahdiya', Sudan Notes and Records,
»959> 4°. 107-12.
The Mahdist state {see under ch. 1).
Studies in the history of the Near East. London, 1973. (Therein:' Holy Families
and Islam in the Sudan';' Modernization and Reaction in the nineteenth-
century Sudan'.)
Holt, P. M. and Daly, M. W., The history of the Sudan {see under ch. 1).
'Ismat Hasan Zulfu, Karart: tahlil 'askari li ma'raka Umdurmdn {A military
analysis of the battle of Omdurman). Khartoum, 1973.
English trans. (P. Clark), Karari: the Sudanese account of the battle of Omdurman.
London, 1980.
Jackson, H. C , Black ivory: or the story of El Zubeir Pasha. Oxford, 1913;
reprinted New York, 1970.
Osman Digna. London, 1926.
Johnson, D. H., 'Tribal boundaries and border wars: Nuer-Diuka relations
in the Sobat and Zaraf valleys, c. 1860-1976', / . Afr. Hist., 1982, 23, 2,
183-203.
Kapteijns, L. E. M., 'The religious background of the Mahdi and his mov-
ement', African Perspectives (Leiden), 1977, 2, 1, 61-79.
'Mahdist faith and Sudanic tradition: history of Dar Masallt, 1870-1930'.
D.Litt. thesis, Amsterdam, 1982.
Magnus, P., Kitchener: portrait of an Imperialist. London, 1958.
Mekki Abbas, The Sudan question. London, 1952.
Mounteney-Jephson, A. J., Emin Pasha and the rebellion at the Equator. London,
1890; reprinted New York, 1969.
Muddathir 'Abd al-Rahlm, Imperialism and nationalism in the Sudan. Oxford, 1969.
878

I I N I L E BASIN A N D T H E EASTERN HORN
Muhammad Ahmad al-Hajj, 'HayatG b. Sa'Id: a revolutionary Mahdist in the
Western Sudan', in YQsuf Fadl Hasan (ed.), Sudan {see under ch. i).
Muhammad Fu'ad Shukri {see also Shukry, M. F.), al-Hukm al-Misrifi-l-Sudan
{i820-188j). Cairo, 1945.
Muhammad Ibrahim Ahmad Abu Salim, Mansburat al-Mahdiyya {official publi-
cations of the Mabdij/ya). [Beirut], 1969.
Al-mursbid ila wathaiq al Mabdi {Guide to the records of the Mabdi). Khartoum,
1969.
Muhammad Sa'Id al-Qaddal, Al-Mahdiyya wa'l-Habasba {The Mabdyya and
Ethiopia). Khartoum, 1973.
Musa al-Mubarak al-Hasan, Ta'rtkh Ddr Fur al-siyasi, 1882-1898 {Political
history of Dar Fur). Khartoum, 1971.
Na'um Shoucair (NaOm Shuqayr), Ta'rikb al-Sudan al-qadim wa''l-baditb... {The
ancient and modern history of the Sudan). Cairo, n.d. [1903].
O'Fahey, R. S., State and society in Dar Fur. New York, 1980.
Ohrwalder, J. {see Wingate, F. R., with Ohrwalder, J.).
Paul, A., A history of the Beja tribes of the Sudan. Cambridge, 1954; reprinted
London, 1971.
Rehfisch, F. (ed.),' Omdurman during the Mahdiya' (trans, from P. Rossignoli,
I mieidodiciannidiprigionia in meigp ai Dervisci, Mondovi, 1898), Sudan Notes
and Records, 1967, 48, 33-61.
Silih Muhammad Nur,' A critical edition of the memoirs of Yfisuf Mikhail'.
Sanderson, G. N., ' European powers and the Sudan' {see under ch. 2).
'Contributions from African sources to the history of European competition
in the upper valley of the Nile', / . Afr. Hist., 1962, 3, 1, 69—90.
'The modern Sudan, 1820-1956: the present position of historical studies',
/ . Afr. Hist., 1963, 4, 3, 435-6i-
' Negus Menelik' {see under ch. 2).
England, Europe and Upper Nile {see under ch. 2).
'Ethiopia and the Mahdist State' {see under ch. 2).
'Aspects of resistance to British rule in the Southern Sudan, 1900-1928',
in Etudes africaines offertes a Henri Brunscbwig. Paris, 1982.
Sanderson, L. M. Passmore, and Sanderson, G. N., Education, religion and
politics in southern Sudan, 1899-1964. London, 1981.
Santandrea, S., A tribal history of the western Babr el Gbaâl. Bologna, 1964.
Luci e ombre dell'amministraîone Britannica nel Babr el Ghaâl {1898—19JJ).
Cairo, 1967.
A popular history of Wau {Babr el Gbaâl, Sudan). Rome, 1977.
Santi, P. and Hill, R. L., The Europeans in the Sudan, 1834-1878. Oxford,
1980.
Schweinfurth, G. et al. (eds.), Emin Pasba in central Africa. London, 1888.
Schweitzer, G., Emin Pascba. Eine Darstellungseines Lebensund Wirkens... Berlin,
1898.
English trans., Emin Pasba: bis life and work (introd. R. W. Felkin). London,
1898.
Shaked, H., The life of the Sudanese Mabdi (edition, with condensed paraphrase,
879

BIBLIOGRAPHY
of a contemporary hagiography by Isma'Il 'Abd al-Qadir al-Kurdufanl).

New Brunswick, N.J., 1978.
Shibeika, M. [Makkl Shibayka], British policy in the Sudan, 1882-1902. London,
1952.
Shukry, M. F. (see also Muhammad Fu'ad Shukrl), The Khedive Ismail and slavery
in the Sudan {jt6)—iiyf). Cairo, 1957.
Equatoria (see under ch. 2).
Slatin, R. C , Fire and sword in the Sudan. London, 1896.
Smith, I. R., Emm Pasha relief expedition (see under ch. 2).
Stevenson, R. C , 'Old Khartoum, 1821-1885', Sudan Notes and Records, 1966,
47. i-3«-
Sudan Notes and Records, Subject and author index, vols. 1—55 (1918-1974).
Khartoum, 198}.
Theobald, A. B., 'The Khalifa Abdullahi', Sudan Notes and Records, 1950, 31,
2, 254-73.
The Mahdiya. London, 1951.
'Alt Dinar, last Sultan of Darfur, 1898—1916. London, 1965.
Thuriaux-Hennebert, A., Les Zande dans I'Histoire du Bahr et Ghaâl et de
rEquatoria. Brussels, 1964.
Trimingham, J. S., Islam in the Sudan. London, 1949; reprinted London, 1965.
Triulzi, A., 'Northwestern Wallagga' (see under ch. 1).
Voll, J. O., 'A history of the Khatmiyyah Tariqah in the Sudan'. Ph.D. thesis,
Harvard, 1969.
Historical dictionary of the Sudan. Metuchen, N.J./London, 1978.
Warburg, G. R., 'The Governor-General of the Sudan and his relations with
the British Consuls-General in Egypt, 1899—1916', Asian and African
Studies (Annual of the Israel Oriental Society), 1969, 5, 97-152.
' Slavery and labour in the Anglo-Egyptian Sudan', Asian and African Studies
(Journal of the Israel Oriental Society), 1978, 12, 221-45.
'Popular Islam and tribal leadership in socio-political structure of North
Sudan', in Milson, M. (ed.), Society and political structure in the Arab World.
New York, 1974.
The Sudan under Wingate. London, 1971.
'Religious policy in the Northern Sudan: 'Ulama' and Sufism, 1899—1918',
Asian and African Studies (Annual of the Israel Oriental Society), 1971, 7,
89-119.
'Sudan, Egypt and Britain' (see under ch. 11 A).
'British rule in the Nile valley' (see under ch. 1 iA).
Ward, J., Our Sudan: its pyramids and progress. London, 1905.
Wingate, F. R., Mabdiism and the Egyptian Sudan. London, 1891; 2nd edn, introd.
P. M. Holt, London, 1968.
(with Ohrwalder, J.,), Ten years' captivity in the Mahdfs camp, 1882-1892.
London, 1892.
Wingate, R., Wingate of the Sudan. London, 1955.
Yusuf Fadl Hasan (ed.), Sudan in Africa (see under ch. 1).
Zaghi, C , Vita di Romolo Gessi. Milan, 1939.
Gordon, Gessi e la reconquista del Sudan. Florence, 1947.
880

C. Ethiopia
Bartnicki, A. and Mantd-Niecko, J., Gescbicbte Atbiopiens-von den Anfangen
bis ûr Gegenwart, German trans, from the Polish, 2 vols. Berlin, DDR,
1978.
Berkeley, G. F. H., Tbe campaign of' Adowa and the rise oj' Mtntlik. London, 1902
(reprinted, 1955).
Brown, C. F., Ethiopian perspectives: a bibliographicalguide to tbe history of Ethiopia.
Westport, Conn./London, 1978.
Caulk, R. A., 'The origins and development o f the foreign policy of Menelik
II, 1865-1896'. Ph.D. thesis, London, 1966.
'Menilek and the Ethio-Egyptian war of 1875-76: a reconstruction of
source material', Rural Africana, 1970, 11.
' Yohannes IV' (see under ch. 2).
'The occupation of Harar: January 1887', / . Ethiopian Studies, 1971, 9, 2.
' Firearms and princely power in Ethiopia in the nineteenth century', J. Afr.
Hist., 1972, 13, 4, 609-30.
'Religion and state in nineteenth-century Ethiopia', J. Ethiopian Studies,
1972, 10, 1, 23-41.
'Harar Town' (see under ch. 1).
Conti Rossini, C , Italia ed Etiopia (see under ch. 2).
Crummey, D., 'Abyssinian feudalism', Past and Present, 1980, 89, 115-38.
'Family and property among the Amhara nobility', J. Afr. Hist., 1983, 24,
2, 207-20.
Darkwah, R. H. K., Menelik of Ethiopia. London, 1972.
Sbewa, Menilek... (see under ch. 1).
Darley, H., Slaves and ivory. London, 1926; reprinted (with a foreword by
A. G. Freeman) Northbrook, 111., 1972.
Douin, G., Regie du Kbe'dive Ismail (see under ch. 1).
Erlich, H.,' Alula," The son of Qubi": a " King's Man " in northern Ethiopia',
/ . Afr. Hist., 1974, 15, 2, 261-74.
Gabre-Sellassie, Z., Yohannes IV (see under ch. 2).
Garretson, P., "The history and development of Addis Ababa, 1886-1910'.
Ghada Hashcm Talhami, Suakin and Massawa (see under ch. 1 iB).
Giglio, C , L'Italia in Africa (see under ch. 2).
Gilmour, T. L., Tbe Ethiopian railway and tbe Powers. London, 1906.
Gleichen, A. E. W. (Count v.), With tbe mission to Menelik. London, 1898.
Great Britain, War Office, Precis of information obtained by tbe British mission to
Abyssinia. London, 1897 (restricted circulation).
Greenfield, R., Ethiopia: a new political history. London, 1965; 2nd edn
1967.
Guebre Sellassie, Cbronique du regne de Menelik II, Roi des Rois d'Etbiopie (ed.
M. de Coppet), 2 vols. Paris, 1930-1.
Haberland, E., Galla Sid-Atbiopiens. Stuttgart, 1963.
Holt, P. M., Mabdist State (see under ch. 1).
881

BIBLIOGRAPHY
Huntingford, G. W. B., The Galla of Ethiopia; the kingdoms of Kaffa and Jan/ero.
L o n d o n , 1955.
Jesman, C , The Russians in Ethiopia. London, 1958.
Keller, C , Alfred Ilg, sein Leben und sein Wirken. Frauenfeld (Leipzig), 1918.
Levine, D. N., Wax and gold: tradition and innovation in Ethiopian culture. Chicago
and London, 1965.
Greater Ethiopia: the evolution of a multi-ethnic society. Chicago, 1974.
Lewis, H. S., Galla monarchy (see under ch. 1).
Loepfe, W. Alfred Ilg und die athiopische Eisenbahn. Zurich, 1974.
Mantel-Niecko, J., The role of land-tenure in the system of Ethiopian imperial
government in modern times. Warsaw, 1980.
Marcus, H. G.,' Ethio-British negotiations concerning the western border with
Sudan, 1896-1902', J. Afr. Hist., 1963,4, 1, 81-94.
'A preliminary history of the Tripartite Treaty of December I J , 1906', / .
Ethiopian Studies, 1964, 2, 1, 21-40.
'The foreign policy of the Negus Menelik, 1896-1898: a rejoinder',/. Afr.
Hist., 1966, 7, 1, 117-22.
'Menilek II* in Bennett, N. R. (ed.), Leadership in Eastern Africa. Boston,
Mass., 1968.
'Imperialism and expansionism in Ethiopia from 1865 to 1900', in Gann,
L. H. and Duignan, P. (eds.), Colonialism in Africa (see under General).
The modern history of Ethiopia and the Horn of Africa: a select and annotated
bibliography. Palo Alto, Calif., 1972.
Menelik II (see under ch. 2).
Mathew, D , J., Ethiopia. The study of a polity, i) 40-19jj. London, 1947.
Michel, C , Vers Fachoda.. .a travers I Ethiopie. Paris, 1900.
Muhammad Sa'Id al-Qaddal, Al-Mahdiyya (see under ch. 1 iB).
Pankhurst, R., 'Menelik and the foundation of Addis Ababa', / . Afr. Hist.,
1961, 2, 1, 103—17.
'The trade of Northern Ethiopia in the nineteenth and early twentieth
centuries', J. Ethiopian Studies, 1964, 2, 1.
'The Ethiopian slave-trade in the 19th and early 20th centuries', J. Semitic
Studies, 1964, 9.
State and land in Ethiopian history. Addis Ababa, 1966.
'Menelik and the utilization of foreign skills', Ethiopia Observer, 1967, 5.
Economic history of Ethiopia (1968) (see under ch. 1).
Perham, M., The government of Ethiopia. London, 1948; reprinted (with
introduction by C. Clapham) 1969.
Portal, G. H., My mission to Abyssinia. London, 1892. Earlier, privately-printed,
version: An account of the English mission to King fohannis in I8SJ.
Winchester, [1888].
Prouty, C. and Rosenfeld, E., Historical dictionary of Ethiopia. Metuchen,
N.J./London, 1981.
Ram, K. V., 'British government, finance capitalists and the French Jibuti -
Addis Ababa railway 1898-191}', / . Imp. C'wealth Hist., 1981, 9, 2,
146-68.
Rosenfeld, C. P., A chronology of Menilek II of Ethiopia. East Lansing, Mich.,
882

Rubenson, S., 'Some aspects of the survival of Ethiopian independence in the

period of the Scramble for Africa', in Historians in tropical Africa, Salisbury
(Rhodesia) i960; also, University College Review, Addis Ababa, 1961.
WicbaleXVU (1964), (see under ch. 2).
'The Protectorate paragraph of the Wichale treaty', J. Afr. Hist., 1964, 5,
2, 243-8}.
Survival (1976), (see under ch. 1).
Sanderson, G. N., 'African Sources' (see under ch. 11B).
'Negus Menelik' (1964), (see under ch. z).
England, Europe and Upper Nile (1965), (see under ch. 2).
'Conflict and cooperation' (see under ch. 1 iB).
' Sudanese factors in the history of Ethiopia in the nineteenth century', in
Yusuf Fadl Hasan (ed.), Sudan (see under ch. 1).
Silberman, L., 'Why the Haud was ceded', Cabiers d'etudes africaines, 1961, 2,
37-83.
Soleillet, P., Obock, le Choa, le Kaffa. Paris, 1886.
Trimingham, J. S., Islam in Ethiopia (see under Ch. 1).
Triulzi, A., 'Northwestern Wallagga' (see under ch. 1).
Ullendorff, E., The Ethiopians. London, i960.
Weissleder, W., 'The political ecology of \mhara domination'. Ph.D. thesis,
Chicago, 1965.
Wylde, A. B., Modern Abyssinia. London, 1901.
Zaghi, C , Crispi e Menelich nel diario inedito del conte Augusto Salimbeni. Turin,
1956.
I Russi (197}), (see under ch. 2).
D. Somalia
'Abd al-Sabur Marziiq, TbeTir min al-Sumdl: al-Mulld Muhammad' Abdalldh
Hasan (A Somali revolutionary: the Mullah M. 'A. H.). Cairo, 1964.
Andrzejewski, B. W. and Lewis, I. M., Somali Poetry.. Oxford, 1964.
Brockett, A. M., 'The British Somaliland Protectorate to 1905'. D.Phil, thesis,
Oxford, 1969.
Caroselli, F. S., Ferro efuoco in Somalia. Rome, 1931.
Cerulli, E., Somalia (see under ch. 1).
Douin, G., Regne du Khedive Ismail (see under ch. 1).
Drysdale, J., The Somali Dispute. London, 1964.
Great Britain, War Office, Official history of the operations in Somaliland,
1901-1904, 2 vols. London, 1907.
Hess, R. L., 'The "Mad Mullah" and Northern Somalia', / . Afr. Hist., 1964,
S» 3.4IJ-33-
Italian colonialism (1966), (see under ch. 2).
'The Poor Man of God - Muhammad Abdullah Hasan', in Bennett, N. R.
(ed.), Leadership in East Africa. Boston, Mass., 1968.
Jaamac Cumat Ciise (Jama' 'Umar 'Isa), Diiwaanka Gabayadii Sayid Maxamad
Cabdulle Xasan (Collected poems of Muhammad 'Abdullah Hasan).
Mogadishu, 1974.
883

BIBLIOGRAPHY
Taariikbdii Daraawiishta iyo Sayid Maxamad Cabdulle 'Hasan (History of the
Dervish movement of Muhammad' Abdallah Hasan). Mogadishu, 1976.
Jama' 'Umar 'Isa (Jaamac Cumar Ciise), Ta'rikb al-Sumal fi"l-'usur al-wusta
wa'l-hadttha (History of the Somali in medieval and modern times). Cairo,
1965.
Jardinc, D . , The Mad Mullah of Somaliland. London, 1923; reprinted New York,
1969.
Lewis, I. M., A pastoral democracy. London, 1961; reprinted 1967.
'Conformity and contrast in Somali Islam', in Lewis, I. M. (ed.), Islam in
tropical Africa. London, 1966.
Modern history of Somalia (1980), (see under ch. 1).
Martin, B. G., Muslim brotherhoods in nineteenth-century Africa. Cambridge, 1976.
Samatar, S. S., Oral poetry and Somali nationalism: the case of Sayyid Mohammad
'Abdille Hasan. Cambridge, 1982.
Somali Government, The Somali peninsula: new light on Imperial motives. Revised
edn, London, 1962.
Swayne, H. G. C , Seventeen trips through Somaliland. London, 1895.
Turton, E. R.,' The impact of Mohammad Abdille Hasan in the East African
Protectorate', 7. Afr. Hist., 1969, 10, 4, 641-57.
12. THE EUROPEAN SCRAMBLE AND CONQUEST IN AFRICAN

HISTORY
Abu-Lughod, I., 'The transformation of the Egyptian elite: prelude to the

'Urabl Revolt', Middle Eastern J., 1967, ai, 3, 325-44.
Ajayi, J. F. A., Christian missions in Nigeria 1841-1891: the making of a new elite.
London, 1965.
Ajayi, J. F. A. and Smith, R. S., Yoruba warfare in the nineteenth century.
Cambridge, 1964.
Akintoye, S. A., Revolution and power politics in Yorubaland, 1840-189}. London,
1971.
Amin, Samir (ed.), Modern migrations in Western Africa. Oxford, 1974.
Arhin, K., 'The economic and social significance of rubber production and
exchange in the Gold and Ivory Coasts, 1880-1900', Cahiers d''etudes
africaines, 1980, 20, 1-2, 49-62.
'Rank and class among the Asante and Fame in the nineteenth century',
Africa, 1983, 53, 1: 2-22.
Austen, R. A., Northwest Tanzania under German and British rule. New Haven
and London, 1968.
Awe, B . , ' Militarism and economic development in nineteenth century Yoruba
country: the Ibadan example', J. Afr. Hist., 1973, 14, 1, 65-78.
Baier, S., An economic history of Central Niger, Oxford, 1980.
Ballard, C , 'The role of trade and hunter-traders in the political economy of
Natal and Zululand, 1824-1880', Afr. Eicon. Hist., 1981, 10, 3-21.
Beach, D . N., '"Chimurenga": the Shona rising of 1896-97', / . Afr. Hist.,
1979, 20, 3, 395-4*°-
Beinart, W., The political economy of Pondoland i860—19jo. Cambridge, 1982.
884

12 SCRAMBLE AND CONQUEST IN AFRICAN HISTORY
Berry, S. S., Cocoa, custom and socio-economic change in rural western Nigeria. Oxford,
>975-
Berque, J., French North Africa: the Maghrib between two world wars. London,
1967.
Biobaku, S. O., The Egba and their neighbours 1842-1872. Oxford, 1957.
Boahen, A., 'Politics in Ghana, 1800-1874', in J. F. A. Ajayi and M. Crowder
(eds.), West Africa, 11 {see under General).
Bonner, P., Kings, commoners and concessionaires {see under ch. 7/8).
Bozzoli, B., ' Marxism, feminism and South African studies', J. Southern Afr.
Studies, 198}, 9, 2, 139—71.
Brockway, L. H., Science and colonial expansion: the role of the British botanic gardens.
New York, 1979.
Bundy, C , The rise and fall of the South African peasantry. London, 1979.
Burke m , E., Prelude to protectorate in Morocco {see under ch. 1).
Cain, P. J. and Hopkins, A. G,., 'The political economy of British expansion
overseas, 1750-1914', Econ. Hist. Review, 1980, 2nd series, 33, 4, 463—90.
Caulk, R. A., 'Firearms and princely power in Ethiopia in the nineteenth
century', / . Afr. Hist., 1972, 13, 4, 609-30.
Churchill, Winston, My African journey. London, 1908.
Cobbing, J., 'The absent priesthood: another look at the Rhodesian risings
of 1896-1897',/ Afr. Hist., 1977, 18, 1, 61-84.
Conrad, Joseph, Heart of darkness. Penguin edn, Harmondsworth, 1973.
Cooper, F., Plantation slavery on the east coast of Africa. New Haven and London,
'977-
From slaves to squatters {see under ch. 10).
Curtin, P. et al., African History {see under General).
Curtis, L., With Milner in South Africa. Oxford, 1951.
Dachs, A. J., 'Missionary imperialism - the case of Bechuanaland', / . Afr.
Hist., 1972, 13, 4, 647-58.
Dawson, M. H., 'Disease and population decline of the Kikuyu of Kenya,
1890—1925', in African Historical Demography, vol. 11. Edinburgh, 1981.
De Kiewiet, C. W., History of South Africa {see under chs. 7/8).
Delius, P. N., "The Pedi polity' {see under chs. 7/8).
Denoon, D., 'Participation in the "Boer War"' {see under chs. 7/8).
Dias, J. R., 'Black chiefs, white traders and colonial policy near the Kwanza:
Kabuku Kambilo and the Portuguese, 1873-1896', J. Afr. Hist., 1976,
17, 2, 245-65.
'Famine and disease in the history of Angola r.i83o-i93o', J. Afr. Hist.,
1981, 22, 3, 349-78-
Dike, K. O., Trade and politics in the Niger Delta, I8JO-I88J. Oxford, 1956.
Dorward, D. C, 'Ethnography and administration: a study of Anglo-Tiv
"Working misunderstanding"',/. Afr. Hist., 1974, 15, 3, 457-77.
Dumett, R., 'The rubber trade' {see under ch. 1).
Dunn, J. and Robertson, A. F., Dependence and opportunity: political change in
Abafo. Cambridge, 1973.
Dunn, R. E., 'Bu Himara's European connexion: the commercial relations of
a Moroccan warlord', J. Afr. Hist., 1980, ai, 2, 235-53.
885

BIBLIOGRAPHY
Ellis, S., 'Collaboration and resistance in Madagascar 1895-1899, with special

reference to the Kingdom of Imerina'. D.Phil. Oxford, 1980.
Etherington, N., Preachers, peasants and politics in southeast Africa, i8jr-t88o:
African Christian communities in Natal, Pondoland and Zululand. London,
1978.
' Theories of imperialism in Southern Africa revisited', African Affairs, 1982,
81, 324, $85-407.
Fieldhouse, D. K., Unilever overseas: the anatomy of a multinational, 189^-196).
London, 1978.
Ford, J., The role of the Trypanosomiases in African ecology, a study of the tsetse-fly
problem. Oxford, 1971-
Gellner, E., Saints of the Atlas. London, 1969.
Gertzel, C. J., 'Relations between African and European traders in the Niger
Delta, X880-1896', J. Afr. Hist., 1962, 3, 2, 361-6.
Giliomee, H., 'Historical introduction' to Andre du Toit and H. Giliomee
(eds.), Afrikaner political thought, vol. n: itjo-ip/o. Berkeley,
forthcoming.
Girouard, Sir Percy, Memoranda for Provincial and District Commissioners.
Nairobi, 1910.
Goodfellow, C. F., Great Britain and South African Confederation, 1870-1881.
Cape Town, 1966.
Goody, J., Technology, tradition, and the state in Africa. Oxford, 1971.
Gordon, C. T., The growth of the Boer opposition to Kruger, 1890-189 j . Cape Town,
1970.
Goucher, C. L., 'Iron is iron 'til it is rust: trade and ecology in the decline
of West African iron-smelting', J. Afr. Hist., 1981, 22, 2, 179-89.
Gray, R., A history of the Southern Sudan 18)9—1889. Oxford, 1961.
Guy, J., The destruction of the Zulu kingdom {see under ch. 1).
Hancock, W. K., Smuts, vol. 1: The sanguineyears, 1870-1919. Cambridge, 1962.
Harms, R., 'The end of red rubber: a reassessment', J. Afr. Hist., 1975, 16,
1, 73-88-
Harms, R. W., River of wealth, river of sorrow: the central Zaire basin in the era of
the slave and ivory trade, 1)00-1891. New Haven and London, 1981.
Harries, P., 'Slavery, social incorporation and surplus extraction: the nature
of free and unfree labour in South-East Africa', / . Afr. Hist., 1981, 22,
3, 309-30.
Harris, J. H., Dawn in darkest Africa. London, 1912.
Headrick, D . R., The tools of empire: technology and European imperialism in the
nineteenth century. New York and Oxford, 1981.
Hill, P., The migrant cocoa-farmers of southern Ghana: a study in rural capitalism.
Cambridge, 1963.
Population, prosperity and poverty: rural Kano, 1900 and 1970. Cambridge, 1977.
Hobson, J. A., The war in South Africa: its causes and effects. London, 1900.
Hodgkin, J., Nigerian perspectives, an historical anthology. London, i960.
Holt, P. M., The Mabdist State {see under ch. 1).
Hopkins, A. G., 'The Lagos strike of 1897: an exploration in Nigerian labour
history', Past and Present, 1966, 35, 133—55.
886

12 SCRAMBLE A N D CONQUEST IN A F R I C A N HISTORY
'Economic imperialism in West Africa: Lagos, 1880-1891', Econ. Hist.
Review, 1968, 2nd series, 21, 3, 580-606.
An economic history of West Africa. London, 1973.
'Imperial business in Africa, Part 2: Interpretations',/. Afr. Hist., 1976, 17,
2, 267—90.
'Innovation in a colonial context: African origins of the Nigerian cocoa-
farming industry, 1880—1920', in C. Dewey and A. G. Hopkins (eds.), The
imperial impact: studies in the economic history of Africa and India. London,
1978.
Huxley, E., White man's country. Lord Delamere and the making of Kenya, 1:
1S70—1914. L o n d o n , 1935.
Hyam, R., Elgin and Churchill at the Colonial Office, jyoj-iyoS: the watershed of
the Empire-Commonwealth. London, 1968.
Britain's imperial century, iSij—if/4. London, 1976.
Hynes, W. G.,' British mercantile attitudes towards imperial expansion', Hist.
]., 1976, 19, 4, 969-79.
Iliffe, J., Tanganyika under German rule ipoj-ipu. Cambridge, 1969.
A modern history of Tanganyika. Cambridge, 1979.
The emergence of African capitalism. London, 1983.
Isaacman, A., 'Social banditry in Zimbabwe (Rhodesia) and Mozambique,
1894-1907: an expression of early peasant protest',/. Southern Afr. Studies,
1977,4, 1, 1-30.
Isichei, E., The Ibo people and the Europeans. London, 1973.
Kanya-Forstner, A. S., The conquest of the western Sudan: a study in French military
imperialism. Cambridge, 1969.
'Military expansion in the western Sudan - French and British style', in
P. Gifford and W. R. Louis (eds.), France and Britain in Africa {see under
General).
Kay, G. B., The political economy of colonialism in Ghana. Cambridge, 1972.
Kimble, D., A political history of Ghana: the rise of Gold Coast nationalism,
i8;o-if28. Oxford, 1963.
Kitching, G., Class and economic change in Kenya: the making of an African petite
bourgeoisie ipoj-ipyo. New Haven and London, 1980.
Klein, M. A., Islam and imperialism in Senegal: Sine-Saloum, 1S47-1914. Edin-
burgh, 1968.
Law, R., The horse in West African history. Oxford, 1980.
Lloyd, T., 'Africa and Hobson's imperialism', Past and Prei , 1972, 55,
'3O-53-
Lonsdale, J. M., 'The politics of conquest: the British in Western Kenya,
1894-1908', Hist. J., 1977, 20, 4, 841-70.
Lonsdale, J., 'States and social processes in Africa: a historiographical survey',
Afr. Studies Review, 1981, 24, 2/3, 139—225.
Lonsdale, J. and Bcrman, B.,' Coping with the contradictions: the development
of the colonial state in Kenya, 1895-1914', / . Afr. Hist., 1979, 20, 4,
487-505.
Lovejoy, P., ' Slavery in the context of ideology', in P. Lovejoy (ed.), The
ideology of slavery in Africa. Beverley Hills and London, 1981.
887

BIBLIOGRAPHY
Lovejoy, P. E., Transformations in slavery: a history ofslavery in Africa. Cambridge,

1983.
Low, D. A . , ' Warbands and ground-level imperialism in Uganda, 1870-1900',
Hist. Studies, 1975, 16, 65 (Melbourne), 584-97.
Low, D . A. and Pratt, R. C , Buganda and British overrule: two studies. London,
i960.
McCaskie, T. C,'Innovational eclecticism: the Asante empire and Europe in
the nineteenth century', Comp. Studies in Soc. and Hist., 1972,14, 1, 30-45.
'Office, land and subjects in the history of Manwerc fekuo of Kumase: an
essay in the political economy of the Asante state', / . Afr. Hist., 1980,
21, 2, 189-208.
'State and society, marriage and adultery: some considerations towards a
social history of pre-colonial Asante', / . Afr. Hist., 1981, 22, 4, 477-
94.
' Accumulation, wealth and belief in Asante history. I: to the close of the
nineteenth century', Africa, 1983, 53, 1, 23-43.
McNeill, W. H., The pursuit of power: technology, armed force and society since
A.D. 1000. Oxford, 1983.
Mann, K.,' The dangers of dependence: Christian marriage among elite women
in Lagos Colony, 1880-1915', / . Afr. Hist., 1983, 24, 1, 37-56.
Manning, P., Slavery, colonialism and economic growth in Dahomey, 1640-1960.
Cambridge, 1982.
Markakis, J., Ethiopia, anatomy of a traditional polity. Oxford, 1974.
Marks, S. and Trapido, S., 'Lord Milner and the South African State', History
Workshop, 1979, 8, 50-80.
Matson, A. T., Nandi resistance to British rule, 1890-1906. 1. Nairobi, 1972.
Mayer, P., Townsmen or tribesmen. Cape Town, 1961.
Miller, J. C , 'Cokwe trade and conquest in the nineteenth century', in R. Gray
and D . Birmingham (eds.), Pre-colonial African trade. London, 1970.
Kings and kinsmen: early Mbundu states in Angola. Oxford, 1976.
'The significance of drought, disease and famine in the agriculturally
marginal zones of West-Central Africa'. J. Afr. Hist., 1982, 23, 1, 17-62.
Nasson, W. R., 'Black society in the Cape Colony and the South African War
of 1899—1902: a social history'. Ph.D. Cambridge, 1983.
Newbury, C. W., The western Slave Coast and its rulers. Oxford, 1961.
Newbury, C , 'The economics o f conquest in Nigeria, 1900-1920: amal-
gamation reconsidered', in Etudes africaines offertes a Henri Brunscbwig.
Paris, 1982.
Nicholls, C. S., The Swahili coast: politics, diplomacy and trade on the East African
littoral, 1798—18}6. London, 1971.
O'Brien, D . B. C , The Mouridts of Senegal: the political and economic organisation
of an Islamic brotherhood. Oxford, 1971.
Ochieng', W. R.,' Undercivilisation in Black Africa', Kenya Hist. Review, 1974,
a, «, 45-57-
Oliver, R., Sir Harry Johnston (see under ch. 1).
Oloruntimehin, B. O., The Segu Tukulor empire. London, 1972.
Owen, R., The Middle East and the world economy 1800—1914. London, 1981.
888

Packard, R. M., Chief ship and cosmology: an historical study of political competition.
Bloomington, 1981.
Pankhurst, R., "The great Ethiopian famine of 1888-92: a new assessment',
/ . Hist, of Medicine, 1966, a i , 2, 95-124; and 3, 271-94.
Parsons, N., 'The economic history of Khama's country in Botswana,
1844-1930', in R. Palmer and N . Parsons (eds.), The roots of rural poverty
in Central and Southern Africa. London, 1977.
Peel, J. D . Y., 'Conversion and tradition in two African societies: Ijebu and
Buganda', Past and Present, 1977, 77, 108-41.
Perham, M., Lugard, vol. n: The years of authority 1898-194}. London, i960.
Person, Y., Samort {see under ch. 1).
Phillips, A. M., 'The makeshift settlement: colonial policy in West Africa'. City
of London Univ. Ph.D. thesis, 1982.
Prins, G., The hidden hippopotamus. Cambridge, 1980.
Ranger, T. O., Revolt in Southern Rhodesia 1896-7. London, 1967.
'The invention of tradition in colonial Africa', in E. Hobsbawm and
T. Ranger (eds.), The invention of tradition. Cambridge, 1983.
Richards, A., Land, labour and diet in Northern Rhodesia. London, 1939.
Richardson, P., Chinese mine labour in the Transvaal. London, 1982.
Roberts, R., 'Long distance trade and production: Sinsani in the nineteenth
century',/. Afr. Hist., 1980, 21, 2, 169-88.
'Production and reproduction in warrior states: Segu Bambara and Segu
Tokolor, c. 1712-1890', Int. J. Afr. Hist. Studies, 1980, 13, 3, 389-419.
Roberts, R. and Klein, M.,' The Banamba slave exodus of 1905 and the decline
of slavery in the Western Sudan', / . Afr. Hist., 1980, 21, 3, 375-94.
Robinson, R. E., 'European imperialism and indigenous reactions in British
West Africa, 1880-1914', in H. L. Wesseling (ed.), Expansion and reaction:
essays on European expansion and reactions in Asia and Africa. Leiden, 1978.
Robinson, R. and Gallagher, J., with A. Denny, Africa and the Victorians (see
under General).
Robinson, R. and Gallagher, J., "The partition of Africa', in F. H. Hinsley
(ed.), The new Cambridge modern history, xi. Cambridge, 1962.
Ross, R., Adam Kok's Griquas: a study in the development of stratification in South
Africa. Cambridge, 1976.
Rubenson, S., The survival of Ethiopian independence. London, 1976.
Sanderson, G. N., 'The European partition of Africa: coincidence or con-
juncture?', J. Imp. C'wealtb Hist., 1974, 3, 1, 1—54-
Schapera, I., Tribal innovators: Tswana chiefs and social change, 179J-1940. London,
1970.
Scholch, A., 'Constitutional development in nineteenth-century Egypt: a
reconsideration', Middle Eastern Studies, 1974, 10, 1, 3-14.
'The "Man on the Spot" and the English occupation of Egypt in 1882',
Hist. ]., 1976, 19, 3, 77 3-85.
Seddon, D., Moroccan peasants: a century of change in the eastern Rif, 1870—1970.
London, 1981.
Shea, P.,' The development of an export oriented dyed cloth industry in Kano
emirate in the nineteenth century'. Ph.D. Wisconsin, 1975.

BIBLIOGRAPHY
Shenton, R. W., 'The development of capitalism in Northern Nigeria'. Ph.D.
Toronto, 1981.
Silver, J. B., 'Class struggle and class consciousness: an historical analysis of
mineworkers in Ghana'. Ph.D., Univ. of Sussex, 1981.
Slama, B., L.'insurrection de 1864 en Tunisie. Tunis, 1967.
Smith, M. G., Government in Zavâu. Oxford, i960.
Southall, A., Alur society. Cambridge, 1956.
Steinhart, E. I., Conflict and collaboration: the kingdoms of western Uganda
1890-1907. Princeton, 1977.
Stuart, J. and Malcolm, D. McK. (eds.), The diary of Henry Francis Fynn.
Pietermaritzburg, 1969.
Swanson, M. W., 'The sanitation syndrome: bubonic plague and urban Native
policy in the Cape Colony, 1900-1909',/. Afr. Hist., 1977,18, 3, 387-410.
Swindell, K.,' Serawoollies, Tillibunkas and strange farmers: the development
of migrant groundnut farming along the Gambia river, 1848—95',/. Afr.
Hist., 1980, 21, 1, 93-104.
Szereszewski, R., Structural change in the economy of Ghana, 1891—1911. London,
1965.
Thompson, L., Survival in two worlds: Moshoeshoe of Lesotho, 1786-1870. Oxford,
'975-
Trapido, S., 'The origins of the Cape franchise qualifications of 1853 '>./• Afr-
Hist., 1964, 5. '. 37-54-
'Landlord and tenant in a colonial economy: the Transvaal, 1880-1910', J.
Southern Afr. Studies, 1978, 5, 1, 26—58.
'"The friends of the natives": merchants, peasants and the political and
ideological structure of liberalism in the Cape, 1854-1910', in S. Marks
and A. Atmore (eds.), Economy and society (see under ch. 1).
'Reflections on land, office and wealth in the South African Republic,
1850—1900', in S. Marks and A. Atmore (eds.), Economy and society {see under
ch. 1).
Trebilcock, C , 'War and the failure of industrial mobilization: 1899 and 1914',
in J. M. Winter (ed.), War and economic development. Cambridge, 1975.
Twaddle, M., 'The Bakungu chiefs of Buganda under British colonial rule,
1900—1930', J. Afr. Hist., 1969, 10, 2, 309-22.
'The Muslim revolution in Buganda', African Affairs, 1972, 71, 282, 54-72.
Uzoigwe, G. N., 'Kabalega's Abarusura: the military factor in Bunyoro',
Proceedings of the University of East Africa Social Science Conference 1968/69.
Nairobi, 1969.
'The Mombasa-Victoria railway, 1890-1902: imperial necessity, humani-
tarian venture, or economic imperialism?', Kenya Hist. Review, 1976, 4, 1,
11-34.
Vail, L. and White, L., Capitalism and colonialism in Mozambique: a study of
Quelimane district. London, 1980.
Valensi, L., Fellahs tunisiens: I'economie rurale et la vie des campagnes aux i8e et
/pe sticks. Paris, 1977.
Van Onselen, C , 'Reactions to rinderpest in Southern Africa', / . Afr. Hist.,
1972, 13, 2, 473-88-
890

Chibaro: African mine labour in Southern Khodesia 1900-19}}. London, 1976.
Studies in the social and economic history of the Witwatersrand, vol. 11: New Nineveh.
Johannesburg and London, 1982.
Vansina, J., The children of Woot: a history of the Kuba peoples. Madison, 1978.
'Towards a history of lost corners in the world', Econ. Hist. Review, 1982,
2nd series, 35, 2, 165—78.
Waller, R., 'The traditional economy of Buganda'. M.A. (SOAS) London,
1971.
'The Maasai and the British 1895-1905: the origins of an alliance', / . Afr.
Hist., 1976, 17, 4, 529-53.
Warwick, P., Black people and the South African war 1899-1902. Cambridge, 1983.
Webster, J. B., The African churches among the Yoruba, 1888—1922. Oxford, 1964.
Weiskel, T. C , French colonial rule and the Baule peoples: resistance and collaboration
1899-1911. Oxford, 1980.
Welsh, D., The roots of segregation: Native policy in colonial Natal, 184J-1910. Cape
Town, 1971.
White, G., 'Firearms in Africa: an introduction', J. Afr. Hist., 1971, 12, 2,
173-84.
Wilks, I., Political bi-polarity in nineteenth century Asante. Edinburgh, 1971.
Asante {see under ch. 1).
'Land, labour, capital and the forest kingdom of Asante: a model of early
change', in J. Friedman and M. J. Rowlands (eds.), The evolution of social
systems. Pittsburgh, 1978.
'The golden stool and the elephant tail: an essay on wealth in Asante',
Research in Economic Anthropology, 1979, 2, 1-36.
Willis, R., A state in the making: myth, history and social transformation in pre-colonial
Ufipa. Bloomington, 1981.
Wilson, C , The history of Unilever: a study in economic growth and social change, 1.
London, 1954.
Wilson, F., Labour in the South African gold mines 1911-1969. Cambridge, 1972.
Wright, M., Buganda in the heroic age. Nairobi, 1971.
Wrigley, C. C , 'The Christian revolution in Buganda', Comp. Studies in Soc.
and Hist., 1959, 2, 1, 33-48.
Crops and wealth in Uganda: a short agrarian history. Kampala, 1959.
'Neo-mercantile policies and the new imperialism', in C. Dewey and
A. G. Hopkins (eds.), The imperial impact: studies in the economic history of
Africa and India. London, 1978.
891


The Cambridge History of Africa, Volume 6 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Cambridge History of Africa, Volume 6 PDF

Uploaded by

Copyright:

Available Formats

THE CAMBRIDGE HISTORY

Cambridge Histories Online © Cambridge University Press, 2008

i From the earliest Times to c. 500 B.C.

Cambridge Histories Online © Cambridge University Press, 2008

Cambridge Histories Online © Cambridge University Press, 2008

© Cambridge University Press 1985

This book is in copyright. Subject to statutory exception

First published 1985

Printed in the United Kingdom at the University Press, Cambridge

Library of Congress catalogue card number: 76-2261

ISHN o 521 22803 4

British IJbrary Cataloguing in Publication data

Cambridge Histories Online © Cambridge University Press, 2008

Africa on the eve of partition 10

Cambridge Histories Online © Cambridge University Press, 2008

Western Africa, 1870—1886 208

Cambridge Histories Online © Cambridge University Press, 2008

Western Africa, 1886-1905 257

Western Equatorial Africa 298

Southern Africa, 1867-1886 359

Cambridge Histories Online © Cambridge University Press, 2008

8 Southern and Central Africa, 1886-1910 422

9 Portuguese colonies and Madagascar 493

Cambridge Histories Online © Cambridge University Press, 2008

10 East Africa, 1870-1905 539

Cambridge Histories Online © Cambridge University Press, 2008

1 Northern Africa on the eve of partition page 12

Cambridge Histories Online © Cambridge University Press, 2008

Cambridge Histories Online © Cambridge University Press, 2008

In the English-speaking world, the Cambridge histories have

Cambridge Histories Online © Cambridge University Press, 2008

of Africa as an integrated whole rather than — as it had usually

Cambridge Histories Online © Cambridge University Press, 2008

continent, and the setting up of the colonial structures between

Cambridge Histories Online © Cambridge University Press, 2008

work as it progressed from draft to draft, thus protracting the

September 1984 j. D. FAGE

Cambridge Histories Online © Cambridge University Press, 2008

This volume deals with a period of about thirty-five years, from

Cambridge Histories Online © Cambridge University Press, 2008

conquest was, as Dr Lonsdale makes clear in the concluding

Cambridge Histories Online © Cambridge University Press, 2008

feathers and rare skins, and of course the continuing traffic in

Cambridge Histories Online © Cambridge University Press, 2008

Niger. All these hegemonies reached the peak of their power

Cambridge Histories Online © Cambridge University Press, 2008

those of eastern Zaire, the porters and camp followers of the

Cambridge Histories Online © Cambridge University Press, 2008

inadequate, to improve their own defences. Firearms, long

Cambridge Histories Online © Cambridge University Press, 2008

1897, to German East Africa in 1888 and 1905, to German South

Cambridge Histories Online © Cambridge University Press, 2008

of the continent that a different kind of colonialism, initiated much

Cambridge Histories Online © Cambridge University Press, 2008

of the period 1870—1905 in African history arises from local

Cambridge Histories Online © Cambridge University Press, 2008

AFRICA ON THE EVE OF PARTITION

A. NORTH OF THE EQUATOR

Cambridge Histories Online © Cambridge University Press, 2008

THE SHORES OF THE DESERT

The metaphor of the Sahara as a sea, separating the lands to the

Cambridge Histories Online © Cambridge University Press, 2008

i Northern Africa on the eve of partition