The Archaeology of Mesopotamia (History Arts Ebook)

SETON LLOYD
The Archaeology
of Mesopotamia
^/t
from the
Old Stone Age
o the Persian
lonquest
REVISED EDITION
rHvxJ-yf
^^^ ^^^^X
(^M^^P^%Bcceui\>^^^^Sf)
^^^^-"'M W:^"^^!^
LJi
WwM \iecel
\ummiiL
Î^B
^mM ^HrP'
BOSTON
-"f
{'i
IIBRARY
;t, s:
m^^
SETON LLOYD
The Archaeology
of Mesopotamia
From the Old Stone Age
to the Persian Conquest
REVISED EDITION
with 174 illustrations
THAMES AND HUDSON

Any copy of this book issued by the pubhsher as a
paperback is sold subject to the condition that it shall
not by way of trade or otherwise be lent, re-sold, hired
out or otherwise circulated, without the publisher's
prior consent, in anyform of binding or cover other than
that in which pubHshed and without a similar
it is
condition including these words being imposed on a

subsequent purchaser.
© 1978 and 1984 Thames and Hudson Ltd, London

First pubHshed in the USA in 1978
by Thames and Hudson Inc., 500 Fifth Avenue,
New York, New York loi 10
First paperback edition 1980

Revised edition 1984
Library of Congress Catalog Card Number 84-50036
All Rights Reserved. No part of this publication may be

reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopy,
recording, or any information storage and retrieval
system, without permission in writing from the publisher.
Printed and bound in the German Democratic Republic

Contents
Preface to the First Edition 9
Preface to the Revised Edition 1
1 The Land and its Rivers 12

The Two Regions
The Coastline
The Rivers
Climate and Irrigation
Salinization
Northern Iraq
2 The Twilight of Neolithic Man 21

The Old Stone Age
Garrod and Solecki
American Excavators in Kurdistan
The State of Neolithic Research
Jordan and the Levant; Anatolia; Iran
The Iraq-Jarmo Project
3 The Threshold of Written History 37

The Sequence of Discoveries
Al 'Ubaid; Warka; Eridu; Khafaje
The 'Ubaid Period
Architecture; Religion; Pottery and Small
Objects; The Cemetery
The Uruk Period
Warka; The Anu Area; The Eanna Precinct;
Tell 'Uqair
Buildings of the Jemdet Nasr Period
Warka; Khafaje
The Protoliterate Period
The First Writing; Sculpture; Cylinder-seals;
Pottery
SuMERiAN Antecedents
4 Pre-literate Peoples of Northern 65
Mesopotamia
Excavations in the North
Arpachiyah; Tepe Gawra; Kiiyiinjik; Hassuna;
Other Hassuna Sites; Samarra and Sawwan
Architecture
The Pre-Halaf Period; The Samarra Period; The
Halaf Period; The 'Ubaid Period; The 'Gawra'
Period; Tell Brak
Pottery
Pre-Halaf; Tell Halaf; Northern 'Ubaid; Post-
'Ubaid
Burials
The 'Gawra' Period
Small Objects
Figurines; Seals
Foreign Relations
Susiana
5 The Early Sumerian Dynasties 88

The King-list
The Earliest Written Texts
Archaeological Phases
The 'Flood'
Early Dynastic Sites
Khafaje; Tell Asmar; Tell Agrab; Ur-of-the-
Chaldees; Al 'Ubaid; Kish; Lagash; Nippur;
Ashur and Mari; Shuaira
Variants in Terminology
6 Pre-Sargonid Art and Architecture 111
Sculpture
Statues; Relief Carving
Architecture
Building Methods; Temples; Palaces
Cylinder-seals
Metallurgy and Composite Craftsmanship
Inlaid Ornament; Composite Objects
Pottery
7 The Dynasty of Akkad and the Sumerian 135

Revival
Semites in Mesopotamia
Sargon and his Successors
Archaeology
Buildings; Tell Asmar; Khafaje and Brak;
Sculpture; Cylinder-seals; Gasur
GUTIANS AND LaGASH
De Sarzec; Telloh; Gudea; Al-Hiba
The Third Dynasty of Ur
Buildings at Ur; The Ziggurat; The Terrace;
The Mausoleum; Temples and Sculpture
8 The 2nd Millennium BC 157

Conflicting States
Buildings of the Isin-Larsa Period
Private Houses
The Palace at Mari
Mural Paintings
Tell Rimah
Sculpture
The Kassites
Mitanni (Nuzi)
The Middle Assyrian Period
City of Ashur; Fortifications; Temples and
Palaces; Some Finds at Ashur; Building
Practices
9 The Late Assyrian Period 187
Imperial History
Early Excavations in Assyrian Cities
Nimrud (Kalhu)
Khorsabad
Nineveh
American Excavations at Khorsabad
Building Construction
Late Assyrian Sculpture
British Return to Nimrud
The Nimrud Ivories
10 Babylon: The Last Mesopotamian 222

Monarchy
A Dynastic Revival
The German Excavation
The City
Buildings
Postscript
Notes on the Text 233
Bibliography 239
Photographic Acknowledgments 245
Index 246
To the memory of Henri Frankfort
Preface to the First Edition
Anyone book may well be

already familiar with the subject of ;his
disconcerted at by the extravagant claim impHcit in its title, and
first
justifiably feel that it calls for some immediate quahfication.

Throughout Mesopotamia, archaeological excavations have been
in progress almost continuously for more than lOO years, and the
literature which their results have engendered is by now sufficient
to fill a From field reports and typological
fair-sized library.
analyses, to epigraphical commentaries and stylistic art studies,
every facet is represented of a complex and far-reaching enquiry.
Indeed, one has observed without surprise that a recently published
bibliography could list over 5,000 relevant books and articles. Since
it would then clearly be impossible to summarize even the central
themes of all these writings, the purpose of the conspectus which

follows has been confined to one specific aspect of the research with
which it is concerned, and other minor Hmitations have been
imposed on its coverage. This selective process has in fact primarily
been made possible by a conspicuous duality in the categories of
evidence provided by excavations in a country where writing was
invented at an extremely early age.
The function of archaeology has been rather arbitrarily defined
by one writer as 'a way of learning about the past through things
instead of words' a ludicrous over-simplification in the case of
:
Egypt or Mesopotamia, where a significant proportion of the

'things' found by excavators have themselves been written
documents. On the other hand, the purpose which these documents
have served, far from being restricted to the recovery of narrative
history, has embraced the much wider task of re-creating in detail
the anatomy of ancient civilization. To this remarkable achieve-
ment the written texts have largely contributed, by perfecting the
recorded patterns of social or economic organization and of
intellectual development in literate ages. Yet, where excavations
are concerned, it is the vestigial remains of the physical setting in
which the documents themselves were written that has completed
the revelation of evolving humanity and its adaptation to
environmental influences. In a word, our astonishingly wide
knowledge of Mesopotamian civilization in 'historical' times is
derived in almost equal proportions from two different sources:
ancient literature on the one hand and, on the other, the study of
material remains. It should also be remembered that these
'historical' ages were preceded by a long era of illiteracy, a
formative period in human development of which our increasingly
explicit understanding has been derived exclusively from the results
of 'spade archaeology'.
Preface to the First Edition
As may already have been inferred from these observations, it is

wîth the material remains and with the progress of excavations
which have revealed them that this book is intended to be
concerned, rather than with the philological contribution. Other
limitations have been imposed upon it for a variety of reasons. I have
dealt at length with the pioneer activities of early Mesopotamian
explorers in another book {Foundations in the Dust, rev. edn, London
1980), and therefore I have not wished here to become involved in
anecdotal accounts of primitive digging in Victorian times. I have
preferred that my point of departure should
coincide with the first
introduction of discipline and method into archaeological
procedure. This is well known to have taken place at the turn of the
19th century, and something further should be said about it.
In Mesopotamia, the beginnings of systematic excavation and
proper recording must be credited to the two German scholars,
Robert Koldewey and Walter Andrae, whose work at Babylon
began in 1899. The method which they rapidly perfected of tracing
mud-brick walls, enabled them to expose and study the buildings
and fortifications in a manner which had never before been
attempted. In 1903 Andrae transferred his activities to the old
Assyrian capital at Ashur, whose ruins he proceeded to explore in
the same ingenious manner. Both excavators continued to deal in
this way with buildings immediately beneath the surface; but
Andrae went further. Finding that one particular temple showed
signs of having been repeatedly rebuilt at successive epochs in the
history of the city, he was able to examine the remains at each level
in turn, down to an earliest shrine which he attributed to the
Sumerians a people about whom almost nothing was then known.
:
But of even greater importance was the fact that, in doing so, he
mastered the art of 'stratified' excavation: a practice whose
understanding became the key to effective research in all
Mesopotamian settings. In the third decade of the present century,
when 'professional' archaeologists of other nationalities began to
arrive in Iraq, these German methods were adopted by them to
great advantage. Needless to say, they have been elaborated and
improved upon as time has gone by; but they can still be seen to
have provided the basis for a developing technique of excavation,
without which the discoveries recorded in this book could not have
been made.
Preface to the Revised Edition
These introductory paragraphs to the First Edition were written

at a time, late in the 1970s, when the excavation of major sites
in southern Iraq seemed temporarily to have lost impetus. Work

among the ruins of great cities in the alluvial plain had become
almost prohibitively expensive owing to the increased cost of
manual labour, and interest seemed likely for the present to be
concentrated on more manageable prehistoric settlements in the
10
Preface to the Revised Edition
north, where excavations on a smaller scale could produce profit-

able results more economically. Since the discoveries made in this
vây at Iraqi sites were, as expected, closely related to those in
neighbouring countries, I found myself hampered m writing about
them, by the original intention that my coverage should be restric-
ted to the frontiers of the modern state. It had of course already
become clear that exceptions must be made in the case, for instance
of Andre Parrot's monumental discoveries at Mari on the Middle
Euphrates, or those of Sir Max Mallowan in the Khabur cities.
But when news began to arrive of Paolo Matthiae's marvellous
finds at Ebla, near Aleppo, and of their Sumcro-Akkadian connec-
tions, it became evident that the cultural frontiers of Mesopotamia
were beginning to extend northwestward.
Also, in the prehistoric field, a new factor now contributed to
the northward of archaeological interest. More recent research
drift
at sites in northern Syria, supplementing simultaneous work in
upper Iraq and Iran, was producing new and important contribu-
tions to the clarification of Mesopotamian origins to the long- :
standing enigma regarding the antecedents of the Sumerians,

whose creative genius laid the foundations of Western civilization.
New directions were thus indicated for the migratory movements
of peoples, whose disparate elements, converging on the Meso-
potamian delta, merged their varied talents in a common identity.
A brief summary of the most recent developments in this far-
reaching enquiry will be found in a Postscript appended to my
final chapter.
Here also, it is perhaps desirable to repeat that the book itself
is primarily intended as an introduction to further reading. For

some care has been taken in the Bibliography, where
this reason,
more specialized subjects are concerned, to emphasize references
to works which are most easily accessible, at the same time giving
priority to those in English.
It remains only for me to express further gratitude to Dr Joan
Oates, who has once more been at pains to help me in up-dating
and correcting original text, and to my niece, Dr Dominique
my
CoUon, prompting on subjects more familiar to her arch-
for her
aeological age-group than my own.
II
Chapter One
The Land and its Rivers
The name 'Mesopotamia' is an archaism, traditionally more often

applied to the setting of ancient history than to a clearly defined
geographical area. The Greek translators of the Old Testament
thought of it as the homeland of the Patriarch Abraham around
the ancient city of Harran, which lies between the middle courses of
the Euphrates and Tigris. Strabo also used it to denote only the
northern part of the interfluvial lowland, and referred to the
southern part as 'Babylonia'. It was Pliny who extended its limits to
the Arabian Gulf, making it approximately the equivalent of
modern But the use of this Arabic name, meaning 'the Cliff, is
Iraq.
also slightly equivocal. It was applied by the Arab conquerors only
to Babylonia, and requires some explanation. If one approaches the
\\, TURKEY LAKE

VAN'
Catal Huyuk ^?\lake
^Hacil^ ^ S \urmia
T. ,, )erde
'^Suberde Sakcagozu
^^^^s MTS
TS .Tarsus "
Mersin^ 'Hajji Firuz & Dalma
muk hanidar
C'/ ALEPPO
V.Mardikh
o
SYRIA
Canj Da
Tepe Asiab'
ChogaMami ^TepeGura
i^Mallaha y^.^
ISRAEL/,.-
, , 'MrCarmel
"-..^ .
^ _ ~
s.
^
TepeSabz\
v.AaIjI^sI
'Jericho <^ [^^ X/ SusaA^
-1 ^\
ABeidha '.
SAUDI ARABIA
200 m/
river valley, as they did in the 7th century ad, from the Hamad or
high Arabian Desert to the west, one fmds oneself suddenly on an
escarpment about 30 m
above the plain, looking out over the tree-
tops of the Euphrates cultivation to that of the Tigris beyond and,
in the remoter distance, to the thin line of the Zagros Mountains,
which form the eastern limits of Mesopotamia. At Kufah, on the
edge of the 'cliff, there is a shrine called As-Safinah, 'the Ship',
marking the place where Moslems think that the Ark rested.
The Two Regions
Already, these few sentences will have drawn attention to the

disparity between northern and southern Mesopotamia, and this
will become more apparent if we now consider the geophysical

characterof the country, its geological formation, climatic
and all the other environmental elements which
peculiarities
combined to make a distinctive pattern in the lives of its earliest
inhabitants. As a geographical unit - not yet satisfactorily defmed
'
- it consists of a broad and shallow depression, running

northwestward from the head of the Gulf, of which geologically it
is a prolongation. The allocation of its limits on either side present
Httle difficulty. To the northeast they correspond to the diminish-

ing foothills of the Iranian Mountains, to the southwest, to the
I Map showing some Near

Eastern sites mentioned in the
U.S.S.R text
CASPIAN
SEA
I RAN
Tepe Yahya.
13
2 Map showing Mesopotamian fringes of the tremendous desert which geologists call the 'Arabian
sites mentioned in the text
Plateau'. What is less apparent on an ordinary map is a horizontal
dividing-line, running approximately through Hit on the Euph-
rates and Samarra on the Tigris, which marks the uppermost
limit of the alluvial plain. To the north of this line, the rivers are
separated by a barren limestone plateau called Al-Jazirah, 'the
Island',which restricts the Euphrates to a narrow valley. The
Tigris, however, profiting from its eastern tributaries, passes
through a wide upland of undulating ploughland and pasture: the
productive countryside which was once Assyria. To the south of
the Hit-Samarra line, where the two rivers reach their common
14
delta, there an entirely different landscape, a country which is the

is
creation of rivers. Here, from time immemorial, the Tigris

two
and Euphrates have been depositing alluvium over a bed of
sedimentary rock; and this has brought into being an immense and
entirely flat plain, of which, from the point of view of potential
fertility, there is no equivalent in the whole of the Near East.
The Coastline
Having reached this point in a description of Mesopotamia,' we are

confronted by one of the most outstanding problems regarding the
geological formation of the southern delta. For it has been
tempting of the present
in the past to consider that a great part
alluvial plain has, in the course of time, actually been reclaimed
from the sea. Throughout the first half of the present century, the
concensus of opinion among historians and geologists alike
favoured this interpretation. It was clear of course that any
recession of the coastline must have started at a very early period,
since the upper two-thirds of the plain is dotted with ancient
mounds representing the great cities of Sumer and Akkad, whose
history goes back to the earliest human settlements in the 6th
millennium bc. But here again there is a line across the country at
about the latitude of Nasiriyah, southward of which no mounds are
to be seen ^ and between this line and the present coast there is a
;
wide and sparsely populated region of marshes and lakes. It was

supposed therefore that the Gulf, which had once perhaps extended
over the whole of the delta, by early historic times had receded as
far as the Nasiriyah line; the rivers Karkheh and Karun, flowing
westward from Khuzistan, had built forward their own delta,
closing off the head of the Gulf and converting it into a huge lake,
which the Tigris and Euphrates had since partially contrived to fill
in with their own burden of alluvium.
In 1952, however, this easily accepted hypothesis was demolished
by two geologists 3 who, after a lengthy investigation with
improved equipment, were able to show that the position of the
coastline had in fact not varied greatly since the earliest historical
times. Their initial observations confirmed the fact that very little
of the alluvium carried by the two rivers ever reached the sea, and
that the consequent rise in level of the southern plain could be
calculated at almost 60 cm in every 100 years. No marshes or lakes
in the south could therefore have continued to exist unless this
infilling were countered by a corresponding subsidence of the basic
rock beneath the delta. The occurrence and approximate rate of this
subsidence they were then able to substantiate.
Archaeologists at first found all this a little difficult to reconcile
with certain, fully established historical facts. The Sumerian city of
Eridu for instance (Tell Abu Shahrain), which, according to
Mesopotamian tradition, ranks as 'the oldest city in the world', is
explicitly described by ancient writers as 'standing upon the shores

of thesea' and Ur, situated only a few miles away, had quays at
;
which ocean-going vessels discharged their cargoes. Both of these "*
today are almost 100 miles from the seashore; so some acceptable
compromise had to be arrived at. One small detail of archaeological
15
evidence has helped to make this possible. In the temple at Eridu,

were made to the god Enki, and among them were
fish-offerings
found bones of a sea-perch species which can only live in the
brackish water of tidal estuaries. Perhaps the shallow depression in
which Eridu was then part of the present lake system which, in
lies
its turn, connected by deep channels with the Euphrates estuary.

Similarly, Ur would have been served by an ancient course of the
Euphrates itself.
The Rivers
Perhaps one should now return to the character of the two rivers,
on which the whole economy of the country depends.
The Tigris rises in a small lake (Hazar Golii), .about lOO miles
west of Lake Van, and flows down easterly and southeasterly
towards Nineveh and the Assyrian uplands. The Euphrates, which
is a much longer river (1,780 as opposed, to 1,150 miles), has two
sources between Van and Erzerum. Both branches flow at first

westward, to unite near Elazig, where the Keban dam now creates a
lake. The main stream then follows a winding course southward,
crossing from Turkey into Syria near the ancient city of
Carchemish and so eventually into Iraq. At this point it is separated
from the Tigris by some 250 miles of steppe country and the two do
not draw together again until reaching the neighbourhood of
Baghdad and Ramadi. Here the Euphrates is flowing at a level 9 m
higher than the Tigris, and a sequence of ancient irrigation canals,
draining from one river into.the other, in earlier days made the belt
of country between them extremely fertile. Beyond this, their
streams separate again, and the plain is served by a more
complicated system of canals and diversions.
In their passage through Iraq, before reaching the Hit-Samarra
Line, the two rivers are flowing through beds which they have
themselves cut into hard limestone and shale, so that their courses
have hardly changed at all since prehistoric times. For this reason,
cities like Carchemish, Nineveh, Nimrud and Ashur are still to be
seen, standing as they always have beside the river banks. South of
the same line the situation is quite different. Here rivers meander
through the alluvial plain, frequently changing course and
throwing off side-branches. Also, like all sediment-bearing rivers
which flow at a very low gradient, they gradually raise the level of
their own beds, so that, for the greater part of the time, they flow at
a higher level than the surrounding plain. If, as inevitably happens
on occasion, they overflow their banks in floodtime, great
permanent lakes and swamps tend to be created, and occasionally
the river changes its course. This explains why some of the great
cities of the alluvial plains, which once stood beside the Tigris or
Euphrates, are now huge de cated ruinfields, far out in the

unwatered desert.
This picture, of the two rivers running between semi-artificial
banks rather higher than the surrounding plain, will serve to
introduce the subject of irrigation in Mesopotamia, and its primary
contribution to the country's economy. For this purpose, water has
to be drawn off in such a way that the banks are not damaged or
16
The system by which this is done is naturally

floods released.
dependent upon the seasonal regime of flooding, itself dictated by
climatic conditions.
Climate and Irrigation
Where the climate of Mesopotamia is concerned, it is well to

remember that, according to the findings of geologists, there has
been no perceptible change since very early times. The country has
summer temperatures ranging from no to 130 degrees in the
shade, and eight months in the year without rainfall. By the end of
the dry season, the rivers are reduced to sluggish brown meanders
in a waste of dried mud. Then comes the winter, with pale sunshine
at midday and cold nights, bringing intermittent rainstorms. But
the rivers do not receive their full volume of water until the spring,
when the melting snows in the Taurus and Zagros Mountains feed
their tributaries. Then come the spring floods which, a generation
ago, were considered practically uncontrollable and through
all
history represented an ominous of the

threat to the inhabitants
lower plain. Paradoxically also, this happens between April and
June, which from an agricultural point of view is too late to water
the main crop, usually harvested in April.
This was the climatic regime and seasonal fluctuation with which
the ancient farmers of southern Mesopotamia were faced : rain in
inadequate quantities at the wrong time; river water, also at the
wrong time, and concentrated around the river-beds in almost
unmanageable quantities. So in the end the country had to be
supplied with water, simply by the contrivances of human
ingenuity: a complex system of canals, reservoirs, dykes and
regulator-sluices. This needed a great deal of organization and also
a great deal of patience. The canals themselves very rapidly filled up
their beds with silt, and consequently required repeated dredging.
Then, as this process went on, the banks became too high to throw
out the spoil, and consequently a new canal had to be dug, parallel
to the first. Today, looking at the country from the air, one can see
these extraordinary networks of canals, some of them with as many
as three parallel channels. The overall impression is that of a single
complicated pattern, and this extraordinary sight could lead to the

illusion that all these channels have been in use at the same time. In
fact, for aprevious generation it created an historical picture of
Iraq asone vast granary, whose fabulous prosperity survived until it
was destroyed by the Mongols in the 13th century ad. This was of
course a complete misconception, and a reverse situation has been
revealed by agricultural research in recent years. One specialist
investigation after another has added to the record of declining
productivity, to be explained not by some isolated disaster in
comparatively recent times, but by a longstanding and radical
deficiency in the basic system of cultivation, going back to a
surprisingly early period. ^
Salinization
Next to the effects of flooding, the most disastrous danger to

Mesopotamian agriculture generally has been the salinization of the
17
soil and the consequent practice of what is known as 'extensive

cultivation'. Travelling in Iraq today,one cannot fail to see the
results of the mishandling of the soil. Wide areas, no longer
cultivated, are covered with a white incrustation caused by over-
prolonged cultivation. The water comes from the river with a
strong saline content and, as it evaporates in the hot sun, the salt is
deposited, ruining the fertiUty of the soil. When this happens to

land in an 'extensive economy', the farmer simply transfers his
cultivation to new ground and starts irrigating all over again. But
an even more serious sourceof salinization is the rise in level of the
ground-water as a result of prolonged irrigation, which pushes the
salt up to the surface. Only efficient drainage can counteract this
effect; but there is no inducement to go to such lengths as long as
'extensive' agriculture can be practised.
In the late 1950s much new light was thrown on this subject
durmg the course of a study by Thorkildjacobsen, one of the great
Assyriologists of our time.^ In a variety of cuneiform texts, he
found unmistakable references to the results of salinization, and
could study its effects over a long period of history. Jacobsen was
able for instance to gather that soil deterioration was particularly
serious in the city state of Lagash, where salinization began in about
2400 BC and spread westward towards the Euphrates. A thousand
years later, it had reached as far as Babylonia. He was able to
calculate that wheat at first accounted for 16% of the total crop.
Three centuries later, the percentage had dropped to 2% and,
between 2000 and 1700 bc his reports contained no mention of
wheat at all. Even the barley, whose greater toleration of
salinization had long made it the principal crop, could now be seen
to have a greatly reduced yield per acre in many southern districts.
Information of this sort gave substance to a picture of diminishing
agricultural prosperity, moving continually northward as a result
of soil impoverishment in the south. This, he thought, could even
account for the sequence of major changes in Mesopotamian
history, by which political ascendancy transferred itself first from
Sumer to Babylonia and later to the Assyrian kingdom in the
north, where the problem of salinization did not arise. Neverthe-
less, there is much evidence to suggest that the diminution of
agricultural productivity during these centuries was only a

temporary affair. For Jacobsen and his colleagues, examining the
Diyala area east of Baghdad, were able to detect already in the late
3rd millennium bc improved techniques for combating saliniz-
ation, or at least delaying it. Long experience had resulted in the
contrivance of better practices for extracting a maximum yield
from the soil. A Sumerian agricultural manual of about 2100 bc
describes the use of fallow agriculture and elementary forms of
drainage.
It seems therefore that, even if we discard Herodotus'
exaggerated account of the harvest in Babylonia at his time, we
should be justified in concluding that some sort of recovery took
place in the agricultural productivity of southern Mesopotamia
during the 2nd and early ist millennia b c. Undoubtedly there must
have been periodic disasters, due to flooding and unpredictable
changes in the course of the river-beds. For that matter, the
18
The Land and its Ri vers
northern provinces too must occasionally have suffered, as they do

today, from abnormal shortages of rainwater. But, taken as a
whole, Mesopotamia may be said to have been a country rich in
agricultural products and well able, during the greater part of its
early history, to feed its own population. In addition, it has as a rule

been able to barter its surplus cereals for stone, metals and other
materials, which had to be obtained by trade with neighbouring
countries. Nor was its agricultural productivity limited to grain
alone. We learn from ancient texts that as early as the 3rd
millennium b c southern Iraq had extensive palm-groves' and that
dates were already being cultivated by artificial pollination. In fact,
flour and dates formed the staple diet of ancient Mesopotamians,
though cattle and sheep were also bred and grazed in uncultivated
areas, while rivers, lakes, canals and the sea produced fish in
abundance. Fruit and vegetables were grown in gardens, sheltered
from the sun by palm-trees and irrigated by very simple water-lift
devices, of a sort which are still used in Iraq today.
Northern Iraq
We should now complete the geophysical picture of Mesopotamia

with a short description of the northern uplands; beyond the Hit-
Samarra line. To the west of the barren Al-Jazirah plateau, as we
have said, the Euphrates irrigates a narrow strip of agricultural
land, punctuated by small market-towns such as Hit, Rawa and
Anah. North of the present Syrian border the cultivation widens out
into a considerable agricultural province, dependent in early times
on the purely Mesopotamian city called Mari. Beyond the Jazirah
and its tributaries. Greater Zab, Lesser Zab
to the east, the Tigris
and Adhaim, form the agricultural arteries of the Assyrian
countryside. This is undulating gravel steppe, with rich ploughland
in the valleys and a plentiful supply of building stone. The rainfall
in an average year is here sufficient to produce a single crop of
wheat without irrigation. Only gardens and plantations are
artificiallywatered from the rivers or wells. For much of the year
the country is bare, but spring covers it with grass and flowers. ^
Finally, enfolding Assyria to the east and north is the
mountainous country known today as Iraqi Kurdistan a crescent :
of highlands, with one corner resting on modern Khanikin and the

other on the crossing of the Tigris at modern Faish Khabur, where
the Syrian, Iraqi and Turkish frontiers now meet. The country here
differs only in minor respects from the eastern vilayets of Turkey.
There are stone-built villages, terraced into the hillsides, with tall
Lombardy poplars and terraced cultivation including vines and
tobacco. Large areas of mountainside are covered with scrub-oak
or more rarely conifers, and there is much game below the snow-
line in winter. We shall later be visiting this area (sometimes called
the 'hilly-flanks' zone of Breasted's 'Fertile Crescent'), in search
of the earliest human settlement.
After this general description of the background against which

archaeological research in Mesopotamia has taken place, a word
may be added regarding the historical relationship between its
19
inhabitants and those of the countries by which it was enclosed on

either side of the Syrian desert, that is, and of the tribal areas of the
:
Zagros Mountains. The Sumerians and Babylonians were essen-

tially town-dwellers and peasant folk. Unlike the nomads of the
desert and the migratory herdsmen of the Iranian uplands, they

preferred to turn their backs on the open spaces and to concentrate
upon a settled life among the amenities of the river valleys. In doing
so, however, they were seldom free from the attentions of their less
fortunate neighbours, whose practice was to interfere with their

trade-routes and raid their outlying villages. As we shall see, long
chapters in the history of their country are primarily concerned
with the struggle to preserve their sedentary way of Hfe and to
protect themselves against these avaricious nomads. At times, the
active hostility of such raiders became less significant than their
peaceful penetration of river-valley society and their eventual
establishment of a formidable majority. It was on such occasions
that the hegemony of indigenous rulers could be replaced by a
hitherto unfamiliar Dynasty-of-Akkad or by a largely Semitic line
of Babylonian kings. These in their turn might in due course be
replaced by from the east or the intrusion of a
a tribal invasion
Kassite aristocracy. Such events are the substance of Mesopotamian
history.
20
Chapter Two
The Twilight of Neolithic Man
The subject with which this chapter is intended primarily to be

concerned is theoutcome of research into the origins and
sociological development of peoples whose descendants comprised
the inhabitants of Mesopotamia in historical times. The thread
which connects them with humanity of Palaeolithic (Old Stone
Age) times is still tenuous and controversial. It cannot however
be ignored; and, since we have already set arbitrary limits to the
geographical area with which we are dealing, it would be wrong to
neglect the evidence that has been found within those frontiers of
human occupation preceding the end of the Old Stone Age.
Naturally this subject merges with, or rather emerges from, the
geological history of southwestern Asia. Yet it is only with its
terminal or most recent phases that we need be concerned. It is the
order of these phases or the pattern which they are thought to have
created which we should perhaps first endeavour to recollect.
The Old Stone Age
The Palaeolithic period then, which saw the evolution of man Table I
from ape-like ancestry to the true semblance of Homo sapiens,
corresponds roughly with the Pleistocene or 'most recent' phase in
geological history. It started over 2,000,000 years ago and ended
between 20,000 and 12,000 years ago. It was during this period that,
at least four times in succession, huge glaciers which were an
extension of the polar ice-cap crept southward to cover large parts
of Eurasia and North America. ^ The earliest phase of the Old Stone
Age, the Lower Palaeolithic, extends from well before 2,000,000
years ago to about 80,000 years ago. In Europe during the latter
part of this long era, the first evidence is found of primitive cultures
(Abbevillian, Clactonian and Acheulian), material remains of
creatures already related to the forebears of modern man. With the
Middle Palaeolithic (80,000-30,000 years ago) we shall be more
closely concerned in our present context, since in northern Iraq it is
represented by cave-occupations of the Mousterian period, and also

by human remains which include the bones of Neanderthal man,
that strange, extinct offshoot from the tree of human heredity.
The Upper Palaeolithic (30,000-12,000 years ago) will also be of
great interest to us here. In Iraqi Kurdistan, as in other parts of the
Near East, its remains have been closely studied whenever an
opportunity occurred, because they constitute a preface to the great
change which took place at the end of the Pleistocene epoch. The
phases corresponding to the Aurignacian and Gravettian in Europe,
21
I
BEFORE CTT L
WEST IRAQ LEVEL OF
PRESENT b 1 Aor, EUROPE SHANIDAR OTHER CULTURE
CAVE SITES
Layer B Zawi- FOOD
10,600 BP Chemi PRODUCING
12,000 Layer B2
12,000 BP KarimShahir
M'lefaat
Hazar Merd
Magdalenian Palegawra
U
15,000
X
H
Hiatus
2
< 1
N-l
<
0-
ce: Solutrean
0.
20,000 a. Zarzi
Gravettian Z
X
H
<
Perigordian
Top Layer C Q
Aurignacian
30,000 28,000 BP 2
' ^ ' '
1
Q
<
Bot torn
Lay erC
35,0(DO BP 7
n
1I
40,000
Hicitus
>
TopL ayerD
Mous terian 46,0 DO BP
Up per
Lay 2rD
50,000 50,0(DOBP
Table I Generalized scheme for

Other Caves
the Palaeohthic in Europe and
> and
the Near East. (After Solecki,
80,000 Base fCave Open Sites
1971)
represented by discoveries on Mount Carmel in Palestine and in

Iran at sites ranging from the Zagros Mountains to Lake Urmia, are
of interest in themselves, as shovîng the geographical distribution
of these cultures; but the terminal or transitional stage which
follows has even greater significance, in that it bridges the gap
between the study of man's imperfectly developed ancestors and
the fuller revelation of Homo sapiens as we know him today,
profiting as we may from ampler evidence of his early accomplish-
ments and aspirations. The chronology of this final phase of the
Palaeolithic is still a little uncertain. Its terminal limit, however,
must be associated with the appearance of the first farming
communities at the beginning of the Neolithic period, soon after
10,000 BC.
22
It is imagine the geological and climatic changes

a little difficult to
which took place Western Asia during the Pleistocene, when
in
early man made his appearance for the first time. Some traces of
periodic glaciation have been recognized in the high mountains of
Anatolia and, to a lesser extent, in western Iran yet it has become;
apparent that the encroaching ice-sheet itself did not reach as far as
the East. On the contrary, throughout most of the Ice Age,
Near
moist from the Mediterranean blew around the lower contours
air
of the mountains in an easterly and southeasterly direction, creating

grassy steppes and uplands, which benefited from a comparatively
temperate and uniform climate. For the greater part of the time,
therefore, the southward-facing lower slopes of the Taurus and
Zagros (what has been called the 'upper piedmont zone of foothills
and intermontane valleys'), offered good hunting and favourable
climatic conditions for Palaeolithic food-hunters. It may accord-
ingly have been no great surprise for Old World prehistorians to
discover traces of their flint industries in the caves of Iraqi
Kurdistan, or for that matter in Palestine, southwest Anatolia or
northwest Iran. Nor need it have seemed strange that the cultures of
such hunters proved to have affinities with those already recognized
in Europe. For all these areas corresponded to a geographical line of
migration, which may have been more obvious in Pleistocene
times than it is today. Speaking of changes in the inland seas of
western Asia one authority suggests for instance that,
at this time,
during the last accumulation of pluvial waters
glacial age, the
raised the Caspian Sea something like 76 m
above its present level.
The Black Sea by contrast was reduced to a brackish lake, cut off
from the Mediterranean by a dry Bosphorus, and so leaving an easy
connection for early man between Europe and Asia. Further east,
however, the Caspian and vast swamps around the Aral Sea
inhibited his movements. ^
Garrod and Solecki
The first were made as

Palaeolithic discoveries in northern Iraq
early as 1928 by Dorothy A. whose name afterwards
E. Garrod,
became well known for her more prolonged excavations on Mt
Carmel. Her first sounding was made in a cave called Zarzi, near
the headwaters of the Lesser Zab river, about 20 miles north of
Sulaimaniya. She found undisturbed deposits corresponding to the
Gravettian culture in Europe, and Zarzi became the type-site for
material of that period. ^^
Garrod's second experiment was in a cave nearer to Sulaimaniya
called Hazar Merd, where a brief sounding revealed an earlier flint
industry which could be identified as Mousterian, partly by the
presence of characteristic unifacial points and scrapers. Garrod was
also able to list of contemporary animal remains,
publish a
identified for her by her colleague Dorothea Blake. ° After her '
departure for Palestine, Palaeolithic research in Iraq remained in

abeyance until 1949, when a succession of American scholars took
up the thread of her investigations. Most successful of all was the
work of Ralph S. Solecki in the cave called Shanidar, which he
selected from a numerous group on the southern flank of the
23
TURKEY
USSR
Mt Ararat )*y
^^^
Lake Van ....-.''
Tabriz
y JamtamaQ
Tar
J-^ ,^ ., P ^Rizaiyeh'
Lake Urmia
ShanidarJ Ushnavieh
Mahabad
«Hamadan
Bisitun and Ghar-i-Khar

^ _
r~ . PaSangar ^
Yafteh Cave^ Khorra'mabad ^
GarArjeneh^ Kunji Cave
Map showing major • Dezful

3
Palaeolithic sites in northern Iraq
Baradost mountain range, above Rowanduz. His sounding there

reached a depth of 13.7 m, and the earhest sequence of occupations
(level D) he was able to attribute to a Mousterian phase, wîth
predominantly flake culture, as opposed
artifacts 'characterising a
to the blade culture of the Upper Palaeolithic (level C) horizon'.
But it was here also that, between the years 1953 and 1957, he had
the great good fortune to uncover the remains of four human
skeletons, unmistakably exhibiting the characteristics of Neander-
thal man. '
^
In western Iran, material comparable to that found by Solecki at
Shanidar had already been reported by Carleton Coon, an

anthropologist from Pennsylvania, in 1949. Following the
supposed line of migration we have Coon had
already mentioned.
come upon a cave site at Kermanshah, with traces of a
Bisitun, near
Mousterian industry similar to that found at Hazar Merd. He also
claimed to have discovered human skeletal material which he
described as 'characteristically Neanderthaloid'. Similar parallels
were later found by Turkish archaeologists at Beldibi and
elsewhere, near the Mediterranean coast of Anatolia; so the pattern
of diffusion began to be better defined.
It appears that the two final phases of the Late Palaeolithic in
Europe, the Solutrean and the Magdalenian, which accounted

among other things for most of the famous rock-paintings of
24
western France, did not reach the Near East. Garrod's assemblage
of late Aurignacian or 'extended Gravettian' from Zarzi accord-
ingly came to be regarded as representing the terminal phase of the
Old Stone Age in Iraq. Since it already comprised some features of
the transitional period which followed, it is perhaps worth
examining a little more closely.
American Excavators in Kurdistan
The rather limited range of Garrod's finds at Zarzi were

conveniently supplemented by those from a cave at Palegawra,
some miles to the southeast, excavated by H. E. Wright and Bruce
Howe in 1950, and the combined results consequently give a fuller
picture. One important innovation is the use of microlithic flints, a
newly refined blade industry which included the first chipped stone
arrowheads, in addition to a wide variety of microlithic bladelets,
scrapers and burins. Among other stone implements was a polished
celt, fragments of querns for grinding and some obsidian (volcanic
glass), for which the nearest source would have been the area of
Lake Van in eastern Anatolia. There were also beads and pendants
of shell. Animal bones showed that the onager or wild ass was the
most commonly hunted species but wild goats, sheep, cattle and
;
gazelle also indicated the nature of the terrain in which 'Baradostian

Man' lived. Wood remains included oak, tamarisk, poplar and
conifer, all of which are to be found in Kurdistan today. They also
provided a convenient radiocarbon date, between 13,060 and
14,210 years ago. The cultural situation envisaged from all this
evidence could therefore be treated point of departure for the
as a
intensified investigation which now took place, of the transition
from a Palaeolithic way of life to the vastly changed economy of
the Early Neolithic.
It will now be important to recollect the state of this enquiry
during the fifth decade of the present century, when R.J.
Braidwood of Chicago initiated the Iraq-Jarmo project, supported
by an inter-disciplinary team of specialists, whose composition was
in itself an innovation. ^ ^ In Iraq at that time, archaeologists of
many years been devoting themselves
various nationalities had for
to study of pre-Sumerian cultures and the origins of
the
Mesopotamian civilization. Following Leonard Woolley's dis-
covery of primitive marsh-dwellers at the site called Al 'Ubaid and
his deep sounding in the prehistoric levels at Ur-of-the-Chaldees,
his younger colleague, M. Mallowan, had made an even
E. L.
deeper penetration beneath the Kiiyiinjik mound at Nineveh, and
had recognized a sequence of occupations, throughout which the
smelting of copper appeared to have been understood and
increasingly practised. It was not however until 1943 that an Iraq
Government excavation at Hassuna, on the western periphery of
the Assyrian uplands, revealed in its deepest level a nomadic camp-
site, which could be regarded stratigraphically as Late Neolithic.
The period of some five millennia which separated these remains

from the terminal Palaeolithic horizon at Zarzi and Palegawra
represented a hiatus in the prehistoric sequence, whose completion
became the purpose of Braidwood's expedition during the 1950s.
25
Braidwood's enterprise in Kurdistan made a number of

worthwhile contributions to the solution of the problem in hand.
But today it must be considered in terms of the very plentiful
information since accumulated from similar investigations in all
parts of the Near East. For the moment, therefore, our review
cannot be confined to Iraq alone. We must rather examine the
pattern created by discoveries at a score or more of ancient sites in
the Levant, western Anatolia and Iran. First, however, a word
should be said about the term 'NeoHthic Revolution', first used by
the doyen of British prehistorians, Gordon Childe.
Childe was primarily concerned with the transition from
hunting and food-gathering to a food-producing economy. In his
time {c. 1927 onwards) 'Neohthic' had already long been accepted
as a name for the phase in human development during which this
change took place. He himself qualified the additional word
'Revolution' by explaining that it should not of course imply any
violent upheaval, but 'the culmination of sudden progressive
change in the economic structure and social organisation of
communities. .'.'^ Childe accepted the two existing criteria for
.
the Neolithic development - namely, the practice of agriculture

and the domestication of animals. He also saw the manufacture of
pottery as a hallmark of the period; although the more recent
discovery of pre-ceramic food-producing cultures has shown this
to be too broad a generalization, in northern Syria at least pottery-
making and the collection of wild cereals do seem to have
coincided. Yet the most recent assessments of the great 'revolution'
have in fact come to suggest an evolutionary rather than a
revolutionary process of change, one that was neither rapid in time
nor geographically uniform. It has been pointed out, for instance,
that 'the earliest species of domesticated animals and plants do not
appear in one particular area or at one point of time, but rather at
different sites at different times'. Conclusions like these have tended
to refute in turn theories propounded by at least one of Childe's
successors in the Neolithic field. Braidwood, over a long period,
had attempted to define geographically a 'Natural Habitat Zone',
in which plant and animal domestication took place, forming a
crescent around the 'hilly flanks' of the Zagros and Taurus
Mountains with its western point on the Mediterranean. Since
many newly discovered sites, showing Neolithic communities
practising agriculture and herdsmanship, plainly lie outside this
'zone', this concept has also had to be revised.
The State of Neolithic Research
We should now briefly summarize those excavations in countries of

the Near East other than Iraq which have, in the past years, made
major contributions to our knowledge of Neolithic cultures and
their immediate antecedents. In doing so, we should perhaps start
with the Levant where, as long ago as 1928, settlements were found
showing the earliest evidence of food production. And here once
more we return to the work of Dorothy Garrod, this time in the
Wadi-al-Natuf on Mt Carmel. Her 'Natufian' culture, there
26
ANATOLIA
BC LEVANT IRAN
OTHER SITES MERSIN
Susa®
2500
a-c
Bronze Age
Susa ©
c
3000
b
t
Uruk-'Ubaid a
3500
Susa®
'Ubaid xi-xv
4000 Susa ^
(Old Susa I)
t
PNB
Fortress xvi Susiana e
4500
PNA
Susiana
^ Halaf xvii-xix a-d
1
5000
-2 X
G T
Gi\P
'Neolithic'
><
X X Guran
1
6000
1
m pp NB
' ^ >
'
>
7000 .
Qayonii ,
> '
'
>
PP VA >
h
Mureybet Asiab Table II Generalized chronology

for sites in the Levant, Anatolia
8000 '
(Syria)
1
and Iran
27
recognized in caves and shelters, has since served as a point-of-

departure for the identification of phases and sub-phases in
NeoHthic development. But it was also in itself a remarkable
revelation of a new way of life, departing in important respects
from that of the terminal Palaeolithic which preceded it. The
microhthic flint industry was of a sort at that time provisionally
described as 'Mesohthic'. Hunting and fishing were the main
sources of food but flint sickle-blades, showing evidence of use,
;
pointed to the reaping of wild wheat or barley. Craftsmanship

extended to the carving of animal figures and this was applied to the
handles of bone hafts in which the blades were set. Personal
ornaments were also found, including headdresses of dentalia and
necklaces. Other sites in the Levant, especially Mureybet, Abu
Hureyra, Eynan (Ain Mallaha), and Jericho described below, have
more recently added new features of this comparatively sophisti-
cated culture, including primitive architecture and evidence of
organized religion. In the words of one authority, the Natufians
'founded the earliest permanent settlements in the world'.
Jordan and the Levant

At Jericho in Jordan, where the deepest levels of the great 'oasis'
city-mound were excavated with such sensational results by the
late Kathleen M. Kenyon between 1952 and 1958, there were two
earliest occupations, termed respectively 'Mesolithic' and 'Proto-

Neolithic', both corresponding to the Natufian period elsewhere.
Kenyon describes the earhest occupants (in about 9000 bc) as
hunters and food-gatherers, but a conspicuous feature of their
settlement was a stone-built shrine or sanctuary. The Proto-
Neolithic was succeeded by two more productive phases, referred
to as Pre-Pottery NeoHthic (PPN) 'A' and 'B'. The first of these
shows a rapid advance of a communal society.
in the organization
There are circular or rectangular houses, now
built of sun-dried
bricks whose shape is described as 'hog-backed'. The town, which
now covered an area of some 10 acres, was defended by a massive
stone wall with a circular tower measuring over 12 min diameter,
and a rock-cut ditch. Understandably, Kenyon conjectured that a
defensive system of this sort must imply a surprising degree of
social solidarity and leadership an idea supported by the evidence
:
of organized agriculture and a far-reaching trade in commodities

such as obsidian. Radiocarbon dates for this period fall between
8350 and 6770 BC.
The PPNB stage at Jericho, like its predecessor, must have lasted
a long period of time, since no less than twenty-six structural phases
were distinguished. During this phase, domestic architecture was
greatly improved, each house having large intercommunicating
rooms. Walls were still of curiously shaped mud-bricks, the floors
being carefully paved with gypsum plaster and in some cases
covered with reed matting. Again, one symmetrically planned
building was thought by Kenyon to be a temple. But the most
curious artistic feature of this period were the human skulls, with
remodelled in plaster and inlaid eyes. No less than ten
faces skilfully
of these were found, suggesting some cultic practice for the
posthumous commemoration of individuals. For the rest, there
28
4 Three examples of human

skulls, with features carefully
modelled in plaster and inlaid
eyes of shell, found in the PPNB
levels at Jericho (yth millennium
Bc), perhaps implying a form of
ancestor worship
were the bones of wild animals, among which the gazelle was
preponderant, and of goats which were plainly domesticated. The
carbonized remains of food-grains were also found, the emmer
wheat of the previous period having now largely been replaced by
einkorn. The Pottery Neolithic phase, which followed the
temporary abandonment of the Jericho settlement, was distin-
guished as its name implies by the appearance of plain or clumsily
painted ceramics, of a sort which are better represented at other
sites.
At Jericho, Kenyon's excavation had been hampered to some

extent by the limited area of her deep sounding. Soon afterwards,
Diana Kirkbride (Mrs Hans Helbaek) had the good fortune to
discover a site on the banks of a dry wadi to the north of Petra,
where the buildings of a Neolithic settlement were accessible
directly beneath the surface. At Beidha (Seyl Ahlat) she excavated a
wide area during the years between 1958 and 1967, revealing an
attractively detailed picture of primitive village life. After a short
occupation in the early Natufian period, the site was temporarily
deserted and then reoccupied sometime after 7000 b c by a people
corresponding to the later Pre-Pottery inhabitants ofJericho. Their
houses, developing from round to rectangular shapes, were
constructed with heavily built, dry-stone walling and the rooms
repeatedly plastered. Clusters of small rooms were later replaced by
larger, single compartments with plastered walls, and pavements a
little below ground-level were decorated with lines of coloured
paint. A baker's and a butcher's shop were recognized and there

was other evidence suggesting a precocious development of
specialized trades. Stone implements were variously contrived for
every conceivable purpose, while toys or cult-objects were
modelled in clay.
The evidence of sites like these in Jordan and the coastal Levant
has been supplemented by new and impressive exposures in inland
Syria, in process of publication. The implications of at least one
have been disconcerting. At Mureybet, 50 miles southeast of
Aleppo, M. van Loon and J. Cauvin have excavated a settlement
29
5 Excavations at Beidha, near ot the 9th millennium bc, where a people living in substantially
Petra, exposing the stone-built
built clay houses may already have cultivated einkorn, herded
houses of a settlement founded ir
Natufian times, but reoccupied goats, and made pottery (see D. and J. Oates (1976), p. 74 etc.)
from about 7000 bc onwards by
a characteristically Neolithic Anatolia
people
If we must now choose from a dozen relevant excavations in
modern Turkey, one in particular immediately suggests itself as
being likely to provide the Anatolian equivalent of the Pre-Pottery
Neolithic at Jericho. This is Qayonii, in the Diyarbakir area, at
which Braidwood has more recently been collaborating with a
Turkish scholar, Halet f ambel. The material remains here can now
be regarded as characteristic of a developing society in its 'aceramic'
stage, with an adequate knowledge of agriculture and the
exploitation of domestic animals. But it has two unusual
characteristics, one of which is architectural. During the second of
four phases, the building remains consist of stone foundations or
'sleeper-walls', forming a regular 'grill' with spaces between too
small for any purpose other than storage ;
while, in the third
'
"^
phase, an extraordinary concrete floor was found, decorated in the

manner of a 'terrazzo' pavement with a pattern of stone chippings,
ground to a smooth surface. A second and equally surprising
innovation was the manufacture of simple implements by cold-
hammering native copper - the earliest use of metal yet known.
The Cayonii site represents 1000 years of occupation {c. "JSô to
6500 bc), immediately preceding the earliest levels of James
Mellaart's now-famous settlement at Qatal Hiiyiik on the Konya
Plain, where the record is prolonged through a further 800 or 900
years of spectacular progress and invention. '
^
30
It would be out of place here to describe in detail the many 6 The restored interior of a
shrine-room in the NeoHthic
aspects of the Qatal Hiiyiik settlement, so well published and
township of Qatal Hiiyiik, near
publicized in recent years. A township covering more than 15 acres
Konya in AnatoHa, ornamented
is composed of brick-built houses, arranged contiguously like a with mural paintings of bulls and
honeycomb and entered by ladders from the communal roof- a stag-hunt {c. 5800 Bc). (After
space. Of unique interest are buildings apparently associated with a Mellaart, 1967)
religious cult, their walls decorated with coloured murals, recalling

the cave-paintings of an earlier era, and with the heads or horns of
animals. Not only are pottery vessels in general use, but their
prototypes in wood or basketwork have survived - no less well
preserved than the fragments of woven fabrics which accompany
them. Human and animals figures are carved in stone or modelled
in clay, weapons and implements fashioned from delicately shaped
cores of flint or obsidian. Hunting still takes its place as an
important occupation, but the economy is now mainly sustained
by agriculture, with a surprising variety of food-plants under
cultivation and indications that even irrigation was beginning.
One would be reluctant to believe that this crescendo of cultural
advancement was an isolated phenomenon, unparallelled at this
time in other parts of the Near East. Yet, for the present this appears
to be the case, and there is little evidence of its having served to
accelerate the tempo of development in Neolithic societies
elsewhere. Indeed, during the Late Neolithic phase which
followed, many of its attainments seem to have been forgotten, and
signs are apparent of renewed intellectual inertia. In Turkey, this
period is well represented in the deeper levels of sites like Hacilar,
Sakâgozu, Mersin and Tarsus, where stone-built houses are
associated with characteristic black- or brown-burnished pottery
31
and finely made obsidian artifacts. But its culture seems no more
than a prelude to that of the more advanced ChalcoHthic period
that followed, about which there will be much to say when we
return to Mesopotamia.
Iran
For the moment, however, a few words must finally be said about
contemporary developments in Iran. For this purpose we should
first mention the particular which discoveries have been
areas in
made and note wide distribution. The search for
their significantly
prehistoric origins of Iranian culture hardly began until the 1950s
and many contributory discoveries have therefore been made too
recently yet to be recorded in the form of definitive publications.
An admirable summary published by P. Singh draws mainly on
preliminary communications; but in the present context it can be
highly recommended. Singh is able to distinguish geographically
three clusters of newly discovered sites, which have produced
'cultural assemblages' relevant to the Neolithic enquiry. '
^ The first
is in the river valleys to the south and of Kermanshah the

east ;
second in the plain of Khuzistan, within a 50-mile radius of Susa,

while the third lies to the south and southwest of Lake Reza'iyeh
(Urmia) in Azarbaijan. In each of these areas major and minor
excavations have taken place, resulting in discoveries which
amplify rather than supplement those we have already referred to
in Anatolia and the Levant.
Here are some details of primary sites in the above groupings and
their contributions to the sequence of Neolithic developments.
First, in the Kermanshah group, Tepe Asiab contains 'the vestige
of a semi-permanent settlement of food collectors ... on the verge

of achieving food-production' (estimated date 7100—6800 bc).
A further phase of the early food-collecting culture is seen at Tepe
Sarab, where there is still no mud-walled architecture, though the
manufacture of pottery, plain and painted, is now in evidence.
Another use of baked clay at this site is for various forms of figurine,
one of which, known as the 'Sarab Venus', shows particularly
skilful modelling. The goat is now domesticated, but there is no
proof of the presence of cereals (earliest carbon date 6000 bc). The
first signs of solid architecture are found at Ganj Dareh Tepe mud- :
brick, rectiUnear houses of the 'cluster-type', built against each

other, with walls of long plano-convex bricks laid in mortar. The
flint industry represents a blade-flake tradition and a total absence
of obsidian is still evident. Signs of polishing on some blades suggest

the harvesting of grain. Clay figurines already include a 'stalky-
headed' type which we shall meet with again in a later setting.
(Time bracket of this site, 7300-6900 bc). A well documented site
in this group is Tepe Guran, which again illustrates 'transition
from the hut to house and from aceramic to the ceramic stage'.
Plain pottery appears at the fourth occupation level and painted
or burnished wares The stone industry now includes obsidian
later.
and marble is used for making vessels by grinding. A notable archi-

tectural feature is the making of pavements of feldspar laid in clay
which is coloured with red ochre. (Neohthic culture dated
6500-5500 bc).
32
The second group of sites, in the Deh Luran plain west of

Susiana, also creates a sequence of cultural phases, now distin-
guished by site-names. The first two phases (Bus Mordeh and All
Kosh), are aceramic, but pottery appears in the third (Jaffar) phase.
Together they show that 'the era of early dry farming and caprine
domestication, from around 7000 bc, was gradually
starting
succeeded and replaced by the era of irrigation farming and cattle
dornestication, which began around 5000 bc', and finds a parallel
atChoga Mami in Iraq.
The third group of early village cultures, represented by sites
such as Hajji Firuz, Dalma and Pisdeli in southwest Azarbaijan, are
of a more-nearly Chalcolithic type and have less relevance in the
present context. The same may be said of Tepe Yahya, halfway
between Kirman and Bandar Abbas, whose remote situation alone
serves to emphasize the extended geographical pattern of such
sites, and may indicate a direction for further exploration.
^
'
The Iraq-Jarmo Project
Having attempted a conspectus of major contributions to the

Neolithic enquiry, made over a quarter-century by workers in
three important areas of western Asia, we are now in a position to
appraise the results of similar operations in northern Mesopotamia.
For this purpose, we should return to the Iraq-Jarmo project,
initiatedby Braidwood in the 1950s; and if in doing so we discover
that some of his conclusions have been superseded, we should
remember in extenuation the pioneer character of his work.
The results of operations at two of Braidwood's sites may be
selected, as taking a modest place in the sequence of cultural
developments now so well documented elsewhere. One of these is
the open site called Karim Shahir, to the north of Chemchemal in
the Kirkuk Liwa. This appears to have been one of several sites in
the neighbourhood, occupied only as a 'seasonal camp', at a time
imprecisely fixed (perhaps contemporary with Kenyon's 'Meso-
Jericho), when the movement from caves and shelters to
lithic' at
open settings was becoming more general. Hunting and fishing

were the main sources of food and there were no permanent
dwellings. Animals were not yet domesticated, and sickle-blades
were thought only to have been used in reaping reeds for building
purposes. Polished-stone celts and simple clay figurines were
among the few positive criteria of emancipation from a Palaeolithic
way of life. There is a long gap in time between Karim Shahir and
the site of Jarmo itself, where Braidwood excavated on a larger
scale between 1948 and 1955. The radiocarbon dates are
unsatisfactory; but, in modern charts of chronology, Jarmo
occupies a short period before and after the beginning of PPNB at
Jericho, with a mean date of about 6750 bc. In Braidwood's
terminology, it still takes its place in the category called 'Primary
'^
Effective Village Farming Communities'.
The village itself occupied about on an eroded shoulder
3 acres,
of conglomerate, overlooking a deep, dry wadi. In a total deposit of

7.6 m
there were twelve building levels with walls of pise (slabs of
clay superimposed on each other), latterly on stone foundations.
33
w:rk
7 The site of Jarmo, on the Pottery was found in the upper third of these strata, but there were
brow of an escarpment in the 'baked-in-place' floor-basins throughout. The impHcations of flint
Kurdish hill-country. In this
or stone implements for reaping or grinding were confirmed by the
excavation a system of
stratigraphic soundings was used presence of food-grains, including emmer wheat and two-row
to locate the Neohthic barley. A flint industry comprised both blades and microliths, with
settlement, where pottery a good deal of obsidian from Lake Van in use.
appeared only in the later levels
The pottery at Jarmo presented something of an enigma. Rather
{c. 5800 BC onwards)
well-made stone jars in the deeper levels gave way in mid-
occupation to a range of clay vessels, competently finished with a
burnished slip or painted with simple designs; but their quality
seemed to deteriorate nearer to the surface. Braidwood had
expected, through this medium, to establish a link with the earliest
pottery at Hassuna. In this however he was unsuccessful and the
most recent assessment of his painted wares'^ finds a parallel with
those at Tepe Guran (one of the best published sites in the
Kermanshah group of Iran), thus suggesting an association with the
Zagros Mountains, rather than the Mesopotamian lowlands.
34
Braidwood's dating of Jarmo may in fact be taken to suggest a
point in time for the abandonment of his settlement near the end of
the Late Neohthic elsewhere. If so, the earliest evolution of the
Hassuna culture should be sought in the lowlands themselves: at
Umm Dabaghiyah and sites excavated by the Russians in Sinjar
(p. yoff.). Here, however, the Neolithic phase merges so amor-
phously with that until recently called the 'Chalcolithic', that it
may better be dealt with in another chapter.
Before leaving Braidwood's site, there is something further to be
said about his findings.
We have earlier mentioned the inter-disciplinary element in the
group of research- workers which he brought to Iraq in the 1950s.
For the first time in the Near East, geologists and climatologists,
together with specialists in flora and fauna, were to be found
collaborating with archaeologists in their search for the en-
vironmental background of ancient man. From their reports, a
picture emerged of an ecological situation remarkably different
from that of the present day. The contrast which they have revealed
between 'then' and 'now' in the physical aspect of northern Iraq is
one outstanding result of their work.
As Braidwood says.
During the period which has elapsed since Jarmo was a village, man has
been the pre-eminent environmental influence, and the effects of his
handiwork are to be seen throughout the Near East. In general the role of
man, of his agriculture and of his flocks has been destructive, and this
without any one man wishing to be destructive.
'Today', Braidwood says,

throughout much of the once-wooded plain and foothill area of the
Chemchemal valley, hardly a shrub remains. The scrub-oak is rarely
allowed to reach more than six feet in height before it is hacked away by
the charcoal-burners. With the trees and the bush-cover gone, and the grass
eaten down to its roots each spring, the soil has largely gone too to silt up
the rivers. ... In winter it washes away on every slope almost as fast as it
can form and the rains rage off the land in chocolate torrents. ^°
Here then is the source of the alluvium which created the

Mesopotamian plain.
35
DIYALA SITES
BC PERIOD UR URUK ERIDU UBAID NIPPUR KISH 'UQAIR
ASMAR KHAFAJE AGRAB
Akkadian
Akkadian Graves Buildings Monument Palace Build ings Houses
Z
Graves Graves
Inanna Single
First
iiib Nm-Khursag Cemetery I
Shrine
Dynasty Temple
2500 Temple A I
Temple
Royal
Tombs
Temenos
Survives
i I _l_
- Oval Palace A
Terrace Ziggurat
Palaces
Square Shara
Enmerkar Houses
Temple Temple
Seal Archaic
Houses '-n
ipressions Temple
White
Temple
3000 Jemdet Nasr Chapel
Chapel
Temples
—i- 1
Painted
i-ii
Late Uruk Temple
Temenos
3250 Raised I
I-
III-V
Early Uruk
Walled
Temenos
m.
VI-VII
Settlement Settlement
'Ubaid0' Cemeterv
4000 Temples
VIII-XI
'Ubaid(3)
Levels
XII-XIV
4500 •Ubaid i
Hajyi
(Hajji
Muhammad
Muhamma
No Building
Temples
'UbaidT XV-XVIII
(Eriduy
Eridu Ware
Table III Chronology for sites in

southern Mesopotamia
36
Chapter Three
The Threshold of Written History
We must now return to the alluvial plain of SQUthern Mesopotamia

and the marshland which separates it from the head of the Arabian
Gulf in order to follow successive stages in the evolution of
Sumerian civilization. In that area, we shall fmd ourselves dealing
with a period of time whose beginning at least is easy to defme
because it corresponds to the earliest pre-Sumerian settlement of
which traces have yet been found, dating from a little before 5000
BC. It ends in about 2900 bc with the foundation of the Sumerian
dynasties and the first written references to the names of kings who
are now shown to have been historical characters. In the strictest
sense, the period could therefore be called prehistoric but since its ;
fmal phases cover the earliest use of writing and other major
inventions of the Sumerians, the word has seemed inappropriate
and other labels have had to be found for the chronological stages in
its development.
The Sequence of Discoveries
As early as 1929, among excavators in Mesopotamia, the practice

was adopted of naming sub-periods after the sites at which they
were first recognized. Later some attempt was made to subdivide
or re-group them under names more culturally meaningful. But
since by this time the sequence had become a subject for debate
among archaeologists of several nationalities, each with his own
conception of the stratigraphic implications, a point had been
reached where no single system of terminology seemed to be
universally accepted. With this difficulty in view, it may be well
here first to summarize the original discoveries in their chronologi-
cal order.
Al 'Ubaid
When excavations were resumed in Iraq after the First World War,
little was known about the Sumerians and even less about their
antecedents. The first indication of prehistoric occupation in the
south was found by Leonard WooUey at a small site called Al
'Ubaid, 4 miles to the west of Ur, where H. R. Hall had previously
located the brick platform on which a Sumerian temple had once
stood. Woolley
noticed, as Hall had done, that another part of the
mound was covered with sherds of painted pottery and, during a
brief excavation there, he was able to associate them with the
remains of reed-built houses, which for him created a picture of
primitive settlers 'on an island in the marshes'. Later in a deep
sounding which he made in the great city-mound of Ur itself, he
37
found the same painted pottery, underlying the Sumerian remains,

in a sequence of occupation-levels which suggested that it had been
in use over a long period of time. Styhstically he was able to divide
this series into three phases, though the terms by which he described
them have since been superseded by more recent discoveries.
Warka
While in the 1920s WooUey was making these discoveries,
excavators of other nationalities were experimenting with similar
soundings beneath Sumerian of which the most significant
cities,
was that undertaken by the German

Oriental Society under N.
Noldeke and J. Jordan in the great temple-precinct of Eanna at
Uruk (modern Warka).-' Here, eighteen 'archaic' occupation-
levels were identified, of which the five earliest (XVIII-XIII)
produced pottery corresponding to Woolley's 'Ur 'Ubaid'
assemblage. After a short transitional period (XIV-XII), these
painted wares disappeared and were replaced by a totally different
pottery, often wheel-made and without ornament, except for an
occasional red or grey 'slip' with a burnished surface. Apart from
the complete repertory of new shapes that now appeared, another
notable change could be seen in the complete absence of small
objects (such as terracotta figurines, baked-clay sickles and nail-
shaped wall-plugs), which had accompanied the painted wares
both here and at Ur. Phenomena such as these could be taken to
reflect some ethnic change in the city's population.
From level V upwards in the Eanna stratification at Warka, the
Germans soon had less reason to rely on the findings in their shaft.
Nearby a new excavation had begun which, in the years that
followed, was extended over a large part of the precinct. ^^ This led
by degrees to the astonishing exposure in levels V to III of
architectural remains, sculpture, cylinder-seals and inscriptions,
through which the marvellous attainments of this pre-dynastic
people were first revealed. Meanwhile, a further discovery had
been made at a site called Jemdet Nasr, a few miles from Kish,
where another Anglo-American expedition under S. Langdon had
been excavating on similar lines. ^^ Here, the architectural remains
were neither well clarified nor well published, but their interest was
greatly increased by an entirely new type of painted pottery which
accompanied them. This was a polychrome ware, with geometric
designs and a distinctively glossy surface. It now appeared to
represent the final phase of the pre-dynastic period with which we
are dealing.
It was at this stage that a conference of Mesopotamian
archaeologists, meeting at Leiden in 1929, agreed upon site-names
to designate the three main chronological phases so far recognized
in the pre-dynastic sequence.The 'Al 'Ubaid' phase was to cover
the whole period associated with painted pottery in the deepest
levels at Ur and Warka; the name 'Uruk' could provisionally be
applied to the post-'Ubaid, 'archaic' levels in the Warka sounding;
and 'Jemdet Nasr' to probably recognizable at Warka
a final phase,
as well as at the name-site itself. The primary need at that time was
for a further exposure and analysis of the lengthy 'Ubaid period. As
it proved, it was almost twenty years before this hope was fulfilled.
Eridu 8 The prehistoric temple at
In 1948, F. Safar and S. Lloyd, on behalf of the Iraq Antiquities Eridu. An impression of the final
rebuilding in ProtoUterate times
Directorate, opened a first season of excavations at Eridu (now
('Temple I'). Above the stone
called Tell Abu Shahrain), which the Sumerians considered to be retaining-wall of the temenos,
the oldest city in the world. ^"^ Here were the remains of a late- the platform had survived up to
Sumerian ziggurat and, beneath one corner of its ruins, the walls of the base of the columned
portico. Above this, the
a small temple, dated by pottery to the very end of the 'Ubaid
reconstruction is based on fallen
period. This proved to be the latest in a long series of rebuildings,
facade ornament and the
whose plans could be traced in turn through eighteen occupation- appearance of temples
levels, down to a primitive chapel founded on a dune of clean sand. elsewhere. (M. E. Weaver after
The evidence provided by the contents of these buildings was of Lloyd)
course invaluable for dating purposes. In particular, successive

stages in the development of 'Ubaid pottery could be noted,
form of the painted vessels so
the earliest of these, the technique and
little resembled the standard product of later times that they came
provisionally to be known as 'Eridu Ware'. Later we shall see how
this and other sub-phases of the Eridu sequence have been assessed
and numbered.
The succession of temples at Eridu did not of course end with the
'Ubaid period. A further five rebuildings corresponded to the
Uruk period, but of these little remained, save for the brickwork of
their repeatedly extended emplacements and fallen fragments of
their fa9ade ornament. But here there was a supplementary form of
dating evidence. At Eridu, early in the Uruk period, a spreading
cluster of subsidiary religious buildings had grown up around the
temple. At some point in time this whole complex was abandoned
and its rooms filled almost to roof-level with drifting sand. Then
(perhaps at the time of Uruk's own great period of architectural
invention, in levels V and IV) the whole of these ruins at Eridu had
been enclosed in a stone retaining-wall and paved over, to create a
raised temenos on which the temple could finally be reconstructed
39
URUK
XIV-IX VIII VII VI V IVc IVb IVa IIIc Illb Ilia ( ARCHAIC
LEVELS
EARLY URUK PROTOLITERATE Delougaz
WARKAN PROTOLITERATE Perkins
MIDDLE URUK LATE URUK PROTOLITERATE C, D

EARLY URUK Ehrich
PROTOLITERATE A PROTOLITERATE B JEMDET NASR
URUK
XII < LRl K Mallowan
(jEMDET NASR)
PROTOHISTORIC Moortgat
. '
,
EARLY URUK LATE URUK JEMDET NASR Lloyd
Table IV Pre-dynastic in a really magnificent form. The contents of the old buildings
terminologies according to
beneath the temenos, when compared with the remains of those
different authorities
afterwards superimposed upon it, now established a primary
division of the Uruk period into 'Early' and 'Late' phases.
Khafaje
Long before during the 1930s, new hght of another sort had
this,
already beenthrown on pre-dynastic chronology by excavations

undertaken by the Oriental Institute of the University of Chicago
in citiesof the ancient state called Eshnunna, in the Diyala region
east of modern Baghdad. These had been mainly concerned with
the Sumerian dynastic and later periods but the predecessors of one
;
Sumerian temple at Khafaje had been explored in depth, down to

an original foundation at the end of the Uruk period. ^= In
publishing the contents of this building, its excavator was the first
to reaUze that this late phase of the Uruk period (levels V and IV at
Uruk itself), with its great cultural accomplishments, must be
separated, as it afterwards was at Eridu, from the rather
undistinguished phase which preceded it (levels XIV-VI at Uruk).
He therefore proposed to combine the former with the equally
productive Jemdet Nasr phase, making a single period to which he
gave the name 'Protoliterate'. This term has in fact never been
Table IV easily accepted and several alternative systems of dating have since
been suggested. For the moment however it may best serve our
purpose.-^
In these few paragraphs, then, a preliminary account has been given

of the earliest attempts to create a chronological framework for the
pre-dynastic period, by rationalizing the stratigraphical evidence
40
from a number of sources.
In doing so, priority has been given to 9 The guesthouse (mudhij) of a
discoveries in southern Mesopotamia, mainly because the cultural modern Marsh Arab village a :
traditional form of reed

sequence established there long provided a system of criteria by
architecture already to be seen
which parallel findings in northern Iraq and neighbouring illustrated in carvings of the
countries could be judged. Before extending our review to such Protoliterate period. (From
regions, less directly connected as they are with Sumerian origins, it Leacroft, 1974)
may be well to study in greater detail the character of the southern

Mesopotamian cultures already enumerated.
The'Ubaid Period
Architecture
The text of a Sumerian inscription tells us that Shulgi, a king of the
Third Dynasty of Ur, 'cared greatly for the city of Eridu, which
was on the shore of the sea'. This description need not be taken too
literally. It might, as we have already shown, be concluded that the
city was merely connected to the sea by a system of tidal lagoons,

and that these formed part of a marshy area, similar to that which
today separates the alluvial plain from the head of the Arabian Gulf.
In any case, it seems certain that the first inhabitants of Eridu and
some other Sumerian sites were themselves marsh-dwellers, living
in an environment comparable to that of the modern Marsh Arabs
of Iraq. It is not therefore surprising that the earliest (symbolical)
representations of architecture on Sumerian cylinder-seals and
reliefs depict reed-built structures, resembling in their details the
ingeniously designed mudhifs or guest-houses to be seen in the
marshland villages today. ^^ This being so, it is a little disconcerting
found in
to find that the earliest identifiable religious buildings,
level XVIof the temple sounding at Eridu and elsewhere, were
already constructed of sun-dried bricks. Admittedly, in another
sounding at Eridu, unmistakeable remnants were found of a
modest reed structure; but this later proved to have been no more
than an outbuilding of a normal mud-brick house. ^^
The prototype 'temple' in level XVI at Eridu was in fact a single 10
compartment, no more than 3 m
square; yet its plan already
showed features, such as a cult-niche and central offering-table,
41
10 Temple plans characteristic of
the three main phases recognized
in the sounding at Eridu "Ubaid :
4' (temple VII) (top) "Ubaid 3'
(temple IX) (above) and "Ubaid i'

(temple XVI) (n;^/if), dated
respectively to c. 3800, 4100 and
4900 BC. Temple XVI is no more
than a chapel, but already has a
niche for the altar and
offering-table
42
which are characteristic of temples from that time onwards. In the 1 1 Temple VII at Eridu after
excavation. Overlying it behind
levels immediately above this (XV-XII), the siting of the building
can be seen the great bulk of the
seemed to be beyond the reach of the sounding but it reappeared in
;
Third Dynasty ziggurat, built
level XI and was twice rebuilt (XI-IX) to a more ambitious plan, by Shulgi (2095-2048 Bc) and in
having a central sanctuary and projecting lateral wings. In this case the foreground the extended
platforms of Protoliterate temples
its extremely narrow brick walls were strengthened at intervals
I-V. The walls are of rectangular
with thicker piers or buttresses, in a manner perhaps reminiscent of
mud-bricks
earlier reed structures. Next came a sequence of more substantially
built and sophisticated temples (VIII-VI), accounting for the
remainder of the 'Ubaid period. The long central sanctuary was
still entered through a lateral chamber, but it also tended to have
more ceremonial doorways at one end and a raised altar at the

other. Once more there was a free-standing pedestal for votive
offerings, apparently including fish, whose bones were deposited in
an adjoining chamber. ^^ Facades were now formally decorated
with alternating buttresses and recesses, a feature rarely absent from
religious buildings from that time onwards.
Religion
Having thus already encountered the prototype of Mesopotamian
religious buildings, about which so much will presently have to be
said, some impression should at once be given of the purpose for
which they were built: of the forms of worship and ritual
performance for which they provided a setting, and of their
significance as a primary expression of abstract thought and
spiritual consciousness. At the early stage in their evolution which
43
we have so far reached, there understandably Httle to be learnt on

is
this subject from their physical or architecturalremains as revealed

by excavation. A few elementary features can already be
recognized by analogy with subsequent developments - we have
spoken of an 'altar', 'cult-niche' and 'offering-table'. But a fuller
picture of the cults themselves to which the buildings were
dedicated, and of the actions by which they were expressed, can
only be reconstructed in the light of information provided by the
written texts of later times. Here then, in the briefest possible form,
are some elements of Sumerian religious worship and ritual
performance.
Duties and observances practised in the temple can be grouped
under two principal headings: offerings and sacrifices. High
amongst the services which the gods required of their worshippers
was the provision of food, drink and oil for anointing. According
to H. W. F. Saggs (1962), 'The gods enjoyed regular meals ., . .
which were placed on tables before the divine images'. Their food
included bread in large quantities, the meat of sheep or cattle and
drink in the form of beer, which was greatly favoured by the
Sumerians. Among provisions listed in later times were honey,
ghee, fme oil, milk, dates, figs, salt, cakes, poultry, fish and
vegetables.
The meal of the gods was technically a banquet to which other deities
were invited and at which the human worshippers and even the dead
might be present. The gods themselves received specified parts of the
animals, the remainder going to the king, the priests and the temple staff.
Sacrifices were These were actually made

in a different category.
on special altars or on the roof of the temple, by a qualified priest
who cut the animal's throat while reciting an incantation, the
welling blood itself being a libation. Other forms of libation are
12 Some painted designs used by frequently represented in Sumerian art, the wine or other liquid
the makers of 'Eridu ware', the being poured over an altar, on to the ground or over an animal or
earhest pottery yet found in plant. Saggs writes:
southern Mesopotamia. These
areflat dishes, but tall cups and The incense-burner was a very common
feature of the temple ritual, and
bowl-shapes were similarly the burning of aromatic woods could beapplied either as a purification rite
decorated. (Now known as or as a service to the deity, for the gods delighted in sweet odours.
"Ubaid i'). (After Lloyd and
Safar) Something more is to be learnt from the names of priests who
performed special parts of the ceremonies. Some were concerned
especially with incantations, and there were exorcists whose task
involved the playing of music. Others dealt with washing and
anointing rituals. Finally, in a more important category, were the
Baru priests, specialists in omen interpretation, who greatly
influenced pohtical decisions. It can well be understood that the
requirements of these varied religious performances are reflected in
the planning and installations of the buildings where they took
place.
Pottery and Small Objects

At each of the levels at Eridu so far mentioned (XVI-VI) pottery
was plentiful. The gradual transformation of the early 'Eridu'
designs into those associated elsewhere with Al 'Ubaid proved to be
separable into four distinct phases, which also conveniently
44
mm 13 Shapes and patterns

characteristic
pottery in
C'Ubaid 4').
its
of 'Ubaid painted
latest
At this
phase
period the
'Ubaidian culture seems to have
spread northwestwards as far as
Syria and Cilicia. (After Lloyd

and Safar, JN£S, 1923)
corresponded to the major changes in architectural planning,

recognized throughout the series of rebuildings. From 'virgin soil'
to level XVthe early 'Eridu' pottery could be seen to be 12
predominant, 'a fine monochrome, usually chocolate-coloured
painted ware, decorated with small-scale rectilinear patterns'. This
coincides in time with the modest building activities which
culminated in the level XVI chapel, and has been designated 'phase
i'. A second phase accounted for levels XIV-XII above, where no
architecture was found, and it was distinguished by the intrusion of

an alien pottery, whose peculiar 'wickerwork' patterns had
previously been found at a riverside site called Hajji Muhammad,
near Warka.^" 'Phases 3 and 4' corresponded to earlier and later 13
stages in the development of the conventional 'Ubaid culture,
which reached its peak at the time of Temples VII and VI, and it is
in this stage at Eridu that a new and rich source of evidence becomes
available.
The Cemetery
on the southwest side
Fifty metres outside the stone retaining wall,
of the raised temple-precinct, a cemetery was found containing
perhaps 1000 graves of the late 'Ubaid period; some 200 of these
were excavated. The burials were made in rectangular shafts, lined
45
14 Typical small objects and with mud-brick, and their position must have been marked on the
sherds with figured designs from surface, since a secondary interment was often made in the same
an "Ubaid 4' settlement (Tell
shaft as the first. The body of a dog was occasionally laid across the
'Uqair). The former include
sickles, wall-nails, net-weights grave of its master.
and socketed axes (all of baked The Eridu cemetery, which may be dated to the first half of the
clay), polished celts, a hammer 4th millennium bc, represents a fairly advanced stage in the
and digging stones
development of funerary practices. As we shall presently see, earlier
sites have been found where burials are concentrated in particular
areas, often in the vicinity of holy places. But here, for the first time,
the number and uniformity of the graves create the impression of a
genuine necropolis. The location chosen at Eridu seems to have
been just outside the 'Ubaidian settlement and its temple precinct.
No corresponding graveyard of any later period has yet been found
there, though early excavators at this site, having come upon small
deposits of votive pottery within the raised precinct of the Uruk
period, did mistake them for human burials (with the bones
missing). At Ur, WooUey found a dozen or so 'Ubaid burials in the
area which later became the 'Royal Cemetery' - once more, likely
to have been just outside the contemporary temple precinct. He
observed that, as at Eridu, 'the body was placed on its back, fully
extended with its arms by its side or slightly bent so that the hands
could be folded across the pelvis', and he comments that 'this was an
Opposite
15, 16 Baked-clay figurines from attitude peculiar to the 'Ubaid people, never to be adopted by any
'Ubaidian graves at Ur (female) of the subsequent inhabitants of Sumer'. Woolley also recorded the
and Eridu (male). The 'coffee- fact that,
bean eyes', conical headdress and
shoulder ornament (tattoo ?) are In two body was covered with a fine red
graves, the upper part of the
widespread conventions at this powder, and in one by the head a lump of red haematite
case there lay
and earher periods (cf ill. 45) paint. Whether the bodies had been painted it was impossible to say; but
46
there was no doubt that the powder was the same as the paint in the
lump. . . .
It is interesting that a similar phenomenon was observed in our own
cemetery at Eridu. Almost all the bones were pigmented with a

dark orange colour. But as this applied equally to the bodies of dogs
- and even in one case to a meat-bone laid beside the mouth of the
dog - we are inclined to discount the possibility of a deliberate or
ritual pigmentation with ochre, and to attribute the colour to some
chemical action of the soil.
The contents of these burials included a rare collection of

complete painted pots, matching very closely the contemporary
sherds collected from the temple sounding (levels VI and VII). The
shapes, painted designs and technical peculiarities of 'Ubaid pottery
at this time are of considerable archaeological importance, because
the 'Ubaid culture extended itself, now or later, far beyond the
frontiers of Mesopotamia to the east and northwest. It is a hand-
made ware and its curvilinear or other designs are freely painted
with a soft brush, usually in black or dark brown. Foliate patterns
often occur and occasionally there are stylized figures of animals or
birds. At Eridu, the paint is generally applied over a buff or cream-
coloured slip. Elsewhere the slip is more often omitted and the
vessel fired at so high a temperature that the clay acquires a dark
green colour with the black paint biting deeply into it. The fact that
such vessels frequently suffer from over-baking, almost to the point
of vitrification, and are often deformed in the process, suggests that
these 'Ubaid potters had very little control over their kilns.
The fragmentary vessels found in Woolley's 'reed huts' at Al
'Ubaid itself were in other ways poorly made and seem now to
have represented a decadent phase immediately preceding the end
of the 'Ubaid period. They, like some examples from level VI at
Eridu, show signs of being made on a 'tournette', or hand-turned
wheel, thus anticipating the wheel-made pottery of the Uruk
period. ^ Surface finds suggest that a deeper penetration would
'
have encountered earlier 'Ubaid occupations.

Objects other than pottery, found to be characteristic of the
'Ubaid period, also have an interest of their own. First and foremost
are the painted terracotta figurines, described as 'mother-goddesses'
male counterpart was found in a woman's tomb at Eridu.
until a
Their 'lizard-shaped' heads and bitumen headdresses, which
created so much interest when examples were first found by
Woolley, can today be recognized as a convention widely accepted
in the cult-imagery of the prehistoric Near East. Some earlier stages
in the evolution of their design are discussed below in connection
with finds made at sites in northern Iraq. In another category are the
implements made of baked clay, such as sickles and shaft-hole axes,
whose manufacture must have been made practicable by the high
temperatures attained in kilns of that time (the sickles were
evidently baked in bundles for they are often found adhering to
each other in a vitrified mass). In this class also are the large clay nails
with bent-back points, used as we think for fixing reed matting to
the face of a mud-brick wall. For the rest there are simple artifacts,
such as flint knives and hoes or bone implements with bitumen
handles, which could belong to any other pre-dynastic phase.
47
UrGate
The Uruk Period
17 Site-plan of the city of Warka

Uruk/Warka, showing the Anu
The most copious source of information about this period has been
and Eanna precincts in which
the work of German archaeologists, whose excavations at the
ProtoHterate temples were
found. The huge buildings, Bit name-site have been carried on intermittently for almost half a
Resh and Irigal, are temples built century,more recently under the direction of H.J. Lenzen. Uruk,
at the time when the site was whose modern name is Warka, is also the 'Erech' of the Old
reoccupied during the Seleucid
Testament. In Sumerian literature it is associated with the name of
Dynasty (331-329 bc). The
outUne of the city may have Gilgamesh, one of the earliest dynastic rulers credited today with
been much the same at the time having been an actual historical character. But already in pre-
of Gilgamesh. (From Hawkes, dynastic times the city is thought to have had an area of some 200
1974) acres, about one third of which was covered with temples and other
48
public buildings. Soundings in widely separated parts of the site

have suggested that even in the 'Ubaid period the settlement had
attained an impressive size. In the famous epic of which Gilgamesh 17
was the hero, it is said of him that: 'In Uruk he built walls, a great
rampart and the temple of blessed Eanna for the god of the
firmament Anu, and for Ishtar the goddess of love'. Sure enough,
when the Germans came to excavate the temple area in the centre of
the city, they found their work concentrated on two primary
building complexes, which they named respectively the 'Anu
Ziggurat' and the 'Eanna Precinct'. It was greatly t6 their
advantage that, in both cases, ruins of the pre-dynastic period
proved to be accessible at no great depth beneath the surface of the
mound.
The Anu Area

The earliest discoveries in the Anu area are easier to understand
today than they were when first made. The building called by its
excavators the 'White Temple', whose walls, painted white
outside, appeared standing upon an irregular-shaped brick
platform, was in fact nomore than a late reconstruction of a shrine
very much like that at Eridu, in that its origins could, as we now
know, be traced back far into the 'Ubaid period.^- Nevertheless,
from
the planning and arrangement of this final rebuilding, dating
thejemdet Nasr period (3200-2900 bc), are of special interest, in
that all the main characteristics of the 'Ubaid temples are still
preserved. Only the area of its platform is greatly increased. Its
panelled facades slope inwards at a slight angle and approached

it is
by an impressive triple stairway. Since no 'platform temple' of a

later period than this has survived, we shall presently see that it
gains significance as a prototype of the Sumerian ziggurat.

Fifty metres or so to the east of the White Temple were the outer
18 A 'cut-open' reconstruction
walls of the Eanna Precinct, a broad area of courtyards and terraces
of the 'White Temple' at Warka,
surrounding a true ziggurat, dedicated to the goddess Inanna and standing on its high platform
repeatedly rebuilt in historical times. It was early discovered that known to the excavators as the
this monumental layout had replaced a more ancient complex of 'Anu Ziggurat'. Beneath it they
found traces of earher temples
religious buildings, dating from the final phases of the pre-dynastic
going back in date to the 'Ubaid
period. We have already seen how the Germans, by means of a period. (From Leacroft, 1974)
'pilot' sounding in the middle of this area, were able tentatively to
construct a chronological sequence for these 'archaic' buildings and

to relate them stratigraphically with occupation-levels of an earlier
period beneath. During the years that followed, the Eanna
excavation was widely extended and the full attention of the
excavators was devoted to the task of disentangling their plans.
The problem which this presented can well be imagined if it is
realized that such buildings were constructed almost exclusively of
perishable mud brick. In the end the excavators were faced with
half-a-dozen temples, complete with their various appendages,
founded, extended, destroyed, rebuilt and generally overlapping
each other, the stumps of their walls often standing no more than a
few inches high. If one remembers that the shape and direction of
these walls could only be determined by patiently articulating each
individual brick, one cannot but regard the clearance of Eanna as a
major accomplishment of archaeological technology.
49
;'Mv.^MAV,\
AOVhU tAn
Y *^
'j
i
.-râili
19 A German architect's The Eanna Precinct

reconstruction of the 'Pillar Hall'
It would be impossible here to describe each individual building in
at Warka ('Uruk IVc', c. 3200
Bc). Its brick columns, half-
this section of the Warka excavations, or to appraise the
columns and walls are entirely reconstructions of their plans, some of v^hich have been ingeniously
covered with coloured cone- made from exiguous surviving fragments. Buildings of special
mosaic in geometrical patterns.
interest are not however difficult to select. One of these - better
Between the central stairways, a
miniature temple is represented.
preserved than most- is the so-called 'Pillar Hall' (level IVc), which
(I. Mackenzie-Kerr after provided the first in situ example of the 'cone-mosaic' facade
Heinrich, 1932) ornament, so characteristic of the period. It takes the form of a
gigantic portico, 30 m wide, comprising a double row of circular,
free-standing columns, 2 m in diameter, with corresponding half-
columns at either end. This is approached at a lower level through a
long rectangular courtyard. The lateral walls of the courtyard too
are faced with contiguous half-columns, and from it the colonnade
is approached by three separate stairways. The entire surface of
columns and side-walls alike is decorated with a mosaic of

terracotta cones, thrust into a bed of clay, their coloured ends
forming a variety of geometrical patterns. This remarkable
architectural composition appears to have created a monumental
approach to some building beyond, either uncompleted or totally
destroyed. The face of the double stairway in the centre is decorated
to resemble in miniature the facade of a building, also ornamented
with mosaics.
Cone-mosaic ornament is again more sparingly used on the walls
of actual temples built at this period. Their plans seem to adhere to
two quite different conventions. One of these is represented by the
standard 'tripartite' form, with which we have become familiar at
Eridu and again in the White Temple. The second type of plan is a
complete innovation, having transepts opening off the sanctuary at
one end, with a doorway between them leading to a separate cella
on the main axis. The finest individual example of this type is to be
20 seen in the restored plan of Temple D in level IVa. Its very
dimensions (80 x 55 m) make this building remarkable, while its
planning (if the excavator's restoration may be accepted) is an
astonishing tour deforce. The great T-shaped sanctuary is entered
through its transepts and surrounded by lateral chambers on all four
sides. The outer facades are decorated with elaborate chasing and,
on the two longer faces, ornamental niches deeply recessed in the
50
facades alternate with stairway chambers, some accessible from

outside. In anotherexample (Temple C) in the same level, a simpler
plan of the same sort is extended at the cella end to include a second,
'tripartite' element, corresponding exactly in shape and dimensions
to the White Temple, as though the building in some way fulfilled
a double function.
We shall presently learn that the T-shaped or 'cruciform' sanctu-
ary, as a featureof temple planning, is not restricted to southern
Mesopotamia. Many interesting examples of the same arrangement
have more recently been found at sites in north or east Iraq as well
as in north Syria. Some appear to be contemporary with those
just described, others to belong to an even earlier date (see

Postscript).
One other building of interest - though of doubtful purpose -
20 Plans, reliably reconstructed,
was the awkwardly named 'Stone Cone Mosaic' temple, built on of Temples C and D in the
an isolated site to the west of the main Eanna grouping and assigned Eanna Precinct at Warka (level
in date also to level IV. Enigmatically planned, it was surrounded IVa, c. 3100 BC). In these two
by a curious, doubly-buttressed protective wall, whose inner face, buildings the T-shaped sanctuary
is an innovation, elsewhere
like the facades of the building itself, had been decorated with a
found only in northern
mosaic of cones, in this case shaped out of coloured stone. Stone Mesopotamia. Annexed to
cones of this sort have been the subject of much speculation, since Temple C is a smaller unit,
they appear in considerable quantities, displaced from their original resembling the White Temple.
setting, at Eridu, Al 'Ubaid and elsewhere. They have at times been These buildings, the largest of
which measures 80 x 55 m,
thought of as a more primitive form of ornament, for which
emphasize the precocious
terracotta cones were later substituted, in the cause of greater ingenuity of Protoliterate
economy. This seems probable, if one considers the variety and architects. (After Lenzen, 1949)
21 Architectural components of
a Late Uruk period temple
('Uqair) : prismatic bricks
(riemchen) of clay or gypsum,
elements of cone-mosaic fa(;:ade
ornament, rain-water spout and

round pedestals of unknown
purpose
elaboration of such ornament towards the end of the Uruk period.

In some cases, for instance,hollow-ended terracotta cones, up to
30.5 cm in length, formed an ornamental band around the parapet
of a temple platform, while in a late building at Eridu similar cones
of gypsum were used, their heads sheathed in copper. Sometimes
also, figured ornament carved on small plaques of stone was
incorporated in the pattern of the mosaic and other devices were

used to enliven it.
Tell 'Uqair
At many sites in lower Mesopotamia, a scatter of these small mosaic
components appears on the surface, and can be recognized as a
reliable testimony to the occupation of such sites during the Uruk
period. One site discovered in this way was Tell 'Uqair, about 50
miles due south of modern Baghdad, excavated by archaeologists

from the Iraq Museum, led by F. Safar and S. Lloyd, in 1940-1.^^
On the outer edge of a characteristic 'Ubaid settlement, a slightly
higher mound covered the remains of an Uruk-period temple,
with a well-preserved platform and some parts of the walls still
22 standing several metres high. The plan in this case • precisely

matched that of the White Temple at Uruk, except that the
ceremonial doorways at one end of the sanctuary were replaced by
a high altar on the main axis, approached by a miniature flight of
steps. It was, however, through one of the lateral doors that the
excavators were (for once) able to enter the building, and it was
then immediately discovered that the whole of the internal wall-
faces were covered with mural paintings in several colours. A
'dado' of wine-coloured paint was surmounted by a band of
geometrical ornament, and above this there was a frieze of human
and animal figures, of which only the lower parts had survived.
Better preserved was the decoration on the faces of the main altar,
which, like the central stairway in the 'Pillar Hall' at Uruk, was
painted to represent the facade of a miniature temple with its
vertical panels of mosaic. On the sides facing the steps, guardian

2j figures of two spotted leopards were represented in red and black
52
22 Surviving walls and platform
of the Late Uruk period temple
at Tell 'Uqair. Wall-paintings
in the sanctuary are being
recorded. In the foreground are
remnants of the ground-level
chapel of thejemdet Nasr
period, where archaic tablets
were found
23 Painted leopard, forming part

of the altar decoration in the
sanctuary of the Late Uruk
period temple at Tell 'Uqair
53
paint. Apart from the lime-washed facade which had given the
White Temple at Uruk its name, and a 'Red Temple' in the Eanna
Precinct, of which only a few courses remained standing, this mural
painting at 'Uqair was the only example known in a building of the
pre-dynastic period, and it continues to be remarkable for that
reason.
The Painted Temple at 'Uqair has been provisionally dated to
the fmal phase of theUruk period (c. 3250 bc). Again like the White
Temple, chambers had been packed solid with rather large mud
its
bricks to create a higher platform, on which the shrine could once

more be rebuilt on a more ambitious scale. The date at which this
took place could not of course be ascertained, but a discovery at the
foot of the platform afforded some indication. Here was a small
subsidiary chapel, packed with the remains of ex-voto vessels: a
collection ofJemdet Nasr pottery, fmer in quality than any found at
the name-site itself. Here, furthermore, chronological evidence of
the most reliable sort was provided by the discovery of four clay
tablets, which seemed to be items from the temple accounts,
inscribed with the pictographic signs typical of the Jemdet Nasr
period.
Apart from the ground-plans and wall-ornament of the temples
hitherto described, somethingmore can be gathered about their
general appearance from the representations of architecture
sometimes to be found on Sumerian cylinder-seals and carved
High up in the space between their lateral buttresses,
reliefs. 3^
triangularwindows - either real or ornamental - are often shown.

But, where internal lighting is concerned, they leave little doubt
that the walls of the central sanctuary rose higher than those of the
subsidiary chambers, and that accordingly space was left for
clerestory windows. In the case of the great temples of the Eanna
Precinct, it has sometimes been doubted whether their sanctuaries
could ever have been roofed at all. Today, however, most authorities
are agreed that suitable timber could probably have been obtained
by river transport.
Buildings of the Jemdet Nasr Period
Warka
We should now remind ourselves of the proposal made in 1942,
that the 'Late Uruk period',
which the most impressive buildings
to
in the Eanna Precinct belong, should be combined chronologically
with the Jemdet Nasr period to create a new major phase called
'Protohterate'. Clearly, if this is accepted, no discussion of
Protoliterate architecture would be complete without some further
reference to buildings of Jemdet Nasr times. So far, only one of
these has been mentioned at Uruk itself, namely the fmal rebuilding
of the White Temple; and we should next perhaps glance at
contemporary developments in the Eanna Precinct during that
period. Here, the great temples of Uruk IV were now in ruins and
their sites covered with the confused remains of courtyards and
terraces, whose walls were occasionally enriched with panels of
mosaic ornament. One curious feature of this layout was the
frequent appearance of installations called by the Germans
54
opferstdtten. These were long plastered troughs, evidently used for

incinerating the remnants of sacrificial offerings in the form of
animals, birds or fish. Sometimes they were laid out in orderly rows
and showed signs of having been re-used many times. They have
occasionally been recognized at sites other than Uruk at about this
same period. ^^
One actual building, dedicated to some ritual purpose of the
same sort, was the so-called Riemchengebdude,^^ whose foundations
were dug down into the ruins of the earlier 'Stone-Cone-Mosaic'
temple on the west periphery of the Precinct. Its innermost
chamber, to which there was no apparent access, was entirely
surrounded by a corridor, showing signs of some violent
conflagration. Within the building, disposed in great confusion as
though discarded intentionally, was a rich collection of objects,
including 'hundreds of pottery and stone vases, alabaster bowls,
copper vessels, clay cones, gold leaf and nails with heads covered in
gold leaf, weapons, animal bones and broken components of
furniture'. All this was taken by the excavators to imply the ritual
destruction of fittings and sanctified objects belonging to some
earlier building which had to be demolished.
Khafaje
A great deal more, well-published information about the Jemdet
Nasr period was obtained from the earliest remains of the so-called
'Sin Temple', dedicated to the moon-god of that name, excavated
by Americans in the 1930s at Khafaje in the Diyala region, east of
modern Baghdad. ^ ^ This little shrine was a ground-level building,
yet its plan closely approximated to that of the White Temple.
Only the ceremonial doorways at the ends of its sanctuary were
missing. Its five earliest building-levels all fall within the period
with which we are concerned, and something will be said about
their contents if we now discuss aspects of the Protoliterate period,
other than architecture.
The Protoliterate Period
The First Writing

Having now watched the crescendo of architectural activity and
invention which marked the course of the Protoliterate period, we
are able to infer the strength of religious belief by which it was
inspired. From buildings alone, however, little can be learnt about
the detailed anatomy of Sumerian religion, or about the social
background against which it had been conceived. Further
information on this subject was of course to be revealed in due
course by the written records of later times but for the moment the
;
elementary forms in which writing was available limited the

contribution which could be expected from this source. ^^
The earliest tablets as yet discovered come from level IV at Uruk
and use a pictographic script, recognizable as the ancestor of the 24
later cuneiform. ^ 9 The degree of competence which had already
it
attained suggests that earlier stages in its development may

eventually be recognized elsewhere, perhaps in levels correspond-
55
24 Reverse side of a pictographic

tablet from Jemdet Nasr [c. 3000
BC), giving a list of commodities.
Such tablets v^êre at first written
in vertical columns, starting top
right; later they were turned
ninety degrees, anticlockwise, to
read horizontally, left to right
ing to Uruk V and VI (which have for that reason been

provisionally included in the Protoliteratc period). ^^'' Regarding
the language for which this script provided a vehicle, the first texts
demonstrably written in Sumerian are found in a Jemdet Nasr
setting. Since, however, some authorities have been tempted to
conclude that an 'ethnic substructure', in part Semitic, may have
existed in Sumer previous to an (hypothetical) immigration of true
Sumerians, they would claim that a different language may be
expressed in the texts dated to Uruk IV. Speculations of this sort
form a part of the long-standing debate on the problem of
Sumerian origins (see below).
Some signs used in the Protoliteratc script have a recognizable
equivalent in the developed cuneiform of later times, and their
meaning is accordingly known. Animals such as sheep and goats,
cattle and donkeys could all have been expected. Words connected
with fishing and hunting also occur, while commerce is implied by
the names of merchants. Pottery techniques themselves confirm
that the wheel was known, and a 'sledge' mounted on wheels could
become a 'chariot'. Other signs suggest metal objects, one of
which, the shaft-hole axe, is made in a closed mould. Like the
copper sheathing of mosaic cones already mentioned, this would
suggest an improved understanding of metallurgy. In a different
sphere, it is significant that, although such words occur as 'elder'
and 'council', the concept of a 'king' seems still unfamiliar. Such
rudimentary scraps of information are fortunately supplemented
by the subjects depicted in sculpture and designs in relief carving.
Sculpture
In sculpture surviving from the Protoliteratc period, the scenes
depicted are predominantly religious. At this point, therefore, a
56
digressionmay not be out of place regarding the character and

composition of the Sumerian pantheon.
From the earhest written formulations of rehgious belief, the
names of three male gods, Anu, Enlil and Enki emerge as
dominating figures. Anu, the sky-god, whose temple we have
noticed at Uruk, was originally recognized as the highest power in
the universe and sovereign of all Sumerian

the gods. Later in
history, he seems to have been replaced in this capacity, first by
Enlil, patron god of Nippur and secondly by Marduk, tutelary
deity of Babylon. Enki.of Eridu was in a class by himself as god of
wisdom and learning. Generally speaking, though the whole
country worshipped a common pantheon, each individual city
retained its own patron god and its own set of legends. For the rest,
in the words of George Roux,
The heavens were populated with hundreds of supremely powerful, man-

like beings, and each of these gods was assigned to a particular task or a
particular sphere of activity. One god for instance might have charge of the
sky, another of the air, a third of the sweet waters, and so forth, down to
humbler deities responsible for the plough, the brick, the flint or the
pickaxe.
These gods had the physical appearance as well as all the qualities
and defects of human beings. 'In brief, as Roux concludes, 'they
represented the best and the worst of human nature on a
superhuman scale'. "^^
Examples of those associated with particular cities, who became
also the objectof a general cult, include the moon-god, Nanna (Sin)
of Ur and his son, the sun-god Utu (Shamash) of Sippar and Larsa;
Ninurta, the warrior-god; Nin-khursag (Nintu), mother and wife
of Enlil; Inanna (Ishtar), goddess of love and her husband Dumuzi
(Tammuz). Inanna is of course the Great Mother, worshipped
throughout the land, to whom the Eanna precinct at Uruk was
dedicated: a female principle of creativity, expressing godhead
through fecundity. As for Dumuzi, Roux points out that, though
undoubtedly associated in the Sumerian mind with productivity in
the vegetable and animal world, so that his union with Inanna must
symbolize fertility, the theory which long associated him with the
'dying and resurrected' god of Frazer's The Golden Bough has more
recently been rejected. Other minor gods and goddesses are too
numerous to mention.
A well-known stone vase from Warka, 90 cm high, has three 23
registers of figures very finely carved in relief. Above, the goddess
Inanna appears in front of two reed bundles terminating in loops 26
and streamers, which are her perpetual symbol, and a ritually naked
priest offers her a basket of fruit. Behind this the relief is damaged,
but a small fragment remains of a figure which may well be that of a
'king' or leader, and an attendant supports the tasselled girdle
which he is perhaps about to present. Inanna is supported by minor
deities, mounted on model temples and appropriate beasts, with
other symbols including a pair of vases like that on which they are
carved. In the second register, naked priests bring further offerings
and in the third, beasts and plants represent her two 'kingdoms'.
Another sacred symbol present here, is the stylized rosette with
57
25, 26 {Above left) Sculptured eight petals. Again, on the sculptured sides of a stone trough in the
vase of alabaster, found in a
British Museum, sheep and rams from the 'sacred herd' return to
Jemdet Nasr setting at Warka (c.
3000 BC). In the upper register,

her temple, here represented as a marsh-dweller's reed building,
the 'king' figure is missing, but a and from it lambs emerge to meet them. This also is a scene
their
servant is seen supporting his perpetually repeated, for example on the faces of a stone bowl from
heavy girdle (perhaps an
Khafaje, where cattle replace sheep, and much later in an
offering? Cf. ill. 28, third row
architectural relief from the Early Dynastic temple at Al 'Ubaid.
down). The reverse side {above
right), repaired in antiquity, Secular subjects are less common. There is the fragmentary stela
shows the goddess Inanna, to of black granite from Warka, carved with difficulty, to show two
whom offerings are being scenes in which the same 'king' figure hunts lions; and there is the
brought by a procession of
extraordinary life-size head in limestone, found in a Jemdet Nasr
priests, extended over the second
register. Ht of vase, 90 cm
setting, but dated stylistically to Uruk IV. It is only a component
part of a composite figure, the remainder of which could have been
made from other materials such as wood and bitumen. The eye and
eyebrow inlays are missing, as is the metal-foil covering of the hair.
With these restored, the head looks a little outlandish but the bare
;
stone mask, as found, is strikingly beautiful. Animals carved in the

round - a couchant ram from Warka and a wild boar from Ur -
seem to have been ornaments attached by metal to larger
contrivances, while stone vases also are sometimes decorated with
animals - lions or bulls - carved partly in the round and partly in
58
relief. Dating from the Jemdet Nasr period at Warka, there are
crudely carved 'king' figures, foreshadowing the ex-voto 'personal'
statues of Early Dynastic times. Nearer still to these is a single
votive statue of a woman from Khafaje. Of the same period are jars
of animal form with an opening on top, usually of dark-coloured
stone.These are often decorated with coloured inlays in geometrical
patterns.
Cylinder-seals
Another innovation which runs parallel to the pictographiC'Writing
of Protoliterate times is the cyhnder-seal, a device used for
establishing ownership or recording an agreement. These small
objects of stone or shell, pierced through the centre for suspension,
were delicately carved on their curved sides with a variety of
devices or designs, to create a tiny frieze of ornament when rolled
over soft clay. They have been found in great numbers and their
"^
'
impressions on tablets or clay jar-stoppers are even more plentiful.

The cylinder-seal is a device which makes its first appearance in
levels V and IV at Uruk. ^ In the examples dating from this period,
"^
therefore, we should expect to see the Sumerian craftsman's first,

and perhaps awkward, experiments in an extremely difficult form
of carving. On the contrary, though it is clear that he had not yet
completely mastered the intricacies of repetitive design, the variety
of his subjects and the ingenuity of their decorative treatment
testify to an achievement seldom rivalled during the centuries that
followed. Already his work has the quality of relief carving, rather 28
than mere linear drawing, as is shown by the vigorous modelling
and articulation of the individual figures. "^^^
The pictorial themes and mythical symbols contributing to the
composition of these designs have been the subject of much study
and are today partly understood. Most evident in the religious
scenes is the invocation of two deities never personally depicted,
but symbolized by ideograms implying their attributes. There is a
god who, like the biblical Tammuz, personifies the generative force
in nature, graphically represented by
certain animals and plants.
Other symbols of the god include an eagle, sometimes lion-headed,
and a snake. Equally prominent is Inanna, whose curiously shaped
emblem (identified by some as the 'gatepost' of a reed-built
temple), afterwards became her name-sign in pictographic writing.
A decorative combination of these symbols alone can provide the 27 Carved alabaster trough in
subject of a seal design. Alternatively, when the performance of the British Museum, of the
Protohterate period. Animals of
some religious ritual - sacrifice or oblation - is portrayed, human
Inanna's 'sacred herd' beside a
figures appear. A
bearded man with diadem and 'chignon',
reed-built byre, surmounted by
wearing a long folded skirt and attended by a naked or kilted priest, her 'doorpost' symbols. Ht
can be identified as a 'king'. He brings ritual objects to a reed-built 20 cm
59
_\ N ^'r -
28 Cylinder-seals of the Uruk

period. The 'king' figure
appears, feeding Inanna's beasts
or facing her portable shrine on a
reed boat. Priests bring

ceremonial gifts (one of them by
boat) to a reed-built temple with
unidentified 'ring-pillar'
symbols. Ibexes and other beasts
provide alternative motifs
temple or presides over the ceremonial feeding of sacred cattle.

Flocks and herds make frequent appearances and are defended
against lions by such figures as the 'bull-man' and 'lion-headed
eagle', familiar features of Sumerian imagery in later times. Game
animals also appear boars, stags, ibex and moufflon, and from their
:
shapes heraldic patterns are composed to supplement the seal-

cutters repertory of designs. Occasionally there are secular subjects,
including the 'king' figure, either hunting or at war.
Types of stone favoured by the seal-cutters are often of a sort
unobtainable in Mesopotamia and suggest trade connections with
neighbouring countries.
29 During the Jemdet Nasr period, there was a slight deterioration
in the care and precision with which seals were carved. The drill, for
instance, which had previously been used sparingly and its effects
when possible concealed, now became conspicuous and the
modelling of animal figures declined in accomplishment. A new
60
form of seal also appeared, longer in proportion to its diameter and

entirely covered with sharply defmed geometric ornament.
Pottery
While tracing successive phases in the evolution of pre-dynastic
culture, mention has frequently been made of pottery, stressing its
significance as stratigraphical evidence. The pottery of the
Protoliterate period is for the most part undecorated. Its technical
have been the subject of much specialized study, which
peculiarities
would not here be possible to summarize. One exception to this
generalization however is the fme painted ware which appeared
towards the end of the Jemdet Nasr period."*^ The commonest 30
form of vessel decorated in this way is a stoutly proportioned jar,
with a broad shoulder and rimmed neck. Geometric ornament in
red and yellow is usually confmed to the shoulder, sometimes
forming panels in which animals, foliage or even human figures
appear. The remaining surface is covered "with plum-coloured
paint, forming a glossy surface. This distinctive ware has provided a
useful criterion for dating, when found in alien settings.
29 Cylinder-seals of the Jemdet

Nasr period. In those with
animal motifs (above), the quahty
of the carving is impaired by the
use of a bov^^-drill. The
geometrical designs below are
peculiar to this phase of seal-
'// ; s cutting
61
30 Painted pottery of the Jemdet

Nasi period. Designs are in
plum-red, yellow and black,
mostly on the shoulders of jars,
and the vessels are given an
attractive 'glossy' finish. (After
Lloyd and Safar, 1943)
Parallels between pottery of the pre-dynastic period in

Mesopotamia and that of contemporary cultures in Iran and
elsewhere will be discussed later under a separate heading; but the
subject of early relations between Mesopotamia and Egypt may be
mentioned here. In doing so, we are concerned less with pottery
than with certain cultural developments, apparently shared at this
time between the two great centres of incipient civilization. Several
aspects of this subject have been ably examined by scholars in recent
times. '^'^ One aspect is the influence of the earliest Mesopotamian
brick temples on the design of pre-dynastic 'mastaba' architecture
in Egypt. Another is the contention that hieroglyphic writing in
Egypt was stimulated by a knowledge of Sumerian or pre-
Sumerian experiments in the same field. This has today been made
to seem less improbable by the discovery of pictographic writing,
similar to that of Mesopotamia, in settings as remotely placed as
Romania on the one hand, "^^ ^nd the frontiers of Baluchistan on the
other. "^^ But the most tangible evidence of communication
between the two countries at such an early period is the discovery
of Jemdet Nasr cylinder-seals in graves at Naqada in Egypt,
indicating some form of trade connection, whose existence at that
time still requires further investigation.
Sumerian Antecedents
Finally, we must return to the problem of Sumerian origins, which

has in the past been the subject of so much deliberation among
scholars. Here there has been a conspicuous conflict between the
62
reasoning of philologists on the one hand and the conclusions of

archaeologists on the other. Broadly, the problem arises from
uncertainty regarding the exact point m the sequence of pre-
dynastic 'periods' at which the inhabitants of lower Mesopotamia
can rightly be called 'Sumerians'. It has been said that 'the
innovations of the Protoliterate period established the identity
which Mesopotamian civilization retained throughout its long
and this was at one time accepted as a strong argument for
history',
the 'arrival' of the Sumerians at the beginning of that period. The
epigraphists' contention to this effect is on a single
largely based
premise - namely that some of the older Mesopotamian cities
mentioned in historical texts and founded before the Protoliterate
period could be thought to have non-Sumerian names. This has
been used to justify a whole tissue of speculation about the
hypothetical migration of earlier, perhaps Semitic peoples
throughout the Near East."^^ Archaeologists, on the other hand,
point to the uncertainty of the original premise and to the absence
of proof that these names were actually in use before the invention
of writing. Finally, they are impressed by the overwhelming
evidence for cultural continuity between the 'Ubaid and the Uruk
-^^
periods.
The evidence alone for continuity of religious beliefs and
practices to an archaeologist seem particularly convincing. Three
sites already mentioned make impressive contributions to the
argument. The sequences at Al 'Ubaid itself, Tell 'Uqair and Eridu
start chronologically with a conventional 'Ubaid settlement and.
3 1 Map showing the known

distribution of Ubaidian sites in
I50m/
the Arabian Gulf. (S. Ebrahim
300 km after J. Oates, 1976)
SAUDI ARABIA
. RIYADH
63
by analogy with Eridu, the two former may also be assumed to

have had an 'Ubaid-period temple, frequently rebuilt. At these two
sites, the settlements themselves were abandoned at a point in time
corresponding to the end of the so-called 'Ubaid period. Not so,

however, the temple which, like that at Eridu, was repeatedly
rebuilt and enlarged throughout the Protoliterate period. At
'Ubaid itself it was again rebuilt in Early Dynastic times and at
Eridu its ruins were chosen as the site for a ziggurat by the Third
Dynasty kings of Ur. One also observes that, both at 'Ubaid and at
'Uqair, the old 'Ubaidian settlement mound was used as a cemetery
by the Sumerians of Dynastic times. "^"^
With this situation in mind, the substitution at any time of alien

immigrants for the indigenous population becomes an improbable
hypothesis.
A fmal question remains, regarding the origmal provenance of
the southern Mesopotamia. Much interest has now
first settlers in
been aroused by the discovery of a thriving enclave of 'Ubaidian

31 settlements along the southern shore of the Arabian Gulf and inland
west of Bahrein. 5° Their earliest appearance in this quarter is at
present dated to the middle phase of the 'Ubaid period. More
recently, however, neutron activation evidence has become
available,showing that the pottery in use by these settlers was in
fact imported from Sumer, to which their relationship must
accordingly have been that o{ colonists or traders. Speculation
regarding an 'earlier home' o{ the Ubaidian culture would thus
appear to become increasingly pointless.
64
Chapter Four
Pre-literate Peoples of Northern

Mesopotamia
During the previous chapter our attention has been concentrated

on the results of excavations in the southern Mesopotamian plain,
and on the evidence v^hich they produced of cultural developments
during the pre-dynastic period. We must next turn to consider
parallel researches v^hich had been taking place among the uplands
of northern Iraq, and examine the rather different sequence of events
v^hich they revealed. In doing so, we shall find that our thread of
association with the evolving civilization of the Sumerians may
become increasingly while cultural influences from
tenuous,
beyond the frontiers of Mesopotamia correspondingly gain in
importance. For this reason, the term 'pre-dynastic' will clearly no
longer be appropriate and, for the period intervening between the
Neolithic and Early Bronze Ages, when the smelting of metal
began to be understood, the name 'Chalcolithic' should perhaps be
substituted, as hasbeen usual in other countries of western Asia. ^ ^
Once more then in this case, it may be well to sketch the sequence
of discoveries, before discussing in detail the succession of cultural

phases which they have made it possible to establish. Table V
Excavations in the North
Arpachiyah
In Assyria the initial investigation of prehistoric settlements was
undertaken simultaneously by British and American archaeologists
during the early 1930s. As a point-of-departure one would select
the small site called Arpachiyah, 4 miles to the northeast of
Nineveh, excavated by M. E. L. Mallowan in 1933.^^ Here on the
surface Mallowan found a little village composed of rather poor
huts built o£ pise clay. But he noticed that the pottery which they
contained could be recognized as a northern variant of the 'Ubaid
painted wares previously found in Babylonia. There was a wider
variety in the colours used - brown, red and even purple paint
being more common than black; but the designs and the shapes
remained substantially the same. This was confirmed when, outside
the village, a cemetery was found containing more than forty
graves, containing many complete vessels, some of them unbroken.
Also, in the houses themselves were found incised terracotta beads
and bent clay both generally characteristic of 'Ubaid sites in
nails,
the south. Finally, Mallowan found one open-cast copper axe

suggesting that, though still rare, copper was already in use side-by-
side with the usual flint implements, as one would expect in a
Chalcohthic setting.
65
Pre-literate Peoples of Northern Mesopotamia
BC SOUTHERN GAWRA NINEVEH GRAI RESH BRAK

SEQUENCE
Akkadian Palace
iiib
2500
ma Ninevite
l-Q
Five
Gawra
3000 Jemdet Nasr Period IV 'Eye'
VIII-X ii-iii
Temple
Late Uruk
IV I
XI-XII
ARPA-
Early Uruk III CHIYAH
3500
XIII
Late VI-IX
Acropohs 'Ubaid
'Ubaid I
Houses
® :2
r}
xiv-xvi and
Graves
.D
4000 I-IV
Late
'Ubaid
HASSUNA
XVII-XIX
®
'Ubaid
Transition
4500 Hajyi
Muhammad VI-X
Halaf
® Halaf
Halaf
Round
Houses
Endu
5000 . SamarraT
t i
o
Hassuna
Camp-sites
Table V Chronology for sites 6000 Umm

northern Mesopotamia Dabaghiyah Jarmo
Similar houses vêre found in four successive building levels (1-4

dov^n) ; but Mallowân did not think that these represented more
than five or six generations, because the graves in the cemetery did
not overlap, suggesting that the positions of the oldest graves wêre
still remembered w^hen the latest ones vêre dug. In level 5 the
character of the pottery began to change, and throughout the
remainder of his excavation (6-10), he found himself dealing wîth
an entirely different culture, characterized by a finely finished and
brilliant wâre, painted for the most part in several colours. This
pottery was already knov^n by the name 'Tell Halaf. It had first
been found by the German excavator, Baron Max von Oppen-
66
heim, before the First World War, at a site of that name on the
Turko-Syrian frontier. But von Oppenheim had encountered it
while excavating the great Iron Age palace of an Aramaean prince,
in the city called Guzana - piles of painted potsherds, thrown out
when the foundations of the palace were being dug - and he could
find no indication of its age. At Arpachiyah it was for the first time
found both accurately stratified and associated with contemporary
architecture.
More will need to be said later about this beautiful Halaf pottery, 43-4
on which the elaborate designs with an
are executed in glossy paint
overall burnish. Its character is sufficiently distinctive to have
provided an invaluable dating criterion when found at the further
extremities of its distribution in northern Syria and southern
Anatolia. In the same sense, it is important to notice that it has no
exact equivalent anywhere in southern Mesopotamia. At Ar-
pachiyah also, the first examples were found of small objects in
other categories - stylized cult-figurines of terracotta, stamp-seals
with figured designs of men or animals and a variety of beads and
amulets - all of which came soon to be considered representative of
the 'Halaf period'. Yet the feature by which this site has been
primarily remembered are the circular buildings, found in its deeper
levels and known at the time by the Greek word 'tholoi'. Built of j6
pise on stone foundations and apparently domed, they were
sometimes approached through a rectangular ante-room or
'dromos', thus increasing their resemblance to the 'bee-hive' tombs
of a much later age at Mycenae.
Tepe Gawra
While Mallowan was excavating Arpachiyah, an American
expedition had begun work on a much larger mound called Tepe
Gawra, some miles to the northeast. ^ 3 As was usual in those days,
the Americans were more liberally financed than the British, and
their leader, E. A. Speiser, was consequently able to plan an
excavation covering the whole summit of the Gawra mound.
During the years that followed, therefore, its silhouette was
dramatically changed, from an initial height of over 18 m to rather
less than 9 m. The shortcomings till then associated with the
restricted area of stratigraphical soundings were thus eliminated

and the exposure of the entire settlement made possible. During
their earlier years, Speiser and his successors were engaged in
clearing levels of habitation contemporary with the Sumerian cities
in the south, or corresponding with the Protoliterate phase of the
pre-dynastic period. The latter proved to represent a culture so
it was referred to as
distinctive in character that, for the time being
the 'Gawra period' of northern Iraq. There were rich graves with
gold jewellery; temples with curious, unfamiliar planning;
strikingly carved amulets and everywhere evidence of the large-
scale use of copper. These were some of the characteristics of
Gawra, levels VIII to XI (down), which will presently be referred
to at greater length, since Gawra serves as a key-site to the
stratigraphy of northern Mesopotamia.
It was not until the Arpachiyah excavations were over that the
workers at Gawra began themselves to encounter Mallowan's
67
32 Tepe Gawra, plan of the

'acropolis' at level XIII (c. 3600
BC), with its 'Ubaidian temples

facing an open courtyard. The
fragile wall construction is
reminiscent of the ''Ubaid 3'
temples at Eridu, levels XI-IX

(cf. ill. 10). There are no ritual
installations in this case, a,
northern temple; b, central
temple; c, eastern shrine; dotted
line indicates edge of mound.
(S. Ebrahim after Tobler, 1950)
'northern 'Ubaid' painted pottery. Here they were able to study its
development over a period represented by no less than eight

building-levels (XII to XIX), and to learn a great deal more about
32 it. It W2is in level XIII that the culture seemed have reached its
to
state of maximum sophistication. At this time, the greater part of
the mound's summit was occupied by what amounted to a
religious acropolis; an open space surrounded by a trio of large and
elaborately built mud-brick temples, satisfactorily corresponding
in many respects to those in the 'Ubaid levels at Eridu, with which
they were later shown to be contemporary.
All through the deeper 'Ubaid levels at Gawra, traces were
found of pottery shapes and forms of ornament surviving from the
previous Tell Halaf period. When level XX was reached, the
transition between the two cultures was found to have been passed.
Polychrome pottery predominated and all other fmds could be seen
to be characteristic of the Tell Halaf period. There was even an
interesting circular building which could, not quite conclusively,
be identified as a temple. But at this point the Gawra sounding
ended and no further clue was found to an earlier occupation.
It remains to be added that, already known at this time, was one
other form of painted pottery, which seemed to have been

contemporary with or to have preceded the manufacture of Tell
Halaf ware. This had first been found before 1914 by Ernst
Herzfeld, when he was excavating the huge, 9th-century ad city of
Samarra, on the Tigris, halfway between Mosul and Baghdad, ^^
41-2 and came to be known as 'Samarra ware'. Among the foundation
68
of Islamic buildings, Herzfeld came upon traces of a prehistoric
settlement or cemetery, where this beautiful and unusual pottery

had been in general use. Herzfeld being at that time unable to detect
any architectural remains or stratigraphical indications, the age of
the settlement temporarily remained a mystery, and it is only in
much more recent times that the date of its first appearance - a little
before that of the earliest Halaf pottery - has been ascertained. This
too will later be explained.
KiJYUNJIK
The combined results of excavations so far mentioned constituted
our total knowledge of northern Mesopotamian stratigraphy up to
the outbreak of the Second World War. It remains only to discuss
an immensely deep sounding made by Mallowan beneath the
Assyrian Ishtar Temple in the Kiiyiinjik mound at Nineveh. ^ s This
operation had a number of extremely important results. First, in an
upper level, Mallowan had the good fortune to discover the hfe-
size bronze head of an Akkadian king, now considered one of the
great masterpieces of Mesopotamian art (see chapter 7) ; secondly,
he found textual proof that in Sumerian times Nineveh had been a
city with temples. Thirdly, he was able to confirm the sequence of
prehistoric phases so far known and, more significantly, to prove
the existence of an earlier, pre-Halaf culture. Directly above virgin
m
soil, at a depth of 27 beneath the summit of the mound, he found
an unknown type of pottery with finely scratched incisions which
he provisionally labelled 'Ninevite I'. For ten years after Mallowan
had left Iraq to excavate in Syria, a small bag of these curious
potsherds remained in the Iraq Museum, to represent the earUest
form of pottery as yet found in Mesopotamia.
Hassuna
The next development came in 1943, when inspectors of the Iraq
Antiquities Directorate located a small mound called Hassuna, on
the fringe of the cultivated upland 20 miles due south of Mosul,
whose surface was littered with Ninevite I pottery. The excavation
of Hassuna involved a renewal of the 'Uqair partnership between
the present writer and Fuad Safar: its result far exceeded our
expectations. ^ ^ A new and earliest chapter was added to the history
of Chalcolithic settlements in northern Iraq, beginning with the
arrival of a still-nomadic people, whose camp-sites were almost
immediately overlaid by the primitive houses of a small farming
community. An excavation covering an area about 20 msquare
enabled us to follow the development of this village through six
successive building-levels, and to associate its final occupation with
the arrival of new inhabitants at the site, using an early form of Tell
Halaf painted pottery.
Counting from the bottom upwards, levels III to V represented
what we considered to be the 'Standard' Hassuna culture. The
pottery was all unburnished; some incised with pin-scratched 40
designs, some painted and some combining both forms of
ornament. In the three levels below (II, Ic and lb), a so-called
'Archaic ware' appeared, burnished or decorated with glossy paint.
69
Throughout these six sub-phases, the general features of the village

remained fairly consistent; small houses of pise clay were grouped
around courtyards resembling farmyards. There were reasons to
think that they had pitched roofs of wood and thatch, exactly as are
found in neighbouring villages today. There were also ample
indications of primitive agriculture, sickles with flint teeth set in
bitumen on a wooden haft, showing that at least these people were
reaping the -wild grain which still grows in the vicinity. But there
were also elaborate grain-stores, protected with gypsum and
bitumen, which proved that seed-grain was kept for re-sowing.
Curious oval 'husking-trays' of baked clay were used for the
process of winnowing. Bones of domestic animals were plentiful,
while obsidian arrowheads and javelins indicated that hunting was
still practised. Some sort of religious belief was already suggested
by a damaged 'goddess' figurine of a now well-known type.

In levels lb and Ic, plain burnished pottery was mostly of a sort
associated elsewhere with the late stages of the Neolithic; and then
finally, directly above virgin soil in level la, there was a complete
change. Shelters, if any, must have been of a perishable material, for
all we found were several widely separated domestic hearths, each
with an identical assemblage of primitive artifacts around it. There

was crudely made pottery, always including a peculiar, tall-sided
storage-jar, apparently for milk, and implements such as heavy
chert hoes or handaxes. In one case a skeleton still lay crouched
beside the ashes of a fire. Here then, as it seemed to us, were the
symbols of a transition from the nomadic life of herdsmen and
hunters to the economy of a settled farming community.
Other Hassuna Sites

During more recent years, two major attempts have been made to
amplify our knowledge of the Hassuna culture and if possible to
trace its immediate antecedents. One was the work of a Soviet

expedition at a site called Yarim Tepe, of Tell
to the southwest
'Afar in the Sinjar northwest of Mosul. ^^ Here, one of
district,
several mounds was excavated from 1969, with exemplary attention

to method, and the results recorded with almost exaggerated
precision. They served both to amplify and to supplement those of
our own excavations at Hassuna. Many small houses - over 100
rooms in all - were examined, at thirteen successive occupation-
levels, covering the whole sequence of developments originally
recognized by ourselves at the name-site, and some features
previously lacking made welcome additions to our knowledge.
These included bracelets of lead, a new and interesting category of
human and also prototypes of the stamp-seals familiar in
figurines
Soundings at other sites in the same area (e.g. Tell
later periods.
es-Soto and Maghzaliyah) produced proto-Hassuna or aceramic
material.
Meanwhile, in the early 1970s, a second investigation with a
similar purpose was undertaken by British excavators led by Diana

Kirkbride in the Jazirah steppe country to the west of Hatra. In this
area, now totally desiccated, 87 sites were found, of which no less
than 40 showed signs of occupation in the Hassuna/Samarra period.

One of them, named Umm Dabaghiyah, was selected for
70
excavation down to virgin soil, and for the excavators the deeper 33 Onagers (wild asses) depicted
levels had a surprise in store. ^ ^ Here they were faced with abundant in a primitive wall-painting from
the pre-Hassuna settlement at
parallels to the earliest occupation (level la) at Hassuna, but, in
Umm Dabaghiyah in the Jazirah
contrast to the nomadic camp-sites of the first arrivals at that site,
(c. 5700 Bc). These animals were
substantial buildings appeared observed at Yarim

of a sort already hunted for their skins. (After
Tepe (level I) and clearly designed for some purpose other than Kirkbride, 1975)
domestic habitation. Rows of rectangular compartments, too small

for habitation, were grouped in a grid-shaped plan, as though for
storing some commodity suitable for trade. The evidence of animal 34
bones found in profusion nearby, of which onager and gazelle
accounted for 68% and i6% respectively, led the excavators to
consider some sort of commerce in skins - possibly those of the
onager, which survived in that region until late in the 19th century
AD. This theory was rather dramatically confirmed in the final
season of excavation when, on the walls of dwelling-houses in the
deeper levels, rudimentary mural paintings were found, clearly
representing onagers in flight, apparently towards a Hne of netting 33
staked to the ground. ^^
Diana Kirkbride proposed a date for the Umm
Dabaghiyah
settlement in the early 6th millennium bc, contemporary with the
deepest level at Hassuna, thus attributing it to the phase which we
have elsewhere labelled 'Pottery Neolithic'. She could recognize no
relationship between these skin-trading hunters and the hill-people
ofJarmo in the northeast, though the two cannot have been widely
separated chronologically.
Samarra and Sawwan

During the years following the finds at Hassuna, a secondary
problem which occupied Mesopotamian prehistorians concerned
the status of the distinctive and finely painted Samarra pottery,
which appeared side-by-side with the fully developed 'standard'
Hassuna wares, appearing to be an extraneous product, perhaps
iiTiported from elsewhere. At Matarrah, south of Kirkuk, where
Braidwood excavated soon after the publication of Hassuna, this
pottery seemed to be preponderant he therefore concluded that he
;
was dealing with a basically 'Samarran' culture. ^° Few distinctive

criteria of such a culture were, however, there apparent and its final
71
revelation in an authentic setting had to await the discovery in

1964, by an of a new and extremely
Iraqi scholar, B. Abu-al-Soof,
productive site in the immediate of Samarra itself. ^ This
vicinity '
was the settlement now known as Tell-es-Sawwan, where five

occupation-levels were subsequently excavated, all but the earliest
exclusively associated with the intriguing painted wares, originally
found unstratified in this area by Herzfeld. Like Kirkbride's site in
the Jazirah, Tell-es-Sawwan held many surprises. In addition to
dwelling-houses of sun-dried brick, there were strange, T-shaped
buildings with many rooms, identified by the excavators as
55 granaries. For a contemporary example of a fortification, at Choga
Mami, see Curtis (1983), figs. 12-13.
During the third occupation, the whole area of habitation had
been enclosed by an outer wall, irregularly buttressed and protected
by a formidable ditch. Other discoveries at Tell-es-Sawwan included
more than 100 burials, mostly beneath the floors of the houses. Grave-
goods from this source were indeed remarkable: quantities of fine
alabaster vessels, and, of the same material, a striking collection of
45 human figurines, almost totally dissimilar from those of baked clay,
found elsewhere in a Chalcolithic setting. ^^ In another sphere,
food-grains and other vegetable remains together with a plentiful
variety offish and animal bones, have made it possible to envisage
the life of the settlement a mixed economy based on agriculture,
:
aided by rudimentary irrigation, hunting and herdsmanship.

Here then was an indigenous manifestation of the hitherto
elusive Samarran culture, centred apparently upon the Tigris at the
northern extremity of the alluvial plain. Excavations at related sites
have indicated its geographical extension, in the east to the frontiers
of Iran (Choga Mami, near Mandali) and westward to the
Euphrates (Baghouz, near Abu Kemal).^^ The fact that at Tell-es-
Sawwan hardly a single sherd of Hassuna ware was found
emphasizes the disparity between the two cultures.
Architecture
The Pre-Halaf Period

At the earhest village sites in northern Iraq, house walls are most
commonly of pise, and rendered in mud plaster. In rarer cases

built
(e.g. Umm Dabaghiyah), gypsum plaster was preferred for wall-
faces and pavements alike. Sun-dried bricks, shaped in a wooden
mould, do not appear until the end of the Hassuna period. Stone
foundations occur only at Jarmo and Maghzaliyah, where stream-
beds could have provided the material. At Jarmo also, pavements
were of 'clean mud packed over beds of reeds'. From the begin-
ning, multi-roomed dwelling-houses seem to have been planned
on a rectilinear principle. At Umm
Dabaghiyah each consisted of
one main room, often measuring no more than 2 x 1.5 m, and
others even smaller, reached through archways less than i high.m
Here, unexpected features were fireplaces with hooded chimney-
flues, connected through the base of the walls with external bread
ovens. A single example of a similar contrivance was found at

Jarmo. A built-up step and toe-holes in the walls apparently made
access to these rooms possible through an opening in the roof At
72
«P*.W
Yarim Tepe, the size and number of the rooms was considerably 34 Plan of the Umm
Dabaghiyah settlement at level
greater, one single house measuring 14 x6 m. Houses were built
IV, with storage buildings,
in contiguous groups (Umm Dabaghiyah), lining narrow alleys
perhaps for drying onager skins.
(Yarim Tepe) or facing their own open courtyards (Hassuna). In (After Kirkbride, 1975)
addition to the usual domed bread ovens, many houses were pro-
vided with large circular grain-bins, partly buried in the ground
and lined with gypsum or protected by a coat of bitumen.
In the village at UmmDabaghiyah, domestic quarters were
subordinated to the two great cellular blocks of storage chambers, 34
whose hypothetical association with a trade in hides has already
been mentioned. Their vsâlls were of 'strongly tempered clay',
50 cm thick and unplastered. The individual cells, of which over
seventy were excavated, had an average dimension of
1.50 X 1.75 m and no communicating doorways.
The Samarran Period

Architecture of this period is represented at two sites on the
periphery of the southern Mesopotamian alluvium: Choga
Mami^3 q^ the Iranian frontier and Tell-es-Sawwan on the
Tigris.^' The former was a large, agricultural village
(350 X 150 m), apparently profiting from some primitive system of
Houses had up to twelve rooms, with walls
artificial irrigation.
builtof 'cigar-shaped mud-bricks' and plastered with clay. In the

more sophisticated settlement at Tell-es-Sawwan, building
methods had improved and prismatic bricks (Arabic, lihn) were
used, with average dimensions of 80 x 30 x 8 cm. The large T-
shaped buildings at this site, built to a standard plan with up to 35
fourteen rooms, are not yet fully explained. Doorways between
rooms showed that they were for habitation and some of their
contents suggested normal domestic occupation yet they were first
;
identified as granaries.
73
35 Tell-es-Sawwan, layout of
the Samarra period settlement at
level III {c. 5350 Bc), with
distinctively planned buildings in
a walled enclosure. (After Abu
al-Soof, 1968)
The Halaf Period

In the Halaf levels at Arpachiyah, an architectural innovation takes
J6 the form of circular buildings, called by the excavators 'tholoi'. At
various levels in the mound itself or in outlying parts of the site as
many as ten of these were found, though no more than two seem to
have existed simultaneously. The earliest are isolated circular
structures, 4 m or so across; but later their diameter increases to
over 6 m, and the thickness of their walls to 1.65 m. They are now
approached through a rectangular 'drOmos', like the 'bee-hive
tombs' at Mycenae, and in one case there is a transverse
antechamber. As a rule only their stone foundations had survived;
but one provided evidence of api5e superstructure- probably a low
dome. Owing to the number of graves and cult-figurines found in
their immediate vicinity, Mallowan was inclined to attribute a
religious purpose to these buildings. But recent excavations in
Sinjar, Syria and Iran have demonstrated that at most Tell Halaf
sites, unlike Arpachiyah, the circular structures are predominantly
dwelling houses. ^"^ Within the frontiers of Iraq, a contemporary

parallel to the Arpachiyah 'tholoi' occurs at Tepe Gawra (level XX)
and there are earlier examples at Yarim Tepe.
74
The 'Ubaid Period

The earliest building in northern Iraq to be confidently identified as
a temple occurs in level XIX numbering downwards at Tepe
Gawra, and coincides with the beginning of the 'Ubaid period at
that site. It has a rectilinear plan, with flimsily built mud-brick
walls, but is reconstructed on a more impressive scale in level
XVIII, with no less than twenty rooms arranged around a central
sanctuary (?). This has a podium in the middle but no recognizable
altar. Circular buildings reappear in level XVII, but they show no
evidence of cult-practices and temples are not again found until
level XIII, when, as has already been mentioned, the summit of the
mound was cleared to create an emplacement for a group of three
obviously religious buildings, with distinctive plans and wall-
treatment. They were built around an open courtyard, almost 20 m
wide, overlooked on three sides by their facades, which were
ornamented with a complicated system of multiple recessing and
painted in different colours. Of the so-called 'Northern Temple'
the entire plan was recovered : a rectangular sanctuary running the
whole length of the building, with two lateral chambers on either
side, separated by deeply recessed niches whose purpose is not
apparent. As is usual in a pre-literate setting, there is no evidence to
associate the building with a particular cult. The mud-brick walls
are surprisingly thin, but strengthened at regular intervals by
stouter piers, upon which the ceiling beams must be presumed to
have rested. This system of building is one which we have already
seen in temples XI-IX at Eridu in the south, though the sub-period
which they represent would appear to antedate the 'acropolis' at
Gawra. In the north, Gawra XIII seems to represent the
culmination of the 'Ubaidian period. In level XII there were signs
of a great conflagration which had destroyed the entire settlement.
(But see references to more recent discoveries of 'Ubaidian archi- 36 Alternative reconstructions of
the 'tholoi' at Arpachiyah:
tecture in Postscript on p. 231.)
circular buildings, entered
through a rectangular 'dromos'
The 'Gawra' Period (Tell Halaf period, c.4800 bc).
It is from a sequence of four later occupations at Gawra (levels
They may, or may not be
religious shrines. (After
XI-VIII), that most is developments in northern
to be learnt about
Mallowan and Rose, 1935)
Iraq during the phase which in the south has been called the Uruk
period. Here, the cultural transformation following the end of the
'Ubaidian epoch can be seen as clearly in the architectural remains
as in the character of the pottery and the increasing profusion of
metal objects. Rather surprisingly, the most conspicuous building
in level XI was a circular affair known to the excavators as the
'Round House'. Its metre-thick outer wall has a diameter of over
1 8 m and internally the plan can be seen to comprise no less than
seventeen rooms. Their arrangement has no ritual significance and 37
there is no justification for attributing to the building any religious
purpose. In this connection we are reminded by D. and J. Oates
(1976) that
round houses are found in some of the earliest settlements in the Levant
from the Natufian period onwards and represent, with rectangular houses,
one of the two simple forms of permanent structure that might have
evolved from different traditions of construction in the temporary
dwellings of a mobile population.
75
There accordingly seems to be little reason for attributing to the
'Round House' in level XI any non-secular function. Indeed, it is
during this same occupation that elsewhere the first example
/ appears of an actual temple, planned in a manner which was to
persist throughout the remainder of the Uruk period. These
peculiar buildings, of a type found until now only at Gawra, have
aroused a good deal of interest. ^^ They consist of a rectangular
sanctuary, approached through an open portico, with two lateral
chambers on either side. One example in level VIII has a free-
37 The heavily buih 'Round
standing altar or offering-table at one end of the cella, while
House, in level XI at Tepe
Gawra : a building of unknown another has facades ornamented all round with doubly recessed
purpose, corresponding in date vertical panels. It is notable that more than one such building was
to the Protohterate period in often found in the same level and that burials seemed to be
southern Mesopotamia. (M. E.
concentrated around them.
Weaver after Tobler, 1950)
Tell Brak
The only other temple building of the Uruk period in this northern
y area of Mesopotamia was found by M. E. L. Mallowan in 1938 at
Tell Brak, a site near the River Khabur beyond the present Syrian
frontier. ^^ This large building (30 x25 m) differed completely
from those at Tepe Gawra described above, having many features
in common with the Uruk temples in the south. The central
sanctuary extended from one side of the building to the other and
38 Projected reconstruction of a
had small transepts at one end, similar to those mentioned earlier in
'Gawra period' temple in level
VIII. Several examples were level IV at Uruk itself A row of lateral chambers on one side was
found of temples built to this balanced by a more complicated arrangement of chambers on the
plan, which is peculiar to the site other, and the whole enclosed in a stout mud-brick wall on stone
and period. A comparison with
foundations. Its decoration, with coloured clay cones and inlaid-
the pre-Greek 'megaron' plan
has been tentatively suggested.
stone rosettes, provided another parallel with the southern temples.
(P. P. Pratt after Speiser, 1935) But most surprismg of all was the elaborate ornament in the
39 An artist's reconstruction of
the main shrine-room in the 'Eye
Temple' at TeU Brak (Jemdet
Nasr period). The altar and wall-
faces were ornamented with rich
mosaic and inlay, partly of gold.
The floor was Httered with hiany
thousands of beads and small >^X
amulets. (Cf. ill. 46). (P. P. Pratt)
76
sanctuary itself. Its whitewashed walls were 'decorated with

coloured stone rosettes, strips of red limestone inlay and copper
panelling'. There was an altar at one dnd, bordered at top and
bottom with sheet-gold between bands of coloured stone, attached
by gold-headed silver nails. The temple was raised upon the ruins of
two other similar buildings and it was named by the excavators the
'Eye Temple', on account of the 'eye' or 'spectacle' motif appearing 39
everywhere in the ornament and in the form of innumerable votive 46
images of alabaster, built into the structure of the temple platform
in a medley of beads, amulets and stamp-seals. Since these had
clearly survived from the lifetime of earlier temples Mallowan
attributed the Eye Temple itself to a late stage in the Uruk-Jemdet
Nasr period. Regarding the significance of the 'eye' symbol, the
excavators could reach no definite conclusion.
Pottery
Pre-Halaf
In northern Iraq, the task of correlating the finds from a dozen
important excavations and of analysing the interplay of successive
cultural changes which they imply has now been prolonged for
almost half a century. It has involved a long process of meticulous
recording, while the patient consideration of logical inferences has
moved from one stage to another. In this milieu, the study of
pottery and small finds has assumed an even greater importance
than was the case in the south, where other forms of evidence were
available and the influence of intrusive cultures from beyond the
frontiers of Mesopotamia less evident. As a result, and particularly
in the realm of pottery, the field of study has assumed a high degree
of complexity and may well appear intimidating to the student or
layman who must approach it through a labyrinth of technical
description and diagrammatic illustration. In the present context,
therefore, a simplified commentary on variants in ceramic practices
and some other artifactual peculiarities may be sufficient to convey
an impression of the varied evidence which has contributed to an
orderly interpretation of prehistoric developments.
At the earliest sites hitherto discussed in this chapter, we have
been largely concerned with the transition from what has been
called Pre-Pottery Neolithic to Chalcolithic cultures. Con-
veniently, this has been found to coincide with the first use of
painted ornament on pottery. The initiative in this respect must be
attributed to the first settlers in the Jazirah and Sinjar districts. At
Umm Dabaghiyah the painted patterns are primitive - spots,
stripes and chevrons - but there are also crude figures of men and
animals in plastic relief. At Hassuna burnished and coarse, straw-
tempered pottery in level la gives way to 'archaic' painted wares
(levels Ib-III), with parallel lines of glossy paint or a burnished
finish. The remainder of the pre-Halaf occupations produced the
so-called 'Hassuna standard' wares, with pin-scratched or painted 40

ornament, used together or separately on an unburnished slip to
decorate a very limited range of characteristically shaped bowls and
jars. Side-by-side with these from level III upwards, Samarra
pottery appears, distinctive in itself and clearly a luxury 41
commodity imported from elsewhere, or copied locally.
77
40 Painted and/or incised

'standard wares' and a 'husking-
tray': all characteristic of levels
III-VI at Hassuna. A few sherds
of this pottery were collected
from the deepest level in
Mallowan's sounding in the
Kiiyiinjik mound at Nineveh
(over 30 m beneath the surface).
(After Lloyd)
41, 42 (Below right) Typical

Samarra painted wares from
Hassuna, including (bottom left)
crude local imitation and (above)
a so-called 'face-urn' similar to
those found in the deepest levels

at Troy. (After Lloyd and Safar,
1945). (Below left) A photograph
of one of the bowls, with a

design showing stylized ibexes
78
The repertory of Samarran designs has been exhaustively studied 41-2

at Baghouz and elsewhere. ^^ Perhaps the most characteristic form
is an open dish with internal border ornament and a centripetal
design, often incorporating human or animal figures. Samarran
designs are painted in black or reddish brown on a thick cream or
pinkish with multiple chevrons and other geometric devices,
slip,
in which the painted lines are wider than the spaces between them.
At Hassuna and the main Samarran sites there are also 'face-urns',
like those found at Troy, with the principal features modelled in
relief. The skill and good taste of the Samarran potters is
emphasized by the ineptitude of crude imitations occasionally

"made locally at sites like Hassuna.
Tell Halaf
The elaboration of painted designs on pottery reaches its peak in the
polychrome wares of the Tell Halaf period. ^^ The colours used are 43-4
in fact mainly restricted to red, reddish or purplish brown and
black; but when applied for instance to an apricot-coloured slip.
43 Polychrome and other

painted wares of the Tell Halaf
period (c. 4800 bc). Patterns in
this 'fully developed' style
include (bottom left) the stylized
'bucranium' motif. The Halafian
culture extendednorthward into
Anatolia and westward through
Syria to the Mediterranean.
(After Mallowan)
79
(h
44 A polychrome dish from

Arpachiyah, of the Tell Halaf
period
the tones vary greatly and their contrasts are emphasized by the
sparing use of white paint. Where the Samarra patterns contrived
to give an impression of movement, those of the Halaf pottery have
been described as 'static' or 'architectonic'. They cover a larger
proportion of the vessel's surface than is the case vîth most other
styles, and there is a good deal of rectilinear panelling, checker
patterns with alternating detail like tile- work, dotted areas and fish-
scale ornament. The development of these designs can be traced
through successive stages, of which only the last is truly
characteristic. One individual motif may be taken to illustrate this
process, namely the bull's head or 'bucranium'. In its earliest,
naturalistic form, it appears isolated and empanelled. Later it is
simplified, turned sideways and repeated to make a running
pattern, which is finally stylized to a point at which its original
significance is forgotten. Among the various forms of Halaf pottery
there are also characteristic shapes, of which the most popular is
perhaps the shallow open dish, decorated inside with border

ornament and a central panel. But there are also flat-bottomed
bowls, carinated bowls with flaring rims and distinctive squat jars.
The distribution of Halaf-type pottery is again interesting. It has
been found as far north as Diyarbakir in Anatolia, and westward
from the Euphrates crossings to the 'Amuq and the Mediterranean
coast at Ras Shamra. Its southern limits seem to correspond
approximately to the latitude of Baghdad. It has no relationship
with contemporary ceramic developments in Iran, except with the
Kermanshah area, and must be considered indigenous in northern
Mesopotamia.
Northern 'Ubaid
The formative stages of the 'Ubaid culture are still something of a
80
mystery; but its attainment of a mature form seems to have

coincided with the earhest unification of upper and lower
Mesopotamia. Whether it evolved in the two areas simultaneously
or (which appears more probable), spread northward from the
direction of southern Mesopotamia, the characteristics by which it
can be recognized archaeologically are as easily distinguishable at
Gawra or Arpachiyah as they were at Ur and Eridu. The earliest
'Ubaid levels at Gawra from one culture
suggest that the transition
to another was a gradual one. The retains some
pottery at first
characteristics of the Halaf period or combines them with- new

conventions. Yet, when the new culture has reached its full
maturity, in -Gawra XIII, the differences between northern and

southern 'Ubaid pottery are no more than could be occasioned by
local disparities in the raw materials available or animproved
understanding of technical processes, such as the firingof clay and
the handling of paint. One particular type of vessel, found in both
settings, is of so eccentric a shape that it must be assumed to
symbolize some short-lived but universal convention. Variously
described as a 'lenticular' or 'tortoise' jar, it is a flat hole-mouthed
affair with trumpet-shaped spout rising almost vertically
a single
from its shoulder. Its stratigraphical contexts in the north and south
respectively are of some significance, since they link the first two
'Ubaid levels at Gawra (XIX-XVIII) with the post-Hajji
Muhammad phases (3 and 4) at Eridu. This could perhaps be taken to
equate Samarra in the north with the 'Eridu' phases in the south.
Post-'Ubaid
Less need perhaps be said about the pottery of the post-' Ubaid
phase, which has logically been taken to correspond with the Uruk
period in the south. Its outstanding characteristics are, first, the
absence of painted ornament, and secondly the fact that, unlike the
'hand-made' pottery of earlier periods, most vessels now showed
tracesof having been turned on a fast-moving wheel. At levels
XI-VIII at Gawra, pottery of brown or buff clay with straw
tempering replaces the fine, greenish-grey fabric of the 'Ubaid
period. The burnished or polished monochrome wares, so
common in the south at this time, are also to be found here and
lightly burnished grey pottery contemporary
predominates at the
site of the Tigris. ^^ As in
called Grai Resh, in the Sinjar district west
the south, however, one of the most distinctive forms of pottery at
these and other northern sites during this period is the 'bevelled-rim
bowl' or glockentopf, a small, very coarsely made vessel, produced in
great numbers and easily broken. In the present writer's opinion, it
was given away free with the food which it contained or was used
for votive offerings.
As a symptom of the great cultural change which can be detected
Tepe Gawra, ceramic innovations are hardly less
in these levels at
conspicuous than the evidence of a sudden interest in metallurgy. A
principal source from which the metal was derived seems to
have been the frequent groupings of richly furnished graves, to
which we must now refer.
81
Burials
The 'Gawra' Period
Where burials are concerned, we have till now mentioned only in
passing those for instance at Tell-es-Sawwan, with their votive
treasure of alabaster vessels or figurines
and the 'Ubaid cemetery at
Arpachiyah which was the source of so much painted pottery. At
other northern sites, the few simple shaft-graves and urn burials
dating from the early periods were of less interest. At Gawra,
however, the sudden multiplication of interments during the final
'Ubaidian occupation (level XII) seemed to anticipate the more
sophisticated funerary practices of the Uruk period. In levels
XI-VIII, burials were of two sorts.
There were no less than 200 simple interments, of which about
80% were infants, buried indiscriminately beneath the floors of
houses or in the vicinity of religious buildings. Some adults were
perhaps buried in a cemetery beyond the limits of the village; but
the more distinguished members of the community were provided
with carefully built tombs, among the buildings on the mound
itself. There were as many as 80 of these structures rectangular :
chambers of stone or mud-brick, roofed with timber or stone slabs.

Each as a rule contained a single body in a contracted position,
apparently fully dressed, though sometimes also covered or
wrapped in reed matting.
Apart from the structure of these tombs, their interest centred on
the wide variety of personal ornaments which they contained.
Beads, in particular, had been used to decorate every part of the
body, so that over 25,000 could be recovered from a single burial.
Listed among the materials of which they were made are turquoise,
jadite, cornelian, hematite, marble, limestone, quartz, obsidian,
lapis lazuli and diorite, as well as shell and ivory, implying trade
with countries as remote as Afghanistan. Perhaps most surprising of
all was the profusion of ornaments in gold. Most of these were
rings, rosettes, small 'studs' and crescentic decorations, cut from

sheet metal and applied to the cloth of garments or diadems but ;
there was one small object, a finial ornament of electrum in the

form of a wolfs head, composed of separate elements skilfully
welded together.
The main significance of these tombs in levels XI to VIII lies first
in the implication they suggest that the mound at Gawra was now
occupied by a people with new skills and increasing sophistication.
Above all, the advance in the practice of metallurgy is most
remarkable. In the 'Ubaid levels beneath, a mere half-dozen metal

objects were found, hammered or cast from pure copper without
the addition of tin. In the post-'Ubaidian levels, metals generally,
including bronze, seem suddenly to be in general use. For the
excavators were able at one point to record that 'the whole ground
seemed to be tinged green with decaying copper or glinting with
gold'. If the Gawra settlement at this period may be thought of as an
unpretentious rural community, the discovery of a site at which the
metropolitan aspect of the same culture could be examined and
compared with the contemporary achievement of Protoliterate
peoples in the south would teach us much about the chronological
and other relationships between them.
82
45 Below: a male figurine in
terracotta from level I at Tell-es-
Sawwan. In this level, upright

alabaster figurines of a quite
different type were also found in
large quantities, with many
bowls of the same material. Ht
6.7 cm. Above: a Samarra-period
terracotta head from Choga
Mami. Ht 4.8 cm
Small Objects
Figurines
Cult objects are comparatively rare among the small fmds at
Chalcolithic sites in northern Iraq. An interesting exception is the
class of female figurines, loosely associated with the term 'mother- Overleaf
goddess'."^^ Unlike the 'lizard-headed', standing figures of the 46 'Eye' or 'spectacle' idols, a
'Ubaid period in the south, these are usually steatopygous creatures, great variety of which were
found in the Jemdet-Nasr-period
seated or squatting, with exaggerated breasts and thighs. In the
temple at Tell Brak. One
earliest examples, for instance from Yarim Tepe, the surviving
particularly large one was
heads have pinched faces on thin necks and are prolonged in a form thought by Mallowan 10 have
which suggests a conical headdress.^' But here also there are stood upon the altar in the main
already traces of the painted ornament characteristic of later times, sanctuary (cf ill. 39). Examples
were also found from the late
when features such as 'trousers' with crossed braces or belts are
'Ubaid period onwards at
represented, and some body ornament which may be tattooing. In Gawra. Their significance is
several beautiful examples from Choga Mami of the Samarra unknown. Ht of tallest, 11 cm
83
84
!
period, heads are modelled and painted in much greater detail,
showing, in addition to the conical hairdress, 'coffee-bean' eyes
with painted eyelashes, and facial ornaments such as nose- or lip-
plugs. "^^ Figurines of this sort of the Halaf period are common to
sites in Syria, such as Chagar Bazar and Tell Halaf itself, ^^ but they
are also found for instance at sites as far apart as Qatal Hiiyiik in
Anatolia and Tepe Sarab in western Iran (Singh 1974, figs. 43 and
63), emphasizing the closer cultural contacts between northern Iraq
and the neighbouring countries from the Neolithic period
onwards. Their precise significance as cult objects has' been
variously interpreted. At Ur and Eridu they were found in graves,
while at Arpachiyah and Tell-es-Sawwan a religious purpose was
tentatively attributed to the buildings in which they were
discovered.
Another of figurine, important for dating purposes, is the
class
symbol, most notably associated with Tell
'spectacle' idol or 'eye'
Brak. Two dozen examples of these, in clay or stone, were found at
Gawra, mostly in levels XI-IX, which should make them
^^?^
contemporary with the earlier building-levels of Mallowan's Eye
Temple.
Seals
Other significant objects, which make an early appearance in the
painted-pottery levels, for instance at Gawra, are personal seals.
Primitive 'seal-pendants' with geometric designs in the Halaf levels ^^^^î^-^

give way to hemispherical stamp-seals, whose pictorial designs
attain an interesting degree of sophistication in the terminal phase
of the 'Ubaidian occupation. A particular group of steatite or Stamp-seals and impressions
47
serpentine seals, attributed to the first post-'Ubaid period (level from the 'Protoliterate' levels at
XI), show figured designs of great interest. Spidery engravings of Tepe Gawra. Other motifs
include hunting dogs, masked
horned animals and hunting-dogs alternate with ritual scenes in
men and human figures in an
which masked men appear. Two figures are sometimes combined attitude of copulation. (From
in an attitude of copulation. ^^^ Speiser, 1935)
Foreign Relations
SUSIANA
It would be out of place, and even impracticable here, to follow or
even summarize the parallel developments and interaction of

Chalcolithic cultures in the neighbouring countries most closely
associated with Mesopotamia. "^^ jn northern Syria, Anatolia and
particularly in Iran - not to mention their remoter extension
towards the Caucasus and Afghanistan - so much has been added to
our knowledge of the period by recent excavations that they have
in a sense usurped the priority of interest earUer accorded to the
great river valleys. Nevertheless, there is one area, beyond the
modern frontier of Iraq though adjacent to the ancient home of the
Sumerians, which cannot escape our attention. This is the Iranian
province known as Khuzistan, and named by archaeologists
'Susiana' after the great city which was its capital in Elamite times.
Long before the beginnings of written history, this country was
closely linked to southern Mesopotamia, of which it can be seen as a
physiological extension. Here, due north of modern Basrah, the
85
48 Two examples of the finely

painted pottery, called by de
Morgan 'Susa I' (now known as
'Susa A'). It is dated to the
transition between the 'Ubaid
and Protoliterate periods, c. 4000
BC. Ht of pot above, 26.8 cm
line of the Zagros Mountains recedes eastward, leaving a wide area

of seasonally fertile uplands, watered by the tangled courses of three
rivers - Karkheh, Sha'ur, Ab-i-Diz - and connected by a fourth, the
Karun, to the Shatt-al-Arab.
The first evidence of habitation here in prehistoric times was
provided in 1 897 by French excavations in the great city-mound at
Susa itself. A huge cemetery, adjoining the later town-wall,
produced much of the marvellous painted pottery now in the
Louvre, which thus anticipated by almost twenty years Woolley's
discovery of the 'Ubaidian culture in Mesopotamia. Most striking
of all was a class of beakers and tall goblets, painted with stylized
patterns of birds and animals, which the excavator, Jacques de
Morgan, designated by the term 'Susa I' and which are still
48 considered by some to be the highest accomplishment of ceramic
86
craftsmanship in the Near East.^^ With no comparative material

yet available, the age of this pottery was at first considerably
overestimated. In fact, a further fifty years vêre to elapse before its
correct placing in the sequence of prehistoric periods could be

properly understood. During that time, many new discoveries in
Iraq and a dozen excavations at smaller sites in Khuzistan itself
(which have been more recently multiplied) furnished the material
for a comparative study. This was undertaken by the late L. Le
Breton and published in 1957."^^ It was now proved that de
Morgan's 'Susa I' (in future to be called 'Susa A') belonged to the
last of five phases into which the Iranian painted pottery cultures
were divided ('Susiana a, b, c, d and e'), and that it corresponded

chronologically with the transition from the 'Ubaid to the Uruk
period in Mesopotamia. For the previous four periods ('Susiana
a-d'), convincing parallels could be found with the whole sequence
of developments in Iraq, from Hassuna and Eridu onwards. Le
Breton was able to recognize three further phases ('Susa B, C and
D') corresponding to the Mesopotamian Protoliterate. To 'Susa C
he was able to attribute the so-called 'Proto-Elamite' pictographic
writing, found at Susa itself, where its use would have been a logical
result of the close ties between Susiana and Iraq in the Uruk and
Jemdet Nasr periods. Since it has now been found at such remote
sites as Tepe Yahya, in the Soghum valley 140 miles south of
Kirman, and at Sialk, Godin, Malyan and Choga Mish, it must

have been widely employed in Iran at that time."^^
87
Chapter Five
The Early Sumerian Dynasties
The Protoliterate period must be regarded, not as a prelude to

Sumerian civilization, but as a first formative phase in its
II
development, during which major contributions were made to the
establishment of its identity. All that we know about it has been
learntfrom the results of excavations; but once it is ended, we fmd
ourselves on the threshold of written history, and from now
onwards must check our archaeological conclusions against the
testimony of contemporary documents. The phase with which we
shall next be dealing is generally referred to as the 'Early Dynastic'
period: a title which is explained by the contents of a single,
primarily important written document. This is the so-called 'king-
4g in which the political anatomy of the Sumerian com-
list',
monwealth and the succession of its rulers were at some time

committed in writing to the memory of Mesopotamian pos-
terity. Several slightly varying versions of this text have survived,
"^"^
and among them is one which can be dated (partly by the point
which it stops), to the beginning of the 1 8th century b c. The earlier
part of the same list even reappears in Greek, among the writings of
Berossus, an obscure historian of the Hellenistic age.
Before speaking further about this document, in which the
concept of 'kingship' is at least implicit, it may be well to enlarge a
little upon the titles variously attributed to the rulers of Sumerian
city-states during successive phases of their political evolution. In

the early inscriptions, a number of different terms are used to imply
forms of state leadership which are not precisely defined. Among
these are en ('lord') and ensi (steward?), lugal ('great man' or
'king'), and the distmction between them has been the subject ot
much thought and some controversy. The fluctuating conclusions
of authorities on the subject are usefully summarized by Saggs and
Gadd, both writing in 1962. Saggs recollects the interesting
proposal presented byjacobsen in 1943. He reached the conclusion
that 'Sumerian kingship was not primitive but evolutionary and
that its action was, at least sometimes, controlled by an assembly of
elders and community heads or even by the mass of free men'. This
theory was based on a study of early literary sources: the '. . .
myths, the tales of the behaviour and exploits of the gods, which are
generally thought to reflect the sociology of the times at which
Sumerian society crystallized in the city-state .', probably very
. .
early in the 3rd millennium bc. According to Jacobsen writing in

1943, one function of the 'general assembly' was to choose an
official called En, who was 'primarily a cult-functionary, being the
consort of the city's patron deity, and playing a vital role in the
Sacred Marriage upon which the fertiHty of the city state 49 Part of the 'king-list'
depended'. His secondary administrative functions, in connection document, found at Khorsabad

by Loud. It is a late Assyrian
with the temple lands, gave him great political importance, and at
copy of an earlier text
one stage he became virtually the ruler. Very early, there came
about some division between the priestly and secular functions of
the En, who no longer lived in the temple but in a pretentious
palace of his own. Later, during the historical period, the actual
ruler
was no longer the En, but an official concerned in the

originally
agricultural operations, who bore the title Cultic duties were
Ensi.
delegated to a special priest or priestess. ... In the event of an attack from
the outside, the assembly had to choose a war-leader or king {Lugal). . . .
The of neither En nor Lugal was originally hereditary or permanent,

office
that of the Lugal at least being granted only for the duration of the
emergency.
The En or Lugal might attempt to perpetuate his position of

authority and the two functions might then be vested in the same
man. He could not be an absolute ruler, and could act only as
authorized by the assembly.
Jacobsen's view of Sumerian society and its political structure,
which aroused much interest when originally presented, is today less
generally accepted, owing to a lack of further confirmation by new
historical texts. The bare facts of what we are justified in
concluding about the status of Sumerian rulers are that lugal
remained the general designation of a 'king' and that an ensi,
89
'
though a governor
in another city.
in his own right, might be subordinate to a lugal 1
The King-list
Here are some of the king-hst as we know it today. It

features
presents us with the names of eight semi-legendary rulers
first
'before the Flood', and of the cities with which they are believed to
have been associated. 'After the Flood', we are told, 'Kingship was
sent down from on high'. A more
factual chronicle then follows,
introducing us to of city-states, only one of which at
a federation
any given time held supremacy over all the others. A succession of
dynasties is accordingly Hsted, each based upon an individual city,
whose hereditary rulers on such occasions became 'Kings' of 'the
Land' and accepted responsibility for its welfare. Occasionally
genealogical information is given, or mention is made of some
happening which explains the transfer of hegemony from one state
to another. This secondary list, as has often been pointed out, 'was
not composed for the benefit of modern scholars', and from their
point of view has manifest defects. The stated lengths of individual
reigns are, for instance, fanciful and it has long been realized that
certain dynasties in fact ruled concurrently in their own cities. Yet it
does serve to emphasize the Sumerians' overall conception of the
Land as an entity. Also it embodies for the first time a catalogue of
the principal Sumerian cities - Sippar, Shuruppak, Kish, Ur, Adab,
Mari, Akshak, Lagash, Isin, Larsa and others about which less is
known.
The king-list does not of course come under the heading of what
we have 'contemporary documents'. It is rather a re-
called
trospective record, assembled at a later date from traditional
information. The same is true, for instance, of the great Sumerian
epics, like the story of Gilgamesh, which have survived through
repeated recopyings in the Akkadian and other languages of later
ages.^°We should now therefore glance at the actual historical
information gleaned from the increasingly articulate writings of
early Sumerian scribes, and note the extent to which their
prolonged study in more recent times has served to authenticate
some parts of these Mesopotamian classics. ^
The Earliest Written Texts
The earliest pictographic writing found in the archaic levels at

Uruk has been mentioned in a previous chapter, where it was noted
that, even if the language of the 'Uruk 4' tablets could be
considered as inconclusively identified, that of the 'Uruk
(Jemdet 3'
Nasr) texts was demonstrably Sumerian. Following these chro-

nologically in the 'archaic' series were, first, a group of tablets
found by Woolley during the 1920s, in a stratum preceding his
'Royal Tombs' at Ur, and secondly, inscriptions discovered by the
German excavators, W. Andrae and R. Koldewey, at Farah
(ancient Shuruppak), at the beginning of the present century. ^^ At
this point, however, we must recall that, practically without
exception, the subject of all the writings so far mentioned are no
90
more than and quantities of commodities or persons. The first

lists
instances of individuals (usually kings) adopting the use of writing

for the purpose of recording such things as religious dedications or
personal accomplishments are contemporary with or slightly later
than the Farah group mentioned above, though they were found
more recently, by Woolley at Ur. It is therefore true to say that,
until late in the second decade of the present century, the royal
names in the king-list, including even that of Gilgamesh himself,
could have been considered creations of the Sumerian imagination.
It is accordingly understandable that something of a sensation was
caused by Woolley 's discovery in 191 9, among the ruins of an Early

Dynastic temple at Al 'Ubaid, of a foundation tablet bearing the
names of Mesannipadda, first king of the first Dynasty of Ur, and of
his son, A'annipadda.^3
By that time, methodical excavations by German, French and
American archaeologists were in progress at the sites of other
Sumerian cities mentioned in the king-list, including some from
which large quantities of tablets had been unsystematically
extracted by earlier excavators. In the decades which followed,
therefore, many new Sumerian texts were discovered, not only by
the diggers themselves but by philological scholars who were now
at work studying museum collections. ^"^ In this way, further royal
names were added to the list of kings who now proved to have been
genuine historical characters. Before discussing these further, it will
be necessary to anticipate the archaeological record, by citing
briefly the broad conclusions which have come to be accepted
regarding the chronology of the Early Dynastic period.
Archaeological Phases
This so-called Pre-Sargonid era (preceding the unification of

Mesopotamia under Sargon of Akkad), has come conventionally to
be divided into three phases. 'Early Dynastic I' (ED I), following
directly upon the end of the Protoliterate, is approximately dated
to the years between 2900 and 2750 bc; 'Early Dynastic 11' (ED II)
lasted until 2650 bc; while 'Early Dynastic III' (ED III), divided
into two sub-phases, 'a' and 'b', is taken to account for the greater
part of three further centuries. This system of chronology was
constructed largely from evidence obtained in the 1930s during
excavations by the Oriental Institute of the University of Chicago
at sites in the Diyala region, east of Baghdad. ^^ j^ -^vas based, as we
shall presently see, on progressive variations in the architecture,

sculpture, pottery, seal-cylinders and other small objects associated
with several temples, founded in most cases at the end of the
Protoliterate period and repeatedly rebuilt in Early Dynastic times.
Its validity has been confirmed, with only minor reservations, by
subsequent soundings of the same sort at Nippur and elsewhere (see

below).
The Flood
The task of adapting this archaeological hypothesis to the

chronological implications of the Sumerian texts has on the whole
proved less difficult than one would have imagined. One initial
91
problem concerned the date and significance of 'the Flood', which

figures so prominently in Sumerian tradition and whose memory
has indeed been bequeathed to ourselves through the medium of
Hebrew scriptures. The archaeological evidence in this connection
was unfortunately extremely equivocal. Great floods were a
commonplace of Mesopotamian history until quite recent times;
and it was therefore less than surprising to find that, in deep
soundings ajt relevant Sumerian sites, clean strata of water-borne

sand or clay appeared in stratigraphical contexts which varied in
time from the 'Ubaid period at Ur to the end of the Early Dynastic
phase at Kish.^^ At Farah (Shuruppak), however, a stratum of this
sort occurs at the end of Early Dynastic I, and in this single case it
could, as we shall now see, be cited (without much conviction) as

supporting evidence for an inference from the Sumerian textual
evidence.
Where the king-list is concerned, there can no longer be any
doubt that the semi-historical epoch, represented by a succession of
'rulers before the Flood', must be equated with the archaeological
series defined as 'Early Dynastic I'. The individual names of these
'rulers' are of little interest, since only the last to be listed has any
historical significance. And here there is a connection of some
importance with an episode in the Epic of Gilgamesh, when its hero
made a journey to consult Utnapishtim, the Babylonian Noah,
about the secret of eternal life. It had always been a matter of some
surprise that this individual should receive no mention in the king-
list, the last name before 'the Flood' appeared as Ubartutu, a
where
ruler of Shuruppak. In another version of the Deluge story,
however, its hero is given the alternative name, Ziusudra, and a
surviving fragment of the text makes it clear that he was the son of
Ubartutu. ^7 The implication here that Ziusudra and Gilgamesh
were contemporaries is unfortunately refuted by the fact that the
former was deified after the Flood, and was accordingly already a
god when Gilgamesh met him.
On the subject of Gilgamesh himself, other sources of textual
evidence were to prove more rewarding. One of these was an
important text of which versions were found both at Ur and at
Nippur, giving the names of kings who had piously repaired the
structure of a shrine called 'Tummal', not yet located, at the latter
city. These included the names of three kings, Agga, Gilgamesh
and Mesannipadda, belonging respectively to the first dynasties of
Kish, Erech (Uruk) and Ur. ^^ The king-list, as we have said, would
have us believe that these three dynasties followed one another in
chronological succession. In this case, however, such a claim is
clearly refuted by other evidence, which proves the three kings
mentioned have been much more nearly contemporary. We
to
know for instance from a Sumerian epic that Gilgamesh of Erech
fought against Agga of Kish, but also that it was Mesannipadda of
Ur who put an end to the Kish dynasty. As for the latter king, his
name appears not only on the foundation tablet from Al 'Ubaid,
but on seal-impressions found in the Royal Cemetery at Ur. Armed
with this and other evidence of the same sort, we are safe in dating
all three kings to the third phase of the Early Dynastic period,
perhaps between 2650 and 2550 bc.
92
Between the Flood and Gilgamesh, we are now left with a period
of time (computed by Mallowan at about lOO years), correspond-
ing archaeologically with Early Dynastic II. Many of the kings'
names allotted to this period are either Semitic intrusions or
recognizable divinities; but two of them have proved historical.
The first of these is Enmebaragisi, father to Agga of Kish, whose
name has been significantly found in an Early Dynastic II setting at
one of the Diyala sites. ^ ^ The other is Enmerkar of Erech, ^ ° subject
of a very early Sumerian epic, who appears in the Greek version of
the king-list as grandfather of Gilgamesh. With these two figures
now authenticated, our line of enquiry is rapidly approaching its
terminus post quern in the person of Ziusudra, the -Babylonian Noah.
Early Dynastic Sites
Khafaje
We must now return to the archaeological record and enumerate
some of the Sumerian sites which have contributed to our now
very extensive knowledge of the Early Dynastic period. For this
purpose, it may be well to start with the so-called 'Diyala' sites, if
only because the finds there played so large a part in the

establishment of a chronological structure on which subsequent
calculations could be based. And in this area, priority should
perhaps be accorded to the site called Khafaje (ancient Tutub), to
which attention was drawn in the years preceding 1929 by the great
quantity of fragmentary Sumerian sculpture which illicit digging
was bringing into the hands of Baghdad dealers. ^^ The source of
these antiquities was a mound near the east bank of the River
Diyala, some 15 miles north of its confluence with the Tigris, and a
single season's excavation by a team from the Oriental Institute of
the University of Chicago revealed the presence, directly beneath
its summit, of at least two Sumerian temples, one of which,
occupying a central position, was dedicated to the moon-god. Sin.

During the years that followed, numerous rebuildings of the shrine
were carefully traced, through levels representing every phase of
the Early Dynastic period, down to an original foundation in
Jemdet Nasr times.
The Sin Temple at Khafaje, like those of the Protoliterate levels
in the Eanna Precinct Warka, belongs to the category of
at
'ground-level' temples. That is to say, it was not at any time raised
upon an artificial platform, but occupied a site whose shape, if not
its size, was dictated by the layout of residential buildings around it.
Following the already age-old tradition of temple building, its basic

element took the form of a long rectangular sanctuary, with a
podium altar at one end and an entrance on the cross-axis at the
other. To this, in the case of the earliest Sin Temple, lateral 50 Projected plan of Sin Temple
chambers were added on either side, creating a conventional VIII at Khafaje dating from the
tripartite plan. Little was added to this simple structure until its fifth Early Dynastic II period {c.
rebuilding at the beginning of the Early Dynastic period (Temple 2750-2650 Bc). Its plan adapted
IS
to the irregular shape of the site

VI), when the shapeless forecourt through which its worshippers
originally available when
approached was extended and converted into a walled enclosure, Temple I was Jemdet
built in the
with a formal gatehouse and various outbuildings. Four further Nasr period. (After Delougaz)
93
1
rebuildings during later periods left the asymmetrical shape of the

complex unaltered. Only in level IX was the outer entry given
greater dignity and the courtyard provided with an open-air shrine.
P. Delougaz, who was in charge of the Khafaje excavations,
found the whole summit of the mound honeycombed with the pits
and tunnels of illicit diggers. One whole season had to be spent in
removing the loose earth from them and in tracing the walls of
which fragments had survived in undisturbed areas. In this way, his
skill as an excavator enabled him to reconstruct the plans of the two
latest Sin temples (IX and X) and to recover such objects as the
looters had missed. The additional collection of votive statuary

from this source, though impressive in itself, was overshadowed by
the temple fittings, ritual vessels and small objects found
undisturbed in the deeper levels. This valuable assemblage of finds
was also supplemented by the contents of three smaller temples,
found among the dwelling-houses to the west and south of the site,
which had escaped the attention of the looters. One was a large
shapeless enclosure containing two separate shrines, dedicated, as
Delougaz understood from an inscription, to the mother-goddess,
Nintu. There were also two smaller, single-shrine temples.
Since all these buildings appeared to be 'ground-level' temples, it
was perhaps less surprising that, in the course of the first season's
work, a wide area on the lower slope of the mound to the west was
found to be covered by the scanty remains of a vast oval-shaped
temenos which had once provided a setting for a typical 'high
51-3 temple' raised on its artificial platform. The excavation of the 'Oval
Temple' at Khafaje presented Delougaz with a technical problem
which would have daunted many excavators. The area enclosed by
its double line of outer walls had a maximum dimension of almost
exactly 100 m; but neither the outer walls themselves nor those of
the buildings which they enclosed remained standing more than a
few brick-courses high. Delougaz was compelled to adopt a
method of excavation earlier perfected by the Germans at Warka,
53 which involved articulating every individual mud brick and
thereby recreating the pattern of walls which they composed. After
much labour in training Arab workmen to do this, the whole
ground-plan of the complex was eventually recovered.
The plans and reconstructions show three stages of building in
the second and third phases of the Early Dynastic period. First,
there is an almost symmetrical oval enceinte, around which
outbuildings are arranged to form a rectangular inner courtyard.
5 Reconstruction of the Oval
Half filling this space, the outline only could be traced of the temple
Temple at Khafaje, as rebuilt in
the Early Dynastic III period {c.
platform, with a single stairway leading to it. To the first enclosure
2650-2350 BC). Only the wall a second is added, distorted somewhat in shape to allow room
foundations of the temple for an outer courtyard and a roomy dwelling-house, perhaps for a
platform and its oval enclosure
priest. In the second building phase (ED Ilia), the outer wall
walls had survived the ;
appearance of the raised shrine increased in and was strengthened with external
thickness
itself is conjectural. (After buttresses, while in ED lllb a new and pretentious portal was
Darby) added, occupying most of the outer courtyard: A section cut by
Delougaz through the two enclosure walls showed their offset
52 Air photograph of the Oval
foundations beneath pavement level; but it also revealed a
Temple, from the south,
showîng a later tow^n vâll and phenomenon which has rare parallels in other periods of temple
buildings in the background building. Before the foundations were laid the entire area of the
94
^R ^
- '
95
IM'
53 Excavating the foundations

of the Oval Temple enclosure at
Khafaje (looking northwest).
The individual mud-bricks (of
the 'plano-convex' type) had
laboriously to be articulated. The
excavators in the foreground
have discovered the deposit of
sand on which the foundations
rested
temenos had been excavated to a depth of 4.6 m

and then filled
with clean sand, brought from some source outside the city.
Delougaz estimated that 64,000 cubic metres of sand had been
moved for the purpose, and assumed that ritual conventions could
alone account for so great a labour.
Complementary to the fmds from these temples at Khafaje vêre
the contents of graves, almost 200 of which were found in the town
area, mostly beneath the floors of dwelling-houses.^^ They varied
from simple shaft burials to walled tombs, at least two of which
were built of kiln-baked brick and covered with corbelled vaulting.
The grave furniture consisted mainly of pottery vessels, whose
great number and variety contributed effectively, as we shall later
see, to the final analysis of Early Dynastic stratigraphy.
Tell Asmar
In post-Sumerian times, the city whose remains were found at
Khafaje became part of a politically important state called
Eshnunna, the capital of which was at Tell Asmar, 50 miles
northeast of modern Baghdad. Here too there had been a city in
Early Dynastic times, and a small temple was excavated by the
Oriental Institute of the University of Chicago concurrently with
96
; ;
The 'Abu Temple' (as it came to be known after

those at Khafaje. ^ ^
the discovery in of a statue bearing the 'plant' insignia of that god,
it
elsewhere described as 'Lord of Vegetation') was founded, like the

Sin Temple, in the Jemdet Nasr period, and its architectural history
could again be traced up to a final rebuilding early in Akkadian
times. From a small and shapeless 'chapel', its plan developed first
on conventional lines, into a building with a rectangular sanctuary

and one row of lateral chambers. This 'Archaic Temple'
corresponded in time to the first phase of the Early Dynastic period.
In the second, it was replaced by a 'Square Temple',' with 54
rectangular chambers, three of which were miniature sanctuaries,
arranged around a central court. In the third phase, the plan
reverted to a 'Single Shrine Temple', similar to the smallest of the
series at Khafaje.
The finds in these buildings at Tell Asmar were plentiful, and of a
sort with which the Diyala excavators soon became familiar beads, :
carved amulets, seals and even

bronze mirror, in addition to stone
a
vessels and a wide variety of pottery. Where sculpture was
concerned, a striking find was made in a sanctuary of the Square
Temple. Carefully buried beneath the pavement beside the altar
was a cache of twenty-one stone votive statues, remarkably well
preserved. Furthermore, these were distinguished by a formalized
style of carving, perhaps characteristic of the Early Dynastic II
54 Projected plan of the Square

Temple at Tell Asmar (Early
Dynastic II period). A cache of
21 votive statues were found
buried beside the altar in shrine
II. The building differs httle
from plans of contemporary
dwelling-houses. (After
Delougaz and Lloyd, 1942). a,
priests' room ; b. Shrine I ; c,
hearth ; d, Shrine II ; e, entrance

/, ablution room ; g. Shrine III
h, courtyard
10m
97
55 A view of the Shara Temple phase, which contrasted interestingly with the more naturahstic
(Early Dynastic II period) after
Early Dynastic III sculpture from Khafaje.
excavation. The main sanctuary,
with its altar and offering-tables,
can be seen in the centre. Secular Tell Agrab
buildings of later times The third of the Diyala sites, and almost the last to be excavated,
accumulated around the once once more dramatically illustrated the capabilities of these
consecrated site, when the
Sumerian temple-builders. This was Tell Agrab, a large city-
temple itself had fallen into ruins
at the end of the Early Dynastic mound far out in the now-empty alluvial desert (chol), 1 5 miles east
period of Tell Asmar.^* The temple here, whose walls appeared directly
beneath the surface, was again square in plan but measured no less
than 60 m from side to side. Part of the building nearest to the city-
wall had been denuded by rainwater, but the surviving half
55 contained not only the impressive main sanctuary, but two
subsidiary shrine-chambers and living accommodation for priests;
their walls survived in some cases up to 2 m
high. The most prolific
source of removable objects was the sanctuary itself and a small
'sacristy' chamber adjoining its high altar. As had been the case in
the Abu Temple at Tell Asmar, discarded or damaged cult-objects
and votive offerings had been buried beneath pavement level, or
even built into the structure of the altar itself. In the sacristy alone,
many weeks were spent in recovering delicate products of
Sumerian craftsmanship, deposited in layers and covered with hard
earth. Among many thousands of beads and small objects,
including carved amulets and seals, were unique art-works in
sculptured stone or bronze and some hundreds of stone maceheads,
perhaps used for processional purposes. More will be said later
about these finds in dealing with the categories to which they
belong. As for the stratigraphy of this temple, apparently dedicated
to Shara, the patron god of Umma, the bulk of the surviving ruins
could be dated to the second Early Dynastic phase; but small
98
sections of an earlier building (Early Dynastic I) were excavated

beneath it, while finds on the surface suggested a later occupation in
Early Dynastic III.
Ur-of-the-Chaldees
In 1930, when work on the Diyala sites began, Woolley's
excavation Ur-of-the-Chaldees had already been in progress for
at
more than ten years and his spectacular finds there had provided a
great volume of comparative material for the benefit of other
workers in the same field. ^^ From prehistoric times onwards, Ur
had been a great cultural and religious centre of the Sumerian
people, and it continued to be so, long after its political importance
had diminished.
At the site, whose modern name is Muqayyar, Woolley found
himself dealing with a walled city, roughly oval in shape and
having a maximum dimension of over ^ mile.^^ It was surrounded 35
on its north and west sides by an ancient bed of the Euphrates and had
been served by two harbours for shipping. Near to the surface in the
northwestern quarter were the walls of a very extensive religious
precinct, built by Nebuchadnezzar in the 6th century b c to enclose its
temples, temple-palaces and subsidiary buildings, most of which had
been piously maintained and frequently rebuilt by a long succession
of Mesopotamian kings. It is understandable, therefore, that little 37
remained on the surface of their original foundations, which were by
now for the most part deeply buried beneath the structural accretions
oflater times. This applied equally to the city's great ziggurat, built in
part by kings of the Third Dynasty, lated in the 3rd millennium bc.
Woolley satisfied himself that, somewhere enclosed within its later
fabric, were the remains of a much earlier tower - a modest ziggurat
rather than a mere temple platform - and, deeply buried beneath the
later pavements on either side of it, he found the denuded walls of
other Early Dynastic buildings, clearly associated with it.
But was in another part of the site that Woolley made the
it
sequence of discoveries for which this Sumerian city became most

famous. In the time of Nebuchadnezzar, the sacred temenos had
been considerably extended at its southeastern end and the
foundations of its new enclosure wall had penetrated into an
important burial-ground of earlier times. It was here that Woolley
made a deep sounding and was rewarded by the discovery of a
cemetery, dating from the Early Dynastic period. Among many
hundreds of more modest burials he encountered a group of 'Royal
Tombs', whose accompanying display of contemporary riches
astonished the world.
Woolley found altogether sixteen of these tombs, each
distinguishedby the construction in their shafts of stone-built
chambers, sometimes with more than one compartment. They
were roofed with corbelled vaulting of stone or brick and, in at least
one case, the construction of a dome had been attempted. A more
significant characteristic was the evidence which they provided of
ritual, involving some sort of human sacrifice.
an elaborate funeral
The major tombs were approached from the surface by a sloping
ramp, and itwas at the foot of this, beside the tomb-chamber, that
the bodies were found of soldiers and female attendants, as well as
99
Schematic plan
of Ur
emple
a>Fortress
Houses *
Ziggur (Urilland w
later) V
S»iA>>
\V~'
D\A/elling quarter
(Urlllandlate||^
arter^
Dwelling quartefj
56 The of Ur at the time of
city (Ur III and
its Third Dynasty and later. Ur- \\ Western later
V (Harbour
nammu's city-wall was
Temple"
destroyed by the Elamites in of /
;
2006 BC. Woolley found it to Enki
have been over 27 m

in thickness Late B abylonianquarter\<
U'''
and traced the remains of two
harbours accessible from the
y i»ii\
Euphrates. The inner Temenos
Wall was rebuilt by ' !''
Nebuchadnezzar of Babylon "^•-^ Outer //''/
fifteen hundred years later. ^V^""-- City wallv''?.'
(From Hawkes, 1974)
wheeled vehicles with their draft-animals, which had formed part

of the funeral cortege. All these were disposed in an orderly manner
in the shaft, while some privileged individuals shared the burial
chamber with its principal occupant. The personnages whose
itself
deaths had been the occasion for so much ritual lay as a rule on a
wooden bier, surrounded by a great wealth of personal possessions.
The number and variety of these - ornaments, weapons, musical
instruments and other treasures - create an inventory of beautiful
objects, whose archaeological interest even exceeds their intrinsic
value. Many of them have been made familiar to a wide public
through the medium of book illustration and museum display; but
the aptitude of their design and high standard of craftsmanship
remain one of the great marvels of antiquity.
There is space here only to recollect some examples of the more
characteristic burials, and for this purpose one must use the
catalogue numbers by which Woolley distinguished them.
The burial whose contents are perhaps best known of all is that of
'Queen Shubad' (no. 800B), whose name, inscribed on her lapis-
lazuH cylinder-seal, is now read as Pu-abi.
Unlike some others, this
grave had escaped the attention of tomb-robbers, so that both its
stone chamber and shaft, with the retinue of attendants, remained
intact. In the dromos (an extension of the shaft) were 5 soldiers, a
wagon drawn by 2 oxen and 10 court ladies, one of whom was a
harpist."^"^
In the tomb-chamber, the queen was accompanied by
two companions. She wore splendid jewellery, including the
100
57 The great Temenos or sacred

enclosure at Ur dating from the
time of the Third Dynasty
(2 1 1 3-2006 Bc), showing some
later features. (From Hawkes,
1974)
elaborate headdress of gold and semi-precious stones of which

reconstructions are to be seen in the British Museum and elsewhere.
Around her were vessels of gold and silver, a harp decorated with
a cow's head, an inlaid gaming-table, an electrum rein-ring
surmounted by the figure of a wild ass, and 267 other objects of
great value. A chest containing the queen's clothes concealed a hole
in the floor, through which the workmen preparing her tomb had
penetrated into and partially looted another chamber beneath. This
Woolley assumed to have been the 'King's Tomb', since again its 58
own shaft and dromos were occupied by appropriate sacrifices,
formally disposed. There were in all 59 bodies, including 6 soldiers
who had led the cortege, 2 chariots drawn by 6 oxen, 19 court ladies
lOI
Mag. N
58 The so-called 'King's Tomb', in gold headdresses and, elsewhere in the shaft, a lyre with an inlaid
no. 789, in the Early Dynastic
sounding-box and the head of a bull in gold and lapis. Despite the
cemetery at Ur, found by
hurried looting of the chamber itself, a silver model of a boat and a
Woolley beneath tomb 800 (ill.
59). Against the tomb wall is a shell-inlaid gaming-table remained in place.

row of court ladies and near them Another grave whose occupant was identified by his name on a
guards with spears. (S. Ebrahim cylinder-seal was that of Akalamdug (no. 1050). He was
after Woolley, 1935). a, door; b,
accompanied by the bodies of 40 attendants and from the shaft
dromos ; c, wagons
came two of the fmest ceremonial daggers, one with a lapis-lazuli
59 The tomb of 'Queen Shubad' haft and granulated gold ornament. Another tomb was that of
(or 'Pu-abi' as she is now known), Meskalamdug (no. 755), owner of the famous golden 'wig-
no. 800, built over the 'King's
helmet', now in the Iraq Museum, whose name was also inscribed
Tomb', no. 789 (ill. 58).
(S. Ebrahim after Woolley,
on a gold lamp. But the expenditure of human lives seemed to
1935), ^, tomb 800B; b, chest; c, reach its climax in the great, anonymous shaft-grave (no. 1237),
chariot; d, dromos; e, pit;/, lyre which contained 74 bodies - 68 of them women in full regalia, some
with lyres. Mallowan, who was present when this burial was
exposed, has described the impression which it created: '. the
. .
ghastly of human sacrifice, a crowd of skeletons so

scene
gorgeously bedecked that they seemed to be lying on a golden
carpet'. Major treasures from this source included the two gold and
lapis he-goats, rampant against bronze foliage, which reminded
02
Woolley of the biblical phrase, 'a ram caught by its horns'. Like the
magnificent 'royal standard' with its inlaid scenes of 'war and
peace', found in a plundered tomb elsewhere (no. 779), the purpose
of these too was uncertain.
In the years since these discoveries were made, some authorities
have questioned Woolley's use of the term 'Royal Tombs',
suggesting an alternative identification of their occupants as
participants in some fertility cult-practice.^^ Today, however, the
evidence in favour of the excavator's own conclusion has come to be
generally accepted. On the seal-cyhnder of Akalamdug he is given
the specific title, 'King of Ur', while that of Meskalamdug similarly
designates him as lugal (king). As for the practice of human sacrifice,
in a passage from the Epic of Gilgamesh a hero is 'accompanied in
death by some of his retainers'. But Mallowan and others have also
pointed out that at least half-a-dozen of these 'royalties' appear to
be members of a single 'Kalam' family, and that they may well
represent a 'dynasty', preceding that associated by the king-list with
Mesannipadda and his successors. In the disturbed strata overlying
Woolley's cemetery, inscriptions have been found, naming these
kings of the canonical 'First Dynasty of Ur', and the circumstances
suggest that their own tombs may have been more effectively
plundered than those beneath. This being the case, the two
dynasties have been respectively associated with the archaeological
phases Early Dynastic Ilia and Illb.
Al 'Ubaid
At this point something should be said about Woolley's earlier
(1922) excavation at the small neighbouring site of Al 'Ubaid,
where H. R. Hall had already discovered a rich deposit of Sumerian
antiquities.^^ Here, the prehistoric settlement to which we have
alluded in a previous chapter seems to have been abandoned at the
beginning of the Protoliterate period. In Early Dynastic times,
however, a temple had been built nearby, probably on the site of a
much earlier religious shrine, and dedicated by King A'annipadda
to the goddess Nin-khursag, mother and wife of Enlil. Its platform,
whose facades were faced with kiln-baked brick, had survived
almost intact, and it was in the angle between this and the stairway
approaching its summit that Hall had made his most striking find.
At some time the temple itself had been dismantled and destroyed,
but much of its architectural ornament had been removed and was
found still lying where it had fallen or been deposited. A great
bronze lintel, with the projecting figure of a lion-headed eagle
between two stags, may have decorated the main doorway and a
pair of columns, encrusted with coloured inlay, perhaps helped to
support it. The wall-faces nearby had also been ornamented with
friezes of animals, modelled in high relief and sheathed in copper,
while others, inlaid with limestone, depicted formal scenes similar
to those on cylinder-seals. The actual form of the building to which
these and other decorations had been applied unfortunately remains
obscure.
In the mid-i930s, when the remnants of a contemporary temple
platform were found at Khafaje, Delougaz suspected a close parallel
with its counterpart at Al 'Ubaid. Further soundings which he was
103
104
then able to make in the vicinity of the Nin-khursag temple 60 ThesiteofTell-ArUbaid.
Plan showing surviving platform
confirmed the fact that it too had once been surrounded by an oval-
°°
of an Early Dynastic temple in
shaped enclosure wall. '
an oval enclosure. It is built on
the outskirts of a prehistoric
KiSH settlement, dating from a
The only examples of temple platforms (described by their thousand years earlier,which the
Sumerians used as a graveyard.
excavators as 'ziggurats'), dating from the Early Dynastic period,
(From Delougaz, 1938)
are those at the site of Kish. It will be remembered that this city, first
in the series capitals mentioned in the king-list, was held
of dynastic 61 The great hntel, now in the
in great respect by the Sumerians; and indeed, the eponymous title BritishMuseum, from the
'King of Kish' was afterwards sometimes assumed by rulers of Temple of Nin-khursag at Al
'Ubaid. Found with other
other cities which attained the hegemony of the Land (Sumer).
architectural ornament, displaced
Today, its site is represented by a wide grouping of mounds, once and lying at the base of the
more bordering an ancient bed of the Euphrates, 9 miles east of platform, it is composed of
Babylon. In the course of the city's history, its principal buildings copper sheeting over a wooden
core and represents Imdugud, the
were alternately concentrated in areas over 2 miles apart which
Uon-headed eagle with two stags
are today known as Uhaimir and Ingharra (ancient Harsagkalama). (c. 2500 Bc). Ht 107 cm
Both were excavated in the 1920s and 1930s by English or French

archaeologists. '
° ^
The earliest finds were made by E. Mackay in an outlying

mound of the Ingharra group, afterwards known as Area 'A'. Here
there was a cemetery, with graves mostly dating from the final
phase of the Early Dynastic period; but, in clearing them, Mackay
found that the earlier burials penetrated into the ruins of an
extensive and important building belonging to a slightly earlier
period. This was the so-called 'Palace A', the first example to be
found of a monumental building devoted to other than religious
purposes in the early days of the Sumerian dynasties. With its
towered portico, buttressed outer walls and columned chambers, it
once more emphasizes the dignity of contemporary architecture.
The excavation which was next undertaken in the main mound
at Ingharra was in the end extremely productive, though it suffered
from some deficiency in archaeological method, which served to

detract from the clarity of its subsequent publication. Many years
later, however, its stratigraphy was more effectively elucidated
by a modern scholar, P. R. S. Moorey, after a painstaking re-
105
examination of the actual finds in relation to the contents of

surviving field notebooks. 'o- His conclusions were broadly as
follow^s.
At the summit of the mound vêre two Early Dynastic ziggurats
of different sizes. The smaller of these had been partly demolished
to provide an emplacement for a Neo-Babylonian temple; but the
excavators were able to trace the brick facâde of an artificial terrace
upon which it had stood. In doing so, they located its original
foundation, at a level demonstrably contemporary with that of the
palace in Area A (ED Ilia) ; but they could also distinguish a later
building-level, from the first by a deep deposit of
separated
discarded brick debris, into which graves had been dug - some in
the subsequent phase (ED Illb) and others in Akkadian times. These
burials could therefore be equated with the cemetery in Area A.
The original ziggurat terrace was founded directly upon an easily
recognizable 'flood stratum' which covered the whole site, and
beneath this an important sounding was made. It revealed part of a
residential area (Settlement Y), with dwelling-houses on either side
of a narrow street.
Before reaching ground-water, the Y-sounding passed through
three distinct occupation-levels, dated by pottery to Early Dynastic
I and II, and at each of them there were further traces of flooding. In
these levels also, there were many simple graves, dug beneath the
floors of the houses; but there were also a number of so-called
'chariot burials', containing up to three wheeled vehicles, though
the grave-goods which accompanied them - weapons, vases and
even a rein-ring - were mostly of copper. It has been rightly argued
that such burials would require deeper shafts than ordinary graves
and that accordingly these must belong to an occupation above the
main 'flood' level. This could make them contemporary with the
Royal Graves at Ur, with which their contents are comparable.
Lagash
The site called Telloh halfway between the Tigris and
lies
I
Euphrates, near the modern town of Shatra. Ever since the
beginning of the present century, when E. de Sarzec brought from
101-2 it to the Louvre diorite statues of the governors (here called ishakku)
of Lagash, it has been comfortably identified as the site of that

city. '°^ It was not until 1953 that a study by Thorkild Jacobsen
drew attention to a neighbouring site called Tell Al-Hiba, 15 miles
to the southeast, where excavators from the Metropohtan Museum
and New York University have now located the true Lagash,
capital of a state, one of whose religious was ancient Girsu,
centres
loj now Telloh. ^^'^ Wehave therefore to accept that 'Tell Al-Hiba
(ancient Uruku) is Lagash city, of the state of Lagash, comparable
to New York city of the State of New York'."^5
Nevertheless, it is the finds made early in the present century by
the French at Telloh which have provided us not only with
important collections of tablets, but with treasures dating from the
time of the great rulers of Lagash, mentioned in the king-list, and in
some cases bearing their names. '^^ Sculptured stone plaques,
^S weapons and seals are associated with Ur-Nanshe, who may have
been approximately contemporary with Mesannipadda. There are
106
fragments of the famous 'Stela of the Vultures' which celebrated 6g

the victory of Eannatum, ruler of Lagash, over the neighbouring
city o{ Umma, and on which he is depicted wearing a helmet
similar to that buried with Meskalamdug at Ur, and there is the
finely engraved silver vase of Entemena, Eannatum's successor.
Some of these objects are known to have contributed to the
endowment of a famous temple dedicated to Enlil at Girsu, but the
early excavations at Telloh were such that its ruins are unlikely to
have been recognized. More will now be learnt about the state
capital from the American excavations in progress there.
Nippur
This is another great city of the Sumerian epoch, first excavated by
Americans led by H. V. Hilprecht at the end of the last century.

The mounds, today known as Niffer, extend for almost half-a-mile
on either side of the old Shatt-al-Nil river-bed, northeast of
Diwaniyah. Beneath them on the northeast side are the remains of a
fortified religious quarter, with a ziggurat and temple dedicated to
Enlil at the highest point. "^^ Renewed excavations during the
1960s under R. C. Haines partly centred on an
area between the
ziggurat and the old water-course, where an Early Dynastic temple
was found, in this case associated with the name of Inanna, 'Queen
of Heaven'. '^^ Like similar temples at the Diyala sites, it had been
maintained or rebuilt throughout all three phases of the Early
Dynastic period, and most effectively amplified the evidence
already available, regarding contemporary developments in
architecture, sculpture and the design of pottery. '°^ Of great
interest in this respect are the exposures in two levels, VIII and VII
(numbered downwards), respectively representing Early Dynastic
II and the transition from that phase to Early Dynastic Ilia, the era
of the Royal Tombs at Ur.

The temple in level VII showed several remarkable features. Its 75
twin sanctuaries were sited at one end of the long, straggling plan
and were approached through a succession of courts and
antechambers, two of which had circular brick columns to support
the roof Furthermore, contrary to the general practice at this time,
one of the sanctuaries was entered through a doorway at one end,
on the main axis of the building, thereby anticipating the
convention of later times. In these levels the Inanna temple also
produced a rich harvest of sculpture, partly in the form of votive
statues, whose clearly indicated stratigraphy provided new
evidence for the study of stylistic development. But relief carving
too was represented by an important series of square 'wall-plaques'
of the sort whose purpose is still disputed, though their figured
designs are common to almost all sites in these and later periods. ° '
'
Of pottery types a preliminary study has already been made, and

confirms to a large extent the chronological conclusions reached in
the publication of the Diyala sequence.
ASHUR AND MaRI

Mention must now be made of two other sites, situated beyond
the limits of the alluvial plain, yet both clearly to be considered
107
as outposts of the cultural and political enclave which we call

Sumerian. These are, first, Ashur (Qal'at Sharqat) on the Tigris,
1 88 miles north of Baghdad and Mari (Tell Hariri) on the
Euphrates, 7.5 miles north of the present Syrian frontier.
Ashur was later to become the first capital of the state called
Assyria. The was founded on a high rocky promontory,
city
overlooking the river at a point where its main stream was joined
126-7 by a subsidiary waterway. During the 2nd millennium b c it was
'
'
^
surrounded by a powerful fortress wall, to which a second outer

wall was later added, and extended to enclose a residential suburb,
increasing the river frontage to a length of more than 1.8 miles.
Among the public buildings, including three ziggurats, which
occupied the northern heights of the old town in these later days,
M'li
the German excavators under W. Andrae discovered a temple
dedicated to the goddess Ishtar, whose earliest foundation dated
back to Early Dynastic times. After recording and removing many
later rebuildings, they were able to study this 'Archaic Ishtar
Temple' and, in the best-preserved building-level, to expose its
complete plan and to recover the greater part of its contents. ^ '
'
Many years later, a very similar find was made by the French at
Mari. Here it was to be expected that remains of the Early Dynastic
period would be found, since in the king-list it is named as one of
the cities from which a dynasty of kings ruled over Sumer. Sure
enough, in a position adjoining one of the principal gates in the
rampart surrounding the town, the excavator, A. Parrot, found a
temple, again dedicated to Ishtar, of which the first three building-
levels (a-c) dated from Pre-Sargonid times. In the years that
followed, other temples of the same period were found nearer to
the centre of the town, and these, like the Ishtar Temple, produced
a rich harvest of sculpture and other objects. Meanwhile, Parrot
and his team had become preoccupied with the clearance of an

1 1 1 enormous palace, dating frohi the first quarter of the 2nd
millennium bc, and this too was afterwards found to have replaced
an earlier palace, contemporary with the Ishtar Temple. It was
during clearances near this earlier building that they came upon a
cache of valuable objects which included a seal-cylinder presented
to a local ruler by 'Mesannipadda, King of Ur'.' '^ The building
was thus firmly dated to the third phase of the Early Dynastic
period and the relations between the two Sumerian cities
"^
established. '
'
The shrines at these two northern sites can now be added to the
category of small Sumerian temples sites and at the Diyala
elsewhere, to whose and other conventions they
architectural
almost exactly conform. Each has one or more rectangular
sanctuaries, with an altar at one end and an entry on the cross-axis.
In each, well-preserved examples of votive statues had survived
and, against the side walls of the sanctuary, clay benches could be
seen upon which they had rested. At Ashur there were other sorts of
votive objects and ritual fixtures in the form of miniature buildings
in terracotta. Fragments of a small painted gypsum plaque,
showing the recumbent figure of the goddess Ishtar in reUef, were
thought by the excavators to suggest the form which the cult statue
might have taken.
108
62 A pseudo-Sumerian statue
from Chuera in northern Syria, a
city whose inhabitants 'had
adopted the garb and manners of
the ahen peoples dominant in
their world at the time'
Chuera
A final word should be said about this site, which is remotely placed
in northern Syria, between the Khabur and Balikh rivers, eastern
tributaries of the Euphrates. Excavated since 1958 by Anton
Moortgat, it has revealed a curiously outlandish reflection of
Sumerian civilization, much affected by environmental differences
and alien influences, = Fully occupied, as the excavations appear
'
'
to show, both in the Early Dynastic and Akkadian periods, its

buildings are of undressed stone, sometimes surmounted by
brickwork, and its craftsmanship much affected by the proximity
of metal sources. The temples have open porticoes, reminiscent of
those in the Protoliterate levels at Tepe Gawra, which have been
compared with the 'megaron' dwelling-houses of Bronze Age
Anatolia (see above) and there are strange, un-Sumerian burial
chambers. Yet there are also many votive statues - albeit small in 62
109
sizeand rather crudely made - and pottery vessels with figures in

reliefand other objects which are unmistakably Early Dynastic.
Of the inhabitants themselves Mallowan says 'They need not have :
been Sumerians they need not even have had any Semitic affinities.
;
It may be assumed that they were wealthy natives, who, like many
before them and many after, had adopted the garb and the manners
of the alien peoples dominant in their world at the time.' '^ '
Variants in Terminology
Readers of Moortgat's reports on this excavation '

'
^ should be
warned that some German scholars adopt a different terminology
for the chronological sub-divisions of the Early Dynastic period,
substituting names with a more epigraphic flavour. The terms they
use are applroximately as follows:
1 There is a Mesilim period, called after a titular 'King of Kish'

whose name has been found on a sculptured macehead. This is
associated with a group of cylinder-seals in the 'Diyala linear style'
(Frankfort), which Strommenger calls 'the early abstract style' of
Early Dynastic II (ED II). It also corresponds to the level VIII
Inanna Temple at Nippur.
2 Next comes a phase identified by Moortgat with 'cylinder-seals
grouped around an inscription [formerly] read Imdugud Sikurru .
Strommenger calls this the Farah period, and sees in these seals the
'later more naturalistic style' of ED II, perhaps continuing into ED
Ilia, like the level VII Inanna Temple at Nippur.
3 After this Moortgat couples together a Meskalamdug style with
UR I (Mesannipadda) period, which Strommenger equates with ED

IIIb."8
There have been criticisms both of these terms themselves and of
the chronological conclusions which they imply. Mallowan, for
instance, points out that Mesilim's name is not included in the king-
list and that macehead could be as late as ED III. (Rowton places
his
him a little before Ur-Nanshe of Lagash, in ED Illa). Mallowan also
reminds us that the Royal Graves at Ur were dug into a layer
containing tablets older than those from Farah. Generally, British
and American scholars prefer the simpler system of numbered
phases.
no
Chapter Six
Pre-Sargonid Art and Architecture
We have now summarized the main results of excavations at some

of the principal ancient founded or occupied during the
cities
earliest historical periods in Mesopotamia.Our total knowledge of

the civilization created during these centuries and of the way of life
which It engendered has been acquired from two principal sources:
first, from the study of material remains, exposed or recovered by
archaeologists, and secondly from the contents of the written

documents which their finds have made available to philologists. It
is with the former source of information that we are here primarily
concerned; so we must now consider the archaeological finds in

greater detail. This may again be done under separate headings.
Sculpture
Statues
It might today seem strange to us that sculpture in stone should
feature so significantly in Mesopotamian art. Perhaps the very
scarcity of the material and the remoteness of the sources from
which it could be obtained themselves invested it with a rarity
value, appropriate to the purposes for which it was generally used.
For a fact that must initially be remembered is that all forms of
Sumerian sculpture, whether statues in-the-round or relief

carvings, were of a religious character and intended to perform
some ritual function within the confines of a temple. In this respect,
the commonest artform, and perhaps the most characteristic of the
Early Dynastic period, is the category of votive or 'personal'
statues, already frequently mentioned in describing the contents of
early Sumerian temples. Fortunately the dedicatory inscriptions
often found carved upon them leave little doubt as to their
intention. The effigy of an individual worshipper, translated into
stone and placed in the sanctuary of a religious building, could be
expected to intercede on his behalf with an appropriate deity. '
'
"^
In almost every one of the many temples mentioned in the

previous chapter such statues were found. Whether lying displaced
or broken beside the brick 'benches' on which they had stood,
carefully buried beneath the floor of a sanctuary or even built into
the structure of an altar, their recovery has contributed to the great
volume of Sumerian sculpture now available for study. All that is
now lacking in this field of discovery are the actual cult-statues

themselves, which must have stood upon the altar-platform,
creating a focal point for the liturgy ofSumerian religious ritual.
Among the cache of discarded sculpture found in the Square
Temple at Tell Asmar, a pair of male and female figures, almost half
life-size, were tentatively identified by Henri Frankfort as god and

goddess, by symbols carved on the base of the male figure. But the
evidence of these and of the exaggerated size of their staring eyes
have proved unconvincing to most other critics, who note in them
an attitude of prayer common to all the smaller and less ostentatious
votive figures: the folded hands, holding a cup from which a
libation is about to be poured, and other features seeming more
logically associated with a worshipper than with the object of his
veneration. At Ashur too, in the 'Archaic' Ishtar Temple, W.
Andrae imagined a gypsum figure of the deity, in high relief above
the altar, since a miniature version of such a figure had been found
among the debris of the sanctuary. ^° This proposal too did not
'
prove wholly acceptable, since no imprint of it had been left upon

M'
the wall. It is in fact not improbable that cult-statues generally were
too valuable and too easily removed to be likely to have survived.
It was Frankfort again who first attempted a stylistic analysis of
these sculptures, using mainly as his study-material the 200-or-so

complete or fragmentary statues recovered from temples at the
Diyala sites. ^ Bearing in mind that their peculiarities might be
' '
slightly provincial, he was able to distinguish two separate stages in

the development of their conventional design: an earlier and a later
style of carving, corresponding as it proved, at least in the Diyala
region, to the second and third phases of the Early Dynastic period.
Regarding the earlier of these two styles of carving, if we are to
look first for its antecedents, they can be found only among the few
surviving works of Protoliterate sculptors, to some of which we
63 Stone figure of a worshipper have already referred. Of these the only object comparable to our
from Khafaje (Jemdet Nasr votive statues of later times is a rather crudely carved female figure
period) : the earhest known
from a Jemdet Nasr provenance at Khafaje, whose hands, folded
prototype of votive statues found
in great numbers among the around some missing object, already suggest an attitude of
sanctuaries of laterSumerian worship. '^^ Chronologically, however, this object is separated
temples. The style of carving is from Frankfort's 'earlier' dynastic sculpture by a period of time
still undeveloped. Ht 10 cm
corresponding to the entire Early Dynastic I period, to which no
single work of sculpture has yet been firmly attributed. We shall
therefore be justified in concluding that the Early Dynastic II style
which we are about to consider, developed independently of any
traditional influence. Its characteristics may perhaps best be seen in
64 the group of statues from the Square Temple at Tell Asmar, all of
which appear to be attributable to a single 'school' of sculptors. ^^ '
Of the 'worshippers' in this group, nine are men, dressed in the

conventional garment of the period - a simple skirt of wool with a
girdle and long fringe. One, who is bald and clean-shaven, can be
recognized as a priest the remainder have square beards and long
;
hair, both neatly corrugated and painted black. Their eyes are inlaid
with shell and lapis-lazuli. The twelfth sculpture is the kneeling
figure of a priest carved in alabaster, ritually naked and wearing a
toque-like headdress. The two female figures also wear a
conventional garment, passing diagonally across the breast and
draped over the left shoulder. As to their hairstyle, they show two
variations of an arrangement common to both the second and third
Early Dynastic phases: a 'halo-plait' passing vertically over the
crown of the head with a 'chignon' behind. But many other devices
are to be seen at this time. The styHstic peculiarity of these figures
112
consists mainly in the distinctive rendering of the human body, by 64 Group of votive statues of the
Early Dynastic II period, found
the reduction of its component shapes to abstract forms. Whatever
in the 'Square Temple' at Tell
the difference between individual statues, the formal principles are Asmar. The tallest pair were at
the same: the sculptor has somehov^ rationalized its combination first thought to be cult statues of
of miscellaneous features into a manageable formula of semi- a god and goddess. They
represent Frankfort's 'earlier
geometric equations. It can of course be contended that the general
style' of carving, with
simplification of forms may have been dictated in part by lack of
formahzed geometric shapes. Ht
technical skill among these early stone-cutters but the artistry with
;
of tallest figure, 76 cm
which it was achieved testifies rather to an interest in design, as
opposed to actual representation. An interesting contrast is
incidentally to be seen in the contemporary figures modelled for
casting in metal. In the Early Dynastic II period metallurgy was
already well understood and some striking works in copper were
among the finds at Khafaje and Tell Agrab. But the designs are
purely naturalistic and show no signs of deliberate formalization.
For the purpose of Frankfort's study, the Early Dynastic III style
of sculpture was represented mostly by votive statuary from the
Temples at Khafaje. But it is also abundantly illustrated by
later Sin
examples from other sites, both in Sumer itself and in the
dependent provinces of upper Mesopotamia (where the distinction
between the two stylistic phases is less clearly defined). The
sculptors of this later period have gained in confidence. They have
discarded the devices of formal simplification and are no longer
of physical appearance.
afraid to interest themselves in the details
The subtler contours of the body are carefully modelled; mouths
and cheeks are shaped to give the face expression; the corrugation
113
of hair and beards are replaced by a pattern of curls, sometimes

separated by drill-holes. An even more conspicuous change has
taken place in the design and portrayal of clothing. Now for the
first time the so-called kaunakes appears: a system of weaving
which covers the whole garment with petal-shaped tufts of wool in

an overlapping pattern, thought of by some as an attempt to
simulate sheep-skin. ^'^ Certainly it transforms the appearance of
'
these statues and is freely represented in other forms of pictorial art

at this time.
There is now in fact a striking uniformity of style and artistic
convention throughout Mesopotamia. Fashions in dress and
personal appearance are governed by a rigid convention which
ignores any regional difference of race or tradition. The
idiosyncrasy, for instance, of shaving the head while growing a
luxurious beard, which is seen in statues from Khafaje or Warka, is
equally common Mari on the upper Euphrates, where the

at
65 The seated statue of Ebih-il population is predominantly un-Sumerian. It is in fact from the
from Mari, representing the Ishtar Temple and other Early Dynastic shrines at this northern site
'later style'of the Early Dynastic
that some of the fmest and best-preserved Early Dynastic statues are
III period. He wears the
characteristically Sumerian
derived, several of them inscribed with their names and occu-
kaunakes garment with 'petals' of pations. '^^ Some also, like the famous Ebih-il, superintendent of
wool. His beard indicates the use the temple, are shown seated upon a chair or stool: once more a
of a drill: a hallmark of Mari
convention which is to be seen at Khafaje and elsewhere. Only the
sculpture. The figure is dedicated
to the goddess Ishtar. Ht 52.5 cm women in some cases wear a 'polos' headdress, shaped like a biretta,
and one of them has a kaunakes outer garment draped over the top
of it. By contrast, the 'singer', Ur-Nanshe, is seated crosslegged on
a cushion and has straight hair, parted in the middle. ^^ '
Equally rare are sculptured figures other than those of human

beings, such as the bearded cow in alabaster from the Nintu Temple
at Khafaje, which Delougaz suspected of being in fact a cult-
statue. ' -^ We should accordingly now turn to the subject of relief

carving, since it is plentiful and of great interest.
Relief Carving
In Early Dynastic times this form of carving is at first best illustrated
by a of wall-plaques in stone or slate. They are square,
class
perforated in the centre and decorated in relief with a variety of
pictorial scenes. Their purpose has been widely discussed and a
number of ingenious suggestions put forward. '^^ The most
obvious inference - that they were affixed to an interior wall-face
by a central peg, perhaps with an ornamental head - is confirmed
by some examples in which flanges of undressed stone at the sides of
the square are clearly intended to be covered by the surrounding
plaster. As for the scenes depicted, they are usually arranged in two
or three registers, showing rows of figures engaged in religious
ritual or ceremonies associated with royalty. Some are intended to
66 Seated statue of an individual
commemorate a particular event and in one well-known example,
;
called Ur-Nanshe, from the

Ishtar Temple at Mari.
both the nature of the occasion and the names of the participants are
According to an inscription, she recorded by inscriptions. ^^ Ur-Nanshe, 'divine bailiff of Lagash', is
^
(or he?) was a singer at the court seen carrying bricks for the construction of a temple, accompanied
of King Iblulil, which may by his sons and But there were other, more standardized
attendants.
account for the unconventional
pictorial compositions,one of which had clearly at this time
hairstyle. (Early Dynastic III
period). Ht 26 cm become so stereotyped among Sumerian craftsmen that a missing
114
67 Sumerian wall-plaque from

Khafaje (Early Dynastic III),
showing servants attendant upon
a king, with his chariot in
readiness below\ A missing
fragment (bottom left), has been
replaced by part of a similar
plaque from Ur, proving that
the design was standardized at
this time. The plaque is 51.5 cm
square
68 Wall-plaque carved in relief,
from Telloh. Ur-Nanshe of

Lagash is seen carrymg a basket
of bricks for the building of a
temple - and again, seated
among his children. The archaic
inscriptions are an unusual
feature of this plaque. Ht 40 cm
115
Prt'-Sargonid Art and Architecture
fragment of a plaque from Khafaje could be restored by reference

found at Ur, some 100 miles away.'^^ This was the
to a replica
well-known 'banqueting scene', where a royal person in the top
register being served with food and drink, while below his empty
is
chariot is being prepared. Musicians and domestic animals

complete the pattern.
As we have mentioned earlier, an important sequence of these
plaques was found at Nippur in levels VIII and VIIB of the Inanna
Temple. They have been very carefully studied by D. P.
Hansen, ' who points out that, in the earliest examples of all, the
-*
'
figures are merely drawn with incised lines, whereas they are later
made to project from a recessed background. The designs at this
stage are restricted to simple motifs, reminiscent of cylinder-seals. It
II*
f' is not until the second phase that they are improved by detailed
modelling and that subjects like those mentioned above begin to be

depicted. One may mention in passing that a musician depicted in a
banqueting scene from level IIIB at Nippur carries an eight-
stringed lyre with a bull-headed sounding-box, precisely similar to
those found in the Royal Tombs at Ur (Early Dynastic Ilia). -'^ '
Sumerian relief carving reaches a high degree of proficiency in

the final phase of the Early Dynastic period (ED Illb). Unfor-
tunately few examples of major works from this period have
survived, and our judgment must be based primarily on the
69 monument known as the of the Vultures', fragments of
'Stela
which were recovered by the French excavators of Telloh and are
now in the Louvre. The scenes, which are carved in horizontal
registers on both sides of the stone, commemorate the victory of
Eannatum, ruler of Lagash, over the neighbouring state of Umma.
The king, wearing a helmet resembling that found in the Tomb of
69 Part of the so-called 'Stela of

from Telloh,
the Vultures'
showing Eannatum, 'Divine
Baihff of Lagash', at the head of
the Sumcnan phalanx, and
(below) driving his chariot. He
wears helmet similar to that of
a
Meskalamdug, found in his tomb
at Ur. Ht 182 cm
116
Pre-Sargonid Art and Architectw
Meskalamdug at phalanx of his spearmen into battle,

Ur, leads a 70 Part of a complicated pot-
drives a chariot at the head of his light infantry and afterwards stand from Tell Agrab (Early
Dynastic II), carved in high relief
presides over the ceremonial burial of the dead. On the reverse side,
with a scene familiar in cyhnder-
the victory is symbolically attributed to his god, Ningirsu, the seals of a 'hero' figure holding
vârrior-god, son of Enlil, who gathers his victims in a net, sealed spotted lions by their tails. Ht
with the image of Imdugud, the lion-headed eagle. The objective 20 cm
record of a tumultuous event thus culminates in the expression of a
religious abstraction. This monument, with its individual style of
carving and ingenious pictorial composition, may well be
considered one of the great documents of human history.
Returning for a moment to the Early Dynastic II phase, there are
other forms of stone sculpture found in temples vases, for instance,
:
or vase-holders, with figures partly carved in-the-round. An

elaborately ornamental example from Tell Agrab depicts the naked 70
and bearded 'hero', familiar on cylinder-seals, grasping two lions
by their tails. In another, a kneeling priest, also naked, bears a vase
on his head. But there is another category of vessels which have
long remained an enigma on account of their wide distribution in
countries other than Mesopotamia. These are flat-bottomed bowls
with almost vertical sides, made of a soft greenstone, formerly
referred to as 'steatite' (now more accurately identified as chlorite).
Their carved designs are unique. To take a typical example from
Khafaje, now in the British Museum, standing or seated human
figures, Sumerian in appearance, hold spotted snakes or stylized 71
streams of water, which undulate around the face of the vase, above
and between mythical beasts and symbols. ^^ These include lions,
'
leopards and small bears, with birds of prey and occasional

scorpions. But also prominent and more unusual are humped bulls
117
71 A 'steatite' vase, imported

from Baluchistan, showing a
seated Sumerian figure, a spotted
snake and a humped bull (zebu),
which is not native to
Mesopotamia. These vases have a
wide distribution throughout

Mesopotamia and Persia. Ht
1 1.4 cm
of the zebu breed, which is not native to Mesopotamia. Some of

these figures are sparingly enriched with inlays of paste or coloured
stone. Most remarkable of all, however, is their wide geographical
distribution. Almost identical examples have been found as far
north as Mari, south of Sumer itself at sites in the Arabian Gulf and,
finally, as far away as Mohenjo-daro in the Indus Valley.
Some light has at last been thrown, at least on the subject of local
manufacture, by finds at the extraordinary site called Tepe Yahya
south of Kerman, which lies on a main route from Sumer to north-
ern India. Here, in a level dated to the late 3rd millennium bc,
C. C. Lamberg-Karlovsky recovered over 1,000 fragments of these
'steatite vases', including some unfinished vessels and 'wasters',
together with large pieces of the raw material from which they
were being carved. He later located a nearby source from which
chlorite was obtained. ^^'* The excavator was also able to confirm
that later the designs deteriorated into more simple representations
of reed architecture. ^^^ But their symbolic significance and the
talismanic(?) properties which made them so internationally
popular at present remain a mystery.
Architecture
Building Methods
The Sumerians, like other inhabitants of Mesopotamia, constructed
their buildings of mud bricks, shaped in a four-sided mould and
yz dried in the sun. Up to the end of the Protohterate period small
rectangular bricks laid flat in horizontal courses (German, riemchen)
seem to have been the rule but, after a transitional period, covering
;
the Early Dynastic I period, these disappeared completely and a

new method of building was adopted. Bricks were now slightly
118
larger and 'plano-convex' in shape, having one rounded face often

marked with the impression of a fmger or thumb. '
^^ They were
laid 'on-edge' like books on a shelf, successive courses leaning 73
sideways in opposite directions to create a herring-bone pattern
(easily recognizable as a criterion of the second and third Early
Dynastic phases). Kiln-baked bricks of the same shape proved more
suitable for pavements and wall-faces in parts of a building where
water was in use, or for the revetment of an external facade. At a
few sites, such as Eridu, Ur, and Mari, some poor-quality stone was
locally available and could be used for foundations or powdered to
make gypsum plaster. For the rest, buildings of the Early Dynastic
period are assumed to have had flat roofs, constructed of palm-
trunks, brushwood and clay. Doors pivoted on an indented stone,
while windows - generally small and high up in the walls - could be
protected by perforated terracotta grilles. In the absence of suitable
wood, columns for supporting the roof were rare; but circular
pillars, composed of segmentally shaped bricks, were not
uncommon in major buildings. '

^^
Temples
A good deal has already been said about the planning of temples at
this period. We have seen how they can apparently be divided into
two classes: first, the 'high' temples, raised on brick platforms or
actual ziggurats, and secondly those at ground-level, sometimes
enclosed by private dwelling-houses. Unfortunately, for reasons
which have been explained, very little is known about the planning
and appearance of the former, and we are able only to assume some
similarity to their prehistoric prototypes. In the second category,
however, we are now provided with a dozen examples, so widely
distributed among the Sumerian cities and sharing so many
characteristics in common that a general analysis can be made of
their design and function.
72 In the modern villages of Iraq

bricks are made of mud,
tempered with chaff and cast in a
bottomless wooden mould,
smoothed on top and dried in
the sun
73 Method of laying the 'plano-

convex' bricks, used exclusively
in the Early Dynastic II and III
periods.
119
The nuclear element in these temples, then, is a rectangular

sanctuary with a door in one of the long sides ('bent-axis
approach'). At one end there is a raised brick altar and behind it a
wall-niche marking the position in which the cult-statue must have
stood. Elsewhere in mid-sanctuary are smaller brick platforms or
pedestals usually referred to as 'offering-tables', while brick benches
along the bases of the walls served, as we know, as emplacements
for votive statues (Ashur). More portable forms of furniture
included tall pottery stands, in which some sort of foliage was
placed; figures of carved stone or copper, supporting vases or
bowls of incense (Khafaje, Agrab, Ur, etc.); 'rush-lights' of copper
(Kish) and larger offering-stands of stone or terracotta, sometimes
simulating buildings (Ashur). Explicit indications of actual ritual
are to be found only in the actual structure of the altar. Its upper
surface was sometimes protected by an inlaid slab of stone or slate
(Asmar), from which ran a small terracotta drain, discharging into
a concealed pottery jar (Agrab). These provisions could suggest
either animal sacrifices or the pouring of libations, both of which
are portrayed on cylinder-sealsand elsewhere.
In its simplest form, this sanctuary chamber needed only a
buttressed outer facade, small towers flanking the entrance and
a modest annexe containing a bread-oven (Asmar, Single Shrine
Temple) to fulfil the requirements of a religious building. '^^ As
larger and more elaborate temples developed, a main sanctuary as
described above continued to be the nuclear element of the plan,
around which extensions of various sorts could be added in ways
varying according to the shape and other limitations of the site. The
first addition (reflecting traditional precedents), would be rows of
lateral chambers on either side, including one with a stairway

giving access to the flat roof. Next, a forecourt could be contrived
by enclosing the open space which usually existed at the approach
to the main entrance. In due course this would itself be surrounded
by subsidiary chambers, forming a small precinct, itself entered
through an outer gateway with flanking towers.
This process of growth is well illustrated by the Sin Temple at
Khafaje which, in the final stages of its development, still occupied
an irregular-shaped site, surrounded by private dwellings. ^9 A '
complete contrast is presented by other temples for which an open

site was available, making it possible for the building to be pre-
planned and symmetrically composed. An unpretentious example

is the Square Temple at Tell Asmar, which is neatly arranged
around a central court in such a way that two minor shrines can be
added to the main sanctuary. This more formal type of design,
unhampered by site restrictions, reaches perfection in the great
74 Shara Temple at Tell Agrab, whose massive outer walls enclose a
number of self-contained units, each with chambers grouped
around a separate courtyard. '"^^ In addition to the generously
proportioned main sanctuary, with its two-tiered altar and ranges
of offering-tables, two minor shrines are again added, together
with elaborate accommodation for resident priests.
A comparative study of all these buildings and the characteristics
which they share has now made it possible to visualize rather clearly
the appearance and function of a typical place of worship in early
120
;
74 Plan of the 'Shara Temple' at

TellAgrab and adjoining city-
wall, partly reconstructed by
reasoned conjecture. The
surviving part comprises {top left)
a unit with two minor shrines

(centre) the main sanctuary, with
dependent chambers; and (below)

a priests' resicTjnce.The high
altar was two tiers, with
built in
a miniature stairway. (From
Delougaz and Lloyd, 1942)
Sumerian times. One example only seems to deviate from the

formula most generally accepted. In the Inanna
architectural
Temple complex at Nippur, in addition to a sanctuary of the
conventional 'bent-axis' type there is a second, free-standing shrine, 75
isolated in its own courtyard and entered by a doorway centrally
placed at the end opposite the altar. '
"^
' This arrangement was to
become the basic principle of all Mesopotamian temple planning
from the end of the Early Dynastic period onwards.
Before leaving the subject of Sumerian temples, something may
well be said about the apparent discrepancy between the modest
scaleof their architectural remains and the magnitude of the
purpose which they served. Quite apart from their function as a
75 Two sanctuaries of the 'Ishtar

Temple' at Nippur (level VII,
Early Dynastic II). That on the
left has the long-axis approach,
more common in later times.
Elsewhere in complex were
this
of brick columns. (From

pairs
Hansen and Dales, Archaeology,
1962)
121
setting for seasonal ceremonial and religious ritual, their paramount

function economic and political administration of
as centres for the
the state would surely lead one to expect far larger and more
pretentious buildings than those which we have till now been
describing. Let us, for instance, glance at a brief account given by
H. W. F. Saggs of their status in this respect and the part which they
played in the life of a Sumerian city. Bearing in mind that his views
are not universally accepted, he says:
The government of the Sumerian city-states at this time [mid-3 rd

millennium bc], was hieratic. . . . The city and its lands, with all its
inhabitants, was the estate of the city-god, with the ruler or priest-king (of
whom Gilgamesh was a prototype) as his steward. The gods created Man
to do their service, wherefore the free citizens were the servants of the god.
Their primary function was thus,

that they should serve the temple estates, or(what was the same thing in
ancient eyes) the human who acted as the god's steward, namely the Ensi
(governor or prince) of the city. In return, each free citizen received an
allotment from the god's estates, that is from the temple lands.
It isclear that, in early days, the greater part of the land - or perhaps
all of it - belonged to the temple, and it was let out to the people on
a share-cropping basis. The Ensi was the bailiff of the city-deity and
administered the temple estates with a hierarchy of officials to serve
him. Private ownership of property in land continued to be very
rare until the fmal years of the Early Dynasties, when the more
powerful Ensis tended to divert public funds for the aggrandise-
ment of their own families. This led to the rise of a secular power,
distinct from the religious authority (and perhaps to the kind of
confrontation which took place under certain 'governors' of
Lagash). By that time, state rulers had begun to acquire, as we shall
see, large palaces of their own; and these perhaps relieved the
pressure of space in the temples themselves. But one must also

remember that, through the whole of the Early Dynastic period,
the service of scribes was still at a premium and the actual bulk of
written documents accumulated in the process of state adminis-
tration still on a very modest scale. Let us now therefore consider
what is known about such buildings from excavations.
Palaces
Where secular buildings are concerned, those loosely described as
'palaces' form the most obviously important group. And here we
find that any commentary on their architecture is hampered by our
imperfect understanding of their function and practical require-
ments. In the earliest phase of the historical period, the concept of a
'temple' seems to have implied a seat of royalty, as well as the
administrative centre for the ruling hierarchy. Later, however, a
more effective separation must have taken place between church
and state; for the king now had a residence of his own, presumably
of a sort which could meet the requirements of political activity and
ceremonial. It is these characteristics therefore that we should
expect to find in our so-called 'palaces'. There are in fact four
notable examples, of which three only - at Kish, Eridu and Mari -
have sufficient features in common to suggest a conventional
formula for such buildings. The fourth is the Northern Palace at Tell
122
76 Twin palace buildings of the
(D Early Dynastic
monumental gateway
Ilia period and a
Here again, circular brick

at Kish.
columns are an unusual feature.

35m (After Mackay)
Asmar, which seems at this period to be composed merely of

domestic units, each grouped around its own courtyard. '^^
The palace complex at Kish (Early Dynastic Ilia) is composed of 76
two buildings separated by a narrow alleyway. '"^^ The larger of
them, approached through a towered gateway, is protected by
duphcate outer walls and has a single interior courtyard, around
which it is formally planned. Of the less heavily-walled building
annexed to it, the only distinct features are at one end a long
hallway with four central columns and at the other some sort of
columned 'loggia'. Beyond this, neither plan is in any way self-
explanatory. Equally enigmatic in most respects are the twin
palaces of the Early Dynastic period, located outside the raised 77
temple-precinct at Eridu. ^-^"^ Once more these are protected by
77 Outside the raised temenos at

Eridu, one of two palaces dating
from the Early Dynastic period.
EH The unit consisting of a main
reception room and square court,
sets a precedent for palaces of
\\ U later times. (After Safar, 1950)
mm U
123
double walls, buttressed like those at Kish in a manner normally

considered more appropriate to a temple. Yet here also we see the
earliest known example of an 'audience-hall' or 'throne-room',
opening off a square central courtyard an arangement which we
:
shall fmd adopted from now onwards as the basis of the main
'reception-suite' in all Mesopotamian palaces.
This combination of square courtyard and rectangular throne-
room even more strikingly illustrated in the 'Palais Pre-
is
sargonique"", which French archaeologists have partially excavated

beneath the southeast corner of the huge 18th-century bc palace of
Zimrilim at Mari. "^^ Whether or not this will prove to be a single
element of a much larger building, its importance as an
independent unit is emphasized by the duplication of its outer walls
1^1 on all four sides. It is also confirmed by another aspect of the
building, which has led the French to speak of it as a zone sacre. The
rectangular hall has a small sanctuary at one end; the square court
has ornamentally recessed wall-faces and is provided with ritual
installations of various sorts; foundation deposits had been laid
beneath the pavement of corridors separating the outer walls. If
such discoveries seem strange in a building apparently dedicated to
a secular purpose, it should be mentioned in advance that similar
features occur in the main reception suite of the great palace built
on the same site by Zimrilim some seven centuries later. As we shall
see, they have been subject to diverse interpretations.
Cylinder-Seals
The transition from one phase of the Early Dynastic period to

another is changing designs of cylinder-
clearly reflected in the
seals. '"^^ We have already observed a certain deterioration in the
artistic quality of seal-cutters' work towards the end of the
Protoliterate epoch, and little improvement is to be seen during the

century-or-so which followed. All interest in mythology and
religious symbolism seems to have been abandoned, in favour of
decorative compositions based on the shapes of animals and plants,
supplemented by meaningless ornament curiously reminiscent of
textile designs. This 'brocade style', as it has been called, is the
hallmark of the first Early Dynastic phase. A reversion to more
interesting and imaginative subjects begins to take place early in
phase II. One subject which is revived and soon gains in popularity is
that of 'animal contests' attacks by Hons on cattle and their defence
:
by human or half-human 'guardian' figures. In this role the bearded

78 'hero' reappears and the 'bull-man', now wearing a girdle and
'"^^
ornamental side-locks beneath his horns. Intricate patterns are
made from these figures, which
sometimes interlaced or
are
inverted; but there is little attempt at plastic rendering and the
linear designing is undisciplined. At a later stage experiments are
made with pictorial subjects such as the well-known 'banqueting
yg scene', but the results are hardly more successful. A notable
innovation, however, is the occasional introduction of a picto-
graphic inscription.
In the third Early Dynastic phase, seal-making finally attains its
full status as miniature relief carving. The figures become more
124
78 Cylinder-seal of the Early

Dynastic II period. Familiar
figures of the naked 'hero' and
the 'bull-man' protect horned
animals from attacking lions
79 A 'banqueting scene' of the

Early Dynastic III period. The
seated female figure in the top
c? register is identified by an
^J^f^ ^
empanelled inscription
a
(previously known
queen whose body occupied
as
as Pu-abi
'Shubad'),
one of the richest tombs in the

Royal Cemetery at Ur. Opposite
her is presumablyher husband
Abargi
80 Improved carving in the

Early Dynastic III period. Bulls,
now shown full-face, are
protected from lions by the usual
mythical figures
81 A seated figure, identified as

Shamash, the sun god, in his
boat, with other deities and
appropriate animal symbols
125
\lumiJiliim„„u,u„. ^^, """"" Iff
82 Metal tools and weapons

from Ur tombs. Top row. tanged
and ribbed spearheads; second
row: leaf-shaped spearhead,
harpoon, knives and daggers;
row drill-bits, straight and
4
third :
curved pins with eyeleted /^

shanks; bottom row: a and b,
scrapers; c, chisel; d and e,
socketed adzes and/, axe cast in a

two-piece mould; g and h,
crescentic battle-axes
I
^
f^^
1 '^ \7
I
L\
126
massive and are skilfully modelled; lions and 'guardians' alike are
often depicted full-face and more ornamental species of ruminants 80
are introduced; scenes of combat are punctuated by empanelled
inscriptions ;designs are occasionally divided horizontally into two
registers. Perhaps the most intriguing subjects chosen for these
Early Dynastic III seals are the mythological or ceremonial scenes in
which men and gods take part. Their interpretation has always
depended largely on the evidence provided by Sumerian literature,
and since more of this became available, many themes have been
reliably identified. One sees for instance the sun-god, Shamash, 81
voyaging in his boat through the cycle of night and day,
surrounded by symbols which the texts explain. Then there is the
fertility god on a throne, before which priests pour libations or pile
offerings on an ornamental stand. More rarely portrayed is the

ritual 'marriage', which took place between a god (Ningirsu) and
goddess (Bau) at the New Year's Festival. But secular 'banquet'
scenes make an alternative to religious subjects, sometimes attended
by a burlesque 'orchestra' of animals. Many of these motifs now
appear in other forms of ornament, especially on objects from the
Royal Tombs at Ur.
As for the materials from which the seals are made, many
of semi-precious stone are now in use. The cylinder could
varieties
even be capped with silver at either end or, in rare cases, made of
solid gold.
Metallurgy and Composite Craftsmanship
Treasures recovered from Early Dynastic levels at Sumerian sites 83 Copper stand for an incense-
testify to the ability of specialists in several different fields of bowl representing a naked priest.
Ht 55.5 cm. Three were found in
craftsmanship and, in some of the most strikingly beautiful objects,
the Oval Temple at Khafaje
one sees how their skills have been combined. '"^^ In a class by (Early Dynastic III), smaller ones
themselves are the metalsmiths, whose services must at this time at Tell Agrab
have been so greatly in demand at all levels of society. Their
products, in the form of weapons, implements and utensils, have
survived in great numbers, side-by-side with pictorial repre-
sentations of the purposes for which they were used. In a well-
known relief, spears, shields and helmets of copper are to be seen in ;
another battle-axes 'guardian' figures in seal designs wield daggers

;
with crescent-shaped handles; 'rein-rings' with their animal

mascots appear on chariots in battle-scenes; and at a 'banquet'
copper 'drinking-tubes' are used.
For votive and ceremonial purposes, replicas of such objects
were made in precious metals and delicately ornamented. The
work of Sumerian goldsmiths ranks with that of the best craftsmen
in any later age and there were few technical processes with which
they were still unfamiliar. In fine hammered work, for instance, as
well as repousse, chasing, engraving, granulation and filigree, they
were entirely proficient. The perfection of their skill is seen in an
object like the wig-helmet of Meskalamdug, which WooUey has
described as '. . . a up from the flat,
veritable tour -de-force', beaten
with locks of hair hammered and individual hairs
in relief
represented by fine chased lines'. At the same time, some more
simple objects, like the fluted cups and vases from the same tombs.
84 A reconstructed figure in the show something more than mere technical competence. In their
Iraq Museum of Meskalamdug, combination of shape and ornament one recognizes a real talent for
wearing the ceremonial wig-
elegant design. The work of craftsmen in gold and silver is again to
helmet, found in his tomb at Ur.
Beaten from a single sheet of be seen in the ornaments of female attendants in the Ur tombs 'the
:
gold and delicately chased, its gorgeous beech-leaf headdresses, the gold and silver florally
perforations are for attachment decorated combs; the wreaths composed of Httle amulets in the
to a leather linmg. {c. 2500 Bc)
shape of bulls, rams, birds and fish' and the huge boat-shaped
'"^^
earrings.
85 Reconstructed figure in the
Iraq Museum of a court lady In another sphere, the most remarkable accomplishments in
from a royal tomb at Ur, casting must be attributed to the coppersmiths of the Early
wearing an elaborate headdress Dynastic period. Examples of their work are already in evidence at
and jewellery of gold, lapis-lazuH
Tell Agrab during its second phase and they include the remarkable
and carnehan. There were 68
similarly dressed women in the
group in miniature of a chariot drawn by four onagers and driven
'dromos' of a single tomb (no. by a bearded 'king'.'ô Technologically this work is of special
1237), known to the excavators interest because it shows evidence of casting by the so-called cire
as the 'Death
perdue or 'lost-wax' process which now appears for the first time.
Pit'
More common finds in the Diyala temples were copper statues in

«5 the form of naked priests (or 'heroes'), with attachments enabling
them to be used as offering-stands. One of these, from a group of
three found at Khafaje, measured 76 cm high, while fragments of
another from an Early Dynastic III provenance at Tell Agrab
showed it to have been life-size.
128
Inlaid Ornament 86 Miniature group in copper of

Another realm in which Early Dynastic craftsmen excelled was that a chariot drawn by four onagers
and driven by a bearded
of inlaid ornament in coloured stone and other materials. Its uses,
Sumerian king. It was found in
perhaps developing from the cone-mosaic technique of an earlier the 'sacristy' room near the high
period, varied from pictorial friezes decorating the facades of altar in the 'Shara Temple' at
buildings to the miniscule enrichment of precious objects. In the Tell Agrab and is the earHest
former case, carved figures of white limestone were set against a known example of the cire perdue
('lost wax') process. Details of
background of grey slate. For small-scale ornament a wide variety the chariot and harness have been
of coloured materials were used to form a surface incrustation, laid reconstructed after careful study.
in bitumen over a wooden core. Of figurative designs executed in Ht 7 cm
this technique, individual features could be carved in shell, bone or
mother-of-pearl, to be fitted into a coloured background. A few
detached figures of this sort were first found in Palace A at Kish.
But the full possibilities of the medium were revealed by the
abundant and often undamaged examples surviving among the
treasures of the Ur tombs. Gaming-boards, toilet boxes and the
sounding-boards of harps were some of the objects decorated in this
way, often with spirited scenes recalling those on cylinder-seals.
Individual figures cut from shell or nacre could be engraved with
additional detail and the incisions filled with red or black paste.
Flakes of lapis-lazuli would then be fitted together to form a
background, and the design enclosed in a formally patterned
29
'
I"
87 The so-called 'royal standard' border of other colours. At Ur, the tour-de-force in this category of
from a tomb Ur: a mosaic of
at
ornament was the famous 'royal standard' with its scores of busy
shell figures on a background of
figures on either side, representing scenes of 'war and peace', and
lapis-lazuli, depicting on this side
a 'peace' scene of banqueting and thought by some to be the sounding-box of some musical
agricultural activity. Ht 20 cm contrivance. '
^
Less well-known than the Ur 'standard', owing partly to the

88 Reverse side of the 'royal
incomplete state in which it was found, is a similar and equally
standard', showing a 'vvâr' scene,
with the king {primus inter pares),
magnificent mosaic panel from Mari, composed of elements in
his helmeted infantry and ivory, shell, schist, lapis and gold. ^2 In the uppermost of its three
'
chariots in action. This object registers, a procession of conventionally dressed male figures
may have been the sounding-box
attends a libation ceremony, carrying appropriate vessels; but the
of a musical instrument
scenes beneath them are composed entirely of women, whose dress
and occupation are of unique interest. Above, two figures face each
other at either end of a couch supported by legs resembling those of
a bull, and bend forward to arrange its kaunakes draperies.
Approaching them from behind are other women, wearing the
'polos' headdress and carrying votive vessels. Below them are pairs
of women with unfamiliar coiffures, one of each pair seated on a
three-legged stool while the other assists her in a task for which a
'distaff appears to be changing hands. Without exception, these
130
female figures wear, over their usual lower garment, a fringed

shawl, conspicuously fastened by a long metal pin with its head
bent commonly found object, whose purpose has
at rightangles (a
till remained obscure). From a perforation at the head of the

now
pin hangs a string of curiously shaped amulets or charms.
Supplementary details add to the interest of this intriguing
composition.
Composite Objects
It remains for us to consider some examples of the composite
objects which figure so conspicuously among Early Dynastic art
treasures. In a category of their own are musical instruments found
in the Ur tombs, which WooUey extracted from the earth with
such meticulous care. In referring to these the word 'harp' is
sometimes used rather loosely, since in fact only two answered to

this description. The most common instrument was a lyre, with
rectangular wooden sound-box inlaid in coloured ornament,
upright members at either end and a cross-piece to which the
strings were attached with wooden pegs for tuning. Usually the
89 Shell inlay warrior with a
body of the lyre had a frontal ornament of metal in the form of a battle-axe from the Ishtar
bull's head; but there was also a 'boat-shaped' lyre, with the copper Temple at Mari (Early Dynastic
figure of a stag set against the foremost upright. As we know from III)
pictorial inlays found both here and at Mari, these instruments

could be held at chest-level by a standing player, or themselves
stand upright on the ground between the player's knees. ^3 There
^
was also a so-called 'harp-lyre', with twelve strings attached to a

single uprightand a bull's head, reconstructed by Woolley for the
British Museum. But R. D. Barnett has now shown how a study of
its actual remains in WooUey's photographs and notebooks
90 A lyre from the tomb of

Meskalamdug at Ur, with the
head of a bearded bull in gold
and ornament of other materials.
An instrument of this sort
appears in the 'peace' scene on
the 'royal standard' (ill. 87 top
right). It is held at waist height
by a standing man, accompanied
by a singer. Ht 122 cm
131
revealed the components of two separate instruments: the

sounding box of a conventional lyre and the single upright member
of a harp. ^54 Combined vîth the figured inlay of their frontal
panels and the gold or silver sheathing of their framev^ôrk, the
bull's head ornaments wîth their luxurious beards of gold or lapis-
lazuli add a fmal element of grandeur to a characteristic Sumerian
design.
one more example is required, we should remember the so-
If
called 'ram caught by its horns', which is in fact not a ram but a
9^ he-goat, symbolizing the generative vitality of the animal king-
dom. 5 5 jts head and legs are of gold, its belly of silver, its fleece part
'
lapis and part shell. The stylized plant against which it stands is also
sheathed in gold. This object is one of a pair which were taken to be
llIN
ornamental supports for some sort of vessel or cult-object.
Pottery
Painted pottery with geometric and figured designs reappears for

the last time in Sumer during the first Early Dynastic phase, espe-
92 cially at the Diyala sites. This is the 'scarlet ware', which retained
some characteristics of the fine polychrome pottery that preceded
it thejemdet Nasr period. ^^^ Technically, however, the decora-
in
tion is less competently applied. The designs are in bright red paint,
outlined in black over a buff or yellowish slip ; but the paint is
friable, lacking the 'soapy' burnish of its earlier prototype. The

commonest form of vessel is a broad-shouldered jar, with a shallow
neck and one or more triangular 'lugs' beneath the rim. The
painted designs are in many ways unique. Filling the panels into
which the surface of the vase is divided are groups of human and
animal figures. In one example from Tell Agrab, three naked
women are shown beating tambourines for the benefit of a tethered
animal which appears to be a sacred bull. Other animals,
conspicuously male, are of the antelope type with twisted horns. A
tme example from Khafaje, now in the British Museum, has a more
ambitious design, in which a wheeled chariot is represented. I
After the brief appearance of these attractive vessels in the Early
Dynastic I and II phases, painted decoration on pottery seems to
have been completely abandoned, nor does it appear again at any
period in southern Mesopotamia. In the second and third Early
Dynastic phases, painting is replaced by incised ornament in a wide
variety of forms which, in combination with the changing shapes
of the vessels themselves, provide useful criteria for dating
purposes.
In southern Iraq, the production of 'scarlet ware' seems to have
been confined to the Sumerian provinces east of the Tigris,
extending to the present frontiers of Iran. It is of some interest to
91 He-goat in composite remember that to the north also, in an area reaching from Nineveh
materials: gold, silver, lapis- to the confines of Syria, the final phase of the 'painted pottery era'
lazuli etc., on a wooden core. was marked by the appearance of a new and remarkable form of
One of a pair found in a royal
design. This was the elaborately decorated ware first found by
tomb at Ur, it symbolizes the
fertility of plant and animal life
Mallowan in his deep sounding beneath the greatest of all Assyrian
and was used as a support for mounds, and since known as 'Ninevite Five'. " He has described'
some other object. Ht 50 cm these vessels as 'tall fruit-stands with pedestal bases, high-necked
132
133
vases with angular shoulders and ring- or pedestal-bases, carinated

bowls and other types'. He adds: 'Many of these vessels are
overcrowded with designs, especially the bigger ones which depict
long-necked giraffe-like goats, waterfowl, fish and many mon-
otonous geometric patterns'. With its deep purple designs on a
neutrally coloured ground, this pottery seems unrelated to the
'scarlet ware' of the south, with which it is now known to be
contemporary. '5^ Here in the north also, in the phase which
followed, it was gradually replaced or accompanied by unpainted
pottery whose quality depended upon the use of fme clay and the
elaboration of incised ornament. Current excavations at Eski Mosul
and elsewhere are throwing new light on this subject. See Post-
script, p. 231.
\\m
92 'Scarlet ware' of the Early

Dynastic I period, developed
from the polychrome pottery of
Jemdet Nasr. The method of
painting is here technically
inferior,though ambitious figure
designs are sometimes attempted
134
Chapter Seven
The Dynasty of Akkad and the

Sumerian Revival
The term 'Pre-Sargonid', whichsometimes apphed to the Early

is
Dynastic period, serves in emphasize the importance of the

itself to
event which brought that period to an end. Sargon-of-Akkad's

accession to power in about 2370 bc, and his rapid subjection of the
old Sumerian city-states, resulted for the first time in the temporary
supremacy of a Semitic element in the population of Mesopotamia.
The transition from Sumerian to Akkadian rule must itself have
portended a revolution in political ideas. Its magnitude will become
apparent if we recollect for a moment the unchanging pattern of
Sumerian history during the preceding centuries and the peculiar
conventions by which it was dictated. These latter are indeed
implicit even in the formal statements of which the king-list is

composed, and they can be spelt out in a few words. The primary
assumption here is that the land of Sumer was an entity made up of
a number of city-states and that only one of them at any one time
was supreme over the others. The overall kingship remained in
this one city until, by force of arms it was removed to another.
Also, each city had its own patron god, who owned the state; but
these tutelary deities remained unaffected by inter-state conflicts.
They were not held responsible for the rivalry of kings who were
their 'agents', though disputes among themselves were equally
frequent. Finally, although the cities were continually fighting,
either for supremacy or for adjustments of their frontiers, there
remained among their rulers a strong sense of pride in the solidarity
of 'the land' itself. And here one must remember that Sumer was
regarded as a finite entity. The dozen-or-so city-states of which it
was composed occupied a restricted territory (hardly larger than
the modern Irish Republic), with northward extensions reaching
no further than adjacent cities in north Syria, which shared the same
culture. Sumerian ambitions did not reach beyond these limits,
remoter lands being of little interest except where trading was
concerned.
With the foundation of an Akkadian kingdom in the north, all
this was changed. In the dynamic authority and statesmanship of
its first ruler a new factor became apparent. Unhampered by
Sumerian tradition and the limitations which it had imposed on

earlier kings, Sargon's own conception of monarchy was absolute
and his ambition pointed beyond the bounds of an united Sumer
and Akkad. It may be well, therefore, before returning to the
archaeological record, to summarize such facts as are known
regarding the origin and history of the Akkadian dynasty.
135
The Dynasty of Akkad and the Sumerian Revival
\\m
93 Life-size head in bronze of an

Akkadian king, perhaps Sargon
himself (2370-2316 bc), found
out of context in ruins of the
Ishtar Temple at Nineveh by
Campbell-Thompson and
Mallowan. Note the refinement
of modeUing characteristic of this
period. Ht 30 cm I
Semites in Mesopotamia
First, then, a word must be said about the Semites, whose presence
in Sumer at a much earlier date has already been inferred from the
linguistic peculiarities of certain traditional names of Mesopot-
amian cities. '^9Regarding their country of origin, the one-time
conception of the Syrian desert as a centre of diffusion for nomadic
peoples has first to be discarded - if only because the greater part of
it almost certainly remained uninhabited from the end of the
Palaeolithic period until late in the 2nd millennium bc, when the
camel first came into use for desert transport. Lacking the mobility
of the modern Bedouin, nomadic herdsmen of those early days
must have confined their movements to the peripheral grasslands,
adjoining the territories of more settled agricultural peoples, whose
society they gradually infiltrated. Certainly this was the case on the
Mesopotamian side of the Syrian desert, and here there is some
evidence to indicate the direction from which the infiltration took
place. Semitic personal names appear in written texts throughout
136
the Early Dynastic period, and their geographic distribution

suggests that, whereas Semites were in a minority in the most
southerly Sumerian cities, their numbers increased as one moved
northward. In Mari and Ashur they already accounted for the
greater part of the population.
In the case of the Akkadians, their nomadic origin (if any) seems
to have been quickly forgotten, and in the early texts there is no
suggestion of the Sumerians having either opposed their intrusion
into the country or resented their presence. It seems that, by the
time of Sargon's accession, the central part of Mesopotamia, from
the region of Nippur northward to the Hit-Samarra Line,
including the Diyala district, had long been known as the 'Country
of Akkad'. For this reason the name Akkadian came to be applied to
the non-Sumerian peoples of Mesopotamia generally. Culturally
the most conspicuous distinction between the two ethnic groups
was a linguistic one. The Akkadians had retained their own Semitic
form of speech and, for purposes of writing, were compelled to use
an awkward adaptation of the old cuneiform script. Nevertheless,
once it was perfected, this written form of Akkadian was destined
to become the lingua franca of the Near East where commercial and
diplomatic correspondence was concerned. For the rest, Sargon's
subjects continued to share the religious beliefs and social practices
of the Sumerians.
Sargon and his Successors
The legend of Sargon's humble origin became a familiar theme in 95

the literature of later times : the upbringing of an orphan boy by
benevolent peasants and his employment as cup-bearer to a king of
Kish whom he eventually replaced upon the throne. These were
prehminaries to a spectacular military career, which started with his
defeat of Lugal-zaggesi of Uruk, temporarily the paramount ruler
of all Sumer. Having then subdued the cities individually and
'washed his weapons in the Lower Sea', Sargon founded a new
capital at a place called Agade. Unfortunately this city is one of the
few important political centres of Mesopotamia whose site has not
yet been located. It seems in any case to have been well to the north
of the main Sumerian enclave, though still accessible by river to
sea-going vessels. One notices that at this point there is some
evidence of discrimination in favour of the Akkadian element
among Sargon's supporters. Akkadian governors were installed in
the other Sumerian and the Sumerian language ceased to be
cities
used for administrative purposes. On the other hand, much time

and energy was devoted to the restoration or rebuilding of old
Sumerian religious monuments, and one remembers for instance
that Sargon's own daughter became priestess of Nannar, the moon-
god of Ur.
For Sargon, as we have said, the domination of Sumer was no
more than a preface to the extension of his conquests beyond the
natural frontiers of Mesopotamia. His first venture of this sort
carried him eastward into Elam, where he defeated the combined
forces of four rulers, led by the King of Awan, and estabhshed a
viceroyalty in the city of Susa, which from now onwards acquired
137
a new political status. His second expedition followed the course of

the Euphrates into northern Syria where, according to his own
account, 'the God Dagan gave him the Upper Region'. This would
imply that he gained access to the cedar forests of the Amanus
Mountains and to the silver mines of the Taurus. Later expeditions
assured the allegiance of settlements around or beyond Nineveh,
and there is a famous text, known as the epic of 'The King of
Battle', which shows the king advancing deep into the heart of
Anatolia, to protect his own 'merchants' frOm the exactions of a
local rujer, described as the King of Burushanda, somewhere in the
region south of modern Kayseri. Also, accepted with some reserve,
is an account of an expedition by sea, which carried him as far as the
southernmost extremity of the Arabian Gulf, and some credit is

given to his claim to have crossed the 'Sea of the West', to reach
Cyprus and Crete.
It is of this sort led to the creation,
clear that successive enterprises
for the first time of a Mesopotamian empire, in the
in history,
political sense, facilitating the commercial exploitation of countries
hitherto almost unknown. Apart from the new sources of timber
and essential metals which the opening of northern routes made
accessible, and the many commodities unobtainable from the east,
by way of Elam, the possibilities of seaborne trade in the Arabian
Gulf became increasingly apparent. We are told for instance that,
in Sargon's time, 'the ships of Dilmun, Magan and Meluhha were
moored at the quayside in front of Agade'.^^° The first of these
names, identified with the island of Bahrein, is already mentioned
in texts of the Early Dynastic period as a centre of commerce; but
if Magan can be recognized as Oman or the Makran coast, Meluhha
must be taken to be further afield and may well have served as

a link with the contemporary civilization of the Indus Valley. By
trade, at least, the Akkadian empire had reached the limits of the
known world.
Sargon's reign lasted for fifty-five years, but before he died, he
found it necessary to repress the first of a series of revolts, which also
darkened the reigns of his successors. Of his sons, Rimush was killed
in a palace revolution, but Manishtusu and to a greater extent his
grandson, Naram-Sin, were able to continue the aggrandisement
of the Akkadian empire. Naram-Sin seems to have been a ruler of
the same calibre as Sargon and like him became a hero of legend.
His long reign (2291-2255 Bc) was occupied by a series of military
operations, some of which are reflected in his own surviving
monuments. In the north, for example, a campaign against a Hurrian
king was commemorated by a royal relief, carved on a rock-face at
Pir Hussein, near Diyarbakir (now in the Istanbul Museum).
Another rock-sculpture at Darband-i-Gawr in northwest Iran
(perhaps post-Akkadian) records a victory over the Lullubi, one
of the tribal peoples of Luristan, who presented a continual threat
to the Mesopotamian frontier ;^î
and another similar victory is
celebrated in the design of the famous 'Stela of Naram-Sin', dis-
covered at Susa and now in the Louvre.
As for written records, references to the Akkadian kings might
be expected in the archives of Ebla a city once sacked by Sargon's
:
armies (see Postscript).
138
But, in the end, was the Guti or Gutians, neighbours of the

it
Lullubi in the north, who

effectively overran the homeland of the
Akkadians. This happened in the reign of Naram-Sin's successor,
Shar-gali-sharri and, since their onslaught coincided with a
concerted uprising of the old Sumerian states, the centre of
government in Agade was destroyed and the empire disintegrated.
The political chaos which ensued must have been considerable for ;
the compilers of the king-list content themselves with the rhetorical

question, 'Who was king? Who was not king?' The fact remains
that for the greater part of a century, Mesopotamia was ri;led by
semi-anonymous barbarians, who have left few monuments or
comprehensible inscriptions.
Archaeology
We have mentioned earlier in the present chapter the energy which

the Akkadian kings devoted to the rebuilding of temples and
sanctuaries in the old Sumerian cities. With this in mind, it is
curious to observe how few traces of their work have actually been
revealed by excavations at the sites themselves. One explanation to
be considered is that, in almost every case, the process of rebuilding
had in fact to be repeated a couple of hundred years later by the
Third Dynasty kings of Ur who, perhaps intentionally, concealed
all evidence of their predecessors' accomplishments. At the same
time, it has to be remembered that these sites were some of the first
to be excavated, at a period when the recovery of written texts and
other removable antiquities took precedence over the study of
architectural remains. The shortcomings of these early systems of
excavating no longer need to be emphasized. Their relevance in the
present context may however be illustrated by some selected
examples.
Buildings
H. V. Hilprecht, excavating at Nippur in 1 899-1900, discovered
that the great Third Dynasty ziggurat covered the remains of an
earlier one, founded by Naram-Sin of Akkad, but its investigation
he considered to be impracticable. ^^^ Vincent Scheil, at Sippar
(Abu Habba) in 1893, encountered a great religious temenos,
originally founded by Sargon of Akkad, but could trace and record
only part of the buildings which replaced it a thousand years
later. '^3 £ ^ Banks, at Bismaya (Adab) in 1903-4, during his
search for tablets or statues, noted the existence of a palace, private

houses and a cemetery, all of the Akkadian period, but made no
records of them. ^^^ gy 1923-33, things were a little better, when
Ch. Watelin, working in the Ingharra mound at Kish, could
distinguish repairs and additions made to an older ziggurat in
Akkadian times, and record contemporary graves in the 'A'
Cemetery. '^5 ^t Ur, any traces of buildings dating from the
Sargonid period were deeply buried beneath the ruins of later
times; but Woolley was at least able to study and minutely to
record as many as 400 Akkadian graves. '^^ It remained for other
British and American excavators in the 1930s to throw light on
some peculiarities of Akkadian architecture.
139
140
Tell Asmar 94 Plan of the 'Akkadian Palace'
at Tell Asmar, with its royal
Reference has already been made to the discovery at Tell Asmar in
apartments, elaborate drainage
the Diyala region of a large and well-planned residential building
and separate wing for women.
dating from the Early Dynastic III period. At a slightly later date it The adjoining Abu Temple
had been rebuilt on a much more pretentious scale, with walls over (bottom left) was also rebuilt at
2 m thick and an overall dimension of 73 m. '^^ This so-called this time 2350 bc) with a
{c.
double sanctuary. (Lloyd 1933)

'Northern Palace' was adapted, like its predecessors, to an irregular
shaped site, and an analysis of its plan suggests its division into three
distmct units. Centrally placed, one sees a suite of major residential

and reception rooms, whose function is easy to infer. Annexed to
this in the south, and accessible from it by a narrow doorway, is a
self-contained unit with its own courtyard; and one may recollect
that the contents of its surrounding rooms (mirrors, ornaments and
toilet accessories), implied its occupation by women. Lastly to the
north, and separated from the central by a system of
suite
courtyards leading to the main entrance, there was a third unit with
service accommodation. But perhaps the most curious feature of
this building was a range of variously shaped rooms on the east side,
with built-in installations of kiln-baked bricks. These appear to

have served some purpose involving the use of water, since each of
them had a drain leading to a brick-vaulted sewer in the narrow
alleyway outside the building. One notices that there are indeed
small lustral places or toilets annexed to larger rooms at other points
in the plan; but these longer chambers to the east seem to have had
some other function, which has led one authority even to suggest
that the building has been wrongly identified as a 'palace'. ^^^
Chronologically it is dated by brick-shapes and other evidence to
the earlier years of the Akkadian epoch.
Khafaje and Brak

There was other evidence at the Diyala sites of extensive
occupation during the Akkadian period. Near the Northern Palace
at Tell Asmar, private houses covering a wide area were exposed
95 Foundation plan of the

'Palace of Naram-Sin' at Tell
Brak, a military stronghold on

the northern frontier of
Mesopotamia. (P. P. Pratt after
Mallowan)
I4.I
and carefully studied. ^^ At Khafaje, in an outlying part of the site,

'
an enclosure wall had been built around a group of Akkadian

buildings, of which only the foundations survived; '^^ and there
were traces of an Akkadian occupation directly beneath the surface
at Tell Agrab. But one of the most impressive and historically
significant buildings dating from this period was discovered by
Mallowan at Tell Brak, '^' a little beyond the modern Syrian
frontier in northern Mesopotamia. This again has been called a
pj 'palace', but would rather seem to have been a military outpost or
fortified entrepot, occupying a strategic position on the trade-route
to southern Anatolia in the time of Naram-Sin. Built upon a site
overlying the prehistoric 'Eye Temple', referred to in an earlier
chapter, only its foundations had in part survived but from these ;
ill'M'
the plan could be restored of an approximately square building,
with immensely thick outer walls and a maximum dimension of
over 100 m. The plan itself, as one would expect, consisted largely
of long storage chambers, grouped around a system of open courts.
Plundered after the fall of the Akkadian empire, and destroyed by
fire, it had been replaced by a less substantial building in the time of
Ur-Nammu (211 3-2096 bc).
Sculpture
Well-preserved examples of Akkadian sculpture are hardly more
plentiful than surviving buildings of the period, a fact which must
be deplored if one considers the remarkably high quality of their
achievement. Outstanding in this respect are two conspicuous
monuments which chance circumstances have preserved for us.
One is Akkadian king, discovered

the life-size head in bronze of an
out of context in the Assyrian ruins of the Ishtar Temple at
Kiiyiinjik (Nineveh) and now in the Iraq Museum. ^^ Mallowan, '
who had the good fortune to find it, provisionally identified it as a

portrait of Sargon-of-Akkad, perhaps dedicated by his son
Manishtusu, whose name is recorded as founder of the temple. The
king is depicted wearing an ornamental coiffure, whose style
resembles the wig-helmet of Meskalamdug or that worn by
Eannatum in the 'Stela of the Vultures'. On the other hand, his
moustache and divided beard have a new air of sophistication and
the modelling of his face shows a notable improvement on the
work of Sumerian sculptors. One of his eyes has been damaged
by the forcible removal of its valuable inlay.
The second most important monument of the Akkadian period
is the 'Stela of Naram-Sin', found by de Morgan at Susa, whither it
96 The Stela of Naram-Sin,
King of Agade (2291-2255 bc), had been brought by the Elamite king Shutruk-nakhkhunte as part
celebrating a victory over the of 'the booty of Sippar'.'^^ The purpose of this relief is to
Lullubi and carried as booty to commemorate an Akkadian victory over the Lullubi tribesmen. In
Susa by an Elamite ruler. In
a composition skilfully suggesting a campaign among wooded
mountainous and wooded
mountains, one's eye is drawn to the dominant figure of the king,
country the Akkadian monarch
is depicted at the head of his
standing high above his troops and protected by the symbols of his
troops protected by the symbols gods. Wearing a horned headdress to signify his own divinity and
of his deities. The expressive carrying a bow, he tramples the enemy beneath his feet.
freedom of the design and its fine
Aesthetically, this is a magnificent design, and would alone serve to
carving shows a notable
improvement on early Sumerian emphasize the creative abiHty of Akkadian artists. If further
relief sculpture. Ht 198 cm testimony is needed in this respect, it must be sought among more
142
143
fragmentary remains from Susa and elsewhere. Pieces for instance

survive in the Louvre of other diorite stelae, depicting scenes from
the triumphant aftermath of Sargon's victories over anonymous
enemies. ^"^ Had these been less firmly dated, one wôuld still have
'
attributed them to an earlier phase in the stylistic development of

Akkadian carving. With figures arranged in horizontal registers,
these reliefs retain many characteristics of their Sumerian
prototypes, though the kaunakes garment is less in evidence and the
nude bodies of defeated enemies show an improved understanding
of musculature. A striking advance in the mastery of such designs
does however appear to have taken place in the time of Sargon's
successors. For there are fragments of a similar stela in the Iraq
97 Museum, dating almost certainly from the reign of Naram-Sin,

||i«|.
in which one sees an ultimate refinement of relief carving,

comparable with that of contemporary Egyptian sculptors. '^^
Cylinder-seals
Some compensation for the paucity of Akkadian sculpture is
provided by the rich harvest of contemporary cylinder-seals,
recovered from graves and private houses at some of the sites to
which we have referred. The skill of 'Akkadian seal-cutters and
their talent for design set a new standard for glyptic art in
Mesopotamia.'^^ In the first place, some innovations are to be
noticed in their choice of subjects but more important than these is
;
an overall change in the principle of their designs and style of

carving. The aim of earlier artists had been to link their action-
related figures together in a continuous frieze. This is now
abandoned, and the new designs are contrived in such a way that
the seal's rotation produces a succession of self-contained tableaux,
often isolated from each other by empanelled inscriptions. The
figures themselves are also much changed. Generally larger and
more widely separated, the depth of carving gives higher relief to
the impression and much attention is paid to ornamental detail.
Unfilled spaces between them enhance the purity of their outline,
while emphasizing by contrast the elaboration of their modelling
and the felicity of the patterns which together they compose.
Amongst the mythical figures of earlier times, the 'naked hero'
and his companion the 'man-headed bull' are still to be seen in
99 combat with horned animals or lions; but the contestants are
separated and the animals often arranged antithetically, hke the
'supporters' in a heraldic design, with a central motif in between.
Mythical and religious scenes are also again popular, but must now
be integrated into the framework of a static design. They are
widely varied and of considerable interest. Apart from the more
conventional subjects of worship and ritual, reference can easily be
detected to the familiar myths of Sumerian literature. Once more
the sun-god appears in his boat, or the water-god with streams
97 Unprovenanced fragment of flowing from his body. There is the bird-man, Zu, who stole the
an Akkadian relief, showing 'tablets of destiny', Etana, who attempted to reach heaven on the
naked prisoners after a victory
: a
back of an eagle, and many other figures of legend, which have
rare example emphasizing the
been identified in the course of long studies devoted to the subject
high attainments of
Mesopotamian sculptors at this of these seals. '^^ Unique among them is an example dating from
period the Akkadian occupation of the Abu Temple at Tell Asmar, whose
144
45
war^WfF-^^Ti
146
;
imagery, characteristic of the Indus Valley civilization, established 98 Akkadian cylinder-seal. The
ornamental 'combat' motif is
one of the first chronological links with the contemporary cities of
^^
isolatedbetween vertical reeds
Mohenjo-daro and Harappa. '
and an empanelled inscription.
Both design and modelling here
Gasur reach a high standard
Mention should be made finally from which

of one further site
99 'Bull-man' and 'hero', each in
important material of the Akkadian period was recovered by an
combat with wild bulls, are
American expedition in the late 1920s. This was Nuzi in the arranged antethetically, with a
neighbourhood of Kirkuk, which became a city of some mountain/tree symbol between
importance under its Hurrian rulers in the 15th centu/y bc.
100 Mythical scene showing the
Excavations beneath the Hurrian palace area produced evidence of
sun-god Shamash, rising between
continuous occupation throughout the 3rd and'early 2nd millennia, two mountains, and a water-god,
during which time the place seems to have been known as Gasur. with other deities in attendance
Already in the Early Dynastic period a temple had existed here,
dedicated to Ishtar like the contemporary shrine at Ashur which it
closely resembled. Important finds in the vicinity of this building
included a large collection of tablets, throwing much light on
commerce and trade connections in the Akkadian period. Among
Old Assyrian texts also, were found references to mercantile estab-
hshments in Anatolia, which lent new substance to the Sargonid
legends associated with this subject. Other finds contributed to the
hitherto unfamiliar typology of domestic and military equipment
in theAkkadian period. Metal objects, seal-cylinders and pottery
were plentiful, and could afterwards be compared with those from
contemporary levels at Tell Brak, with which they had more in
common than with their counterparts in the south. '
^^
Gutians and Lagash
The Mesopotamia during the long interregnum which

situation in
resulted from the Gutian occupation is veiled in obscurity by the
rarity of written texts. But it seems at least unlikely that any
systematic control was maintained over the whole country. The
interlude was, as we know, finally brought to an end early in the
22nd century b c by a military insurrection, led by a ruler of Ur
yet, some sixty years before this occurred, our attention is already
drawn to the survival in the city of Lagash of a Sumerian ruler, who
had attained sufficient independence and wealth to restore its
temples and reorganize the irrigation of its territory. This was Ur-
Baba, whose successor, Gudea, has been ensured a prominent place
in Mesopotamian history, first, by the great number of inscriptions
dealing with his accomplishments which have survived and,
secondly, by his patronage of a school of sculptors in hard stone,
who have bequeathed to posterity some of the most striking
masterpieces of Mesopotamian sculpture. From Gudea's records,
we even learn of a successful military campaign against Elam and
the dedication of its spoils to his god Ningirsu, the embellishment
of whose temple, E-ninnu^ is described in much detail. '^°
De Sarzec
The modern 'spoils' of Lagash, Louvre, are the
installed in the
results of excavations conducted by the
at the site called Telloh,
French scholar-explorer Ernest de Sarzec during the final quarter of
147
IHl'
loi Fragments of a diorite

statue, now in the British
Museum, probably from Telloh.
The style of carving is well
adapted to the quality of the
material. The bald head is not
unusual in this setting. Ht
c. 73 cm
the last century. '

^
' To understand the sensation created by the
arrival of his finds in Paris, one must recall that they provided the
first substantial evidence of a civilization in Mesopotamia
antecedent to that of the Assyrians, and that they remained unique

in this respect until the resumption of excavations in southern Iraq
after the First World War. A brief summary of de Sarzec's
discoveries may not therefore be out of place. '^^
Telloh
Telloh is a very large site, near the Shatt-al-Hai canal, with a
maximum diameter of about 0.9 miles. Of the principal mounds
grouped near its centre (identified by the French alphabetically),
the most productive were Tell A ('Palais'), Tell K ('Maison des
and Tell V ('Tell des Tablettes'). Attracted to the site by the
Fruits')
findsof illicit diggers, de Sarzec's first two seasons of excavation
(1877-8) amply confirmed the importance of Tells A and K, where
he reported the discovery of a 'superb fragment de statue'.
148
foundation deposits with stone tablets and bronze figurines, two

and
large inscribed clay cylinders, a fragment of the 'Vultures' stela
a 'statue colossale'which he left in place. Early in 1879, during his
absence on leave, H. Rassam, representing the British Museum,
slipped in and made some soundings of his own, but was frustrated
by labour difficulties. A year later de Sarzec was back again and
began a third campaign, resuming his excavation in Tell A. This
time his success was phenomenal ('. . toute la collection des
.
grandes statues de Gudea et Ur-Baba, plusieurs morceaux de la stele

des Vautours, des statuettes en pierre, des figurines de bronze des
vases etun grand nombre d'inscriptions. Butin enorme, qui arriva
en France en mai 1881').
Up till now and during his fourth season, de Sarzec had been
dealing with relics of the post-Akkadian phase in the history of
Lagash. In his fifth and sixth campaigns, having transferred his
activities to Tell K, he became involved with the remains of an
earlier period. Inscriptions were found mentioning the names of
governors of Lagash - Ur-Nanshe, Eannatum, Entemena,
Urukagina - who had ruled the state in the Early Dynastic period,
and an outstanding find at this time, bearing the name of
Entemena, was the now-famous 'silver vase', with its finely
engraved symbolic design. Four further campaigns were to follow
before de Sarzec's death in 1901 but, after the discovery in 1893 of
;
the source (Tell V) from which illicit diggers had been obtaining
large quantities of cuneiform tablets, the French excavators became
primarily occupied in attempts to prevent the total pillage of their
site. Unfortunately, owing to de Sarzec's failing health and other
circumstances, there were long periods when it remained

unguarded and the robbers were consequently given free rein. The
number of tablets which found their way into the markets of
Baghdad during those years have been computed at between
35,000 and 40,000, as compared with the 3,800 which were
recovered by the French from their excavations.
102 Diorite statue found near

Gudea
Telloh, probably depicting
The statues of Gudea and his son Ur-Ningirsu, seated or standing Gudea himself, to judge by the
and sometimes more than half life-size, are all identified by characteristic headdress. The
inscriptions. They are carved from boulders of hard diorite, quality of these sculptures is
brought by ship from Magan at the southern end of the Arabian partly dictated by the hardness of
the stone, which will take a high
Gulf. The exposed parts of the body show a refinement of
polish. Ht 105 cm
modelling well suited to this material and the sculptor's treatment
of his subject conveys an impression of strength in repose. Beardless
and sometimes bald, the heads are set low in the shoulders, while 101-2
the eyes profit from the absence of coloured inlay. '^^ While
admiring the masterly design of these sculptures, one is compelled
to regret the loss of the temple in which they stood, and of other
rich furnishings which Gudea records having dedicated to his god
Ningirsu. During his excavations, de Sarzec paid httle attention to
architectural remains. After his death, however, the work at Telloh
was resumed and continued intermittently until 1909 under the
direction of Gaston Cros, who seems rapidly to have acquired the
rudiments of the wall-tracing technique. One gathers from A.
149
\ V
.
\
• \ AGirsu (Telloh)
— -^
^V
—
A.,
AUruk
^..^Â
<
Larsa
V--^îk_Lagash (Al-Hiba)
\\ XXANina(Surghul)
V.
^
^
1^
103 Map showing the relative

positions of Telloh and
in the ancient state
Ebrahim after W. S.
Al-Hiba
of Lagash. (S.
Thomas)
3pm/
sokm
^^'
AEridu
^s^./-^^^^
"X
Parrot's summary of his reports that he did much to clarify the
topography of the site. In Tell A, beneath the ruins of a palace built
by an Aramaean ruler in the 2nd century bc, he located what he
considered to be some remains of the Ningirsu temple and, in Tell
K, a fortification built by Gudea. Among his fmds elsewhere (actu-
ally purchased) was a headless statue, subsequently re-united in the
Louvre with the fragment till then known as the 'Tete-au-Turban'.
Al-Hiba
More than seventy years after de death, in 1972
Sarzec's
extraordinary discoveries were made duringsurvey undertaken
a
by Th. Jacobsen and F. Safar, which led to an excavation sponsored
by the Metropolitan Museum and University of New York, under
the leadership o^V . E. Crawford, at a site called Al-Hiba, 15 miles
southeast of Telloh. A variety of Sumerian inscriptions were found
on tablets,stamped bricks and other objects (some of them
associated with a temple building). Their content provided
103 indisputable proof that Al-Hiba was in fact the site of the ancient
city of Lagash. Further excavations may throw new light on this
paradoxical situation; but for the present it is only possible to
conclude that the city discovered by de Sarzec at Telloh and now
identified by its ancient name, Girsu, '^"^ was a secondary political
or religious centre in the state of Lagash.
The Third Dynasty of Ur
The Sumerian revival itself must be thought of as having begun in

about 2120 BC, when Utu-hegal, King of Uruk, initiated the first
He was supported by the

large-scale revolt against the tribal rulers.
Sumerian governors of several other cities, and one of these, Ur-
Nammu, seems to have replaced him after his first military
successes. It was he who completed the liberation of Sumer and
founded the great Third Dynasty of Ur, thereby inaugurating what
is sometimes called the Neo-Sumerian period. For a little over a
century, under this king and his four successors, Ur, like Agade,
150
: :
was the capital of an empire; here as elsewhere much time was

devoted to ambitious building programmes, which have left their
mark on the cities in the form of ziggurats, temples and palaces. For
it may be well to concentrate on Ur
our present purpose therefore,
whose site (Tell Muqayyar) was so thoroughly excavated by
itself,
WooUey between the years 1922 and 1934. '^^
Buildings at Ur
Woolley published an overall plan of the city of Ur as he found it,
with details of the fortifications partly restored. '^^ Represented
here are of course the city-walls by the Neo-Babylonian
as rebuilt
kings in the 6th century bc; but there no reason to suppose that
is
their outline differs greatly from those built in the time of Ur-
Nammu fifteen centuries earlier. The town which they enclose takes
the form of an irregular oval with a maximum dimension of almost
exactly 1200 m, and in Ur-Nammu's time was already surrounded
by a wall and a rampart. Woolley describes it as follows
The rampart was of mud brick with a steeply sloping outer face the lower
;
part was in fact a revetment against the side of the mound formed by the
older town, but the upper part of it extended inwards over the top of the
ruins to make a solid platform. Along the top of this ran the wall proper
. . .
built of baked bricks. .This massive fortification was further

. .
strengthened by the fact that the river Euphrates (as can be seen from the
line of its old bed), washed the foot of the western rampart, while fifty
yards from the foot of the eastern rampart had been dug a broad canal,
which left the river immediately above the north end of the town. On
three sides therefore Ur was ringed by a moat and only from the south
could it be approached by dry land.
Itwas presumably from this side that the walls were assaulted by an
Elamite army in 2006 bc when the city was destroyed. For
Woolley adds
... of Ur-Nammu's wall not a trace remained. We would come on
examples of very large bricks specially moulded with the king's name and
titles, reused in some later building, but none of them were in situ. Just
because the defences of Ur had been so strong, the victorious enemy had
dismantled them with special care.
The Ziggurat
A second heavily buttressed enclosure wall surrounded the sacred
temenos in the northwest part of the town. In WooUey's plan of
this precinct one sees that it was considerably enlarged in the time of
Nebuchadnezzar, whose new wall passed across the ruins of the

Third Dynasty mausoleum, which had been sited just outside the
old temenos, near the earlier Royal Tombs. Of the buildings inside
the original enclosure, by far the most conspicuous was the colossal
ziggurat, built by Ur-Nammu and completed by his son Shulgi.^^"^ 1
04
This remarkable monument, which is better preserved than any
other of its sort in Mesopotamia, was studied and recorded with
great care by Woolley, whose well-known perspective recon-
struction of its original appearance has recently facilitated the
partial restoration of its ruins by the Iraq Government. It is a soHd
structure, built in three 'stages'. The core is of mud brick, probably
laid around and over the ruins of an earlier tower, and it was faced
151
fM"
104 The Third Dynasty with a skin of baked brick, set in bitumen, about 2.4 m thick. The
ziggurat of Ur-Nammu at Ur,
lowest stage, which is best preserved, measures 61 x 45.7 m at
convincingly reconstructed by
Woolley except for the form of ground level and is about 15 m high. The initial approach to the
thesummit temple, which is summit was by a triple stairway, with three converging flights and
hypothetical. The overall squat tower-buttresses in the angles between them. Where they
dimension at ground level is
met, at the first terrace level, the remnants of four brick piers
61 X45.7 m suggested some sort of portico, which Woolley hypothetically
provided with a domed roof. The height and dimensionsof the two
upper stages were ascertainable, but they were also much denuded.
Woolley observed three peculiar aspects of the ziggurat
structure, for which he has offered tentative explanations. The
fabric of the tower consisted of sun-dried brickwork, reinforced
with thick layers of woven reeds at intervals of six or eight courses.
Penetrating into its core, he found that bricks and mortar alike were
hardened and discoloured by fire, a phenomenon which he
attributed to the saturation of the structure by damp (perhaps while
temporarily in ruins), and the consequent occurrence of internal
combustion in the decaying vegetable matter. Conversely he was
by the multiplication of 'weeper-holes' penetrating the
surprised
baked-brick outer shell of the lowest stage; and he wondered
whether the terrace above could have been planted with trees
which would have required irrigation. He had in fact found
carbonized tree-trunks near the base of the tower. '^^ Finally he
discovered a slight outward deviation in the line of the main facades
at pavement level. This he was tempted to compare with the subtle
distortion of perspective contrived by the builders of Greek temples
fifteen centuries later. Remembering the great weight of the tower
and the pliable quality of mud brick, others have mistrusted this
explanation.
The Terrace
With the exception of the mausoleum which we have mentioned,
all the major buildings originally founded by the kings of the Third
152
Dynasty within the area of the old temenos as it existed at that

fell
time. For the most part, they had been so frequently rebuilt, with
adjustments to their planning and arrangements, that the analysis of
their ruins required much skill and patience. Indeed, had it not been
for the convention by which, from Akkadian times onwards,
baked bricks were stamped with royal inscriptions, the task would
have been impossible. Ur-Nammu built his ziggurat on a raised
terrace, surrounded by double walls with intramural chambers ('E-
temen-ni-gur'). To this he added a second courtyard, with a
monumental gateway at temenos level, which he dedicated to
Nannar, patron god of Ur. Southeast of the ziggurat terrace was a
huge and heavily fortified building called 'Gi-par-u', dedicated to
Ningal, the consort of Nannar but also containing several minor
shrines, incorporated during its long and complicated architectural
history. This is probably to be regarded as the tieftempel or ground-
level sanctuary. A smaller, square building ('E-nun-makh'), in the
angle between the two ziggurat courtyards, has been variously
identified as a palace, a temple or a 'treasury'.
The main entrance to the ziggurat terrace in Ur-Nammu's time
was by a rather inconspicuous gateway in the eastern corner,
known as 'E-dub-lal-makh'. Its outer portico, looking southward
over the more public part of the precinct, contained a statue of
Nannar and was used by the king as a 'Seat of Judgment'. Later
rulers separated it from the terrace and made it into a court-of-law,
with its own subsidiary chambers grouped around it. Yet, almost
seven cenruries later, the Kassite king Kurigalzu still referred to it as
'The Great Gate, the \ncient One'. Of the remaining Third
Dynasty buildings, two only survived in their original form. One
of these, in the southeast angle of the old precinct, was 'E-kharsag',
clearly the residential palace of Ur-Nammu and his successors. The
other was the mausoleum, about which there is more to say.^^^
This formidable conglomeration of buildings comprised Ur-
Nammu's own tomb and those of his two immediate successors.
Below the contemporary ground-level were the burial-chambers
themselves with their corbelled brick vaults and long flights of steps
leading down to them. Above these were the remains of elaborate
and formally planned funerary chapels. See ill. 57.
The Mausoleum
One should say at once that the looting of these tombs in antiquity
has to be regarded as a major tragedy. Judging from the astonishing
wealth of the older dynastic tombs, which were otherwise
conceived on a much more modest scale, these first great imperial
monarchs must have been provided with immeasurably richer
grave-deposits and buried with even more elaborate ceremony. All
that is now left is the shell of the buildings with their vaulting
partially intact; and Woolley found no more than fragments of
gold treasure which was removed from them.
leaf, testifying to
Almost without exception, the tombs of subsequent rulers in

Mesopotamia have remained unlocated.
Regarding the vaulted chambers themselves, they had long 103
flights of stairs leading down into them and were roofed with
corbel-type vaulting composed of baked bricks. In estimating the
153
I /-**
"'
s '-^^
105 Entrance to the brick height of the chambers, WooUey became aware of a curious
tomb-chamber of Ur-
vaulted
circumstance. was evident that, in the case of Ur-nammu's tomb,
It
Nammu in the royal mausoleum
the pavement on which the burial was actually laid out seemed a
atUr
somewhat shoddy affair and left a space between itself and the
vault, in which there was only just room for a man to stand. He
then found that the stairway went down several metres beneath
this, to a proper pavement laid with several courses of baked bricks
set in bitumen. What had apparently happened was that, when the
tomb was actually constructed, the builders had not taken into
account the flood-level of sub-surface water; so that, when the king
died and the time came for the tomb to be used, it was found to be
filled with water to a considerable depth. The only remedy
therefore was to fill it up to above water-level with rubble and then

repave it. Evidently, since that time, the water-level had receded,
because Woolley was able to excavate all the tomb chambers down
pavement level, and, after shoring up the vaults with

to the original
timber, it became possible for visitors to walk' about in them.
Today, a more permanent reconstruction of the vaults has been
achieved and they have become a tourist attraction.
154
Where the Ur mausoleum was concerned, Woolley as usual

made an extremely thorough study of the surviving remains. He
was even able to record details of a temporary superstructure,
connected solely with the ritual ceremony of inhumation, which
was later replaced by permanent memorial chapels.
Temples and Sculpture

The period of Sumerian revival seems to have coincided with a
notable change in the planning of ordinary temples. The 'bent-axis'
approach of earlier times was now abandoned in favour of ^ more
obvious symmetry. The new plan, which was to survive with only
minor variations throughout the remaining history of Babylonia, is
to be seen in its simplest form at Tell Asmar (Eshnunna), where a
subject prince dedicated a temple to Shu-Sin, the deified King of
106 Tell Asmar (Eshnunna), cut-
Ur. '^° Tower-flanked portal and vestibule, ante-cella and inner
away plan and section of a
sanctuary were now aligned on opposite sides of a central court, to
temple (A) dedicated to Shu-
create a single vista which terminated in the cult-statue itself. Ante- Sin, deified King of Ur, and the
cellas could be duplicated and dependencies multiplied; but the adjoining 'Palace of the
cardinal sequence of features around which the plan developed Governors' with its private
chapel, in the time of Illushuilia
remained constant from then onwards. As for the cult-statue itself,
(2317-2283 BC). Vaulting over
upon which the liturgy of Sumerian ritual was focused, some clue the 'Great Hall'is not impossible
to its appearance may be sought in relief sculptures where deities are at this period. (Drawing by R.
depicted. A rare example is to be seen in the so-called 'Stela of Ur- Leacroft) 1, sanctuary; 2, altar; j,
great hall; throne-room;

Nammu', fragments of which were found by Woolley among the 4, 3,
courtyard; private court;
earliest remains of 'E-dub-lal-makh'. This was a monument 3 m ablutions; 8,
6,
palace chapel;
7,
9,
high, divided horizontally to depict scenes connected with the ante-chamber; 10, sanctuary; 11,
building of a temple. In the only undamaged register, the king is toilet
155
twice depicted in confrontation, first with a god and then a goddess
Ur-
(perhaps Nannar and Ningal), before whom he pours a hbation
107 Detail of the Stela of
Nammu found at Ur, over some sort of plant. Fragments of a terracotta vessel, similar to
reconstructed from fragments. that in which the plant grows, were found beside the altar in the
The King is instructed by seated Shu-Sin temple. In contrast to the secular dynamism of the
gods to build temple (perhaps
a
Sargonid reliefs, the tranquility of this religious scene is once more
the ziggurat). The 'Sumerian
owed much characteristically Sumerian. Similar subjects depicted on the seal-
Revival' to his
initiative. Ht 304 cm cylinders of this period have a correspondingly static quality.
!
Mr-
lt\,ul Lirt.,/ r
156
'
Chapter Eight
The 2nd Millennium bc
In the early years of the 2nd millennium, with the Sumerian empire
disintegrating and a number of new states becoming actively
independent, we are faced with an era of change in the political

anatomy of Mesopotamia, which may once more justify a short
^
historical digression. '
Conflicting States
Of the two peoples who were jointly responsible for the downfall
of Ur's Third Dynasty, the Amorites from Syria settled in Babylon
and other cities to the north, while the Elamites returned eastward
across the Zagros, leaving a governor of their own choosing in the
Diyala state, Eshnunna, to guard their interests in Mesopotamia. In
Sumer itself, meanwhile, throughout the whole of the 20th century
BC, one watches two states, Isin (modern Bahriyat) and Larsa
(modern Sinkara), contending for control of the ancient cities and
their sacred shrines. At first Isin is in the ascendancy, but its enemies
are many, including not only Larsa but Eshnunna with its Elamite
allies. Furthermore, by the end of the century, two formidable new
states have emerged from their previous obscurity to become
prominent in the political scene: Mari (modern Tell Hariri) and

Ashur (Qal'at Sharqat), situated respectively on the middle courses
of the Euphrates and Tigris. Also in about 1900 bc, when Larsa
appeared to be gaining the upper hand, a new dynasty was founded
in Babylon, whose sixth ruler, Hammurabi, was destined to
reshape the whole political pattern.
Of the more northerly states which we have mentioned above,
Mari is one at which we should take a closer look. Its ruins as we
have said lie on the west bank of the Euphrates, a little to the north
of the modern Syrian frontier. The Sumerian king-list attributes to
it the status of a city with a dynasty of rulers and sure enough, when
;
a French expedition led by A. Parrot began excavating there in
1933, their first find was an Early Dynastic temple, dedicated like
that at Ashur which it closely resembled, to the goddess Ishtar. By
the beginning of the 2nd millennium bc, the place had acquired a
special importance as a station on the great trade-route which
brought timber and metal from Syria or from the mines of the
Taurus Mountains. At the time when the First Dynasty of Babylon
was founded, its rulers were consequently rich and prosperous.
Their palace, which Parrot had the good fortune to find and the
opportunity to excavate, was an enormous building which will
presently be described, and it contained an archive of more than
20,000 tablets, revealing the city's history prior to its destruction by
Hammurabi in about 1760 bc. '^^
57
The 2nd Millennium BC
The Mari tablets first introduce us to several

obscure Semitic
kings, and then to one more lakhdunlim, early in
historical ruler,
whose reign the city temporarily lost its independence and was
annexed to the state of Ashur. Mentioned in this connection is one
of the first really notable Assyrian kings, Shamshi-Adad I,
concerning whom a great deal is to be learnt from the Mari archive.
The city of Ashur occupied a strategic position on a shoulder of
rock, overlooking the Tigris at a point well beyond the Hmits of the
alluvial plain. '
^^ Li]^^ was an important trading centre, in
Mari, it
contact both with the southern cities and with the sources of raw
I
materials in Anatoha. Between about 1940 and 1 800 b c it had even

j
I maintained a commercial colony or karum as far away as Kanesh

I (modern Kiiltepe) in Cappadocia. Under Shamshi-Adad its own
,
territory extended northward to include the uplands around
Nineveh; so its extension westward as far as the Euphrates now
created a formidable kingdom. From the Mari archive one learns in
great detail how this came to be administered by Shamshi-Adad
and his two sons, one of whom, Ishme-Adad, he established as
governor of Mari. He himself preferred to be on the move from
one minor city to another, and seems to have spent a good deal of
his time at a place called Shubat-Enlil, probably on the River
Khabur and now equated by some scholars with the site called
Chagar Bazar, where Mallowan came upon tablets dealing with his
financial affairs. '^'* Other contemporary letters in which the
Assyrian royal family is mentioned have been found more recently
at Tell Rimah, the site of a provincial capital in the Sinjar area. The
combined contents of all these texts have helped to create an
extremely clear (and sometimes entertaining) picture, both of
public administration and of personal relations between the ruling
families. For this reason alone they are invaluable documents.
The end of Shamshi-Adad's reign is less easy to reconstruct. His
southern garrisons seem to have been overrun by the armies of
Eshnunna and Elam, which then drove westward to the Euphrates,
and it may well have been in resisting this attack that the Assyrian
king lost his life. When becomes clear, Ishme-
the situation again
Adad has succeeded him
King of Ashur (178 1 bc) but Mari has
as ;
regained its independence and is ruled by lakhdunlim's son,

Zimrihm, who had been exiled to Aleppo. It was in Zimrihm's
time that final improvements were made to the palace which
Parrot excavated.
Zimrilim's political relations with the south, dependent on
ephemeral alliances with Babylon and Eshnunna, were for the time
being satisfactory. In the northwest, Syrian states such as lamkhad,
of which Aleppo was a principal city, Alalakh and even Qatnah in
the plain of Homs, all remained friendly. Only to the east of the
Euphrates at this time (c. 1800 bc), the country had begun to be
disturbed by the arrival of new immigrants. These were the
Hurrians a people of unknown extraction, neither Semites, nor
:
Indo-Europeans. Ever since the Third Dynasty of Ur, there had

been small centres in northern Mesopotamia whose rulers had
Hurrian names. In Zimrilim's time they were already to be found
at Carchemish, and later as far afield as Alalakh and the Orontes
Valley. Nevertheless, it is from the American excavations at Nuzi,
158
The 2nd Millennium B C
1 08 Head of a basalt stela

recording Hammurabi's code of
laws, carried away to Susa
during an Elamite invasion and
now in the Louvre. The King is
seen confronting the seated

figure of Shamash, god of
Beneath are 16 columns
justice.
of cuneiform inscription in
which the traditional laws of
Babylonia are definitively re-
formulated. Ht of top of stela,
71 cm
near Kirkuk, that our rather scanty knowledge of them is mainly

derived.
The Hurrians, in any case, were for the time being disunited and
presented no serious threat to thecities of upper Mesopotamia. In
the end was the accession of a great statesman to the throne of

it
Babylon which put an end to this peaceful interlude. In 1759 bc, the
armies of Hammurabi defeated a coalition consisting of Larsa,
Eshnunna, Elam and Mari. Two year's later Mari revolted, but was
then completely destroyed, while Assyria, now much reduced in
size, became a vassal of Babylon. As in the time of Sargon-of-
Akkad, the whole country was now again united under a single
ruler and an appropriate system of administration devised.
Hammurabi's achievements in this respect are reflected in his
famous law-code, inscribed on a basalt stela 3mhigh, which is to be 10(
seen in the Louvre.
159
. During the reigns of Hammurabi's successors, three unexpected

? developments foreshadowed the disintegration of his empire. In the
south a people inhabiting the so-called 'Sea-Land' at the head of the
j
Arabian Gulf, encroached upon the old Sumerian cities and created
j
from them a new kingdom. Secondly, from the northeast came a
I
host of Kassites, displaced from their homeland by the same
I
movement of Indo-European peoples which brought the Hittites
I
into Anatolia. And finally, in 1595 bc, the Hittites themselves
i under their king Mursilis I, Carchemish and
after sacking in turn
1 Mari, swept on down Babylon itself, which
the Euphrates as far as
they plundered and burnt, thereby putting an end to Hammurabi's
dynasty.
The Hittite raid on Babylon was one of the stranger phenomena
!
of Near Eastern history; for Mursilis was almost immediately
summoned home, to deal with enemies who were threatening his
own capital in Anatolia. It was accordingly the Kassites who
profited most, because they were thus enabled peacefully to take
over the government of Babylonia. They adopted the religion and
culture of Mesopotamia so thoroughly and unquestionably that,
for the next 360 years, one would hardly be aware that much of that
country was now being ruled by a dynasty of completely alien
kings. To the north, meanwhile, in the hilly country beyond
Nineveh, and west to the Euphrates, much the same situation had
arisen in regard to the Hurrians. Such rulers as they possessed had
now been replaced by a new, Indo-European aristocracy, whose
political acumen enabled them at last to acquire for their people a
national identity. Between Assyria and the Hittites there emerged a
powerful kingdom, thereafter known as Mitanni, with its capital at
Washshukanni (a site to the west of Nisibin which has not yet been
located). ^ = In the early years of the 14th century, one finds its kings
'
corresponding almost on equal terms with Egyptian Pharaohs of

the 1 8th Dynasty.
It was also during this century that Assyria, still with its capital at
Ashur, began to reappear as a major power. As one able ruler

followed another, its southern frontier was stabilized, while the
Mitannians, under pressure from the Hittites, soon began to relax
their hold on northern Mesopotamia. Finally, weakened by
internal disputes, their royal family lost authority and its army
ceased to be a match for that of Assyria. In 1250 bc, after several
defeats, the last Mitannian king was slain by Shalmaneser I and his
country became a province of the Assyrian state. Shalmaneser's
successor Tukulti-Ninurta I (1244-1208) conquered Babylon and,
during the remainder of the 2nd millennium one watches the
earlier stages in the creation of an Assyrian empire.
With these historical events in mind, it may be easier to
understand that the archaeology of the 2nd millennium can be
divided into four principal phases, each of which is, as it happens,
illustrated by the results of at least one major excavation. To begin
with there is the 'Isin-Larsa' period (c. 2020-1763 bc), ending with
Hammurabi's unification of Mesopotamia. This period is well
represented at the Diyala Asmar (Eshnunna), Ischali
sites: Tell
(Neribtum), and Tell Harmal (Shaduppum). Next there are the
reigns of Hammurabi and his successors, the Old Babylonian
160
period, ending with the fall of Babylon in 1 595 b c. For this we have
Parrot's wonderfully comprehensive finds at Mari. Thirdly, in the

south there is the time of the Kassite Dynasty (1595-1235 bc),
documented primarily by the Iraq Government's excavations at
Dur-Kurigalzu ('Aqar Quf). In the north this partly coincides with
a period of Hurrian-Mitannian domination, of which we shall fmd
evidence at Nuzi and elsewhere. And finally, there is the growth of
Assyrian power: the so-called 'Middle Assyrian' period, which will
bring us to recall the excavations made by the Germans at Ashur
itself
Buildings of the Isin-Larsa Period
Eshnunna ceased to be a vassal of Ur in 2027 b c and some years later

independent governor, Ilushuilia added a palace to the great
its first
temple which one of his predecessors had dedicated to Shu-Sin, a

deified king of Ur. ^^ ' To the palace he annexed a smaller religious
building, probably now dedicated to the local god, Tishpak. This
'Palace Chapel', like the older temple, adheres to the plan now
uniformly adopted in Babylonia, with entry, courtyard, antecella
and sanctuary all on the same axis. The palace itself is of greater
interest, because it takes its place in a long line of secular buildings
planned for the same purpose. ^^ Facing the square court is a broad
'
rectangular throne-room or audience chamber, with a staircase

leading to the flat roof near one end. It is separated by smaller
chambers from a 'great hall' (or bitanu court), perhaps used for
assemblies. We
shall see something of the same sort at Mari and,
more than thousand years on, in the standard 'reception suites' of
a
the Late Assyrian palaces. In the Eshnunna palace, remnants of the
state archive were found in the stair-well; but some of the most
important historical evidence was derived from the stamped bricks,
used during each successive reconstruction of the building, which
provided a complete genealogy of local rulers. (See ill. 106.)
We learn in this way that it was a later Governor, Ibiqadad II
('enlarger of Eshnunna'), who

immensely greater palace,
built an
facing the earlier buildings across a broad street. ^^ Of this only the ^
foundations have survived; and it may well have remained

unfinished. Yet here again one may recognize the standard
arrangement of court, throne-room and 'great hall', now much
enlarged and even duplicated. Elsewhere to the north, Ibiqadad's
son (confusingly called Naram-Sin), built a separate audience hall-
in this case differently planned, supposedly because he had 'assumed
the prerogatives of divinity'.
Nearer to the River Diyala itself is Ischali (ancient Neribtum),
where, 'in the year Rapiqu was sacked', Ibiqadad built a temple
which is perhaps the finest monument of the Isin-Larsa period. "'^ log
The building is a precise rectangle, measuring approximately
100 X 60 m, and incorporates three separate shrines, the largest
dedicated to Ishtar-Kititum, a local form of the great goddess. The
whole complex is raised 3 m above ground-level on a platform or
kisu, faced with kiln-baked bricks set in bitumen.Baked brick is
again generously used elsewhere, for instance in the construction of
a wide stairway, leading from the outer courtyard to that of the
i6i
109 The Ishtar-Kititum major shrine which stands at an even higher level, and for the three
Temple at Ischali (ancient
towered gateways. An unusual feature is to be seen in one of the
Neribtum in the state of
Eshnunna). Raised above
subsidiary shrines, whose sanctuary is set lengthwise on the main
its
surrounding buildings on a axis: a practice later adopted in Assyrian temples (German

terrace faced with baked brick, langraum). Elsewhere the hreitraum arrangement is maintained. A
itcomprised three separate striking reconstruction o^ the whole building is to be seen in a
shrines, of which the largest was
much-published drawing by the late Harold Hill. It makes an
raised even higher. A fine
example of Mesopotamian interesting contrast to Shaduppum (Tell Harmal), which is a small,
architecture in the Isin-Larsa outlying heavily fortified administrative station, on the outskirts of
period (2017-1794 Bc). modern Baghdad. ^^^ But it too was provided with no less than
(Watercolour by W. Suddaby)
four comparatively modest temples, the largest of which has outer
and inner gateways, flanked by the almost life-size figures of lions
in terracotta. Archives of tablets found in its administrative
buildings included, among other important texts, a code of laws
preceding that of Hammurabi.
Private Houses
Under this heading, a word should be said about ordinary

dwelling-houses of the Larsa period. These are well represented in
contemporary levels at the Diyala sites, but perhaps even better
illustrated at Ur, by WooUey's excavation in a residential area to
110 the southeast of the temple precinct. ^^^
Here there was a tangle of
narrow streets with modest houses on either side, and occasionally,
at important cross-roads where several lanes met, small rehgious
shrines occupying sites of their own. The houses themselves so
closely resembled those of any small town in Iraq early in the
162
mag. N
present century, that they hardly merit description: blank walls no A residential quarter of Ur
in the 20th century BC. A tangle
facing the street, rooms in two storeys opening on to an unroofed
of narrow lanes lead off wider
central court, and in one, of which Woolley drew a well-known roads. The richer houses have
reconstruction, a sheltered wooden gallery at first-floor level, two storeys of rooms, Hghted by
supported on posts. ^^^ Among their few were
distinctive features an open court, and closely
resemble those in the older cities
the graves of their occupants, often buried beneath the ground-
of modern Iraq. (After Woolley)
floor pavements.
163
;
Enlrance
Palace
III The great palace at Mari, as

completed by Zimrilim
(i 779-1 761 Bc). The royal
archives were found in two
rooms near the approach to the
king's apartments, where there
was also accommodation (a
school ?) for scribes. (After
Hawkes, 1974). a, outer
courtyard; b, audience hall (with
murals) ; c, ceremonial courtyard
(with murals) ; d, throne-room
e, inner sanctuary and hall
The Palace at Mari
1 1 1 The Mari palace had doubtless been the residence of Zimrilim's

father, lakhdunlim, and afterwards used by Shamshi-Adad's son
during the Assyrian interlude. But, as much of it was built of bricks
stamped with the name of ZimriUm himself, he must have
considerably enlarged and rebuilt it. It was, as we know, finally
destroyed when the city-walls were razed to the ground in the 35 th
year of Hammurabi (1757 bc). Nevertheless, when Parrot found it,
its 4-m thick walls remained standing in some cases as much as 5 m
high and the lintels of some doorways were still intact. ^^^ The
whole building measures approximately 200 x 120 m - twice the
dimensions of the Ishtar-Kititum temple. Much of the plan consists
of a multiplication of what is really the Babylonian dwelling-house
unit - a number of chambers grouped around an open court; but
these are subsidiary to the great central system of reception rooms
and religious shrines.
From the main entrance with its guardrooms, one passes into a
huge outer courtyard (a), of which the main feature is a three-
sided audience chamber, approached by a flight of steps (b),
thought by some to be a surviving feature of an earlier building.
But it is the inner, and slightly smaller courtyard (c), on to which
the main suite of reception chambers faces. It is composed of the
now-traditional elements: first a throne-room (d) with a pod-
ium facing the central doorway, and then, separated from it by
vestibules, a larger element corresponding to the 'great hall' (e).
But the latter now has a clearly religious function for, at one end,
;
steps lead up to a sanctuary and, facing it at the opposite end of the
164
;
hall, is a throne emplacement. As Parrot has observed, this gives the
whole central suite a non-secular character, and wê shall see that this
is borne out by the relics of sculpture and mural paintings which are
associated with it.
More simply decorated are the living apartments of the royal

family in the northwest corner of the building, whose outer walls
on both sides are enormously strengthened. The king's own
chamber and that of the queen were recognizable, and in the court 112
between, some sort of game had been marked out on the pavement.
Closer than one would expect to this domestic unit were two
chambers with rows of clay benches, identified as a school or offices
for scribes. Accessible from these were two of the store-rooms in
which archives were kept - originally arranged upon shelves. The
third and most important archive room was conveniently placed
between the inner and outer courts. Elsewhere, among the 200-or-
more rooms clustering around the central unit, ranges of other
store-rooms could be recognized, including one group centred
upon a chamber known to the excavators as the Cercle des officiers,
which was distinguished by its painted walls. Also near the
southeast corner of the building were two rooms composing a
112 Palace of Zimrilim at Mari
minor shrine, connected by a 'processional way' to the main toilet and terracotta baths in the
audience hall. royal apartments
165
/VWyWY
Mural Paintings
More must now be said about the mural paintings, which have
been the subject of much speciahzed study. ^^'^ As Parrot himself has
observed, they are of special significance, in that they clearly
illustrate 'the spirit of synthesis which animated the artists of the
middle Euphrates at this period': the curious combination of
Sumerian heiraticism, surviving from an earlier age, with the
distinctive naturalism of the Semitic approach to art. As he says,
these influences seem to have contributed to creating a new school
of decoration, which, had it been given time to mature, might have
1 1 3 (
Top) Mural painting on an
outer wall of the throne-room in produced one of the high artistic epochs of Near Eastern history.
Zimrihm's palace at Mari (V in The murals were painted directly upon a thin layer of mud plaster,
ill. III). The 'Investiture' scene in a manner which has been taken to suggest that the true 'fresco'
(above) shouts the King before
technique was understood.
Ishtar, 'taking the oath' with his
hand upon the divine emblem in
The largest and most ambitious composition, LTnvestiture du Roi
the presence of other deities. de Mari, was found, still in place, on an outer wall adjoining the
Below, two figures of the central entrance to the throne-room, in a part of the courtyard
'goddess with flowing vase' are
(c) which was protected from the weather by a roof or awning
in attendance. On either side are
supported on posts. The central design is composed of upper and
and other symbohc
'sacred trees'
The Mari scene has been
figures. lower panels. In the former, the king, draped in an elaborately
compared with the entrance to fringed costume and polos headdress, receives or touches a divine
an Assyrian temple at Khorsabad emblem held by the goddess Ishtar, whose raised foot rests upon the
(above), which again is flanked
back of a lion. Other gods and goddesses are in attendance. In the
by 'flowing-vase' figures and
palm trees. (S. Ebrahim
artificial
lower scene, facing each other, are two identical figures of the
after Parrot, 1958) 'goddess with flowing vase', wearing a flounced dress and holding
166
an ariballos vessel, from which streams of water, full of fish,
undulate around the picture. Both scenes are framed on either side
by tall panels with palm and other stylized trees, among which
birds, mythical animals and divinities are to be seen. This
composition has been ingeniously explained by one scholar as a
ceremony taking place in the cella and antecella of a temple, of the
sort which in later, Assyrian times was entered between 'flowing- i ij
vase' statues and artificial palms. ^°5 Whether or not this is so, the
decorative effect of such a design, with its ornamental border and
bright colours, must have been magnificent.
Another such painting adorned a wall of the audience chamber
in the outer courtyard, likewhich it may be of slightly earlier date.
Known to the excavators as Le sacrifice de I'eau by dufeu, it is a scene
of worship and libation, attended by monstrous figures of
Sumerian mythology: Tiamat of the Enuma Elish epic and
Shamash, 'rising behind his mountain to scatter the stars'. The king
(Shamshi-Adad I ?), again in his fringed garment, also appears in the
surviving fragment of a 'sacrificial scene', whose original position
in the building could not be positively determined.
We have noted above that, in the centre of the city at the
southeast approach to the palace, a cluster of minor temples had
already existed in Early Dynastic times, dedicated to such gods as
Nin-khursag, or more local deities like Ishtarat and Ninni-Zaza. ^'^^
Some of these still survived in the 1 8th century but they were now
;
dwarfed in size by the Temple of Dagan (a grain-god worshipped

along the middle Euphrates), to whose sanctuary a small ziggurat
was at some time annexed. Dedicatory tablets showed that this
temple was in fact build by Ishtup-ilum, a governor of the city in
the late Larsa period {c. 1 890 Bc), of whom a fine portrait statue was
found, not in the temple but in the 'great sanctuary' of Zimrilim's
palace. This and other Mari statues may therefore presently be
discussed together with the combined products of the Isin-Larsa
and Old Babylonian periods.
Tell Rimah
During the excavation of buildings to which we have so far
referred, information could be recovered regarding either
little
their facade ornament or method of roofing. In the 1960s, a British

expedition led by David Oates, and working at Tell Rimah, in the
Sinjar district west of Mosul, was able to throw much new Ught on
both these matters. ^^^ Here, where a largish mound already
covered the remains of earlier settlements, a local ruler in the time
of Shamshi-Adad I had built himself a city, surrounding it with
fortifications and creating from the mound itself an emplacement
for a remarkably fine temple. This building, as described by the 114
excavator himself,
was approached by a free-standing stair carried on vaults, and from its roof
further stairs or ramps led to a high terrace, perhaps surmounted by a
higher shrine. The whole, three- or four-tiered structure must have
resembled a ziggurat. The temple itself on its platform high above the city,
was laid out on the Babylonian plan and decorated in a style that also has
southern parallels, although as a complete system of ornament it is unique.
All the external and courtyard facades were adorned with engaged
167
5
1 14 Plan of a temple at Tell

Rimah in the Sinjar district, at the
foot of a small ziggurat (early 2nd
millennium bc). The facâdes are
elaborately ornamented with
half- and three-quarter- columns
of mud-brick, some moulded
palm
spirally or to represent
Remnants of compHcated
trunks.
mud-brick vaulting had survived
in this building. (After D. Oates,
1964). a, ziggurat; b, shrine; c,
antechamber; d, courtyard
1 1 Plan and section of 'pitched-

brick' vaulting on pendentives at
Tell Rimah, late 3rd millennium
rc'X TaWo
BC. A simpler form of 'pitched-
brick' vaulting was used 2000
years later in the great iwan
behind the Arch of Ctesiphon.
(After D. Oates, 1973)
100 cm
68
columns, set singly or in groups, 277 in all the 50 large columns were built
;
of carved bricks, laid in complicated patterns to represent spirals or palm-

trunks.-^^
But perhaps the more extraordinary feature of this building was the
prevailing use of mud-brick arches and vaulting, even for trivial 115
purposes where wood could have been substituted. ^°^ Com-
plicated vaulting systems, including the use of 'pitched-brick'
construction and even pendentives, were contrived with such
casual facility that a long and widespread tradition of building in
this manner seemed to be implied.
Sculpture
Of the Mari statues, the only one undoubtedly contemporary with

the palace is the 'flowing-vase' goddess, fragments of which were
found both in the 'throne-room' and in the courtyard outside.
(This must have been a fixture, since, by an arrangement of pipes, it
actually dispensed water.) Stylistically however, it is of little
significance, since its design in most respects suggests a traditional
treatment of this particular subject. More characteristic are the
several statues of local princes: Ishtup-ilum, found in the 'great 116
sanctuary'; Idi-ilum from the southeast shrine and Puzur-
Ishtar, which was part of a 'collection' of antiquities assembled 117
by a Late Babylonian king (and now in the Istanbul
116 Statue in black stone of

Ishtup-ilum, Governor of Mari,
from the palace of Zimrilim (c.
2200 Bc). Ht 152 cm
117 Statue of deified king

dedicated to a temple by Puzur-
Ishtar, Governor of Mari, found

in the 6th-century 'Palace
Museum' at Babylon. Like
Ishtup-ilum, he wears a toga-like
on contemporary
dress, also seen
statuesfrom Eshnunna, found at
Susa and now in the Louvre. Ht
175 cm
169
8
fjtn|
^^^r IdjH ^^^^J

1
^L^B^m ^1
^^^^^^^^^^HKt aH
H *^ti^^^H
1
mS^*^
s
Museum). These compare interestingly with the statues of two
Eshnunna governors, one of them seated, which were carried as
^^^^^^^
1 1
'
i«
B
Painted terracotta plaque
showing the goddess Lilith,
'Bringer of Death', winged and
booty to Susa. In a world where the crudity of the kaunakes was holding symbols of justice. Her
hardly remembered, these statues are dressed in a toga-like name is associated with owls,
garment, whose fme texture and restrained modelling are both in the Gilgamesh epic and
effectively contrasted with the obtrusive detail of ornamental theOld Testament {Isaiah,
XXXIV, 14). Ht 50 cm
fringes and the pattern of formally dressed beards. They are by no
means lacking in dignity. ^ ° '
119 Guardian Uons in terracotta
Other sculpture of the period is scarce and often of poor quality. from temple gateways at Tell
Where relief carving is concerned, the scene depicted at the head of Harmal (ancient Shaduppum), a
minor walled city in the state of
Hammurabi's 'law-code' stela is perhaps unique. The king, whose
Eshnunna. Reconstructed from
profileand headdress are familiar from fragmentary sculptures in- some hundreds of fragments
the-round, stands facing a seated god, with the emblem of divinity (c. 1900 Bc). Ht c. 63 cm
(justice?) in his hand, like Ishtar in the 'Investiture' painting. In the

words of Henri Frankfort, the scene 'conveys, not only a sense of
confrontation but of communication between the lord of justice
and the law-giver'. But there is at this time another form of relief,
modelled in terracotta. Of this, a fme example is the (privately
owned) panel showing the naked goddess, Lilith, with her 118
supporting lions and owls. ^ More common still are votive
'
'
plaques of baked clay, cast in an open mould, which were probably

sold to worshippers at a temple. Their subjects vary from the
ubiquitous 'mother-goddess', now grotesquely conventionalized,
to sensitively modelled figures of musicians, temple-women or
more trivial motifs such as a bitch suckling her puppies.
Corresponding to the fine terracotta lions from the temple 119
gateways at Tell Harmal, similar beasts of bronze, with inlaid eyes,
guarded the sanctuary of the Dagan temple at Mari. ^'^
171
The Kassites
120 Wall ornament from the In view of the fact that the Kassites ruled Babylonia for more than
temple of Karaindash at Uruk: a
four centuries, it is surprising to observe how little their national
rare example of Kassite
characteristics are reflected in the material remains of their
architecture (14th century bc).
Ht of brickwork, 198 cm occupation. These make it abundantly clear that, in religion,
administration and technical practices, they adopted and faithfully
121 The eroded ruin of a great maintained the age-old conventions of Mesopotamia, rebuilding
ziggurat at 'Aqar Quf, west of
temples and honouring the shrines of Sumerian deities. An almost
modern Baghdad and site of the
Dur-
Kassite capital called
unique exception in this respect is the temple of Karaindash at Uruk
Kurigalzu. The mud-brick core which, with its exaggerated corner buttresses, seems at first to be
isreinforced with layers of reeds sharply differentiated from the buildings around it. Even so, its
and heavy cables (Kassite elaborately decorated facade embodies one motif with which we
Dynasty, I5th-i3th centuries bc)
between its projecting pilasters are
are already familiar. Tall niches
filled with the alternating figures of Babylonian gods and
120 goddesses, holding in both hands the traditional 'flowing vase'. For
the rest, they provide us with a very early example of the
'moulded-brick' ornament which was later perfected by Assyrian
and Neo-Babylonian craftsmen, ^ ^ '
The grandiose new capital, which was founded by the

121 Kassites themselves and called Dur-Kurigalzu, is distinguished
from other Babylonian cities only by the nature of its chosen site. It
was built upon a low outcrop of soft limestone, at the northern
172
*'«!
^.2*'4
.'Hi.
extremity of the alluvial plain to the west of modern Baghdad,
where the eroded core of its great ziggurat (now partially restored)
still creates a conspicuous landmark. Iraq Government excavations
in 1942-5 exposed a complex of temples at the foot of the tower
and, further out on a peninsula extending into the neighbouring
^
flood-basin, the remains of at least four palaces. '
"^
Peculiar features
of the temple area were rectangular platforms faced with baked
brick, supporting the principal shrines. Inscribed pivot-stones and
tablets found in this area were mostly dated to the reign of a king
called Kurigalzu; but the stratification of the palaces, which had
been rebuilt at more frequent intervals, implied the existence of
one king at least bearing the same name.
earlier
One of these palaces ('H'),^'= unfortunately in part denuded
Iliri-
owing to the weathering of the mound, did show some signs of an
architectural arrangement perhaps characteristically Kassite. The
central courtyard was surrounded by multiple doorways, giving on
to long galleries reminiscent of cloisters. The reveals of these
doorways were decorated up to a height of about i with mural m
paintings, depicting processions of court officials.^ '^ They wore a
short-sleeved garment with a fringed girdle and a fez-shaped
headdress over their long hair in the Assyrian manner. It has been
suggested and is entirely possible that this figured ornament,
122 Kisske-type kudurru or

boundary stone, assigning land to
an official called Marduk-nasir.
The armed figure of a king(?)
is surmounted by religious
symbols. Ht 53 cm
123 Head in terracotta of a

bearded man from Dur-
Kurigalzu. The style recalls that
of the 1 8th Dynasty in Egypt.
Ht 4 cm
174
forming a 'dado' at the base of the walls, foreshadowed the use of

sculptured slabs for the same purpose in later Assyrian palaces. In
Kassite times, relief sculpture was mainly restricted to boundary
stones (kudurru), bearing the emblems of gods invoked as witnesses
of a land-settlement, or of a king who had authorized it.^'^
Another palace ('A') of a much earlier period, had one long
gallery, painted white, with triple rows of small plastered pedestals,
The many fragments of gold leaf and other valuables lying on the
pavement between these, strongly suggested its use as a treasure-
chamber. Later it had been replaced by ranges of brick-vaulted
repositories, probably intended for the same purpose. ^'^ It was
here, and in the stair-wells leading to the roof, that other valuable
objects had been scattered, perhaps at a time when the palace was
looted. Among many damaged gold ornaments and beads, one
intact gold bracelet was found, on which 'granular' ornament
alternated with inlays of blue paste. Many fragments were also
found o£ mosaic ornament in coloured glass. This would be
contemporary with or older than the i8th-Dynasty Egyptian glass
found at Tell-el-Amarnah. The craftsmanship of that period in
Egypt was also recalled by the painted head of a man with a full 123
beard in terracotta and the equally well-modelled figure of a
^^
lioness.-^ 124
124 Sensitively modelled head of

a lioness from Dur-Kurigalzu.
Our knowledge of Kassite art
depends mainly on small objects
of this sort. Ht c. 5 cm
175
Mitanni Nuzi
We have now to deal with the Mitannian occupation of northern
Mesopotamia, which was in part contemporary with the Kassite
dynasty in Babylonia. Mitannian kings reigned from their capital at
Washshukanni, near the headwaters of the River Khabur, from
about 1500 BC until the mid-i4th century, when their kingdom
disintegrated. During that time, one of their southernmost
strongholds was the city now known as Nuzi, 12^ miles to the
southwest of modern Kirkuk (Arrapha), and its site was excavated
by Americans from Pennsylvania, under R. F. S. Starr and others,
during the late 1920s and early 1930s. ^^° To anyone seeking the
criteria of Mitannian culture, the results of this excavation are both
rewarding and disappointing. A palace and temple were found, as
well as a large number of important private dwellings. In the
absence of conspicuous works of art, innumerable small finds give
an impression of a people whose way of life was much enriched by
trade and familiarity with the countries of the Levant and even with
Egypt. But undoubtedly the major discovery of the expedition was
an abundance of tablets: over 5,000 texts in all, including in certain
cases the entire archives of a single family over periods of as much as
five generations.^^'
The main mound was called Yorghan Tepe, and beneath it lay
the actual walled city with its palace and temple ;
^ ^ ^ but there were
also outlying mounds covering the remains of other palatial
residences. To some reference has already been made,
the temple
since was originally founded in the Early Dynastic period, at a
it
time whenthe place was known as Gasur. It was finally rebuilt in

Mitannian times, in a form which comprised two sanctuaries, and
its remains were conveniently dated by a letter from the Mitannian
king Shaushattar {c. 1460 Bc). One shrine had internal walls faced
with wooden panelling in the form of 'clapper-boards', pegged to
the brickwork. Elsewhere the wall-faces were decorated with nail-
shaped bosses of glazed terracotta and, suspended from these, some
sort of woven fabric enriched with beads. Brightly coloured
glazing, of the sort common in Egypt at this time, was also applied
to other objects, such as small guardian figures of lions and, in one
case, to a wall-ornament in the form of a boar's head. As for
sculpture, the excavators had to be content with the eye-inlay of a
life-size statue.
The palace was built around the usual broad open courtyards. In
the largest of these, at the approach to the main reception suite, the
walls seemed to have had some sort of painted wooden entablature

beneath the eaves and, before the main entrance to the throne-
room, a pair of free-standing brick piers may have supported a
portico or ornamental baldachin. Everywhere there were elaborate
provisions for drainage and sanitation, including at least one
example of a toilet, flushed by piped water from above. A
luxurious bathroom and lavatory were annexed to the royal
apartments, and the vestibule through which these were approached
was decorated with impressive mural paintings, incorporating
foreign motifs such as guilloche ornament, acroteria, bucrania and
'Hathor heads' ^ ^ ^ Main doorways had pivot-stones, in one of which
.
the bronze 'shoe' of the door-pivot itself remained in place. The
176
12$ Fragile pottery of the style

known as 'Nuzi ware', with
designs painted in white on a
dark ground. Such vessels date
from the time of the Mitannian
kingdom {c. 1475-13 50 BC),
based in northeast Syria. (From

Starr, 1938)
great thickness of the walls in the principal reception units could

have implied an increase in their height to allow for clerestory
lighting but the excavators preferred to see in this a
; mere provision
for security.
Several mansions in the northern suburb, including that of the
'Crown were the
Prince' ('Governor of the City of Arrapha'),
subject of special study. Those of two brothers called Tehip and
Shurki Tilla between them produced 1,000 tablets. That of an
individual called Zigi contained a formidable collection of copper
objects, comprising arrowheads, spearheads, knives and much
laminated armour plating, fused into heavy masses. But there were
tools as well as weapons scythes, punches, chisels and adzes - all
:
discarded apparently at a time when the buildings were destroyed

by fire.
Among the minor fmds made in these palace buildings at Nuzi,

one distinctive feature was an unusual type of pottery. ^^"^ These 125
fmely made vessels were elegantly decorated with designs in white
paint on a dark-coloured background. The commonest form was a
tall cup with 'button base', and the patterns, usually in horizontal
registers, included motifs till now unfamiliar in Mesopotamia, such

as running spirals, guilloches and rosettes, as well as formalized
birds and plants. Though this ware has been found as far south as the
Kassite capital Dur-Kurigalzu and to the north at Tell Billa beyond
Mosul, its affinities are with northern Syria, where it has appeared
at such sites as Tell Atchana (Alalakh), and was supposed by
Woolley (perhaps wrongly) to have originated in Crete. Mitannian

cylinder-seals too, have more in common with those of northern
Syria, and even Cappadocia. The designs are often overcrowded
and confused; but certain motifs such as the stylized 'sacred tree'
and the winged griffin survived in the imagery of later Assyrian
ornament. ^^5
Finally, there is the famous collection of texts from Nuzi. The
excavations brought to light no less than 4,000 cuneiform tablets, of
which almost 1,000 were found in the house of Tehip-Tilla. Their
contents create an extraordinarily detailed picture of daily life in
this part-Hurrian community of citizens. Legal and commercial
transactions which they often record are of interest in themselves;

but it is their incidental references to people, places and things that
promote one's understanding of the social or environmental
background against which they were written, and amplify the
evidence provided by archaeological fmds.
177
In an appendix to R. F. S. Starr's publication of the Nuzi

excavation (1939), E. R. Lacheman summarizes most interestingly
the 'evidences of material culture' to be derived from the texts: a
most welcome addition to an archaeological report of this sort. It
I
appears under headings whose variety alone prepares one for the
I wealth of information which they contain.
First then, Nuzi emerges as a walled city, with towers, bastions
and gates whose names are mentioned. The palace is a great
complex of buildings with annexes serving many purposes and a
long architectural history. The temple known to the excavators
was one of several shrines, piously served by the leading families, by
suitable offeringsand the gift of slaves. Accounts of the building or
repair of private houses provide a glossary of architectural terms
and structural details. Streets are frequently mentioned by name,
and the topographical picture is supplemented by much detailed
reference to the distribution of water: a small river serving a
network of canals, reservoirs and bridges, each of them carefully
named. Other landmarks were individual wells. Dealings in real
estate and the transfer of land ownership reveal details of
agricultural practices in a well-irrigated province, surrounded by
uncultivated steppc-land, and there is a precise vocabulary of words
commonly used for such purposes. One learns the relative
popularity of various crops, of which barley once more the most
is
common, followed by wheat, emmer, poppy and flax. In a

separate category are orchards and gardens, from which fruit,
vegetables and flowers are listed by name, in careful inventories,
and the wood of fruit-trees is recommended for specified purposes.
Much information of a more technical character is provided under
further headings, such as 'Chariots and Wagons', 'Furniture' or
'Clothing', and there is a section on tools, weapons and armour.
In the catalogue of subjects dealt with by the Nuzi texts, one
regrets above all the paucity of references to political subjects and
contemporary history. Indeed, the only piece of inscriptional
evidence by which the Nuzi culture can be positively dated is, as we
have said, a document from a room in the great house of Shilwi-
Teshup a well-known letter from Shaushattar, King of Mitanni,
:
who ruled in the latter part of the 15th century bc.
The Middle Assyrian Period
For our knowledge of Assyrian archaeology during this period we

are indebted almost exclusively to the work of the German
excavators at Ashur (Qal'at Sharqat) on the Tigris. The decision to
embark on major enterprise was made in 1903, when, in the
this
south, Robert Koldewey's expedition had already spent four years
wrestling with the problems presented by the great ruinfields of
Babylon. They had by then largely elucidated the site-plan,
identified most of the principal buildings and were assured of
success in their efforts to resurrect the remains of Nebuchadnezzar's
capital city. Only in that year, it became evident that the
unexpectedly high level of the sub-soil water-table was likely to
preclude the investigation of any occupation level earlier than the
7th century bc. A supplementary operation was therefore
178
suggested, at asite where such remains might be more accessible
and, when Ashur was chosen, it was Walter Andrae who was
entrusted with the new undertaking. From then onwards, until
1 91 3, the two excavations continued simultaneously at sites 350
miles apart. Their combined accomplishment, both in the recovery

of historical information and in the perfection of archaeological
method, has since come to be generally recognized. ^^^
City of Ashur
We have already mentioned the situation of Ashur, on an outcrop
of limestone rising to a considerable height in the angle formed by
two courses of the River Tigris. This gave the city natural 126-7
protection on two sides and, onlandward side, its defences were
the
completed by a crescent-shaped array of powerful walls, enclosing
an area three-quarters of a mile wide.^^^ Andrae's principal
excavations were carried out in the elevated part of the city to the
north and northwest, where its temples and palaces were situated.
He had assured himself that this was so by the now-standardized
German method of cutting search-trenches (suchgraben) clean across
the site from east to west at intervals of 100 m, corresponding to the
grid-lines of his initial survey. At the same time the full circuit of
the fortifications was traced, exposing the principal gates, river-side
quays and other features. His deepest penetration was the sounding
which revealed the Pre-Sargonid foundation of the Ishtar Temple.
The latest architectural remains he encountered could be dated to
the final phase of the Parthian period, about ad 256.
Andrae and Koldewey were both architects, and at Babylon
their tendency to over-correct the evident faults of earlier
excavators resulted in their concentration on the study of buildings
rather than written texts. ^^^ At Ashur, the function of Andrae's
epigraphical advisers gained in importance for reasons which soon
became obvious. Over a period of some 2,000 years, successive
kings had contributed to the restoration and multiplication of
public buildings or fortifications and the chronological sequence
;
of their building-levels could only be determined with the help of

associated inscriptions. These showed that many of them had
continued to be rebuilt by Late Assyrian kings, long after the city
had ceased to be their administrative capital. In the paragraphs that
follow, this will explain the mention of royal names belonging
more correctly to a later chapter.
Fortifications
The landward rampart of the inner city was built and rebuilt by a
number of kings, including Shamshi-Adad I in Hammurabi's time.
Tukulti-Ninurta I (i 244-1 208 Bc) provided it with an external
moat, 20 mwide. A foundation inscription also records that

Shalmaneser III (858-824 bc) completely reconstructed it, adding
an outer wall which ran parallel to it but was diverted at its southern
end to enclose an extension of the city. Meanwhile, in about 1300
BC, Adad-Nirari I had constructed a great quay of baked brick
along the eastern river front, and this had been strengthened by
several later kings as a secondary protection against flooding. A
description of the town in the time of Sennacherib (704-681 bc)
179
126 The of Ashur on the

city
Tigris, in the Middle Assyrian
period, as seen from the
northwest. A drawing by Walter
Andrae himself, towards the end
of his ten-year excavation
127 Town plan of the city of

Ashur, on its eminence at the
junction between two branches
of the Tigris.The Bit Akitu is
outside the town to the
northwest. (From Hawkes, 1974)
180
mentions thirteen gates in these fortifications, though the

excavators were able to locate only eight (distinguished by
numbers in the site-plan). The most important of these, used by
travellers to the north, was known to the Assyrians as 'Gurgurri' (i
& 2 in Andrae's plan). Between this and the western river-bed there
was a kind of citadel (referred to by the Germans as Aussenhaken),
with gates at either end (3 & 4) through which a great procession
passed out of the city at the time of the New Year's Festival (sec
below). Next there were gates in the north wall, a western gate
known as 'Illat', and finally minor gates at the southern extremity
of the city and its later extension (7 & 12).
Temples and Palaces

Approaching the city by river from the north, its massed buildings
must have created an impressive silhouette - as one realizes from a
well-known reconstruction drawn by Andrae himself. ^-^ The 126
temples rose to a maximum height on the northernmost shoulder
of rock, and there were three ziggurats. The greatest of these,
which had been given its final form by Shamshi-Adad I, measured
60 m square and was originally dedicated to the Sumerian god
Enlil, who, as a patron of Assyria was later replaced by the city-god
Ashur himself. In its exposed position the tower was much denuded
by weather and nothing remained of its stairways or auxiliary
buildings. Out of 38 temples mentioned in the inscriptions, only 4
were fully excavated, the largest of which was again dedicated by
Shamshi-Adad to Ashur and occupied the most prominent position
on the northern height, with an enclosed precinct extending
southward. Here, as in the later rebuildings of the old Ishtar
Temple, something is to be learnt about the changing conventions
that dictated the planning of sacred buildings. The Ashur Temple,
as built by Shamshi-Adad, seemed to have two main sanctuaries, one
at least conforming to the breitraum arrangement currently adopted

in Babylonia. Many centuries later, Sennacherib added to it an
eastern annex, with its cella set lengthwise on the main axis in the
Assyrian manner. As for the Ishtar Temple, after the destruction of
the 'Archaic' shrine, the Germans found traces of no less than six
rebuildings, in the course of which even its dedication was changed.
Yet, always annexed to it, was a small secondary shrine, where
Ishtar continued to be worshipped, and which still retained the old
langraum cella, with its 'bent-axis' approach following the
Sumerian tradition.
Of the two other principal temples, both had dual dedications
and unusual plans. One was the Anu-Adad shrine, with its twin
ziggurats, identical tieftempel units in the Assyrian manner, and an
enclosed forecourt: all contrived as a single architectural com-
position. The second double temple, dedicated to the moon- and

sun-gods, Sin and Shamash, was first built by Ashur-nirari I (c. 1 500 1 28
Bc), but rebuilt to a different plan by Sennacherib. Both have twin
Assyrian sanctuaries, approached through breitraum antecellas. A
fifth cult-building, hardly to be regarded as a temple, was that
known in Assyrian as the Bit Akitu, to which the cult-statues were
brought in procession during the New Year's Festival. It was
located 400 m outside the city to the north-west, and consisted of a
181
llHri'
128 The Temple of Sin and

Shamash at Ashur above,
:
foundation plan at the time of

Ashur-nirari I (1500 Bc); below,
plan at the time of Sennacherib
(704-681 Bc). The long, axial
sanctuaries are common to most
Assyrian temples. (After Parrot,
1946)
great square courtyard with lateral colonnades and a sanctuary on

the main axis,more closely resembling a throne-room.
A further word should here be said about the Akitu or 'New
Year's Festival', which acquired complete ceremonial form
its
during the early centuries of the millennium bc. By that time it

ist
had replaced or assimilated the earlier Sumerian 'Spring Festival',

associated with the 'Sacred Marriage' of god and goddess,
symbolizing the recurrent cycle of creation and fertility. The feast
now took place at the time of the spring equinox and lasted eleven
days, each day marking a stage in a prolonged and complicated
182
ritual performance. It was for the culminating phases of this
ceremony House provided a setting.

that the Bit Akitu or Festival
Thither the statues of the gods wêre escorted, mounted on wagons
or boats, having been temporarily removed from their temples, to
be present at the celebration of a rite known as the 'Fixing of
Destinies', inwhich the king himself played a leading role. Many
details of the ritual used on these occasions are known from
surviving texts; and at Ashur the paved way along which the
procession passed out of the city has been revealed by excavation.
In contrast to the temples, there is little to be said about the
secular buildings at Ashur. The so-called 'Old Palace', sited next to
the main ziggurat, embodies a labyrinth of rectangular chambers
and ranges of store-rooms, entered and lit from a succession of open
courtyards. In this way it resembles the palaces of the Larsa period
at Eshnunna, with which indeed its original foundation was
contemporary. Here and there also, as in the palaces of Mari and

Ur, private religious shrines are incorporated. The Ashur palace
was still in use at the time of Tiglath-pileser I (ii 15-1077 bc); but
during the Late Assyrian period, what remained of it seems to have
been converted into a mausoleum. Ashur-nasirpal (883-859 bc)
embellished one of its gateways with winged-bull sculptures and
beneath it he constructed elegant tomb-chambers. These were
found by the Germans to contain huge, monolithic sarcophagi,
looted in antiquity but still bearing the names, among others, of
Ashur-nasirpal himself and of Shamshi-Adad V (823-81 1 bc).
Meanwhile, on an artificial terrace in the northwest angle of the
city wall, Tukulti-Ninurta I, after his conquest of Babylon, had
built a 'New Palace' of which only the stone foundations remained.
Not content with this, he also provided himself with a new
residential suburb, i-8 miles to the north of the city on the left bank
of the Tigris, and named it 'Kar-Tukulti-Ninurta'. Here too he
built a miniature ziggurat and, beside it, a temple with a
Babylonian sanctuary and a cult-niche actually penetrating the
facade of the tower itself Andrae thought that in this case access to
the upper terraces of the ziggurat was obtained by building a bridge
over the street behind. For the rest, some of the most interesting
discoveries at Kar-Tukulti-Ninurta were examples of figured
ornament in coloured glaze on terracotta, sometimes in the form of
upright wall-slabs ('orthostats'), and a panel built up out of
polychrome glazed bricks. ^2°
Some Finds at Ashur

We should now consider some of the finds at Ashur and the
symptoms among them of ideas or innovations which can be
accepted as characteristically Assyrian. In a study of this subject,
Henri Frankfort claims that
... in the fourteenth century an art emerged which, for all its derivations,
possessed an individual character, not only in style but also in subject
matter. It depicted secular subjects with an interest in actuaHty for which
no incident seemed too trivial. In religious matters on the other hand, it
displayed a cold formalism, which did not allow man to meet the gods face
to face, but only to perform the established rites before their statues and
emblems. ^^'
183
129 An altar of Tukulti-Ninurta An example, frequently quoted, of this last tendency is the carved
I 244-1 208 Bc) from Ashur,
(i
relief on an altar set up by Tukulti-Ninurta I. If one remembers the
with a relief in which homage is
paid only to the symbol of a

between man and god in the scene carved on
direct confrontation
god, in this case Nusku, set upon Hammurabi's famous stela, here is a direct contrast. The Assyrian
an image of the altar itself king is seen to approach and then kneel before an altar similar to
Ht 53 cm that on w^hich the picture is carved, and it is surmounted, not by a
statue but by the traditional symbol of a particular god, in this case
Nusku, one of the three fire-gods manifest in the sacrificial flame.
Perhaps this attitude to divinity is equally reflected in the planning
of Assyrian temples described earlier, where the cult-statue is
withdrawn to the furthest extremity of an elongated sanctuary. As
for the god Ashur himself, he too more often appears as a symbol
rather than in human form. His figure acquires wings, like the sun-
disc of Egyptian Horus, and the disc has a bird's tail.
Another contemporary development can be seen in the carving
of cylinder-seals, which now show a new upsurge of interest in
design. ^3 2 jj^ ,.j^£ Larsa and Old Babylonian periods 'presentation
IJO scenes' had reached such a degree of monotonous uniformity that
an inscription had often to be added giving the owner's name. This
convention was now discarded and replaced by distinctive
compositions in tabloid form. Some traditional 'combat scenes' can
be seen to survive, but new subjects are abundant. A king hunts
131 ostriches ; a winged horse protects its foal from a lion, or stags graze
184
130 A late example of the

'presentation scene'. According
to the inscription, the seal's
owner, Hashhamer, Governor of
Ishkun-Sin, is brought by a
minor goddess into the presence
of a king of Ur (early 20th
century b c)
131 Middle Assyrian seal design

showing a king or genius hunting
ostriches (13 th century bc)
132 Another Middle Assyrian

scene, described as 'the noble
Pegasus defending his wingless
foal from a Hon . .
.'
133 Seal design marking the

transition from Middle to Late
Assyrian styles {c. 900 Bc). Kings
and supporting genii face a

sacred 'tree of Hfe', with th«
bird-Hke symbol of the god
Ashur above (cf. ill. 148)
^'Ji
Wiii'm':^^:Jj,r
/
mmm mv'mmMm'î^^SW^.
'•
185
among stylized and mountains. Added to these are the

trees
hitherto aUen motifswhich we have seen at Nuzi, imported by way
i55 of Mitanni from Syria: the 'sacred tree' and the 'winged griffin'.
Other Mediterranean forms of ornament are to be seen in the
glazed panels from Kar-Tukulti-Ninurta.
Building Practices
Our earlier account of buildings at Ashur was, for obvious reasons,
largely confined to variations in their ground-plans. Where fa9ade
ornament and external appearance are concerned, occasional clues
are to be found among contemporary seal designs. This applies in
particular to the parapet treatment of buildings, whether religious
or secular. We suspect that the use of crenellations was introduced
Ilhrl
1^4
at about this time from Egypt; and one 13th-century seal from
Ashur shows a double fortress wall, its parapet crenellated with

round-headed merlons. ^^^ Two other seals from the same site,
dated perhaps a hundred years later, show for the first time merlons
in the 'stepped' form familiar today. ^^^ One of these is of special
interest, first because the symbol of the dog-deity, Gula, proves the
building to be a temple rather than a fortress; which would mean
that the crenellations are probably ornamental. Secondly, it shows
architectural details, such as towers rising above the tops of the
walls, doorways and rectangular panels which might be windows.
Finally, in seals of this period, genii of the sort known as apkallu
13} sometimes appear: hybrid creatures with two pairs of wings,
winged and hawk-headed monsters, or bearded men sheathed in
the skins of fish. But these are to be seen more frequently in the
form of terracotta plaques or figurines, such as are sometimes found
buried beneath the floors of houses for apotropaic purposes. They
too take their place in the imagery of later Assyrian times.
134 Fortifications of an Assyrian

from a relief of Tiglath-
city,
Pileser III, showing crenellated
parapets and towers as finally
standardized in Late Assyrian
times. (Drawing by R. Leacroft)
186
Chapter Nine
The Late Assyrian Period
Imperial History
Early in the loth century at Ashur, a new dynasty was founded by a

king called Ashur-rabi II and marked the beginning of
in a sense this
an important epoch in Mesopotamian history. The reign of his
grandson, Adad-Nirari II (91 1-89 1 bc) saw the earliest stage in the
revival of economic stability and military initiative which
eventually led to the creation of an Assyrian empire. Adad-Nirari's
concern was to establish once and for all the essential frontiers
first
of his kingdom, and this he accomplished by a series of successful

campaigns and poUtical manoeuvres. Cities like Arrapha (Kirkuk)
in the east (Tell Halaf) on the headwaters of the Khabur
and Guzan
became strongpoints in a new defensive system, while
in the north,
small Aramaean principalities on the middle course of the
Euphrates were annexed to protect his trade connections with
Syria.
At an early stage in these operations, Assyria's southern frontier
with Babylon was fixed by a treaty, the text of which has been
preserved in a rather famous inscription known as the
'Synchronous History'. This document conveniently incorporates
a chronological account of all previous border disputes between the
two states. It may also be considered to correspond in time with the
beginning of accurately dated history, since from now onwards we
have the supplementary evidence of the so-called '//mmw-lists'.
These lists record in chronological order the names of particular
officials appointed annually to preside over the New Year's
Festival. Since they had come to be used in dating legal agreements

and other important documents, their historical significance is
obvious.
Adad-Nirari's successor, Tukulti-Ninurta II made the first move
in extending his father's frontiers to the north and, in doing so,
found himself for the first time in contact with the Muski (or
Phrygians?) who had replaced the Hittites on the Anatolian
plateau, and also with the 'Nairi peoples', later consolidated into
the state of Urartu. After this he contented himself with an
operation of the sort which was later repeated at intervals by most
of his successors: a general sweep around the eastern and southern
provinces, checking on subject states from whom tribute had not
been forthcoming, and making an admonitory gesture towards
Babylon. On this occasion, he paused to inspect the remnants of the
old Kassite capital at Dur-Kurigalzu.
187
But, from an archaeological viewpoint, the most important of

thesefirst 'Iron Age' kings of Assyria was Ashur-nasirpal (883-859
Bc) ; for it was he who transferred the centre of government from

Ashur itself to a hitherto less-important city at the junction of the
Tigris and the Upper Zab. This was Kalhu, the Calah of Genesis,
now known as Nimrud, which he enlarged and embellished to
create a new Assyrian capital. Once this was done, and an improved
military headquarters established, he set out on a campaign directed
towards the Euphrates crossings in the northwest. Until now, the
main barrier to progress in this direction had been a powerful state
centred on Til Barsip (Tell Ahmar), near Carchemish, and its
capture opened the way to Aleppo. From there his armies marched
almost unopposed as far as the Mediterranean, and tribute was
exacted from the coastal cities. Inaugurated in this way was the long
saga of foreign conquest which, under Ashur-nasirpal's successors,
brought Assyrian armies westward as far as Egypt.
For the time being, his hold on the Levantine cities at least must
have been oppressive, for, in the reign of his son, Shalmaneser III,
we hear o£ an anti-Assyrian coalition, headed by a prince called
Adad-iri of Damascus (who is suspected to have been Benhadad of
the Old Testament). This revolt was suppressed and the inland
cities of Syria brought into subjection, along with Tabal, north of
Aleppo and Que in Cilicia. Economically these conquests must

have been greatly to the advantage of Assyria, since they ensured
the freedom of trade-routes to the west and access to metal-sources
in the Taurus Mountains. Similarly, in the south, Shalmaneser's
suppression of a revolt by Chaldean chieftains against the friendly
government in Babylon (celebrated on his throne base at Nimrud),
enabled him to maintain contact with traders in the Arabian Gulf,
who brought goods from India and Arabia by way of Dilmun
(Bahrein).
It would be out of place here of Late
to pursue further the threads
Assyrian history in any great detail. Of its framework
be it may
necessary to recall only those elements which lend coherence to the
archaeological record. The five reigns which followed that of
Shalmaneser III represent a period of comparative inertia, during
which imperial aspirations seem to have been temporarily in
abeyance and little effort was made to counteract the erosion of
newly acquired territories. To the north in particular, the now-
powerful kingdom of Urartu, with its system of fortresses in the
region of Lake Van, was pursuing a policy of expansion which
threatened to deny the Assyrians access to Anatolia. Its territories
now extended eastward to the shores of Lake Urmia and, by 745 bc
when Tiglath-pileser III succeeded to the throne in Kalhu, the
Urartian armies were engaged in a westward thrust towards Syria.
Firm action was needed and Tiglath-pileser proved himself a ruler
equal to the occasion. During his reign one sees Assyria recovering
its lost territories and re-establishing itself as the leading economic
and military power in the Near East. Contributary to this revival

were the successful administrative reforms whereby he reorganized
the whole system of government, both in the homeland and in the
provinces. These were made effective by the establishment of
posting-stages along the main caravan routes which enabled the
king to be in continual contact with his provincial governors. The 135 Map of the Assyrian empire.
vassal states which they controlled were further protected by an (S. Ebrahim after Postgate, 1977)
outer tier of 'client kingdoms', with a greater measure of

independence. By the time of his death in 727 bc Tiglath-pileser
ruled an empire extending from the borders of Egypt as far as
southern Babylonia, the only region still afflicted with political
unrest. It was in this quarter that dissident tribesmen, now under
the leadership of the formidable Marduk-apil-idinna (called in the
Bible Merodach-baladan), were for many years to persist in
defying the authority of Assyria.
We are now confronted with the names of four great rulers, in
whose reigns one sees, as it has been said, 'not only the culmination
of Assyrian power but the seeds of its disintegration'. Where

Sargon is concerned, the circumstances of his accession and his
subsequent military triumphs must, for our present purpose, be
subordinated to his enterprise in providing himself with a new
capital.At the site known today as Khorsabad, 12 miles northeast of
Nineveh, he built himself a heavily walled city, more than i mile
square, and named it Dur-Sharrukin. Its pubUc buildings and
fortifications were completed in a remarkably short time; but there
is some doubt whether the accumulation of residents which would
have created a populous city, could have been accomplished in the
sixteen years of his reign. For he was seldom there and after his
189
136 Map showing the Hne of

Sennacherib's Canal, from the
barrage and sluices at Bavian,
passing over thejerwan aqueduct
and finally joining the course of
the River Khosr to Nineveh. (S.
Lloyd)
death in 705 bc the place was abandoned. His successor

Sennacherib preferred to adopt the older city of Nineveh as his
centre of government, thereby profiting from the obvious
advantages of its central position on the east bank of the Tigris, and
it was there that, for the third time in succession, a new and more
splendid setting was contrived for the Assyrian court.
In a well-known text, Sennacherib reproaches his forbears for
having neglected the upkeep of Nineveh. He adds a lengthy and
rather detailed account of his own efforts to remedy this situation,
and the magnitude of his undertaking is made apparent by the
immensely extended area of the city whose ruins can be seen today
from the west-bank eminences of modern Mosul. Its walls form an
irregular rectangle almost 2j miles long, and on the side facing the
river one sees the older mound whose shape and size he adapted to
form an emplacement for his palaces and temples. Outside the
walls, as we know from the same text, he laid out orchards and
plantations with rare trees brought from distant provinces, and he
added to these a zoological garden. Not content with the quality of
lj6 water in the Tigris he also constructed a canal, more than 50 miles
long with stone aqueducts, to bring a cleaner supply from springs in
the mountains to the northeast. At its remotest source, on a cliff-
190
face near Bavian, completion was commemorated by rock-

its
carvings of the Assyrian gods and inscriptions, to which he added

an account of his other poHtical and mihtary accomphshments.^^^
In these texts and others from his palace at Nineveh, Sennacherib is
convincingly presented as perhaps the greatest soldier-statesman
since Hammurabi.
Only two notable figures remain in Late Assyrian history. One is
Esarhaddon, who succeeded Sennacherib after his eventual murder
by members of his own family. He it was who reached the ultimate
goal of his country's imperial ambitions by invading Egypt.
During the temporary absence of its Ethiopian ruler, Tarku or
Tirhakah, he besieged and captured Memphis, thereafter adopting
the title of the Pharaohs, 'King of Upper and Lower Egypt'. At
Nineveh he had time to build himself a new administrative palace,
inwhich he installed some of the loot from Egypt, before receiving
the news that Memphis had been retaken by Tarku. He was on his
way to deal with this situation when he died. In a document which
has survived, Esarhaddon had long ago assured the accession to the
Assyrian throne of his son Ashur-banipal, and it was he who was
now compelled to reconquer Egypt. This time Memphis was
destroyed. Ashur-banipal next turned his attention to the southeast
frontier, where Elam was in revolt, having formed an alliance with
Chaldea. Here too he scored a notable victory and it was his
capture on this occasion of Susa, the Elamite capital, which was
commemorated by a well-known relief carving in his palace at
Nineveh.
Twenty-five years later, the Chaldeans played some part in the
events which led to the final destruction of Nineveh. In 614 bc, a
Median army under Cyaxares invaded the Assyrian homeland,
capturing Nimrud and totally destroying the more ancient capital
at Ashur. This brought them into contact with the Babylonians and
a treaty was signed between Cyaxares and the Chaldean king

Nabopolassar. During the following year, the Medes were also able
to enlist the help of Scythian tribes to the north of their realm and
the three peoples made common cause against Nineveh. The city
fell 612 bc and was largely
after a surprisingly short siege in
destroyed. The remnants of the Assyrian court took refuge at
Harran on the northwest frontier.
Early Excavations in Assyrian Cities
It has seemed necessary to present this rather threadbare account of

the latest phase in Assyrian dynastic history in order to clarify the
chronology and location of the discoveries which we are about to
discuss. Their abundance and variety are today made apparent by
the heavy volumes in which they are recorded, and by the
accumulation around them of commentary hterature. Here again it
becomes clear that a selective analysis of the knowledge which they
have helped us to acquire is the most that can usefully be attempted.
For our present purpose, then, a distinction must first be made
between two classes of excavation, to which the ruins of Assy-
rian cities have on occasions been submitted - widely separated
as they were in time, and variously dependent on the contemporary
191
lifi
137 Austin Henry Layard

(18 1 7-1 894) in Bakhtiyari dress.
A drawing by Ulrica Lloyd from
portraits in Layard's Early
Adventures and Autobiography and
Letters
understanding of archaeological intention. The first category of

course comprises the work of the great pioneer explorers in the
19th century, whose objective was the discovery and acquisition of
removable antiquities, with little attention to the setting in which
ij8 they were found. It started in 1842 with Botta's discovery of
1J7 Khorsabad, followed two years later by Layard's first excavations
at Nimrud. It lasted, with a brief interval during and after the
Crimean War, until the death in 1877 of George Smith, who had
found and identified the cuneiform version of the 'Deluge' story.
Excavations in the second category - of the sort with which we
are familiar today, paying proper attention to stratigraphy and
environmental evidence - were resumed in 1927, when American
archaeologists returned to Khorsabad, and later by British
excavators at Nimrud. To the lives and adventures of the earliest
excavators little attention will be paid here, except to admire in
retrospect their dedication and persistence in the face of sceptical
discouragement or deliberate obstruction. They were rewarded in
the end by popular reaction to the romantic appeal of their
192
138 Paul Emile Botta, discoverer

and excavator of Khorsabad.
Botta's pioneer work in Assyria
received little recognition from
the government of the Second
Republic in France. He was
relegated to a consular post at
Tripoli in Syria and died in 1870
discoveries and could forget the ill-humoured contention and

international rivalry which had marred their relations with each
other, and which later deteriorated into an undignified scramble for
archaeological loot. Their story has been told elsewhere, in
narrative form by the present writer and, more objectively by
French and German scholars. ^^^ Our present aim will be to
examine the sites at which they dug, and to identify the monuments
or buildings that they encountered, in the light of subsequent and
more systematic investigations.
Nimrud (Kalhu)
At Kalhu, the greatly enlarged city designed by Ashur-nasirpal II
(883-859 Bc) during the early years of his reign, was approximately
square in shape with a maximum dimension of almost 1.25 139
miles. 2 3*7 was bounded on the west side by the Tigris and to
Its site
the south by a canal, into which water was diverted from the Upper
Zab river. In the southwest corner, a mound created by the remains
193
^he Late Assyrian Period
Plan of City Walls
Palace
139 Diagram showing the

fortifications of Nimrud and its
principal groups of buildings.

The city was protected to the
west by the Tigris and to the
south by a wide canal. (After
Hawkes, 1974). A, Fort
Shalmaneser Original bed of Tigris
of had been reshaped and revetted to form a

earlier settlements
140 which the main palaces and temple buildings

raised citadel, within
could be concentrated. But there was also, in the southeast corner, a
walled enclosure containing the imperial arsenal.
It was the citadel mound which first attracted the attention of
Layard when he started his excavations in November 1 845 ^ ^ ^ As it .
exists today, partly eroded by the drainage of rainwater from the

summit, one sees the rhomboid of its retaining-walls, dominated by
the remnants of a small ziggurat in its north-west corner.
Dramatically revealed by Layard's first trenches were the walls and
sculptures of Ashur-nasirpal's own official residence (later known
as the 'Northwest Palace'), overlooking the river directly south of
the ziggurat. It was here that, in addition to wall-reliefs, he exposed
for the first time doorways flanked by guardian figures of winged
bulls or lions with tall 'genii' behind them. His work was for the
time being confined to chambers around the ceremonial court of
the building, and they contained figures of bronze, alabaster vessels
and fallen fragments of mural paintings which he had no means of
140 Plan of the citadel mound at
conserving.
Nimrud after the conclusion of
Layard's first period of work at Nimrud lasted until 1847, by
Mallowan's work there in 1963.
Previous excavations by Layard, which time he had located and examined a number of other
Rassam, Loftus and others took buildings. Again on the river-front was a so-called 'Western Palace'
place intermittently between of Adad-nirari III and a 'Southwest Palace', built or restored by
1845 and 1854. Principal finds
Esarhaddon, using sculptures intended for a building of Tiglath-
were made in the northwestern
III. The name of this last king was also associated with that of
pileser
and southeastern groups of
buildings. (From Mallowan, Shalmaneser III on many inscriptions from a 'Central Palace' and :
1966) itwas in the environs of this building Layard came upon the
that
famous 'Black Obelisk' (now in the British Museum), with its
scenes of tribute-bearers among whom 'Jehu, King of Israel' is
named. Finally, beneath an eminence in the southeast corner of the
mound, he cleared some rooms of a building attributed to a very
late Assyrian king called Ashur-etililani, in which Mallowan, over a
century later, made important finds. In 1847, with Rawlinson in
Baghdad acquiring improved knowledge of cuneiform, Layard
194
195
was able to satisfy himself that the site he was excavating was
Kalhu, rather than Nineveh.
Layard returned to Nimrud in 1849, after pubHshing his book,
Nineveh and its Remains, and he brought with him an artist (Cooper,
later replaced by Bell), to take over the task of drawing the reliefs
which had been occupying too much of his time. During the
following two years he made further finds in the Northwest Palace,
including some hundreds of bronze objects, with ornaments of
ivory and" even of glass. At the foot of the ziggurat he also
discovered a temple dedicated to Ninurta, with inscribed paving
slabs presenting the annals of Ashur-nasirpal's reign and, in its
sanctuary, an almost undamaged statue of the same king, which has
i
tken its place in the British Museum as a rare example of Assyrian
'•f^ •
sculpture in-the-round. But, during 1850 as we shall presently see,
he had begun a secondary excavation in the palace mound at
Nineveh, where his discovery of Sennacherib's palace brought his
archaeological field-work to an end.
From 1852 to 1854 Layard's work in Assyria was taken over by
his assistant Hormuzd Rassam, brother of the locally born British
Consular representative in Mosul, who from then onwards became
known for the ruthless pragmatism of his archaeological methods
and disregard for the finer points of international ethics. ^^^ At
Nimrud, in the southeast part of the citadel he located the Nabu
Temple complex, more recently and more methodically excavated
by Mallowan's expedition. There he found statues, two of which
were respectively dedicated 'for the life of Adad-nirari III and of
his mother Sammuramat (Semiramis) : also a stela of Shamshi-Adad
V. But he was by now too preoccupied with excavations at
Nineveh to absent himself for long from Mosul. In 1854 Rawlinson

entrusted the work at Nimrud to W. K. Loftus, who is better
known of Sumerian cities in the
for his earlier soundings at sites
south. ^•*'^ Loftus made one more important find in a building to the
west of the Nabu Temple, now known as the 'Burnt Palace'. This
was a fine group of carvings in ivory, which, after a belated study
by the British Museum in the 1950s, made possible the first stylistic
analysis of craftsmanship in this medium.
Khorsabad
Three years before Layard's discovery of Nimrud, Paul Emile

ij8 Botta was appointed French Consul in Mosul. Already in 1842, he
had started without much success to make soundings in the
Kiiyiinjik mound at Nineveh, when his workmen told him of
ancient sculptures recently found at Khorsabad a large mound on
:
the River Khosr, 12 miles to the northeast. In March 1843 he

was soon busily excavating
transferred his activities to this site and
in the building which we now know to have been the palace of
King Sargon II. His success was immediate and his finds very
similar to those which were to astonish Layard two years later at
Nimrud. In the following year he had the good fortune to be joined
by E. Flandin, a remarkable draughtsman, who systematically
recorded the reliefs and other sculptures now brought to hght in
almost overwhelming numbers. ^"^^
196
At Khorsabad Sargon had selected a site where, as at Nimrud,

there already existed a small ancient mound of the sort which could
conveniently be reshaped to form a raised emplacement for his
The city which he laid out took the form of a
palace buildings. 143
square, with sides measuring rather more than a mile each, and was
surrounded by towered walls with seven gateways. The palace
platform was set astride the northwest wall and its inner part was
surrounded (as American excavators discovered in the 1930s), by an
inner citadel at ground level, containing other public buildings.
Also as at Nimrud was an 'imperial arsenal' in the southern
there
corner of the town. conformity with Assyrian tradition,
In
Sargon's palace itself was planned around two main courtyards, the
innermost of which was used for ceremonial purposes. Where
Botta's excavations are concerned, it is made clear in the five-
volume record of work, which he and Flandin published in
his
1849-50, that they were limited to the projecting suite of state-

rooms on the northwest side. They lasted until 1846; but after the
political events of 1848 in Paris, Botta was for some reason
disgraced and sent to an obscure consular post in the Levant. By
then, like Layard, he had come to understand that the city which he
had been excavating was not, as he had thought, Nineveh.
At Khorsabad in 1852, Botta was replaced by Victor Place. The
artist Felix Thomas accompanied him as principal assistant, and it is
to him that we owe the only surviving records of major finds made
by the French expedition during that and the following year. ^^^
The entire plan of the palace was now traced and the better- 143-4
preserved sculptures extracted for transport to Paris, together with
many smaller finds. According to Place's own calculation, at the
end of 1853, he had cleared 209 chambers, grouped around 31
courts, in addition to 3 temples and a small ziggurat. He had traced
the circuit of the city walls - 24 m
thick on stone foundations - and
he had examined 7 gateways, 3 with sculptured portal figures and i
(no. 3 on his plan), with its vaulted archways and archivolts
ornamented in coloured glaze, almost intact. By 1855, with his
records completed, he was ready to return to France, and it was in
that year that the great disaster took place. The story is today
generally well known. The Khorsabad sculptures travelled safely as
far as Baghdad, where the 235 cases were loaded, together with
antiquities from other sources, on to a large country boat and two
kelek rafts, bound for Basrah. Near Kurnah, where the Tigris is
joined by an effluent from the Euphrates, the convoy was attacked
by hostile tribesmen and all five vessels were capsized. Fortunately,
an earlier consignment, shipped to France by Botta in 1847, had
arrived safely, but today this modest collection, together with two
winged-bull figures in the British Museum, represent all that
remains of the French finds at Khorsabad. ^"^^
Nineveh
The ruins of Nineveh, as we have mentioned, received attention

from both English and French excavators during the period of
activity which we have been discussing. Its city walls, now some 141
distance from the Tigris, form an irregular rectangle, almost
jj
197
141 The and

city plan miles in circumference, with an outer rampart and ditch on the
of Nineveh in Late
fortifications eastern side.^'*'^ The palace mound, known by the Turkish name,
Assyrian times. Excavations in
KUyiinjik, marks the line of the western wall, facing the river. One
the second half of the 19th
century took place principally on
mile to the south, a second, smaller mound covers the ruins of the
the main citadel mound, Assyrian arsenal, and at its summit today the houses of an Arab
Kuyiinjik. The mound,
smaller village cluster around a Moslem shrine, associated with the name of
Nebi Yunus, is still inhabited,
the Prophet Jonah {Nebi Yunus). Entrances to the city firmly
being the site of an important
identified are, in the north wall the Nergal Gate, with its guardian
Moslem shrine. An earUer course
of the Tigris skirted the city wâll figures still preserved, and to the east the Shamash Gate.^'^^
to the southwest, which now These and others, together with the adjoining fortifications, have
marks the limit of modern recently been exposed and partially restored.
Mosul. (From Hawkes, 1974)
During his first campaign at Nimrud, Layard had found time to
continue Botta's soundings in the Kuyiinjik mound at Nineveh;
but it was not until June 1847 that his expectations in that quarter
were realized. In the southern corner of the citadel, he came upon
the walls of an enormous building which proved to be
142 Sennacherib's principal palace; but he was compelled to postpone
its further exploration while he returned temporarily to London.
When he resumed work there in 1849, he was at first disappointed

to find evidence that the palace had been destroyed by fire - perhaps
during the sack of Nineveh in 612 bc. The damage, however,
proved to be less than he had expected, and much of the sculpture
198
V'-W'
had survived. In the 71 chambers which he investigated, over 2,000 142 Plan of Sennacherib's palace
sculptured slabs were exposed, while two small rooms (numbered atNineveh. Unhappily no
complete record has been kept of
40 and 41 in his plan), were found to contain great quantities of
all the excavations in the
tablets, comprising a part of Sennacherib's library - a prodigious
Kiiyiinjik mound and even the
store-house of contemporary knowledge, with detailed records of relative positionsof some major
contemporary events. ^"^^ At this time, Layard also made a less- buildings are today uncertain.
successful attempt to excavate in the unoccupied part of the Nebi (After Paterson, 19 12)
Yunus mound. Here he found inscriptions of Adad-nirari,

Sennacherib and Esarhaddon, but was forced by the hostility of the
inhabitants to discontinue his work. In 1851, he finally returned
home, leaving the excavations in charge of Hormuzd Rassam.
Layard had reached an understanding with Victor Place that
British excavations should be restricted to the southern half of the
Kiiyiinjik mound and Place now asserted his right to explore the
northern part. Rassam, who returned to Mosul in 1852, contested
this agreement, and, having eventually resorted to the expedient of
working under the cover of darkness, had the good fortune to

discover the great northwest palace of Ashur-banipal, which he
claimed for the British Museum. In addition to the famous 'Lion
Hunt' sequence of sculptures, the building also proved to contain
the remaining half of the royal library. This was the triumphant
culmination of the British excavations at Nineveh. They were
199
CiW Wall
143 Ground plan of Sargon II's Khorsabad

Dur Sharrukin, at
capital city,
Khorsabad, with fortifications
enclosing an area over one mile
square. The detailed plan (above)
includes the royal palace on its
platform, level with the city

walls, excavated by Botta and
Place and later re-examined by
Gordon Loud. Beneath this at
ground level is the fortified

citadel excavated by Loud in the
1930s, comprising a number of
minor palace and other public
buildings.(From Hawkes, 1974)
200
completed in 1874 by George Smith, whose methodical search

through the spoil from Rassam's excavations resulted in the
recovery of a missing fragment from the 'Deluge' text.
It should be added, howêver, that a highlight in Rassam's own
subsequent explorations was his discovery in 1878 of Imgur-Enlil, a

'country seat' of the Assyrian kings at Balawat, 25 miles east of
Mosul. From a low mound at this site he extracted two pairs of
huge bronze gates, set up respectively by Ashur-nasirpal and
Shalmaneser III at the entrances to a palace and temple. ^"^^ These,
together with a third pair found by Mallowan, will presently be
referred to in greater detail.
American Excavations at Khorsabad
From the above account of excavations in the ruins of Late Assyrian

cities, it will already have become evident that secular buildings
now take precedence over temples. The king-emperors of this age

were more concerned with the construction of fortifications and
the planning of pretentious royal palaces than with the religious
shrines which were occasionally annexed to them. As one ruler
succeeded another, the palaces themselves multiplied in numbers
and were enlarged or reconstructed, usually without respect to the 144 A reconstruction of the
overall planning of the great platforms on which they stood, and Royal Palace and citadel
their ultimate destruction added to the confusion of their surviving Dur Sharrukin by
buildings at
ruins. This was the case at Nimrud, and to an even greater extent at
William Suddaby after Loud and
Altmann. On the left, the raised
Nineveh, where the architectural record has to this day remained
temple of Nabu can be seen,
regrettably incomplete. If, by contrast, we turn to Khorsabad, we connected to the palace platform
fmd a city built, occupied and abandoned in the space of a single by a bridge
201
generation. Its remains, uncomplicated by problems of stratig-

raphy, have been excavated and recorded vîth great thoroughness
over a long period of time. For our present purpose, therefore, it
may be accepted as a model most effectively illustrating the
principles of planning and construction favoured by contemporary
builders.
143-4 The American excavators who returned to Khorsabad in 1928
did much to amplify the fmds of their French predecessors. After
clearing an 'undecorated' city-gate (no. 7 in Botta's plan), they set
about re-excavating parts of the palace, at the same time removing
certain sculptures rejected by Botta for lack of transport
facilities. ^"^^ In 1930, a second gate ('A') was found, ornamented
with winged bulls and genii, which led in its turn to the discovery
of an inner citadel, surrounding the palace at ground level. Under
the direction of Gordon Loud, the following years were spent in
exposing the many buildings enclosed by the citadel wall, while at
the same time examining the great 'arsenal' estabhshment in the
southern corner of the city (known to Place as 'Palace F').^4 9
In the course of Loud's work in the citadel, a new survey was
undertaken, which involved checking and correcting Place's palace
plan. - 5 From this a striking inference could be made regarding the
inadequacy of Assyrian building methods in their initial stages.
Even the platform itself proved to be asymmetrically planned, while
the layout of buildings in the citadel appeared haphazard to the
point of inconvenience. At the approach from Gate 'A', a large
temple, dedicated to the god Nabu, was sited at an awkward angle
beside a paved street. This building was raised on a platform of its
own to the level of the palace, from which it could be reached
directly by a bridge over the road. Beyond it there was an open
space from which a broad ramp led up to the main portal of the
palace, and facing this - again off-centre - a secondary citadel
gateway. Elsewhere within these inner walls, Loud excavated five
minor palaces, whose plans had been adapted with obvious
difficulty to the sites available. This curious absence of preliminary
planning would appear to have been characteristic of monumental
building at all periods in Mesopotamian history.
In the main palace, as we have said, the ceremonial apartments
and state-rooms were planned around an inner courtyard,
dominated on one side by sculptural composition of winged bulls
and other figures, guarding the three entrances to the throne-room.
This huge rectangular chamber had walls about 12 high, m
decorated with mural paintings from pavement to ceiling, and at
one end a throne emplacement with relief carvings. At the other, a
vestibule led to a stairway, giving access to the flat roof, and a
doorway near the throne brought one through a long robing-room
to the private courtyard around which the residential apartments
were grouped. This arrangement of a throne-room with subsidiary
chambers and stairway seems to have been an architectural
convention universally adopted in all Assyrian palaces, and here at
Khorsabad it can again be recognized in the 'reception suites' of
minor buildings in the lower citadel. A second convention of the
same sort dictates the composition of an isolated group of state-
rooms on the northwest side of Sargon's palace, where doorways
202
2 3
lead out on to an open terrace. Loud found an identical

arrangement of chambers in the residential part of the southern
arsenal. ^5
1
Where religious buildings are concerned, the Nabu temple with

its two outer courtyards and conventionally planned sanctuary is
the most conspicuous. But Loud also re-excavated the assemblage
of three smaller temples annexed to the palace on the southwest
side; and one of these - the Sin Temple - revealed a well-preserved
facade. Reeded panels on either side of the arched doorway^ sprang
from small projecting platforms, faced with glazed-brick ornament
in bright colours. Beside each of these was a tall artificial palm-tree
and a female statue of the 'flowing-vase' type - all extremely
reminiscent of the mural painting from Mari on which we have
earlier commented. The small ziggurat adjoining these temples was
too denuded to examine further. But if Place's reconstruction can
be relied upon, it had a spiral ascending stairway dividing it into
successive stages, each painted a different colour. ^^
Regarding the arsenal-palace {ekal masharti) at Khorsabad, which
Loud took to be the residence of Sennacherib at the time when he
was crown-prince, one point of interest could now be established.
Stone column-bases found by Place in its ruins had brought to
mind an inscription in which Sargon claimed to have built a palace
of the Syrian kind, known as bit hilani, with a columned loggia in
front of the throne-room. But Loud was able to show that the
position of the columns in the building did not conform to this
theory, and Sargon's hilani must therefore be sought elsewhere. ^^
Little more need be added about the plan, since, as we shall
presently see, an arsenal building of the same sort was later
completely excavated by Mallowan at Nimrud.
Building Construction
At Khorsabad, walls of every description were built of mud brick.

Contrary to the usual practice, mortar was here very seldom used,
the bricks being only partially dried after casting, and laid in a soft,
pliable condition. Kiln-baked bricks were used in great quantities
for facings and pavements. The city walls, which were over 20 m
thick, were revetted at their base with dressed stonework up to a
height of 1. 10 m. Behind this facing, undressed stone was roughly
laid to form a base for the brick upper structure, which terminated
in a crenellated parapet with stone merlons. Here, as elsewhere in
Assyria, stone which may be described as gypseous alabaster (now
known as 'Mosul marble'), was easily obtained from local quarries,
if necessary in very large blocks. The palace platform had a facing
of stone in blocks up to 2.7 m long, weighing as much as 23 tons

apiece. For the rest, stone was mainly used for portal sculptures and
for the 'orthostat' slabs on which the reliefs were carved. These
latter were exclusively placed inside the building, to form an
ornamental 'dado' along the bases of the walls. The Americans

discovered that the rows of slabs were set in place and their reliefs
carved in situ, before the brick upper structure was built. ^^ 4 Where
slabs occurred on both faces of a wall, the space in between was
filled with rubble, composed partly of masons' chippings. Another
203
5
145 One ot a pair ot winged and use oi stone was for the thresholds of important doorways, where
human-headed bulls with
huge slabswere laid, carved with designs representing carpets. ^^
attendant genii from the
southwest gate into the citadel at
With regard to roofing, traces were found in the city gates of
Khorsabad (now at the entrance barrel-vaulting in brick, and 'pitched-brick' vaulting was used in
to the Iraq Museum in Baghdad) the drainage system of the palace platform. But in the palace itself
Loud satisfied himself that flat ceilings were the rule, sometimes
with painted beams. For wood was comparatively plentiful. Cedar,
cypress, juniper and maple were among the species recognized, and
from fallen beams he calculated that it was available in scantlings
capable of spanning the widest chambers.
Late Assyrian Sculpture
Sculptures in-the-round of this period are for some reason

remarkably rare. A few isolated portraits of royalty have survived,
the best-preserved examples being rather more than half life-size.
That of Ashur-nasirpal II, found in the Ninurta Temple at Nimrud,

which may be taken as typical, has been described as 'dull and
impersonal' yet the figure of this king, bare-headed and draped in
;
the simple fringed shawl which was the court-dress of Assyria, has an
indisputable dignity. ^ ^ 6
204
:
By contrast, the most conspicuous and characteristic class of

sculptures in this category are the lamassu guardian figures with H5
which the gateways and palaces of Assyrian cities were adorned.
These hybrid colossi - winged and human-headed bulls or lions -
were usually 'double-aspect' figures, set facing outwards against the
revealsof doorways, with a fifth leg, intended to rationalize their
appearance from in front as well as from the side. Each of these
monolithic monsters was carved, partly in relief and part in-the-
round, from a single slab of stone measuring up to 5.5 m square.
Roughly shaped in the quarry, it was transported to its destination, 146

often by river, and set in place for the final carving to be done. ^^^
The most spectacular achievement of Late Assyrian sculptors was
in the realm of relief-carving. Their contrivance of pictorial designs
in this medium for the decoration of interior wall surfaces was in
itself a great technical accomplishment; but their capacity for
abstract expression elevated the products of their craftsmanship
into the realm of creative art.
As for the subject of the reliefs, from the time of Ashur-nasirpal
onwards, they were primarily concerned with the king's victorious
campaigns against the armies and cities of dissident dependencies.
This theme tends to become almost monotonously repetitive. As
Frankfort has put it: 'We see the march of armies, subjugating,
burning and punishing in country after country.' The war-chariot
is a recurrent motif, and one sees them advancing against a
146 Relief from Sennacherib's
retreating enemy, while infantry dispatch the wounded, left upon southwest palace at Nineveh,
the field. 'Opposition is centred on a city; it is taken, its leaders are depicting the transport by raft of
impaled or killed in other ways, then the inexorable chariotry a winged bull {lamassu) figure
presses on again.' Variety is obtained by suggesting the geographi- from a quarry in wooded
country. The sculpture has
cal background against which these events take place, and the already been roughly shaped and
national characteristics of Assyria's various enemies. Details of these will be completed in situ.
are carefully and vividly presented, with a reality which is the more Ht 135 cm
m 'r> m ^«^»^m€i t4^mmm.M4r..
\^
\\'^
/('
K
205
;
147 Relief from the 'Central remarkable one remembers the limitations of the medium in
if
Palace' of Tiglath-Pileser III
which these worked, and their ignorance where the
artists
(744-^27 Bc) at Nimrud,
principles of perspective were concerned. Under these circum-
showing an attack on a walled
city. Scaling ladders and a stances, the ingenious devices by which spatial recession could be
battering-ram are used, with suggested, particularly in depicting landscapes, are unparalleled in
covering fire from archers. In a
pre-Greek art.
mountainous setting (below)
enemies are slain and (above)
On the whole, the king's military campaigns took place during
naked captives impaled on stakes fixed seasons of the year. During the intervals between these, he
would seek distraction in hunting, and this was another subject
148 Rehef from the 'Northwest much favoured in the reliefs. It is perhaps best illustrated in the
Palace' of Ashur-nasirpal II (883-
is one good sequence of
carvings of a later historical phase, but there
859 BC) at Nimrud. Figures are
shown of the King, worshipping hunting pictures from the time of Ashur-nasirpal. ^^8 xhese do not
before a 'sacred tree', supported in fact represent an event taking place in open country, but in an
by winged gods, each holding a enclosure formed by the shields of soldiers, within which lions were
cone and situla (metal bucket)
released to be shot at by the king from his chariot. After the hunt he
above, the god Ashur in a winged
is seen pouring libations over the bodies of his prey, attended by
sun-disc (compare the slightly
earher seal design in ill. 133)
courtiers and musicians. Usually at this time, the slabs were about
2 m
high and the pictures arranged in two horizontal registers. But
149 Reliefs from the 'North the culminating scene in this sequence, where the king rests and
Palace' of Ashur-banipal (668-
takes refreshment after the hunt, occupies the full height of the
627 Bc) Nineveh. Above, a
at
Hon-hunt below, a
in progress ;
stone. It is a sort of epilogue, in which we are reminded of the
ceremony in which the king divine protection which he enjoys, by the figures of benevolent
pours libations over his victims genii on either side. Similar figures, facing a 'tree-of-life', appear
again in a great self-contained panel behind the actual throne in the
throne-room of the Northwest Palace at Nimrud. ^^q ^^s the jj-
148 character of a rather splendid wall-tapestry, and its composition is

repeated elsewhere in the embroidered design on Ashur-nasirpal's
mantle. ^^*^ Strangely enough religious svmbolism of any sort
appears only rarely in these sculptures.
In the reign of Shalmaneser III, a variant form of relief ornament
130-1 is seen for the first time, in the great bronze gates of Balawat. These
206
207
im tmsi-^g,jH^ Siii mm^'î-^.r^^m- %mS&»^<^^&Mm
«m
m\ irw«r«.m » » • n
^^
*
F^-ifci JhiiEX-'F z. m 4„
% «
m^
2£ M M •># >. , .gaMtta
T~- -
<
T TT .
1
t 1
T I
f
#
f^^JS
A
'
f'l -
\: «
k
'
!
»
ki-
re i»*t
«- ^
..JH %. %. a ,.,'
'
*T* ••t»
- . -
tl-±\
«*
'%tr%s:iA.
1 ^jm-miy-^
'^'i* tl"^ ''^'.Â'%
ITS
n 1^4
mi
• •• «r-<
150 Reconstruction of the are double doors of wood, with each 'leaf measuring 1.8 wide m
Balawat Gates (lower half shown
only) in the British Museum
by 6. 1 m high, and each
is supported at the side by a circular pivot-
shaft of the same material. They are ornamented with separate,

horizontal bands of bronze plating, 28 cm wide and hardly more
than 2 mm
thick, running from side to side and continuing around
the pivot-shaft. These bands are modelled by a repousse process with
scenes in relief similar to those elsewhere carved in stone, and, as
there are two registers to each strip, the height of the actual design
cannot exceed 13 cm. The scenes here chosen are once more of a
narrative or episodic character, but lively and revealing. ^^' A
victory is won over the Chaldeans among the palm-groves of
lower Mesopotamia on the Mediterranean coast, the island fortress
;
151 of Tyre is captured and the tribute of its ruler brought ashore in
boats to the Assyrian camp in southern Urartu, the army's advance ;
is hampered by hills and forests. But here in the north, a more

strange episode is also depicted. The king has discovered what he
takes to be the source of the Tigris, in a mountain cave. With the
soldiers up to their waists in water he makes appropriate sacrifices,
while a sculptor carves a rock-relief to commemorate the
occasion. ^^^ The third pair of gates from Balawat, now restored
and returned to the Iraq Museum, add new instalments to the same
pictorial history. ^^^
After the Assyrian revival in the mid-8th century bc, Tiglath-
pileser III built himself a provincial palace at Til Barsip (Tell
132 Ahmar) on which mural paintings were
the Euphrates in Syria, in
substituted for relief sculptures in stone. ^^"^ The designs were in red,
208
-^<^^S?»*«««^^s!S£!r''»"
151 A detail from the Balawat

Gates. An episode in
Shalmaneser Ill's campaign of
858 BC on the Levant coast. The
city of Tyre, on its rocky island
(top left), by boat
sends tribute
to placate the king. Below, he is
seen in his chariot, passing on to
further conquests, and his camp
(bottom left) is left empty.
Ht 27 cm
152 Detail from a mural

painting in the 'Governor's
on the
Palace' at Til Barsip
Middle Euphrates (reign of
Tiglath-Pileser III, 744-727 bc).
The murals (substitutes for stone
reUefs) are painted in black, red,
and blue on a white plaster
background
209
^F^ ,
3^ . -'^^^
i i f
7 \
Wl — L/.^~>
,^:•,^•- r
:-
1
n
tj!Xx.^\V^\^v;A^^ A \^^W/A\<.\\\x^ <^ ^^^^v^^C'Â%C<<^3^:S^SÂ\î vMWf^^''
210
blue and black on a background of white plaster, and their subjects 153 Relief carvings of a winged
god and gift-bearers from the
very similar to those of the stone reliefs. Episodes in the king's
Palace of Sargon II at Khorsabad.
military campaigns alternated with hunting scenes, and paintings of In this case the orthostat slabs are
winged genii took the place of portal sculptures. Reconstructed in over 3 m high
the Louvre, these murals were not reproduced in colour for
pubhcation until comparatively recently. ^^^
Stone reliefs from palaces of Tiglath-pileser's successors in the
8th and yth centuries show changing characteristics. In the state-
rooms at Khorsabad, narrative is replaced by static compositions, 153
representing court ceremonial. The huge figures of the king and his
courtiers, some 2.75 m high, follow or confront each other in
hierarchical formality, and one notices the coloured paint which is
sparingly applied to their hair, beards and the exposed parts of their
bodies. The possibilities offered by these larger slabs were later to be
exploited by the artists of Sennacherib's time for pictorial 154 (Right) Part of a sculptured
compositions of another sort. Historical episodes could now be scenefrom Sennacherib's palace
elaborated on a much less restricted scale and concurrent incidents at Nineveh, identified from
inscriptions as the capitulation of
added to a central theme. For this purpose, the division of a picture
Lakish in Palestine. In a hilly
into horizontal registers was often abandoned. A single design
landscape wîth vines and olives,
might now extend to the full height of the slabs, while new the King, seated on an elaborate
formulae could be devised to imply relative distance or dramatic throne before his tent, receives
priority. New details of landscape added to the panoramic effect of his commanders while Jewish
prisoners bow before him. His
these compositions.
chariot awaits him below
It would be impossible here to enumerate the many and various (Left) The captive people of
subjects chosen by the artists of this period. ^^^ Almost all are Lakish are tortured or deported,
concerned with violence and the destructive effects of punitive together with their families and
belongings. Early in the yth
warfare. Some are lent special interest by their accompanying
century BC attempts of this sort
inscriptions, which identify historical events and even individual
were made to evoke pictorially
personalities. After the capitulation of Lakish in Palestine, the setting in which events took
Sennacherib may be seen receiving his officers and their suppliant place
211
155 Relief from the 'North captives or he

; is shown, during one of his campaigns in the marshes
Palace' of Ashur-banipal at of southern Mesopotamia, watching from an island while his

Nineveh. In a vine-arbour, to the
soldiers pursue the enemy in boats through the reed-beds and fishy
sound of music, the king and
queen celebrate a victory over
waters of the lagoon. Ashur-banipal celebrates the capture of Susa
the Elamites. Hanging in a tree to the sound of music, with a quaint bird's-eye view of the city
(left) is the severed head of their behind him, contemplating the gruesome aftermath of heavy
enemy Te-umman, King of
fighting or accompanied by his wife, he relaxes on a couch in an
;
Elam
arbour of vines. ^^"^
156 Sandstone stela of Ashur- In this last phase of Assyrian relief carving, the drawing of
nasirpal II, whose portrait animals receives particular care and attention. It reaches a peak of
appears inset beneath the accomplishment in the time of Ashur-banipal, when hunting
symbols of his gods. Sin, Ashur,
scenes, depicting lions, wild asses or gazelles, became popular
Shamash etc. Around this figure
subjects. For here - perhaps deliberately - a note of pathos is
are 153 lines of inscription
celebrating the completion of his introduced, and one's sympathy is aroused for the 'dying lioness' or
Nimrud, in 879 BC and
capital, for the 'onager mare' who is compelled to abandon her foal.
recording the attendant festivities The more practical value of the reliefs, from an archaeological
viewpoint, is the wealth of detailed evidence which they provide
about the appearance and characteristics, not only of the Assyrians
but of neighbouring peoples with whom warfare brought them
into contact. Painstaking study of these pictorial records has by
now acquainted us with every smallest detail of military
equipment, and of common practices in its use which the written
texts alone could never have made clear. ^^^ Regarding those
aspects of Assyrian life which the reliefs illustrate, little remains to
be learnt.
British Return to Nimrud
Some account must now be given of the finds made by a British

expedition under the leadership of M. E. L. Mallowan, which
139-40 returned to Nimrud
in 1949 and worked there for a further
fourteen years. ^^^
Mallowan at first concentrated on the
Northwest Palace, where the 'central ceremonial' block of
buildings, excavated by Layard, had included a great throne-
room. ^^° His first important find was indeed made in a side-
chamber adjoining the throne itself This was a commemorative
136 stela of Ashur-nasirpal II, bearing an effigy of the king in relief and
212
213
an inscription of 153 lines covering both sides of the stone. It is a

monument of special interest, because it records the ceremonies and
arranged to celebrate the formal opening of the palace in
festivities
the year 879 bc, andit includes details of the food prepared for the
entertainment of 69,574 persons over a period often days. Other

information of this sort enabled Mallowan to calculate the
approximate population of Calah itself, had
after Ashur-nasirpal
made it his new capital. The which he arrived (a
figure at
minimum of 86,000 including children), makes an interesting
comparison with the 'six score thousand' inhabitants which are
attributed to Nineveh in the Old Testament {Jonah, IV, 11).
Meanwhile, in the eastern 'domestic wing' of the palace, two
deep wells were investigated. In one of these ('AB'), which had
been partly excavated by Layard, a fmd was made whose
significance became clear after it had been examined in the British
157 Museum laboratory. This was a 'book', composed of fifteen or
more 'leaves', joined by golden hinges to open inwards and
outwards, rather like a Japanese screen. The 'leaves', of wood and
ivory, had raised edges to protect a surface of wax, on which
inscriptionswere engraved in cuneiform. Together they composed
a written document of several thousand lines. The 'cover' bore the
name of Sargon II, together with the title of the text and a note to
say that the 'book' should be kept in the 'King's new palace at Dur
Sharrukin [Khorsabad]'. Until that time, contemporary evidence
of writing on a wax surface had been found only in Phrygia. But
Mallowan was reminded of a later Babylonian inscription, in
which the officer who read the omens to the king was instructed,
when the reading was over, to 'close the book' a phrase which was
;
difficult to understand, supposing that it was a tablet from which he

157 Reconstructed part of a was reading.
'book', found in a well at
It was from a second well in this part of the palace ('NN') that
Nimrud, consisting of wax-
covered 'boards' of ivory, hinged
Mallowan recovered several objects, now considered to be the
to unfold. The wax was of 'Nimrud Ivories', presently
greatest masterpieces in the category
darkened with a yellow
in colour to be discussed. Firstcame the strikingly beautiful female head,
'orpiment' (sulphide of arsenic)
perhaps an ornament from some piece of furniture, which came to
to make the inscription more
be known by the excavators as 'Mona Lisa'.^^' Measuring
easily readable. The of each
size
board was 33.8 x 15.6 cm. (From 16 X 13.3 cm, the ivory of which it was made had matured to a
Mallowan, 1966) warm brown colour and the features, lightly stained with
214
158 Ivory head of a woman,

perhaps from of
a piece
ornamental furniture, found in a

well beneath the 'Northwest
Nimrud. The hair,
Palace' at
eyebrows and pupils are stained a
darker colour. Known to the

excavators as the 'Mona Lisa',
this carving probably dates from

the late 8th century bc.
Ht 16 cm
159 Ivory plaque from the same

appropriate colouring, have been compared by Mallowan
well at Nimrud as ill. 158,
stylisticallv to those of archaic Greek sculptures. There was a second
depicting a lioness killing a
head of the same sort, much more crudely carved, and both are young negro in a thicket of lotus
dated to about 700 bc. Even more aesthetically effective than either and papyrus plants. The flowers
of these objects were the two ivory (or rather chryselephantine) are inlaid with lapis and
carnelian, mounted in gold by a
plaques, representing a lioness killing a negro, against a formal
cloisonne process ;
prominent
background of lotuses. ^'^^ They are about 10 cm high, elaborately features such as the boy's curls
enriched with gold leaf and coloured incrustations. The cloisonne and his trousers are gilded.
process by which they are decorated has been closely studied in the Ht 10.5 cm
British Museum. ^"^^
Other buildings of the citadel which Mallowan excavated
included the northern wing or Chancery of the Northwest Palace,
and the Ninurta Temple at the foot of the ziggurat; the southwest
complex, consisting of Loftus' Burnt Palace and the Nabu Temple
(Ezida) with the neighbouring Governor's Palace; a large group of
private dwellings and the great Quay Wall, supporting the western
or river-side of the citadel platform. All these building remains
were methodically recorded and successive phases in their
architectural history distinguished. During these years, Mallowan
215
S.E. PALACE AND

NABU TEMPLE
i6o Projected plan of the 'Burnt was not primarily searching for large statues or reliefs, but his work
Palace' at Nimrud {c. 612 bc),
was amply rewarded by fmds on a smaller scale bronze vessels or
:
where ivories were first

weapons, fme ivory carvings and deposits of unbaked tablets,
discovered by Loftus in 1854.
(From Mallowan, 1966) objects of a sort which earlier excavators had been incapable of
preserving. But his most notable discoveries were made after the
161 Plan of Fort Shalmaneser at expedition had transferred its activities to the huge, and hitherto
Nimrud. To the north and west
unexcavated, group of buildings in the remote southeast corner of
are three courtyards, surrounded
by workshops, stores and offices.
the outer town. This was the 'Imperial Arsenal' (ekal masharti) of
One of these (southwest) is Shalmaneser III.^"^^
divided into long magazines, one Buildings answering tothis description have been found to exist
of which was the main Source of in all three of the Late Assyrian capitals. The arsenal at Nineveh
ivory. Separated from these by a
remains buried beneath the modern buildings of Nebi Yunus; but
parade ground is the king's
residence, composed of a throne- an inscribed prism from that quarter gives a description of its
room and standard suites of royal reconstruction in the reign of Esarhaddon.^^^ Another passage
apartments. defines the purpose served by such a building, stating that it was 'for
the ordinance of the camp, for the maintenance of the stallions,
mules, chariots, weapons, equipment of war and spoils of the foe of
every kind'. It also records how, annually at the time of the New
Year's Festival, the establishment was subject to inspection by the
king himself, for whom residential quarters were provided.
The Arsenal at Nimrud, known to the excavators as Fort
Idl Shalmaneser, occupied a rectangular site measuring 300 x 200 m,
enclosed on the east and south sides by the city wall and elsewhere
by its own massive fortifications. David Oates, who took charge of
itsexcavation in 1958, thought that it had also been provided with

an 'outer bailey', and there is evidence of a similar arrangement at
Khorsabad. The was divided into five main sections.
fortress itself
In the centre there was
broad parade-ground, with a throne-dais
a
on one side for the king. Two large courtyards to the north of this
were surrounded by the functional accommodation of an official
216
217
162 Part of the huge stone base

for the king's throne at Fort
Shalmaneser, sculptured in reUef
allround with processions of
tribute bearersand court officials.
Here in the centre Shalmaneser
himself is seen greeting with a
handshake Marduk-Zakir-
Shumi, King of Babylon
entrepot: workshops for carpenters, smiths, leather-workers or

armourers, often with their tools and appliances preserved among
the litter of scale-armour and, in one case, a royal statue under
repair. Elsewhere, stores for food and wine adjoined the
administrative offices and adjutant's quarters. A third court to the
west of the parade was divided into spacious magazines, arranged
around smaller light-wells. The entire south side of the building
was occupied by the royal residence, of which the ceremonial
section, with its entrance from the parade-ground, was composed
of two elements, their plans conforming precisely to architectural
conventions which we have already observed at Khorsabad and
elsewhere: namely, a 'reception suite' with its huge rectangular
throne-room and a projecting wing containing state apartments.
In the throne-room the stone dais itself had survived, and proved
to be decorated with spirited relief carvings representing the receipt
of tribute from Syria and Chaldaea. The principal subject was an
encounter between Shalmaneser and the contemporary ruler of
162 Babylon, in which the two kings actually clasp hands. ^'^^ Almost
equally impressive was the pictorial panel in glazed brick, fallen
from above the doorway in an adjoining chamber, in which the
figure of the Assyrian king appears in duplicate beneath a winged
disc. 2 77
At Fort Shalmaneser, however, the most rewarding discovery of

all was the contents of storage chambers in and around the
'southwest courtyard', on the western side of the parade-

ground. ^^^ They contained a vast collection of ivory carvings,
mostly of the sort which were used to decorate furniture, chariots
or parade harness. Piled or scattered at all levels in the debris which
filled the rooms, they consisted of plaques in high or low relief,
ajoure figures of animals or men and some subjects carved in-the-

round - the work of Phoenician or Syrian craftsmen, looted from
the cities of the west. From the thousands of fragments recovered,
some hundreds of items have now been restored and many of them
^"^^
published.
218
The Nimrud Ivories
We have already had reason to mention a study of Late Assyrian

ivories made by R. D. Barnett during the years preceding
Mallowan's excavation at Nimrud. ^^° This was based on a
collection which had been retained for almost a century in the
British Museum, and it proved to be derived from two sources:
Layard's finds in the Northwest Palace and Loftus' discoveries in
the Burnt Palace. Unlike these early excavators, Barnett was now
able to compare these ivories with more recent finds of tjie same
sort from sites in Palestine, Syria, Anatolia and even from Cyprus.
At an early stage in his study of their stylistic characteristics, it
became abundantly clear that in both groups the majority of
designs were the work of Phoenician craftsmen in the coastal cities
of the Levant, This was notably so in the group from the
Northwest Palace, which consisted mainly of furniture enrich-
ment, like the parts of a 'royal throne' said by Layard to have been
found in 'Well AB'. Some examples of 'standard designs', such as
the 'woman-at-the-window' (the Phoenician goddess Astarte) or 16
the 'cow turning its head to lick a suckling calf, were marked by 163
their makers with a sign corresponding to a letter in the Phoenician
alphabet. In others, 'Egyptianizing' designs pointed to the same
origin. By group Barnett professed also to
contrast, in the Loftus
recognize a 'Syrian' forming an independent unit associated
style,
with some more inland centre, in which non-Phoenician motifs
were used and the designs applied to different forms of object.
These included caryatid figures, model shrines and pyxis vessels for
ointment. He also suspected the presence in this group of objects
manufactured in Assyria itself, perhaps by imported craftsmen.
163 Ivory ornament with tenons

for attachment to furniture. The
Phoenician motif known as
'courtesan at a window' was
popular in the Levant and
elsewhere. This fragment, dated
c.700 B c is from the
,
'Northwest Palace' at Nimrud
219
164 Ivory figure, carved in the Barnett's Nimrud was eventually published in 1957 and
Ivories
round, of a Nubian leading an
during the years that followed, as we have seen, a great volume of
oryx while carrying a live
monkey and a leopard-skin. A new material became available. Mallowan, in the study of his own
'luxury object' from Fort fmds which appeared nine years later, was able to confirm many
Shalmaneser earlier conclusions; but he was cautious in accepting Barnett's firm
distinction between Phoenician and 'Syrian' work, on the grounds
165 Another motif, popular
that a free interchange of craftsmen may have taken place at that
among Phoenician ivory-carvers
over a period of several centuries time in markets west of the Euphrates. On the other hand, he
all
and with Egyptian overtones, is also was able to detect a whole school of imported artists and
the 'cow suckUng its calf. This indigenous carvers, owing little except their technique to foreign
ajoure (open-work) example is
influence. Regarding the purpose and setting of the ivories,
from Fort Shalmaneser {c. 700
BC)
Mallowan gained the impression that this highly prized, luxury
material was applied as a veneer to every conceivable form of
furniture. One room at Fort Shalmaneser (no. SW7) contained a
great number of panels, juxtaposed in their original setting, and
these at least could be identified as the elaborate ornamental backs
166 of beds, couches and chairs. In this connection, he could point to
the furniture so carefully depicted in a relief, already mentioned,
in which Ashur-banipal and his wife are resting after the capture
of Susa.He lists some of the other purposes and classes of objects
164 for which ivory carving was used, and makes an interesting com-
parison with a surviving inventory of spoil, brought back by
Sargon II from the cities of Urartu, where ivory was as popular
as in Assyria itself.
In considering the Levantine style of carving and its background,

one is reminded that the Phoenician craftsmen by whom it was
220
perfected were renowned, even in the Old Testament, for their skill
and ingenuity. Nevertheless, where design was concerned, one also
notices that they lacked any authentically native tradition. Just as in
Europe the artists of the Renaissance turned to ancient Greece for
their inspiration, so for the Phoenicians, Egyptian art provided the
source of a ready-made idiom. Egyptian art, however, was based
on religious imagery and on a huge repertory of symbolic and
mythical forms. For the Phoenicians neither the religion nor the
symbolism was their own, and little attempt was made to
understand them. Their adaptation of Egyptian imagery .was
accordingly often incorrect and occasionally clumsy. Surprisingly,
^^'
its visual effect remained unimpaired.
Chronologically, the Nimrud ivories could logically be dated to

any period between Ashur-nasirpal's reign and the eventual 166 A magnificent ivory bed-
destruction of his capital. Adad-nirari III is known to have acquired head from Fort Shalmaneser,
an ivory bed and a couch from the king of Damascus in about 800 composed of 12 panels, each
with the figure of a bearded
BC. But Mallowan considers the bulk of his fmds to be later than
warrior and magical tree, those
this and dates much of it to the reign of Sargon II. The extinction of
in the centre surmounted by sun-
the Syrian elephant towards the end of the 8th century may have discs. Dated by Mallowan to
increased the shortage of ivory. 730 BC
221
Chapter Ten
Babylon: The Last Mesopotamian

Monarchy
A Dynastic Revival
After the fall of Nineveh, the Medes withdrew to consolidate their

conquests in eastern Anatolia, leaving the Babylonian ruler in full
control of Assyria. For the moment, however, Nabopolassar was
compelled to devote his energies primarily to the revival of the
southern Mesopotamian economy and the restoration of cities
whose civic amenities had been neglected during the period of
Assyrian domination. Meanwhile, the most formidable threat to
the newly acquired Assyrian provinces was from the Egyptians,
who had already established a military outpost on the Euphrates
crossings at Carchemish. It was the crown-prince, Nebuchadrezzar
('Nebuchadnezzar' of the Old Testament), who was entrusted with
the task of confronting their armies and recovering the cities of
northern Syria. In this he was initially successful but, long after he
had himself become king, he continued to be involved in a
perpetual struggle to protect the trade-routes on which the
prosperity of Babylon depended. It was during one of his successful
campaigns in the Levant that the city of Jerusalem was taken and
some thousands of Jews deported to Babylonia. At home, during
the intervals of peace, he devoted much of his time to completing
the work of his father Nabopolassar, in rebuilding the old
Sumerian cities; and at Babylon itself, the magnificent layout of
pretentious public buildings and fortifications which have been
revealed in our own time can be attributed for the most part to
Nebuchadnezzar's ambitious planning and precocious ingenuity.
As for Nabonidus, the last independent ruler of Babylon, he
appears by contrast to have been both eccentric and ineffectual a :
'deviationist', whose enigmatic behaviour and frequent absence

from his kingdom remain to this day something of a puzzle to
historians. Apart from his rejection of Marduk, the time-honoured
patron of Babylon, in favour of the god Sin, whom he worshipped
at Harran, recently found inscriptions have provided an account of
^^^
his prolonged and totally inexplicable sojourn in central Arabia.
A group of historical relics, found at Ur, were identified by
WooUey as Nabonidus' private collection of antiquities.
At home in Babylon during Nabonidus' reign, political events
were overshadowed by the ominous expansion of the Persian
empire under its first great Achaemenid ruler. In 539 b c, the armies
of Cyrus II, supported by dissident peoples of Assyria, marched
upon Babylonia. Frontier defences were hastily manned by the
222
Babylon : The Last Mesopotamian Monarchy
crown-prince (Belshazzar of the Old Testament), who had

deputized for Nabonidus during his long absences but these were;
soon overrun, and within days the city of Babylon itself was
peacefully occupied, almost without opposition, by the Persian
army.^^^ On instructions from Cyrus, violence was avoided and
where possible the inhabitants persuaded to accept their new ruler
as some sort of 'liberator'. Three thousand years of self-rule in
Mesopotamia thus ended in a curious anti-climax.
We have three different sources of information regarding the
city of Babylon as it existed during these fmal years of
independence. One is to be found in the writings of the Neo-
Babylonian rulers themselves, which include (in a document
known as the Steinplatten inscription), Nebuchadnezzar's own
account of how he built the city. But at this time in the 6th century
BC there is another form of evidence which would not previously
have been available namely the eye-witness accounts, written by
:
contemporary travellers of other nationalities who had actually

visited Babylon. From the Hebrew writers of the Old Testament
little can be learnt in this respect, since their short and unwelcome
association with the city apparently left no time for detached

observation. In the case of the Greek travel-writers, however, the
situation is quite the reverse. Herodotus, and to a lesser extent
Ctesias, have bequeathed to us long and detailed descriptions, not
only of the city and its buildings but of the 'manners and customs of
^^'^
the Babylonians'.
Textual information of this sort was of course available to the
earliest explorers of the actual ruins in the 19th century, from
Claudius James Rich onwards. ^^^ gm; their actual excavation had
to await the turn of that century, when Robert Koldewey and his
German colleagues submitted the site to a systematic investigation,
which lasted thirteen years. The
of their excavations are the
results
thirdand by far the most important source of our present
knowledge about Babylon. ^^^
The German Excavation
The patience and ingenuity which the German excavators devoted

to their prolonged task Babylon have been remembered with
at
admiration by later generations of archaeologists and are manifest
in the results of their work. Koldewey was well provided with
architectural assistants, and there was some good-humoured rivalry
between them and the epigraphic specialists upon whom he relied
to interpret the evidence of the written texts. Yet it is to the
collaboration of both that we owe the remarkably detailed
information available today, regarding every aspect of a city of
which, hardly a century ago, few traces remained visible above
ground. Koldewey and Andrae between them were the first
excavators in the Near Eastern field properly to understand the
problems of stratification, and they were the first to train a gang of
expert wall-tracers. From most points of view, their methods
attained a higher standard of efficiency than anyone could have
expected at that early period in the history of archaeological
technique. Yet, with all this industrious activity and the expense
223
which it involved, Koldewey, at least, could never afterwards

regard his excavations at Babylon as having been in any sense
completed. When he started he had estimated that five years would
be necessary for the work, yet, fourteen years later in 191 3, one
finds him writing in one of his reports that 'approximately one half
of the work considered necessary or desirable has now been
completed'. The German excavations at Babylon were in fact never
afterwards resumed on any considerable scale.
It should'be remembered that Koldewey's interest in the site was
not originally confined to the ruins of the Neo-Babylonian town.

He had hoped to find, buried beneath them, remains of the 2nd-
millennium city which had been the capital of Hammurabi's
empire. By the end of his first campaign, however, he already
realized that in this respect a disappointment was in store, owing to
the high level of the sub-surface water-table. Instead, therefore, he
resolved to extend his excavations laterally over the whole area of
the later city, in the hope of recovering as much of its plan as
possible. This of course proved to be an immense task, which
involved continuous residence at the site in all seasons, interrupted
only by rare visits to his home.
Some idea of the enormous scale on which the 6th-century
fortifications of Babylon were built can be inferred from
Herodotus' own first-hand account. He gives an overall description
of the area enclosed by the walls, adding a characteristically careful
explanation of their disposal and construction, with detailed
dimensions which have proved to be approximately correct. One
needs only to glance at the site-plans in the German publication to
see the Greek traveller's impressions transformed into an accurate
topographical record. ^^^
The City
The old 'Inner City' can be seen on the right bank of the Euphrates,
making an irregular square, with sides rather more than a mile
167 Babylon at the time of

Nebuchadnezzar II (605-562 bc)
with fortifications extended to
enclose the Summer Palace (a)
in the north. (After R. Leacroft)
224
Babylon
J
long. In Nebuchadnezzar's time these walls were extended to 168 Plan of the old Inner City of
Babylon, with the western
enclose a further large area on the western side of the river, and the
suburb, showing the Procession
square thus became a rectangle with a longest dimension of i^
and principal buildings.
Street
miles. In the western sector the excavations were restricted to (From Hawkes, 1974)
soundings, whereby it was possible to establish the extent of the
walls and to obtain some idea of the street planning. For the rest, the
work was concentrated on the older part of the town to the east;
and here a very careful study was made of the fortifications. They
consisted of a double line of walls, the innermost of which had a 168
thickness of 6.5 m and was probably higher than the outer one,
which measured no more than 3.7 m. Both were built of sun-dried
bricks and strengthened at intervals by projecting towers, doubtless
with turrets projecting above the crenellated battlements. The
space between them, as Herodotus had recorded, gave room for a
protected military roadway at parapet level. At the foot of the
outer wall there was a moat, which varied from 20 m to 80 m wide,
leaving and returning to the river at either end. The 'scarp' or inner
face of the moat was strengthened by a wall of kiln-baked bricks,
set in bitumen. Koldewey also confirmed the testimony of
Herodotus, that 'there was free passage for boats and craft of all
kinds around the moat', and one is left to assume that it must
225
;
therefore have been bridged where the main gateways occurred.

There were in all nine of these, named after the patron gods of other
cities to which the roads led, and they were provided with outer
guard-rooms and roomy inner gate-chambers. Four of them were

excavated.
The whole of this inner fortification was built by Nabopolassar
and Nebuchadnezzar between the years 625 and 562 bc. Along the
east bank of the river, where streets came down to the waterside, a
new wall was later built by Nabonidus with a broad quay for
shipping at its foot.
1 72 The mainpalace of the Babylonian kings was situated in the
northwest corner of the old city, at a point where the walls
themselves were further protected by a colossal bastion overlook-
ing the river and an external citadel. It was here also that the great
171 highway from the north passed through the famous 'Ishtar Gate'
into the inner city, of which it became the main artery. This
'Procession Street', along which the statues of the gods were carried
at the time of the New Year's Festival, continued southward along
the eastern wall of the palace, crossed a canal called Lihilhegalla and
skirted the walls of the ziggurat enclosure, Etemenanki. At the city
centre it turned westward, passing on its left the great temple of
Marduk, Esagila, and came down to the Euphrates, which it crossed
upon a bridge supported by five stone piers, to enter the western
sector of the city.^^^
Finally, we must refer to the outer enceinte, which created a
huge extension of the city on the eastern bank. This great moated
rampart was built to enclose, in the far north, Nebuchadnezzar's
1 67 'Summer Palace', and it returned to the river south of the inner city.
It has sometimes been thought that it had been intended to repeat
this outer triangle on the west side of the river, making the enlarged
city into a gigantic square; but there is no evidence that this was
ever done. The two short sides of the existing triangle measure 2^
miles each ; so the complete square would have enclosed 6^ square
miles of country.
Buildings
The northern highway rises steeply before reaching the Ishtar Gate
so that the gate itselfand the Procession Street are paved at a level
several metres higher than the surrounding buildings. The
pavement itself was of limestone slabs, over one metre square, and
bordered with flags of red breccia. Beneath this, the Germans were
able to expose the foundations of the gate and street-walls and they ;
found that Nebuchadnezzar (as mentioned in his Steinplatten

inscription), had carried them down to a depth of 15 m: almost as
far as the clean soil beneath the mound. Since the quarrying
activities of local builders before Koldewey's arrival had left
practically nothing standing above pavement level, these foun-
dations are today all that remains to be seen by modern visitors to
the site. They did, however, enable the excavators to recover the
plan of the gate itself, with its imposing towers and high vaulted
inner chambers, as can be judged, by the remarkable reconstruction
afterwards made in the Berlin Museum.
226
^^/^^/^•7^^•^/v/ \--Â A/'-y V \/\ v \y 'y

/ '-y^v/^v/ v / 's7
-^ Drawing of the restored

"f-^-^^ b^#l^p^^^' •lf^ 169
Ishtar Gate, as seenfrom the
north. The raised Procession
t ^ '
:
'
Street is shown in section, with

i^-iSS^*""; '^^ ^. -:-;.. ^
f- -
J decorated foundations beneath it.
f;^;.~r
"^
1^=.^;^:"'
-^
.1 ,. '::'^ : # Xy 'M (From Koldewey, 1926)
70 Relief figure of a
,, "mr, :
•
f: ^-v-^..'r 1
dragon in
sirrush
moulded brickwork,
BiW^^^-î-r^B^^BII^BBBBI^WBI^^BMBBl from the facade of the Ishtar
Gate at Babyloii. The rehefs
^^HBI^B^^^fei^^- ,-':.^. '^j^y:^^^^^^^^!^^^^^^^ were covered with brightly
coloured glaze brown and
Bf
^f^^^^^K"' j, ;
yellow on a background of
:
^^--.W-W^^^^^-^^-, peacock blue. Ht 92 cm
^:~^^^«-Ti.^' K-#-"I^
"'-!»V.
j^*-#:"^
"•
i:»~^"~^
;
~-î[^ '^'\^>^'-A
l--''' :
- 1 - T - f-l
Now,regarding the facade ornament for which these buildings
are so justly famous: thewhole of the gate itself and the adjoining
walls of the Procession Street had outer faces ornamented with
designs in brightly coloured glazed brickwork. Repeated at regular
intervalswere the figures of Hons, bulls and sirrush dragons, 170
sometimes represented in flat glaze - at others in glazed rehef The
reconstruction of these designs was made possible by the survival of
a small section which remained standing at the base of one tower.
As to the process by which this brick ornament was glazed and
227
^4!
m^-.?^
171 Reconstructions in the assembled, Koldewey in his report speculates at great length, but
Berlin Museum of (left) the with a certain lack of clarity. One is left to suppose that it resembled
ornamental facâdes of the Ishtar
in most respects that used at Khorsabad and in Fort
by the Assyrians
Gate, and (right) a glazed brick
panel from the outer face of the Shalmaneser Nimrud.
at
great throne-room, facing the The same form of facade ornament seems to have been used in
thirdand largest courtyard in certain parts of the royal palace ('Southern Citadel', as the Germans
Nebuchadnezzar's palace
refer to it). A tall panel recovered from a wall outside the throne-
room is of special interest, in that classical motifs are adapted to the
design, suggesting that some contact now existed between Babylon
and Greece. Otherwise, the planning and architecture of this
building show no departure from Mesopotamian tradition except
for its colossal size. There were five successive courtyards, arranged
on a single axis, and the largest of these (63 x 58 m) in the centre of
the building, was entered through a sort of 'Sublime Porte'. On its
south side, and occupying rather more than its complete width, was
the principal throne-room, which, Hke all the other important
reception rooms, was thus given a northward exposure for the sake
of coolness in summer. The remaining space was filled with
domestic and administrative offices. One other feature, in the
northeast corner of the building, aroused special interest. Here was
a self-contained group of subterranean store-chambers, heavily
vaulted and containing much fallen stonework. It was thought by
the excavators that this curious structure might represent an
emplacement on which the famous 'Hanging Gardens' could have
been built, as described by Herodotus. Their theory was given
228
increased credibility by the discovery at one point of a deep well,

surmounted by a triple-shaft chamber, which might have accom-
modated apparatus for drawing water for the gardens above. In
more recent years, however, scholars have found reasons to doubt
this interpretation.
As we have already mentioned, the Procession Street, in its
passage westward towards the bridge, passes between the ziggurat

enclosure, Etemenanki and the Marduk Temple, Esagila, which
and tieftempel of Mesopot-
respectively represent the hochtempel
amian religious convention. If one remembers that these two
buildings must have demonstrated the culmination of a great
architectural tradition, it is distressing to realize that for all practical
purposes their physical remains are today non-existent. The
ziggurat enclosure, more than 500 m square, has an inner and an
outer temenos wall. Around these a complex of administrative
buildings were successfully excavated ; but in the case of the great
staged tower itself its structure had been totally quarried away by 172 Plan of Nebuchadnezzar's
brick-robbers, so that all that could be learnt about it archaeologi- palace, adjoining the Ishtar Gate.
cally had to be inferred from the negative impression left in the The throne-room, with its three
ground - today a deep, marshy pit. Based on Herodotus and other portals and dependent suite of
apartments, faces the great
ancient writers, a variety of reconstructions have been proposed, of
courtyard, which is approached
which the most convincing are those which most closely resemble from the through a 'sublime
east
the Third Dynasty tower at Ur.^^^ Esagila, on the other hand, porte'. The group of vaulted
presented a different problem. Its ruins were buried beneath 21m store-houses, once thought to
of later debris, topped by the Islamic shrine of Hajji Amran, have supported the 'Hanging
Gardens', is located in the
and its excavation, which involved the removal of 30,000 cubic
northeast corner of the building.
metres of earth, could be achieved only by tunnelling along the (After Koldewey, 1925). a, Ishtar
base of the walls. At least it was possible to determine the size of the Gate; b, Ninmakh Temple
/. - *- .--, V% -
229
building, which measured 86 x 78 m. But conflicting conclusions
about its planning become apparent when one observes that the
reconstruction in the final report and the great scale-model in the
Berlin Museum disagree, even in regard to the position of the main
sanctuary. As to its traditionally rich fiarnishings, one must once
more be content with the stories, among other marvels, of a seated
statue of gold, 6 m high, and a throne of gold, weighing i8y tons.
In addition to the Ninmakh Temple adjoining the Ishtar Gate,
which has already been mentioned, a number of other minor
shrineswere excavated. They can all be seen to conform to the now
generally accepted arrangement of entry, courtyard and sanctuary,
all on the same axis, but with variously developed subsidiary
chambers. In some respects more mteresting were the private

dwelling-houses, some with pretentious appointments of a sort
which one would expect in a capital city. Like town-houses in all
periods of Mesopotamian history their ground-floor rooms had no
outward exposure and obtained light from a central court. A
strange form of external facade ornament - if so it may be called - is
frequently to be seen in buildings of this sort, and would appear to
173 Pl^n of the Ninmakh
Temple, also near the Ishtar
have been an innovation in Neo-Babylonian times. Successive tiers
Gate: a building characteristic of of bricks are set at a slight angle to the true line of the wall-face,
the Late Babylonian period, creating a 'dog-tooth' pattern of vertical ridges. It is a practice
which has recently been which may originally have been adopted when an irregular shaped
reconstructed in situ. (After
site resulted in a disparity between the oblique angle of the outer
Macqueen, 1964)
wall and the rectangular rooms which it contained. For the rest,
Herodotus had spoken of houses with 'two or three storeys' and,
though the excavators found this difficult to confirm, it could at
least be seen that, like the much earlier houses excavated by
Woollcy at Ur, these too must have been provided with rooms on
an upper floor, entered and lighted from a wooden gallery around
the court. -'^'^ Private chapels, sometimes with burials beneath
them, suggested another aspect of Babylonian domestic life.
Evidence of large-scale building activities in the time of the Neo-
babyloman kings is by no means limited to the capital itself
Something has already been said about Nebuchadnezzar's
ambitious reconstruction o{ the city walls at Ur, with their
imposing gateways and protected harbours for ships trading on the
174 Facades of Neo-Babylonian Euphrates. On a new site in the city, a huge building known as the
temples at the summit of the Palace of Bel-shalti-nannar also appeared at this time;^^' while at
Harsagkalama mound at Kish,
Kish, on the Harsagkalama mound colossal temples were built,
still standing up to 5.6 m high.
Shown by foundation deposits to whose walls still remain standing many metres high. At almost
be the work of Nebuchadnezzar every other Sumerian city derelict stacks of kiln-baked bricks.
^ %i»¥l
stamped with Nebuchadnezzar's name, testify to his repair of some

ancient shrine and the belated renewal of its splendour.
In summarizing the results of the prolonged excavation at
Babylon, it is indeed surprising to find so httle to record in the
realm of sculpture or among the various products of contemporary
craftsmanship. The German excavators would have been the first to
admit their considerable disappointment in this respect; and it is
reflected in the commentary of a more recent writer. He says
'There is a scarcity of works of this period, which makes it
impossible to estimate its artistic achievement', and he addsf
There was an intense literary activity, and it was from Neo-Babylonian

sources that the Hellenistic world acquired its knowledge of astrology and
other Mesopotamian sciences; but Neo-Babylonian art did not affect the
West. It was from Assyria that Greece and Etruria obtained their models
during their 'orientalizing period', through the intermediacy of the
Phoenicians. ^^^
Postscript
In more recent years, new discoveries related to the theme of this

book have continued to be reported with gratifying regularity,
though the locations at which they were made - often beyond
the immediate territories of the old Mesopotamian kingdoms -
may point to a new widening of historical perspective. One pri-
mary instance may here be cited of an excavation, still in progress,
whose relevance in this respect may be thought to have particular
significance.
At the Mardikh, 3 1 miles south of Aleppo in Syria,
site called
the ruins have beenfound of the ancient city of Ebla, some of whose
rulers in the 3rd millennium bc were already known by name.
At a slightly later period, contemporary with the Larsa and
Amorite dynasties of Babylon, architecture and sculpture have
been found, which rivals in interest and accomplishment that cur-
rently known in Mesopotamia. But an even greater discovery has
now been added to these: an archive of 3rd millennium tablets
numbering many thousands, some of them each with over a
thousand lines of writing. Here then is a completely new field of
philological study. Written in a hitherto unfamiliar language of
the West Semitic type, their transcription has been facilitated by
the inclusion among them of plentiful vocabularies with Sumerian
equivalents. But, quite apart from their linguistic significance, the
archaeological setting in which they were found may in itself sug-
gest a genuine 'breakthrough' into a new realm of discovery. The
point which they serve to make clear is, in the excavator's own
words - that 'south Mesopotamia was not the sole progenitor of
literacy and civilization for the whole Near East similar processes;
were operating over a much wider area, from the Levant to

Elam'.293
Meanwhile, one new source of information, often relevant to
the earlier periods of Mesopotamian cultural development, has
recently been the rescue operations, conducted over areas about
to be flooded by new dams on the great rivers and their tributaries.
A typical instance of a site discovered in this way is Tell Moham-
med Arab, in the Eski Mosul dam area, which promises to be the
231
Postscript
type-site for further study of the controversial 'Ninevite Five'

period (see p. 132),contemporary with the Sumerian cities in the
south. Similarly, in Syria excavations above the Tabqa dam on
the Euphrates, east of Aleppo, have revealed strikingly impressive
sites of the Uruk period. At Habuba Kebira, German excavators
have found the remains of a huge enceinte, extending at least half-a-

kilometre along the river and enclosing a dense network of streets.
A major north-south roadway leads up to an acropolis (now
known as Tell Kaunas), where their Belgian colleagues have
exposed the remains of temples, planned like the 'White Temple'
at Uruk (ill. 18) or the 'Painted Temple' at 'Uqair (ill. 22). Others,
by contrast, are of the so-called 'cruciform' type, recalling the now-

famous examples at Uruk itself (ill. 20).
These discoveries, together with the appearance in south
Anatolia of plentiful Uruk-type and 'Ubaidian material, have once
more been taken to suggest a northern origin for the Uruk culture.
This issue has however been further confused by a simultaneous
discovery in the Hamrin basin, north of Baghdad. At Abada, near
Khaniqin (once more during a rescue operation), a prehistoric
settlement, contemporary with the earliest occupations at Eridu
and Samarra, revealed strikingly well-preserved architecture,
including a temple complex, incorporating multiple units of the
'cruciform' type, which appears to be repeated nearby in less
pretentious buildings - apparently common dwellings. The dis-
crepancy in date between the occupation of this village and the
occurrence of 'cruciform' sanctuaries mentioned above, presents
a problem which will now need to be considered. (Jasim (1983).)
Over the past century, as we have attempted to show in these
pages, the Mesopotamian contribution has been studied with
devoted thoroughness. As a specialized subject, the possibilities
which it offers are still unlimited but clearly it has everything to
;
gain from collateral research among the remoter byways of Near

Eastern antiquity.
232
Notes on the Text
1 The best commentary on the 21 An analysis and discussion of 36 Riemchen in German, meaning
geography of Iraq is in Brice all these soundings is presented 'prismatic', is the name used
(1966, chs. II and 21). in Perkins (1949), and by A. by the excavators for the
Porada in Ehrich (1965). small, rectangular mud bricks
2 But see Roux (i960, pp.
characteristic of the Uruk
20-31). 22 The German reports have
period.
been published at great length,
3 Lees and Falcon (1952, pp.
mostly in the series Uruk Vor- 37 Delougaz and Lloyd (1942).
24-39).
I'dujige Berichte (titles vary),
38 Mostly accounts and in-
4 Jacobsen (i960), pp. 174-83). and Abhandlungen der Deutsche
ventories.
His theory is at present sub Orient Gesellschaft, in progress
judice. since 1929. An almost up-to- 39 A full discussion of this script
date summary in English is to will be found in Falkenstein
5 See Adams (1958, pp. 101-4).
be found in Mallowan (1970, (1936).
6 See Jacobsen (1958a, p. 1251). ch. 8/1, p. 360 ff.), and in
Frankfort and Davies (1971, 39a See Schmandt-Besserat (1977).
7 See Oates, D. (1968, fig. 2, p.
ch. 12). But note also the more 40 Roux
14). (1964, p. 82).
recent find of a level IVa-b
8 See chronological table in So- palace, in Lenzen (1974, p. 41 General reference books for
lecki (1971, fig. 8, p. 182). III). cylinder-seals are Frankfort
(1939), Porada (1948) and
9 See Van Zeist (1977) and 23 Mackay (193 1) and elsewhere. Amiet (1961).
(1982).
24 See Lloyd and Safar (1947, p.
42 Some fine examples are illus-
9a Subsequent excavations at 84 ff. and 1948, p. 115 ff.); also
trated in Frankfort (1955, pi. i
Zarzi are reported in Wahida Lloyd (1947 and 1948).
ff.).
(1981).
25 Delougaz and Lloyd (1942).
10 Garrod (1930,
42a A basic work on the subject is
p. 8).
26 These and other proposed Porada (1948).
11 Solecki (1971); note in parti- changes in stratigraphical ter-
cular map on minology are well shown 43 Well illustrated in Delougaz
2, p. 17.
diagrammatically by E. Porada (1952).
12 Their work is fully reported in in Ehrich (1965, p. 176).
Braidwood and Howe (i960). 44 E.g. in Frankfort (1951).
27 Heinrich (1957).
45 See Falkenstein (1965, p. 269
13 Childe (1950), quoted in
28 Safar (1950, p. 27 ff).
ff.).
Singh (1974, p. 4).
29 Van Buren (1972). 46 See Lamberg-Karlovsky
14 âmbel and Braidwood
30 Ziegler (1953)- (1974)-
(1980).
31 Oates, Joan (i960, pp. 32-50). 47 Cf Kramer (1963, p. iii ff.).

15 Mellaart (1967).
32 See 'Excavations in Iraq,

48 Cf D. and J. Oates (1976, p.
16 Singh (1974); for a fuller ac-
136).
count see Mellaart (1975) or 1971-72', Iraq, 34, pt 2 (1972),
D. and J. Oates (1976). p. 149. 49 See Lloyd (i960).
33 Lloyd and Safar (1943, p. 131 50 Oates, Joan (1976, pp. 20-31),
17 Lamberg-Karlovski (1974).
ff.).
with map, fig. i, p. 23, and
18 Braidwood (1974b). Burkholder (1972, pp. 264-9).
34 Heinrich (1957).
Mortensen (1964,
The writer is indebted to Dr
19 p. 32 ff.).
35 Van Buren (1952, p. 76 ff.). Oates for the further infor-
20 Braidwood (i960, p. 173). Also see Moorey (1975). mation which follows.
233
1
Notes on the Text
51 For comparative analysis in

a 71 Munchajef and Merpert native dating in Ehrich (1954,
detail of the evidence from (1971, pi. V, fig. 2 and pi. VII, table on p. 176).
these excavations, in northern fig. 10).
86 This subject has been exhaus-
Iraq, see Mellaart (1975). Also
72 Oates, Joan (1969a) and illus- tively studied in Mallowan
to be recommended is the
trations reproduced in Singh (1964, p. 62 ff.). At Kish he
site-by-site commentary in
Mallowan (1974, figs. 59a-d, 60 and 61). noted several alternative
(1970, ch. 8/2, p.
'Flood-levels'.
376 ff.).
Typical figurines from these
73
52 Mallowan and Rose (1935). A two sites are illustrated side- 87 See Kramer (1956, p. 290).
subsequent excavation at by-side in Strommenger
88 See Sollberger (1954-6, p. 10
Arpachiyah by Ismail Hijara is (1964, pi. 6).
ff.).
reported in Iraq 42/2 (1980).
74 See Speiser (1935, pi. CXLII,
89 See Edzard (1959, p. 9 ff.).
53 See Speiser (1935) and Tobler p. 36) and Barnett (1966).
(1950). 90 See Kramer (1952).
75 Syria and Iran are covered
54 Herzfeld (1950). rather more fully in Mellaart 9 The relevant final pubHcations
Thompson (1975)- of the Khafaje excavations are,
55 and Mallowan
Delougaz (1940) and De-
(1932, p. 71 ff.). 76 Two of these are illustrated in
lougaz and Lloyd (1942).
colour in Strommenger (1964,
56 Lloyd and Safar (1945, p. 255
pis. Ill and IV). 92 For burials, see Delougaz et al.
ff.).
(1967).
57 Summary reports in English: 77 See Le Breton (1957, p. 79 ff.),
with diagrams of comparative 93 Delougaz and Lloyd (1942).
Munchajef and Merpert (1969
onwards). chronology.
94 Delougaz and Lloyd (1942).
58 Prehminary reports: Kirk- 78 See Lamberg-Karlovsky
95 Woolley's preliminary reports
bride (1972 onwards). (1970). It should be noted that
were published under the
recent excavations in Susiana
Of the animal bones 68-5% heading 'Excavations at Ur',
59 and at Deh Luran have now
were onager and 15-9% gain the Antiquaries Journal, 3
complicated the Susiana se-
zelle. (1923), to 14 (1934), and re-
quence and its relationship
pubhshed in a single volume
60 See Braidwood et al. (1952, p. with Iraq, as may be gathered
by Benn, London (Woolley
Iff.). from new synchronisms men-
1954). Of the final publi-
tioned by Oates in Curtis
61 Preliminary reports by B. cations, Woolley (1934),
(1983).
Abu-al-Soof and others in Woolley (1939) and Woolley
Sumer, 21 (1965) to 24 (1968). (1974) are relevant to this
79 Jacobsen (1939). Simply pre-
sented in Woolley (1965, pp. chapter.
62 See Oates, Joan (1966).
21-6).
and 96 Site-plans are reproduced in
63 See Oates, Joan (1969)
colour in Hawkes (1974, p.
Braidwood (1944, pp. 47-72). 80 A useful Enghsh translation is
173)-
Sandars (i960).
64 E.g. Yanik Tepe in Azar-
at
97 For the identity of these anim-
baijan Burney (1964). But see
: 81 There is a summary of the
als, see Dyson (i960, p. 102
Hijara (1980). evidence in Gadd (1971, ch.
ff.).
13, P- 93)-
65 Cf Mallowan (1970, ch. 8/2,
footnote no. 2 on
98 Cf Gadd (i960, p. 51).
p. 383). 82 Sollberger (i960,p. 69 ff.).
Recently found tablets from 99 See Hall and Woolley (1927)-
66 See Mallowan (1947, p. i ff.).
Abu Salabikh are closely com-
100 See Delougaz (1938, p. iff.).
67 See Braidwood (1944, pp. parable with those from Farah
47-72). (written by scribes with Ak- loi Mackay (1929) also Waterlin
:
kadian names). Cf Postgate and Langdon (1930-4).

68 Typical examples of Samarra
in Curtis (1983), p. 50.
and Tel Halaf painted pottery 102 See Moorey (1966, p. 83) and
are well illustrated in colour in 83 Hall and Woolley (1927)- Lloyd (1969, p. 40).
Strommenger (1964, pi. II).
84 See Kramer (1965). 103 De Sarzec and Heuzey
69 Lloyd (1938, p. 123 ff.). (1884-1912).
85 Delougaz and Lloyd (1942,
70 See Ucko (1968). fig. 115, p. 134), and alter- 104 Jacobsen (1958, pp. 127-9).
234
Notes on the Text
105 See Crawford (1974, P- 29 ff-)- 122 Illustrated in Frankfort (i954» 138 Delougaz and Lloyd (1942, pi.
pi. 9b). 23)-

106 Parrot (1948).
123 Illustrated in Frankfort (1954, 139 Delougaz and Lloyd (1942, pi.
[07 For easy reference, see site-
pi. 13). 12).
plan in Hawkes (i974, P- I74)-
See Frankfort (1939, footnotes 140 Delougaz and Lloyd (1942,

124
108 For preliminary reports, see
on pp. 51 and fig. 203, p. 263).
54).
Haines (1956, 1958, 1961), and
Hansen and Dales (1962). 141 Haines (1961).
125 Well illustrated in Strom-
SeeHansen (1965, pp. 201-13). menger (1964, pis. 88-100 and 142 Delougaz, Hill and Lloyd
109
in colour, pi. XX). (1967, pi. 36).
no See Hansen (1963).
126 Strommenger (1964, p. 401) 143 Strommenger (1964, fig. 17, p.
1 1 For plan in colour, see Hawkes refers to this individual, prob- 391)-
(1974. p. 181). ably rightly, as 'he'. But see 144 Safar (1950, fig. 3).
1 1 See Andrae ( 1 922) . The results also Barnett (1969, p. 96). A

145 Parrot (1972, fig. 2, p. 285).
of this excavation are sum- statue from the same prov-
marized in English in Mal- enance, showing a headless 146 The seals of all three phases are
lowan couple holding hands, repre- described and illustrated in

(1971, ch. 16, p. 298),
and in French (with plan) in sents a convention occasion- Frankfort (1939).
ff. and fig.
ally seen elsewhere.
Parrot (1946, p. 224 147 These two figures were as-
52). 127 Frankfort (1954, pi. 26b). sociated by an earlier gen-
Cf Edzard eration of archaeologists with
113 (1959, p. 9 ff-)-
128 See Hansen (1963, p. 146).
Gilgamesh and his companion
114 Preliminary on A.
reports
Enkidu of the famous epic;
129 Frankfort (1954, pi. 33b).
Parrot's excavations at Mari
but the equation is today trea-
appeared annually in Syria and 130 Frankfort (1954, pi. 33a).
ted with some reserve. Com-
other journals, from 1935 on- 131 Hansen (1963). He detected pare Frankfort (1939, p. 62).
wards. Their final publication two main phases in their stylis-
is in the series Mission development, equating the
tic 148 The most important com-
Archeologique de Mari, of with the earlier half of the
first mentary on objects under this
which vol. I (1956), vol. 2, pts 'ED ir period, and the second heading is to be found in
1-3 (1958), vol. 3 and vol. 4 with a transitional phase ex- Woolley (1934).
(both 1967) are relevant. Vol. tending to cover 'ED llla'.
149 The most satisfying recon-
2, pt 2 and vol. 4 are reviewed This suggests a sub-division of
struction of such a headdress is
in English in Mallowan the 'ED ir period, which
perhaps to be seen in the Iraq
(1969). The Ishtar Temple is would conform to the Ger-
Museum exhibit, mounted by
described (with plan) in Mal- man system of chronology
an Iraqi sculptor on an imag-
lowan (1971, p. 291 ff.). outlined above on p. no.
inary head and shoulders. It is
115 See Moortgat (1959 et seq.). 132 Hansen (1963, pi. VI). See also shown here, ill. 85. See also
The results are discussed in below, p. 131. Maxwell-Hyslop (i960).
Mallowan (1966a) and Mal- Illustrated in detail in Strom- 150 Frankfort (1954, pi. 20a).
133
lowan (1971, ch. 16).
menger (1964, pis. 66-9)
151 Best illustrated in Strom-
116 Mallowan (1971, ch. 16). 134 Lamberg-Karlovsky (1974, p. menger pi. and
(1964, 72,
117 E.g. Moortgat (1959 et seq.)

283). colour pis. X and XI).
and elsewhere. 135 Strommenger (1964, pis.
152 Parrot (1962, pis. 11 and 12
38-9). See also Delougaz and fig. II, p. 164).
118 Cf Ehrich (1965, p. 162). (i960).
136 The making and use of these 153 Compare the top left-hand
119 Cf Frankfort (1954, p. 23 ff.).
figure in the 'Peace Scene' of
All Early Dynastic sculpture bricks is explained in De-
the 'royal standard', ill. 87.
from the Diyala sites is pub- lougaz (1933)-
lished by the same writer in 137 E.g. in Sin Temple VIII at 154 Barnett (1969). The anatomy
Frankfort (1939 and 1943). Khafaje, in level VIII of the of these instruments, in re-
120 Andrae (1922).
Inanna Temple at Nippur and, lation to the known principles
most notably, in Palace 'A' at of Sumerian music, was first
121 SeeFrankfort (1939 and 1943). Kish. discussed in Galpin (1929). A
235
Notes on the Text
more up-to-date commentary 169 Delougaz, Hill and Lloyd 185 Woolley 's preliminary re-
on them is to be found in (1967, pi. 64). been said, were
ports, as has
Rimmer (i960). pubhshed immediately after
170 Delougaz, Hill and Lloyd the end of each season's
155 Strommenger (1964, pi. XIV (1967, p. 54 ff. and pi. 20).
!
in colour).
digging. Mallowan, in his '
Mallowan and introduction to the Memorial

156 A long and very detailed study 171 (1947, p. 26 ff.
63 ff.). Plan also reproduced in Volume, Iraq, vol. 22, pub-

of Early Dynastic pottery,
Strommenger lished soon after Woolley 's
with colour illustrations of (1964, fig. 18,
p. 403).
death in i960, recalls how
painted wares, at this and
these articles were often 'writ-
earlier periods, may be found 172 Mallowan (1932, pi. L), also on board
ten ship as he travel-
in Delougaz (1952). See also (1936), p. 104 ff. and pis.
led home'. For the titles of
Strommenger (1964, pis. VIII V-VII. Illustrated in colour in
volumes in the final publi-
and IX). Strommenger (1964, pis.
cation {Ur Excavations), see
XXII-XXIII).
157 SeeMalIowan(i964, p. I42ff.). Bibliography.
173 Originally pubhshed in
158 Strommenger (1964, pi. 41). 186 See Woolley (1974, pis. 53
Memoires du Delegation en
and 61).
159 George Roux (1964) com- Perse, vol. I, 1900, pi. X.
petently summarizes earlier 187 See Woolley (1939).
information on 174 E.g. two fragments published Woolley's several perspective
this subject.
Further evidence is now avail-
in Strommenger (1964, pis. restorations have been widely
able, e.g. in the form of texts

114-15). published and are well
from Abu Salabikh, mention- known.

175 Strommenger (1964, pis.
ed in note 82. 188 The only parallels in architec-
1 18-19), also discussed in Bar-
160 Saggs (1962, p. 272 ff.). The same might tural history for this practice
relet (1959).
are to be seen in the mausolea
be said of a fragmentary nude
161 These two rock reliefs are of Augustus and Hadrian at
statue, bearing a dedication to
discussed stylistically in Bar- Rome. See Bannister-Fletcher
Haram-Sin, more recently
relet (1959, p. 20). See also (1961, p. 237, figs, b and
found near Dohuk and il- c).
Strommenger (1936, p. 83 ff.).
lustrated in Oates (i979), See Woolley and
J. 189 Mallowan
162 Hilprecht (1903). P-35. (1974, pl- 54)-
163 Scheil (1902). 1 76 Compare Akkadian seals illus-

190 See Frankfort, Lloyd and
trated in Frankfort (1939),
164 Banks (1912). Jacobsen (1940, pl. i passim).
Moortgat (1940) and Wise-
165 Waterlin Langdon
and man (n.d.). 191 A good account of Mesopot-
(1930-4). See also Moorey amian history in the 2nd mil-
177 See Frankfort (1934).
(1966, pp. 18-51) and Lloyd lennium BC is to be found in
(1969, pp. 40-8). 178 Frankfort (1933, pp. 47-53, Roux (1964).
166 Woolley (1934)-
figs. 30-3), also Gadd (i933),
192 Published as the Archives
and Corbieau (1937, p. i ff.).
167 Delougaz, Hill and Lloyd Roy ales de Mari, Paris 1950
(1967, p. 186 ff. and pi. 37). 179 Starr (i939)- onwards (transliterations and
translations).
168 Delougaz, Hill and Lloyd 180 Lambert and Tournay (195 1).
(1967, p. 196 ff.). The present 193 See D. Oates (1968, p. 31 ff.).
181 De Sarzec and Heuzey (1884-
writer, who was responsible
1912), also Parrot (1948).
for the excavation of the Nor- 194 See Parrot (1938, pp. 308-10).
thern Palace at Tell Asmar, 182 What follows is partly taken
See McEwan
195 (1958).
finds the argument for this from an abbreviated account
theory unconvincing. De- in Parrot (1946, pp. 127 ff.). 196 Frankfort et al. (1940).
lougaz dates the building to a
183 These statues were first pub- 197 Turner (1968).
so-called 'Proto-Imperial'
lished in de Sarzec and Heuzey 198 Frankfort
period, defined by Th. Jacob- et al. (1940, pl. 7).
(1884-1912). They are well
sen, in Assyriological Studies,
illustrated in Strommenger 199 Frankfort (1936, p. 74 ff.).
no. II, table II, as lasting from
reign of the Lagashite
(1964, pis. XXVI and 133-7).
the 200 See Smith and Baqir (1946);
king, Entemena, to the first 184 Crawford (1974) and the ac- Baqir (1948 and 1959); Gur-
years of Sargon-of-Akkad. companying footnotes. ney (1950).
236
Notes on the Text
201 See Woolley (1954, fig- 12, p. cation (Starr 1938, vol. I, p. 528 238 Layard's two principal books
176). ff.). are Layard (1950 and 1953).
202 E.g. in Woolley (1935. pl- 222 Starr (1938, general plan in 239 Rassam's excavations are re-
12).
folder). corded, as a rule rather briefly,
Strommenger in Rassam (1897).

Preliminary reports in Syria 223 (1964, fig. 40,
203
(1935-67) and Annales Ar- P- 430). 240 Loftus (1857).
cheologiques de Syrie (1951-67). 224 Frankfort (1954, fig. 64, p. 241 Botta and Flandin (1849-50).
Final publication see Bibliog- :
142); also Mallowan (1939,
raphy under Parrot. pp. 887-94), and Ceccini 242 Place (1867-70).
(1965). Gadd (1936, p. 57). These
204 Illustrated in colour, e.g. in 243 See
Strommenger (1964, pis. 225 Strommenger (1964, fig. 179, two figures were purchased
XXVIII and XXIX) and in bottom row). from the French Consular
Agent in 1849 and sawn into
Parrot (1961). 226 Pubhshed by W. Andrae in
four pieces each. They were
Wissenschaftliche Veroffen-
205 See Barrelet (1950). transported safely by sea from
lichungen der deutschen Orient
Basrah to London, though the
206 See Parrot (1967a). Gesellschaft, no. 10 (1909),
ship was reported to be 'well
no. 23 (1913), no. 24 (1913),
207 Annual reports by Oates, D. no. 39 (1922), no. 58 (1935),
down in the water'.
in Iraq, 27-34 (1965-72). and summarized by him in 244 An early survey was pubhshed
208 See Oates, D. (1967, p. 70 ff.).
Andrae (1938). in Jones (1857).
He refers to similar facade 227 Andrae's site plan is repro-
ornament at Ur, 'flanking the 245 See Finch (1948, p. 9 ff.) and
duced in Strommenger (1964,
entrance to the bastion of Madhloum (1967 or 1968).
fig. 46, p. 434).
Warad-Sin, which was built
246 See plan in Layard (1853, vol.
228 Earlier excavators at Ashur
c. 1830 Bc'. Cf Woolley (1939, I, facing p. 67).
had included Rich, Layard,
P- 42-3).
Place, Rassam and finally 247 See plan of Ashurbanipal's
209 See Oates, D. (1973, p. 183). George Smith of the British palace in Rassam (1897, facing
Museum. See Lloyd (1955). p. 8); Balawat, King
for
210 All these sculptures are well
illustrated in Strommenger 229 Reproduced in Strommenger (191 5), also Barnett (1973) and
(1954, figs. 148-54). But see (1964, fig. 47, p. 435), from D. Oates on more recent ex-
R. Kupper in Revue Andrae (1938).
cavations in Iraq, 36 (1974).
J.
d'Assyriologie, 15 (1971), p.
230 Andrae (1925). Examples il- 248 See Loud (1936).
113 ff.
lustrated in Frankfort (1954,
211 The 'Burney Plaque', illus-
249 SeeLoudand Altmann (1938).
pis. 74 A-B).
trated here, ill. 118, and in
Loud and Altmann
250 Since
Frankfort (1954, fig. 56). 231 Frankfort (1954, p. 65).
(1938) is a heavy volume,
212 Strommenger (1964, pl. 232 See Frankfort (i939) and Loud's plans may more con-
xxvii). Porada (1948). veniently be consulted in e.g.
Strommenger (1964, pp.
213 Strommenger (1964, fig. 170).
233 Weber (1920, p. 105, no. 531). 445-7)-
214 Baqir (1944, 1945 and 1946a).
234 Moortgat (1944, P- 43, figs. 251 See Turner (1970, P- I77 ff-)-
215 Baqir (1946a). 45b and 46).

252 Reproduced in Frankfort
216 Baqir (1946a, pis. XI-XIV). 235 See Jacobsen and Lloyd (1954, fig. 32, p. 79)-
217 E.g. Frankfort (1954, pl. 71).

(1935). For other rock-reliefs
253 For the bit hilani, see Frankfort
of Sennacherib, see Bachmann
218 Baqir (1945, pis. 17 and 21). (1952)-
(1927)-
219 Frankfort (1954, pis. 70b and 254 See Loud and Altmann (1938,
236 Lloyd (1947 with illustrations,
70c). p. 19)-
and 1955).
220 Starr (1938). 255 Reproduced in Frankfort
237 For the site of Nimrud and its
221 For once, a very revealing (1954, fig. 40, p. 103).

surroundings, see Mallowan
analysis of their contents is (1966b, vol. I, ch. i) and 256 Reproduced in Frankfort
appended to the main publi- Oates, D. (1968, ch. 3). (1954, pl. 82).
237
Notes on the Text
257 The pair of bulls from Gate 1963. Finally pubhshed in this sort did in fact happen.
'A', now at the entrance to the Mallowan (1966, vol. 2). Others have doubted it.
Iraq Museum, measured 44 m

275 See Turner (i960, p. 68 ff.). 284 Herodotus' account (Bk i, ch.
square and weighed over 20
tons. The single bull, with 276 Well illustrated in Illustrated
178 ff.), is very fully discussed
and compared with the exca-
I
head facing sideways, from London News, i December vators' findings in Ravn
the throne-room entrance, 1962. (1942), a book in English
now in the Oriental Institute
which is unfortunately
Museum, Chicago, was 47 m 277 See Mallowan (1966, vol. 2,
Koldewey's
difficult to obtain.
square, and the largest frag- fig- 373, P- 453)-
semi-popular account of his
ment weighed 14 tons. Other
278 In the second volume of work appeared in 1914,
fragments remain of an
in situ
Mallowan's major work on poorly translated into EngHsh.
even larger bull, measuring
the excavations at Nimrud, he A better summary is to be
5-5 m square. found in MacQueen
deals with two groups of ivor- (1964).
258 Frankfort (1954, pi. 87). ies, one from rooms NW.15,
285 See Lloyd (1947).
SW.2 and SW.7, the other
259 Frankfort (1954, pi- 9o)-
from rooms SW.12, SW.37. 286 Koldewey's own reports ap-
260 Frankfort (1954, fig. 41, p. This last is one of the eight peared in the Mitteilunger der
104). magazines - each almost deutschen Orient Gesellschaft
261 Strommenger (1964, pi.

30 m
long - of which the (MDOG), in the years bet-
storage space in the South- ween 1899 and 1932.
209-14).
west courtyard composed. is
287 For the site plan before exca-
262 King (1915). The remainder are marked on
vation, see Koldewey (1925,
the plan 'partly excavated'.
263 See Barnett (i973, P- 442 and fig. i), and for the inner city
notes).
See Mallowan (1966, folding
after excavation, ibid., fig. 256.
plan no. 8).
264 Thureau-Dangin and Dunand
288 Herodotus {History, Bk i, ch.
(1936). 279 The final publication of the
ivories (Commentary, cat-
186), describes how this
265 See Parrot (1961, pis. I-IV and bridge was built by 'Ni-
alogue and Plates), has ap-
figs.109-20, 266 and 336-48).
tochris'.
peared in a succession of fas-
266 Some of the best photographs cicles under the title Ivories
289 Various reconstructions are
are those taken by Max Hir- from Nimrud (1966, 1970, illustrated side-by-side in
mer to illustrate Strom- 1974). Many more are to be
Ravn (1942, pis. 14 and 15).
menger (1964). expected as the Iraqi excava-
267 Strommenger tions continue to be pubhshed. 290 Discussed fully in Ravn (1942,
(1964, pi. 241).
p. 67 ff.).
268 E.g. Madhloum 280 Barnett (1957).

(1964).
291 Woolley's plan of the Neo-
269 See Mallowan (1966). 281 It has been compared to the
Babylonian temenos at Ur
'Chinoiserie' phase of Enghsh
270 See plan of the citadel in appears in Woolley and Mal-
ornament in the i8th century.
Mallowan (1966, fig. i, p. 32). lowan (1962, pi. 60), and that
282 See Gadd (1958, p. 35 ff.). of the Bel-Shalti-Nanuar
271 See Mallowan (1966, vol. i,
palace in ibid., pi. 70. For the
facing p. 128). 283 Herodotus {History, Bk i, ch.
plan of the outer city wall, see
lOi ff.), and Xenophon {Cyr-
272 See Mallowan (1966, vol. i,
Hawkes (i974, P- i73)-
opaedia, Book VII, ch. V, 10),
Frontispiece).
maintain that Cyrus diverted 292 Frankfort (1954, p- 108).
the course of the Euphrates, so
273 See Plenderleith's report, Matthiae (1980).
293
that his troops could enter the
quoted by Mallowan: Mal-
city along its dry bed. Kol- 294 The journals Annales
lowan (1966, p. 139). Iraq,
dewey (in F. Wetsel, Die archeologiques de Syrie and
274 Preliminary reports by D. Stadtmauern von Babylon, Leip- Anatolian Studies pubHsh re-
Oates in Iraq and Illustrated zig 1930, p. 53), was inclined gular reports on current arch-
London News from 1957 to to beUeve that something of aeological work.
238
Bibliography
General Recommendations
Beek, M. a. Atlas of Mesopotamia, London 1962.

Curtis, J. (ed.) Fifty Yearsof Mesopotamian Discovery, London 1983.
Ehrich, R. W. (ed.) Chronologies in Old World Archaeology, Chicago 1965.
Ellis, S. A Bibliography of Mesopotamian

R. Archaeological Sites,
Harrassowitz, Wiesbaden 1972.

Hawkes, Jacquetta The First Great Civilizations, London and New York
1973-
Kramer, S. N. The Sumerians, Chicago 1963.
Lloyd, Seton Foundations in the Dust, rev. edn, London and New York
1980.
Mallowan, M. E. L. Twenty-Five Years of Mesopotamian Discovery, British
School of Archaeology in Iraq, London 1956.
MooRTGAT, A. The Art of Ancient Mesopotamia, London 1969.
Oates, J. Babylon, London and New York 1979.
Oppenheim, a. L. Ancient Mesopotamia, Chicago 1964.
Parrot, A. Archeologie Mesopotamienne, vol. i, Les Etapes, Paris 1946; vol.
2, Technique et Probemes, Paris 1953.
Perkins, A. L. The Comparative Archaeology of Early Mesopotamia, Studies
in Ancient Oriental Civilization, no. 25, Chicago 1949.
Roux, George Ancient Iraq, London 1964.
Saggs, H. W.
The Greatness that was Babylon, London 1962.
F.
Strommenger, E. The Art of Mesopotamia, London 1964.
Other References
Adams, R. M. 'Survey of Ancient Water Courses and Settlements in

Central Iraq', Sumer, 14 (1958).
Adams, R. M. and Nissen, H.J. The Uruk Countryside, Chicago 1972.
Amiet, p. La Glypti que Mesopotamienne, Paris 1961.
Andrae, W. Die archdischen Ischtar Tempel, Mitteilungen der deutschen
Orient-Gesellschaft, Berhn 1922.
Farbige Keramik aus Assur (Coloured Ceramics from Assur), Berlin
1925.
Das Wiedererstandene Assur, Leipzig 1938.
Bachmann, W. Felsreliefs in Assyrien, Leipzig 1927.
Banks, E.J. Bismaya or the Lost City of Adab, New York 1912.
Bannister-Fletcher A History of Architecture, London 1961.
Baqir, T. 'Iraq Government Excavations at 'Aqar Quf, 1942-43', Iraq,
Supplement (1944).
'Iraq Government Excavations at 'Aqar Quf; Second Preliminary
Report', Iraq, Supplement (1945).
'Iraq Government Excavations at 'Aqar Quf; Third Interim Report',
Iraq, 8 (1946a).
'Tell Harmal: A Preliminary Report', Sumer, 2, pt 2 (1946b), 226.

Baghdad 1959.
Tell Harmal,
'A New Law Code from Tell Harmal', Sumer, 4, no. (1948). i
Barnett, R. D. a Catalogue of the Nimrud Ivories in the British Museum,

London 1957.
239
Bibliography
'Homme Masque ou Dieux-Ibex', Syria, 43 (1966).

'New Facts about Musical Instruments from Ur', Iraq, 31, pt 2 (1969)-
'Le Palais et le Royaute', Proceedings of the XIXe Rencontre

Assyriologique Internationale, Paris 1971.
'More Balawat Gates', Studies Dedicated to F. M. T. de L. Bohl, Brill,
Leiden 1973.
Barrelet, M.-Th. 'Une Peinture de la Cour 106 du Palais de Mari', Studia
Mariana, Brill, Leiden 1950, pp. 9-35-
'Quelque Sculptures Mesopotamienne de I'Epoque d'Akkad', Syria,
36 (1959), 20 ff.
BoTTA, and Flandin, E. Monuments de Ninive (5 vols.), Paris 1849-50.

P. E.
Braidwood, R. J. 'The Iraq-Jarmo Project', in G. R. Willey (ed.),
Archaeological Researches in Retrospect, Cambridge (Mass.) 1974(b).
Braidwood, R. J. et al. 'Baghouz; New Chalcolithic Material of the
Samarra Type and its Antecedents', JoMrwij/ of Near Eastern Studies, 3
(1944)-
'Matarra: A Southern Variant of the Hassuna Assemblage, Excavated
in ig^%\ Journal of Near Eastern Studies, 11 (1952).
Prehistoric Investigations in Iraqi Kurdistan, Chicago i960.
'The Beginnings of Village Farming Communities in South-eastern
Turkey', Proceedings of the National Academy of Sciences, U.S.A., 1974(a).
Brice,W. C. South-west Asia. A Systematic Regional Geography, no. 8,
London 1966.
Burkholder, G. "Ubaid Sites and Pottery in Saudi Arabia', Archaeology,
25 (1970).
Burney, C. a. 'Yanik Tepe', Iraq, 23, pt 2 (1961) and 26, pt i (1964).
ÂMBEL, H. and Braidwood, R.J. Prehistoric Research in Southeastern Asia
Minor, Istanbul 1980.
Ceccini,J. M. La Ceramica di Nuzi, Rome 1965.
Childe, Gordon. 'The Urban Revolution', in Town Planning Review, 21
(1950), 3-17-
Coon, Carleton. The Origin of Races, London 1963.
CoRBiEAU, S. 'Indus Parallels', Iraq, 4, (1937), i ff.
Crawford, V. E. 'Lagash', Iraq, 36 (1974).

Delougaz, p. 'Plano-convex Bricks', Studies in Ancient Oriental Civiliza-
tion, no. 7, Chicago 1933.
'A Short Investigation of the Temple at Al 'Ubaid', Iraq, 5 (1938).

The Temple Oval at Khafaje, Oriental Institute Publications, no. 53,
Chicago 1940.
Pottery from the Diyala Region, Oriental Institute PubHcations, no. 63,
Chicago 1952.
'Architectural Representation on Steatite Vases', Iraq, 22 (i960).
Delougaz, P. and Lloyd, S. Pre-Sargonid Temples in the Diyala Region,
Oriental Institute Publications, no. 58, Chicago 1942.
Delougaz, P., Hill, H. and Lloyd, S. Private Houses and Graves in the
Diyala Region, Oriental Institute PuWications, no. 88, Chicago 1967.
Dyson, R. H. 'Shub-ad's Onagers', Iraq,/22 (i960).
Edzard, D. O. 'Enmer-baragisi von Kis', Zeitschrift fur Assyriologie von
Vorderasiatische Archaeologie, no. 53, Berlin 1959.
Falkenstein, a. Archdische Texte aus Uruk, Berlin 1936.
'Zu den Tafeln aus Tartaria', Germania, 43 (1965).
Finch,J. P. G. 'The Nergal Gate', Iraq, 10, pt i (1948).
Frankfort, Henri. Archaeology and the Sumerian Problem, Studies in
Ancient Oriental Civilization, no. 4, Chicago 1932.
Tell Asmar, Khafaje and Khorsabad. Oriental Institute Com-
munications, no. 16, Chicago 1933.
'Gods and Myths on Sargonid Seals', Iraq, i, pt i (1933).
Progress of the Work of the Oriental Institute in Iraq, 1934-35, Oriental
240
Bibliography
Institute Communications, no. 20,Chicago 1936.

Sculpture of the Third Millennium B.C. from Tell Asmar and Khafaje,
Oriental Institute Publications, no. 24, Chicago 1939(a).
Cylinder Seals, London 1939(b).
More Sculpture from the Diyala Region, Oriental Institute Publications,
no. 60, Chicago 1943.
Birth of Civilization in the Near East, London 195 1.
'The Origin of the Bit Hilani', Iraq, 14 (1952), 120-31.

Art and Architecture of the Ancient Orient, Harmondsworth 1954, 4th
rev. impression 1970.
Stratified Cylinder Seals from the Diyala Region, Oriental Institute
Publications, no. 72, Chicago 1955.
Frankfort, Henri, Lloyd, S. and Jacobsen, Tn.'The Gimilsin Temple and
Palace of the Governors at Tell Asmar, Oriental Institute PubUcations, no.
43, Chicago 1940.
Frankfort, Henri and Davies, L. 'The Last Predynastic Period in
Mesopotamia', in Cambridge Ancient History, vol. i, pt 2, ch. 12,
Cambridge 1971.
Gadd, C.J. 'Seals of Ancient Indian Style Found at Ur', Proceedings of the
British Academy, vol. 18, 1933.
The Stones of Assyria, London 1936.
'The Harran Inscriptions of Nabonidus', Anatolian Studies, 8 (1958).
'The Spirit of Living Sacrifices in Tombs', Iraq, 22 (i960), 51 ff.
'The Cities of Babylonia', in Cambridge Ancient History, vol. i, pt 2, ch.
13, Cambridge 1971.
Galpin, F. W.
'The Sumerian Harp of Ur', Music and Letters, 10 (1929).
Garrod, D. a. W. The Caves of Zarzi and Hazar Merd, in Bulletin of the
American Schools of Prehistoric Research, no. 6, 1930.
Gurney, O. R. 'Laws of Eshnunna', Symbolae Hrozny (1950).
Haines, R. C. 'Nippur', Illustrated London News, 18 August 1956, 6
September 1958 and 9 September 1961.
Hall, H. R. and Woolley, C. L. Ur Excavations, vol. i: Al 'Ubaid,
London and Philadelphia 1927.
Hansen, D. P. 'Votive Plaques from Nippur', Journal of Near Eastern
Studies, 22 (1963).
Hawkes, Jacquetta (ed.) Atlas of Ancient Archaeology, London 1974.
Heinrich, E. Bauwerke in der altsumerischen Bildkunst, Berlin 1957.
Herzfeld, E. E. Die Ausgrabungen von Samarra, vol. 5 Die Vorge- :
schichtlichen Tbpfereien von Samarra, Berlin 1930.

HijARA, I. 'Arpachiyah', Iraq, 42, pt 2 (1980).
Hilprecht, H. V. Exploration iti Bible Lands during the Nineteenth Century,
Philadelphia 1903.
Jacobsen, Th. The Sumerian King-list, Assyriological Studies, no. 11,
Chicago 1939.
'Salt and Silt in Mesopotamian Agriculture', Science, no. 128 (1958a).
'Girsu', Revue d'Assyriologie, 52 (1958b).

'The Waters of Ur', Iraq, 22 (i960).
Jacobsen, Th. and Lloyd, S. Sennacherib's Aqueduct at Jer wan. Oriental
Institute Pubhcations, no. 24, Chicago 1935.
Jasim, S. a. 'Excavations at Tell Abada', Iraq, 45, pt 2 (1983).
Jones, Felix. The Topography of Nineveh, Bombay 1957.
King, L. W. Bronze Reliefs from the Gates of Shalmaneser, London 191 5.
KiRKBRiDE, D. 'Umm Dabaghiyah', Iraq, 34 (1972) onward.
KoLDEWEY, R. The Excavations at Babylon (Engl transl.), London 1914.
Das Wiedererstehende Babylon, Leipzig 1925.
Kramer, S. N. Enmerkar and the land of Aratta, Philadelphia 1952.
'Dilmun: Quest for Paradise', Antiquity, 27, no. 146 (1963).

From the Tablets ofUr, Indianapohs, Colorado 1965.
241
Bibliography
Lamberg-Karlovsky, C. C. Excavations at Tepe Yahya, Iran, Cambridge

(Mass.) 1970.
'Excavations at Tepe Yahya', G. R. Willey (ed.), Archaeological
Cambridge
Researches in Retrospect, (Mass.) I974-
Lambert, M. and Tournay, J. R. 'La Statue B de Gudea', Revue
d'Assyriologie, 45 (1951)-
La yard, a. H. Nineveh and its Remains (2 vols.), London 1850.
Discoveries in the ruins of Nineveh and Babylon, London 1853.
Leacroft, H. and R. The Buildings of Ancient Mesopotamia, Leicester 1974.
Le Breton, L. 'The Early Periods of Susa Mesopotamian Relations', Iraq,
:
19(1957)-
Lees, G. N. and Falcon, N. L. 'The Geographical History of the
Mesopotamian Plains', The Geographical Journal, 118 (1952).
Lenzen, H.J. 'Die Architecktur in Eanna in der Uruk IV Period', Iraq, 36,
pt 1/2 (1974), P- III ff-
LoFTUS, W. K. Travels and researches in Chaldaea and Susiana, London 1857
(republished 1971).
Loud, Gordon. Khorsabad, vol. i. Oriental Institute PubHcations, no. 38,
Chicago 1936.
Loud, Gordon and Altmann, C. B. Khorsabad, vol. 2, Oriental Institute
PubHcations, no. 40, Chicago 1938.
Lloyd, S. 'Some ancient sites in the Jebel Sinjar District', Iraq, 5 (1938).
'The Oldest City: A Pre-Sumerian Temple at Eridu', Illustrated
London News, 31 May 1947.
'The Oldest City of Sumeria The Origin of Eridu', Illustrated London
:
News, II September 1948.

'Ur-Al 'Ubaid, 'Uqair and Eridu', Iraq, 22 (i960).
'Back to Ingharra', Iraq, 31, pt i (1969).
Lloyd, S. and Safar, F. 'Eridu', Sumer, 3 (1947) and 4 (1948).
'Tell 'Uqair: Excavations by the Iraq Government Directorate of

Antiquities in 1940-41', Journal of Near Eastern Studies, 2, no. 2 (1943), p.
131-
'Tell Hassuna : Excavations by the Iraq Government Directorate of
Antiquities in 1943-44', Jowm^/ of Near Eastern Studies, 4 (1945).
Mackay, E. a Sumerian Palace and the 'A' Cemetery at Kish, Mesopotamia,
Chicago 1929.
'Excavations at Jemdet Nasr', in Field Museum: Anthropological
Memoirs, nos. 1-3 Chicago 1931.
Macqueen,J. Babylon, London 1964.
Madhloum, T. 'Excavation of Nineveh', Sumer, 23 (1967).
'Nineveh', Sumer, 24 (1968).
The Chronology of Neo- Assyrian Art, London 1970.
Mallowan, M. E. L. 'White Painted Subartu Pottery', in Melanges Offerts
a M. Rene Dussaud, vol. 2, 1939.
'Excavations at Brak and Chagar Bazar', Iraq, 9 (1947).
'Noah's Flood Reconsidered', Iraq, 26, pt 2 (1964a).
'Ninevite 5', in Vorderasiatische Archdologie Studien und Aufsatze,
Festschrift Moortgat, Berlin 1964(b).
'Tell Chuera: Fourth Campaign', Iraq, 28, pt i (1966a).
Nimrud and its Remains (2 vols, and plans), London 1966(b).
Review of Mission Archeologique de Mari, vol. 4, in Bibliotheca
Orientalis, 26, no. 1/2 (1969).
Cambridge Ancient History, vol. i, pt i, Cambridge 1970, ch. 8, 'The
Development of Cities from Al 'Ubaid to the end of Uruk V : section i,
'Babylonia and xMesopotamia', section 2, 'Assyria', section 3, 'Syria',
section 4, 'Iran'; vol. i, pt 2, 197 1, ch. 16, 'The Early Dynastic Period in
Mesopotamia'.
Mallowan, M. E. L. and Rose, J. C. 'The Excavations of Tell
242
Bibliography
Arpachiyah', Iraq, 2, (i935)-
Matthiae, p. Ebla: an Empire Rediscovered, London 1980.

McEwAN, C. et al. Soundings at Tell Fachariya, Oriental Institute
Publications, no. 79, Chicago 1958.

Mellaart, J. Catal Hidyiik, London 1967.
The Neolithic of the Near East, London and New York 1975.
MooREY, P. R. S. 'A Reconstruction of East Kish', Iraq, 26, pt i (1966).
'Terracotta Plaques from Kish', Iraq, 26, pt 2 (1975).
MooRTGAT, A. Vorderasiatische Rollsiegel, BerHn 1940.
'Assyrische Glyptik des 12 Jahrhunderts', Zeitschriftfur Assyriologie, 48
(n.f 14) (1944)-
'Tell Chuera in Nord-Ost Syrien', Preliminary Reports in
Abhandlungenfur Arheitgemeinschaf^, Koln 1959 et seq.
Wissenschaftliche
MoRTENSEN, P. 'Additional Remarks on the Chronology of Early Farming
Communities in the Zagros Area', Sumer, 20 (1964).
MuNCHAjEF, R. and Merpert, N. in Sumer, 25 (1969) onwards.
'Early Agricultural Settlements in the Sinjar Plain, Northern Iraq',
Iraq, 35, pt2 (i973), P- 93 ff-

Gates, David. 'Tell Rimah, 1966', Iraq, 29, pt 2 (1967).
Studies in the Ancient History of Northern Iraq, British Academy,
London 1968.
'Early Vaulting in Mesopotamia', in Archaeological Theory and
Practice: Essays Presented to W. F. Grimes, London 1973.
Gates, Joan. 'Ur and Eridu: the Prehistory', Iraq, 22 (i960).
'Tell-es-Sawwan Figurines', Iraq, 28, pt 2 (1966).
'Goddess of Choga Mami', Illustrated London News, 28 September
1969(a).
'Choga Mami, 1967-8': A Preliminary Report, Iraq, 31, pt 2 (1969b).
'Prehistory in North-eastern Arabia', Antiquity, 50, no. 179 (1976).

Gates, D. andj. The Rise of Civilization, Gxford and New York 1976.
Parrot, Andre. 'Mari et Chagar Bazar', Iraq, 19 (1938).
Tello Vingt Campagnes de Fouilles, 1871-ig^j, Paris 1948.
:
Mission Archeologique deMari vol. Le Temple d'lshtar, Paris 1956 vol.

: i , ;
2, pt. I, Le Palais: Le Palais:

Architectur, Paris 1958(a); vol. 2, pt 2,
Peintures Murales, Paris 1958(b); vol. 2, pt 3, Le Palais: Documents et
Monuments, Paris 1958(c); vol. 3, Les Temples de Ishtarat et de Nini-
Zaza, Paris 1967(a); vol. 4, Le Tresor d'Ur, Paris 1967(b).
Nineveh and Babylon, London 1961.
'Les Fouilles de Mari: Vingtieme Campagne', Syria, 39 (1962).
Place, V. Ninive et I'Assyrie, avec des Essaies de Restauration par Felix
Thomas, (3 vols.), Paris 1867-70.
PoRADA, E. Corpus of Ancient Near Eastern Seals in North American
Collections, vol. Washington 1948.
i,
Postgate, N. The First Empires, Gxford and New York 1977.

Rassam, H. Asshur and the Land of Nimrod. New York 1897 (republished
1971).
Ravn, G. E. Herodotus' Description of Babylon, Arnold Busck, Copen-
hagen 1942.
RiMMER, J. Ancient Musical Instruments of Western Asia, London i960.
Roux, George. 'Recently Discovered Sites in the Hammar Lake District',
Sumer, 16 (i960).
Safar, F. 'Eridu: A Preliminary Report on the Third Season's
Excavations', Sumer, 6 (1950).
Safar, F., Mustafa M.A. and Lloyd, S. Eridu (the fmal pubhcation),
Baghdad 1983.
Sandars, N. K. (transl.) The Epic of Gilgamesh, Harmondsworth i960.
De Sarzec, E. and Heuzey, L. Decouvertes en Chaldee, Paris 1 884-1912.
ScHEiL, v., 'Une Saison de Fouilles a Sippar', in Memoires Publiees par les
243
Bibliography
Memhres de l' Academic Frangaise d'Archeologie Orientale du Caire, vol. i,
1902.
Schmandt-Besserat, D. 'Some Characteristics of the Early Tablets', Syro-
Mesopotamian Studies, i, no. 2 (1977).
Singh, P. Neolithic Cultures of Western Asia, London 1974.
Smith, Sidney. 'Diniktim', Sumer, 2, no. 2 (1946), p. i9ff.
SoLECKi, Ralph S. Shanidar: The Humanity of Neanderthal Man, London
1971.
SoLLBERGER, E. 'Sur la Chronique des Rois d'Ur et Quelque Problemes
Annexes', In Archives fur Orientforschung, no. 17, 1954.
'Notes on the Early Inscriptions from Ur and Al 'Ubaid', Iraq, 22
(i960).
Speiser, E. a. Excavations at Tepe Gawra, vol. i, Philadelphia 1935.
Starr, R. F. S. Nuzi (2 vols.), London 1938.
Strommenger, E. 'Akkadian Sculpture', in Baghdader Mitteilungen no. , 2,
pt I, 1936.
Thompson, R. Campbell, and Mallowan, M. E. L. 'The British Museum
Excavations at Nineveh, 193 1-32', in Liverpool Annals of Archaeology and
Anthropology vol. 20, 1932.
,
Thureau-Dangin, F. and Dunand, N. Til Bar sip, Paris 1936.
Tobler, A.J. Excavations at Tepe Gawra, vol. 2, Philadelphia 1950.

Turner, G. 'Tell Nebi Yunus The Ekal Masharti of Nineveh', Iraq, vol.
:
32,
pt I (1970).
'The State Apartments of Late Assyrian Palaces', Iraq, 32, pt 2 (1970).
and their Mesopotamian Antecedents', Ph.D.
'Late Assyrian Palaces
thesis, London Institute of Archaeology, 1967.
UcKO, P. J. 'Anthropomorphic Figurines', Royal Anthropological Society,

Occasional Papers, no. 24, London 1968.
Van Buren, E. D. 'Places of Sacrifice (Opferstatten)', Iraq, 14 (1952).
'Fish Offerings', Iraq, 10, pt 2 (1948), p. loi.

Van Zeist, W. and Bottema, S. in Palaeohistoria, 19 (1977).
Van Zeist, W. and Bintliff, J. (eds.) Palaeoclimates Palaeoenvironments ,
andHuman Communities in the E. Mediterranean Region in Later Prehistory,

BAR, Si 33, Oxford 1982.
Wahida, G. 'The Re-excavation ofZarzi 1971', Proc. Preh. Soc, 47, (1981).
Waterlin, L. C. and Langdon, S. Excavations at Kish, London 1930-4.
Weber, O. 'Altorientalische Siegebilder', in Der Alte Orient, vols. 17-19,
Leipzig 1920.
Wetsel, F. Die Stadtmauern von Babylon, Leipzig 1930.
Wiseman, D.J. Cylinder Seals of Western Asia, London (n.d.).
WooLLEY, C. L. 'Excavations at Ur', Antiquaries Journal, 3 (1923), to 13
(1934a).
Ur Excavations, vol. 2 : The Royal Cemetery, London and Philadelphia
1934(b).
Ur of the Chaldees, London 1935.
Ur Excavations, vol. 4: The Early Periods, London and Philadelphia
1956.
Ur Excavations, vol. 5 : The Ziggurat and its Surroundings, London and
Philadelphia 1939.
Ur Excavations, vol. 6 : The Buildings of the Third Dynasty, London and
Philadelphia 1974.
Ur Excavations, vol. 8: The Kassite Period, London and Philadelphia
1965.
Excavations at Ur (preHminary reports collected in a single volume),
London 1954.
WooLLEY, C. L. and Mallowan, M. E. L. Ur Excavations, vol. 9, The Neo-
Babylonian and Persian Periods, London and Philadelphia 1962.
Ziegler, C. Die Keramik von Haggi Mohammed, Berlin 1953.
244
Photographic Acknowledgments
The photographs used for the undermentioned illustrations are reproduced

by courtesy of:
Aleppo Museum, photo courtesy Director-General of Antiquities, Syria,
ii6; Aleppo Museum, photo Hirmer Verlag, Munich, 152; Alinari-
Giraudon, 153; Director, Archaeological Museum, Istanbul, 117;
Archaeological Museum, Tehran, photo Delegation Fran<;aise en Perse,
48; Ashmolean Museum, Oxford, 24; Berlin Museum, 120, 129, 170, 171
R. J. Braidwood and the Oriental Institute, University of Chicago, 7;
Trustees of the British Museum, 27, 44, 61, 71, 78, 87, 88, 91, 100, loi, 122,
130, 132, 133, 146-51, 155, 159, 163; British School of Archaeology in
School of Archaeology, Jerusalem, 4;
Iraq, 46, 156, 162, 165, 166; British
Brockhampton Press, 9, 18, 106, 134; Damascus Museum, photo Hirmer
Verlag, Munich, 66; Directorate-General of Antiquities, Baghdad, Iraq,
II, 121, 158; Henri Frankfort (1939b), 28, 29; Hirmer Verlag, Munich, 41,
79, 80, 98, 99, 105, 174; Iraq Museum, Directorate-General of
83,
Antiquities, Baghdad, Iraq, 67, 81, 84, 85, 90, 123, 124; Iraq Museum,
Baghdad, photo Hirmer Verlag, Munich, 15, 16, 25, 26, 86, 93, 97; Iraq
Museum, Baghdad, photo courtesy Director, Oriental Institute, Uni-
versity of Chicago, 64, 70; Diana Kirkbride (Mrs Hans Helbaek), 5;
Professor Seton Lloyd, 14, 21-3, 72, 119, 145; Louvre, photo Archives
Photographiques, 96; Louvre, Conservateur-en-Chef des Antiquites
Orientales, 138; Louvre, photo Editions Ides et Calendes, Neuchatel, 65;
Louvre, photo Giraudon, 68 Louvre, photo Hirmer Verlag, Munich, 69,
;
89, 102, 108; Metropolitan Museum of Art, Rogers Fund (i960), 164; Dr
Moortgat-Correns, 62 Dr Joan Oates, 45 Director, Oriental Institute,
; ;
University of Chicago, 49, 53, 55, 63; Professor Andre Parrot, 112;
Trustees of the Pierpont Morgan Library, 131; Sotheby's, 118; University
Museum, University of Pennsylvania, Philadelphia, 107; Visual Publi-
cations Ltd, 109, 144.
245
;
Index
Numerals in italics refer to the Arpachiyah, 65-7, 74, j6 Baradost, 24, 25

illustrations Arrapha, 177, 187, see also Kirkuk Barnett, R. D., 131, 219
'arsenal', 202, 216, 161 basketwork, 31
A'Annipadda, 91, 103 Ashur, 10, 16, 107, 108, 135, Bau, 127
Abargi, 79 I37» I57» i79ff-. i26, 127; Ishtar Bavian, 1^6, 191
Abbevillian, 21 Temple, 108, 112, 120, 127, Bedouin, 138
Abu-al-Soof, B., 72 181 ; other temples and palaces, Beidha, 27, 29, 3
Abu-Shahrain, Tell, see Eridu i8iff., 127, 128 Beldibi, 24
'aceramic periods', 32 Ashur-banipal, 191, 199 Bel-shalti-nannar, Palace of, 230
Achaemenids, 222 Ashur-etililani, 194 Belshazzar, 222
Acheulian, 21 Ashur-nasirpal, 183, 188, 193; Berlin Museum, 226, 230
Adab, 90, 139 'Stela of, 156, 212 Berossus, 88
Adad-Nirari I, 179 Ashur-rabi II, 187 Billa, Tell, 177
Adad-Nirari II, 187 Asiab, Tepe, 27, 32 Bisitun, 24, 3
Adad-Nirari III, 194 Asmar, Tell, 36, 39, 160; Abu Bismaya, 139
Afghanistan, 82, 85 Temple, 94, 97, 120; Archaic bit akitu, 1 8 1-3, 203
Agade, 137-9, 150, see also Temple, 97; Single Shrine Black Sea, 23
Akkad Temple, 97; Square Temple, Botta, E., 192, 138, 196, 197
Agga of Kish, 92, 93 97, 54, 112, 113, 64; Gimil-Sin bow-drill, 61
Agrab, Tell, 36, 98; Shara Temple, Temple and palaces, 155, 106, Braidwood, R. J., 25, 26, 30, 33,
98, 33, 120, 74, 142; sculpture, 161 ; houses, 141 ; Northern 35,71
113, 117, 70, 120, 128, 86 (Akkadian) Palace, 122, 140, Brak, Tell, 66, 76, 147; Eye
Akalamdug, 102, 103 141, 94; sculpture, 112, 113, 64 Temple, 76, 39, 77, 46, 85;
Akkad, 20, 137; Akkadian period, 'assembly, general', 88 Akkadian Palace, 95, 141, 142
36, 66, 6g, 97, 106, 109, 135 ff., Assyria, 19, 65, 107, passim; bricks, 'cigar-shaped', 73
137, 139; archaeology, 139; Assyrian empire, ijf gypsum, 52; 'hog-backed', 28;
graves at Ur, 139; language, Aurignacian, 21, 22, 25 kiln-baked, 119, 152; 'plano-
137; sculpture, 93, 138, 142, g6, Azarbaijan, 32, 33, 158 convex', 119, 73\ riemchen, 21;
144; seals, 144, g8 stamped, 153, 230; sun-dried,
Alalakh (Atchana), 158, 177 41,72
Aleppo, 29, 188, 231 Babylon, 10, 57, 157; city of, 224; British Museum, 58, 59, loi, 131,
Ah Kosh, 33 Northern Palace, 228, 229, 171, 149, 194, 196, 197, 199, 219
altars, 184 172; Summer Palace, 224, 167; 'brocade style', 124
Altmann, C. B., 144 walls of, 223ff., 167, 168; First 'bucranium', 80
Al 'Ubaid, see 'Ubaid Dynasty of, 157 burials, 46, 47, 72, 74, 81, 82, 96,
Amanus, 138 Babylonia, 18, 172 99ff., 139, 183, 230; infant, 92;
Anah, 19 Baghad, 18, 55, 80; antique pigmented, 47
Andrae, W., 10, 90, 108, 112, 126, dealers of, 93, 149 'Burnt Palace', 196, 160, 219
181 Baghouz, 72, 79 Burushanda, 138
Anu Area, see Uruk Bahrein, 64, ji, 138 Bus Mordeh, 33
apkallu, 186 Bahriyat, see Isin
'Aqar Quf, 161, 121, 174 Balawat, Gates of, 201, 206, 130,
Aral Sea, 23 151 canals, 17
Arabia, 188 Balik, 108 Cappadocia, 158, 177
Arabian Desert, 13 Baluchistan, 118 Carchemish, 16, 158, 188, 222
Arabian Gulf, 15,3^, 37, 64, 138, Bandar Abbas, 33 Carmel, Mount, 22, 23, 26
149, 160, 188 Banks, E. H., 139 Caspian Sea, 23
'Ark', the, 13 'banqueting scene', 116, 79, 87 Caucasus, 85
246
;
Index
Cauvin, J., 29 Diwaniyah, 107 'eye' or 'spectacle' idols, 77, 46, 85

cedar, 204 Diyala, 18; Diyala sites, 36, 55, 'Eye Temple', 77, 39
cemeteries, 45, 64, 86 91-3, 99, 107, 108, 112, 141, Eynan (Ain Mallaha), 28
Chagar Bazar, 158 160, 161
Chalcolithic period, 32, 33, 35, 65, Diyarbakir, 30, 80, 138
69, 72, 77 domes, 67, 74, 99 Faish Khabur, 19, 2
Chaldea, 188, 191, 218 domestication of animals, 26, 32, Farah, 90-2; 'Farah period', no
chariot burials, 106 33 Fertile Crescent, 19
Chemchemal, 33, 35 domestication of plants, 26, 31, fireplaces, 72
Chicago, Oriental Institute of, 91, 33, 34 fish, 19; offerings, 16,55
93,96 doors, 119, 208, 130, 131 Flandin, E., 196, 197
Childe, Gordon, 26 Dumuzi, 57 'Flood', the, 90-2, 106; floods,
Choga Mami, 66, 72, 73, 83, 43 Dur-Kurigalzu, 160, 172, iSj see river, 16
Choga Mish, 87 'Aqar Quf 'Flowing-vase goddess', 166, 113,
Chuera, 109, 62 Dur-Sharrukin, see Khorsabad 167, 169, 120, 172, 203
Cilicia, 188 'Fort Shalmaneser', 161, 216;
cire perdue, 86, 128 throne-base, 162, 218; glazed
city-states, 135 Eanna Precinct, see Uruk brick, 218
'Cliff, the', 12 Eannatum, 106, 116, 149 foundation deposits, 124
climate, 17 Early Bronze Age, 65 foundations, sand, 94, 33, 96;
coastline, 15 Early Dynastic period, 36, 66, 88, Babylon, 169
composite objects, 131 99, 107, 108, no; pottery, 132, Frazer, Sir James, 57
'cone-mosaic', 50-2, 21, 54, 129 133, 92; sculpture, 112, 113, French excavations, 86, 105, 108,
continuity (rehgious beliefs), 63 passim, 64 passim; seals, 124, 124, 147, 157, 192, 193, 138
Coon, Carleton, 24 78-81 ; sites, 92ff. ; terminology,
copper, 25, 30 91, no
Crawford, V. E., 150 Ebih-il, 114 Gadd, C.J., 88
Crete, 138, 177 Ebla, 231 gaming-boards, 129
Crimean War, 192 E-dub-lal-makh, 153 Ganj Dareh Tepe, 32
crop-yields, 18 Egypt, 62, 184, 188, 191, 221, 222 Garrod, A. E., 21-3, 25, 26
Cros, G., 149 einkorn, 30 Gasur, 147
Ctesias, 223 ekal masharti, 216, 161 Gawra, Tepe, 66, 67; 'acropolis',
Ctesiphon, Arch of, 115 Elam, 85, 138, 142, 147, 151, 67, 75, 32; 'round house', 75, 37
cult statues, 108, 112, 114 157-9, 191, 155 'Gawra period', 67, 75, 38, 76, 81,
Cyaxares, 191 elephant, Syrian, 221 82, 85, 47
cylinder-seals, 59; ProtoHterate, Elazig, 16 gazelles, 29, 35
59,60, 28, 29; ED II and III, En, 88, 89 German excavators, 38, 108, no,
124, 125, 78-81, 127; Akkadian, Enki, 57 178, 193, 223
144, g8; Mitannian, 177; Enlil, 57, 103, 107, 117 Gilgamesh, 48, 49, 90, 91, 92, 93
Middle Assyrian, 184, 130-j Enmeberagisi, 93 Epic, 103
Cyprus, 138, 218 Enmerkar, 93 Gi-par-u, 153
Cyrus, 222 ensi, 88, 89, 122 Girsu, 107, 103, 150
Cambel, H., 30 Enuma Elish, Epic of Creation, 167 glazed ornament, 183, 218, 227,
Catal Hiiyiik, 27, 30, 6 E-nun-makh, 153 170, 171
Cayonii, 27, 30 Erech, 45 glockentopf, 81
Eridu, 15, 39ff., 57, 66, 81; Godin Tepe, 87
cemetery, 45-7; palaces, 122, Grai Resh, 66, 81
Dagan, 138; Temple of, 167 123, 77; temples, 39, 8, 10, 43, Gravettian, 21-3, 25
Dalma, 33 32,7s; pottery, 36, 39, 12,
11, Greek art, 221, 228, 230; writers, 88
Damascus, 188, 221 44,45 Gudea, 147, 101, 102, 149, 150
Darband-i-Gawr, 138 Eshnunna (Tell Asmar), 96, 157-9 Gula, 186
date-palms, 19 Etana, 144 Guran, Tepe, 27, 32, 34
Deh Luran, 33 E-temen-anki, 226, 229 Gutians, 139, 147
Delougaz, P., 94, 103, 114 Ethiopia, 191 Guzan, see Halaf
'Deluge' text, 201 Euphrates, 12, 16, 19, 72, 99,
democracy, Sumerian, 88, 89 105-7, 114, 138, 158, 222, 230;
de Morgan, J., 86, 142 crossings, 80 Hacilar, 27
Dilmun, 188 'extensive' cultivation, 18 Haines, R. C, 107
247
Index
Hajyi Firuz, 33 Ishtar-Kititum Temple, 161, log Kurigalzu, 153, 174

Hajji Muhammad, 36, 45, 66, 81 Israel, 194 Kurnah, disaster, 197
Halaf, Tell, period, 27, 67, 74; Istanbul Museum, 138 Kiiyiinjik, 25 ; Ishtar Temple, 69,
pottery, 66-8, 43, 79, 44, 80, 81 ivory, 215, 2i9ff. 142; sounding at, 69, 198, 141
Hall, H. R., 107

Hamad, 13
Hammurabi, 157, 171, 108; 171, Jacobsen, Th., 18, 88, 89, 106, 150 Lacheman, E. R., 178
159, 184 'Jaffar phase', 33
Lakish, siege of, 134, 211
'Hanging Gardens', 228, 172 Jarmo, 25 33, 7, 35, 66, 71, 72; lamassu, 141, 202
Hansen, D. P., 116 ancient environment, 35
Lamberg-Karlovsky, C. C., 118
Hariri, Tell, see Mari Jazirah, 2, 14, 19, 70, 77
land, ownership of, 122
Harmal, Tell, 160; laws of, 162; Jehu, 194
langraum, 162
lions, 171, iig Jemdet Nasr, 38; period, 36, 38,
Larsa, 57, 90, 157, 159; period,
Harran, 13, 191, 222 40, 49, 54, 55, 66, 71, 72; seals,
160, 161
Harsagkalama, 105 2g\ pottery, 61, 62, 10
Law, Antiquities, 10
Hashhamer, ijo Jericho, 27, 28, 33
Layard, A. H., 192, 137, 194, 140,
Hassuna, 25, 34, 35, 66, 69-70; Jerusalem, 222
195 passim
pottery, 'Standard', 69, 77, 40, Jews, 222
lead,70
79; 'Archaic', 69, 77, 79 Jonah, 198
Le Breton, L., 87
Hatra, 70 Jordan, 28
Leiden, conference at, 38
Hazar Golii, 16 Jordan, J., 38, 60
Lenzen, H. J., 48
Hazar Merd, 22, 23, 3
Levant, 26, 27, 32, 188, 131, 219
Hebrew scriptures, 92, 223
Lilith, 171, 118
Herodotus, 18, 223, 229 Kalhu (Calah), see Nimrud
limmu-lists, 187
Herzfeld, E., 68, 72 Kanesh, 158
'Lion Hunt', 199
Hiba, Tell al-, 106, ioj, 150 Karaindash, Temple of, 172
Lloyd, S., 39, 52, 69
Hill, Harold, 162, log Karim Shahir, 22, 23
Loftus, W. K., 196, 140, 160
'hilly-flanks zone', 19 Karkheh, 15, 86
Loud, G., 89, 144, 202ff.
Hilprecht, H. V., 139 Karun, 15, 86
Louvre, the, 116, 138, 144, 147,
Hittites, 160, 187 kaunakes, 114, 130, 144, 171
159
Homs, 158 Kayseri, 138
lugal, 88, 89
Howe, B., 25 Keban dam, 6
Lugal-zaggesi, 137
Hurrians, 138, 149, 158, 159 Kenyon, K. M., 28
Lullubi, 138
'husking-trays', 70 Kermanshah, 24, 3, 32
Luristan, 138
Khabur, 10, 76
Khafaje, }6, 55, 58, 92, 94, 103,
lakhdunlim, 158 120; houses, 142; Oval Temple,
Ibiqadad II, 161 36, 94, 51-3; pottery, 132; Mackay, E., 105
Idi-ilum, 169 sculpture, 1 12-14, ^75 Sin Magan, 138, 149
Ilushuilia, 161 Temple, 36, 40, 55, 93, 30, 120 Magdalenian, 22, 24
'Imdugud-Sikurru period', no, Khanikin, 19 Makran, 38
117 Khosr, 196, ij6 Mallowan, M. E. L., 10, 25 passim
Imgur-Enlil, 201 Khuzistan, 15, 85, 87 Malyan, 87
Inanna, 49, 57, 25, 59, 107 'king' figure, 58-60 Mandali, 72
India, 188 King-Hst, 88, 49, 90, 92, 93, 105, Manishtusu, 138, 142
Indo-European, 160 106, 108, 139 Mardikh, Tell, 231
inlaidornament, 129, 130 kingship, 88, 135 Marduk, 226
57, 222,
Iran, 26, 32, 80, 85, 87 King-of-Battle, epic of, 138 Marduk-apal-idinna (Merodach-
Iraq, Antiquities Directorate, 25, Kirkbride, D. (Mrs H. Helbaek), baladon), 189
39, 69, 151, 174; modern, 12, 29, 70, 72 Marduk-nasir, 122
18, 25; Museum, 52, 69, 102, Kirkuk, 33, 7i, H?, i^7 Mari, 10, 19, 90, 107, 108, 135,
142, 144, 203, 208; northern, 19 Kish, 36, 38, 90, 92, 105, 120, 139, 136, 157; archives, 158; Ishtar
Iron Age, 188 230, 174 Temple, 108, 157; inlays, 130,
irrigation, 16, 17, 31, 33, 72 Koldewey, R., 10, 90, 188, 223ff. 8g, 131; murals, 116, 113;
Ischali, 160, 161, log kudurru, 122, 175 palace, 122, 124, i64ff., 111,
ishakku, 106 Kufah, 13 112; 'Palais Presargonique', 124
Ishtar, 49, 108, 100 Kiiltepe, 158 marsh-dwellers, 25, 41, 58
Ishtar Gate, 226, i6g, iji Kurdistan (Iraq), 19, 21, 25, 26, 32 Matarrah, 71
248
Index
mausoleum (Ur III), 15 1-3, 105, New Year's Festival, 127, 182, pottery, Halaf, 66-8, 43, 79, 44,
155 183, 187, 226 80, 81; Eridu, 36, 39, 12, 44,
Medes, 191, 222 New York University, 106 45; S. 'Ubaid, 37-9, 44, 12, 13,
megaron, 38, 76, 109 Nippur
Niffer, see 45, 47; N. 'Ubaid, 65-7, 75, 80,
Mellaart, J., 30, 31 Nimrud, 16, 188, 192, I93ff., 139, 81; Protoliterate, 61, 10, 132;
Meluhha, 138 140, 141 ; buildings, 2i2ff. ED I and II, 132, 92; ED III,
Memphis, 191 ivories, 214, 215, 2i9ff., 163-6 134

Mersin, 31 Nineveh, 16, 25, 65, 189, 190, Predynastic, 65
Mesannipadda, 91, 92, 106, 108, 136, 197, 141; 'Ninevite One', 'Pre-pottery Neolithic', 28, 30, 33
no 69; 'Ninevite Five', 132, 134; Pre-Sargonid period, 91, 108, in
'Mesilim' period, no fall of, 198; population, 214;' priests, 57, 28, 112, 83; ham, 44
Meskalamdug, 102, 103, 107, 6g, sounding at, 66, 69 private houses, 141, 162, 110, 230
117, 127, 90, 142; style, no Nin-gal, 153, 156' Procession Street, 226, 1 68, 227,
Mesolithic, 27 Ningirsu, 117, 127, 147, 149, 150 172
Mesopotamia, 12, 2; northern, 65, Nin-khursag, 57, 103, 167 Proto-Elamite script, 87
67, 107, 108 Ninmakh Temple, 230, 173 ProtoHterate period, 40, 54, 55,
metal, 65, 81, 113, 82, 83, 127, Ninni-Zaza, 167 67, 82, 103, 118, 124
128, 86 Nintu, 57, 114 Proto-NeoHthic, 28
Metropolitan Museum, 106, 150 Ninurta, 57, 196 Pu-abi, 125
Mitanni, 160, I76ff., 186 Nippur, 36, 57, 91, 107, 137, 139 Puzur-Ishtar, 169
M'lefaat, 22 Nisibin, 160
Mohenjo-daro, 118, 147 Noah, Babylonian, 92
'Mona Lisa', 214, 158 Noldeke, N., 38
Moorey, P. R. S., 105 nomads, 26, 29 Qal'at Sharqat, 107
Moortgat, A., 40, 109, no Nusku, 184 Qatnah, 158

Mosul, 68, 70, 167, 189, 196, 197 Nuzi, 177 Que, 188
Mousterian, 21, 22, 24
mudhif, 41
Muqayyar, see Ur Oates, D., 167, 216
ram ornament, 132, 90
mural paintings, 31, 6, 52, 71, 55, Oates, D. and J., n, 30, 75 Rapiqu, 141
166, 113 Obelisk, 'Black', 194
Rassam, H., 49
Mureybet, 27, 29 obsidian, 32, 70
Ras Shamra, 80
Mursilis I, 160 Old Testament, 12, 171, 188, 189,
Rawa, 19
musical instruments, 100, 116, 87, 214, 221-3 Rawlinson, G., 194
131, 90 Oman, 138 'reception suite', 124
Muski, 187 omens, 44 reed architecture, 41, 9, 59, 28, 29
Mycenae, 67 onager, 71, 33, 128, 86
rehgion, 43, 55, 57, 122, 172
opferstdtten, 55
Reza'iyeh, see Urmia
Oppenheim, Max von, 66 Rich, C. J., 223
Nabonidus, 222, 226 orthostats (slabs), 203
rietnchen, 118
Nabopolassar, 191, 222, 226
Rimah, Tell, 158, i67ff., 114, 113
Nabu Temple, Khorsabad, 202
Rimush, 138
Nimrud, 196, 140 palaces, 106, 122, 140, 155, 161,
rivers, 12, 2, 15
Nairi peoples, 187 76, 106, 141, 142, 144, 171,
rock reliefs, 138
Nannar, 57, 153, 156 172, 200
Romania, 62
Naram-Sin, 138, 139; Stela of, Palaeolithic period, 21, 22, 24;
roofs, 119, 204
138, 142, 96 terminal, 27
Rowanduz, 24, 3
Nasiriyah, 15 Palegawra, 22, 3, 25
Rowton, no
Natufian, 26, 28, 29, 5 Palestine, 22, 219
'natural habitat zone', 26 parapet design, 184, 186
Neanderthal, 21, 14 Parrot, A., 10, 108, 150, 157, 164
Nebi Yunus, 198, 141, 199, 216 Parthian period, 179 'sacred herd', 59
Nebuchadnezzar Pennsylvania, 176 'sacred marriage', 88, 127, 182
(Nebuchadrezzar), 99, 151 Perigordian, 22 sacrifices, 44
Neo-Babylonian period, 106, 151, Persians, 222 Safar, Fuad, 39, 52, 69, 150
222ff. phalanx, 116, n7 Saggs, H. W. F., 44, 88, 122
Neohthic man, 21 ; period, 27, 28, Pliny, 12 Sak(;agozii, 31
31-3, 70, 77; 'Revolution', 26 'polos' headdress, 114 salinization, 17
249
Index
n
Samarra, period, 70, 71, 73 ;
Sippar, 57, 90, 139, 142 Thomas, Felix, 197
pottery, 66, 68, 71, 72, 77, 41, 227, 1 70

sirrush, tieftempel, 153, 181, 229
42, 78, 80 Smith, G., 192, 201 Tigl^th-Pileser I, 183; III, 186,
Sammuramat, see Semiramis Solecki, 23 188, 189
Sarab, Tepe, 32; 'Venus', 32 Solutrian, 22, 24 Tigris, 12, 16, 93, 106, 108, 179,
Sargon-of-Akkad, 91, 135, 137, Soviet archaeologists, 70 190, 193
139, 142 Speiser, E. A., 67 Til Barsip (TellAhmar), 188, 208
Sargon II, 189, 196 stamp-seals, 67, 70, 85, 47 Tilla,Tehip and Shurki, 177
Sarzec, E. de, 106, 147, 149, 150 Standard (Ur), 103, 130, 87, 88 timber, 138, 204
Sawwan, Tell-es-, 35, 66, 71, 73, Starr, R. F., 176, 178 'tortoise jars', 81
35, 74, 45 'steatite bowls', 117, 71 'tree-of-life', 155, 206

'scarlet ware', 132, 134, g2 'Steinplatten' inscription, 223 T-shaped buildings, 73, 35
scribes, 90, 122 Stela of Hammurabi, 108 'Tummal' inscription, 92
sculpture, 31, 56; Protoliterate, 56, Stela of Ur-nammu, 155 Tutub, see Khafaje
58, 23, 26, 59, 27, 112, 6j; 'Stela-of-the- Vultures', 106, 116, Tyre, 131, 208
Early Dynastic, 59, 97, 108, 6g, 142
109, 62, III, 112, 64; Mari, Stone Age, Old, 21, 22, 25
114, 65, 66; relief, 114, 67-g, stones, gem-, 82 'Ubaid, Tell Al, 25, 36, 37, 47,
115-17, 70, 71; Akkadian, 95, stone, building-, 119, 203 103 ; culture, 47; Nin-khursag
138, '142, g6, g7, 144; Gudea, Strabo, 12 Temple, 36, 37, 90, 103, 60,
47, 101, 102, 149; Ur III, 107, suchgrahen, 179 105; 'northern', 65-7, 75, 80,
155; second millennium, 108, Sumer, 15, 18 81; period, 27, 36, 38, 63;
1 16-19, 122-4, 184, i2g; Late Sumerians, 37, 39, 107; origins, pottery, 37-9,44, 12, 13, AS,
Assyrian, 204ff., i46ff.; metal, 62, 63, 137 47; settlement, 36, 37, 63, 31,
83, 128, 86, 93 'Summer Palace', 226, 167 103, 60; small objects, 46
Scythians, 191 Susa, 32, 86; pottery, 86, 48, 137, Uhaimir, 105
Sea-Land, 160 142, 144, 171 Umma, 107, 116
'Sea-of-the-West', 138 Susiana (periods), 27, 85, 87 Umm Dabaghiyah, 35, 66, 70, 71,
Semiramis, 196 Syria, 10, 16, 29, 85, 108, 138, 33, 72, 34, 73, 77
Sennacherib, 179, 190, 191 157, 177, 186, 219, 231 'Uqair, Tell, 36; Painted Temple,
Shaduppum, see Harmal Syrian Desert, 20, 136 36, 52, 22, 25; Jemdet Nasr
Shalmaneser I, 160 chapel, 36, 54; 'Ubaid
Shalmaneser III, 179, 188; Fort, settlement, 36, 52, 63; small
216, 161 Tabal, 188 objects, 14
Shamash, 57, 81, 108 tablets, archaic, 55, 56, 24; Ur-of-the-Chaldees, 15, 57, 81,
Shamash Gate, 198, 141 cuneiform, 49, 122, 157; ivory, 90; soundings, 36-8; site, 151;
Shamshi-Adad, 158, 164, 167, 174, 214 walls, 56, 230; First Dynasty,
179, 181, 196 Tammuz, 57, 59 103, no; Royal Tombs, 90, 92,
Shanidar, Cave, 22, 23, 5 Tarsus, 31 99, loi, 58, sg, 102, 103,. 106,
Shar-gali-sharri, 139 Taurus, 17, 26, 138, 188 107, no, 116, 151, 87, 88, go,
Shatra, 106 Tell Afar, 70 127, 129; Third Dynasty, 99,
Shatt-al-Arab, 86, 107 Telloh, 106, 107, 6g, 147, 149, 150 150, 151; ziggurat, 151, 152,
Sha'ur, 86 temples, 119; Eridu, 41, 10, 43, 104; Mausoleum, 15 1-3, 103,
Shaushattar, 176, 178 49, 50; Early Dynastic, 93, 30, 155; other buildings, 153
Shell, v., 139 98, 108, 119; Ur III, 155, io6\ Urartu, 187, 188, 220
Shilwi-Teshup, 178 Larsa period, 161, log; second Ur-baba, 147, 149
ships, 138 millennium, 167, 114; Mid- Urbartutu, 92
Shubad, see Pu-abi Assyrian, 126, 127, 1 8 iff. Late ; Urmia, 22, 32, 188
Shubat-Enlil, 158 Assyrian, 196, 202, 203, 160; Ur-Nammu, 142, 150, 151, 155,
Shuruppak, see Farah Neo-Babylonian, 230, 174; 107, 130
Shu-Sin, 155, 106, 161 'platform', 52, 103,-105, 119; Ur-Nanshe, 106, 68, 149; (singer)
Shutruk-nakhkhunte, 142 size, 121; function, 122 114, 66
Sialk, 87 terminology (Uruk period), 40; Ur-Ningirsu, 149
silver-mines, 138 pre-Sargonid, no Uruk (Warka), 27, 36, 48ff., 17;
Sin, 57, 93; of Harran, 222 'terrazzo pavements', 30 period, 27, 36, 38; terminology,
Sinjar, 70, 77, 81, 158, 167 'Tete au Turban', 150 40, 63, 66; Anu area, 36, 17, 18,
Sinkara, see Larsa texts, written, 9 49; Eanna Precinct, 36, 38, 17,
Sin Temple (Khorsabad), 203 'tholoi', 67, 74, j6, 75 49-51, 20, 54; White Temple,
250
Index
36, 17, 18, 52, 54; Pillar Hall, tracing, 94; wall-plaques, 114, Yorghan Tepe, 176
50, ig, 52; Riemchengebdude, 55: 67, 68
Stone-cone-mosaic Temple, 51, Warka, see Uruk Zab, Greater, 19, 118
55 Washshukanni, 160, 176 Zab, Lesser, 19, 193
Utnapishtim, 92 Watelin, C. H., 139 Zagros Mountains, 13, 17, 20, 22
Utu, 57 weapons, 82, 127, 177 Zarzi, 22, 23, 5, 25
Utu-hegal, 150 wheeled vehicles, 106, 128, 86, 8 Zawi Chemi, 22
178 ziggurats, 39, 49, 203 ; Ur, 99,
windows, 119 107, 139, 151, 152, 104, 156,
winged bulls, 202, 145, 146 157; Kish, 105, 106; Ashur,
Van, Lake, 16, 25, 188
wooden vessels, 3 181, 126; 'Aqar Quf, 121, 174;
van Loon, M., 29
Woolley, C. L., 25, 99ff. and Babylon, 226, 229
vaulting, 115, 169, 204
passim Zigi, 177
Wright, H. E., 25 Zimrilim, 124, 158; Palace of,
writing, 55, 24, 87, 90, iii, 137 i64ff.
j
Wadi-al-Natuf, 26 Ziusudra, 92
walls, enclosure and defensive, 72, Yahya, Tepe, 33, 37, 118 ^ zone sacre, 124
i
151, 179, iî, 224; wall- Yarim Tepe, 70, 72, 83 Zu, 144
251
7 5/3///fe
AHGHAECLOGY OF MESOPOTAMIA
DS69.5 L58/1984
*
85603232-56
WBx
Boston Public Library
WEST ROXBURY
BRANCH LIBRARY
The Date Due Card in the pocket indi-

cates the date on or before which this
book should be returned to the Library.
Please do not remove cards from this
pocket.
Ancient Mesopotamia, the "valley of the twin rivers," Tigris and
Euphrates, was the cradle of civilization in the Near East. Here,
among the Sumerians, the earliest experiments were made in writing
and mathematics, organized religion and the communal administration
of city-states. For over a century, archaeologists have sought to
uncover remains of the first human settlements and the great towns
into which they later developed. Today, several thousand books and
articles concerned with the archaeology of Mesopotamia are in
existence, yet curiously no recent attempt has been made until this
present study to provide an up-to-date synthesis of their conclusions.
The chronological sweep extends from the Stone Age down to the
conquest of Babylon by Cyrus th^ Gre?.t in 39 bc. ^^l.ing in the pre- '
dynastic millennia, the rise andfall of Suiixji and Akkad,
Hammurabi's united kingdom, Assyria and the empire which it

bequeathed to the later Babylonian kings. Professor Lloyd's method is
to analyze systematically the wealth of material produced by major

and minor excavations since 1900 and the developments thus revealed
in the realms of art, architecture, religion and social history.
For the revised edition the text has been updated throughout,
incorporating' an account of recent discoveries and expanded notes
and bibliography.
During a career of thirty-five years in the Near Eastern field, Seton

Lloyd excavated a score of important sites, including several in
Mesopotamia. Among his many books dealing with the archaeology
of this area are Ar" of the Ancient Near East, Foundations in the Dust
(recently reissued in a n^vised edition by Thames and Hudson) and
Mounds of the Ancient Near East. He has been successively
Archaeologies Adviser to the Iraq Government (1959-49), Director
of the British Institute of Archaeology at Ankara (1949-61) and
Professor of \)C estern Asiatic Archaeology in the University of
London (1962-69).
^' His first-hand acquaintance with the subject is unequalled, and the result is a
resounding success . . . a model of lucidity and balance.''
Nicholas Postgate, ^;;//^////)/
^' Sure to become the graduate student's constant companion, as well as a

standard reference for interested laypersons." Eibrary Journal
; ^' i
with 174 illustrations

ON THE cov^K•. Lifc-size head in bronze of
an. Akkadian king (photo Maji Hirmer)
Thames and Hudson —^

500 FIFTH AVENUE, NEW YORK,' NEW YORK lO I lO ^^ *o'
FPT ?
Printed in the German Democratic Republic ISBN 0-SDD-7TDD^-H >$10.T5

The Archaeology of Mesopotamia (History Arts Ebook)

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 Archaeology of Mesopotamia (History Arts Ebook)

Uploaded by

Copyright:

Available Formats

SETON LLOYD

with 174 illustrations

THAMES AND HUDSON

condition including these words being imposed on a

© 1978 and 1984 Thames and Hudson Ltd, London

First paperback edition 1980

Library of Congress Catalog Card Number 84-50036

All Rights Reserved. No part of this publication may be

Printed and bound in the German Democratic Republic

Preface to the First Edition 9

Preface to the Revised Edition 1

1 The Land and its Rivers 12

2 The Twilight of Neolithic Man 21

3 The Threshold of Written History 37

5 The Early Sumerian Dynasties 88

6 Pre-Sargonid Art and Architecture 111

7 The Dynasty of Akkad and the Sumerian 135

8 The 2nd Millennium BC 157

9 The Late Assyrian Period 187

10 Babylon: The Last Mesopotamian 222

Notes on the Text 233

Photographic Acknowledgments 245

Anyone book may well be

justifiably feel that it calls for some immediate quahfication.

themes of all these writings, the purpose of the conspectus which

Egypt or Mesopotamia, where a significant proportion of the

As may already have been inferred from these observations, it is

Preface to the Revised Edition

These introductory paragraphs to the First Edition were written

in southern Iraq seemed temporarily to have lost impetus. Work

north, where excavations on a smaller scale could produce profit-

standing enigma regarding the antecedents of the Sumerians,

is primarily intended as an introduction to further reading. For

The Land and its Rivers

The name 'Mesopotamia' is an archaism, traditionally more often

to Babylonia, and requires some explanation. If one approaches the

\\, TURKEY LAKE

The Two Regions

Already, these few sentences will have drawn attention to the

will become more apparent if we now consider the geophysical

- it consists of a broad and shallow depression, running

Httle difficulty. To the northeast they correspond to the diminish-

I Map showing some Near

delta, there an entirely different landscape, a country which is the

creation of rivers. Here, from time immemorial, the Tigris

Having reached this point in a description of Mesopotamia,' we are

wide and sparsely populated region of marshes and lakes. It was

explicitly described by ancient writers as 'standing upon the shores

which ocean-going vessels discharged their cargoes. Both of these "*

evidence has helped to make this possible. In the temple at Eridu,

its turn, connected by deep channels with the Euphrates estuary.

sources between Van and Erzerum. Both branches flow at first

Euphrates, are now huge de cated ruinfields, far out in the

The system by which this is done is naturally

Climate and Irrigation

Where the climate of Mesopotamia is concerned, it is well to

history represented an ominous of the

complicated pattern, and this extraordinary sight could lead to the

Next to the effects of flooding, the most disastrous danger to

soil and the consequent practice of what is known as 'extensive

deposited, ruining the fertiUty of the soil. When this happens to

agricultural productivity during these centuries was only a

northern provinces too must occasionally have suffered, as they do