Victor Prescott, Gillian D. Triggs - International Frontiers and Boundaries - Law, Politics and Geography (2008) PDF
Victor Prescott, Gillian D. Triggs - International Frontiers and Boundaries - Law, Politics and Geography (2008) PDF
Editorial board
Professor Madeline Caviness, Tufts University (USA)
Dr Eva Frojmovic, University of Leeds (UK) (series editor)
Professor Diane Wolfthal, Arizona State University (USA)
Professor Catherine Harding, University of Victoria (Canada)
VOLUME 2
International Frontiers and
Boundaries
Law, Politics and Geography
by
Victor Prescott and Gillian D. Triggs
LEIDEN • BOSTON
2008
This book is printed on acid-free paper.
Acknowledgements vii
List of Illustrations ix
1. Introduction 1
2. Frontiers 23
3. The evolution of international boundaries 51
4. International boundary disputes 91
5. International Law and Territorial Boundaries 137
6. Maps as Evidence of Territorial Frontiers 191
7. River Boundaries and International Law 215
8. The Americas 233
9. The Middle East 265
10. Africa 291
11. Asia 317
12. Europe 349
13. Land boundaries on islands off Southeast Asia 373
14. Antarctica 383
Annexes 409
Illustrations 411
Bibliograpy 485
Legal Materials 489
Table of Cases 489
Table of International legal instruments 491
13.1 The boundary between East Timor and Indonesia in the east
of the Island 480
13.2 The boundary between Oecussi and Indonesia 481
13.3 Boundaries on Borneo 482
International frontiers and boundaries separate land, rivers and lakes subject
to different sovereignties. Frontiers are zones of varying widths and they were
common many centuries ago. By 1900 frontiers had almost disappeared and had
been replaced by boundaries that are lines. The divisive nature of frontiers and
boundaries has formed the focus of inter-disciplinary studies by economists,
geographers, historians, lawyers and political scientists. Scholars from these
disciplines have produced a rich literature dealing with frontiers and boundar-
ies. A survey of this extensive literature reveals that the following themes have
attracted most attention.
When the histories of nations are unravelled it is plain that most of them did
not emerge at one time within a single set of international limits that have
remained unchanged. It is the case that many modern African states became
independent after 1945 within boundaries that have not subsequently changed.
However, research into their pre-colonial and colonial antecedents reveals a
varied sequence of frontiers and boundaries. A significant part of the history of
several countries concerns the struggle for territory, and the identification of the
succession of national boundaries, on a single map, provides a shorthand account
of stages in the progression to the present pattern of national states. Sometimes
the events, that established new limits, were sufficiently significant to mark the
division between important periods in the political history of countries or the
diplomatic and military history of continents. This point can be illustrated by
Figure 1.1, which shows the boundaries of Greece since 1832, the year when
the modern state of Greece emerged from nearly three centuries of Turkish rule.
The following commentary is based on the volume on Greece produced by the
Naval Intelligence Division (1944). The rebellion began in the Peloponnisos in
April 1821, and this area was quickly cleared of Turks. The Greek sailors also
enjoyed success against the Turkish navy in the Aegean Sea, but the tide of
rebellion had been halted by June 1827 and other foreign powers decided it was
time to intervene to avoid continued instability in the region. The destruction of
2 Chapter 1
the allied Turkish and Egyptian fleets by naval squadrons from Britain, France
and Russia in Navarino Bay on 20 October 1827, paved the way for an enforced
settlement dictated by these countries. In 1830 it was proposed that the northern
boundary of Greece should run southwestwards from the vicinity of Lamia to
Mesolongion, but Palmerston, the British Prime Minister, then persuaded his
allies that the line should run from Pagasitikos Kolpos [Gulf] to Amvrakikos
Kolpos (Figure 1.1) Palmerston defeated proposals to give Greece the islands
of Samos and Kriti on the grounds that the former was too close to the Turkish
coast and the latter was too valuable and had a large Turkish population, that
should not be subject to Greece.
On 29 March 1864, six months after Greece had installed a Danish King,
Britain ceded the Ionian Islands to Greece. They stretch from Kerkira [Corfu] in
the north to Kithira in the south off the west coast of Ipiros and Pelopponisos.
These islands had been formed into the United States of the Ionian Islands in
1815, by the Treaty of Versailles and placed under British protection. Their union
with Greece followed a unanimous vote of support by the legislative assembly
of the islands.
The war between Russia and Turkey in 1877 presented Greece with an oppor-
tunity to claim the Greek provinces in Turkish Europe, but the decision to take
action was delayed so long that the Greek army could not be deployed before the
war was over. The peace treaty signed by Russia and Turkey at San Stefano on
3 March 1878 contained no territorial gains for Greece. Fortunately for Greece
the major European powers were dissatisfied with the terms Russia had forced
on Turkey. This treaty was revised at the Congress of Berlin, attended by all
the major powers in June and July 1878. Although Greece was not represented
in Berlin, Britain persuaded its allies to require Turkey to make concessions to
Greece along their common land boundary. At first it was proposed that the new
boundary should run from Stoupi in the east to the mouth of the Thiamis River
in the west opposite Kerkira Island. This would have given Greece the region
of Thessalia and most of Ipiros. Turkey could scarcely resist the cession of the
former region, with its pronounced Hellenic character, but it managed to retain
the areas of Janina and Preveza, that constituted most of modern Ipiros. The
arguments in favour of continued Turkish control in these areas centred on their
large Moslem minorities. The final treaty was signed on 24 May 1881, although
Greece was not a party to these arrangements.
The Greek authorities overplayed their hand in 1897 when they tried to force
further concessions from Turkey in Kriti and the mainland. The European pow-
ers had to intervene to prevent a Turkish victory, and Greece was forced to cede
11 small areas that had particular strategic interest along its northern boundary
to Turkey. The next major territorial advance for Greece came in 1912 and
1913 in wars, first with Turkey in alliance with Bulgaria and Serbia and then
with Bulgaria in alliance with Serbia. These advances enabled Greece to move
northward to its present boundary and eastwards along the Macedonian coast
Introduction 3
to the Nestos River. In the Aegean Sea Greece gained many Turkish islands,
apart from Gokceada and Bozcaada that Turkey retained, and the Dodecanese
Islands, that had been occupied by Italy in April and May 1912. Italy seized
these islands from Turkey as part of its campaign against Turkey over control
of Tripoli and Cyrenaica.
The role of Greece in World War I was rewarded with Bulgaria’s coastlands on
the northern shore of the Aegean Sea in western Thraki [Thrace] and the Turkish
area of eastern Thraki, as far east as Catalca, only 32 km from Istanbul. A large
area of the Turkish mainland around Izmir [Smyrna] was placed under Greek
control, and it was arranged that a plebiscite in five years could opt for union
with Greece. The Treaty of Sevres, which conferred these gains on Greece, was
not ratified and the replacement Treaty of Lausanne, on 24 July 1924, left only
western Thraki to Greece. This dramatic change was produced by the revival of
a nationalist government in Turkey, agreements between that government and
the Soviet Union, and France, and by Greek military defeats by Turkish forces.
Turkey was helped by the unpopularity of King Constantine with the western
powers because of his pro-German policy in 1915. The Treaty of Lausanne
marked a retreat from the line marking the largest territorial extent of modern
Greece, and Turkey’s boundary along the Meric River was restored.
The final territorial extension of Greece occurred in 1947 when Italy ceded the
Dodecanese Islands. Megisti Island, obtained by Italy from France in 1920, was
also ceded to Greece and marked the most easterly extent of the realm.
In tracing the evolution of national boundaries there is no substitute for the
treaties, protocols, agreements and conventions that specify the formal arrange-
ments between states. Sometimes these can be difficult to find and hard to
translate. Fortunately there is now a comprehensive range of publications deal-
ing with territorial treaties. For the period 1648-1919 which is ‘. . . classically
regarded as the date of the modern foundation of the modern system of states’
(Parry, 1969, vol. 1, 3), Parry has edited a collection of treaties filling 231 vol-
umes. From 1920 until the present many treaties are published in the League of
Nations Treaty Series and its successor the United Nations Treaty Series. These
collections contain only treaties lodged with these organizations and therefore
they are not comprehensive.
One serious problem in interpreting treaties relates to the need to consult maps
or plans that were contemporaneous with the treaty. Maps are often attached to
treaties, but they are not reproduced by Parry and do not always appear in the
other two collections. The importance of using maps that were contemporary
with the treaty turns on the fact that modern place names might be completely
different, and sometimes maps were inaccurate and those inaccuracies were
carried into the written description of the boundary. For example, the boundary
between British Rhodesia and Portugal’s Mozambique south of parallel 18° 30 S
was defined in the following terms on 11 June 1891.
4 Chapter 1
. . . thence it follows the upper part of the eastern slope of the Manica pla-
teau southwards to the centre of the main channel of the Sabi . . . (Brownlie,
1979, 1119)
The demarcation teams sent by both countries to fix this line could not agree on
the location of the upper part of the eastern slope of the Manica Plateau and the
matter was referred to the arbitration of Senor Vigliani, an Italian senator. He
handed down his decision on 20 January 1897 (The Geographer, 1971, 2). The
original negotiators can be excused for producing such an uncertain definition
in view of the maps found in Harare and London. It is clear that the maps and
sections sent by field surveyors show the edge of the plateau to be a prominent
apparently unmistakeable feature.
Fortunately there are regional guides to the literature and treaties dealing with
the evolution of international boundaries. Nicholson (1954), Paullin (1932) and
Garret (1988) have provided useful accounts of international boundaries in North
America. The atlases edited by Paullin and Garrett contain some helpful maps.
Rebert (2001) has provided an excellent, detailed and well illustrated account
of delimitation and demarcation of the Mexican-United States boundary. Ireland
(1938) produced a wonderful account of the stages by which South America’s
international boundaries developed before 1938. Three years later Ireland (1941)
produced a matching volume on Central and North America. Hertslet (1875-91)
produced a monumental record of boundary-making in Europe at the end of
the nineteenth century. Hertslet (1909, 1967) then turned his attention to Africa
and produced three volumes that were reprinted in 1967. Brownlie (1979) built
on Hertslet’s African work, by compiling an encyclopaedia recording the trea-
ties governing the evolution of Africa’s boundaries illustrated by clear maps of
the final results. Brownlie does not provide the diplomatic history of Africa’s
boundaries in the way that Ireland does for South America and Prescott does for
mainland Asia (1975 and 1977).
The Geographer of the United States State Department produced an invaluable
set of studies of nearly 200 separate international boundaries. These studies were
published, on a compact disk, by the United States Department of State’s Bureau
of Intelligence and Research (2002).
In the 1990s the International Boundaries Research Unit at the University of
Durham published two useful journals dealing with international land boundaries.
Each issue of Boundary and Security Bulletin had news sections on boundar-
ies in each of the continents and a number of boundary articles, Boundary and
Territorial Briefing dealt with one boundary subject or one boundary. Thirty-six
volumes of Boundary and Security Bulletin were published between 1993-2002
and 24 volumes of Boundary and Territorial Briefing were published between
1994 and 2002. These excellent series had a current value when published and
now are historically important.
Introduction 5
Boundaries are the lines of physical contact between states and provide oppor-
tunities for cooperation and risks of discord. Lord Curzon of Kedleston, who
was once Viceroy of India, summed up the situation in words that have often
been quoted.
Frontiers are indeed the razor’s edge on which hang suspended the modern
issues of war and peace. (Curzon, 1907, 7)
Most commentators use these words to introduce discussion of boundary conflicts,
but Curzon’s reference to peace should not be forgotten. Regrettably, editors of
newspapers and radio and television bulletins seem to regard boundary disputes
as more worthy of attention than cooperation to solve boundary questions. In
fact both discord and concord regarding boundaries are important subjects for
discussion. Indeed a survey of the United Nations Treaty Series (United Nations
1945) reveals dozens of treaties dealing with cooperation between states sharing
a common boundary as the following contemporary examples show.
In December 2003 Russia and the Ukraine agreed on terms for the use of Kerch
Strait at the mouth of the Sea of Azov. In October problems had arisen when
local Russian authorities started to build a causeway from their shore towards a
small island called Tuzla, that is controlled by Ukraine. Tuzla had been attached
to the Russian coast by a narrow spit until 1925 when a severe storm removed
sediment over a 4 km stretch. Work on the causeway caused the Ukrainian
authorities to protest to Russia and reinforce the population of the small island.
Talks between the two governments resulted in agreements to regard the Sea of
Azov and Kerch Strait as internal waters of the two countries, and to make joint
decisions on their use. The effect of this was that Russia shared control of the
Kerch-Yenikal’skiy Kanal through Kerch Strait.
In November 2003 France and Spain agreed on measures to enable both
countries to act in concert or separately against Basque Separatists. Evidence
gathered in each country is admissible in the other, and police from both sides
may act separately or in combination. In 2003 59 suspects had been arrested in
the borderland.
In March 2005 China, the Philippines and Vietnam agreed to permit seismic
studies in the vicinity of the Spratly islands. These countries occupy some of the
Spratly Islands and assert claims to islands occupied by each other. However,
these counter claims have not prevented scientific examination of the seabed’s
structure.
In February 2004 Azerbaijan, Georgia and Turkey agreed to the construction
of an oil pipeline to convey oil from Baku to Turkey’s Black Sea coast. Finally
in October 2003 Bangladesh and India agreed on measures to share the water
of the Tista and six other rivers.
6 Chapter 1
possible limits of its territorial dominion and he called this idea ‘space concep-
tion’. This notion appears to be similar to the concept of les limites naturelles
(natural boundaries), that Pounds (1951, 1954) explored in respect of France,
Pounds established that for much of France’s history after the 16 th century, suc-
cessive rulers regarded France’s desirable boundaries as coinciding with the sea,
the Alpine watershed, the Pyrenees and the Rhine. History records that briefly
Napoleon had a broader vision of the French state (Zamoyski, 2004, 12-3).
Ratzel’s view on the space conception of states followed logically from his
belief that the state was a living organism that grew and decayed. The bound-
ary and the adjacent area, that is called the border, formed the epidermis of this
organism that provided protection and allowed exchanges to occur. So for Ratzel
the border was a dynamic feature and when it was fixed in position we were
witnessing a temporary halt in political expansion. He enunciated two laws of
territorial growth.
The law of the evolution of boundaries can be defined as a striving towards
simplification and in this simplification is contained a shortening of borders.
(Ratzel, 1897, 555)
In accordance with the general law of growth of historical spatial phenom-
ena the borders of the larger areas embrace the borders of the smaller one.
(Ratzel, 1897, 557)
Evidence for devising these rules can be found in the evolution of Germany by
an amalgamation of small marches, kingdoms and principalities. Many examples
involving other countries can be found in the collection of treaties prepared by
Hertslet (1875-91). For example, by a treaty dated 14 April 1816 Austria and
Bavaria exchanged territories in their borderlands. Bavaria ceded parts of Haus-
ruckviertel and Innviertel to Austria and received in return parts of the Department
of Mount Tonnerre and the city and fortress of Landau (Hertslet 1875, 434-43).
Ratzel would also have found support for his laws in the procedures by which
Britain, Belgium, Germany and France were acquiring colonies in Africa. He
also made strong assertions about the nature of borders.
Political balance [between countries] is to a large extent dependent on the
[characteristics of ] borders between them. (Ratzel, 1897, 584)
We have seen how growth and decline of a region not only find expression
in the areal form and protective measures of the border but also, in a way,
prepare and foreshadow themselves therein. (Ratzel, 1897, 605)
This view that the border was the area within which the growth and decline of
the state was organized and became evident, was taken up by students of Geo-
politik thirty years later. Geopolitik was the name given to a school of political
geography established by Major-General Haushofer, who emphasized the role of
geography in creating policies that would make the Germany of the 1920s strong
8 Chapter 1
that reflected the distribution of important tribes, trade routes and rivers. This
was especially the case in sub-Saharan Africa, where changes were made to
simplify administration rather than as a consequence of the unequal strength of
the two countries involved. Straight-line boundaries tended to survive in desert
conditions. In 1889 the original division between French Dahomey and British
Nigeria followed a meridian northwards to parallel 9° N. In 1906 the two colonial
powers agreed to adjust the boundary, so that in the coastal section particular
settlements were awarded to France and Britain, and further from the coast the
boundary followed the River Okpara (McEwen, 1991).
Of course throughout history states have sometimes used force to capture all
or part of a neighbours territory. For example, in 1975 Indonesia invaded Por-
tuguese Timor after Portugal had abandoned its overseas empire. The territory
was held by Indonesia until 1999, when the inhabitants were allowed to vote
on their future and selected independence. In 1990 Iraq invaded and captured
the territory of Kuwait. The following year, Kuwait was liberated by a United
Nations force.
While scholars have failed to produce laws or even guidelines by which behav-
iour in respect of international boundaries can be predicted with confidence, they
have succeeded in identifying reliable procedures for studying the evolution of
boundaries and boundary disputes.
A French lawyer called Lapradelle and an American geographer called Jones
defined a series of stages through which ideal boundaries would pass during
their history. In both cases, these major analyses, published 17 years apart, do
not bear the imprint of the period in which they were written. This characteristic
sets them apart from studies by Ratzel, Haushofer, Spykman and Ancel.
Lapradelle identified three stages in the evolution of a boundary that he called
preparation, decision, and execution. The equivalent stages described by Jones
are called allocation, delimitation and demarcation.
The processes of preparation precede true delimitation. The problem of the
boundary’s location is debated first at the political level then on the technical
level. The question is, in general, of determining, without complete territo-
rial debate, the principal alignment which the boundary will follow. . . . The
decision involves the description of the boundary or delimitation. . . . The
execution consists of marking on the ground the boundary which has been
described and adopted, an operation which carries the name demarcation.
(Lapradelle, 1928, 73)
In adopting and improving this pioneer proposal, Jones added the final stage
of administration that describes the maintenance of the boundary monuments,
cleared lines and fences.
Neither of these authors insisted that all boundaries would pass through each of
these stages. In some cases the first boundary, that allocates territory, might be a
straight line and that this line is then demarcated, either because there is no more
10 Chapter 1
two values gives a geopolitical index of zero. The accessibility index and the
instability index for Israel and Palestine are both 5. This means that the geopo-
litical index is 25.
The final calculation produces the national land boundary vulnerability. The
national boundary is the sum all of the individual segments with neighbours. The
national index is derived by weighting the individual indexes for each neigh-
bour, according to the proportion of that boundary length to the entire boundary
circumference.
The difficulty with this mathematical approach to boundary descriptions is
that Anderson gives no examples of how he obtained any particular value for
the potential boundary accessibility or potential instability indices. But once
they have been calculated the geopolitical index and the national land boundary
vulnerability index are generated mechanically. If scholars use these indexes to
compile tables ranking boundary segments according to their risks of conflict it
is unlikely that the results will be useful.
Scholars from different disciplines have written many case studies of boundary
disputes. Collectively they enable the essential elements of boundary disputes to
be identified. A lawyer, Johnson (1966), wrote about the Columbia River between
Canada and the United States. Lamb (1966) and Boban (1993) are historians who
respectively wrote about the McMahon Line of 1914 and Croatia’s boundaries
from 1918-1993. Ronald Bruce St John (1994, 1994a and 1998) is a political
scientist who has analysed the boundaries of Ecuador and Peru, Bolivia, Chile
and Peru and Cambodia, Laos and Vietnam. Many political geographers have
examined international boundary disputes. Richard Schofield (1993 and 1994)
has published seminal studies of the historical claims of Iraq and Kuwait and the
borders and territoriality in the Gulf and Arabian Peninsula in the 20 th Century.
Recent important contributions include those by Klemencic (1994) on Bosnia and
Hercegovina and Newman (1995) on the Green Line boundary between Israel
and the West Bank. The large library of boundary disputes reveals that analysis
of new disputes should focus on the cause, the reason for its development at a
particular time, the aims of the countries concerned, the arguments used to justify
particular attitudes and actions, and the results which follow from the settlement
or continuation of the dispute.
The terms boundary, frontier and border are often used as synonyms in conversa-
tion, in newspaper reports and in television and radio broadcasts. For boundary
scholars they are all specific terms that are not interchangeable, and there are
other terms that have precise meanings.
12 Chapter 1
one hand and Indonesia and Papua New Guinea on the other hand, have drawn
two lines, so that the Australian seabed extends under the water columns of the
other states.
Maritime boundaries are more permeable than international boundaries on
land. Unless special arrangements have been made for citizens living close to
a land boundary, there will usually be some formality when a person crosses
the boundary into the neighbouring state. In contrast the master of a vessel has
the right of innocent passage through the territorial waters of another country.
According to Article 19 of the 1982 Convention passage is innocent providing it
is not prejudicial to the peace, good order or security of the coastal state (United
Nations, 1997, 27).
Unlike land boundaries maritime boundaries are rarely demarcated except where
two countries share a comparatively narrow navigable waterway. All maritime
boundaries are defined by straight lines or arcs of circles, whereas some land
boundaries follow an irregular course. One encouraging feature about maritime
boundaries is that they have not been the cause of wars, although there have
sometimes been incidents that caused difficulties. In contrast, states have some-
times gone to war with a neighbour in an effort to achieve boundary adjustments,
as conflicts between Eritrea and Ethiopia, China and India, Iran and Iraq and
India and Pakistan can attest.
Although national claims to land, lakes, rivers and adjoining seas have been
common since Roman times, national claims to the air above these features were
delayed until the 20th century. Matte (1981, 53-5) has described individual legal
rights to airspace from Roman times to the 19th century. The legal principal was
summed up in the gloss prepared by Accursius in the Middle Ages.
Cujus est solum ejus debet esse usque ad coelum. [He who possesses land
possesses that which is above it to the sky.] (Matte, 1981, 54)
This meant that if water draining from one roof fell onto and damaged the home
of a neighbour, then the neighbour could claim compensation. The first flight by
balloon occurred on 21 November 1783, and 32 years later a British judge ruled
that passage of a balloon over private property did not involve any infraction of
ownership rights (Matte, 1981, 55).
National claims to airspace started when aeroplanes began to cross international
boundaries and when they were used for warfare. The earliest use of balloons
in war recorded by Matte occurred in 1793. French Republican forces used a
tethered balloon Entreprenant [Enterprising or Adventurous] as an observation
platform during the battle of Fleurus (Matte, 1981, 93). The first military use
of dirigibles and aeroplanes occurred in a conflict between Italy and Turkey in
Introduction 15
The Paris and Chicago Conventions defined territory as all land and territorial
waters. This means that on land the horizontal airspace is defined by international
boundaries with neighbours, whereas over the sea it is defined by the limits of
the territorial sea. There are about 150 coastal states in the world and 129 states
claim territorial waters 12 nm wide. Wider claims are made by 13 states and
narrower claims are made by 7 states. There are three possible combinations of
these limits. The horizontal airspace of landlocked countries will be bounded
entirely by international boundaries on land and in rivers and lakes. If there is
some dispute over the location of any of these boundaries the airspace above the
disputed territory will also be involved. However, disputes over territory never
seem to refer to the airspace above the territory.
The airspace of coastal states will be bounded by their international boundar-
ies on land, the outer limits of their territorial waters measured from specified
baselines and maritime boundaries with adjacent or opposite states. The baselines
might be a particular low-water line or straight lines closing legal bays or joining
islands and headlands along an indented coast.
The horizontal airspace of states composed of archipelagos, such as the
Maldives or Fiji will be the limit of territorial waters measured from coasts
or archipelagic baselines. The archipelagic states of Indonesia and Papua New
16 Chapter 1
Guinea share the island of New Guinea, so their airspace will be partly defined
by an international boundary drawn over land. Indonesia also shares Borneo with
Malaysia, that is not an archipelagic state, so again part of Indonesia’s airspace
will be bounded by an international land boundary.
The 1982 Convention on the Law of the Sea contains some rules about the
airspace over maritime zones. Freedom of overflight, over waters in an interna-
tional strait, is guaranteed by Article 38. An international strait is one that links
two areas of seas and measures not more than 24 nm in width. This means that
the strait is totally occupied by the territorial waters of the flanking state or states.
An international strait is used for international navigation from one part of the
high seas or an exclusive economic zone to another part of the high seas or an
exclusive economic zone. Examples of such straits include Juan de Fuca Strait,
Saint Vincent Passage, Strait of Gibraltar, Strait of Malacca, Bab el Mandab
and Torres Strait.
Flight over these straits is described as transit passage that should be con-
tinuous and expeditious. However, this requirement does not preclude passage
through the strait for the purpose of landing in a state bordering the strait, in
accordance with the normal rules of entry. The duties of aircraft pilots are set
out in Article 39.
1. Ships and aircraft, while exercising the right of transit passage shall:
(a) proceed without delay through or over the strait;
(b) refrain from any threat or use of force against the sovereignty, territorial
integrity or political independence of States bordering the strait, or in any
other manner in violation of the principles of international law embodied
in the Charter of the United Nations;
(c) refrain from any activities other than those incident to their normal modes
of continuous and expeditious transit unless rendered necessary by force
majeure or by distress;
(d) comply with other relevant provisions of this Part. (United Nations,
1997, 33)
Aircraft in transit passage must also observe rules established by the International
Civil Aviation Organization and continually monitor the radio frequency assigned
by air traffic control authorities. There is an obligation on the states bordering
the international strait to publicize any known dangers to overflight.
The right of transit passage does not apply in one special case. This occurs
when the strait is formed by an island, belonging to the country that owns the
mainland coast, and when there is an alternative route seawards of similar con-
venience with regard to navigational and hydrographical characteristics.
Archipelagic states may draw straight baselines around the outer edge of their
outer islands if they can satisfy certain rules. The waters encompassed by such
baselines are called archipelagic waters and sections of these waters may be
more than 12 nm from islands. In the southern part of the archipelagic waters
Introduction 17
proclaimed by Fiji on 1 December 1960, some waters are more than 60 nm from
the nearest island. Article 49 of the 1982 Convention on the Law of the Sea
extends sovereignty to the archipelagic state of the airspace over archipelagic
waters. Foreign aircraft may fly through the airspace over archipelagic waters
according to the right of archipelagic sea-lanes passage. This right seems to be
very similar to transit passage through international straits.
Archipelagic states may designate air routes suitable for continuous and
expeditious passage by a series of continuous axes from the point of entry into
the archipelagic airspace to the point of exit. Such routes must include all the
normal routes used for international surface navigation, although duplication
of routes of similar convenience between the same entry and exit points is not
required. Planes flying along these routes may not deviate by more than 25 nm
from the defined axis. However, the plane may not pass closer to land than 10
per cent of the width of the strait. This means that the full deviation will only
be possible when the strait is at least 63 nm wide. If the archipelagic state does
not designate archipelagic sea-lanes, planes may fly along the routes normally
used for international navigation.
The Law of the Sea Convention also places an obligation on states to prevent,
reduce and control pollution of the marine environment from atmospheric sources.
Article 212 entitles the country to adopt laws and regulations dealing with this
subject that shall apply to its airspace and to planes registered by that country.
The vertical extent of national sovereignty was not defined in either the Chicago
or Paris Conventions. Nor is there any reference to the vertical limits of airspace
in the Law of the Sea Convention. Escalada (1979, 51-3) discusses the lack of a
precise vertical limit relating to airspace and notes that most writers on the laws
relating to airspace believe that a definite limit should be set. If a specific limit
is agreed it will presumably separate national airspace from international airspace
in the same way that high seas are separate from national maritime zones. High
seas are those seas that do not form part of national claims to exclusive economic
zones, territorial seas and internal waters (United Nations, 1997, 53).
The treaty dealing with outer space, that was opened for signature on 27 Janu-
ary 1967, stresses the international nature of outer space in Article 2.
Outer space . . . is not subject to national appropriation by claims of sov-
ereignty, by means of use or occupation or by any other means. (Martin
et alia, 1985, IV/1)
Various proposals about the vertical definition of national airspace fall into three
categories. First there are those associated with the physical properties of the
atmosphere. Such limits could be related to specified values for air pressure or
18 Chapter 1
density. Unfortunately the surfaces marking equal air pressures or densities would
resemble the surface of the ocean with very large waves. In short, the elevation
of any specified value would not be the same for various places at the same time,
nor the same for one place at different times. Such a variable limit would create
unrestricted opportunities for legal disagreement in the future.
Second, the vertical limit could be related to the performance of aircraft or
satellites. It has been suggested that the limit should be set at the level where
flight based on aerodynamic lift becomes impossible This level is known in the
literature as the von Karman Line (Escalda, 1979, 53; Martin et alia, 1985, IV/1).
Although it is agreed that this level will occur at about 80 km, it will vary above
and below that height, depending on the density of the air at any time. When an
American spy plane, piloted by Gary Power, was shot down over Russia it was
at an elevation of about 62,000 feet or nearly 19,000 metres.
At a meeting of the International Law Association in 1968 at Beunos Aires it
was agreed that the zone of outer space should begin at the perigee of the orbit
of the satellite closest to the Earth’s surface. Many satellites used for the global
positioning system operate at 20,200 km.
Third, there is the option of selecting some arbitrary height above the Earth’s
surface. This would be equivalent to the territorial sea limit of 12 nm, or the
exclusive economic zone 200 nm wide. It would be necessary to fix a height
above the level planes could reach in aerodynamic flight. If the von Karman line
has merit the arbitrary limit could be set at 90 or 100 km above sea level.
Escelada (1979, 53) may be right in asserting that the drawing of a line between
airspace and outer space is imperative, but there is little evidence that the major-
ity of countries in the world give this matter a high priority.
CONCLUSION
Following this introduction the book falls into three sections. The first section
deals systematically with frontiers, boundary evolution and boundary disputes.
The second section considers aspects of international law related to boundaries. It
includes chapters dealing with international law and territorial boundaries, maps
as evidence of international boundaries and river boundaries and international law.
The third section consists of seven regional chapters that examine the evolution
20 Chapter 1
of boundaries in the Americas, the Middle East, Africa, Asia, Europe, islands
off Southeast Asia and Antarctica.
REFERENCES
Ancel, J., 1938. Les frontières [The boundaries], Armand Colin: Paris.
Anderson, E.W., 2003. International boundaries: a geopolitical atlas, The Sta-
tionery Office: London.
Boban, L., 1993. Croatian borders 1918-1993. Skolskaknjiga and Croatian
Academy of Sciences and Arts: Zagreb.
Boggs, S.W., 1940. International boundaries: a study of boundary functions and
problems, Columbia University Press: New York.
Brown, C., B. Jones, G.B. Sperling, and T. White, 1981. Rain of death: acid
rain in western Canada, NeWest Press: Edmonton.
Brownlie, I., 1979. African boundaries: a legal and diplomatic encyclopaedia,
Hurst: London.
Curzon of Kedleston, Lord, 1907. Frontiers, The Romanes Lecture, Oxford
University Press: Oxford.
Dorion, H., 1963. La frontière Quebec-Terreneuve (The boundary between Quebec
and Newfoundland), Centre d’Études Nordiques.
Ekert, R.D., 1979. The enclosure of the oceans, Hoover Press: Stanford.
Escalada, V., 1979. Aeronautical law, Sijthoff and Noordhoff: Alphen aan den
Rijn. 1979.
Garrett, W.E., 1988. Historical atlas of the United States: Centennial edition,
National Geographic: Washington DC.
Gyorgy, A., 1944. Geopolitics: the new German science, University of California
Publications in International Relations, 3, 141-304.
Hartshorne, R., 1936. ‘Suggestions on the terminology of political boundaries’,
Annals, Association of American Geographers, 26, 56-7.
Haushofer, K., 1927. Grenzen im ihrer Geographischen und Politischen Bedeutung
(The geographical and political significance of boundaries), K. Vowinckel:
Berlin.
Hertslet, E., 1875-91. The map of Europe by treaty (5 volumes), Butterworth:
London.
———, 1909. Map of Africa by treaty, HMSO: London [Reprinted by Frank
Cass, 1967.]
Hill, N.L., 1976. Claims to territory in international law, Greenwood Press:
New York.
Hinks, A.R., 1940. Review, Geographical Journal, 96, 286-9.
Ireland, G., 1938. The possessions and conflicts of South America, Harvard
University Press: Cambridge, Massachusetts.
Introduction 21
Ireland, G., 1941. Boundaries, possessions and conflicts in Central and North
America and the Caribbean, Harvard University Press: Massachusetts.
Johnson, R.W., 1966. ‘The Canada-United States controversy over the Columbia
River’, University of Washington Law Review, 41, 676-763.
Jones. S.B., 1945. Boundary-making, a handbook for statesmen, treaty editors
and boundary commissioners, Carnegie Endowment for International Peace:
Washington DC.
Kinahan, J., 1992. By command of their Lordships: the exploration of the Namib-
ian Coast by the Royal Navy, 1795-1895, Namibia Archeological Trust:
Windhoek
Klemencic, M., 1994. ‘Territorial proposals for the settlement of the war in
Bosnia-Hercegovina’, Boundary and Territory Briefing, 1(3), 1-74.
Kroell, J., 1934. Traité de droit international public aerien [Treaty of International
Law dealing with Airspace],Les Éditions Internationales: Paris.
Lamb, A., 1966. The McMahon Line (2 volumes), Routledge and Kegan Paul:
London.
Lapradelle, P. de, 1928. La frontière: etude de droit international [The boundary:
a study of international law], Les Éditions Internationales: Paris.
Martin, P., J.D. McClean, E. de M. Martin and R.D. Margo, 1985. Shawcross
and Beaumont: air law, Volume 1, Butterworth: London..
Matte, N.M., 1981. Treatise on air-aeronautical law, Carswell: Toronto.
McEwen, A.C., 1991. ‘The establishment of the Nigeria/Benin boundary, 1889-
1989’, Geographical Journal, 157(1), 62-70.
Naval Intelligence Division, 1944. Greece, vol. I, Geographical Handbook Series,
BR 516.
Newman, D., 1995. ‘Boundaries in flux: the Green Line boundary between Israel
and the West Bank – past, present and future’, Boundary and Territory Brief-
ing, 1(7), 1-52.
Nicholson, N.L., 1954. The boundaries of Canada its Provinces and Territo-
ries, Memoir 2, Department of Mines and Technical Surveys, Geographical
Branch: Ottowa.
Oliver, R. and M. Crowder, 1981. The Cambridge Encyclopedia of Africa, Cam-
bridge University Press: Cambridge.
Ostmann, R., 1982. Acid rain: a plague upon the waters, Dillon: Minneapolis.
Parry, C., 1969-79. Consolidated treaty series, 231 vols., Oceana: New York.
Paullin, C.O., 1932. Atlas of the historical geography of the United States,
Carnegie Institution and the American Geographical Society: New York.
Pearce, J., 2000. ‘Facts on the ground: war and peace in the Horn of Africa,
May-June 2000’, Boundary and Security Bulletin, 8(2), 74-79.
Peninou, J-L., 1998. ‘The Ethiopia-Eritrean border conflict’, Boundary and
Security Bulletin, 6(2), 46-50.
Pounds, N.J.G., 1951. ‘The origin of the idea of natural frontiers in France’,
Annals, Association of American Geographers, 41, 146-57.
22 Chapter 1
———, 1954. ‘France and “les limites naturelles” from the seventeenth to
the twentieth centuries’, Annals, Association of American Geographers, 54,
51-62.
Prescott, J.R.V., 1975. Map of mainland Asia by treaty, Melbourne University
Press: Melbourne.
Prescott, J.R.V., H.J. Collier and D.F. Prescott, 1977. Frontiers of Asia and
southeast Asia, Melbourne University Press: Melbourne.
Prescott, J.R.V. and C. Schofield, 2005. The maritime political boundaries of the
world, Martinus Nijhoff: Leiden.
Ratzel, F., 1897. Politische Geographie (Political geography), Oldenbourg:
Berlin.
Rebert, P., 2001. La Gran Linea: mapping the United States-Mexico boundary,
1849-1857, University of Texas Press: Austin.
Schofield, R., 1993. Kuwait and Iraq: historical claims and territorial disputes,
Royal Institute of International Affairs: London.
———, (Editor), 1994. Territorial foundations of the Gulf States, University
College London Press: London.
Spykman, N.J. and A.A. Rollins, 1939. ‘Geographic objectives in foreign policy’,
American Political Science Review, 32, 28-50, 213-36.
St. John, R.B., 1994. ‘The boundary between Ecuador and Peru’, Boundary and
Territory Briefing, 1 (4).
———, 1994a. ‘The Bolivia-Chile-Peru dispute in the Atacama Desert’, Bound-
ary and Territorial Briefing, 1(6).
———, 1998. ‘The land boundaries of Indochina: Cambodia, Laos and Vietnam’,
Boundary and Territory Briefing, 2(6), 1-51.
The Geographer, 1971. ‘Mozambique-Southern Rhodesia boundary’, International
Boundary Study, No. 118, Department of State, Washington DC.
The Observer, 1 June 1986, London.
United Nations, 1945-. Treaty Series, United Nations: New York.
———, 1997. United Nations Convention on the Law of the Sea, Division for
Ocean Affairs and the Law of the Sea Office of Legal Affairs: New York.
United States, Department of State, Bureau of Intelligence and Research, 2002.
International Boundary Studies Library (IBSL), Version 1.0 CD.
Waldron, A., 1990. The Great Wall of China: from history to myth, Cambridge
University Press: Cambridge.
Zamoyski, A., 2004. 1812: Napoleon’s fatal march on Moscow, Harper Peren-
nial: London.
2. FRONTIERS
Political geographers use the term ‘frontier’ in two senses. It can either refer to
the political division between two countries or the division between the settled
and uninhabited parts of one country. There is no excuse for geographers who
use the terms ‘frontier’ and ‘boundary’ as synonyms.
This chapter considers those aspects of frontiers that are of interest to political
geographers and it is divided into three parts. First there is a consideration of
settlement frontiers that exist within a single country. Second political frontiers
that are used to separate neighbouring countries are examined. Third there is an
account of some historical political frontiers.
SETTLEMENT FRONTIERS
Two kinds of settlement frontiers are recognized. Primary settlement frontiers exist
where a state is taking possession of its territory for the first time. The classical
example is the westward expansion of American sovereignty through its territory
in North America. Secondary settlement frontiers are found in many countries
today and mark zones that separate settled and uninhabited regions of the state.
The two types of settlement frontiers have different characteristics.
Primary settlement frontiers are historic features, whereas secondary settlement
frontiers are found currently in many countries where an adverse physical envi-
ronment or inadequate farming and mining techniques hinder further advances in
land-use and settlement. The primary settlement frontier marked the limit of the
state’s political authority. The political authority of modern states extends beyond
secondary settlement frontiers and usually can be exerted when necessary. Any
country, such as Australia, that includes large areas of desert provides special
services that can operate in those uninhabited or sparsely populated regions. The
range of potential economic activities is generally larger in primary settlement
frontiers than along secondary frontiers. Fur-trapping, timber-felling, semi-sub-
sistence cultivation, grazing, mining, and manufacturing and service industries
were either found at some points on the American frontier or developed as it
advanced. On the other hand the advancement of secondary settlement frontiers
is likely to be accompanied by the extension of irrigated farming, as in Mali,
24 Chapter 2
Much has been written about the primary settlement frontier. American historians
are mainly responsible for the thorough documentation of the United States’
primary frontier. Turner was able to use early studies of the primary settlement
frontier to establish his hypothesis that ‘. . . the existence of an area of free land,
its continuous recession and the advance of American settlement westward
explains American development’ (Turner, 1953). Historians rather than historical
geographers have provided the majority of studies on this subject. Whittlesey
(1956) writing on the expansion and consolidation of the United States, makes
only passing reference to the frontier and no reference to the detailed histori-
cal studies. Some geographers have contributed to frontier studies beyond the
simple mapping of frontier phenomena. Such mapping has been done for the
American and Canadian frontiers, by Paullin (1932), Adams (1943), Kerr (1961)
and Garrett (1988).
Frontiers 25
The position of the frontier, that represents the zonal limits of political author-
ity, and its width, are of prime interest to geographers. In order to determine
the frontier’s extent some criteria must be developed to distinguish it from
non-frontier areas. A simple estimate of population density is unsatisfactory,
and a better measure is likely to be found in the degree of political and eco-
nomic organization. This is a task calling for training in historical and political
geography. In an examination of South Africa’s primary settlement frontiers,
Christopher (1982) drew attention to the inadequate statistical information for
the 19th century. He proposed a careful examination of population distribution,
land alienation, cultivation and numbers of stock to locate the positions of the
frontiers and their characteristics.
Information about the position of the frontier at any time will give a clue
to the factors that have influenced its rate of advance. This information will
certainly allow the identification of the personalities and policies of the gov-
ernment of the day. The advance of a primary frontier will frequently depend
upon a combination of factors. They can be broadly divided into the forces of
attraction, attributable to the environment beyond the frontier, and the social,
economic and political forces of pressure encountered in the frontier itself. The
role of unusually favourable soil groups, such as those found in the Blue Grass
country of Kentucky and the cotton lands of the Gulf Plains, promoted the rapid
advance of the frontier in the early decades of the 19th century. In South Africa
the discoveries of precious mineral deposits in the Transvaal caused spectacular
advances of the frontier.
Pressures within the frontier hinterland take many forms. Turner (1953) and
Billington (1960) have shown how frontiersmen were seeking to avoid high land
prices, heavy taxation and political and religious disabilities imposed either by
the first, well-established settlers or by the governments of the country of origin.
Further, the experience gained on one section of a frontier, in respect of land
legislation, mining laws and Indian treaties was applied at subsequent frontiers,
allowing speedier settlement of problems. Periods of standstill or retreat along
the frontier resulted generally from the unfavourable nature of the environment
or the inadequacy of techniques for using it, the armed resistance of indigenous
groups and the preoccupation of governments with more urgent considerations
(Figure 2.1).
In these successive frontiers we find natural boundary lines which have
served to mark and affect the characteristics of the frontiers, namely: the
‘fall-line’; the Allegheny Mountains; the Mississippi; the Missouri where
its direction approximates north-south; the line of the arid plains approxi-
mately the thirty-ninth meridian; and the Rocky Mountains. The fall-line
marked the frontier of the seventeenth century; and the Alleghenies that of
the eighteenth; the Mississippi that of the first quarter of the nineteenth; the
Missouri that of the middle of this century [19th] (omitting the Californian
movement); and the belt of the Rocky Mountains and arid tract, the present
frontier. Each was won by a series of Indian wars. (Turner, 1953, 9)
26 Chapter 2
The maps of the battles with Indians show that the fiercest resistance by indi-
genes often coincided with the most difficult terrain, that offered superior stra-
tegic opportunities for defence (Paullin, 1932). It was in the scrub country of
Queensland that Aborigines offered the sternest resistance to the extension of
pastoral activities. In Kamerun, during the early years of the twentieth century,
the Germans faced their most severe problems in pacifying the Chamba and
pagan groups, who were located in the heavily dissected borderland with Nigeria.
The stagnation of the American frontier, during the twenty years before 1795,
resulted from the preoccupation of the colonies with securing independence,
establishing a federal constitution and defeating Indians in areas already settled.
Porter (1979) noted that the delayed settlement of the western areas of Maryland
can only be understood after examination of the contemporary physical, financial
and political conditions.
In some cases attempts were made to halt the frontier’s advance by governments
or interested trading organizations. Some of the earliest coastal states in North
America tried to enforce such restrictions in order to retain political power and
avoid a further loss of population. In the hinterlands of New York and Pennsyl-
vania, the Iroquois Confederation blocked for a century, the route through the
Catskill and Berkshire Ranges, by the Mohawk and Hudson valleys. This was
done so that Indians who supplied the fur trade should not be driven away (Bil-
lington, 1960). Konrad (1981) has described how the desire of the Iroquois to
continue their profitable trade in furs, especially beaver, resulted in an advance
of their frontier from the southern shore to the northern shore of Lake Ontario. In
the second half of the 17th century this area was a devastated no-man’s-land as
far as the British and French authorities were concerned. However, the Iroquois
established their settlements to provide bases, from which winter expeditions
could be conducted north and south of Lake Ontario. They were used as staging
posts for fur brigades travelling south to trade with the English.
Turner maintains that ‘. . . each frontier leaves its trace behind it, and when it
becomes a settled area the region still partakes of the frontier characteristics’
(Turner, 1953, 4). This suggests a fruitful field for geographical research. Is it
possible to attribute any elements of the cultural landscape to the period when
the area was a primary frontier? There is probably a connection between present
property boundaries and the original policies of land allocation and appropriation.
It seems unlikely that the present economy will reveal many features that can
be traced to frontier times, since the earliest economic activities of hunting and
grazing will survive only if the land was unsuitable for cultivation and lacked the
resources on which could be built towns with secondary and service industries.
It has been noted by Clark (1959) that when a period of standstill allowed an
accumulation of population, as in Georgia and Tennessee, eventual advance of
the frontier was more orderly and complete. Rapid advance without resistance
resulted in scattered and discontinuous patterns. It would be interesting to discover
whether settlement analysis reflects this process of development.
Frontiers 27
Secondary settlement frontiers are found in all but the smallest countries. They
include areas of unfavourable environment, such as tropical or temperate desert,
heavily dissected uplands, thick tropical rain forest, and areas that require the use
of advanced and often expensive techniques if they are to be used for purposes
other than mineral extraction. These are often the areas by-passed by the primary
settlement frontier, when the population is concerned with rapid exploitation and
profit. These inferior regions will be assessed later if circumstances require it and
if new techniques or discoveries make it possible to revalue the environment. Burt
has recorded the following interesting observation about the Canadian settlement
frontier in the 19th century.
The expansion of Canadian settlement ran up against the rocky pre-Cam-
brian Shield, with the result that the Canadian frontier movement crossed
the [American] border, where it became merged in the greater movement
to the northern Middle Western States. (Burt, 1957, 71)
Only when the availability of land in the American West was reduced did the
secondary settlement frontier cross the boundary again, promoting the develop-
ment of western Canada. Straight (1978) shows that there was also northern
movement by American settlers across the Canadian boundary in New England.
This movement started after the 17th century and continued without much appar-
ent interference.
28 Chapter 2
In many European states only short settlement frontiers surround small areas
of unfavourable environment. In Australia a long secondary settlement frontier
surrounds the central desert. It occupies about 2 million square kilometres of
dunes formed by the anti-clockwise pattern of winds (The Grolier Society,
1983, volume 3, 202). The longitudinal dunes are about 500 metres apart and
15 metres high in the Great Victoria, Simpson and Great Sandy Deserts. The
western Gibson Desert is composed of laterite high plains with some mesas.
Adjacent to both sides of the boundary between Western Australia to the west
and South Australia and the Northern Territory to the east most of these desert
areas are organized into freehold or leasehold Aboriginal Trusts. The remaining
desert regions in Western Australia are mainly Vacant Crown Land and some
small zones set aside for Nature Conservation Reserves (Australian Surveying
and Land Information Group, 1993). The total population of this large region is
small and tends to be concentrated in tiny settlements. Forty such settlements
were identified around the perimeter of this desert region and a few were located
closer to the centre. There were 23 in Western Australia and 17 in the Northern
Territory and South Australia. The populations at these locations varied from
10 at Yulara in the Northern territory and 29 at Edjudina in Western Australia.
However, it is noteworthy that although the area is lightly populated the entire
area is allocated to a land classification, such as freehold or leasehold, vacant
crown land, Aboriginal trusts, defence land, nature conservation reserve and
Aboriginal national park.
Examples of the role played by improved techniques can be found in many
countries. The waterless areas of the Kalahari sandveld in South Africa were
not settled until after 1903, when drilling equipment allowed underground water
reserves to the tapped. The Mallee scrub region of interior South Australia, Vic-
toria and New South Wales was covered with a hardy eucalypt formed of many
branches rising from a mass of hard wood at, or just below, the surface. It was
very difficult to clear, the vegetation was unpalatable to stock and sheep could not
be mustered because the vegetation was dense and watering points were scarce.
In 1877 the stump-jump plough was developed. It had two shares that moved
separately and on contacting a stone or root they rose over the obstacle and then
re-entered the ground. Much land was cleared for wheat farming but the dryness
of the climate allowed strong winds to remove the light top-soil. Dunes were
formed and sometimes fences and outbuildings were covered. Sheep rearing was
tried in the 1930s but there was a shortage of important trace elements in the
ground and soil erosion continued. Much of the area has now been established
as nature reserves and national parks.
In the middle of the 19th century Russian emigration was threatening Japanese
sovereignty over Hokkaido. The Japanese response was to encourage Japanese
migration to the island. The government identified two groups of immigrants:
those who travelled independently and those who travelled with the aid of a
government subsidy. Independent farmers received implements, seed and 10
Frontiers 29
yen for every quarter acre cleared. Subsidized farmers received a rice ration for
three years in addition to seed and implements. Their bonus for clearing land was
2 yen per quarter acre. Independent merchants and artisans received a gift of
150 yen for three years, which eventually had to be repaid (Harrison, 1953).
The discovery of new mineral deposits in remote areas, or changes in world
prices that make known deposits profitable, encourages advances in the secondary
settlement frontier in tropical and arctic deserts. Such advances occur at specific
locations rather than on a broad front. Generally this advance will last only as
long as the mine is profitable. A major uranium deposit was found at Arlit in
Niger, north of Agades in the 1970s on the southern edge of the Sahara Desert.
In 1999 it contained over 5 per cent of the world’s reserves of uranium (Times
Books, 1999, Plate 44-5). The same map shows a sprinkling of manganese and
nickel mines along the coastal rim of the Arctic Ocean, iron deposits in Xizang
Zizhiqu in central southern China and gold, copper and diamond mines in central
and north Australia.
Stone (1979) provided a very useful map showing the distribution of the world’s
settlement frontiers that cover an area of at least 7,000 square miles. He classified
the land into three categories: continuous settlement, discontinuous settlement
and uninhabited. He then divided the areas of discontinuous settlement into the
inner fringe zone, the middle fringe zone and the outer fringe zone. The inner
and middle fringe zones were considered to constitute settlement frontiers that
might have had a reasonable potential for new settlement. Stone noted that there
might be some areas smaller than 7,000 square miles that may have a reason-
able potential for new settlement. He recommended that they are identified on
maps at a scale larger than 1: 125,000 that he used to create the world map. The
frontiers he identified were described according to population densities and the
recent changes in numbers of people.
Stone’s work builds on the global view of secondary settlements set out
by Bowman in 1931. He called the second settlement frontiers ‘the pioneer
fringe’.
Without wading too deeply into the sea of technology we may define the
remaining pioneer areas of the world as regions of potential settlement in
which man may have a reasonable safe and prosperous life; but regions
in most of which he is required to make certain special adaptations. What
those adaptations are the settler may discover by painful experiment, as in
the past, or by the less painful experience if government and science steps
in to help him. The rainfall of undeveloped pioneer lands is on the whole
not so favourable as in most of the settled areas of the earth, but it is suf-
ficient if the settler uses it rightly. The temperature may be too hot or too
cold to suit him, but it is tolerable. (Bowman, 1931, 51)
The world maps produced by Bowman and Stone are not directly comparable.
However, the pioneer fringes, identified by Bowman in eastern Asia, south of
30 Chapter 2
parallel 60° north, in central Asia east of the Aral Sea, in Africa north of par-
allel 15° north and in central Africa between a line joining the mouths of the
Benguela and the Rovuma Rivers and the northern boundary of South Africa
have been reduced in extent.
POLITICAL FRONTIERS
run the British Royal Navy blockade of Ouidah was the Ciceron in 1863 (Naval
Intelligence Division, 1944, 44-5).
International boundaries have replaced international frontiers throughout the
world. This means that research into political frontiers must have a strong his-
torical and anthropological basis. Political frontiers generally experienced less
intensive economic development than the territories they separated. There were
three possible explanations for this situation. They were the unfavourable nature
of the environment, the satisfaction of all needs within the territory bounded
by the political frontier, or the deliberate neglect of the frontier to enhance its
divisive character. Deserts, mountain ranges, rivers, marshes and woodlands have
all formed frontiers at some time. It follows from this that political frontiers
were usually less densely populated than the flanking states and in some cases
the residents of the frontier experienced a lower standard of living. Tacitus, the
eminent Roman historian of the first century, described the debased condition
of the Slavic Venedi who occupied the wooded and mountainous area between
the Peucini and Fenni. A more recent example was cited by Tilho (Ministère
des Colonies, 1910). He referred to the wretched Beddé pagans, who survived
in the swampy areas between the kingdoms of Bornu and Sokoto only if they
could avoid enslavement by raiding parties from both states.
Yesner described the frontier that separated the Aleut and Inuit peoples in
Alaska, although he does not use the term frontier.
The boundary zone between Eskimo and Aleut populations on the Alaska
peninsula was not a resource-rich buffer zone, but a (relatively) resource-
poor zone that acted as an isolating barrier between two relatively richer
resource zones – one nearly exclusively maritime and one providing a
mixed coastal, riverine, tundra suite of resources . . . Once established, the
boundary between these two regions remained relatively stable over a long
period until the Thule people – either because of environmental change,
increased technological efficiency, or population pressure – were finally
able to penetrate the boundary. (Yesner, 1985, 84)
There is at least one case where the populations living close to the frontier
acquired military and economic characteristics that enabled them to become
paramount throughout the country. The Chou Dynasty of China was paramount
from 1200-770 BC.
The conquering group of the Chous came from the west, but probably
not further west than the region of the Wei Ho valley at the head of the
Kansu corridor. This has been again and again a crucible region of vast
significance to China. It lies within the belt of fertile loess and of normally
sufficient rainfall and so was in touch with the characteristic development
of Chinese agriculture further to the east. But it is closely bordered by the
steppe, and upon it impinged not only the influences coming by way of the
Turkestan oases from Western Asia, but also those of the virile, pastoral
32 Chapter 2
nomads of the grasslands. Here there tended to be formed under able rulers
a strong composite society which has given many masters to China. The
Chou conquest is the first great example of it, and it was as guardians of the
marches, ‘Chiefs of the West’, that the Dukes of Chou acquired the power
to overthrow the Shang dynasty which had become effete and discredited.
(Naval Intelligence Division, 1944a, 305)
Dictionaries do not give one common meaning for the term ‘march’. The mean-
ings given vary from ’borderland’ or ‘frontier’ to a territory, organized on a
semi-permanent military system, within the frontier to defend the state. It is
the latter sense that is used in this text. By 255 BC the central authority of the
Chou dynasty had moved eastwards, to the basin of the Chang Jiang [River]
below the gorges, and it was replaced by the Ch’in dynasty. The Ch’in dynasty
also originated in the western marches. Frequent conflict with mounted nomads
had encouraged the development of a powerful army that included cavalry units
(Naval Intelligence Division, 1944a, 312). Further some nomadic groups had been
incorporated into the Ch’in organization. The Duke of Ch’in became the First
Emperor. After the Chin Dynasty collapsed it was eventually replaced by the
Han dynasty. It lasted from 206 BC to 214 AD and was initially cautious. After
a period of consolidation attention was turned to the northwest frontier. Attention
was focussed on the Gansu corridor that extended from Wuwei, via a number
of rich oases, to Yumen [Jade Gate] where the trade routes diverge to the north
and south of the arid Tarim Basin. The wall was extended along the northern
edge of this panhandle and the constriction of the central area on modern Gansu
Province is the remnant of the Han advance. The policy established by Emperor
Wu Ti was that the wall represented the inner line of defence delimiting China
Proper. However, imperial authority was to be extended as far as possible to
the west and northeast (Naval Intelligence Division, 1944a, 322). One means of
controlling areas beyond Yumen was through the agency of native princes. That
system allowed Han control over the Tarim Basin as far as Kashgar until 23 AD
(Barraclough, 1978, 80-1; Naval Intelligence Division, 1944a, 324-5).
Where there was the threat of invasion or trespass, political frontiers were
selected for their defensive advantages and this point was thoroughly investigated
by Curzon and Holdich. Curzon noted that deserts formed the best defensive
frontiers. However, extensive deserts, such as the Sahara, were often inhabited
by mobile warlike groups, such as the Tuareg, that plagued the surrounding
agricultural communities. Davies made this precise point when he wrote of the
northwest frontier of British India.
So long as hungry tribesmen inhabit barren and almost waterless hills, which
command open and fertile plains, so long will they resort to plundering
incursions in order to obtain the necessaries of life. (Davies, 1932, 179)
Linear mountain ranges and wide rivers provided the advantage that the defend-
ing forces could concentrate their strength at passes and crossing points. The
Frontiers 33
Many states tried to mark their frontiers in some way. The famous Roman and
Chinese walls are the best examples. The Great Wall of China served not only
to exclude nomadic barbarians, but also to restrict the number of Chinese who
adopted a modified agricultural system, that made them more difficult to control
from the Chinese capital (Lattimore, 1940). The journey from Beijing to the
restored portion of the Great Wall, visited by tourists, passes a number of earlier,
local walls, They show the same disregard for steep gradients exhibited by the
famous national equivalent. Unlike China’s Great Wall the walls built by the
Romans did not mark major environmental divides. They were built in order to
exclude barbarians. Where clear linear physical features were not available as good
defensive lines the Romans built walls such as Hadrian’s Wall linking the Solway
Firth to the Tyne valley. Two others were built across the re-entrant formed by
the upper courses of the Rhine and the Danube and east of the Drava-Danube
confluence. The barbarians north of the Roman wall also built earthworks to
delimit the edge of their territory. It is recorded that the Angrivarii constructed a
broad earthwork to mark their frontier with the Cherusci. Baradez (1949) makes
the important point that the walls and ditches formed the first or last line of a
system of defence in depth. The Roman walls were reinforced by establishing
farmers on the land behind the walls, in a zone called agri limitanei. The men
of these families were expected to assist in the defence of the wall when it was
attacked. This was an ancient predecessor to some Israeli Kibbutzim adjacent
to Arab territory in the past fifty years. However, as an exception to this rule,
Collingwood (1923) noted that the vallum [a palisade of stakes] behind Hadrian’s
Wall marked the limits of Rome’s civil government.
The counterparts of the Chinese and Roman walls could be found in parts of
Africa in the 19th century.
The kingdom was surrounded, where there were no natural defences by deep
and wide ditches defended by tree trunk palisades and crossed at intervals by
narrow bridges. The northern frontier was formed by the Gojeb River, called
Godafa by the Kafa. Beiber gives the dimensions of the ditches as 6 metres
in width and 3 metres in depth; he describes the gates, kello, as consisting of
34 Chapter 2
circular fenced enclosures entered by drop gates. Outside the line of fences
was a strip of unoccupied land like the moga of the Galla states. At points
where Galla attacks were expected, the gates were additionally defended by
a high rampart and several lines of entrenchment, a form of defence much
admired by the neighbours of the Kafa. (Huntingford, 1955, 116)
The moga or uncultivated strip, of the Galla states in the Horn of Africa was
inhabited only by fierce brigands, who were encouraged by the Galla rulers, to
attack common enemies and recapture escaped slaves.
Fischer maintained that ‘. . . a rather extensive literature deals with the develop-
ment of boundary lines out of such [frontier] zones or related features’ (Weigert
et alia, 1957, 79). However, the works cited do not treat this aspect of boundaries
in detail. The general impression is that as states separated by frontiers extend
their territory, the unclaimed land diminishes. Eventually property disputes arise
and attempts to solve such problems result in a delimited boundary. No doubt
this series of events has occurred in some cases, but there are variations on this
theme.
Frontiers normally diminish in width and frontiers of separation are replaced
by frontiers of contact. There have been exceptions when frontiers increased
in width. Such a development occurred in Hashtadan on the borderlands of
Afghanistan and Persia, southwest of the great northward bend of the Hari Rud
at Koshan. This area was investigated by General Maclean in 1888-9, prior to
making a boundary award as requested by the Afghan and Persian governments.
He found ruins of villages, faint field patterns and sections of underground canals
that suggested the area had once been occupied and prosperous. He discovered
that an epidemic throat disease had decimated the valley’s population in 1788
and that subsequent attacks by Uzbek, Hazarah and Turkoman raiders were
responsible for the subsequent devastation of the valley.
But from the appearance of the ruins and abandoned fields it is quite evident
that the valley has been deserted for some generations. . . . Upon the whole,
looking to the nature of my present information, it seems to me that neither
Persians nor Afghans can produce proofs of recent possession in support
of their respective claims, neither having felt inclined to stand the brunt
of collisions, in such an exposed locality, with the Turkomans. (Prescott,
1975, 154)
In the more usual process frontiers diminished in width either by the incorpora-
tion of parts of the frontier by one or both flanking states or by the creation of
subsidiary political units within the frontier.
Frontiers 35
Annexation of parts of the frontier might have taken place because of rising
land hunger within the state or through the development of new techniques that
enabled the frontiers resources to be revalued. Fifer (1966) provided an interest-
ing account of the extension of political control by Bolivia and Brazil over the
‘. . . empty, unknown, formerly negative frontier of separation . . .’ that lay between
them. She clearly identified the desire for new lands on which high quality rubber
could be produced during the second half of the 19th century. This process was
accelerated during the 1880s, when a severe drought throughout Brazil’s eastern
province of Ceara encouraged a mass exodus of labourers in search of work.
. . . not until the rubber boom involved the Amazon headwater region, thus
probing beyond Brazil’s undisputed territorial claims, did old negative
frontiers of separation suddenly assume both economic and hence political
significance. (Fifer, 1966, 361)
If the frontier existed because of the internal weaknesses of the neighbouring
states or their preoccupation with other external threats, the removal of the
threats or the resolution of the weaknesses might allow appropriation of parts of
the frontier. Alternatively the frontier might have been invaded and annexed for
strategic reasons. For example, after the Roman successes in Gaul, the eastern
flank of this advance was protected by the annexation of Noricum, Pannonia,
Meosia and Dacia in the Danube Basin. This advance also removed the scene of
conflict from the Mediterranean centres of the Empire (East, 1962). Annexation
for one reason sometimes conferred unexpected benefits. The Romans invaded
the area between the River Rhine, the River Main and the Taunus Range to
stop raids by the Chatti. Later they discovered hot springs and iron and silver
deposits in the region.
The subsidiary territorial organizations that were created within frontiers included
marches, buffer states and spheres of interest or influence. As noted earlier a
march is a territory, organized on a semi-permanent military system within the
frontier, to defend the state. An illustration of the creation of marches, also called
marks, was provided by the policies of Charlemagne and Otto.
From these Marks, intended to safeguard the Frontiers of the Empire from
Slavonic or alien contact, and ruled by Markgrafs or Markgraves, sprang
nearly all the kingdoms and states which afterwards obtained a national
independent existence, until they became either the seats of empires them-
selves, as in the case of the Mark of Brandenburg, or autonomous members
of the German Federation. (Curzon, 1907, 27)
36 Chapter 2
The German Kingdom was protected from the Slavs and Magyars by a series
of marches stretching from the Baltic Sea to the Adriatic Sea (Figure 2.2). The
Markgraves had the responsibility of defending the kingdom and extending the
territory subject to it. The extension of territory resulted in the North and East
Marks becoming the cores of major state. North Mark, founded in 928 by Henry
the Fowler, later became known as Altmark, and it acquired Mittelmark, Vormark,
Ubermark and eventually Neumark east of the Oder River. This enlarged area
was created the Mark of Brandenburg in 1157 and subsequently provided the
basis for the unification of Germany. East Mark, after being almost submerged
beneath Magyar raids, was recreated by Otto the Great in 955. The additions
of the Marks of Styria and Carinthia by 1282 created the core from which the
Austro-Hungarian Empire grew.
One of the most enduring marches in Europe was created in 1578. After the
defeat of the Magyar army by Turkey at the battle of Mohacz in 1526 the Turks
controlled most of the Hungarian plain (Naval Intelligence Division, 1944b, 17).
As a defence against the Turks Ferdinand of Austria created a march called the
Militärgrenze [Military Frontier] in 1578.
This was a land of forts, watchtowers and beacons, and its inhabitants, the
granicari or frontiersmen, held their land on a special tenure in return for
military service. (Naval Intelligence Division, 1944b, 18)
In 1699 Austria secured most of Croatia-Slovenia from Turkey and the Military
Frontier was augmented so that it consisted of the three ‘generalates’ of Carl-
stadt, Varadzin and Petrinja. A generalate is the period of office of a military
general.
The Croatian Diet (and that of Hungary too) greatly resented these [new]
limitations of territorial sovereignty imposed by Austria, and repeatedly
demanded the incorporation of the ‘Frontier’ within the civilian administra-
tion of Croatia. But the frontiersmen were against any change in their status,
and the imperial government saw in them a useful counterpoise against
the unruly nobility of the south. The regime of the ‘frontier’, despite Croat
opposition and the decay of Turkish power, was not finally abolished until
1873-81. The tradition of the frontier remained long after 1881, and a high
percentage of officers in the Austro-Hungarian army continued to be drawn
from the old frontier regiments. (Naval Intelligence Division, 1944b, 18)
Buffer states have been constructed within frontiers when two strong neighbours
decided to reduce the dangers of conflict. In 1872 and 1895 Britain and Russia
fashioned the Wakhan panhandle, that was a sliver of Afghan territory between
Russian territory west of the Caspian Sea and northwest British India. In 1873 a
line between Russian and Afghan territory was agreed as far as the eastern end of
Lake Zorkul and this marked the northern limit of the panhandle (Prescott, 1975,
104-5). In 1893 Sir Mortimer Durand visited Kabul to negotiate the boundary
Frontiers 37
between Afghanistan and British India. The northernmost sector of this line formed
the southern boundary of the panhandle. However, Durand had the delicate task
of explaining to the Emir that the northern limit of the panhandle, with Russia,
would exclude the territories of Roshan and Shignan over which the Emir has
exercised intermittent authority (Prescott, 1975, 137). Fortunately the Emir agreed.
This panhandle is 280 km in length and 55 km at its widest point.
Britain’s prime boundary strategy in northern India was to maintain a series of
small weak states between China and British India. Nepal, Sikkim and Bhutan
were allowed to exist in the Himalayas. In Southeast Asia Britain tried to per-
suade China or Thailand to interpose their territory between Britain’s territory
of Burma and France’s territory of Indo-China across the gap of 120 km that
separated northern Thailand from southern China. British policy was explained
as follows.
We could not have a coterminous frontier [boundary] with France in Bur-
mah. That would involve vast expenditure on both sides and lines of armed
posts garrisoned by European troops. . . . We had proposed a buffer state
in the interests of both countries, for it was evident that if our boundaries
were contiguous, any fussy, or ill-conditioned frontier officer, whether Eng-
lish or French, would have it in his power to magnify every petty incident
into a grave international question, which would be transferred to Europe,
and thus grow into a cause of exacerbation between the two Governments;
whereas if a country like China were in occupation of the intermediate
territory, neither England nor France would ever hear a word of any little
troubles of the sort, which would be settled to the satisfaction of everyone
concerned according to the customs of the country. (British and Foreign
State Papers 1894-5, 272, 379)
In fact France advanced westwards too fast for the British policy to have a chance
and in 1896 the boundary with France was agreed to coincide with the talweg
of the Mekong River.
Some colonial powers employed neutral zones to serve the same function as
buffer states. Britain and Germany separated their spheres of influence in the
Gold Coast and Togoland by a neutral zone. It lay north of the confluence of
the Dakka and Volta Rivers. At various times there were proposals for neutral
zones in southern Africa between British and German territories and between
British and Portuguese territories. In 1965 and 1973 Kuwait and Saudi Arabia
divided their former joint neutral zone on land. The resources of the former
neutral zone and the marine areas attached to that zone are owned in common.
In 1963 an agreement dealing with the offshore continental margin fixed the
southern limit separating the offshore neutral zone and Saudi Arabia’s exclusive
continental margin. However, negotiations involving the maritime boundaries
with Iran could not commence until the northern limit of the Neutral Zone with
Kuwait was settled. This northern limit was settled by an agreement signed in
2000 and ratified in 2001.
38 Chapter 2
The 2 July 2000 Agreement resolves these issues and opens the way (1) for
negotiations between Saudi Arabia and Kuwait on the one hand, and with
Iran, on the other hand, to establish the maritime boundary between the
Offshore Neutral Zone and Iran’s maritime jurisdiction and (2) for negotia-
tions between Kuwai and Iran on their maritime boundary. (Colson, 2002,
2826)
Neutral zones were often convenient solutions to difficult territorial questions,
but they were really only short-term remedies. Usually their continued existence
proved inconvenient and a greater source of friction than a detailed debate about
their division between the neighbouring states.
The concepts of spheres of interest and influence flowered during the 19th
century when the major European powers were establishing actual and potential
claims to parts of Asia and Africa. At no time were the responsibilities and
entitlements assumed under such claims defined precisely. Both concepts were
means of reserving territory from the political interference of another state. It has
been assumed that a sphere of interest is less significant than a sphere of influ-
ence. Holdich (1916, 96) suggested that a sphere of interest became a sphere of
influence when there was the threat of competition by another state. In at least
one case a sphere of influence was advanced to become a sphere of action. Some
spheres were defined in territorial agreements as the following examples show.
The Anglo-French agreement agreed in August 1890 was the forerunner of the
present boundary between Niger and Nigeria.
The Government of Her Britannic Majesty recognises the sphere of influ-
ence of France, to the south of her Mediterranean possessions up to a line
drawn from Say on the Niger to Barruwa on Lake Tchad, drawn in such a
away as to comprise in the sphere of action of the Niger Company all that
properly belongs to the Kingdom of Sokoto. (Hertslet, 1909, 730)
This boundary stretched for 1,120 km and the country through which it passed
was largely unknown to the two parties. Indeed Lord Salisbury, the British Prime
Minister, admitted as much when he commented on the treaty.
We have been engaged in drawing lines upon maps where no white man’s
foot has ever trod; we have been giving away mountains and rivers and lakes
to each other, only hindered by the small impediment that we never knew
exactly where the mountains, rivers and lakes were. (Roberts, 1999, 529)
These comments were made in a speech at the Mansion House soon after the
agreement was signed. When he was criticised in the House of Lords for giving
too much territory to France he made the following justification.
. . . consider its quality as well as its quantity. This is what agriculturalists
would call very ‘light’ land. (Roberts, 1999, 530)
Frontiers 39
The implication that the Royal Niger Company had got the best of the deal
offended the French Ambassador, who was in the gallery. He wrote to show his
annoyance as follows.
[You have] diminished very much the good feeling created by the conclusion
of the arrangement. No doubt the Sahara is not a garden, and contains, as
you say much ‘light land’, but your public reminder of the fact now, perhaps
you will allow me to say, was hardly necessary. You might well have left us
to find it out. (Roberts, 1999, 530)
The Anglo-Portuguese agreement of 1891 was the forerunner of the present
boundary between Angola and Zambia
It is agreed that the line of division separating the British from the Portuguese
sphere of influence in Central Africa shall follow the centre of the channel
of the upper Zambezi, starting from the Katima Rapids up to the point where
it reaches the territory of the Barotse Kingdom. (Brownlie, 1979, 1041)
The degree of interference with the indigenous organizations in the spheres of
interest or influence varied in almost every case. At one end of the scale the
imperial power assumed no responsibility, but claimed the exclusive right of
its nationals to trade in the area. At the other end of the scale there was a high
degree of political control more appropriate to the nature of a protectorate.
It was noted earlier that some geographical research involves identifying elements
or characteristics of the landscape derived from frontier origins. Such studies are
also a proper facet of political frontier studies. Three examples illustrate these
relationships. Cornish (1936) traced the evolution of the language borderlands
of Europe, such as Flanders, Lorraine, Friuli, Istria and Macedonia. In each
case the language frontier coincided with an earlier political frontier between
Christendom and heathen states that had been static for some time. The growth
of polyglot language regions occurred only where the frontier did not coincide
with a divisive physical feature. Cornish called such regions ‘link-lands’, to
emphasise their position between larger state areas. The heathen languages were
eventually reduced to writing through contact with Christianity and their tradi-
tions were preserved. Cornish observes that only during the 19th century, with
improved means of mass communication between the larger state areas and the
link-lands, did the bonds of language become more important than the regional
ties of the link-lands.
Wilkinson (1955) examined the history of Jugoslav Kosmet and explained how
it formed the frontier, at various times, between the eastern and western parts of
the Roman empire, the Bulgar and Byzantine empires, Christianity and Islam and
40 Chapter 2
Yugoslavia and Albania. This situation resulted in some neglect of the economic
resources of the area and hindered its integrated development.
In a number of cases when frontiers were replaced by boundaries, enclaves
of one state or both states were created. An enclave is a territorial fragment of
one country that is totally surrounded by the territory of another country. Whyte
(2002, 2004) has made a definitive studies of political enclaves. He identified 259
enclaves around the world and established that 80 per cent lie astride Bangladesh’s
northern boundary with India and 12 per cent are Belgian enclaves located in and
around the Dutch village of Baarle (Whyte, 2004, 1-2). His meticulous study of
the enclaves on both sides of the Bangladesh-India boundary traces their evolu-
tion from 1500 to 1950; 450 years during which the frontiers were replaced by
boundaries. He establishes that many enclaves had their evolution before the end
of the 18th century. Most of the enclaves survived the hurried partition of the
sub-continent by Sir Cyril Radcliffe, declared in August 1947, that contributed
to the slaughter of Hindus and Moslems who found themselves on the wrong
side of the lines.
Whyte’s account of the Baarle enclaves is based on both archival research and
thorough fieldwork. In remarkable detail he traces the evolution of the enclaves
from 992 to 1995 as frontiers yielded to boundaries. The enclaves survived the
Treaty of Munster in 1648 between the Dutch and Spanish states and Whyte
notes that their survival was unsurprising because ‘. . . enclaves were still com-
mon at many political levels in north-western Europe’ (Whyte, 2004, 13). These
studies serve as models for any scholars who decide to study enclaves around
the world.
The continent of Africa was colonized by European states after parts of Austra-
lia, southern Asia, and North and South America had been both colonized and
organized. Some frontiers between African states persisted until the last decade
of the 19th century and a major political frontier between colonists and indigenes
in southern Africa lasted from 1652 to 1858.
This analysis and description is based on fieldwork from 1956 to 1960 and a
review of the extensive literature on the history of West Africa in general and
Nigeria in particular. The works by Barth (1857), Staudinger (1889) and Hogben
(1930) were the most useful.
Frontiers 41
The largest state in West Africa in the middle of the 19th century was the
Sokoto-Gando Empire, founded by the Fulani Jihad 50 years earlier (Figure 2.3).
It stretched from Libtako in the west to Adamawa in the east and from Katsina
in the north to the latitude of Ilorin in the south. This territory was not subjected
to uniform political authority. It was organized into provinces, each having a
degree of independence that varied directly with their distance from Sokoto or
Gando. In the provinces of Zaria, Bida, Kontagora, Nassarawa, Kano and Muri,
the Fulani subjugated the indigenous tribes. In other areas, such as Bauchi
and western Adamawa, enclaves of pagan groups retained their independence
on defended hilltops. Finally, in Libtako and eastern Adamawa only the main
towns on the principal trade routes were under Fulani authority. These Fulani
towns were exclaves in uncontrolled pagan areas and they might be described
as march-towns.
North and west of Sokoto lay the Habe states led by Hausa chiefs who con-
tinued the struggle against the Fulani from their new capitals. The westernmost
Hausa state was Kebbi that had a narrow frontier of contact with Sokoto and
Gando. Many raids were carried out and some battles waged in this frontier. The
other two Hausa states, called Gober and Maradi, were separated from Sokoto by
a frontier of separation, formed from a devastated zone. The towns of Jankuki,
Dankama and Madawa had been destroyed by Fulani attacks. This depopulated
zone became more thickly wooded than the surrounding areas and served as a
refuge for robbers. On the northern fringes of this frontier Maradi established
the marches of Gazawa and Tessawa.
Northeast and east of Sokoto lay the Bornu Empire and its vassal states that
included Zinder. The reduced power of Bornu after the Fulani conquest increased
the autonomy experienced by Bornu’s traditional northern tributary states, includ-
ing Zinder. Between Kano Province of the Sokoto-Gando Empire and Zinder
there was a deserted frontier of separation. It resulted from the weakness of both
states and there were only occasional raids across this frontier.
The frontier of separation between Bornu and Sokoto can be divided into three
parts. North of the River Gana lived the Bedde pagans protected by a forested,
swampy environment. Raiding parties from both Sokoto and Bornu captured
slaves in this area. Between the River Gana and the Mandara Mountains, the
forested frontier of separation was defended, on the Bornu side, by a number of
quasi-independent marches that had a long history of resistance to the Fulani.
The Mandara Mountains formed the third section of the frontier. This area was
occupied by Marghi pagans, against whom the Fulani exerted intermittent pres-
sure. The continued independence of the Marghi was advantageous to Bornu,
since it prevented possible collision with the Adamawa Fulani and discouraged
slaves from trying to escape southwards.
The southern frontier of the Sokoto-Gando Empire marked the broad division
between the states of the Sudan and those of the forested zone. In the west Gando
had a frontier with the Kingdom of Borgu. Westwards from Yelwa, on the south
42 Chapter 2
side of the River Niger, there was a narrow frontier of contact that the Fulani
attacked frequently without success. Between Yelwa and Jebba the Niger River
flows through a series of deep gorges that effectively separated the two states.
This frontier was continued westwards from Jebba into the forested zone.
There was an unstable frontier of contact between Ilorin Province of the Gando
Empire and the Yoruba states of the south. Both states maintained permanent
armies at the limits of their territory and the location of the frontier depended on
their relative strengths at any time. Eastwards this frontier broadened into one
of separation between the Fulani of Kabba and the Benin Kingdom. At intervals
both states raided the frontier for slaves, further fragmenting the small Yoruba
groups living there. The frontier of separation was continued east of the River
Niger between Nassarawa Province of the Sokoto-Gando Empire and the Igala
tribes. Most of the frontier lay south of the River Benue and was flooded with
the refugees from the north bank of the river, escaping from Fulani forces. The
river protected tribes of the south except at times of low discharge when the
Fulani attackers could cross easily. The Benue also formed a frontier of separa-
tion between the Fulani and Tiv states except for a small foothold that the Tiv
maintained on the north bank of the river. The stability of this frontier resulted
partly from the sturdy independence of the Tiv and partly from their traditional
friendship with the Fulani. The other frontiers of the Tiv group were remarkably
unstable frontiers of contact, resulting from the outward migration of the Tiv.
During this process they annexed the farmlands of the Igala and Ogoja tribes.
The effect of this outward migration caused administrative difficulties for the
colonial administration in 1912. Their response was to build an earth wall, 5 ft
high and 34 miles long to restrict further advance by the Tiv. The plan was
unsuccessful and by 1960 there were many Tiv south of the wall that had fallen
into serious disrepair.
It now remains to describe the frontiers of the four major states in the forested
southern region. Borgu was bounded on the south by a wide forested frontier of
separation with Egba, the westernmost Yoruba state. The frontier was a refuge
for brigands. To the west Egba was separated from Dahomey by a frontier of
separation that narrowed towards the coast.
In the first quarter of the [19th] century the Yoruba state, at its zenith more
powerful than either Dahomey or Ashanti, was breaking up under pressure
from the Fulani and Hausa from the north. In the south several smaller
independent Yoruba states were formed, the chief of which was Egba with
its capital at Abeokuta. In 1820 Gezo [King of Dahomey] repudiated the
tribute which had been regularly paid [to the Yoruba] since 1747, and
resumed the old state of perpetual warfare against Egba, vowing ultimately
to sack Abeokuta. (Naval Intelligence Division, 1944, 43)
After an early sortie that destroyed the Yoruba town of Jenna on the River Yewa,
Dahomey launched a major offensive against Abeokuta in 1851. It was a dismal
failure when the Yoruba of Ishagga, 10 miles west of Abeokuta, promised support,
Frontiers 43
then betrayed Gezo’s forces. The wars between Dahomey and Egba continued
across the frontier.
Gelele [Son of Gezo] continued his father’s campaigns against Egba, and
in 1862 Ishagga was destroyed as a punishment for its treachery of eleven
years earlier. The attack of 1864 found Abeokuta well prepared, with forces
totalling 20,000 against the Dahomeyan army of 12,000. All the villages
between the Yewa and Abeokuta had been evacuated, and all supplies
removed. The invading army, already suffering from hunger, had nothing
but a little cassava for the last 25 miles of its march. Even so, on 15 March
it went boldly to the assault, the Amazons displaying their usual courage,
. . . The Dahomeyans were forced to give ground. This retreat turned into a
rout by a party that took them in the flank. Their losses were 3000 killed
and 1,500 prisoners, whereas the Egba casualties were 50 killed and 100
wounded. (Naval Intelligence Division, 1944, 45-6)
Gelele’s army attacked the small Egba settlement of Ketu on an annual basis and
threatened Abeokuta on a number of occasions. But always the Egba population
west of Abeokuta would fall back into the town and it was never captured by
Dahomey.
The frontier between the Yoruba states and Benin to the east was a peace-
ful frontier of contact that contained a complex intermixture of both groups.
The mingling of peoples shaded from a Yoruba dominance in the west to a
Benin dominance in the east. Bradbury made a useful comment about Benin’s
territory.
It is impossible, at the present time, to determine the extent of the Benin
Empire at any particular period in the past. The frontiers were continually
expanding and contracting, as new conquests were made and as vassals in
the border rebelled and were reconquered. (Bradbury, 1957, 21)
To the south of Benin the delta tribes, such as the Ijaw, preserved their indepen-
dence largely as a result of the defensive attributes of the swamps and creeks.
By the middle of the 19th century a policy of isolation had caused Benin to
withdraw its authority from the western bank of the Niger. East of the Niger the
political organization of the Ibo did not rise above the level of clan or family.
Although some of these groups must have been separated by forested frontiers
they cannot be identified at the scale this analysis has used.
One of the most detailed accounts of the historical evolution of a political fron-
tier was written by Mostert (1993). In 1300 pages he describes, in remarkable
detail, the northward movement of the frontier between Europeans and the Xhosa
44 Chapter 2
People from 1652 to 1858. The book is entitled ‘Frontiers: the epic of South
Africa’s creation and the tragedy of the Xhosa people’. Thompson and Lamar
compared the frontiers of North America and southern Africa on the basis of
the following definition.
[A frontier is] . . . a territory or zone of inter-penetration between two pre-
viously distinct societies, Usually one of the societies is indigenous to the
region. Or at least has occupied it for many generations; the other is intru-
sive. The frontier ‘opens’ in a given zone when the first representatives of
the intrusive society arrives; it ‘closes’ when a single political authority has
established hegemony over the zone. (Lamar and Thompson, 1981, 7)
The Dutch before 1795 and from 1802 to 1806 and the British from 1795 to 1802
and after 1806 were the intruders and the Xhosa were the indigenes.
The year of Governor van Plettenberg’s journey to the banks of the Great
Fish River, 1778, marks the inception of the frontier. It is, as the historian J.S.
Marais says, the year ‘. . . from which we may start the history of continuous
European-Bantu contact’: that is, the start, after a century and a quarter of
slothful and haphazard presence in South Africa, of a far more demanding
situation for the colonial power and its dependents, through which they
arrived at the threshold of the future, as it were. (Mostert, 1993, 228)
Nineteen years later John Barrow, working for the British Governor, reached
the Great Fish River. The high banks and the deep river formed a definite break
in the landscape. But when Barrow journeyed north of the river he travelled for
two days through villages in good order without seeing Xhosa or cattle until he
reached a populated settlement. He discovered that there had been a civil war
between two Xhosa groups and eventually they had disengaged. Before the Great
Fish River had been accepted as the boundary, by the Dutch and British, the
Dutch had sought to locate a boundary at the Gamtoos River and Bushman’s
River but they were breached by Boer settlers, looking for new pastures, in 1770
and 1775 (Figure 2.4).
During the second, short Dutch period the Xhosa crossed over the Great Fish
River to the Zuurveld between the Sunday and Great Fish Rivers. In 1809 the
British developed policies to drive the Xhosa back over the Great Fish River,
populate the Zuurveld with 6000 British settlers, and then insist that the area
between the Great Fish and Keiskamma Rivers should be preserved as a no-
man’s-land.
They should be commanded by treaty to go [north] beyond the Keiskamma
River. The land between the Fish and Keiskamma would then become an
unoccupied bushland belt or green moat between colonists and Xhosa.
(Mostert, 1993, 372)
Frontiers 45
CONCLUSIONS
The study of political frontiers now lies in the realm of historical political
geography. It is possible that new secessionist movements will produce political
frontiers but such attempts are rarely successful. Bangladesh was an exception
for exceptional reasons. It was separated by 2000 km of Indian territory from
west Pakistan, and it was supported by India.
In contrast the study of secondary settlement frontiers is a very appropriate
subject for analysis by both political and economic geographers for the present
and the future. The main areas where such studies will prosper are the tropical
Frontiers 47
deserts in Africa and Australia, the tropical forests of the Amazon Basin, Borneo
and Papua New Guinea, the cold regions of northern Canada and Russia, China’s
Qingzang Gaoyuan [Tibetan Plateau] and Xinjiang Uygur Zizhiqu [Sinkiang]
and Mongolia.
REFERENCES
The missionary, the conqueror, the farmer and, of late the engineer, have
followed so closely in the traveller’s footsteps that the world, in its remoter
borders, has hardly been revealed before one must record its virtually com-
plete political appropriation. (Mackinder, 1904, 421)
Mackinder was speaking at the close of the most intensive period of boundary
construction in world history. It was a period that had enclosed even barren
tropical deserts and unexplored equatorial forests. International boundaries have
now replaced frontiers in all the continents including Antarctica. This chapter
examines the ways by which states created boundaries. The past tense is appropri-
ate because only rarely in the future will two governments have the opportunity
to draw a new international boundary on land.
First this chapter considers the contrasts and similarities of boundary con-
struction on the world’s continents. Second it reviews the procedures by which
boundaries have been drawn between adjacent territories. Third it examines the
possibilities for the determination of new boundaries in the future. Fourth it
provides a case-study of boundary evolution between Thailand and Laos and
Cambodia.
Alsace-Lorraine was acquired by the French through the methods which have
led to the consolidation of most modern States, namely, conquest, trickery
and cession. (Temperley, 1920, 159)
The brutal honesty of this statement would be preserved if names such as Tacna,
Bornu and Primorskiy Kray were substituted for Alsace-Lorraine and the charge
was made against Chile, Britain and Russia respectively. However, it would be
an error to assume that because European states had similar territorial ambitions
in Europe and other continents and sub-continental regions and used similar
polices to achieve them, that there are close similarities between the evolution
of boundaries in Europe and in the other continents. There are more important
52 Chapter 3
differences between Europe and the rest of the world than there are between any
other two continents.
Boundary evolution in Europe was entirely an indigenous process. The chang-
ing patterns of major and minor political divisions of considerable areas of the
continent have been recorded without a break, certainly since the Roman period,
say 510 BC. Useful perspectives can be found in general atlases produced by
Shepherd (1922), Ward, Prothero and Leathes (1924), Engel (1957), Treharne
and Fullard (1965) and Barraclough (1978). Menke (1865) and Smith and
Grove (1874) provided detailed representations of classical Europe. There are
also regional atlases including the remarkably detailed Geschiedkundige Atlas
van Nederland (1913-32) and the monumental atlas of southeastern France by
Baratier et alia (1969).
In the other continents, apart from Antarctica, the indigenous processes of
boundary evolution were generally overlain and halted by colonial activities of
the imperial powers, such as Spain and Portugal in South America, Germany
and France in Africa, China, Britain and Russia in Asia and Turkey, Britain and
France in the southern part of the Middle East.
Using a geological analogy, the advent of imperial power in these lands
may be likened to a political unconformity. The processes by which the Incas,
Amandebeles, Sioux, Turkomans, Karens and Aborigines determined boundaries,
effectively ended when the imperial powers acquired authority in their regions.
In some cases, approximations of the indigenous boundaries that existed at that
time were preserved in comparatively short sections by the imperial powers.
Thus the British drew the boundary, between Northern and Southern Nigeria, in
the vicinity of the most southern advance of the Hausa and Fulani against the
Yoruba kingdom of Oyo.
Even a superficial acquaintance with the evolution of international bound-
aries in continents outside Europe makes it clear that many problems arose from
genuine uncertainties about the distribution of geographical features and the pat-
terns of political authority. These problems may have existed in pre-historic times
in Europe, but for the past 2000 years there is no evidence that such problems
caused significant problems. The high densities of population, the considerable
interregional trade, conquests, migrations, the clearing of forests, the draining of
swamps and the absence of entirely inhospitable terrain ensured that the physi-
cal geography of Europe was generally known to the regional authorities at an
early time.
Regional geographical knowledge necessary for boundary construction was
certainly completed during the feudal age in Europe. Strayer (1965, 17) pointed
out that effective feudal government is local since it requires the performance
of political functions based on personal agreements between small numbers of
people. He and other writers also showed that in any particular parcel of land
different lords might exercise different authorities related to justice, taxes, the
various uses of forests and forced labour. Genicot (1970) provided a very illu-
minating account of this type of situation. Thus it follows that the areas within
The Evolution of International Boundaries 53
which these important rights were available were closely defined. Genicot (1970,
32) is of the view that by the end of the 11th century the growth of population
and the extension of authority had resulted in the remaining frontiers of Europe
being contained within fairly precise lines. There were still disputed zones but
their limits were clearly understood.
Careful study of the treaties contained in the collection of European boundary
treaties by Hertslet (1875-91) shows that boundary construction was simplified
because the continent was divided into a hierarchy of local administrative units
with defined limits. The reader is constantly aware that the international boundar-
ies, described in the treaties, coincide with the limits of cantons, principalities,
bailiwicks, cercles, counties, Commanderies of the Teutonic Order, circuits,
arrondissements, bishoprics, duchies, landgraviates, communes and parishes.
Thus boundary negotiations in Europe were not bedevilled by problems such
as trying to determine the extent of the territory belonging to the Paramount
Chief of Barotseland or the limit of lands occupied by the Saryks on the border
between Afghanistan and Russia.
The situation in Europe also gave a positive advantage that was denied to
negotiators in other continents. In most regions of Europe the land was divided
into small parcels, which were sometimes aggregated into larger administrative
units. These local divisions provided a series of building blocks from which
national territories could be fashioned. The international bargaining centred on
territorial fragments and once their disposition was decided there was a ready-
made boundary that might, at different sections, coincide with known boundaries
between parishes, bishoprics and bailiwicks. This advantage was not available
to the same extent in the Balkans where feudalism, as known and practised in
western Europe, did not exist (Pounds, 1947, 129). Nor did it apply after World
War I, when strategic considerations were a powerful factor in determining the
alignment of boundaries.
Just as the physical and political geographies of Western Europe were known
clearly to negotiators, so the economic geography of disputed regions was under-
stood. Examples of states pursuing important economic objectives include the
determination of Germany to secure as many of the iron-ore fields of Alsace-
Lorraine as possible in 1871; the competition for coalfields in Silesia amongst
Czechoslovakia, Germany and Poland in 1919; Yugoslavia’s enthusiasm for
securing the port of Rijeka by the Treaty of Rapallo; and Rumania’s insistence
on controlling the railway from Timisoara to Arad. Although these are compara-
tively recent examples in Europe’s long history, there is no reason to suppose
that rulers in earlier periods were not fully aware of the economic advantages of
securing the transfer of specific feudal rights from a defeated neighbour.
In a broad swathe of central Europe stretching from the coasts of Germany
and Poland to the shores of the Adriatic, Aegean and Black Seas in the southeast,
present national boundaries were fashioned in a series of major wars. To a much
greater extent than on any other continent, boundaries in this region were created
during the peace conferences that followed widespread conflict. Judged by the
54 Chapter 3
speed with which agreement was reached at the Congress of Vienna in 1815,
the Congress of Berlin in 1878, the London Conference in 1913, and the peace
conferences at the ends of World Wars I and II, it is possible to conclude that
wars simplify the process of drawing boundaries. An example of this conclusion
occurred in 1878. Russia had defeated Turkey and by the Treaty of San Stefano,
signed on 3 March, Russia forced Turkey to disgorge territory and capital.
As Salisbury had predicted, its provisions were far more draconian for
Turkey than the demands the Great Powers had made at the Constantinople
Conference back in January 1877. The Treaty provided for an autonomous
‘Big’ Bulgaria stretching from the Aegean to the Danube and from the Black
Sea to Albania, thereby including much of Macedonia, which Russia claimed
the right to occupy for two years. Romania, Montenegro and Serbia were
to become independent. Turkey had to pay a 700 million rouble indemnity
and surrender Ardahan, Kars, Batoum and Bagazid to Russia, retaining
only Adrianople. (Roberts, 1999, 185)
The major European powers were alarmed at Russia’s major gains and three
months later, in 30 days at the Berlin Congress, both Russia’s gains and Turkey’s
losses were significantly reduced.
When wars end the victors usually have clear ideas of the territorial arrange-
ments they will demand and the defeated state is rarely able to resist most of
the required adjustments. In many cases there is an overwhelming desire on the
part of all parties involved to secure a settlement that will allow the abnormal-
ity of war to be ended. Successful governments want to demonstrate to their
citizens the gains bought by military sacrifice. They also seek to avoid the need
to maintain large armies and navies, and wish to reduce to a minimum the
expenditures involved in administering the territories of the defeated state. The
defeated countries usually seek to regain a level of independence necessary for
the reconstruction of national morale and wealth.
Europeans not only dominated boundary construction in Europe, they also
played the major role in the other inhabited continents and continental regions.
The indigenous population generally played only a minor role in boundary con-
struction. The exceptions were Turkey and Persia in the Middle East and China in
north and central Asia. The minor roles played by indigenes sometimes involved
colonial powers agreeing with each other, to set international boundaries in the
vicinity of frontiers recognized by tribes.
Before examining the evolution of past, present and future boundaries it is neces-
sary to discuss the primary resources that can be examined to reveal the detailed
The Evolution of International Boundaries 55
explanation for the location of any boundary. The primary resources are relevant
government documents, maps used in negotiating the boundary, the reports of
joint demarcation teams and fieldwork in the vicinity of the boundary.
The documentary material may be classified into three sections. First there is
the correspondence between the negotiating parties and between representatives
of the same government. These records are rarely published and usually are found
only in archives. Access by scholars to such material will often be restricted for
a period of 25 or 30 years. These sources are invaluable and are indispensable
in discovering the detailed reasons for the alignment of the boundary. In the let-
ters, minutes and reports can be discovered the geographical, political, economic,
ethnic, strategic and legal factors that played a significant role in producing
agreement on the general location and finally the specific site of the boundary.
In 1957, an examination of the Anglo-French correspondence, in the Nigerian
archive in Ibadan, dealing with the settlement of the inter-Cameroons boundary
in 1920, revealed the following details that would have been difficult to determine
by other means. First the location of the boundary in Bornu resulted from the
incorrect decoding of a British telegram in Lagos in 1916. France benefited from
the error and refused to allow the line to be corrected. This obstinacy allowed
Britain to press successfully for the re-unification of the Higi, Holma and Zummu
pagans in British Cameroons. Previously they had been divided by the Anglo-
German boundary. Second France was keen to secure Garua, a port on the River
Benue, and the land route from Douala on the coast to Garua. Meanwhile Britain
was anxious to re-unite the former Emirate of Yola that had been split between
Britain and Germany in the 1880s. Third the use of inaccurate maps in delimit-
ing the new boundary resulted in two disputes. The first, at the southern end of
the boundary, involved valuable oil plantations, whereas the other in the north,
related to swamplands that were used for cotton production and winter grazing.
Fourth these negotiations were conducted with regard to boundary arrangements
being made in respect of other German colonies (Prescott, 1971, 45-62).
During fieldwork on part of the Benin-Nigeria boundary in 1957 only three of
the twenty surveyed monuments could be found. All three were found near vil-
lages, none was intact since they had all been used to sharpen cutlasses or axes.
There were no significant changes in the cultural landscape within 20 miles of
the boundary; the lives of the people seemed unaffected by the boundary although
some Nigerian farms lay partly in Benin, in accordance with the provisions of
the Anglo-French Agreement of 1906.
The villages situated in proximity to the boundary shall retain the right to
arable and pasture lands, springs and watering places, which they here-
tofore used, even in cases where such arable and pasture lands, springs
and watering places are situated within the territory of one Power, and the
village within the territory of the other. (Brownlie, 1979, 171)
The two crossing points at Ijoun and Idiroko were 32 miles apart and most of
the populations crossed the boundary by uncontrolled paths. At times of tax
56 Chapter 3
collections there was some movements across the boundary in order to escape
fiscal responsibilities. During an interview with Aleketu, who was a Yoruba
chief in Benin, separated from the majority of his people, he said ‘We regard
the boundary as separating the French and the English, not the Yoruba’. All this
information was useful in understanding the situation in the borderland in the
1950s and it could only be obtained by fieldwork.
being established in a location where the forces from each side were neutralized.
Finally, in the ideal stage the boundary became permanently fixed, and a gradual
diminution of state functions applied at the boundary reduced its significance
as an element of the landscape. This altruistic concept of boundary evolution
probably owed much to the world situation when it was published and the hope
that the Great War would bring a long period of peace and stability.
Lapradelle (1928) identified three stages of boundary evolution: preparation,
decision and execution. He emphasized the tentative nature of the first stage com-
pared with those that followed by using the word le trace, translated as ‘outline’
or ‘sketch’. Jones (1945) follows Lapradelle closely in suggesting four stages
of boundary evolution: allocation, delimitation, demarcation and administration.
Each of these stages requires the two states that are drawing the boundary to
work together.
Allocation refers to the political decision on the distribution of territory; delimi-
tation involves the selection of a specific boundary site; demarcation concerns
marking the boundary on the ground; and administration relates to the provisions
for supervising the maintenance of the boundary. Allocation and delimitation
involve drawing lines on maps. Ideally the maps used will be the best available
and they will be at a large to medium scale. Demarcation requires the joint survey
teams, supplied with the best possible maps suitably marked and annotated, to
proceed into the borderland and mark the boundary on the ground by monuments
and, in forested areas by cleared vistas between monuments.
Nicholson (1954, 16) tried to marry the schemes of Brigham and Jones by
carrying the process through from the tribal stage to the demarcated boundary.
However, he admits that the only correlations between the first frontiers and the
final boundaries in Canada were fortuitous and there was no continuous develop-
ment. There seems to be no reason why his ideas should not apply where there
is a continuous history of indigenous development. China and parts of South
America might be territories where Nicholson’s concept could be explored. At
the onset of European settlement in Australia, aboriginal communities knew the
extent of their tribal lands. That knowledge has been lost by some groups for
a variety of reasons, but in the remote parts of Australia clans know precisely
where their boundaries on land and sea are located (Davis and Prescott, 1992).
One boundary that passed through the three stages in sequence was the Anglo-
French boundary between The Niger River and Lake Chad. The first boundary that
allocated territory was signed in 1890. Eight years later a Convention delimited a
boundary. In 1910 the boundary was demarcated and this line generally lay south
of the line of allocation (Figure 3.1). It must not be presumed that all boundar-
ies have passed through the stages of allocation, delimitation and demarcation
in a chronological sequence. In some cases the original allocating line has been
demarcated with no intervening delimitation. In other cases there has been more
than one delimitation before demarcation occurred. There are many international
boundaries that have never been demarcated. There are also many boundaries
The Evolution of International Boundaries 61
that passed through the first three stages of settlement and were then ignored.
Often, especially in Africa, the demarcated boundary was not administered and,
in some cases, sections of the boundary were obliterated by natural processes
and human intervention.
Allocation
Saryks and also their pastures’ (Prescott, 1975, 124). A demarcation commission
started work within two months and progress was rapid until Hauzi Khan was
reached. Yate, a British officer recorded that problems would be faced beyond
that settlement.
Hitherto [up to Hauzi Khan] the line of boundary has been pretty rigidly
defined in the Protocol, and unreasonable claims have been out of the ques-
tion; the further we go through, the less precisely is the Protocol worded, and
should the Russian Commissioner insist on putting forward claims depriving
Maruchak, Kilah Wali, Maimanah and Andkhue, not only of their pastures
but also of their wells, it will be very evident that it is not his intention to
help on the negotiations. (Prescott, 1975, 118)
As Yate noted the basic problem was that there was no clear division between
the cultivation of the Afghan and Russian subjects.
The water supply, cultivation and pasturage belonging to Padjeh [Russian]
is so mixed up with that belonging to the [Afghan] remainder or upper por-
tion of the Kushk Kashan and Murgab valleys that to delimitate a frontier
[boundary] across these valleys was not only a work of great difficulty,
but, when done the boundary is only an arbitrary line based on the circum-
stances of the moment rather than any permanent and natural basis. (Prescott,
1975, 119)
In 1878 the Treaty of Berlin allocated territory between Albania and Montenegro
by a line defined in principle according to the location of specified tribes.
It then coincides with the existing boundaries between the tribes of the
Kuci-Drekalovici on one side, and the Kucka-Krajna, as well as the tribes
of Klementi and Grudi on the other, to the plain of Podgorica, from whence
it proceeds towards Plavnica, leaving the Klementi, Grudfi and Hoti tribes
to Albania. (Hertslet, 1891, vol. 4, 2782)
In May 1908 Ethiopia and Italy defined a boundary in the Horn of Africa that
would separate the tribes subject to Italy from all other tribes subject to the
Emperor of Ethiopia.
ART. I. The line of frontier between the Italian possessions of Somalia and
the provinces of the Ethiopian Empire starts from Dolo at the confluence of
the Daua and the Ganale, proceeds eastwards by the sources of the Maidaba
and continues as far as the Uebi Scebeli following the territorial boundaries
between the tribe of Rahanuin, which remains dependent on Italy, and all
the tribes to its north, which remain dependent on Abyssinia.
ART. II. The frontier on the Uebi Scebeli shall be the point where the bound-
ary between the territory of the Baddi-Addi tribe, which remains dependent
on Italy, and the territory of the tribes above the Baddi-Addi, which remain
dependent on Abyssinia, touches the river. (Brownlie, 1979, 835)
64 Chapter 3
Imprecise lines such as these concern the first stage of boundary definition and
the second stage is designed to eliminate any inconvenient uncertainty.
Delimitation
preserved if the borderlands lacked any economic or strategic value and if the
demarcation of the boundary would have been an unnecessary and unjustifiable
expense. An examination of the political map of the world reveals many geo-
metric boundaries are located in tropical deserts and Antarctica. Second straight
lines persisted when the two countries were unable to agree on any alteration.
Angola and Namibia inherited a boundary with sections of straight lines because
Germany and Portugal could not reach any alternatives to the boundary that
allocated the territory in 1886 (Brownlie, 1979, 1027-8). After South Africa
acquired a mandate over German Southwest Africa in 1919, the Portuguese and
South African authorities contented themselves with clarifying the termini of the
straight sections of boundary, the position of the boundary in the Kunene River,
and the location of the cataracts mentioned in the agreement of 1886. Third it
was usual for straight-line boundaries to be retained when a colonial power came
into possession of the separated territories. The straight sections of boundaries
between Tanzania and Kenya, Tanzania and Uganda and four straight sections
of the boundary of Cameroon with the former French colonies persisted when
Britain was awarded Tanganyika and France received most of Kamerun.
Two types of boundary definitions can be encountered during the process of
delimitation. The first involves complete definition, that requires the surveyors
to proceed into the field and trace the line that has been closely defined. The
second type of definition gives the surveyors power to vary the line, sometimes
up to a specific distance on either side, in order to make the identification and
administration of the boundary easier.
The 1893 Anglo-French boundary between modern Ghana and Cote d’Ivoire
provided an example of complete definition.
The common frontier [boundary] then leaves the River Tanoe and strikes
northward to the centre of the Ferre-Ferrako Hill. Thence passing 2 miles to
the eastward of the villages of Assikasso, Sankaina, Asambosua and Akuakru,
it runs 2 miles to the eastward of the road leading from Suakru to the Boi
River, reaching that river 2 miles to the southeastward of Bamianko, which
village belongs to France. Thence it follows the thalweg of the Boi River and
the line traced by Captain Binger (as marked on the annexed map), leaving
Edubi with territory extending 1 mile to the north of it to France, until it
reaches a point 16,000 metres due east of Yau to a point 1,000 metres to the
south of Aburuferassi, which village belongs to France. Thence it runs 10
kilom. to the westward of the direct road from Annibilekrou to Bondoukou
by Bodomfil and Dadiassi, passes midway between Buko and Adjemrah, runs
10 kilom. to the eastward of the road to Bondoukou via Sorobango, Tambi,
Takhari and Bandagadi, and reaches the Volta at the spot where the river
is intersected by the road from Bandagadi to Kirhindi. Thence it follows the
thalweg of the Volta to its intersection by the 9th degree of north latitude.
(Brownlie, 1979, 232-3)
66 Chapter 3
A number of treaties record the discretion that the demarcation teams were
permitted. In 1885 Britain and Russia were deciding the boundary between
Afghanistan and Russia and it was agreed that ‘. . . in tracing this boundary so
that it conforms with the description in this Protocol, and the points marked
on the annexed maps, the said Commissioners will take due account of local
details and the needs and well-being of the local population’ (Prescott, 1975,
124). In 1888 the British and Liberian government agreed on a boundary with
the qualification ‘. . . with such deviations as may hereafter be found necessary
to place within Liberian territory the town of Boporu and such other towns as
shall be hereafter acknowledged to have belonged to the Republic’ (Brownlie,
1979, 382). In the 1904 Anglo-French agreement in the vicinity of Lake Chad
the following guidance was given to the demarcation parties.
In order to avoid the inconvenience to either party which might result
from the adoption of a line deviating from recognized and well-established
frontiers, it is agreed that in those portions of the projected line where the
frontier is not determined by trade routes, regard shall be had to the present
political divisions of the territories so that the tribes belonging to the ter-
ritories of Tessaoua-Maradi and Zinder shall, as far as possible be left to
France, and those belonging to the territories of the British Zone shall, as
far as possible, be left to Great Britain. (Brownlie, 1979, 448)
In 1927 the Belgian and British governments advised their demarcation teams on
the Northern Rhodesia-Katanga boundary that ‘The Commissioners shall have
the authority, generally, to make such minor rectifications and adjustments, to the
ideal watershed as are necessary to avoid the troubles which might arise from a
literal interpretation of the treaty’ (Brownlie, 1979, 709).
This process involves marking the delimited boundary, or in some cases the allo-
cated boundary, on the landscape. Agreements to demarcate a boundary normally
listed the members of the survey teams from both countries, arranged for the
transport and support of the teams, provided guards when the local population
might be hostile, set out the permissible variation in fixing the boundary and
specified the methods by which the boundary will be defined, such as pillars and
cleared lines. Sometimes demarcation did not follow promptly after delimitation
or allocation and many boundaries delimited in the first half of the 20th century
were never demarcated until major advances in surveying techniques and rapid
transport had occurred,
The commencement of World Wars I and II interrupted planned demarcations
and at their conclusions other matters of reconstruction seemed more important
than boundary delimitation. Laws (1932) and Peake (1934) described how the
The Evolution of International Boundaries 67
boundaries separating the former Belgian Congo from British Northern Rho-
desia and Tanganyika were only delimited when copper and tin deposits were
discovered in the borderland and companies needed certainty about the extent
of Belgian and British authority.
A brief account of the demarcation of the boundary between what are now
Indonesia and Papua New Guinea shows a sample of the problems to be faced
(Figure 3.4). It is based on an account of three Australian surveyors, Cook,
Macartney and Stott (1968). In 1910 A Dutch-German survey team made some
useful maps of the prospective borderland, within which the Dutch hoped to
find natural features that the boundary could follow. When Australia was given
a mandate over German New Guinea after World War I the concept of a natural
boundary was abandoned. Instead there was general agreement that the boundary
would follow meridian 141° E southwards from the north coast of New Guinea
to the Fly River, and the meridian passing through the mouth of the Bensbach
River on the south coast northwards to its intersection with the Fly River. The
boundary would follow the Fly River between these intersections.
In 1933 surveyors from the two countries identified the meridian on the north
coast. When compared with a calculation made in 1928 a difference of 398 metres
was found. The readings did not overlap and accordingly it was decided to divide
the distance between the two positions meaning each territory was increased in
area! New Dutch proposals to complete the demarcation in 1939 were overtaken
by other events. In the three years following 1957 KLM Airways and Adastra
Airways photographed the north and southern boundary sectors respectively. In
1956 a Dutch surveyor fixed a point on the west bank of the Bensbach River and
two years later an Australian made a similar calculation. These were linked by
surveys and a value 141° 01’ 07” E was adopted. In 1962 markers were placed
on the north and south banks of the Fly River on meridian 141° 01’ 07” E at
6° 54’ S and on meridian 141° E at 6° 19’ S.
In terms of scale one of the major demarcations in the past 50 years occurred
in 1962 and involved China and Mongolia. The agreed boundary stretched for
4,672 km from an elevation of 4104 metres in the west to 646 metres in the east.
Pillars were erected at 639 locations that were selected for a variety of reasons.
The first group consists of physical features such as cols, gaps, peaks, rivers
and lakes. The second category was found where a road or railway crosses the
boundary. The third group includes important wells, springs and triangulation
stations that are intersected by the boundary (Department of Commerce, 1971,
2). At some points, for example crossing a road, or following a river two or
three markers bearing the same number may be used. When multiple markers
are included there is a total of 678 pillars. There are 572 concrete pillars and 106
fragmentary rock mounds. The boundary description is divided into 26 sectors
that contain from 3 to 59 pillars. Each pillar is carefully described.
Boundary marker No. 348 is a solitary concrete post, located on the
southeastern slope of the mountain ridge in the northwest of the seasonal
68 Chapter 3
work on rediscovering the boundary. As a start 907 points in the vicinity of the
boundary had been fixed. The limits of Timor’s detached territory called Ocussi
is mainly defined by the courses of rivers. The western terminus on the coast
has been given a double definition.
Proceeding from the mouth of the Noel [River] Besi, from where the sum-
mit of Pulu [Island] Batek can be sighted, on a 30° 47’ NW astronomical
azimuth, following the thalweg of the Noel Besi, that of the Noel Niema and
of the Bidjael, up to its source. (Deeley, 2000, 70)
The specified bearing was determined in 1915. If the mouth of the Noel Besi
has moved along the coast in either direction, in the intervening 90 years, the
two parties will have an interesting problem to solve.
Lambert (1965) and O’Sullivan (2001) have provided very useful accounts
of the work of the International Boundary Commission created by the Canadian
and American governments in 1925. Article 4 authorises the Commission to
inspect the boundary when it is necessary and sets out the actions to be taken
when defects are detected.
. . . repair all damaged monuments and buoys; to relocate and rebuild monu-
ments which have been destroyed; to keep the boundary vistas open; to move
boundary monuments to new sites and establish such monuments and buoys
as they shall deem desirable; to maintain at all times an effective boundary
line as defined by the present treaty and treaties heretofore concluded, or,
hereafter to be concluded; and to determine the location of any point on the
boundary line which may become necessary in the settlement of any question
that may arise between the two governments. (O’Sullivan, 2001, 88)
The Commission submits an annual report on its activities in the field and the
office. In a remarkable and encouraging display of mutual trust a Canadian or
American survey team work on both sides of the sector of the boundary to which
they have been sent. Information about the work done is exchanged at the end of
the fieldwork season. Both countries have created public reservations 18 metres
wide along the boundary. However, if lands had been previously alienated to
private use the reservation does not over-rule it.
and they will generally be related to the duration of these processes. For this
reason it is essential to know how long the boundary has occupied particular
configurations.
Second as the position of the boundary is changed territory will be transferred
from one side of the boundary to the other. This alteration in sovereignty may set
in train changes in settlement patterns, migration streams and the orientation of
local economies. Third transfers of territory from one country to another might
provoke subsequent irredentist movements within the detached areas as reunion
with the homeland is sought.
Changes in the position of an international boundary can occur for three main
reasons. First during the evolution of the boundary there might be changes in
position from the allocated line, to the delimited line to the demarcated line. It
is likely that changes of this nature will decline as successive stages are reached.
For example, the Anglo-French boundary between the River Niger and Lake
Chad passed through two allocations, one delimitation and one demarcation.
The second allocation of 1898 transferred 14,800 square miles to Britain and
4,550 square miles to France, compared with the 1890 allocation. The maximum
displacement of the boundary was 90 miles. The delimitation of 1904 caused
a maximum movement of the boundary of 70 miles and France gained 19,960
square miles. When the delimited line was demarcated in 1907 the commission
made nine small changes, the largest of which involved 17 square miles.
Second the position of boundaries might change at the end of a war to the
disadvantage of the defeated state. Reciprocal examples of this situation involved
Denmark and Germany over a period of 55 years (Figure 3.5). For seven centuries
after 1100 Denmark always consisted of at least the Kingdom of Jutland and
the Duchies of Schlesvig and Holstein. In Jutland the population was Danish, in
Schleswig the population consisted of Danes and Germans and in Holstein most
of the population was German. In about 1060 the Danes had constructed a line of
defensive earthworks that closed the southern end of the peninsula (Naval Intel-
ligence Division, 1944, 102). It was called the Danevirke. In 1807 Denmark’s
navy was captured by Britain and that country then frustrated a Danish assault
on Sweden. When Napoleon was finally defeated Denmark found itself on the
victorious side and still in possession of Jutland, Schleswig and Holstein. It had
also acquired Lauenburg, adjacent to Holstein. In 1848 The Duchies renounced
the sovereignty of Denmark and an inconclusive war left Denmark intact. A
second war started between Denmark and Germany in early 1864. Denmark
offered to withdraw to the Danevirke yielding Lauenburg and southern Holstein.
This offer was refused and the Great Powers and the other Scandinavian states
proposed that Denmark should cede Lauenburg, Holstein and part of Schleswig
(Naval Intelligence Handbook, 1944, 134-5). Denmark declined this suggestion
and was finally forced to abandon Lauenburg, Holstein and Schleswig. Thus in
1864 the Danish Dutch boundary had moved from the location of the Danevirke
to the northern limit of Schleswig.
72 Chapter 3
That situation continued until 1920. During World War I Denmark remained
neutral. Danish citizens of German Schleswig were subject to conscription and
about 30,000 fought against the Allies and one-fifth of this number was killed.
After Germany was defeated some Danes began to demand a return to the
boundaries that existed before 1864 or at least a re-incorporation of Schleswig
(Berdichevsky, 1999, 21). Instead the Danish government made what seemed
a wise decision. It insisted it only wanted to incorporate those areas where the
majority of the population wished to be citizens of Denmark. Elections were
held in two zones. In the northern zone 75 per cent of the voters favoured union
with Denmark. In the southern zone 80 per cent favoured union with Germany
(Naval Intelligence Handbook, 1944, 519). It had been agreed that the vote in
the northern zone would considered as a unity. In the southern zone it would
be decided on a parish basis. The final result showed that four parishes in the
northern zone voted against union with Denmark. Three of those parishes lay
along the southern boundary of the northern zone. Abenra, on the west coast
was the fourth parish to vote against union with Denmark.
In the southern zone it had been decided that if there were majorities for
Denmark and Germany in different parishes then the international boundary
would be related to those parish boundaries (Naval Intelligence Division, 1944,
517). It is uncertain whether this provision could have created Danish enclaves
south of the northern zone. Fortunately the result did not require consideration
of this possibility.
The third possible change in the location of a boundary may occur if the two
countries agree to settlement of a territorial dispute by a court or tribunal. In
June 1973 Libya annexed the Aozou Strip that occupied the entire length of the
Chad-Libya boundary. The strip was about 980 km long and 100 km wide. It
is a barren waterless desert with a varied appearance as blackened sandstone
yardangs rise above a sea of sand dunes interspersed with gravel plains. The
civil war in Chad lasted for fourteen years during which time Libya’s control
was not effectively challenged. After President Habre secured control over most
of Chad in 1987 Libya was approached and the governments agreed that if a
solution could not been found within one year the matter would be referred to
the International Court of Justice. This duly occurred and the matter turned out
to be comparatively simple. Chad relied on the Franco-Italian Treaty of Friend-
ship and Good Neighbourliness of 10 August 1955. That agreement defined the
boundary between the two territories according to several international agree-
ments and treaties. They included the Anglo-French Convention of 14 June 1898
and the additional Declaration of 21 March 1899, the Franco-Italian Accords of
1 November 1902, the Franco-Turkish Convention of 12 May 1910, the Anglo-
French Convention of 8 September 1919 and the Franco-Italian Arrangement
of 12 September 1919 (Brownlie, 1979, 32). Libya relied on the Franco-Italian
Agreement signed on 7 January by Laval and Mussolini. It was not ratified by Italy
and both countries repudiated it in 1938. The Court found in favour of Chad and
the de facto boundary since 1973 was moved northwards for about 100 km.
The Evolution of International Boundaries 73
Because the total process lasted only 17 years it is unsurprising that the posi-
tions of the successive boundaries did not contribute to lasting changes in the
landscape or the population distribution. It is also the case that the low popula-
tion densities in this semi-arid steppe could have contributed to the insignificant
impact of the boundary changes on the landscape or the local population.
There is no formula that would enable the influence of boundary changes,
resulting from any of these three processes, to be calculated. Each case will be
unique. However, when the position of a boundary is changed it is possible to
identify the subjects and factors that should be considered. A short list of topics
would include, for example, the nature of the political relationships between the
two countries concerned, the extent of the transferred areas, the characteristics
of the population dwelling in those areas, the location of boundary-crossing
points and the existence of customs dues on some cargoes. Some authors have
studied the population movements associated with boundary changes. Pallis
(1925) studied national migrations in the Balkans during the period 1912-24
when international boundaries experienced major changes. He concluded that
these population movements were the largest in this region since the break up
of the Roman Empire. The movement of people from areas ceded to a neighbour
formed a large proportion of the migrants. For example, in 1913 the total Greek
population of 5000 left the qazas of Jam’a-i-Bala, Melnik, Nevrokop and Stro-
mittsa, when they were ceded to Bulgaria, by the Treaty of Bucharest. In 1914
about 100,000 Moslems left the portions of central and eastern Macedonia that
had been ceded to the Balkan States by the peace treaty with Turkey, to settle
in eastern Thrace and Anatolia.
Much larger movements occurred during and after World War II. When the
Polish boundary was moved westwards to the Oder-Neisse Line at the expense
of Germany, Wiskemann (1956, 118) estimated that in 1946 nearly 1.5 million
Germans left Polish territory and settled in British-occupied Germany. A further
200,000 moved into the Russian sector of Germany. Over the next three years
another 800,000 Germans moved into the Russian sector. By 1954 the number
of Poles living in Polish-occupied territory had risen from one million to 7 mil-
lions. At the same time there was a redistribution of farmland. All holdings in
excess of 100 hectares were confiscated and distributed to peasants. About 3.6
million hectares were allocated to 605,000 families.
Since 1939 opportunities for drawing a new land boundary normally have arisen
in four different circumstances. First during World War II and at its end boundar-
ies were imposed on defeated countries.
Second the decolonisation of European Empires sometimes required the draw-
ing of new boundaries. However, in Africa the members of the Organization of
74 Chapter 3
Africa agreed in 1964 that existing colonial boundaries would continue to apply
to the newly independent states. In some cases those boundaries were interna-
tional boundaries, as between British Nigeria and French Niger, often negotiated
more than 60 years earlier. In other cases, the internal boundaries of extensive
colonial territories, such as French West Africa, were elevated to the status of
international boundaries. French West Africa included Mauritania, Senegal,
Guinea, Ivory Coast, Mali, Burkina Faso, Dahomey, and Niger. French Equatorial
Africa consisted of Chad, Ubangi-Shari, Cameroon, Gabon and Middle Congo.
Guinea became independent in 1958 and the other twelve colonies followed suit
in 1960. In the case of British India it was decided to divide the former colony
into two states, India and Pakistan.
Third after the fragmentation of an existing independent state, such as Yugosla-
via or the Soviet Union, new countries were created and once again, in most cases,
the internal boundaries were elevated to the status of international boundaries. In
1991 the Soviet Union disappeared and was replaced by the Russian federation
and fourteen independent states that bordered the Federation to the south. Estonia,
Latvia, Lithuania, Belarus, Ukraine and Moldova occupied the zone between the
Baltic Sea and the Sea of Azov. Georgia, Armenia and Azerbaijan bounded the
Russian Federation between the Black Sea and the Caspian Sea. Kazakhstan,
Turkmenistan, Uzbekistan, Tajikistan and Kyrgyzstan lie east of the Caspian Sea.
These newly independent states experienced different histories and are of different
sizes in terms of land and population. Ukraine has the largest population with
a total in excess of 50 millions while Estonia with 1.5 million citizens has the
smallest population. The most extensive state is Kazakhstan with more than one
million sq, km. while Armenia is the smallest state with an area of 29,800 sq.
km. The boundaries of these new states were the boundaries that existed when
they were part of the Soviet Union, however, there are four territorial curiosities.
First the Russian territory of Kaliningrad is now separated from the rest of the
Federation by Latvia and Lithuania. Kaliningrad is Russia’s only ice-free port
on the Baltic Sea and it is of interest to Belarus as a counter to total reliance on
either Lithuania or Poland. Second the territory of Azerbaijan is divided into two
unequal parts by Armenia. Third The Times Atlas of the World (Times Books,
1988, plate 43) shows the boundary between Khazakhstan and Uzbekistan as it
crosses the Aral’skoye More [Aral Sea] as a straight line drawn across water. A
small island called Ostrova Vozrozhdeniya, belonging to Uzbekistan, is shown
south of the line. The equivalent plate in The Times Atlas of the World (Times
Books, 1994) shows that the small island had grown to a large island and it is
divided by the original straight line. By 2005 the island had become a peninsula
jutting from the Sea’s southern shore. Fourth the independence of these new states
has preserved the existence of 13 enclaves (Whyte, 2002, 5). Russia, Armenia,
Azerbaijan, Tajikistan and Kyrgzyzstan each possess one enclave respectively
in Belarus, Azerbaijan, Armenia, Uzbekistan and Uzbekistan. Azerbaijan and
Tajikistan each have two enclaves respectively in Armenia and Kyrgyzstan.
Finally Uzbekistan possesses four enclaves in Kyrgyzstan.
The Evolution of International Boundaries 75
Thailand’s eastern boundary with Laos and Cambodia was settled between Thai-
land and France in the period from 1863 to 1926 (Figure 3.6). In 1862 France
obtained a foothold near the mouth of the Mekong River when Annam ceded
the provinces of Bein Hao, Gia Dinh and My Tho. At the same time Annam
renounced any claims to Cambodia, which at that time was a weak state domi-
nated by Annam and Thailand. France then used a ploy that was to serve it well
The Evolution of International Boundaries 77
in the future. The authorities assumed that France had inherited Annam’s claims
to influence in Cambodia. Admiral de la Grandiere sent a junior officer Lagrée
to collect information about Cambodia and to obtain a treaty giving access to or
control over the Mekong valley (Priestley, 1966, 166; Cady, 1967, 275-6). The
secret treaty signed by Lagrée and the Cambodian ruler, amongst other matters,
gave French protection to Cambodia in return for France having exclusive influ-
ence over Cambodia’s foreign relations. Emperor Napoleon III declined to approve
this secret treaty because he wished to avoid giving offence to Britain.
The Cambodian leader, fearful that Thailand would discover the secret treaty
with France, signed a secret treaty with Thailand that confirmed Cambodia was
a tributary state of Thailand. This treaty was ratified on 4 January 1864. Three
months later the Emperor was persuaded to ratify the secret treaty with Cambodia
and the King had no choice but to ratify it. France took the strong position of
refusing Thailand’s right to perform the coronation of King Norodom in June
1864, but agreed to discuss the problem of reconciling the two treaties that
Cambodia had signed. The discussions produced the first Franco-Thai boundary
treaty on 15 July 1867 (Prescott, 1975, 438-9). The main territorial provisions
were that French protection over Cambodia was recognized and that the Cam-
bodian Provinces of Battambang and Angkor, nominally part of Cambodia were
attached to Thailand. The agreement provided that Siamese and Cambodian
officers, assisted by French officers, would determine the boundary and then
demarcate it by stakes and other markers. Following this work, French officers
would produce ‘. . . an exact map . . .’ of the boundary. There is no evidence that
the boundary was surveyed. Authoritative maps in 1888 still showed French
and Thai versions of the boundary (McCarthy, 1888). In July 1891 a British
representative in Bangkok provided the following information.
. . . the boundary with the French has not been delimited on the coastline, in
fact with the exception of the neighbourhood of the Great Lake [Tônlé Sab], it
has not been fixed anywhere in the south. (United Kingdom, 1895-6, 204)
When France moved into the remainder of Annam and then Tonkin in 1884,
Annam’s influence in Laos was ended and this was to the temporary advantage
of Thailand. Petit Luguenin (1948) provides an excellent account of this period.
Britain was warned of French ambitions in Indo-China.
It is too early yet to say how far French rights might extend, but it is prob-
able that they will claim as the proper boundary the watershed between
the Mekong and the streams which fall into the Gulf of Tonquin. (United
Kingdom, 1895-6, 192)
This did not alarm British authorities because Thai ministers had explained
that their country’s eastern boundary lay along this watershed. However, by
February 1893 the French view had changed dramatically as the British Ambas-
sador discovered.
78 Chapter 3
The Under-Secretary of State [of France] said in reply, that the Govern-
ment were still of the opinion expressed by their predecessors two years
ago [February 1891], to the effect that the left [east] bank of the Mekong
was the western limit of the sphere of French influence, and this opinion
was based on incontestable rights of Annam, which had been exercised
for several centuries. He added that these rights were too important to be
abandoned, and too well established for the Siamese to persist in contest-
ing them in the presence of France to put a stop to their violation. (United
Kingdom, 1895-6, 210)
The British Ambassador had no difficulty in refuting the French arguments.
M. Develle still maintained his two previous theses: first that Luang Prabang
was an actual dependency of Annam; and, secondly, that France ab antiquo
had vindicated her right to the left [east] bank of the Mekong. Upon my part,
I urged him that to adduce Annam’s historical claim to Luang Prabang was
a dangerous line of argument . . . M. Develle knew as well as I did that in
every French Annuaire, in every French map, in every French Geographical
Gazetteer, Luang Prabang, until a year ago, had been described as an inte-
gral part of Siam. It is true that within the last twelve months a mysterious
revolution had occurred in the minds of French geographical authorities, but
as an honest man he must be convinced as I was, that the district in question
was, and had been for nearly a century, bona fide Siamese territory, and
that it could not be confiscated by France without a flagrant infringement of
the formal assurances he had given us not to impair the integrity of Siam.
As for the pretence by France ab antiquo to the left bank of the Mekong,
such a supposition was not only contradicted by M. Waddington’s express
declarations on the subject, but by the further fact that, under the Franco-
Siamese Convention of 1886, the French had claimed the right of sending
a Vice-Consul to Luang Prabang. This in itself is absolute proof that the
locality belonged to Siam. (United Kingdom, 1895-6, 274)
But the policies of the colonial powers never rested on historical accuracy, they
were based on the foundation of national self-interest. A quarrel was forced on
Thailand in February 1893 over charges of Thai aggression against Annam.
Thailand’s offer to submit the dispute to arbitration was rejected and in early
April 1893 Stung Treng was occupied. The first French ultimatum was issued
on 13 April 1893 and thereafter pressure was increased until Thailand signed a
treaty of peace on 3 October 1893. French tactics, during those seven months,
were accurately described by the British Ambassador in Paris.
The Siamese Government were now in possession of an ultimatum, a pen-
ultimatum and an ante-penultimatum. In fact the word ‘ultimatum’ had
completely lost its meaning, for each new one seemed to procreate a suc-
cessor. (United Kingdom, 1895-6, 345)
The Evolution of International Boundaries 79
By the Treaty Thailand renounced all claims of territories on the left bank of the
Mekong; its warships could not sail on the Mekong or the Tônlé Sab; no Thai
fortified posts or military establishments could be established in Battambang and
Siem Reap or in a zone 25 kilometres wide along the right bank of the Mekong;
and France would be able to establish depots of coal or wood for its vessels on
the right or west bank of the Mekong (Prescott, 1975, 440). Finally France was
allowed to maintain a garrison in Chanthaburi until Thailand had complied with
all the terms and conditions of the Treaty. Chanthaburi was a settlement about
215 km southeast of Bankok and 15 km from the Gulf of Thailand.
The Treaty contained no provisions for the demarcation of a boundary. French
and Thai maps of the period agreed on the boundary west of Tônlé Sab and
along the Mekong northwards from Stung Treng, but showed different lines
east of the Lake. To the west of the Lake the boundary followed the Chaine des
Cardamomes before trending east and north to intersect the south bank of the
Lake about 30 km from its northern end. On the east both boundaries extended
northwards from the Lake’s shore for about 15 km. The Thai line then swung to
the east passing 50 km south of Stung Treng. The French version of the boundary
continued northwards for about 75 km before trending south before following a
line due east that passed 10 km south of Stung Treng.
The effect of this boundary was to restrict the areas over which Thailand had
complete authority to areas east of Battambang and Siem Reap and a line paral-
lel to and 25 km west of the Mekong River. Only seven weeks after the Treaty
was signed French and British officials were discussing the prospect of a neutral
zone between British and French possessions in the upper Mekong valley. The
British Ambassador in Paris was told what additional concessions France would
seek from Thailand.
. . . the French Government was precluded by many considerations from
dismembering Luang Prabang. The integrity of Luang Prabang was as valid
and reasonable a cause of solicitude to France as the integrity of Kyaing
Ton was to Britain; nor would the French Chambers or French public
opinion tolerate its disintegration. He thought, however, in the first place,
that when the Commission of Experts examined the question on the spot, it
would be found that the necessary area could be obtained without seriously
infringing the boundary of that province. Its western frontier was at present
uncertain, and there probably would be no difficulty in delimitating it in
such a manner as to secure the result we both desired, namely a substantial
buffer. (United Kingdom, 1895-6, 379-80, emphasis added)
Thailand agreed to a new arrangement with France in the Convention of
13 February 1904. Its two main features involved modifications east of the Tônlé
Sab and in the vicinity of Luang Prabang. East of the Lake the boundary fol-
lowed a parallel for 25 km then turned due north for 120 km to the Dang Raek
escarpment. This well marked feature is composed of horizontal sandstones and
80 Chapter 3
the crest is between 350 metres to 500 metres. It is not difficult to believe that
it would attract the attention of surveyors seeking a site for the boundary. From
the eastern edge of the escarpment the boundary followed the watershed of the
Phnom (Hills) Padang to the confluence of the Mekong and the Mae Nam [River]
Mun. After following the Mekong northwards the boundary left that river at its
confluence with the Mae Nam Huang, to its confluence with a tributary called Mae
Nam Tang. The Tang was then followed to its source on the watershed between
the Mekong drainage basin and the Mae Nam Nan flowing south to the Gulf of
Thailand. The watershed is then followed to the source of the Mae Nam Kop
that is followed back to the Mekong. The Convention provided for the demarca-
tion of the boundary by a mixed commission and in Article 8 the seven French
depots located on the west bank of the Mekong were defined. France gained about
15,000 sq. km of territory according to the terms of the Convention.
A Protocol signed on 29 June 1904 made one alteration to the Convention
boundary and then realigned the boundary west of the Tônlé Sab. The alteration
involved transferring to France an area of 800 sq. km at the head of the Mae Nam
Huang. The realignment west of the Lake transferred a large area from Thailand
to France. The boundary followed a course southwest from the northwest coast of
the Lake to Laem [Point] Ling on the coast of the Gulf of Thailand west of Trat.
This new boundary augmented the area subject to France by 6,500 sq. km. This
boundary was described as ‘une frontière naturelle’. This is a curious description
of a boundary that cuts across the obvious grain of the topography and attaches
the estuary of the Khlong Yai and the port of Trat to a hinterland lying beyond
major mountains through which there was no reliable road.
The 1904 Convention and Protocol were ratified on 9 December 1904 and the
scene was set for the demarcation of the boundary by a joint commission that
started work on 31 January 1905 at Svay Don Kev. The boundary to the sea in
the vicinity of Trat was defined by 31 May 1905. No difficult problems were
encountered until the final section of the line was reached. The line was defined
in the following terms.
The boundary will then follow the Klong-Dja river to its source, which is
situated on the Kao-Mai-See mountain. From here it will follow the chain of
mountains to the Kao-Knun mountain and from this point the chain of moun-
tains as far as the sea at the end of Cape Lem Ling. (Prescott, 1975, 443)
In fact there was no chain of mountains between Kao-Mai-See and Kao-Knun;
they were separated by a valley. Commandant Bernard, the French delegate argued
that since the terms of the agreement could not be satisfied it was necessary to
create a boundary that preserved the characteristics of ‘une frontière naturelle’.
Predictably the French proposal lay north of the defined line and delivered to
France most of the peninsula between the estuaries of the Khlong Yai and Khlong
Welu. Bernard also added the argument that Trat needed an adequate zone of
protection to the west. The Thai delegation agreed reluctantly.
The Evolution of International Boundaries 81
In the north the boundary west of the Mekong was delimited in the period
29 November 1905 to 12 February 1906. In this survey season the French depots
on the west bank of the Mekong south of Luang Prabang were defined. Defini-
tion of the depot at the confluence of the Mekong and Mae Nam Nun was left
to the demarcation from Tônlé Sab to the confluence.
Demarcation of the boundary between the Lake and the confluence of the
Mekong and Mae Nam Mun occurred during the period 2 December 1906 to
18 January 1907. The initial difficulty involved identification of the mouth of the
Stung Roluos on the east shore of the Lake. The river enters the Lake between
low-banks that were inundated during the monsoon season. The Commissioners
decided that they would fix the origin of the latitudinal boundary at the point on
the banks where the obvious banks of the river became difficult to detect. The
latitude of this point was measured. Then the latitude of Phumi Khang Cheung
Vott, a settlement near the dry season mouth of the Stung Roluos, was calcu-
lated. The average of the two parallels was calculated to be 13° 14’ 48.1” N.
This latitude was projected eastwards to the vicinity of the river Prek Kompong
Tiam. The second difficulty arose when it was discovered this was an inlet of
the Lake rather than a river. A river flowed into the inlet and there was evidence
that another river, now dry, had also entered the inlet. The Commissioners agreed
to find the meridian that lay mid-way between a distinct river called the O Run,
located near the former Prek Kompong Tiam, and the abandoned course of the
former river. This meridian was not recorded in the minutes of the Commission.
When this meridian was projected northwards it intersected the Dang Raek escarp-
ment in the vicinity of Chong Ken. This pass had been selected in advance by
the Commission as the route to be followed to the well-watered and populated
Khorat Plateau. The Commission then followed a track about 15 km north of
the crest of the escarpment that had been identified by Captain Tixier, a French
survey officer two years earlier. The survey of the boundary along the watershed
of the escarpment was done by Captain Oum working in the opposite direction
towards Chong Ken. The arrangement for this detached survey had been made
by the Commission on 2 December 1906. There was no formal meeting of the
Commission between 5 December 1906 and 3 January 1907. In those four weeks
the Commission had travelled nearly 400 km from the Lake to the confluence
of the Mekong and Mae Nam Mun.
It was only at the confluence that a meeting, on 3 January 1907, recorded the
solution to the problems encountered on the Lake’s shore. At this meeting there
was no discussion of Oum’s survey. There was one more meeting, on 19 January
1907, dealing with the demarcation of the French depot at the confluence of the
Mae Nam Mun and the Mekong. Although there are French references to the
desirability of holding a final meeting of the Commission on 15 March 1907 no
other meetings were held. This happened because the two countries were engaged
in discussions that produced a major re-alignment of the boundaries from the
coast of the Gulf of Thailand to the Dang Raek escarpment.
82 Chapter 3
In six days from 8-13 March a new boundary west of Chong Ken was nego-
tiated. A treaty was drafted, approved by the King of Thailand and signed on
23 March 1907. On that day the King of Thailand left for a health cure in Europe!
This was the last major boundary change between Thailand and Laos and Cam-
bodia. First France retroceded to Thailand the 800 sq. km at the head of the Mae
Nam Huang acquired by France in June 1904. Second France retroceded 1700
sq. km surrounding Trat, that was gained by France in June 1904. In return France
secured the territories of Battambang, Siem Reap and Sisophon totalling 32,100
sq. km (Prescott, 1975, 444-6). Thailand secured another important concession.
France gave up jurisdiction over French Asians in Thailand. Tate (1971, 519)
considered that Thai authorities were not ceding any vital territory and were more
concerned to make progress restoring Thailand’s legal sovereignty.
The Commission appointed to demarcate the new boundary commenced work
on 8 October 1907 in Bangkok. Five sections of the boundary between the Gulf
of Siam and Chong Ken were identified. They were the watershed of the Dang
Reak escarpment east of Chong Ken, the plain between the eastern end of the
escarpment and Aranyaprathet, the rivers Stung Sisophon and Huai Nam Sai,
the former provincial boundary between Battambang and Chantaburi and the
watershed between Phnom [hill] Thum and the sea. The terminus on the coast
was defined as the point opposite, that means due east, of the highest point on
Ko [Island] Kut. French officials thought this was too close to the nearest Cam-
bodian village and the point was fixed at Laem [Cape] Samit.
The boundary between Chantaburi and Battambang was not plainly marked
in the landscape and it had to be recovered. The Commissioners identified the
villages that were definitely either Cambodian or Thai and proceeded to suggest
lines separating these villages, from Thnom Thum in the south and the source
of the Huai Nam Sai in the north.
The last difficulty involved the straight line linking a point on the Stung
Sisophon, 10 km downstream from Aranyaprahtet. The French Commissioners
cleverly insisted that the distance must be measured along the River not ‘. . . a
vol d’oiseau’ [as the bird flies]. This was agreed and the point from which the
straight line to the escarpment started was only 4 km from Aranyaprathet.
The description of the new boundary was recorded in the minutes of the Com-
mission dated 6 June 1908 and the entire boundary was recorded on five maps
at a scale of 1:200,000. The boundary was marked by 50 pillars constructed of
cement, timber or iron. Article III of the Protocol attached to the 1907 Treaty
allowed the Commissioners to make changes to the described boundaries providing
natural lines were substituted for defined lines and neither country experienced
any detriment. Changes were made at various points to produce a boundary that
was more convenient for administration than the described line.
By Article 3 of the 1926 Convention the boundary along the Mekong River
was defined more precisely ‘. . . with a view to avoiding any dispute concerning
the frontier-line formed by the Mekong between Siam and Indo-China’ (Prescott,
The Evolution of International Boundaries 83
1985, xvii). This Article covered four conditions of the river. First where there
were no islands the boundary follows the talweg. Second if there are islands the
boundary follows the talweg of the channel nearest the Thai shore. Third if there
are islands and the closest branch to the Thai shore dries the boundary will follow
the talweg of the silted branch. However, the Permanent High Commission for
the Mekong may decide that the boundary in the silted channel may be moved
to the talweg of the nearest water-channel. Finally, the river lands, called Don
Khioop, Don-Khioo-Noi, Don-Noi and Dom Somhong, that may be regarded
as forming part of the Thai bank or being alluvial sediments dependent on the
bank, shall be attached to Thai territory.
After war started in Europe in 1939 France sought to secure its position in
Indo-China by signing a non-aggression pact with Thailand in October. Thailand
insisted that a condition in the pact must be the return to Thailand of the Luang
Prabang trans-Mekong area and Bassac ceded by Thailand in 1904. France
accepted these conditions and officials were sent to Indo China to settle the mat-
ter. Flood (1960) wrote an excellent account of these events. He notes that the
French metropolitan authorities were sympathetic to Thailand but the colonial
authorities in Indo-China did not share that view. Events in Europe overtook these
negotiations and Japan’s advance into Tonkin encouraged further demands by
Thailand on France. The retrocession of Battambang, Siem Reap and Sisophon
was demanded and if France ended its colonial rule in Indo-China its Lao and
Cambodian territories would be transferred to Thailand. Some military clashes
occurred but Japan intervened and arbitrated the matter in favour of Thailand.
At the Tokyo Peace Convention of 9 May 1941 Thailand received all areas in
western Cambodia lost in the 1904 and 1907 Treaties and Conventions. This
transfer of territory was reversed under the terms of the Washington Accord of
17 November 1946.
The final chapter in the evolution of this boundary involved the International
Court of Justice. Both Cambodia and Thailand requested the Court to decide
which state had sovereignty over the area on which the temple of Preah Vihear
stood. Cambodia’s case rested mainly on the maps prepared by the French sur-
vey officers in 1904. These were maps that the Thai’s had asked the French to
prepare, that were received in Paris by the Thai Ambassador and that were used
without any disclaimer for many years. Thailand’s main claim was that it had
been in possession of the area for a long time. Alas this claim was trumped by
the fact that Prince Damrong, President of the Royal Institute of Siam, visited
the Temple in 1930, and was greeted by the French Resident of the Preah Vihear
Province under the French flag.
The Court found for Cambodia and perhaps the most important point to come
out of the judgement is that boundary agreements are final once the line has been
delimited and demarcated.
In general, when two countries establish a frontier between them, one of
the primary objects is to achieve stability and finality. This is impossible if
84 Chapter 3
the line so established can, at any moment, and on the basis of a continu-
ously available process, be called in question and its rectification claimed,
whenever any inaccuracy by reference to a clause in the parent treaty is
discovered. Such a process could continue indefinitely, and finality would
never be reached so long as possible errors remain to be discovered. Such
a frontier, so far from being stable, would be completely precarious. It must
be asked why the parties in this case provided for a delimitation [demarca-
tion], instead of relying on the Treaty clause indicating that the frontier line
in the region would be the watershed. There are boundary treaties which
do no more than refer to the watershed line, or the crest line, and which
make no provision for any delimitation [demarcation] in addition. The Par-
ties in the present case must have had a reason for taking this further step.
This could only have been because they regarded the watershed indication
as insufficient by itself to achieve certainty and finality. It is precisely to
achieve this that delimitations [demarcations] and map lines are resorted to.
(Recueil des arrêts, 1962, 2, 34)
Had the Thai authorities complained about the boundary drawn around the temple
as soon as the French maps were received in 1907 the matter could have been
considered by the two governments and adjustments made if necessary. The
error of not checking the maps was then compounded when Thailand agreed to
submit the case to the jurisdiction of the Court. Thailand could have insisted
that the matter be settled by negotiation between the two governments. It could
probably have mounted a case in negotiations on the fact that that Captain Oum,
who was unaccompanied by Thai officers when he surveyed and demarcated this
section of boundary, had deviated from the watershed. But the Court had no
reason to consider that possibility, because the facts that Thailand accepted the
maps without adverse comment and used them without any disclaimer settled
the matter to the satisfaction of a majority of the judges.
This review of the evolution of Thailand’s boundary raises six important
points. First Thailand possessed a boundary with Cambodia and Laos that had
evolved during the period of Chinese suzerainty over Indo-China. Second France’s
intervention in Indo-China coincided with a major threat to China from Russia
in central Asia. Third, in common with the actions of the other imperial powers
such as Portugal and Spain in South America, and Britain and Germany in Africa,
France was too powerful to be resisted by Thailand. The Thai officials who
negotiated the boundaries always seemed to be in retreat. Fourth the advances
were made in stages over 40 years. Fifth Thais appointed to the demarcation
commissions lacked the technical skills to ensure that Thailand’s interests were
not undermined by the French surveyors. Sixth Thailand made the fatal mistake
of agreeing to place the dispute with Cambodia before the International Court
without making a careful assessment of the strength of its case.
The Evolution of International Boundaries 85
CONCLUSIONS
Most of the world’s present international boundaries were drawn in the last 300
years. The Europeans drew the boundaries of Europe and then proceeded to
draw the boundaries of the Americas and Africa and were involved in boundary
construction in the Middle East with Persia, Saudi Arabia and Turkey. China
and Russia were responsible for the boundaries of north, central and east Asia,
with a minor role played by Japan. Britain, France, the Netherlands and Portugal
laid out the pattern of boundaries in Southeast Asia that resulted in the creation
of India and Pakistan, from which Bangladesh was detached. French Indo-China
was succeeded by independent Cambodia, Laos and Vietnam. At first Malaysia
and Singapore was a single political entity, but it divided in 1965.
Scholars undertaking research into the evolution of boundaries in any continent
must discover the archives that contain correspondence regarding the boundary
between neighbouring states, the treaties that resulted from negotiations or arbitra-
tions, and the maps used by both parties and by Courts that become involved. It
is also necessary to engage in fieldwork where that can be done safely. Countries
in North and South America, Africa, Europe, the Middle East and Asia have
shown a willingness to request the assistance of Courts to determine boundaries
on land as well as in the sea.
It is now accepted that boundary evolution in the past could involve the
four stages of allocation, delimitation, demarcation and administration. Today
it is most likely that the last two stages will be relevant. The demarcation and
administration of international boundaries is now simplified by the use of global
positioning systems, helicopters, high quality aerial photographs and joint visits
to boundary markers.
Since World War II several countries have achieved independence and in
most cases they have inherited international boundaries or internal boundaries.
There have been three new states in South America, 34 in Africa, five in the
Balkans, 9 along the European border of the Russian Federation, five along the
Central Asian territory of the Russian Federation, and 11 in Southeast Asia. At
present it appears there are some difficulties in the new states lying between the
Caspian Sea and China.
There have been five attempts by fragments of states to secede and become
independent. Biafra, Katanga and Northern Cyprus have failed in their attempts.
The independence of Eritrea is uncertain at the time of writing. Only Bangladesh
has succeeded as a result of the most favourable circumstances of detachment
and assistance from a powerful friend.
86 Chapter 3
REFERENCES
Hertslet, E., 1875-91. The map of Europe by treaty, 4 volumes with continuous
pagination, Butterworth: London.
———, 1909. Map of Africa by Treaty, (3 volumes), HMSO: London.
Hinks, A.R., 1921. ‘Notes on the techniques of boundary delimitation’, Geo-
graphical Journal, 58, 417-43.
Holdich, Sir T.H., 1909. The Indian borderland, Macmillan: London.
Ireland, G., 1938. The possessions and conflicts of South America, Harvard
University Press: Cambridge, Massachusetts.
Jones, S.B., 1945. Boundary-making, a handbook for statesmen, treaty editors
and boundary commissioners, Carnegie Endowment for International Peace:
Washington DC.
King, L.W., 1895. ‘Report on demarcation operations on the southern section of
the Waziristan-Afghan boundary’, Afghan Boundary Delimitation 1895, 391
F, India Office Library: London.
Lamb, A., 1966. The McMahon Line: a study in the relations between India,
China and Tibet 1904-14, Routledge and Keegan Paul: London.
Lambert, A.F., 1965. ‘Maintaining the Canada-United States boundary’, Car-
tographer, 2, 67-71.
Lapradelle, P. ed, 1928. La frontière: etude de droit international (The boundary:
a study of international law), Les Éditions Internationales: Paris.
Laws, J.B., 1932. ‘A minor adjustment in the boundary between Tanganyika
Territory and Ruanda’, Geographical Journal, 80, 244-7.
Luguenin, P., 1948. ‘Auguste Pavie diplomate, la question franco-siamoise des
États Laotiens 1846-96’, [Auguste Pavie, diplomat, the Franco-Thai question
of the Laotian States 1846-96] Revue d’Histoire des Colonies, 35, 200-30.
McCarthy, J., 1888. ‘Siam’, Proceedings of the Royal Geographical Society,
new series, 10, 117-33, 188.
Mackinder, H.J., 1904. ‘The geographical pivot of history’, Geographical Jour-
nal, 23, 421-44.
Menke, T., 1865. Spruner-Menke Atlas Antiques [Atlas of the ancient world],
Justus Perthes: Gotha.
Mostert, N., 1993. Frontiers: The epic of South Africa’s creation and the tragedy
of the Xhosa People, Pimlico: London.
Naval Intelligence Division, 1944. Denmark, HMSO: London.
Nicholson, N.L., 1954. The boundaries of Canada, its Provinces and Territo-
ries, Memoir 2, Department of Mines and Technical Surveys, Geographical
Branch: Ottawa.
O’Sullivan, M., 2001. ‘The work and value of international boundary commis-
sions: the Canadian/American experience’, Boundary and Security Bulletin,
9 (3), 86-93.
Pallis, A.A., 1925. Racial migrations in the Balkans during the years 1912-24,
Geographical Journal, 66, 315-31.
Peake, E.R.L., 1934. ‘Northern Rhodesia-Belgian Congo Boundary’, Geographi-
cal Journal, 83, 263-80.
88 Chapter 3
Vattell, E. de, 1758. The law of nations or the principles of international law.
Volume 3, published by the Carnegie Institution of Washington, 1916, after
translation by G.G. Fenwick: Washington DC.
Ward, A.W., Prothero, G.W., and S. Leathes, assisted by E.A. Benians, 1924. The
Cambridge modern history atlas, Cambridge University Press: Cambridge.
Whyte, B.R., 2002. ‘Waiting for the Esquimo: an historical and documentary
study of the Cooch Behar enclaves of India and Bangladesh’, Research
Paper 8, School of Anthropology, Geography and Environmental Studies:
University of Melbourne.
Wiskemann, E., 1956. Germany’s eastern neighbours, Oxford University Press:
London.
4. INTERNATIONAL BOUNDARY DISPUTES
The relations between modern states reach their most critical stage in the
form of problems relating to territory. Boundary disputes, conflicting claims
to newly discovered lands, and invasions by expanding nations into the ter-
ritory of weaker neighbours have been conspicuous among the causes of
war. (Hill, 1976, 3)
Boundary disputes between countries have been a popular subject for research
amongst political geographers, lawyers, political scientists and historians. Often
their research benefits because governments publish information stressing the
strength of their case and the weaknesses of their opponents’ arguments. In
the absence of disputes, useful information about the history and contemporary
circumstances of the boundary remain buried in the archives of correspondence
files and secret reports.
with South Africa that had its origins a century before. In 1854 the Orange Free
State was created and attempts were made by the Boer authorities to negotiate
a boundary with the Sotho people to the west of the Caledon River. However,
no permanent settlement was possible and there were three wars in 1858, 1865
and 1867. Only the intervention of Britain that annexed Basutoland forced the
Free State to disgorge territory occupied east of the Caledon. The second Treaty
of Aliwal North in February 1869 provided that the Free State would cede the
conquered areas east of the Caledon River (Eloff, 1979). In May 1973 Lesotho
demanded the return of the lands it lost west of the river but these claims have
been unsuccessful.
Second the circumstances that persuade one or both countries that the latent
problem requires solution must be identified. This trigger action may vary from
active diplomacy to invasion. Schofield (1993) has described how the govern-
ment of newly independent Iraq began to publish propaganda calling for the
unification of Iraq and Kuwait in the period 1933 to 1938. Internal instability in
Kuwait encouraged a formal claim to that territory in April 1938.
Just before the war of 1914-1918 Kuwait was an autonomous qadha of the
Wilayat of Basra. The Iraqi Government, as the successor to the Ottoman
Government in the Wilayats of Mosul, Baghdad and Basra, considers that
Kuwait should properly be incorporated in Iraq. If incorporation should take
place, Iraq would agree to maintain the local autonomy of Kuwait with a
guarantee, in the form of a special statute, but of course without prejudice
as to sovereignty. (Schofield, 1993, 75)
In this case internal instability in Kuwait was the trigger for this attempted
annexation. In July 1990 Iraq was making certain demands of Kuwait. They
included $2.4 billion, the value of oil illegally piped out of the southern portion
of Iraq’s Rumailia Field, the extension of that field into Kuwait and the cancelling
of Iraq’s war debts. The Jedda Summit Conference to solve these issues ended
in failure on 1 August and provided the trigger for the invasion of Kuwait on
2 August. Schofield raises the unrelated but interesting point that if Iraq had
limited its occupation to the islands called Warba and Bubiyan the major powers
might not have used force to evict those invaders (Schofield, 1993, 141).
It happens sometimes that the trigger action is the realization of both par-
ties that agreement seems to be beyond reach and that arbitration is necessary.
Belgium and the Netherlands, Cambodia and Thailand, Burkina Faso and Mali,
Chad and Libya, El Salvador and Honduras and Cameroon and Nigeria have all
taken boundary disputes to the International Court of Justice. Argentina and Chile
have submitted three sections of their boundary to adjudication.
When countries negotiate to solve a territorial dispute or submit it to arbitration
they present the strongest arguments available. Often those strongest arguments
will be found in history, geography, ethnography and the law. When boundary
disputes are solved by negotiation the scholar is generally left to guess about the
International Boundary Disputes 93
arguments pressed by both sides and the extent to which some arguments were
decisive. During the formality of court proceedings scholars have the opportunity
to read the information that has been provided in the exchange of Memorials,
Counter-Memorials, Replies and possibly Rejoinders.
In the case between Belgium and the Netherlands the report of the Mixed
Commission ratified in October 1843 proved decisive in the award of the disputed
enclaves to Belgium. In the dispute with Libya, Chad’s reliance on a chain of
treaties between Britain and France, France and Italy and France and Turkey over
a period from 1898 to 1955 won the case. Libya relied only on the Franco-Italian
Agreement of 1935 signed by Laval and Mussolini. It was not ratified by Italy
and both countries repudiated it in 1938 (Prescott, 1998, 246-8).
If countries negotiate a solution to their boundary dispute it is almost certain
that there will be a compromise and both countries will benefit. When a dispute
is settled in court there is the possibility that one country will win and the other
party will lose.
If a territorial solution is negotiated attention must be paid to the results
of the agreement. The results cannot be predicted but the conditions that will
reduce or increase possible problems can be outlined. Effects are likely to be
less severe when the altered boundary has only existed for a short time, when
few state functions have been applied at the line, when the groups separated
by the previous boundary have a cultural similarity, when the economy of the
transferred area was directed across the boundary and when the economy of the
transferred area is self-sufficient. When an area of Kenya, called Jubaland, was
transferred by Britain to Italian Somaliland, in 1924, the second, third and fifth
conditions noted above were satisfied. The British and Italian governments had
never applied state functions at the previous boundary and the Somali groups
from either side were allowed to cross the boundary to find pastures during their
subsistence stock movements.
Potential difficulties for the populations of areas transferred during the settle-
ment of a new boundary might occur if the following conditions exist. They are
that the boundary has existed for a long time, that the population transferred
is ethnically dissimilar to the indigenous population, that the states have rigor-
ously applied fiscal and security functions at the previous boundary, and when
the economy of the transferred area had been closely integrated with the core
regions of the state from which it has been detached.
Weigend (1950) explored the changes that occurred in the area of the South
Tyrol that was transferred from Austria to Italy in 1919. There was a significant
influx of Italians into the area and some of the Austrian farmers, now in Italy,
changed their activities to supply seed potatoes and Swiss Brown cattle to the
Po valley.
94 Chapter 4
The general term ‘boundary dispute’ includes four different types of disagree-
ment between countries. The first type of dispute can be described as a territorial
boundary dispute. This results from some quality of the neighbouring borderland
that is attractive to the country initiating the dispute. The second type of dispute
concerns the precise location of the boundary. It often involves a disagreement
over the interpretation of terms used in defining the boundary at the stages of
allocation, delimitation or demarcation. This type may be called a positional
boundary dispute. Both of these types of boundary can only be solved to the
satisfaction of the claimant state if the boundary is moved.
The other two types of boundary disputes can be settled without moving the
boundary. The third type of dispute arises over state functions applied, or not
applied, at the boundary. This type may be called functional boundary disputes.
They may arise because one party is being too diligent in enforcing its rules
regarding cross-border traffic, or is being inconveniently negligent. The last type
of boundary dispute concerns the use of some resource that straddles the bound-
ary, such as a river or a coalfield. This type of resource development dispute
may be solved by the creation of an organization that will supervise the use of
the shared resource.
Centuries ago some states were so powerful that they did not need to negotiate
over territorial disputes.
In the south-eastern corner of Europe, the political situation had been steadily
changing during the course of the fourteenth and fifteenth centuries. A new
Power, the Ottoman Turks, Mongolian in race and Mohammadan in religion,
had entered Europe as the natural enemy of its Christian states. Advancing
irresistibly westward, they swept away the kingdoms which had been formed
in the later Middle Ages on the ruins of the East Roman Empire. . . . Though
an Asiatic power in origin, they were, at the end of the fifteenth century,
firmly planted in Europe, and no limit could as yet be seen to their expan-
sion. The order of their conquests had been as follows. Entering Europe in
1354, they captured Adrianople, which they made their capital in 1360. The
Latin principalities speedily succumbed. In 1389, Servia was defeated and
surrendered Macedonia, though she remained independent herself; Walla-
chia became dependent in 1391, Thessaly was annexed in 1393, Bulgaria
conquered by 1398, while the Duchy of Athens, the principality of Achaia
and the despotate of Mistra became vassal States. Thus by the end of the
fourteenth century the Turks had annexed or reduced to dependence all the
International Boundary Disputes 95
First, the boundary might have been drawn without full knowledge of the dis-
tribution of people or topographical features. Many of Africa’s boundaries were
drawn through areas for which no precise information was available. This was
particularly the case during the stage of allocation. There are also examples in
Africa, Asia and South America that some boundaries were based on inaccurate
information. Tribal chiefs and local rulers often exaggerated the extent of their
territory in the hope that the colonial powers would confirm their authority over
claimed areas.
For example, on 4 April 1903 the District Commissioner in Barotseland was
instructed by the British authorities to obtain detailed answers to 16 questions
regarding the extent of the Barotse Kingdom in 1891. This information was
necessary because after Britain and Portugal had signed a treaty in 1891 it was
realized that they disagreed about the extent of Barotseland. The matter had been
submitted to the King of Italy for decision. The Commissioner was asked to sup-
ply the answers in one week and justified his answers in the following terms.
It is extremely difficult to collect reliable evidence and to obtain proof of
statements as to the distance Lewanika’s influence extended in 1891, since
practically the only white men who had lived in Lealui at that time were
Revd. Mr Arnott and Revd Mr Coillard, who were both missionaries and
who therefore had no interest in determining the boundaries of Barotse
influence. . . . It is almost necessary in order to form a correct idea of the
position in 1891, to find out the actual truth at some later date, which would
permit of no doubt, and then to work back on the most probable assumptions
to what the position would have been at the earlier period: here we know
to a large extent the position in 1897, and six years earlier in such large
areas as these, the position could not evidently been very different. (Barotse
Boundary 1903, Archives of Rhodesia, 1960)
To define the boundary the King of Italy established which tribes were in a
position of real dependence upon the Paramount Ruler on 11 June 1891. He
excluded the Bampukush, the Bamarchi, the Mambunda and the Bamakoma.
The boundary was defined by a straight line from the Katima Rapids on the
Zambezi river to the River Kwando; the River upstream to meridian 22° E; the
meridian to latitude 13° S; that parallel to meridian 24°, and that meridian to
the boundary of the Belgian Congo. The King explained why he had relied on
a river and straight lines.
Whereas, as regards the delimitation of the territory over which King
Lewanika reigned as Paramount Ruler, any precise delimitation is impossible,
on account either of the absence of distinguishing geographical features,
or of imperfect knowledge of the country, or of the notorious instability of
International Boundary Disputes 97
the tribes and their frequent intermingling (circumstances which have been
admitted both by the Marquess of Salisbury and the Marquess of Lansdowne),
so that it is indispensable, where the natural lines fail to have recourse to
conventional geographical lines. (Brownlie, 1979, 1068)
The boundary along meridian 24° E was subsequently modified to coincide with
sections of rivers.
Second, at the conclusion of a war, new boundaries may be forced upon the
defeated country and that boundary did not accord with previous distributions of
ethnic groups, administrative areas and economic activities. When the Hapsburg
Empire was dismembered in 1919 Italy secured that part of Austrian Tyrol lying
south of the Brenner Pass. About 70 per cent of the population spoke German
and possessed strong cultural affinities with the Austrians on the other side of the
boundary. Following the surrender of Italy and the defeat of Germany, Austria
tried to reclaim that area of Tyrol but was unsuccessful.
Third it is possible that new distributions of population develop after a bound-
ary has been drawn and give rise to territorial claims. This development is more
probable if the boundary is antecedent. After Peru, Bolivia and Chile had been
established in the period from 1818 to 1824 each held part of the Pacific coastline.
Guano and nitrate deposits were found in the coastal desert. The deposits in the
territory of Chile and Bolivia were developed by private enterprise, while those
of Peru were established under a government monopoly. Agreements in 1866
and 1872, between Bolivia and Chile, fixed their boundary at the coast along
latitude 24° S. Revenue derived from mining between parallels 23° S and 25° S
was shared equally between the two countries. Most of the labourers working
the guano and nitrate deposits came from Chile. When Bolivia increased taxes
without consulting Chile, that country occupied part of the Bolivian coast. When
Peru declined to proclaim its neutrality Chile started the War of the Pacific that
lasted from 1879-1884. Chile occupied Bolivia’s entire coastal tract and the
Peruvian coastal territories called Tarapaca, Tacna and Arica. In the 1920s, after
American mediation, Arica was returned to Peru. In this case the existence of the
Chilean workforce along the littoral of Bolivia and Peru played a role in Chile’s
justification to extend its territory northwards (Dennis, 1931, 37, 73).
number 245 on the problem list compiled by the Foreign Office. Generally the
trigger action will be related to some change in the circumstances of the govern-
ment making the claim or the government that is faced with the claim.
Afghanistan instigated a claim for the formation of Pushtunistan in the border-
land, as a homeland for the Pathans, in 1947, when Britain withdrew from India
after supervising its partition. This was a time when Pakistan was considered
vulnerable. The claim lapsed and was revived when there was an internal revolt
in Balochistan following Bangladesh’s successful secession in 1971. After another
two years the creation of the Republican government in Afghanistan renewed the
call for Pushtunistan. The two most common circumstances that have promoted
territorial disputes are the conclusions of wars and decolonization.
At the end of a war territorial adjustments may be imposed by the victor. In
1919 and 1945 European boundaries were altered in favour of Poland, Czecho-
slovakia and Yugoslavia at the expense of the defeated powers. In addition in
1919, Germany was dispossessed of its colonies, three of which were partitioned.
Civil wars might affect the timing of territorial disputes in two ways. First when a
country experiences a civil war it is unlikely to press territorial ambitions against
its neighbours. Then when the civil war is concluded the successful administration
might begin to assert territorial claims against neighbours. After China’s civil
war ended in 1949, China first recovered Tibet in 1950, then sought renegotiation
of treaties that were considered unequal. These Chinese policies created serious
problems with the Soviet Union and India. Then in 1974 China repossessed the
Xisha Qundao Archipelago [Paracel Islands] that had been claimed by Vietnam.
Once the Cambodian revolution ended in 1975, it began to press territorial claims
against Thailand. At the same time, in Laos, when the new communist govern-
ment ended civil strife it instituted militant policies towards Thailand along the
Mekong River.
The process of decolonisation produced indigenous governments, some of
which regarded the boundaries they had inherited as unsatisfactory. In South
America, after Portugal and Spain had lost control of their colonies, four major
wars were fought over territory in the period 1825-1865. In 1971, when Britain
withdrew from the Persian Gulf, Iran seized the islands Abu Musa and the Greater
and Lesser Tunb, at the western end of the Strait of Hormuz. This was two days
before the independence of the United Arab Emirates that had previously included
these islands. In February 1976, when Spain withdrew from Spanish Sahara, that
territory was annexed by Mauritania and Morocco. When Mauritania renounced
claims to any part of Western Sahara, Morocco acquired the entire territory of
Western Sahara.
Sometimes it is the action of one country that induces another to make a claim.
In 1915 and again in 1927 Guatemala made grants of land to the American Fruit
Company in the area between the Matagua River and the Meredon Mountains.
These actions prompted Honduras to raise claims that had been dormant for a
long time. In a similar fashion, Nicaragua granted the United States a 99 year
International Boundary Disputes 99
lease on the Great and Little Corn Islands. Colombia launched a claim that the
islands were formerly part of the Province of Veragua and therefore properly
part of Colombia under the principle of uti possidetis juris.
When the aims of states initiating a dispute are considered they can be divided
into two classes. First there are claims when the state genuinely wants to acquire
the territory and believes that this might happen. The aim involves strengthen-
ing the state by the accretion of territory. The increased strength might follow
from resources found in the area, from the population that lives there, from the
improved access to the sea or improved lines of communication or because a better
strategic situation is created. When the Southwest Africa People’s Organization
commenced its struggle for independence it demanded that Walvis Bay and the
Penguin Islands should be included in Namibia. Walvis Bay is the major port on
Namibia’s coastline. Independence came to Namibia in 1990 and four years later
South Africa generously gave the islands and the port to Namibia. This transfer
improved the country’s economic potential and transport system and significantly
increased the offshore area of sea and seabed that could be claimed.
Morocco and Mauritania had different reasons for annexing parts of Span-
ish Sahara in 1976. Morocco’s annexation of the northern part of the Spanish
Sahara provided access to valuable phosphate deposits. In contrast Mauritania’s
annexation of the southern sector made available a coastline that allowed addi-
tional claims to the sea and continental shelf. Morocco did not secure a similar
benefit because Spain’s Canary Islands lie close to the Moroccan coast. When
Mauritania withdrew its claim in 1979 Morocco secured all of the Spanish
Sahara and so had the best of both worlds. Long before Iraq invaded Kuwait in
1990, it had coveted the territory for two reasons. First it sought possession of
Kuwait’s oil fields, and second it wanted the greater security that the extended
coastline would bring.
The second class of claims are made, apparently, without hope of a success-
ful outcome! Such claims appear to be made for purposes connected with either
domestic or foreign policy. For example, it is assumed by most commentators
that the Philippines claimed part of Sabah when the Federation of Malaysia was
being formed in an effort to postpone or prevent that political association. The
claim persists and it appears to be as unreasonable as the claim that the treaty
limits, established by the United States and Spain, in 1898 and 1900 and the
United States and Britain in 1930, mark the outer limits of the Philippines ter-
ritorial seas. The widest territorial waters measure 285 nautical miles.
Countries have been known to encourage secessionist movements in neigh-
bouring states. The example of Afghanistan’s attempts to create the territory of
Pushtunistan has already been noted. If this attempt had been successful, and
100 Chapter 4
Samuels review encompasses visits by Chinese fleets in 1902 and 1908, when
tablets were erected, the incorporation of the islands into Kwangtung Province
1909, Japan’s interest in the 1920s and 1930s and the later confrontation between
France and Japan over the archipelago. In the end these various economic,
diplomatic and military actions counted for nothing when China occupied the
Xisha Qundao in 1974. Since then China’s sovereignty over Xisha Qundao has
been consolidated.
After achieving independence in 1980, Vanuatu claimed Matthew and Hunter
Islands. These islands were generally considered to be part of French New Cale-
donia. Matthew Island is formed of basalt and is arid while Hunter island is a
volcanic block with some grassy slopes, a few trees and many seabirds (Hydrog-
rapher of the Navy, 1969, 133). These uninhabited islands have no intrinsic value
but they do allow claims to 59,400 sq. nm. of the Pacific Ocean.
Disputes over islands were common during the search for guano during the
1830s and 1840s. By an Act, dated 18 August 1856, American citizens who dis-
covered guano were authorised to take peaceful possession of any ‘island, rock or
key [cay]’ not claimed by any other government. The discoverer was required to
provide, on oath, details of the feature’s location and evidence that it was unoc-
cupied and not claimed by any other country. If the evidence was satisfactory
and the President decided that the feature appertained to the United States the
discoverer was given the sole right to mine the guano and deliver it for sale to
the United States. The United States acquired many islands under the terms of
this act in the Pacific Ocean. It also acquired Navassa,Island that lies between
Jamaica and Haiti in the Caribbean Sea. In 1980 the United States delimited a
maritime boundary with the Cook Islands. Within the exclusive economic zone
of the Cook Islands there are four islets called Pukapuka, Manahiki, Rakahanga
and Penrhyn. New Zealand and the United States had a low-key dispute over
the ownership of these islands, based on the execution of guano bonds under the
1856 legislation (Smith and Colson, 1993, 986). By this agreement the United
States recognized Cook Island’s sovereignty over the four islands. Spadi (2001)
has provided an interesting review of Haiti’s claim to Navassa Island.
The legal basis of claims in South America was usually the principle of uti
possidetis juris (Ireland, 1938). This principle means that when colonial rule
ended, and indigenous independent states were created, they inherited the colonial
boundaries. The principle should ensure, for example, that colonial powers were
prevented from making claims to strategic fragments of territory between the
new states on the coast. The interpretation of the principle established in 1810
caused some confusion. Some states regarded the rule as applying to the limits
legally in force when the act of decolonization occurred. Others believed the rule
was concerned with the boundaries observed for practical administration by the
colonial authorities. If these two lines existed, then the neighbouring states always
supported the line that provided either the largest area of territory or the best
defensive line. For example, Venezuela split away from Gran Colombia formed
104 Chapter 4
the largest possession, with an area of 282 sq.km. and a population of 200,000
(Smith, 1997, 23). The transfer of territory from France to India was achieved by
peaceful negotiation. In June 1949 Chandernagore was transferred to India and
by November 1954 the remaining French territories had followed. In contrast to
this development Portugal resolved to maintain control of Goa, Daman and Diu
all on the west coast. Goa occupied an area of 3750 sq.km. and had a population
of 624,000. Daman included two tiny enclaves called Dadra and Nagar-Aveli.
They were occupied peacefully by India in July 1954 (Smith, 1997, 25). Next
month peaceful demonstrators on the border of Goa were fired on by Portuguese
troops and about 30 were killed. It seems that Portugal considered that if Goa
was not defended Africans in Angola and Mozambique would be encouraged
to seize independence.
President Nehru continued to search for a peaceful solution. However, India’s
patience ended in November 1961, when its vessels were bombarded by the
Portuguese garrison on Anjadip, a small island off Goa. Despite being urged by
President Salazar to resist the Indian invasion, the Portuguese garrisons surren-
dered on 19 December, and the territories have been part of India since then.
Territory may be claimed on the ground that it was ceded by a neighbour.
Sometimes it may be voluntary. When Britain ceded the Wakhan Strip to Afghani-
stan in 1893 it was done to ensure that there was no common boundary between
Russia and British India. Sometimes the cession might be made under duress.
China found itself under duress in 1858. In 1689 China and Russia delimited a
boundary that placed all the northern tributaries of the Amur River within China.
Territory between this watershed boundary and the River Uda was considered a
neutral zone. Russia negotiated this boundary from a position of weakness. Rus-
sian settlers had been driven out of the lower Amur Valley by 1683 and Albazin,
a major Russian fort, had been captured two years later. The Russian exchequer
was depleted and Russia’s economy was suffering. The Chinese ambassadors to
the negotiations were supported by 10,000 soldiers, three times the force avail-
able to the Russian ambassadors. The relative positions of the two countries were
completely reversed 175 years later. Three Sino-Russian treaties in the period
1858-1860 delivered to Russia all territory lying north of the Amur River and the
coastal zone south of that river east of the Ussuri River. This remarkable Russian
advance was made possible by China’s internal weakness. It was weakened by
the Taiping Rebellion, that started in 1851, and the Panthay Moslem rebellion
in Yunnan, that started in 1855 (Naval Intelligence Division, 1945, 31-4, 40-1).
Although the rebellions were eventually defeated the Manchu regime was fatally
weakened. China was also under pressure from Britain and France in the Arrow
War and was forced to make concessions in the Treaty of Tientsin in 1858 and
the Treaty of Peking in 1860 (Naval Intelligence Division, 1945, 18-19).
The last legal claim to territory mentioned by Hill is based on prescription
that is defined in the following terms.
106 Chapter 4
dividing line between the two countries. This was not a theoretical deduction
based on the rights and wrongs of abstract principles. The fact that this line
had received the sanction of tradition and custom was no matter of accident
or surprise because it conformed to the general development of human geog-
raphy and illustrated that social and political institutions are circumscribed
by physical environment. It was natural that people tended to settle up to and
on the sides of mountain ranges; and the limits of societies – and nations –
were formed by mountain barriers. The Chinese side recognized the fact
that high and insurmountable mountain barriers provided natural obstacles
and suggested that it was appropriate that that the boundary should run
along such ranges. But if mountains form natural barriers, it was even more
logical that the dividing line should be identified with the crest of that range
which forms the watershed in that area. Normally where mountains exist,
the highest range is also the watershed; but in the few cases where they
diverge, the boundary tends to be the watershed range.
. . . it is now a well-recognized principle of customary international law
that when two countries are separated by a mountain range and there
are no boundary treaties or specific agreements, the traditional boundary
tends to take shape along the crest which divides the major volume of the
waters flowing into the two countries. The innate logic of this principle is
self-evident. The inhabitants of the two areas not only tend to settle up to
the intervening barrier but wish and seek to retain control of the drainage
basins. (India, Ministry of External Affairs, 1961, 235-6)
Similarly it is manifest that there are passes all along the high mountains
and that there are always contacts across the ranges. But this does not
invalidate the general conclusion that the watershed range tends to determine
the limits of the settlements of the inhabitants on either side and to form the
boundary between the two peoples. Neither the flow of rivers through the
ranges nor the contacts of peoples across them can undermine the basic fact
that a high watershed range tends to develop into the natural, economic and
political limits of the areas on the two sides. (India, Ministry of External
Affairs, 1961, 237)
This quotation explicitly endorses the concept of natural boundaries. Pounds’
elegant essays (1951 and 1956) described the origin of France’s ambitions to
control all territory bounded by the sea, the Alpine watershed, the Pyrenees and
the River Rhine. However, Harthorne (1936) and East (1937) started to undermine
the concept in the 1930s and in the next decade Boggs (1940) and Jones (1945)
completed the demolition. These writers revealed that all political boundaries
are artificial, because they require the selection of a specific line within a region
where changes in the physical landscape may be more or less perceptible.
It is unclear why a watershed should identify the location for a boundary
rather than the geology of the region, or the vegetation or the terrain’s strategic
108 Chapter 4
The extension of Poland’s territory to the Oder-Neisse line was also intended
to compensate that country for the loss of territory to Russia and the damage
caused by Germany’s invasion and occupation. Allied to such cases some coun-
tries secured promises of territorial reward for cooperating with another state.
The 1915 Treaty of London between Italy and the Allied Powers by which Italy
merged her forces in the general war effort, included the following territorial
promise.
In the event of the total or partial partition of Turkey, Italy was to obtain a
just share of the Mediterranean region adjacent to the province of Adalia.
In the event of France and Great Britain increasing their colonial territories
in Africa at the expense of Germany those Powers agree in principle that
Italy may claim some equitable compensation, particularly as regards the
settlement in her favour of the questions relative to the Italian colonies of
Eritrea, Somaliland and Libya. (Temperley, 1921, volume 4, 290)
Ramirez-Faria, (2001, 9) reports that in the 16th century Adal was a Muslim
Kingdom in the Horn of Africa. A settlement under the terms of this treaty trans-
ferred 90,000 sq. km of northern Kenya to Italian Somaliland on 15 July 1924.
The boundary was shifted west of the Juba River, that had previously marked
the boundary, to the present line, that was demarcated in 1930.
Territorial claims based on ethnic arguments refer to the human qualities
of nationality, race, language, culture, history and religion. From time to time
countries have focussed on one of these characteristics as being the definitive
guide to the political predilection of a region’s population. During the Versailles
Peace Conference some delegations laid great emphasis on language as a reliable
guide to national aspirations of people and therefore to the proper distribution of
territory. Ancel (1936, 76) noted that if this argument was followed to its logical
conclusion linguistic divides would become political boundaries and language
would become ‘. . . the symbol of the nation’. At Versailles efforts were made to
draw boundaries that reduced the number of minorities to a minimum. Fisher
(1940) praised the settlements because they left only 6 per cent of the population
of Europe under alien rule. But the conference records reveal that many decisions
were inconsistent and that tests of nationality were not applied uniformly. It is
difficult to see that this principle could have been applied uniformly given the
complexity of population distributions in Europe. The intermingling of popula-
tions in the European borderlands often made it impossible to define boundaries
that precisely separated different ethnic groups. In some regions, such as western
Banat, there was an intricate mixture of Yugoslavs, Rumanians, Magyars and
Germans (Bowman, 1923, 272). Greece claimed the entire Argyro-Castro area
because of the larger Greek rural populations that surrounded the towns occupied
mainly by Yugoslavs. Yugoslavia used this same argument in the Klagenfurt
Basin, where the rural population was Yugoslav and the urban population was
German. The German-speaking population formed the basis of Austria’s claim.
International Boundary Disputes 111
The consequences of territorial disputes can be classified into three groups. First
there are the consequences of the dispute being initiated, second the consequences
if the dispute cannot be settled by negotiation, and third the consequences of an
agreed settlement.
Invariably, before a country makes any formal claim to additional territory
the neighbouring state will know that it is a possibility. In books, newspaper
articles and parliamentary debates the nature of the existing boundary will be
questioned. Once the dispute is initiated the claimant state will develop in detail
the arguments that support the extension of its territory, while the other state will
marshal the best arguments to ensure that no change is made. Relations between
the governments may continue to be smooth with no significant disruption to
commerce, the trans-boundary movements of citizens or police and military
cooperation. But relations might deteriorate and formal relations might become
acrimonious and involve tit-for-tat administrative irritations.
If the dispute cannot be resolved there are three possible consequences. First
the two governments recognize that the dispute exists and determine that their
good relations are more important than adjusting the boundary. For example,
in 1974 China seized the Xisha Qundao [Paracel Archipelago] that had been
occupied by Vietnam. Although Vietnam still contests China’s sovereignty over
these islands, the two countries negotiated an equitable maritime boundary in
the Gulf of Tonkin in 2000. Territorial disputes between China and Japan, Japan
and Russia, Morocco and Spain, Spain and the United Kingdom and Malaysia
and the Philippines have not undermined good relations with each other. Second
relations between the two countries deteriorate and this is reflected in reduced
trade, lower levels of foreign tourism and possibly hostility in the United Nations
and international conferences. Third the claimant state invades and conquers the
disputed territory. Between 1961 and 1990 India, Indonesia, Morocco, Argentina
and Iraq invaded and occupied the territories of neighbours. Only India, that
secured the small Portuguese coastal settlements, and Morocco, that secured Span-
ish Sahara with an area 273,000 sq. km, have successfully retained their gains.
Argentina and Iraq respectively were driven quickly out of the Falkland Islands
and Kuwait. Indonesia agreed to a United Nations plebiscite in East Timor and
withdrew when the vote was for independence.
Settlements can be achieved by direct negotiations and by judicial arbitration.
Cambodia and Thailand, Argentina and Chile, Burkina Faso and Mali, Chad and
Libya, El Salvador and Honduras, Indonesia and Malaysia and Malaysia and
Singapore are countries that have submitted their land-boundary disputes to the
judicial process. If the dispute is settled by the transfer of territory, from one
country to another, there is a wide range of possible consequences. Some people
in the transferred region, who opposed the change might cross the boundary to
remain citizens of their original country. New patterns of internal administration
might be established both in the transferred territory and the area from which it
International Boundary Disputes 113
was excised. The orientation of the economy of the transferred territory might
turn towards the core of the new state. Different forms of production might be
encouraged to supply new neighbouring markets. It would require careful field-
work in each case to establish the changes that have occurred and those that
unexpectedly, have not occurred. Sanguin (1983) described the problems facing
part of Italy, that was transferred to France in 1947.
One of the most enduring territorial boundary disputes involved Ecuador and
Peru. Ecuador and Venezuela seceded from Gran Colombia in 1830, six years
after Peru ended the period of Spanish control. In July 1832 Ecuador and Peru
signed a treaty of friendship and alliance. Article 14 provided that existing
boundaries should be retained until a final boundary agreement was settled. That
happened in 1942!
This account of the territorial dispute between Ecuador and Peru owes much
to the seminal studies by Ireland (1938, 219-30) and St John (1999).
Five rounds of negotiations in 1841-2 failed to produce a boundary, but the
discussions revealed that Ecuador expected Peru to yield the sub-divisions of Jaén
and Mayna. In 1853 Ecuador proclaimed freedom of navigation, on tributaries of
the Amazon rising in Mayna and Jaén. The tributaries included the Chinchipe,
Santiago, Maranon, Pastaza, Curaray, Napo and Putumayo Rivers. Peru protested
against this action and drew Ecuador’s attention to the existence of the Royal
Cédula of 1802. This proclamation detached the provinces of Maynas and Qui-
jos from New Grenada, that later became Gran Colombia and attached them to
the Viceroyalty of Peru. The cédula fixed the western boundary of Peru along
a line running approximately north-south between 120km and 270 km from the
Pacific coast (Figure 4.1).
In 1854 Ecuador tried to reduce its foreign debt by delivering to English
creditors vacant lands near Canelos in the zone disputed with Peru. Peruvian
forces invaded Ecuador and its navy blockaded the port of Guayuquil. General
Guillermo Franco, who controlled the port, signed a treaty of peace, friendship
and alliance with Peruvian leaders on 25 January 1860. By Article 5 Ecuador
revoked the grant of land to the English companies at Canelos. By Article 6 it
was agreed the two parties would appoint delegates within two years to a joint
commission to delimit the boundary. Pending this development the two parties
would respect the boundaries based on the principle of uti possidetis juris, accord-
ing to the Royal Cédula of 1802. By Article 7 Ecuador was entitled to prove its
title to the territories of Quijos and Canelos within two years. Failing such proof
Ecuador’s claim would lapse and Peru’s claim would not be challenged. Article
23 stipulated that neither side should use force against the other without having
submitted the matter for arbitration to a neutral power.
114 Chapter 4
Over the next 27 years the territorial issue continued to sour relations between
the two countries. For example, in 1875 disputes arose over the issue of naviga-
tion on the Morona and Pastaza Rivers. Then in 1881 Ecuador took advantage
of Peru’s discomfort when Chile occupied Lima and Callao in southern Peru.
Ecuador’s forces occupied areas near the confluence of the Coca and Napa Riv-
ers. In 1881 Ecuador again tried to reduce its foreign debts by awarding land
in dispute to foreign companies. Peru’s protests at Ecuador’s various actions
continued until August 1887 when the parties signed a new treaty to submit the
question to arbitration.
The foreign ministers agreed to invite the King of Spain to settle outstanding
territorial questions by arbitration, and he was asked to indicate his acceptance of
this role within eight months. Both sides would submit written arguments within
one year of the King’s acceptance of the invitation. They agreed to respond
quickly to any questions asked by the arbitrator and to accept the final decision.
Ecuador and Peru noted that they would continue to negotiate and would advise
the arbitrator of any progress.
Since King Alphonso XIII was only 15 months old, the Queen Regent accepted
on his behalf. A proviso was added that consideration of the matter would have to
wait until the arbitrations concerning Colombia and Venezuela and Colombia and
Costa Rica were completed. Fresh negotiations between the two governments were
fruitful and on 2 May 1890 a treaty delimiting a new boundary was completed.
It was called the Garcia-Herrera Treaty. In addition to delimiting the boundary
in considerable detail the parties agreed to free navigation on shared rivers. Ves-
sels registered in Ecuador had the right of passage on the Maranon and Amazon
Rivers subject to Peruvian fiscal requirements and river police regulations. On
5 June 1890 both sides signed a Protocol dealing with the demarcation of the
boundary. In the absence of natural boundaries the line must separate the areas
occupied by each country. The surveyors were instructed to place the boundary
according to the landscape features that were most appropriate. The demarcation
teams were authorised to make minor reciprocal changes in the defined boundary
that would not produce major changes.
Although Ecuador did not gain Tumbes on the coast or the area around Jaén,
it gained a large area bounded by the Chinchipe, Maranon and Pastaza Rivers.
Ecuador also gained a corridor along the right bank of the Putumayo as far east
as its junction with the Cobuya River. St John, drawing on material by Wagner
de Reyna (1964) and Sotomayor (1941), explains the remarkable concessions by
Peru, of about 310,000 sq. km, to the following considerations
First, Peru had not recovered economically or politically from the War of
the Pacific [1879-84 against Chile]; and in just four years, it was scheduled
to participate in the Tacna-Arica plebiscite. Needing all available resources
to protect its southern interests, the Garcia-Herrera Treaty was a means to
neutralise Ecuador while Peru focused on its struggle with Chile. Second,
the Peruvian government lacked the necessary legal documents to prove
International Boundary Disputes 115
This strategy was in vain. In March 1922 Colombia and Peru concluded a Treaty
of Frontiers and Free Navigation. The result of this arrangement was that Colom-
bia received a frontage on the Amazon River and Peru secured all the territory
along the Putumayo that Colombia had received from Ecuador! St John sets out
the consequences starkly.
Overnight the Ecuadorean government found itself confronted by an antago-
nist [Peru] where previously it had an ally [Colombia]. From the San Miguel
River eastward, Ecuador was now enclosed on the north, east and south by
Peruvian territory. In addition to destroying any legal support which the
1916 Colombia-Ecuador Treaty had given Ecuadorian claims, the 1922 treaty
eliminated the possibility of Colombian support either military or diplomatic,
for Ecuador in its dispute with Peru. (St John, 1999, 14-5)
In 1936 Peru made a unilateral declaration of a boundary that it would observe. For
the next two years the two countries pursued a solution through the Washington
Conference that in the end revealed only irreconcilable positions. For Peru the
solution lay in determining the limits of the Peruvian provinces of Tumbes, Jaén
and Maynas adjacent to Ecuador. For Ecuador the solution lay in an equitable
division of the overlapping claims either by direct negotiations or arbitration.
The territorial dispute came to a conclusion in 1942. In mid-1941 conflict
started along the Tumbes border and spread to the Pastaza River. Peruvian forces
quickly occupied about 1000 sq. km of Ecuador. A cease-fire was brokered by
Argentina, Brazil, Chile and the United States in October 1941 and talks in Rio
de Janeiro produced the Protocol of Peace, Friendship and Boundaries in early
1942. A boundary was delimited in the Rio Protocol and provision was made
for the demarcation of the boundary, by expert surveyors (Figure 4.2). That was
the end of the territorial boundary dispute.
For another 56 years positional boundary disputes continued as the demarcation
was completed. For example, the watershed between the Zamora and Santiago
Rivers was much shorter than perceived by the delimitation. In addition, Ecua-
dor raised the complication that there was both a Rio Zarumilla and a Canal del
Zarumilla separated by an area up to 5 km wide.
is faced with the difficulty of matching the boundary definition, whether writ-
ten or marked on a map, with the landscape. A few problems will arise if the
demarcation commission makes an error that is only discovered after some time
has elapsed. There might be a legal distinction between these two situations. If
the demarcated boundary has existed for some years without being challenged,
the satisfied party might use this fact to justify the maintenance of the incorrect
alignment. The majority of difficulties with delimitations fall into two categories.
First some terms will permit more than one interpretation. Second some points
will be defined in two ways that are contradictory.
An example of a phrase in a boundary delimitation that allows more than one
interpretation is provided by the Anglo-German agreement of 1886, The bound-
ary was defined from the Cross River Rapids to the River Benue. The northern
terminus was defined as follows.
. . . starting from a point on the left river bank of the Old Calabar or Cross
River, where the original line terminated, shall be continued diagonally to
such a point on the right bank of the River Benue to the east of and as close
as possible to Yola as may be found on examination, to be practically suited
for the demarcation of the boundary. (Hertslet, 1909, volume 3, 880-1)
Adamawa was an emirate that existed in the Benue valley. The capital was Yola
and it was situated closer to the western frontier than the eastern frontier. The
Emir of Yola had decided to remain in the British sphere, which meant that
Germany gained the larger eastern part of his territory.
The description had intended to refer to the left bank of the Benue River; for-
tunately the Anglo-German team demarcation team agreed on that interpretation.
The Agreement also ran the risk of complicating matters by giving two names for
one river and noting that the line should be ‘continued diagonally’. However, the
phrase that caused difficulties for both sides was ‘. . . to the east of and as close
as possible to Yola as may be found on examination, to be practically suited for
the demarcation of the boundary’ (Figure 4.3). For the German surveyors this
raised two simple technical problems of identifying the most easterly feature of
the walls around Yola, and erecting a beacon on the nearest point of the left
bank of the Benue, above the seasonal flood level. For the British team the word
‘practically’ had political, strategic and economic implications as revealed later in
correspondence from the Royal Niger Company to the British Foreign Office.
No more fruitful source of trouble in these regions can be imagined, than
that the powerful Emir of Yola, who has now thrown in his lot with England,
should be informed that this country has drawn a frontier line with Germany,
almost within sight of his walls. (Prescott, 1971, 32)
It was plainly important that for two reasons the Emir should retain some ter-
ritory, on the south bank of the Benue River. The first concerns his status; his
eastern boundary on the south bank must be more than a few kilometres from
118 Chapter 4
the walls of Yola. Second, because German authority would not be uniformly
exercised across the territory of Kamerun, the Emir needed to be able to estab-
lish forward defensive lines, that would give warning of attack by any other
indigenous polity. The third consideration is that the population of Yola required
access to areas east and southeast of the city for collecting firewood and fruits
and for grazing stock.
Fortunately Germany suggested the solution to the contrary positions of the
two countries. This involved drawing the straight line from the Cross River to
the eastern point of Yola, and then diverting the northern part of that that line
eastwards by an arc with a radius of 20 miles as far as the River Benue. There
was some haggling over the radius. Britain suggested the distance from Yola
to the confluence of the Benue with its left bank tributary the Faro River. This
distance, about 30 miles, was eventually accepted by Germany, but Germany then
insisted that the boundary north of the Benue towards Lake Chad, should start on
the north bank 5 km west of the Benue-Faro confluence, and that was agreed.
Problems caused by uncertainty about geographical definitions fall into three
broad groups, The first involves the identification of a geographical feature such
as the bank of a river, the edge of a plateau or the crest of a range. The second
involves disagreements about place-names, and the third category results from the
location of a tribal or provincial boundary that is elevated to international status.
These problems might be exacerbated if the demarcation is only undertaken long
after the delimitation was agreed.
Hinks (1921) reported a dispute between Bolivia and Peru, north of Lake
Titicaca. It proved impossible to agree on ‘. . . the western source of the River
Heath . . .’ that marked a turning point on the delimited boundary.
In 1823 Costa Rica, El Salvador, Guatemala, Honduras and Nicaragua affirmed
their independence as the United Provinces of Central America. Guatemala and
Honduras declared their independence in 1838. Attempts to define the boundary
between Guatemala and Honduras in 1845, 1895 and 1914 were unsuccessful.
On 16 July 1930 the disputed boundary was submitted to a Special Boundary
Tribunal whose chairman was the Chief Justice of the United States (Marchant,
1944, 317-19). The award was published on 23 January 1933 (The Geographer,
1976, 2). Under the terms of the Treaty of Arbitration the Tribunal was permit-
ted to modify a line based on the principle of uti possidetis juris. This principle
provides that boundaries set by Spain for its colonies should persist when the
colonies become independent states. After an aerial survey of the borderland had
been made the Tribunal defined the boundary between the two states from their
trijunction with El Salvador to the Gulf of Honduras
This boundary is 159 miles long but only the most easterly 25 miles is relevant
to this analysis.
. . . northeasterly straight (11.2 miles) to a point at the center of the Cuyamel
Railroad bridge over the Santo Tomas river, northeasterly straight to the
southernmost point on the right bank of the Tinto river which flows out of
International Boundary Disputes 119
the Laguna Tinto, along the right bank, taken at mean high water mark, of
the Tinto river downstream to its point of discharge into the Motagua river,
along the right bank, taken at mean high water mark, of the Motagua river
downstream to its mouth on the Gulf of Honduras. As thus described, the
boundary is established on the right banks of the Tinto and Motagua rivers
at mean high water mark, and in the event of changes in these streams in the
course of time, whether due to accretion, erosion or avulsion, the boundary
shall follow the mean high water mark upon the actual right bank of both
rivers. (The Geographer, 1976, 5)
This delimitation defines the boundary along the riverbank as the mean high
water mark, but there is no indication of how that level should be determined. It
is possible that hydrographers from each state might set out different conditions
for determining the mean high water line, and that two mean high water lines are
calculated. This would only be a problem when the gradient of the right bank is
gentle. In that case a small vertical difference in the calculations of the water-
lines might produce a significant horizontal shift of the boundary landwards to
the advantage of Guatemala.
This delimitation does not define the point where the mean high water mark
of the river terminates at the coast. That point will be the origin of any maritime
boundary and each country has a vested interest in identifying the optimum
intersection that will give it an advantage.
The Anglo-Dutch boundary between Guyana and Suriname dates from 1799.
It was specified by the Dutch Prime Minister in a letter to the Prime Minister of
Suriname on 25 November 1975, the day the colony became independent.
The western boundary is formed by the low-water line on the left bank of
the Corantijn, from origin to mouth. The boundary therefore runs from a
point to be further determined on the southern boundary to the origin of
the Upper Corantijn, next from this origin along the low-water line on the
left bank of the Upper-Corantijn and the Corantijn up to the point where
the river bank changes into the coastline . . . (Republic of Suriname, 2005,
8, emphasis added.)
There are river mouths, flowing into adjacent seas with low tidal variation, where
a riverbank trending in one direction meets the coastline trending in another
direction. In such cases it would be a straight-forward matter to identify the point
where the river bank meets the coastline. The left bank of the Corantijn River
flowing northwards chances to meet the sea where the river forms an estuary.
The west coast of the estuary and the succeeding coastline has the configuration
of an arc of a circle trending north and west. It is very unlikely that experts
from one or both countries would agree on a single point where the riverbank
becomes the coastline. Hydrographers dealing with closing lines for bays have
devised a mechanical method of defining a point on featureless promontories
(Beazley, 1987, 21).
120 Chapter 4
In 1860 China and Russia sowed the seeds of a positional dispute that remained
unsolved for 145 years. The treaty dealing with their common boundary in
northeast Asia in the following terms
Henceforth the eastern boundary between the two states, beginning from
the confluence of the rivers Shilka and Argun, will follow the course of the
River Amur downstream to the point of juncture of the said river and the
river Ussuri. The lands lying on the left [north] bank of the river Amur
belong to the Russian Empire, while those lying on the right [south] bank,
as far as the river Ussuri, belong to the Chinese Empire. Thenceforward
from the mouth of the river Ussuri to Lake Khinkai, the border-line will go
by the rivers Ussuri and Sungach. The lands lying on the east [right] bank
of these rivers belong to the Russian Empire while [those] on the west [left]
bank [belong] to the Chinese Empire. (Prescott, 1975, 54-5)
Unfortunately, at the junction of the Ussuri and Amur Rivers there is a triangular
island with an area of 1320 sq. km. To the north lies the Amur or Hei-Lung
Chiang; it is a large river with an average width of 2.3 km. The southwest
coast of the island is washed by the Protoka Kazakevicheva, or K’o-tsa-k’ai-
wei-ch’ai-wo Schui-tao. This is a narrow watercourse 30 km long and about
900 metres wide. The southeast margin of the island is bounded by a channel
that the Chinese regard as Wu-su-li Chiang and the Russians called Ussuri. The
Russians interpreted the treaty to mean that the boundary followed the Amur
to its junction with the Protoka Kakakevicheva and then this minor channel to
the Ussuri River. The Chinese view placed the juction of the Amur and Ussuri
Rivers at the easternmost point of the Island. The matter was finally settled on
14 October 2004.
Imprecise placenames have been used in some treaties. Hinks (1921) noted
that the demarcation team sent to delimit the boundary between Bolivia and Peru
was faced with the difficulty of interpreting the name Barraca of Illampu. The
term refers to both the entire estate and the main house on the estate.
In 1872 Britain was trying to establish a boundary between Russia and Afghani-
stan that would keep Russia well away from British India, but in the end Britain
had to accept a boundary much further south than had been envisaged. One of
the main turning points was located where the boundary drawn eastwards from
the Zulfikar Pass at the Aghan-Iran-Russian trijunction reached the Oxus River,
also called Amuda ya. This feature was called ‘the ford of Kwaja Salar’ by the
British, in May 1872. By October British authorities called it ‘the port of Khoja
Saleh’. Later again it was ‘the post of Khoja Saleh’. By 1886 it was realised
that the name Kwaja Salar or Khoja Saleh referred to a ferry, a tomb, a house, a
narrow part of the river and the district east of Khamiab. Until the demarcation
teams reached the Oxus they did not realise that the term Kwaja Salar or Khoja
Saleh could be interpreted in different ways at different locations. The Russians
had argued for starting the survey at Kwaja Salar because this was a known point.
International Boundary Disputes 121
Russia also provided a report, prepared by General Kaufmann, that located the
Chuskhar Guzar ferry opposite Khoja Saleh, while Britain thought the ferry was
77 km east of Khoja Saleh.
According to the India Independence Act of 18 July 1947 the Province of Sind
was awarded to Pakistan. South of Sind there were a number of suzerainties,
including Kutch, Tharad and Santalpur. Kutch lay immediately to the south of
Sind and therefore the boundary between Sind and Kutch became the international
boundary between India and Pakistan. In 1914 the coastal section of the colonial
boundary between Sind and Kutch had been delimited along the Sir Creek to
its source at latitude 23° 58’ N, then following the parallel overland for 35 km
to meridian 68° 41’ E. The land boundary was demarcated by 67 sandstone pil-
lars in 1923-4. At the same time the boundary was continued north along the
meridian for 37 km to latitude 27° 17’ N and marked by 66 sandstone pillars.
This incomplete definition and demarcation left a gap of about 300 km to the
trijunction of the boundaries of Gujarat-Rajasthan and Hyderabad.
India and Pakistan had different views on the location of the boundary that
would fill this gap. India proposed a fairly direct line following the northern edge
of the Great Rann of Kutch. Pakistan’s preference was for a boundary that went,
first south from the terminus at latitude 23° 58’ N and longitude 68° 41’ E, and
after turning eastwards through the middle of the Great Rann swung north to the
trijunction. The two neighbours could not agree and the matter was adjudicated
by a tribunal, in 1968. The boundary award placed the boundary close to the
northern limit of the Great Rann, However, Pakistan was awarded some northern
projections at both end of the line and an area of about 950 sq. km in the vicinity
of Dhara Banni and Chhad Bet (Prescott, Collier and Prescott, 1977, 40-1).
One mistake that can be fatal to boundary delimitations is to define a point
in two ways that turn out to be contradictory. Bolivia and Peru defined a point
of their boundary as the confluence of the Lanza and Tambopata Rivers that
lies north of parallel 14°. The surveyors were able to find the confluence but
it was south of the nominated parallel. The western terminus of the boundary
between the Dutch Timor and Oekussi, a Portuguese outlier, was defined in the
following terms.
Proceeding from the mouth of the Noel [River] Besi, from where the sum-
mit of Pulu [Island] Batek can be sighted, on a 30° 47 NW astronomical
azimuth . . . (Deeley, 2000, 70)
Pulau Batek belongs to Indonesia. This definition was made in 1904. East Timor
and Indonesia have inherited this boundary definition and if the course of the
river has changed so that the nominated bearing is no longer correct this could
complicate negotiations. Klaas Villaneuva (Personal communication, May 2003)
has reported a possible additional complication. He observes that the term Noel
refers to a river that flows throughout the year, whereas a very narrow water
channel, lying west of the Noel Besi is properly called Nono Besi. Deeley (2000,
16) shows the Nono Besi bearing 30° 47’ NW to Pulau Batek.
122 Chapter 4
change was that the telegraph line had originally been alongside Mandarin’s Way,
but it was moved closer to the river when Cambodian rebels sought to cut the
line. In July 1896 the Governor-General of Indo-China created a committee to
enquire into the cartographic annexation and the decision was that the boundary
should be moved back to the telegraph line as it existed in 1876. A postscript to
these events resulted in the boundary north of Ha Tien being moved from the
well-preserved inner rampart of its defences to the older, decayed rampart. This
change transferred 8 sq. km from Cambodia to Vietnam.
Positional boundary disputes will usually involve smaller areas that those
involved in territorial disputes, and this can mean fewer problems for border
residents and international relations. However, positional disputes in two situations
might involve considerable areas. The first situation involves a boundary that is
defined by only a few points, and the location of one of the points is disputed.
This happened in the case of Kwaja Salar where the extreme determinations
were 80 km apart. If the boundary sectors on either side are long, the territory
involved might be considerable. The second situation could occur when the
boundary is defined as following some historic boundary and there are various
representations of its location.
In the middle of the 19th century the barren coastline between the Portuguese
presence on the Angolan coast, from Luanda to Moçâmedes, and Britain’s Cape
Colony was inhabited by small widely scattered groups of Africans. In the 1840s
the discovery of large deposits of guano on Ichabo Island in latitude 26° 17’ S
started a rush that exhausted the deposits on Ichabo and other guano islands in
three years. The birds started to return to the islands in 1847 and it was sug-
gested to the British authorities that they should take possession of the islands.
Ichabo, the largest was annexed on 21 June 1861. In 1866, the remaining guano
islands were annexed. After a series of administrative bungles control passed to
the Cape Colony (Kinahan, 1992, 107-9).
In 1874, British authorities, aware that Boers were trekking northwards, decided
that it would be unwise to allow them to establish an outlet on the west coast
of the continent. Instructions were issued to annex the port and adjacent terri-
tory of Walvis Bay. In March 1876 an explorer and trader called Palgrave was
commissioned to supply information about the resources of the coastlands north
of the Orange River and the political organization of the indigenous communi-
ties (Berat, 1990, 32). Palgrave’s report was encouraging and Commander Dyer
received instructions to proceed to Walvis Bay and annex it (Figure 4.4).
. . . hoist the British flag, to take possession of, and to declare and proclaim
the port, the settlement of Walfisch Bay, and the country immediately sur-
rounding the same for a distance inland to be determined by him [Dyer] to
be British territory. (Berat, 1990, 36)
124 Chapter 4
HMS Industry arrived at Walvis Bay on 6 March 1878. Dyer had been instructed
to consult Palgrave about the extent of territory to be claimed but Palgrave was
not in the area. On 9 March Dyer with two other officers trekked for nine hours
to Scheppmansdorf that was a mission station. Dyer explained the reason for
this expedition in his report.
There being no fresh water or grass at Walfisch Bay, I deemed it indis-
pensable that a place containing both should, if possible, be included in
the annexation and with this object I made a journey by bullock wagon to
Rooibank, taking with me the Assistant Paymaster in Charge and the Sur-
geon, the only available officers, to assist me in viewing the plateau. This
place is variously estimated at from thirteen to eighteen miles east from
Walfish Bay, but nine hours by wagon. It is an oasis thickly covered with
grass and well-watered, and is the nearest available place for water and
good pasturage for supplying the Bay. (Cleverly, 1904, 3)
Dyer was also accompanied by a local resident called Ryden. On 13 March
sailors mustered on the beach beside the flag-pole. The Union Jack was raised
and saluted by 21 guns. Dyer’s Proclamation included the following definition
of the territory of Walvis Bay.
. . . I do further proclaim, declare and make known that the said territory of
Walfisch Bay, so taken possession of by me as aforesaid shall be bounded
as follows: that is to say, on the south by a line from a point on the coast
15 miles south of Pelican Point to Scheppmansdorf; on the east by a line
from Scheppmansdorf to Rooibank, including the Plateau, and thence to 10
miles inland from the mouth of the Swakop River; on the north by the last
10 miles of the course of the said Swakop River. (British and Foreign State
Papers, 1877-8, 1177)
Four years after Walvis Bay was annexed by Britain, Adolf Luderitz from Bre-
men requested German support for any acquisitions he might make along the
coast north of the Orange River. This trader wanted to develop a port that would
enable him to avoid duties on goods imported via Walvis Bay. Luderitz first
bought Angra Pequena a small port in a bay on which the town of Luderitz now
stands. Three months later Luderitz bought the entire coast and 20 miles inland
from the Orange River to parallel 26° S. Then he bought the coast, except for
Walvis Bay from 26° S to Cape Frio, 70 nm from the Portuguese settlements.
This second purchase included Sandwich Harbour about 22 nm south of Walvis
Bay. Germany annexed the territories purchased by Ludertitz on 29 May 1884.
This meant that both countries now had an interest in delimiting the boundary
of Walvis Bay.
During the next twenty years Britain and Germany appointed three joint com-
missions, each composed of two members. None was able to establish a boundary
satisfactory to both sides, but all contributed to the identification of the major
International Boundary Disputes 125
obstacles that were submitted to arbitration in 1909. The two countries agreed
to the appointment of the first joint commission in 1885. It consisted of Mr Jus-
tice Shippard, from the Administrator’s Office in British Bechuanaland and Dr
Beiber, the German Consul-General in Cape Town (Shee, 1978, 25). Shippard
reported in September 1885 that he had advised Beiber of his interpretation of the
proclaimed boundary. Shippard had no doubt that the south and east boundaries
should include the whole commonage of Rooibank or Scheppmansdorf as far as
Ururas (Shee, 1978, 25). However, the commissioners sent a joint letter to their
governments correcting the spelling of two place names in Dyer’s Proclamation
and the Letters Patent of December 1878. It was noted that ‘Scheppmansdorp’
should be ‘Scheppmansdorf’ and ‘Rooibank’ should be ‘Rooikop’. While the
change to Rooikop was necessary both Dyer’s Proclamation and the Letters
Patent refer to Scheppmansdorf. The Commissioners then agreed that further
consideration of the issues should be deferred until the boundary had been sur-
veyed and demarcated by Wrey. Phillip B.S. Wrey was a government surveyor
in Cape Town and he completed his survey in the second half of 1885. His map
was dated 12 December 1885 and his report bore the date 14 January 1886.
The map was drawn at a scale of 1:120,000. The international boundary was
demarcated by beacons at Points B – M. Pillars were also erected at Points A
and N for surveying purposes. The only two points accurately defined by Dyer
were ‘. . . a point on the coast 15 miles south of Pelican Point . . .’ and ‘. . . on the
north by the last 10 miles of the course of said Swakop River’. The map showed
the two important rivers Khuisep [Kuisip] and Tsuachaup [Swakop]. Roads to
Rooibank and the interior are shown, together with four wells called Wortel,
Dorop, Frederick’s and Surveyor’s Dams, between the lagoon to the north and
the boundary heading southwest from Point B. Isolated prominent hills called
‘Kop’ were identified and vegetated areas had tree-like symbols with the words
Tamarisk Trees, Anna Trees and Nara. The seeds of the Anna Trees were a main
supply of food for stock.
At Rooibank and Scheppmansdorf Wrey was faced with Dyer’s reference to
‘. . . including the Plateau . . .’ within the territory of Walvis Bay and reached the
following conclusion.
The word plateau as used in the Proclamation is a misnomer, for the whole
extent of ground between Rooibank and Ururas, excepting the plain and
sand hills, lies in the riverbed. I have therefore taken the word to signify
the area of ground covered by river water in years of flood. This extends
from the plain on the one side [north] to the clearly defined edge of the
sand hills on the other [south], and I have made this edge [sand hills]
the boundary between beacons D and F crossing from thence to G. (Cape
Colony, 1887, 125)
Wrey’s map also identified the location and description of the boundary and
survey markers at Points A-N.
126 Chapter 4
Once Beiber received a copy of Wrey’s map he raised three major objections
in June 1886. First it was pointed out that there was no mention of Ururas in
Dyer’s Proclamation. Then it was urged that it was hardly reasonable to call the
riverbed a plateau. Finally there was reference to a murder that had taken place
on the riverbed north of Scheppmansdorf. Jan Jonker Afrikander had hanged a
mountain Damara near Scheppmansdorf in March 1885 and it was alleged that
the Walvis Bay magistrate has requested the Germans to maintain order on their
territory. Shippard replied briefly along the lines that Wrey had acted in accor-
dance with his instructions in interpreting Dyer’s boundary definition.
There was then a period of three years when there were flurries of correspon-
dence between the Colony and British governments, between German adminis-
trators and the government in Berlin, and between intermediate contacts of the
British and German governments.
Curiously some senior people in the Colony and the Colonial Office had doubts
about the strength of the British case. Torrens the Acting High-Commissioner in
Cape Town expressed reservations in November 1886. He found Wrey’s inter-
pretation of the word ‘plateau’ unusual; and considered that it would be difficult
to resist any German request for a new delimitation of the boundary. Bramston
of the Colonial Office in London wrote to the Prime Minister’s office drawing
attention to weaknesses in Britain’s case. First in his report Dyer mentions ‘the
plateau and Scheppmansdorf to the southeast’ showing that the plateau was north-
west of that place. Second Wrey’s line deviated from the British boundary that
had long been shown on British Admiralty charts.
The German authorities hammered three main points in support of Germany’s
claim. First it was argued that a plateau could not be a riverbed. Second it claimed
that the murder committed by Jan Jonker Afrikander in the riverbed had been
recognised as a German responsibility by the Walvis Bay authorities. Third the
phrase ‘. . . the plateau and Scheppmansdorf to the southeast . . .’ used by Dyer in
his report proved that the plateau was northeast of Scheppmansdorf.
The British Prime Minister’s Office was able to overcome the fears expressed
by the Colony’s Government and the Colonial Office by January 1888. First a
letter from Dyer in September 1887 made it clear that acting on the advice of
missionaries, local settlers and indigenes his principal desire was to include the
grazing areas between Scheppmansdorf and Ururas for the African population.
Second it was pointed out that the phrase ‘. . . the plateau and Scheppmansdorf
to the southeast . . .’ referred to the two features being located in the southeast
portion of the territory of Walvis Bay. The phrase should not be interpreted to
refer to their positions relative to each other. Finally after noting that the murder
occurred before any boundary was demarcated, careful fieldwork had established
that the mountain Damara was hanged beyond the boundary marked by Wrey.
The second joint commission involved Colonel Philips and Doctor Goering,
who visited Walvis Bay in January 1889. Philips issued his report before the end
of the month and Goering followed in mid-June. Goering had the benefit of a letter
International Boundary Disputes 127
from Captain Dyer in April that explained, again, his reasons for including the
area between Scheppmansdorf and Ururas within his boundary. Once again the
commissioners disagreed about Wrey’s interpretation of Dyer’s Proclamation.
In 1890 Britain and Germany agreed on a treaty settling their various disputes
in Africa with the exception of the southern boundary of Walvis Bay (Hertslet,
1909, 899). This matter was reserved for arbitration unless it was solved within
two years. Fourteen years later the matter had not been solved and not submitted
to arbitration. The governments made one final attempt to reach agreement through
a joint commission. Britain appointed John Cleverly, formerly a magistrate in
Walvis Bay and the German delegate was a surveyor called von Frankenberg.
The joint reports of the two commissioners raised only one new matter of con-
tention. Von Frankenberg made two new proposals for the definition of Wrey’s
Point B, at the western end of the southern boundary. First he believed that the
beacon located on Pelican Point should be moved as far north to the edge of the
point as the sand permits of a secure foundation. Second the 15 miles specified
in Dyer’s Proclamation should be English miles not nautical miles. An English
mile measures 1609 metres; a nautical miles measures 1852 metres. If English
miles were used and they were measured from the beacon that existed on Pelican
Point, the new Point B would be 3.6 km north of Point B defined by Dyer and
Wrey. The new line between Point B on the coast and Point C at Scheppmansdorf
would deprive Walvis Bay of about six square miles that contain good water at
Surveyor’s Dam and some vegetation.
Cleverly opposed the changes suggested by Von Frankenberg. The first beacon
on Pelican Point was erected by the crew of HMS Leven in 1825 at latitude
22° 52’ 30” S. In 1878 the crew of HMS Danae erected a new beacon ‘. . . on
the NW corner of the sandspit, as a guide to vessels rounding Pelican Pt. which
otherwise is sometimes difficult to see at 2 miles’ (Kinahan, 1992, 131). Its loca-
tion was at 22° 53’ 32” S, that is one nautical mile south of the original beacon.
That position had been recorded on charts and in sailing directions in the period
1878-1904. Cleverly had no doubt that a naval officer would measure coastlines
in nautical miles and that no other miles had been mentioned in the 26 years
since Dyer’s Proclamation.
After the failure of this third joint commission to produce an agreement the
two countries agreed to submit the matter to arbitration. On 30 January 1909 the
King of Spain was asked to appoint an arbitrator, which he did on 7 March 1909.
Joaquin Fernandez Prida was a Senator and Professor of Law at the University
of Madrid. It was arranged that the British and German Memoranda would be
submitted by 29 November 1909, and their Replies to those documents would be
received by the Arbitrator on 30 July 1910. During the period December 1910
to January 1911 Prida visited Walvis Bay accompanied by the German Commis-
sioner Von Frankenberg and the British Commissioner Lansdown. Prida’s Award
was completed on 23 May 1911 and it was published in the British Parliament
in October 1911 (Africa, 1911).
128 Chapter 4
The Award set out the main points in the Memoranda and the following Replies.
Then Prida identified the two principal questions to be answered.
Considering that there are two fundamental questions which it is necessary
to examine in this award: (1) Whether the southern limit of the territory of
Walfisch Bay ends in the proximity of the mission church of Scheppman-
sdorf, or, on the contrary, whether it should be prolonged to Ururas in
accordance with Mr Wrey’s survey; (2) whether this southern boundary
should begin at a point 15 nautical miles or 15 statute miles from Pelican
Point. (Africa, 1911, 22-3)
The Arbitrator accepted that both parties recognized that the answer to the first
question turned on the interpretation of the phrase ‘. . . including the plateau’. He
focussed first on what the Proclamation or its author Captain Dyer understood by
the word ‘plateau’. Was the plateau the high plain of the Namib or the valley of
the River Kuisip between Scheppmansdorf and Ururas. Prida noted that even if
‘plateau’ is considered to be ‘a high plain’ it is a relative term, and therefore the
greater elevation of the Namib was not sufficient to eliminate the Kuisip valley.
He also noted that the phrase ‘. . . on the east by a line from Scheppmansdorf to
the Rooibank, including the Plateau, . . .’ means that the plateau must be included
by the eastern boundary and that it was in the southeast corner of the annexed
land (Africa, 1911, 23).
Prida also noted the weakness in the German interpretation that the phrase
about the plateau must mean it lay northeast of Scheppmansdorf, because it
could also be interpreted to mean that both features lay in the southeast corner
of the territory. The Arbitrator also noted that if Dyer had been referring to the
Namib he could have reached that territory without traversing the long route to
Scheppmansdorf, and that Dyer’s reference to an ‘oasis’ had to refer to the word
‘plateau’. Germany claimed that Dyer’s phrase of the plateau being ‘. . . above
Rooibank . . .’ disqualified the Kuisip valley from consideration. However Prida
dismissed this view.
. . . the statement in this last report [by Dyer] is perfectly applicable to
the bed of the Kuisip, which rises constantly and gradually towards the
interior from the coast and runs on above Rooibank within the zone in
dispute . . . (Africa, 1911, 25)
The Arbitrator then came to the first conclusion relative to the decision about
whether the boundary should be extended to Ururas.
Considering that if the previous arguments are admitted, and therefore that
with more or less propriety as to the use of the word but with no uncer-
tainty as to the intention, what is called ‘plateau’ in the Proclamation of
Annexation is part of the valley of the Kuisip, the principal problem still
remains undecided, namely, that relative to its extent and limits, in other
words, whether the said plateau should be understood as ending near the
International Boundary Disputes 129
CONCLUSION
REFERENCES
Africa. No. 1, 1911. Translation of the Award of Don Joaquin Fernandez Prida,
American Journal of International Law, 1932, ‘Judicial decisions involving
questions of international law. France-Mexico’, 390-4.
Ancel, J., 1936. Géopolitique, [Geopolitics], Delagrave: Paris.
Ancel, J.J., 1938. Les frontières [Boundaries)], Armand-Colijn: Paris.
Arbitrator in the matter of the southern boundary of the Territory of Walfisch
Bay, HMSO: London
Barotse boundary file 1903. Archives of Rhodesia, Salisbury, seen in 1960.
Beazley, P.B., 1987. Maritime limits and baselines: a guide to their delimitation,
The Hydrographic Society: Dagenham.
Berat, L., 1990. Walvis Bay: the last frontier, Radix: Johannesburg.
Berdichevsky, N., 1999. The German-Danish border: a successful resolution of
an age old conflict or its redefinition’, Boundary and Territorial Briefing,
volume 3 (7).
Boggs, S.W., 1940. International boundaries: a study of boundary functions and
problems, Colombia University Press: New York.
Bowman, I., 1923. The new world, World Book Company: New York.
British and Foreign State Papers, 1877-8. ‘British Proclamation taking possession
of the Port or Settlement of Walfisch Bay, 12 March 1878’ 69.
Brownlie, I., 1979. African boundaries: a legal and diplomatic encyclopaedia,
Hurst: London.
Cape Colony, 1887. Further correspondence respecting the settlement at Angra
Pequena on the southwest coast of Africa, Cape Town.
Chhak, S., 1966. Les frontières du Camboge, Paris.
Cleverly, 1904. Anglo-German boundary enquiry – Walvis Bay, Office of the
Resident Magistrate, Walvis Bay.
International Boundary Disputes 133
Marchant, A., 1944. Boundaries of the Latin American Republics: and annotated
list of documents, 1493-1943, Department of State: Washington DC.
McCaffrey, S.C., 2001. The law of international watercourses: non-navigational
uses, Oxford University Press: Oxford.
Naval Intelligence Division, 1942. France, volume II, HMSO: London
———, 1944. Greece, volume I, HMSO: London.
———, 1945. China Proper, volume II, HSMO: London.
Permanent Court of International Justice, 1933. Legal status of Greenland,
Twenty-sixth session, PCIJ Reports, Series A/B, No. 53, 1-123
Pounds, N.J.G., 1951. ‘The origin of the idea of natural frontiers in France’,
Annals, Association of American Geographers, 41, 146-57.
———, 1956. ‘France and ‘Les limites naturelles’ from the seventeenth to twen-
tieth centuries’, Annals, Association of American Geographers, 44, 51-62.
Prescott, J.R.V., 1971. The evolution of Nigeria’s international and regional
boundaries, 1861-1971, Tantalus Research: University of British Colombia.
———, 1975. Map of mainland Asia by treaty, Melbourne University Press:
Melbourne.
Prescott, J.R.V., Collier, H.J. and Prescott, D.F., 1977. Frontiers of Asia and
Southeast Asia, Melbourne University Press: Melbourne.
Prescott, J.R.V., 1998. ‘Contributions of the United Nations to solving boundary
and territorial disputes, 1945-1997’, in The United Nations at work, edited
by M.I.Glassner, Praeger: Connecticut, 239-84.
Ramirez-Faria, C., 2001. The historical dictionary of world political geography,
Palgrave: New York.
Rebert, P., 2001. La Gran Linea: Mapping the United States-Mexico boundary,
1849-1857, University of Texas Press: Austin.
G. Reilly, 1999. ‘Gibraltar: sovereignty disputes and territorial waters’, Boundary
and Security Bulletin, 7 (1), 67-81.
Republic of Suriname, 2005. Republic of Guyana v. Republic of Suriname, Coun-
ter-Memorial of the Republic of Suriname, volume II, Annex 5, 8.
Ryder, C.H.D., 1926. ‘The demarcation of the Turco-Persian boundary of 1913-
14’, Geographical Journal, 66, 99-114.
Sanguin, A.L., 1983. ‘La bordure franco-italienne des Alpes-Maritimes ou les
consequences de la modification d’une frontière international’ (The Franco-
Italian border in the Alpes-Maritimes or the consequences of change in an
international boundary), Mediterranée, 47, 17-25.
Schofield, R., 1993. Kuwait and Iraq: historical claims and territorial disputes,
Royal Institute for International Affairs: London.
Shee, L.D.A., 1978. The annexation of Walvis Bay and the protracted dispute
over the.southern boundary of the enclave, Research Essay, for the degree of
BA (Honours) in the History Department of the University of Cape Town.
Smith R.W. and D.A. Colson, 1993. ‘Cook Islands-United States’, in International
maritime boundaries, edited by J.I. Charney and L.M. Alexander, Martinus
Nijhoff: Dordrecht.
International Boundary Disputes 135
Samuels, M.S., 1982. Contest for the South China Sea, Methuen: New York.
Smith, R.W. and Colson, D.A., 1993. ‘Cook Islands-United States (American
Samoa), in International Maritime Boundaries, volume I, edited by J.I.
Charney and L.M. Alexander, Martinus Nijhoff: Dordrecht, 985-9.
Smith, S.C., 1997. ‘Conflicting commitments: Britain and the Portuguese pos-
sessions in India. 1847-1961’, South Asia: Journal of South Asian Studies,
XX (1), 17-34.
Sotomayor, A., 1941. Posicion internacional del Peru [The international position
of Peru], Imprenta Torres Aguirre: Lima.
Spadi, F., 2001. ‘Navassa: legal nightmares in a biological heaven’, Boundary
and Security Bulletin, 9 (3), 115-30.
St John, R.B., 1994a. ‘The boundary between Ecuador and Peru’, Boundary and
Territory Briefing, 1 (4).
———, 1994b. ‘The Bolivia-Chile-Peru dispute in the Atacama Desert’, Bound-
ary and Territory Briefing, 1 (6).
———, 1999. ‘The Ecuador-Peru boundary dispute: the road to settlement’,
Boundary and Territory Briefing, 3 (1), 1-65.
Taberd, J.L., 1837. ‘Note on the geography of Cochin-China’, Journal of the
Royal Asiatic Society, Bengal, 2, 737-45.
Temperley, H.W.V., 1920-1. History of the Peace Conference of Paris, 5 volumes,
Hodder and Stoughton: London.
The Geographer, 1976. ‘Guatemala-Honduras’, International Boundary Study,
no. 157, Department of State: Washington DC.
———, 1978. ‘Bolivia-Paraguay boundary’, International Boundary Study,
no. 165, Department of State, Washington DC.
Triggs, G.D., 1983. Australia’s sovereignty in Antarctica: the validity of Aus-
tralia’s claim at international law, Programme in Antarctic Studies, 61,
Melbourne University: Melbourne.
Vietnam Foreign Ministry, 1975. White paper on the Hoang Sa [Paracel] and
Truong Sa [Spratly] Islands, Government Printer: Saigon.
Wagner de Reyna, A., 1964. Historica diplomatica del Peru, 1900-1945, [Dip-
lomatic history of Peru], vols I-II, Ediciones Peruanas: Lima.
Ward, A,W., G.W. Prothero and S. Leathes, 1924. The Cambridge modern his-
tory atlas, University Press: Cambridge.
Ward, F.K., 1932. ‘Explorations on the Burma-Tibet frontier’, Geographical
Journal, 80, 465-83.
Weigend, G.G., 1950. ‘Effects of boundary changes in the south Tyrol’, Geo-
graphical Review, 40, 364-75.
Zamoyski, A., 2004. 1812 Napoleon’s fatal march on Moscow, Harper Peren-
nial: London.
5. INTERNATIONAL LAW AND TERRITORIAL BOUNDARIES
1. INTRODUCTION
Few issues strike so profoundly at the heart of national sovereignty than the deter-
mination of conflicting territorial claims by states and of the land boundaries that
lie between them.1 It is therefore unsurprising that territorial and boundary disputes
have been a primary stimulant to the evolution of international law itself. The
major international courts, the Permanent Court of International Justice (PCIJ) and
the International Court of Justice (ICJ), along with international arbitral tribunals,
have identified and articulated legal principles to delimit boundaries and decide
on the relative strength of disputed claims; principles that derive from custom-
ary and treaty-based law and have a general application well beyond individual
disputes. The docket of the ICJ has been and continues to be occupied by what
Oxman (2006:830) has termed the irresistible ‘territorial temptation’ both to land
claims and to an extended maritime jurisdiction. Indeed, it has been estimated that
about one third of the contentious cases before the ICJ have dealt with boundary
disputes of one kind or another (Ratner 1996:814).
This chapter is concerned with the legal principles applicable to territorial,
as distinct from maritime, boundary delimitations that have evolved from the
decisions of international courts and tribunals. While the aim is to search for
the core legal principles of international law, it should be admitted at the outset
that many boundaries established during the colonial period were drawn by ‘a
blue pencil and a rule’ (Anene 1970:3). As the British Foreign Minister Lord
Salisbury was reported to have said in The Times on 7 August 1890:
We have been engaged . . . in drawing lines upon maps where no white man’s
feet have ever trod; we have been giving away mountains and rivers and
lakes to each other, but we have only been hindered by the small impedi-
ment that we never knew exactly where those mountains and rivers and
1
Parts of this chapter build upon and update materials previously published in
G. Triggs, International Law: Contemporary principles and Practices, (Lexis Nexis
Butterworths, 2006) Chapter 5.
138 Chapter 5
• nature of a boundary,
• principle of uti possidetis,
• distinction between delimitation and a demarcation,
• modes of acquisition of territorial sovereignty,
• recognition, acquiescence, estoppel, geographical and political factors,
• inter-temporal law and the critical date,
• right to self-determination, and
• common heritage of mankind.
Since the fundamental principles of international law with respect to boundary
delimitation were developed by the PCIJ and arbitral tribunals during the 1930s,
there have been significant decisions by the successor to the PCIJ, the ICJ, and
other arbitral tribunals, over the last 20 years that provide contemporary examples
of judicial practice. Case studies are included of the decisions of the ICJ in
Benin/Niger (2005) and the Boundary Commission in Eritrea-Ethiopia (2002).
For the most part, territorial boundaries have been settled by routine diplomatic
negotiations. The ‘tea and macaroons’ approach to drawing boundaries is viv-
idly described by Harold Nicholson in his book Peacemaking 1919 (1933, cited
in McCorquodale & Pangalangan 2001:869), when he observes the Versailles
Conference after the First World War on 8 May 1919:
During the afternoon [at the Quay d’Orsay] . . . the fate of the Austro-Hungar-
ian Empire is finally settled. Hungary is partitioned by these five distinguished
gentlemen – indolently, irresponsibly partitioned – while the water sprinkles
on the lilac outside – while the experts watch anxiously – while AJB, in the
intervals of dialectics on secondary matter, relapses into somnolence – while
Lansing draws hobgoblins on his writing pad – while Pichon, crouching
in his large chair, blinks owlishly as decision after decision is actually
recorded . . . They begin with Transylvania, and after some insults flung like
tennis balls between Tardieu and Lansing, Hungary loses her south. Then
Czechoslovakia, and while the flies drone in and out of the open windows
Hungary loses her north and east . . . Then the Jugo-Slav frontier, where the
Committee’s report is adopted without change. Then tea and macaroons.
That the boundaries thereby agreed should prove to be both unstable and impre-
cise over the following 90 years is hardly surprising if this accurately describes
the process. Once agreed, however, the terms of settlement will usually be
reflected in a formal treaty between the respective territorial sovereigns. For this
International Law and Territorial Boundaries 141
reason, boundaries, and sovereignty in the surrounding areas, will often depend
upon interpretation of the terms of the treaty which forms the basis of title for
the state parties.
Disputes over territorial sovereignty and over the precise location of a boundary
are a common feature of international affairs. States will usually resolve such
disputes through diplomacy at a bilateral or multilateral level. On occasion, they
may resort to third party negotiations or to mediation. The Good Offices of the
United Nations Secretary-General may be offered or states may agree, by special
agreement or compromis, to the appointment of an ad hoc court of arbitration.
Where states have accepted the jurisdiction of the ICJ, it is also possible to sub-
mit the dispute to that Court for determination according to international law.
Territorial and boundary disputes, particularly with respect to maritime delin-
eation, have dominated the work of the Court throughout its history and have
been the vehicle for the articulation of much of the Court’s jurisprudence to the
significant benefit of general international law.
2. UTI POSSIDETIS
The overriding importance of the need for stability in international relations has
been confirmed in recent jurisprudence of international tribunals to the effect
that uti possidetis cannot be ‘ousted’ by the principle of self-determination.
This is so even though the ICJ in the 1995 East Timor case recognised the erga
omnes character of self-determination. Controversially, the European Community
Arbitration Commission on Yugoslavia (Badinter Commission) has also applied
uti possidetis beyond the colonial context in relation to the right of the Serbian
population of Croatia and Bosnia-Herzegovina to self-determination. The Com-
mission concluded in Opinion No. 2 that ‘the right to self-determination must not
involve changes to existing frontiers at the time of independence (uti possidetis
juris) except where the States concerned agree otherwise’ (at 168). In Opinion
No 3 (at 171) the Commission confirmed its earlier view that:
Except where otherwise agreed, the former boundaries become frontiers
protected by international law. This conclusion follows from the principles of
respect for the territorial status quo and, in particular, from the principle of
uti possidetis. Uti possidetis, though initially applied in settling decolonization
issues in America and Africa, is today recognized as a general principle.
The Badinter Commission appears to have concluded that only a strict application
of uti possidetis could prevent the attacks by one part of the former Yugoslavia
upon another. United though the ICJ and EC Commission have been in considering
that the principle of uti possidetis prevails over that of self-determination, they
have recognized that the rights of the peoples involved remain protected by the
‘peremptory . . . norms of international law [that] require respect for minorities’
(Opinion No. 2 at 168). Of contemporary concern, for example, are the rights of
the Dniestrians in the Republic of Moldavia and Armenians in Nagorno-Karabakh.
It is doubtful, however, that the rights of minorities yet amount to a peremptory
norm at customary international law. However, they will be protected by interna-
tional instruments such as the 1966 International Covenant on Civil and Political
Rights and the European Community Guidelines on the Recognition of New States
in Eastern Europe and the Soviet Union of 1991 (Marston 1991:559).
Commentators have been critical of the overly simplistic ‘default’ rule of uti
possidetis that, in effect, mandates the conversion of administrative boundaries
into international borders (Ratner 1996:590). While the rationale for the rule-
predictability, avoidance of conflict over unclear borders and simplicity – is oft
repeated, an automatic application of uti possidetis can potentially encourage
separatist movements, compound historical injustices and deny human rights
such as the right to self-determination and the rights of minorities. While the
retention of colonial boundaries may be justified on pragmatic grounds, to apply
uti possidetis to contemporary issues such as the dissolution of the former Yugo-
slavia can have the effect of blocking attempts to redefine boundaries to respond
to the needs and wishes of local peoples and prevent the recurrence of age old
injustices. The wisdom of the Badinter Commission’s adoption of uti possidetis
to determine the internal borders may thus be doubted.
International Law and Territorial Boundaries 145
When applying the uti possidetis principle in the context of treaties establishing
the boundary between two states, tribunals have looked at the ‘photograph of
the territory’ at the moment of independence. The principle thus determines the
date, known as the critical date, at which the factual evidence is to be assessed
(Frontier Dispute case, p. 568). The ICJ has not however confined analysis of
the uti possidetis principle to the application of title from the relevant treaties.
Rather, the Court has applied the principle to all evidence that pertains to deter-
mining which claimant has the better title, including acts of effective occupation
and acquiescence.
The uti possidetis principle has some limitations in resolving boundary disputes.
It will not necessarily identify the precise location of the frontier. Formal title at
the time of independence, for example, may not be sufficient. The Chamber of
the ICJ in the Land, Island and Maritime Frontier Dispute case in 1992 observed
that the boundaries could not accurately be determined on the basis of colonial
titles and effectivités in 1821. Rather, the Chamber looked to other evidence of
effective occupation and control such as the absence of protest and implications of
acquiescence by El Salvador and Honduras. Moreover, ‘post-colonial effectivitiés’
were considered as relevant evidence. The relationship between effectivités and
the principle of uti possidetis is considered in further detail below.
The following case study of the Frontier Dispute case (Burkina Faso/Mali)
illustrates the application of the legal principles.
The origins of the frontier dispute between Burkina Faso (formerly Upper
Volta) and Mali (formerly French Sudan) lay in French colonial decrees prior
to their independence in 1960. By 1922, France had created three colonies
of Upper Volta, French Sudan and Niger. By a French presidential decree of
5 September 1932, Upper Volta was divided between French Sudan and Niger.
Upper Volta was, however, recreated in 1947 on the basis that the boundaries
were to be those that had existed prior to the division in 1932. While Mali and
Burkina Faso were able to agree on the demarcation of most of the boundary
between them, part of the frontier remained in dispute. A Mediation Commis-
sion, established in 1975, failed to reach a final solution. Mali and Burkina
Faso then agreed, by Special Agreement of 16 September 1983, to submit
the dispute to a five-member Chamber of the ICJ. The Chamber was asked
to delimit the frontier ‘based in particular on respect for the principle of the
intangibility of frontiers inherited from colonization’ (para 19, p. 564).
Armed conflict broke out in December 1986 and the Chamber indicated
provisional measures, calling on the states not to extend the dispute or prejudice
the rights of the other state and to withdraw their armed forces to lines to be
146 Chapter 5
Unusually, in the Eritrea-Ethiopia case, the parties agreed to give the delimita-
tion Commission the additional mandate to achieve a final and binding ‘expedi-
tious demarcation’ as a second phase of its work. (Art 4 (13) of the December
Agreement.) As Shaw (2007:786) points out, this ‘interesting experiment’ in
combining the tasks of delimitation and demarcation was stymied by the lack of
cooperation between the parties between 2002 and 2006 when the Commission
ended its work. Over that time, the Commission issued various Demarcation
Directions and Instructions in an attempt to complete its work in the face of
unilateral troop and population movements across the delimited line. The Com-
mission took the strict view that it did not have the power to vary the delimitation
line as determined in the earlier stage of its deliberations, concluding that if the
boundary ‘runs through and divides a town or village, the line may be varied
only on the basis of an express request agreed between and made by both Par-
ties’ (Demarcation Directions of 8 July 2002, para 14A). Moreover, Ethiopia and
Eritrea had expressly agreed to deny the Commission any right to decide matters
ex aequo et bono. The Commission thus had no power to vary the boundary to
reflect local human needs. It concluded (Observations of the Eritrea-Ethiopia
Boundary Commission, 21 March 2003, para 8) that
a demarcator must demarcate the boundary as it has been laid down in the
delimitation instrument, but with a limited margin of appreciation enabling
it to take account of any flexibility in the terms of the delimitation itself or
of the scale and accuracy of maps used in the delimitation process, and to
avoid establishing a boundary which is manifestly impracticable.
The President of the Commission, responding to Ethiopia’s criticism of the
Commission’s demarcation, defended the implementation of the treaty-based
boundary, observing that:
Where villages have sprung up or spread in recent times, and in so doing
transgress boundaries previously established by older treaties, it is fully
consistent with international law for the treaty-based boundary to be main-
tained and for the resolution of any consequential human problems to be
left for the Parties to resolve by agreement . . . [T]hat result is precisely what
the ICJ decided, in comparable circumstances, in its recent Judgment in the
Cameroon v. Nigeria case. (President’s letter of 7 October 2003, para 10)
While the approach taken by the Commission and supported by dicta of the
ICJ seems overly restricted by the earlier delimitation, some sympathy for
the magnitude of the task is indicated where local populations move across the
delimited lines, possibly in an attempt to influence the final demarcation. Some
flexibility was nonetheless adopted by the Commission in the Eritrea-Ethiopia
case where the line as delimited proved to be ‘manifestly impracticable’. Here,
demarcations were to be issued at a later date to respond to this and other practi-
cal difficulties.
148 Chapter 5
It has been observed that the delineation of boundaries is, in practice, necessarily
an outcome of the determination of territorial sovereignty. For this reason, it is
useful to examine the legal principles governing the means by which a state can
acquire sovereignty. These principles lie at the core of traditional international
law for they determine the framework within which the power of the state is
exercised. (Triggs 2006:210) Title to territory is fundamental to the idea of state
sovereignty and the extent of title to territory typically defines the jurisdictional
reach of the state.
The principles of acquisition were developed during the period of colonial
expansion from the sixteenth century and broadly reflect the needs of the
nineteenth century colonial powers for order and stability. Today, most of the
earth’s territory has been claimed and is recognized as subject to state sover-
eignty. Exceptions, beyond Antarctica, are few. Contemporary disputes generally
concern apparently insignificant islands or territories that have strategic value
or the promise of natural renewable and non-renewable resources. Unresolved
territorial disputes include the Spratly Islands in the South China Seas, claimed
in total or in part by five nations, and the Matthew and Hunter Islands claimed
by Vanuatu and France. Of those submitted to international adjudication, the ICJ
has began its deliberations in the dispute between Singapore and Malaysia in
the Sovereignty over Pedra Branca/Pulau Batu Puteh, Middle Rocks and South
Ledge case in November 2007. Of note is the fact that this is a rare example of
recourse to the ICJ by a state from the Asian region.
International legal authorities typically identify five modes of acquiring ter-
ritorial sovereignty, which may in theory be presented as distinct means of
acquisition. They are:
• occupation of terra nullius, that is, territory that is owned by no one,
• prescription, by which title is gained by possession adverse to the abstract
titleholder,
International Law and Territorial Boundaries 149
It is trite to observe that where there are two rival claimants to territorial sover-
eignty, an international tribunal in an adversarial proceeding will make a decision
in favor of one or the other of them (Triggs 2006:214). Less often observed is
that a judgment determines only which of these states has the stronger or bet-
ter title, the practical effect of which is to confer title erga omnes. In principle,
determination by a tribunal that title rests with one state rather than another cannot
foreclose the rights of a third state. It would always be open to any other state
to contest the outcome of an adversarial proceeding. Moreover, in the Frontier
Dispute case between Burkina Faso and Mali (map 5.6), the ICJ pointed out that
its delineation of a land boundary would not be opposable to Niger, a neighboring
state, because no decision of that court has binding force other than between the
parties (para 50, p. 580, citing article 59 of the ICJ Statute).
An award or decision of an international tribunal cannot, for these reasons, be
a formal means of acquiring title. In practice, however, it is true that no state has
contested a title to territory that has been recognized by a tribunal in this way.
Thus the victorious state emerges from an international adversarial determination
with the full trappings of absolute sovereignty. In the Minquiers and Ecrehos
case, for example, the ICJ equated possession over the islands in dispute in the
English Channel with sovereignty and considered that its task was ‘to appraise
the relative strength of the opposing claims to sovereignty’ (p. 67). As this case
illustrates, many of the classic decisions on sovereignty have concerned apparently
obscure islands, miles from the claimants, where a more compelling conclusion
might have been that neither claimant satisfactorily established sovereignty and
that the territory remained terra nullius. The readiness of international tribunals
to ‘discover’ title may be explained by the terms of the ‘compromis’ or agreement
between the disputing states by which the jurisdiction of the court was accepted
and the need to promote finality, stability and effectiveness in international rela-
tions (Triggs 2006:215).
The jurisprudence that has been build upon the Island of Palmas decision has
proved to be largely uncontroversial, is oft-cited and demonstrates how little was
required to meet the flexible and comparative standard for title to be recognized
(Triggs 2006:217) The views of Judge Huber, the sole arbitrator, on effective
occupation have been accorded an almost oracular importance and remain per-
suasive authority. The factual background to the legal dispute between the US
and the Netherlands arose in 1906 with respect to sovereignty over an isolated
island located between the Philippines and Indonesia. Palmas Island had been
ceded to the US by Spain under the Treaty of Paris in 1889. The US’s claim
was thus derivative from the earlier Spanish title based upon original discovery
and subsequent control over the island. The Netherlands argued, to the contrary,
that it had established a prior title through treaties negotiated on its behalf by
the Dutch East India Company with the princes and chieftains of the island.
These conflicting claims were submitted to the arbitration by Special Agreement
requiring the arbitrator to find in favor of one claimant or the other and thereby
to ‘terminate’ the dispute.
Judge Huber avoided the language of ‘effective occupation’, finding that
(p. 840):
Manifestations of territorial sovereignty assume, it is true, different forms,
according to conditions of time and place. Although continuous in principle,
sovereignty cannot be exercised in fact at every moment on every point of
territory. The intermittence and discontinuity compatible with the mainte-
nance of the right necessarily differ according as inhabited or uninhabited
regions are involved. . . .
[T]he actual, continuous and peaceful display of State functions is in case
of dispute the sound and natural criterium of territorial sovereignty.
Having set out the legal principles, Judge Huber then examined the historical
evidence of sovereignty prior to the date of Spain’s purported cession. He found
International Law and Territorial Boundaries 153
that although the island was isolated, it was permanently inhabited so that it was
‘impossible that acts of administration could be lacking for very long periods’
(p. 855). On the evidence there was no trace of Spanish activities, but the Neth-
erlands had demonstrated a peaceful and continuous display of authority over
the island such that it was a ‘vassal State’ of the Netherlands (p. 867). Over
the next five years or so prior to the Treaty of Paris, the Dutch authorities had
taxed the local people and distributed the Dutch coat of arms and flags on the
island. Judge Huber considered that sovereignty could be founded on such scant
evidence in the ‘epoch immediately preceding the rise of the dispute’ (p. 870). In
concluding that the Netherlands could demonstrate the better title, Judge Huber
emphasized that no state had contested the existence or protested against the
exercise of territorial rights by the Netherlands. Moreover, there were no acts of
sovereignty by another state to counter-balance the weight of the manifestations
of sovereignty by the Netherlands. In these circumstances the ‘unchallenged
acts of peaceful display of Netherlands sovereignty in the period from 1700 to
1906 . . . may be regarded as sufficiently proving the existence of Netherlands
sovereignty’ (pp. 870-871).
The Island of Palmas case illustrates each of the elements that have proved
typical of determinations of title by subsequent international tribunals. Title was
awarded over an island which, although populated, required relatively nominal
acts of sovereignty; the territory in issue was isolated and little known; relevant
evidence of sovereign functions was that close to the time of the dispute; the
absence of protest by other states or of any contrary evidence was influential;
and the arbitrator recognised that his task was to assess the relative strengths
of the titles invoked by each state. While it was reasonable for Judge Huber to
conclude that manifestations of sovereignty could vary depending upon the nature
of the territory and the needs of its people, the Island of Palmas case rested on
such slight evidence as to suggest that neither state had properly established
sovereignty. Judge Huber appeared to concede this point when he suggested
that limited acts of state authority on behalf of the Netherlands could create an
‘inchoate title for completing the conditions of sovereignty’ (p. 870).
later when a Mexican gun-boat visited the Island, sparking a diplomatic row in
which Mexico claimed sovereignty based on prior Spanish discovery. The sole
arbitrator, King Emmanuel III of Italy, found (pp. 393-394):
It is beyond doubt that by immemorial usage having the force of law, besides
the animus occupandi, the actual, and not the nominal, taking of possession
is a necessary condition of occupation. This taking of possession consists in
the act, or series of acts, by which the occupying state reduces to its posses-
sion the territory in question and takes steps to exercise exclusive authority
there. Strictly speaking, and in ordinary cases, that only takes place when
the state establishes in the territory itself an organization capable of making
its law respected. But this step is . . . but a means of procedure to the taking
of possession, and, therefore, is not identical with the latter. . . . Thus, if a
territory, by virtue of the fact that it was completely uninhabited, is, from
the first moment when the occupying state makes its appearance there, at
the absolute and undisputed disposition of that state, from that moment the
taking of possession must be considered as accomplished, and the occupa-
tion is thereby completed.
Recognizing that the island was terra nullius, the arbitrator found that France had
effectively occupied the island and did not have the animus to abandon it. France,
therefore, rather than Mexico, had the better title. The decision demonstrates the
minimal evidence required to satisfy the test of effective occupation of isolated
and uninhabited islands in an adversarial proceeding and is best understood as
reflecting the particular facts of the dispute.
The PCIJ considered the territorial dispute between Denmark and Norway in the
Eastern Greenland case, and substantially adopted the analysis of the arbitra-
tors in the Island of Palmas and Clipperton Island cases (Triggs 2006:219). A
dispute arose in 1931 when Norway issued a Royal Resolution placing part of
the east coast of Greenland under Norwegian sovereignty, naming it after the
fabled Eric the Red. Two days later, Denmark made an application to the court
claiming sovereignty overall of Greenland founded on continuous and peaceful
occupation, uncontested by any other state. Norway conceded Denmark’s claim
to its colonies in ice-free pockets on the west, south-east and south-west coasts
of Greenland but asserted that the rest of Greenland was terra nullius. The long
history of Norwegian activities in Greenland, beginning with its colonization in
1000 AD by Eric the Red, demonstrates that its interests were limited to fishing
and hunting. These interests could, for practical reasons, be protected at interna-
tional law only by making the more radical claim for sovereignty over areas not
settled by Denmark. As in the Palmas Island and Clipperton Island cases, the
PCIJ was required to determine which of the competing states had the better title.
This, the court agreed to do, probably to avoid a scramble to settle the interior
International Law and Territorial Boundaries 155
The PCIJ in the Eastern Greenland case identified two essential elements of
territorial sovereignty: the intent and will to act as a sovereign and the actual
exercise of such authority. It is questionable whether there is any significant dif-
ference between these two limbs of the test of effective occupation (Triggs 2006:
221). If the animus occupandi is a subjective criterion, it ultimately depends
for evidence upon objective manifestations of state authority. The requirement
of intent does have a function, however, insofar as it emphasizes the need to
demonstrate that the activities have been undertaken on behalf of the state rather
than as private acts, that is, the acts must have taken place à titre de souverain.
Private persons do not have the capacity at international law to appropriate
156 Chapter 5
The subjective element of intent also requires that the evidence is referable to the
hypothesis that sovereignty exists (Triggs 2006:222). The exercise of criminal
jurisdiction, for example, is usually exercised on a territorial basis and might
reasonably be construed as an indication of the intent to act as sovereign. By
contrast, the hydrographic surveys were rejected by the ICJ in the Minquiers and
Ecrehos case because they did not necessarily demonstrate an assertion of sov-
ereignty (pp. 70–71). The court similarly rejected evidence that a French citizen
had built a house on one of the contested islands and evidence of tidal power
plants in the area as insufficient evidence of intent (ibid.). By contrast, a powerful
indication of the intent to act as sovereign is the enactment of legislation appli-
cable to the disputed territory. The promulgation and enforcement of Norway’s
International Law and Territorial Boundaries 157
baselines from which the territorial sea was measured in the Anglo-Norwegian
Fisheries case was both a manifestation of sovereignty and an act to which third
states might object. Municipal legislation will not, however, be conclusive of
the validity of the sovereign acts; that will be a question for international law.
Rather, such legislation meets the requirement that the legislating state intends
to act as sovereign in the territory.
The enactment of legislation does not, however, have any absolute value when
determining which of two claimants has the better title. The inability of Norway
to adduce evidence of many activities referable to its intent to act as sovereign
in Eastern Greenland had the effect of tilting the balance in favor of Denmark’s
legislation in relation to the territory. The same is true of other evidence by the
contending states. The fewer the activities by one state, the greater is the weight
to be accorded to those of the other.
The enactment of legislation will provide evidence of the intent and will to act as
a sovereign, but the actual exercise of state authority must also be demonstrated
by acts of administration and the execution of government functions (Triggs
2006:222). Statutory powers must also be exercised if they are to be more than
a bolster for a paper claim. In an example of state practice, Japan protested to
Korea about their disputed claims to Takashima, arguing that:
Under international law, the most decisive factor in determining whether
or not a certain area is an inherent territory of a certain state from the
olden times is how effectively the State concerned has controlled and man-
aged the area in question (S. Oda and H. Owada, The Practice of Japan in
International Law: 1961-70 (1982) 69).
The kinds of evidence that satisfy this second limb of the Huber test are virtually
infinite and it is by no means clear why international tribunals have favoured some
acts above others. Examples of acts that have proved weighty in determinations
are asserting criminal jurisdiction, maintaining a registry of fishing boats, con-
structing mooring facilities, signal posts and lighthouses, promulgating hunting
and fishing licenses, commissioning magistrates, establishing a meteorological
station, granting a grazing licence and taxation.
Such manifestations of state authority have little value in themselves as the
relevant consideration will be how the acts of one state are balanced with those
of another. Indeed, some acts such as the placing of beacons and the installa-
tion of a light boar were relevant in the Grisbådarna case and rejected in the
Minquiers and Ecrehos case (pp. 69-70).
The range of legally relevant state activities is illustrated in the Rann of Kutch
case ((1968) 17 RIAA 1). Pakistan claimed title to the Rann, an area with both
158 Chapter 5
land and river features, as successor to the state of Sind. The president of the
tribunal, Judge Lagergren, stated that evidence of the functions of government
in the areas was limited to:
the imposition of customs duties and taxes on land, livestock and agricultural
procedure in the fiscal sphere, and to the maintenance of peace and order
by police and civil and criminal courts and other law enforcement agencies
in the general public sphere (1968:554).
The tribunal found that the registration of births, deaths and epidemics in Sind
‘comes as close to effective peaceful possession and display of Sind authority as
may be expected in the circumstances’. It was significant to the final conclusion
that Pakistan had made out the better title in sectors where activity was continuous
and intensive and that there had been ‘no effective opposition from the Kutch
side’ (1968:569). The emphasis is upon all the circumstances of the competing
claim, rather than upon any objective value to be attached to the particular facts
to determine ‘in whom the conglomerate of sovereign functions has exclusively
or predominantly vested’ (1968:568).
It is unlikely that it is a necessary condition of territorial sovereignty that a
claimant must full fill the duty of protection of the rights of other states and
their nationals, as suggested by Judge Huber. It is, however, probable that the
more effective a state proves to be in protecting these rights, the more convinc-
ing is its claim to effective control. In summary, the cases demonstrate that state
activities have a relative rather than absolute value and the weight accorded to
them depends upon the strengths of the counter-balancing evidence advanced
by the opposing state. Furthermore, state activities will be accorded probative
weight only if they have been undertaken by and on behalf of a sovereign and
are clearly referable to the exercise of sovereign functions.
Case study
Benin/Niger
International Court of Justice
Judgment of 12 July 2005
The West African Republics of Benin and Niger disputed their extensive com-
mon border from the time of their independence from France in 1960. When
it became clear that efforts by a joint commission to settle the frontier would
not be fruitful, Benin and Niger asked the Chamber of the ICJ, by special
agreement of 15 June 2001, to delimit the whole of the boundary (see map
5.3). Article 6 of the Special Agreement provided that the applicable law
included ‘the principle of State succession to the boundaries inherited from
colonization, that is to say, the intangibility of those boundaries’.
International Law and Territorial Boundaries 159
Torres Bernárdez in the Land, Island and Maritime Dispute case. Allen argues
(2006:736) that the distinction between colonial and state effectivités for the
purposes of the rule requiring sovereign intent is valid on the ground that uti
possidetis is a principle of international law that is determined retrospectively
‘by reference to the colonial heritage bequeathed by a single colonial power’.
However, while colonial acts were not always performed with the intent to
act as sovereign, most do, in fact, meet this criterion for title.
The Chamber also adopted the flexible view that acts subsequent to the
critical period could be relevant in determining the boundary on indepen-
dence. On this ground it accepted the value of an independent study made in
1970 that established the main navigable channel or thalweg. This channel, it
concluded, formed the international boundary, supported by evidence of day-
to-day administration. Adopting the rule that territorial sovereignty extends
in international law to the superjacent airspace, the Chamber was also able to
reach the sensible view that it could determine the boundary on two bridges by
extending a vertical line from the watercourse boundary to the structures.
The Chamber examined a number of administrative documents and ‘arrêtés’
promulgated by the Governor-General and the Governors of Dahomey and
Niger when determining the course of the Western Boundary. Confirmed by
maps and the absence of protest against these instruments, the Chamber con-
cluded that the 1907 inter-colonial boundary established by the river Mekrou
had been validly established by competent authorities throughout 1926 and
1953. As the Joint Benin-Niger Boundary Delimitation Committee was not
able to plot the coordinates of the Mekrou as the river was too shallow to
navigate, the Chamber nominated the median line as the boundary (para 144,
p. 150).
In each of the determinations of title and boundaries considered thus far, the
international tribunal has weighed the evidence in order to determine which of
the competing states had the superior claim. A quite different opportunity to
consider the principles of territorial acquisition arose when the ICJ was asked
for an Advisory Opinion in the Western Sahara case. This time, the ICJ was not
constrained by any compromis to find that the better claim to sovereignty lay with
one state or another (Triggs 2006:228). Rather, the court was open to introduce
fresh thinking about legal interests in territory that was to prove influential with
national courts in the future.
Western Sahara had been colonized by Spain in 1884 and is rich in phosphates
and fishing resources. In 1966 the General Assembly invited Spain, in consulta-
tion with its neighbors Mauritania and Morocco, to hold a referendum to enable
the peoples of Western Sahara to exercise their right to self-determination. A
International Law and Territorial Boundaries 161
The ICJ recognized, also for the first time, that where there are tribal groups in
the territory, any title to it cannot be founded in occupation of terra nullius. Of
contemporary importance to the Court was that any claim to sovereignty over
territory was subject to the customary law principle of self-determination. The ICJ
has subsequently confirmed the status of self-determination in the East Timor case
as a right erga omnes. But, as has been seen, the principle of self-determination
has not given way to the apparently overriding principle of uti possidetis.
Summary
The very existence of that convention precludes the United States from
acquiring by prescription against the terms of their title and . . . the two
republics have ever since the signing of that convention treated it as the
source of all their rights.
Another characteristic of possession serving as a foundation for prescription is
that it should be peaceable (p. 807):
however much the Mexicans may have desired to take physical possession of
the district, the result of any attempt to do so would have provoked scenes
of violence and the Republic of Mexico can not be blamed for resorting to
the milder forms of protest contained in its diplomatic correspondence.
Resorting to force is not, it seems, required to prevent a prescriptive title suc-
ceeding against a state. Moreover, a state is not necessarily bound to seek judicial
determination of a territorial or boundary dispute. As Brownlie (1990:157) put
it, ‘[i]f acquiescence is the crux of the matter (and it is believed that it is) one
cannot dictate what its content is to be.’ While it is true that resolution of each
dispute will turn on its facts, for a state to ignore appropriate dispute resolution
procedures is evidence to be weighed in the balance, especially in light of the
creation of new tribunals over the last decade. As the Foreign Affairs Committee
of the House of Commons recognized in its report of 1985, the fact that neither
the United Kingdom nor Argentina sought to refer their dispute over the Falk-
lands islands to the ICJ or any other form of legal arbitration ‘almost certainly
is evidence . . . of doubts about the solidarity of their respective legal claims’
(cited by Triggs 2006:231). In the more recent dispute over the resources of
the Timor Gap, Australia has withdrawn all maritime disputes from the ICJ or
the International Tribunal for the Law of the Sea (ITLOS) and East Timor has
made no attempt to bring the issue to any international tribunal. While boundar-
ies between neighbors are ideally agreed through negotiations, a failure to seek
readily available international adjudication where negotiations are unsuccessful
may be evidence of a failure to pursue a claim rigorously.
Historical Consolidation
2
(1986) ICJ Reps. at para 64.
International Law and Territorial Boundaries 165
earlier, after which they were implemented without protest. The decrees were
considered by the court to represent a ‘well-defined and uniform system . . . which
would reap the benefit of general toleration, the basis of an historical consolida-
tion which would make it enforceable as against all States’ (p. 138).
In stark contrast to the Anglo Norwegian Fisheries case, the ICJ, some fifty
years later in Cameroon v Nigeria (ICJ Reps. 2002), showed little tolerance for
the concept of historical consolidation. Cameroon had argued that, as the holder
of a conventional title to the disputed areas, it need not demonstrate the effec-
tive exercise of sovereignty because a valid title prevails over effectivités which
might have indicated title lies with another state. The ICJ agreed, noting without
further analysis, that historical consolidation ‘has never been used as a basis of
title in other territorial disputes, whether in its own or in other case law’ (para
65). The Court found that:
the theory of historical consolidation is highly controversial and cannot
replace the established modes of acquisition of title under international
law, which take into account many other important variables of fact and
law. It further observes that nothing in the Fisheries Judgment suggests that
the ‘historical consolidation’ referred to, in connection with the external
boundaries of the territorial sea, allows land occupation to prevail over an
established treaty title. Moreover, the facts and circumstances put forward
by Nigeria with respect to the Lake Chad villages concern a period of some
20 years, which is in any event far too short, even according to the theory
relied on by it. (para 65)
This recent finding of the ICJ confirms that territory subject to a valid title may
not be acquired by effectivités over a period of time, unless the holder of that
title consents to ceding it to another state or has acquiesced to a transfer. On
examination of the evidence, the Court concluded that Cameroon had attempted
to carry out administrative functions in the disputed areas and protested against
Nigerian presumptions. Cameroon had not, therefore, acquiesced in abandonment
of its title in favor of Nigeria. Citing its earlier decision in the Frontier Dispute
(Burkina Faso/ Mali) case, the ICJ found that, where the effectivités adduced by
Nigeria did not correspond to the law, ‘preference should be given to the holder
of the title’ (para 68). Where, however, the effectivités do not co-exist with a
legal title, it should be taken into consideration. On this ground, the Court con-
cluded that, as Cameroon held the legal title to the territory under the relevant
instruments, its conduct was pertinent only in respect of ‘whether it acquiesced
in the establishment of a change in treaty title, which cannot be wholly precluded
as a possibility in law’ (para 68). The evidence suggested that Cameroon had
undertaken some administrative activities in the area and had protested when
Nigeria clamed sovereignty over Darak, expressing ‘its profound shock at the
presumption’. The Court thus concluded that the situation was one in which the
effectivités adduced by Nigeria did not correspond to the law, so that preference
should be given to Cameroon’s claim (para 70).
166 Chapter 5
3.3). On the basis that the boundary treaties were the ‘central feature’ of the
dispute (para 3.3), the Commission applied Article 31 of the Vienna Convention
on the Law of Treaties 1969 to the effect that treaties are to be ‘interpreted in
good faith, in accordance with the ordinary meaning to be given to the terms
of the treaty in their context and in the light of its object and purpose’. Adding
something of a gloss to this well recognised principle of interpretation, the Com-
mission considered it as a tool for determining the ‘common will’ of the states
(para 5.16). This common will was to be understood according to the principle
of inter-temporal law; that is, the treaty should be interpreted ‘by reference to
the circumstances prevailing when the treaty was concluded’ (para 3.5). When
considering the meaning of a geographical name used in a treaty, for example,
the test was the ‘contemporary understanding of the location to which that name
related at the time of the treaty’ (para 5.17). In short, if the ‘name used is incor-
rect, then it is the Parties’ intentions with respect to the reality on the ground
rather than the name that is decisive’ (para 4.14).
The Commission concluded that it is:
required also to apply those rules of international law applicable generally
to the determination of disputed boundaries including, in particular, the rules
relating to the effect of the conduct of the parties. In the present case, the
conduct of the Parties falls into three broad categories: maps; activity on
the ground tending to show the exercise of sovereign authority by the Party
engaging in that activity (effectivités); and a range of diplomatic and other
similar exchanges and records, including admissions before the Commission,
constituting assertions of sovereignty, or acquiescence in or opposition to
such assertions, by the other Party (para 3.15-6).
The Commission took particular account of the December Agreement which
referred to the principle of ‘respect for the borders existing at independence as
stated in resolution AHG/Res. 16 (1) of the OAU Summit in Cairo in 1964’.
This provision confirmed the practice of international tribunals of setting the
critical date at the time of independence, with the exception, in this case, of the
area between Setit and the Mareb which arose earlier in 1935. The importance
of the critical date lies in the legal consequence that developments subsequent to
it are not to be taken into account unless they confirm or continue the conduct
already clearly established (para 3.36).
The Commission adopted customary international law relating to the role
of effectivités, noting that conduct pursued à titre de souverain could, at best,
produce only a title relative to that of the competing state. The process of mea-
suring the conduct of the parties against each other, the Commission concluded,
might lead to a ‘legal result . . . to vary a boundary established by a treaty’. (para
3.29). (The role of maps in this dispute is considered in detail in chapter 6.) This
indeed, was the approach applied by the Commission. On examination of the
three relevant treaties, and in light of the problems experienced in interpreting
168 Chapter 5
For many years, neither Ethiopia nor Eritrea chose to cooperate with the Commis-
sion’s process of placing the pillars marking the line of the Eastern boundary, as
required by the Demarcation Directions of 8 July 2002. Nor have they permitted
pillar site assessments for the Central and Western Sectors. In some frustration,
the Commission decided on 27 November 2006 to adopt the international principle
of ‘effectiveness’ when applying treaties. In light of the failure of demarcation
through the fixing of pillars, being the method of first choice, the Commission
adopted another approach, using image processing and terrain modeling with high
resolution aerial photography to locate boundary points. The boundary points were
considered by the Commission to have ‘a degree of accuracy that does not differ
significantly from’ the pillar emplacement process (Statement of 17 November
2006, para 20). These points also have the signal advantage that they provide a
International Law and Territorial Boundaries 169
practical way in which the Commission could achieve its mandate to establish
a final boundary. If the parties continue to refuse to cooperate, the Commission
has decided (ibid., para 22) that the:
boundary will automatically stand as demarcated by the boundary points
listed in the Annex hereto and that the mandate of the Commission can
then be regarded as fulfilled. . . . Until such time as the boundary is finally
demarcated, the Delimitation Decision of 13 April 2002 continues as the
only valid legal description of the boundary.
As Sir Elihu Lauterpacht, President of the Commission, points out, demarca-
tion of a boundary through coordinates using appropriate technology is entirely
consistent with international precedent (ibid., paras 23–26). In 1993 the United
Nations Secretary-General and the Security Council adopted a list of coordinates
in respect of the Iraq-Kuwait border established by the Iraq-Kuwait Boundary
Demarcation Commission. Similarly, UNCLOS adopts such coordinates when
assessing the maritime claims made by coastal states.
European Powers had entered into many territorial and boundary treaties with
the local chiefs of the Niger delta. One such agreement was that made in respect
of the Bakassi Peninsular by Great Britain under the Treaty of 10 September
1884. The Court decided, however, that the 1884 Treaty was not intended to
establish an international protectorate (para 207). Rather, Britain regarded itself
as administering the territories, not simply protecting them. No other evidence
was adduced to support Nigeria’s claim that a protectorate alone had been cre-
ated. The Court concluded therefore that, under the law as it was in 1884 and in
light of the inter-temporal principle, Great Britain became the sovereign power
over the Bakassi Peninsular and could, in 1913, validly determine its boundar-
ies with Germany in respect of Nigeria. This decision usefully demonstrates the
continuing importance of the fundamental principle that a state must have valid
title before it can purport to cede its rights to another state.
The transfer of territory by cession has frequently arisen in the past where
the treaty is a form of peace settlement after a war. An example is the Treaty
of Paris of 1889 after the Spanish-American war in the Island of Palmas case.
Similarly, after Spain lost the War of Spanish Succession, it agreed in the Treaty
of Utrecht of 1714 to ‘yield to the Crown of Great Britain the full and entire
propriety of the town and castle of Gibraltar’. In light of article 52 of the Vienna
Convention on the Law of Treaties 1969, which provides that a treaty is void if
it has been procured by threat or use of force, any treaty of cession would not
be valid were it to be obtained by force today. Security of title is preserved for
any territory gained by cession which has been gained by the use of force prior
to the 1945 UN Charter prohibition on the use of force. Such titles are protected
by the principle of inter-temporal law; that is, so long as title is valid at the time
it was acquired, it will remain valid.
A treaty will also be employed to transfer title where a colonial or adminis-
tering power cedes authority to the indigenous or local population. An example
arose where the four powers, France, the USSR, the UK and the US, entered
into a treaty with the Federal Republic of Germany and the German Democratic
Republic to transfer their powers to a united Germany in 1990. In a unique
example, the UN Transitional Administration in East Timor (UNTAET) was
granted treaty-making powers on behalf of the peoples of East Timor by the UN.
On independence, UNTAET transferred these powers by treaty to the new state
of Timor Leste. In the Sino-British Joint Declaration of 1984 the UK agreed to
transfer to China that part of Hong Kong that China had ceded to it in the middle
of the nineteenth century. It is also possible, though rare, for a state to sell its
territory to another; a well-known example being the purchase of Alaska from
Russia for $7.2 million in 1867.
Boundary Treaties
recognized the vital role such treaties play in ensuring international stability. In
Eritrea v Yemen (Phase One) the Arbitral Tribunal considered that boundary
treaties constitute an objective territorial regime that is valid erga omnes, with
an independent validity, even if the treaty itself comes to an end (para 153). For
this reason, in a dispute over a boundary established by treaty a tribunal will,
in the first analysis, be concerned with narrow questions of treaty interpreta-
tion rather than wider principles governing the acquisition of territory. The ICJ
recognized, for example, in Cameroon v Nigeria that the dispute ‘is in reality
simply a dispute over the interpretation or application of particular provisions
of the instruments delimiting that boundary’ (para 85, p. 360). In Cameroon v
Nigeria the Court accepted that its task was not to demarcate the boundary but
to specify its course as set out by the instruments of delimitation. The ICJ then
embarked on an extensive examination of each relevant treaty, interpreting and
applying phrases such as ‘source of the river’ and ‘prominent, pointed peak’ to the
available geographical evidence. A specific example of the methodology adopted
by the Court arose when it was required to interpret the Thompson-Marchand
Declaration. Paragraph 18 of this Declaration provides that the boundary is to
follow ‘the Keraua’ river (para 95). In fact, this river splits into two channels,
the Court’s task being to identify which of them was intended by the states par-
ties to define the boundary. The Court found that Cameroon had provided no
evidence in support of its claim that the Keraua River had been diverted by an
artificial channel, allegedly constructed by Nigeria. The Court also rejected the
cartographic and photographic evidence produced on the ground that it did not
confirm the existence of any works to divert the river. Relying instead upon a
map showing the boundary to the east of two villages that were recognized to
be within Nigerian territory, the Court concluded that the boundary lay in the
eastern channel as the accurate location of the boundary.
While treaty interpretation tends to dominate the process of identifying a bound-
ary, where a treaty is ambiguous, an international tribunal will also consider the
customary law principles of territorial acquisition. It remains clear, however, as
illustrated by the Frontier Dispute (Burkina Faso/Mali) case, that acts demon-
strating effective occupation of territory cannot prevail over an otherwise valid
title, particularly one created by treaty.
In the Frontier Dispute (Burkina Faso/ Mali) case, the ICJ recognized the need
to understand the relationship between effectivités and the principle of uti pos-
sidetis by identifying four categories:
1. If the ‘act corresponds exactly to law’ and acts of effective administrative are
additional to uti possidetis, the only role for the effectivités is to confirm the
right that is derived from title (para 63, pp. 586-587).
172 Chapter 5
2. Where the act does not correspond to the law, and the territory is effectively
administered by a state other than the one with title, preference will be given
to the holder of title (para 63, p. 587).
3. Where the acts do not co-exist with any legal title, they ‘must invariably be
taken into consideration’ (ibid.).
4. Where the legal title is not capable of showing exactly the territory to which
it relates, ‘the effectivités can then play an essential role in showing how the
title is interpreted in practice’ (ibid.).
Allen (2006:738) rightly observes that the Chamber’s analysis of the relation-
ship between title and effectivités ‘invites reflection upon the concept of title
itself’. The Chamber in the Frontier Dispute case had recognized that the idea
of ‘legal title’ connotes more than just documentary evidence and that it might
more broadly include ‘any evidence which may establish the existence of a right,
and the actual source of a right’ (para 18, p. 564). Allen (2006:738) argues that
while colonial acts cannot, in principle, affect the application of the right to title
at international law, in practice, they will ‘guide’ the application of the uti pos-
sidetis principle. It seems that, ‘in ambiguous situations, facts rather than “law”
may determine the application of uti possidetis juris’ (ibid.).
Recognition of the vital relationship between title and effectivités does not, how-
ever, deny a role for subsequent activities in interpreting any ambiguous terms
of a boundary treaty. When delimiting the Eritrea-Ethiopia boundary, the Com-
mission accepted that the legal title established by treaty could be adjusted if the
subsequent conduct of the parties indicated that they had accepted the change.
The ICJ in the Cameroon v Nigeria case confirmed that the delimitation agree-
ment could be modified by subsequent state conduct where it was agreed that
the Sapeo area lay within Nigerian territory despite the terms of the boundary
agreement (para 144).
Taking subsequent state conduct a step further, there is also support for the role
of acquiescence in a change in title established by treaty. The ICJ in the Cameroon
v Nigeria case (para 68) and the Land, Island and Maritime Frontier Dispute
case (para 80) recognized that acquiescence by one state in modification of title
agreed by treaty could not be ‘wholly precluded as a possibility in law’.
The traditional capacity of a state to cede its territory to another state is now
subject to the right erga omnes of the local inhabitants to self-determination. As
the ICJ concluded in the East Timor case (para 29, p. 101):
Portugal’s assertion that the right of peoples to self-determination, as it
evolved from the Charter and from United Nations practice, has an erga
omnes character, is irreproachable. The principle of self-determination
International Law and Territorial Boundaries 173
of peoples has been recognized by the United Nations Charter and in the
jurisprudence of the Court . . . it is one of the essential principles of contem-
porary international law.
In the past, the asserted principle that no cession of territory would be valid in
the absence of provision for an expression of opinion by its inhabitants has been
of doubtful credibility, partly because states have accommodated any illegality
by recognition. However, the more recent determinations of the ICJ and practices
of the UN now confirm the status of the principle in customary law.
The right to self-determination may not, however, necessarily prevent third
states from entering into a treaty with a state that has unlawfully gained control
of territory to which the principle applies. The ICJ in the East Timor case rec-
ognized the characterization by the General Assembly of East Timor as a non-
self-governing territory and of the right of its people to self-determination by its
Resolutions 384 (1975) and 389 (1976). The failure of Indonesia to give effect
to the right to self-determination did not persuade the court that Australia was
thereby bound to deal only with Portugal as the administering power over the
territory. It was noted by the court that, despite Portugal’s protests against the
conclusion of the Timor Gap Treaty in 1989, the Security Council and General
Assembly took no responsive action (para 32, p. 104).
Aside from the question respecting the kinds of act that can validly occur
despite a failure to respect the principle of self-determination, it is clear that any
attempt to cede territory without giving effect to the principle will be ineffective.
To this extent, the law governing territorial sovereignty has been significantly
modified.
In practice, most titles to territory have been acquired through some form of
cession as distinct from effective occupation of terra nullius. Indeed, the deci-
sion by Australia’s High Court that Australia was not terra nullius at the time
of colonial settlement has prompted a revision of thinking about the possibility
of pre-existing native title. Many treaties transferring territorial sovereignty were
negotiated with indigenous rulers of identifiable areas of territory. Judge Huber
in the Island of Palmas case (p. 859) observed that such a treaty was:
not an agreement between equals; it is rather a form of internal organization
of a colonial territory, on the basis of autonomy for the natives . . . [a]nd thus
suzerainty over the native State becomes the basis of territorial sovereignty
as towards other members of the community of nations.
The ICJ in the Western Sahara case also recognized that agreements concluded
with local rulers over tribes or people with a social and political organization
‘were regarded as derivative roots of title’ (para 80, p. 39). While the court in
Cameroon v Nigeria noted that this mode of acquisition does not reflect current
international law, it applied the principle of inter-temporal law so that the legal
consequences of the treaties concluded in the Niger delta in the 1880s would
be given effect today.
174 Chapter 5
Changes in territory through accretion, erosion and avulsion are not accurately
described as roots of title (Triggs 2006:248). Rather, states acquire any new
territory formed through such natural processes by effective occupation and
acquiescence. Accretion and erosion, being the gradual addition or reduction of
territory, may result in the acquisition or loss of territory depending on whether
the respective states are in effective occupation of it or acquiesce in title becom-
ing consolidated.
International law will acknowledge that there has been a change in title to
territory if that territory has changed as a result of accretion or avulsion.
• Sovereignty will change in the event of an accretion. For example, if a change
in the course of a river is gradual, the boundary, usually at the thalweg or
middle of the channel, will change with it.
• Sovereignty will remain as it was if the change arises rapidly by an avulsion.
Avulsion refers to the violent change in territory through a flood or creation
of new islands. Where the change, whatever the cause, is within the territory
or maritime zone of a state, international practice assumes title. For example,
when a volcano erupted creating a new island in the territorial seas of Japan,
the UK Government recognized it as Japanese territory (Shaw 2003:420).
These working rules will, in practice, depend upon the presence of other factors
including the relevant treaties, and evidence of occupation and acquiescence. The
leading authority on the legal consequences of gradual change is the Chamizal
Arbitration in 1911. The Rio Grande, as the boundary established by a Treaty
of 1848 between the United States and Mexico, slowly changed its course so
that by 1911 a 600-acre area, known as the Chamizal Tract, had been formed
on the US side of the new river. Both states claimed title and the dispute was
submitted for determination by an International Boundary Commission. While
finally concluding that the United States held title to that part of the tract result-
ing from accretion and Mexico held title to that part resulting from a flood in
1864, the Commission applied the terms of the Treaty rather than any general
principle of international law.
While these rules on accretion and avulsion can thus be useful tools for
resolving a dispute, each boundary dispute will depend upon its particular cir-
cumstances.
International Tribunals will take into consideration many factors that are not
strictly formal roots of title when considering which of the claimant states has
International Law and Territorial Boundaries 175
the better title to territory (Triggs 2006:231). In addition to the relative value of
evidence and the adoption of a balancing process, tribunals have been concerned
to promote international harmony by awarding title to one contender rather
than to create a legal vacuum. Significant among such factors are recognition,
acquiescence and estoppel, which facilitate an international decision based upon
the express or tacit consent or good faith of the states concerned. In particular,
recognition and acquiescence provide valuable probative evidence to supple-
ment competing and inconclusive evidence of effective control and government
activities in the territory.
5.1 Recognition
5.2 Acquiescence
would in any case warrant Norway’s enforcement of her system against the
United Kingdom.’
The Court is thus led to conclude that the method of straight lines, established
in the Norwegian system, was imposed by the peculiar geography of the Nor-
wegian coast. It found that even before the dispute arose, this method had been
consolidated by a constant and sufficiently long practice, in the face of which
the attitude of governments bears witness to the fact that they did not consider
it to be contrary to international law. In his Dissenting Opinion, Judge McNair
observed (p. 162) that:
Governments are not prone to understate their claims. . . . They will usually
give air to their grievances by the formulation and dispatch of a protest,
the purpose of which is to build up an almost instinctive defense mechanism
designed to vitiate any possible interpretation of silence as acquiescence.
Mere silence can, however, be ambiguous and authorities have been reluctant to
treat silence as consent without knowledge of relevant facts (Triggs 2006:232).
As Great Britain argued in the Alaskan Boundary dispute with the United States,
‘you cannot protest against a thing you have never heard of’ (Proceedings of the
Alaskan Boundary Tribunal, 98 Brit. Foreign and State Papers, (1903) 152). In
the Island of Palmas case Judge Huber doubted that relevant information would
not be available in practice because a ‘clandestine exercise of State authority
over an inhabited territory during a considerable length of time would seem to
be impossible’ (p. 868). Not only would it be impossible, it would contradict
any assertion of effective occupation. International tribunals have readily imputed
knowledge of legislative enactments and the published maps of other states. In
the Temple of Preah Vihear case the ICJ found (p. 23) that
the maps were given wide publicity in all technically interested quarters by
being also communicated to the leading geographical societies in important
countries, and to other circles regionally interested; to the Siamese legations
accredited to the British, German, Russian and United States Governments;
and to all the members of the Mixed Commission, French and Siamese. . . .
[I]t is clear that the circumstances were such as called for some reaction,
within a reasonable period, on the part of the Siamese authorities, if they
wished to disagree with the map or had any serious question to raise in
regard to it. They did not do so, either then or for many years, and thereby
must be held to have acquiesced.
The requirement that acquiescence cannot occur in the absence of relevant
knowledge raises the question whether, as Judge Huber asked in the Island of
Palmas case (p. 868), it is a ‘condition of legality’ that a claimant should notify
its pretensions to other states. It is probable that notification is not required at
international law, though states have objected when they have not received noti-
fication of occupation of territory. Notification may, in practice, be made both to
178 Chapter 5
avoid such objections and as evidence of the intent and will to act as sovereign.
When transferring part of its claim to the Antarctic Territory to Australia in 1933,
for example, the UK was careful to advise France accordingly.
It will be recalled that Judge Huber considered that ‘the continuous and peaceful
display of territorial sovereignty (peaceful in relation to other States) is as good
as title’ (Island of Palmas case, p. 839). He has not been taken to have intended
that sovereignty cannot be established wherever there is a protest, but rather,
he was concerned to consider the effects of a lack of protest. Norway’s protests
against Denmark’s displays of sovereign authority in Greenland did not alter the
peaceful character of Denmark’s claim to title in the Eastern Greenland case. It
is also likely that where a state has effectively occupied the territory, the protests
of a competing state could at best be a temporary bar to title where the occupying
state continues to remain in possession. The implication in the Anglo-Norwegian
Fisheries case was that, had the UK constantly and unambiguously objected to the
Norwegian system of delimitation, the baselines would not have been opposable
against it. It is quite another step to argue that protests can prevent a state from
acquiring territory that is terra nullius. It is thus unlikely that protests could be a
permanent bar to acquisition of territory by effective occupation. If, by contrast,
the disputed territory is res communis, persistent objections from a majority of
the international community could prevent consolidation of title.
5.3 Estoppel
the map. . . . It is not now open to Thailand, while continuing to claim and
enjoy the benefits of the settlement, to deny that she was ever a consenting
party to it.
The Temple of Preah Vihear case is an illustration of the difficulties inherent in
isolating bases of title where they are integrated with each other and dependant
on particular circumstances. There is little apparent difference between implied
recognition, acquiescence and estoppel, so that the inaction or silence of a state
may mislead another to its detriment, thereby creating an estoppel.
International law has promoted the stability of international relations through the
principle of sovereignty over territory, the norm of uti possidetis being a high-
water mark of the value of the status quo (Triggs 2006:250). Judge Huber has,
nonetheless, observed that rights in relation to territory are subject to evolving
principles of international law. Foremost among contemporary ideas has been the
development of the right of self-determination by the UN Charter and the 1966
Declaration on the Granting of Independence to Colonial Countries. Of compa-
rable magnitude in moderating traditional approaches to territorial acquisition has
been the development of the concept of the common heritage of mankind and
of rights in rem over territory for the benefit of the international community as
a whole. As the editors of Oppenheim (Jennings & Watts, 9th ed, 1992:716-17,
s275) have observed, one of the most important new factors to be considered
when balancing evidence of territorial sovereignty is the attitude of the interna-
tional community expressed through the UN. Security Council Resolution 662,
for example, states that the annexation of Kuwait by Iraq ‘under any form and
whatever pretext has no legal validity, and it is considered null and void’.
It has been noted that international law does not permit the cession of territory
in the absence of ensuring the local people have had the opportunity to express
their wishes. Generally, the views of the peoples affected by a decision concern-
ing territory, including exploitation of offshore petroleum resources, will now be
relevant to an international tribunal. Judge Vereshchetin in a Separate Opinion
in the East Timor case observed (pp. 135, 137-138):
Besides Indonesia . . . there is another ‘third party’ in this case, whose consent
[was not sought] . . . The ‘third party’ at issue is the people of East Timor.
Since the [majority] Judgment is silent on this matter, one might wrongly
conclude that the people, whose right to self-determination lies at the core
of the whole case, have no role to play in the proceedings. This is not to
suggest that the Court could have placed the States Parties to the case and
the people of East Timor on the same level procedurally . . . This is merely to
say that the right of a people to self-determination, by definition, requires
that the wishes of the people concerned at least be ascertained and taken
into account by the Court . . . [There is a] necessity for the Court at least to
ascertain the views of the East Timorese representatives of various trends
of opinion on the subject-matter of the Portuguese Application.
182 Chapter 5
Indigenous peoples have a special relationship with the land that is increasingly
recognized by international law so as to moderate the rights of the territorial
sovereign. The Advisory Opinion of the ICJ in the Western Sahara case had a
powerful influence on national indigenous interests by unlocking the constraints
imposed by the concept of terra nullius. Outside the context of an adversarial
process, the ICJ was free to consider the possibility that the Western Sahara was
not terra nullius at the time of occupation by Spain and thus was not subject
to acquisition under the principles of effective occupation. The Australian High
Court in Mabo v Queensland (No. 2) adopted the dicta of the ICJ stating (pp.
40, 41-42):
The theory of terra nullius has been critically examined in recent times
by the International Court of Justice in its Advisory Opinion on Western
Sahara . . . If the international law notion that inhabited land may be classified
as terra nullius no longer commands general support, the doctrines of the
common law which depend on the notion that native peoples may be ‘so low
in the scale of social organization’ that it is ‘idle to impute to such people
some shadow of the rights known to our law’ [In re Southern Rhodesia
[1919] AC 211 at 233-4] can hardly be retained. If it were permissible in
past centuries to keep the common law in step with international law, it is
International Law and Territorial Boundaries 183
imperative in today’s world that the common law should neither be nor be
seen to be frozen in an age of racial discrimination.
The possibility that the facts may not support the acquisition of terra nullius
stimulated the High Court to reject the ‘fiction by which the rights and interests
of indigenous inhabitants in land were treated as non-existent’ on the ground that
it has been justified by a ‘policy which has no place in the contemporary law of
this country’ (p. 42). Effective occupation of land was not possible where the
territory was not terra nullius. The court was thus able to conclude that native
title could continue to exist in those parts of Australia where no acts of cession
or conquest had extinguished such title. In this way, the evolving jurisprudence
of the ICJ had an influence on recognizing the native title of ingenious peoples
within the legal framework of territorial sovereignty.
Other developments in international law are influencing the rights of indigenous
peoples in respect of territory, though for the most part they are de lege ferenda;
that is, international law is evolving but not yet binding on states. The Draft
Declaration on the Rights of Indigenous Peoples, for example, seeks to protect,
as a form of self-determination, their right to ‘autonomy or self-government in
matters relating to . . . land and resources management’.
While there has been a growing recognition of the land rights of indigenous
peoples and of the right to self-determination, international tribunals have been
disinclined to moderate established titles to territory when responding to the
contemporary phenomenon of the movement of peoples. In the Cameroon v
Nigeria case (2002), for example, the ICJ noted the influx of people from Nigeria
into the Bakassi Peninsular in 1968 following the civil war, and accepted that
as a result of this influx the Nigerian population had significantly increased in
the area. The Court stated, nonetheless, that ‘these facts of themselves do not
establish Nigerian title over Bakassi territory; nor can they serve as an element
in a claim for ‘historical consolidation’ of title’ (para 221). This decision serves
to underscore the primary value given to titles ceded by treaties that had typi-
cally been negotiated by colonial powers in the 19th and 20th centuries. Such
titles will be upheld by international tribunals and are not capable of amendment
to reflect today’s population movements without the consent of the affected
sovereign states.
The notion that territory and resources beyond the limits of national jurisdic-
tion might be available to all mankind and not subject to unilateral territorial
acquisition evolved through the 1960s and 1970s as the international community
negotiated legal regimes for the regulation of outer space, the deep seabed and
Antarctica (Triggs 2006:252). The concepts of the common heritage of mankind
and res communis share the same core ideas and import a principle rather than
184 Chapter 5
a set of detailed legal rules. The principle of a common heritage does, however,
have certain identifiable characteristics under which territory is subject to com-
munal ownership and control, and is hence res communis and not amenable to
state sovereign title or rights. The central objectives are to safeguard a common
heritage for future generations, to conserve the earth’s resources and, not entirely
consistently, to share the benefits of the use and exploitation of these resources.
The principle of the common heritage has been successfully applied to outer
space, the moon and other celestial bodies, and to the deep seabed of the high
seas beyond the limits of national jurisdiction. There are also suggestions that
the principle will, over time, come to apply to aspects of the environment, such
as global climate, or to ‘iconic’ species of fauna, such as whales.
Of more immediate relevance to consideration of territorial boundaries is the
proposal by some states, such as Malaysia, that the concept of common heritage
should be applied to Antarctica. The consequence would be to deny the validity
of the seven claims currently made to territorial sovereignty in the area south of
60 degrees South latitude. The legal merit of these claims and the legal issues
that arise from differing juridical approaches to Antarctic sovereignty are con-
sidered in Chapter 14.
8. INTER-TEMPORAL LAW
The claimant states had little difficulty with this proposition. The problem
arose with the second limb of Judge Huber’s judgment in which he draws a
distinction between the creation of rights and the existence of rights. He argued
(pp. 845-839):
[I]t cannot be sufficient to establish the title by which the territorial sover-
eignty was validly acquired at a certain moment; it must also be shown that
the territorial sovereignty has continued to exist and did exist at the moment
which for the decision of the dispute must be considered as critical. . . .
The same principle which subjects the act creative of a right to the law in
force at the time the right arises, demands that the existence of the right, in
other words its continued manifestation, shall follow the conditions required
by the evolution of the law.
The Huber doctrine of inter-temporal law thus requires satisfaction of two
apparently inconsistent elements: acts must be judged according to the law
contemporaneous with them and rights validly acquired may be lost if they are
not maintained in accordance with evolving international law. Not only is this
‘gloss’ on the law a departure from the views of other international lawyers at
the time, but also, as Jessup (1928:740) points out, the logical conclusion of
Huber’s extension of inter-temporal law is that ‘[t]itle insurance would be an
impossibility’. Huber’s gloss may not be as alarming as juristic writers have
suggested. If the international rules differ as between the acquisition of title and
maintenance of it, Huber’s approach does not necessarily produce an injustice.
If a state, on validly acquiring territory, carries out the functions of a sovereign
in relation to it, no state could effectively challenge that title. If the state fails to
act as sovereign and has abandoned the territory, no injustice is done if another
state claims to have acquired title by prescription or acquiescence.
Subsequent international tribunals have adopted the Huber approach to inter-
temporal law. In the Island of Palmas case, Judger Huber found that the prior
Spanish title based on discovery could not prevail over the later acts of effec-
tive occupation by the Dutch. The ICJ in the Minquiers and Ecrehos case (at
p. 57) also found that any original feudal title had lapsed as a consequence of
events after 1204 and that title was subsequently acquired in accordance with
the developing customary law of effective occupation.
While many cases considering both territorial and boundary disputes under
treaties have favoured the first limb of the Huber doctrine, the ICJ applied both
limbs in the Aegean Sea – Continental Shelf case. In an application to the court by
Greece against Turkey regarding delimitation of the continental shelf, the Greek
Government argued that the court had jurisdiction under the compulsory proce-
dures established by the General Act of 1928, despite its reservation excluding
all ‘disputes relating to the territorial status of Greece’. Greece maintained that
the reservation could not apply because ‘the very idea of the continental shelf
was wholly unknown in 1928 when the General Act was concluded’ (para 77,
186 Chapter 5
p. 32). The court found (ibid.) that the reservation was of a generic kind and
that:
Once it is established that the expression ‘the territorial status of Greece’
was used in Greece’s instrument of accession as a generic term denoting
any matters comprised within the concept of territorial status under general
international law, the presumption necessarily arises that its meaning was
intended to follow the evolution of the law and to correspond with the mean-
ing attached to the expression by the law in force at any given time.
The court concluded on this basis that the rights were to be interpreted for the
purposes of the General Act ‘in the light of the geographical extent of the Greek
State today, not of its extent in 1931’ (para 78, p. 33). The reservation was thus
interpreted to include disputes relating to the full geographical extent of Greece’s
rights over the continental shelf in the Aegean Sea. The paradoxical outcome of
this analysis was that the court had no jurisdiction over the dispute, even though
Greece herself had brought the application and preferred a restrictive approach
to its reservation.
Article 31 of the Vienna Convention on the Law of Treaties provides that
‘any relevant rules of international law applicable in the relations between the
parties’ are to be taken into account. The principle of inter-temporal law is one
such rule that the courts have continued to apply where relevant.
9. CRITICAL DATE
An important tool for resolving territorial disputes lies in establishing the date at
which the dispute has crystallized between competing claims (Triggs 2006:226).
The function of the so-called ‘critical date’ is to determine the legal weight to be
given to evidence of sovereignty. A rationale for determining the critical date lies
in ensuring that the activities of a state have not been embarked upon with the
purpose of improving its legal position through ‘contrived manoeuvres’. Where
relevant, however, state activities taking place after the critical date may prove
useful in confirming or denying claims. Clearly, the date can be no later than
the institution of legal proceedings.
The choice of the critical date can be the decisive issue between the states. In
Minquiers and Ecrehos France argued (p. 58) that she needed only to demonstrate
title up to the 1838, the year immediately preceding the date of the Anglo-French
Fisheries Convention of 1839. Had the court accepted this as the critical date,
the more favourable evidence of British acts of sovereignty after 1839 would
have been excluded. The Court (pp. 59-60) avoided establishing a single critical
date in the following way:
International Law and Territorial Boundaries 187
A dispute as to sovereignty over the groups did not arise before the years
1886 and 1888, when France for the first time claimed sovereignty over the
Ecrehos and the Minquiers respectively. But in view of the special circum-
stances of the present case, subsequent acts should also be considered by
the Court, unless the measure in question was taken with a view to improv-
ing the legal position of the Party concerned. In many respects activity in
regard to these groups had developed gradually long before the dispute as
to sovereignty arose, and it has since continued without interruption and
in a similar manner. In such circumstances there would be no justification
for ruling out all events which during this continued development occurred
after the years 1886 and 1888 respectively.
In some cases, establishing the critical date poses few difficulties. In the Island
of Palmas case, the title of the US depended on whether Spain had sovereignty
when she purported to cede the island under the Treaty of Paris in 1898. Judge
Huber found that the conclusion and coming into force of the Treaty of Paris
was the ‘critical moment’ where the question of the existence of sovereignty
arose. Similarly, in the Eastern Greenland case, the dispute crystallised when
Norway issued its proclamation of sovereignty on 10 July 1931. It may also be
that the terms of reference to the tribunal state the relevant critical date, as in
the Western Sahara case where the ICJ was asked for its Advisory Opinion by
reference to the date of Spain’s colonization of the territory.
More complex problems arise where, for example, a state needs to demonstrate
that title gained by cession has not subsequently been abandoned. Generally, the
critical date can be set when the dispute arises or crystallises between the parties.
This may occur where one state asserts that it has gained title by prescription
against another with an original but lapsed title, or, as is more likely, where the
original sovereign protests. Where both states claim title on longstanding histori-
cal grounds, it may be necessary to allow evidence over the whole period, as
in the Minquiers and Ecrehos case. A dispute of this kind may not crystallize
for many years so the critical date should be when the dispute arises, to allow
the range of evidence to be considered. The ICJ in this case, while considering
the evidence from feudal times, nonetheless placed the primary emphasis on the
period at which the dispute arose.
The critical date for assessing any challenge to title in cases of new states suc-
ceeding from colonial power would usually be the date of independence, as in the
Frontier Dispute case. In the Benin/Niger case, the Chamber relied heavily on the
Frontier Dispute case and recognized the principle of uti possidetis as ‘freezing’
title to territory at the moment of independence, thus becoming the critical date
for determining the boundaries and, frequently, territorial sovereignty issues.
International tribunals have not, however, always relied upon the date of inde-
pendence as the critical date. The Chamber of the ICJ in the Land, Island and
Maritime Frontier Dispute case found that a later critical date may arise either
from adjudication or from a boundary treaty where the ‘date of the award [had]
188 Chapter 5
become a new and later critical date’ (para 67, p. 401). As the General Treaty of
Peace of 1980 had established parts of the El Salvador/Honduras boundary, 1980
became the new critical date. The Chamber observed (para 67, p. 401) that if the
principle of uti possidetis can be qualified by adjudication and by treaty:
. . . the question then arises whether it can be qualified in other ways, for
example, by acquiescence or recognition. There seems to be no reason in
principle why these factors should not operate, where there is sufficient
evidence to show that the parties have in effect clearly accepted a variation,
or at least an interpretation, of the uti possidetis juris position.
While a tribunal might reject the date of independence as the relevant date at
which to consider evidence of sovereignty, the tribunal might also choose not
to establish a critical date at all. The ICJ in the Frontier Dispute case and the
arbitral tribunal in the Argentine-Chile Frontier case considered the concept to
be of little or no practical value. Similarly, in Eritrea v Yemen (Phase One), the
tribunal decided to examine all the evidence submitted to it, regardless of the
date of the sovereign acts.
The critical date is thus no more than a technique by which a tribunal can
assess the weight of evidence, thus each case will depend on its particular cir-
cumstances. Moreover, effectivités occurring after the notional critical date may
also be considered as relevant evidence. In the Land, Island and Maritime Fron-
tier case the Chamber accepted that the practices of El Salvador and Honduras
subsequent to the date of their independence (the critical date) could supplement
the effectivités to indicate the location of the frontier.
CONCLUSIONS
This chapter has considered the legal principles developed and applied by inter-
national tribunals when determining the respective limits of territorial sovereignty
of disputing states. The reason for considering these principles at some length is
that, while an international tribunal typically has the primary task of identifying
the exact location of a boundary, it can do so only once the respective rights to
territorial sovereignty on either side of the asserted boundary have been deter-
mined. While tribunals defer to the traditional legal modes of territorial acquisi-
tion, they also take into account principles of equity and contemporary norms
of human rights and self-determination. It is notable, however, that any relevant
treaty or other instrument delimiting a boundary is likely to be accorded priority
in settling the final delimitation. Acts of effective occupation over time by one
claimant state, for example, will not displace an otherwise valid title held by
another. Indeed, an examination of the cases indicates that international tribunals
have been preoccupied in practice with interpreting the terms of boundary treaties
International Law and Territorial Boundaries 189
1. INTRODUCTION
Maps and other cartographic materials can provide vital evidence of the intended
location of territorial boundaries and they have been the subject of detailed
analysis by many international judicial and arbitral tribunals. The arbitrators in
the Beagle Channel Arbitration appreciated maps as a source of ‘great beauty’
and gave detailed attention to the over 400 maps produced by Chile and Argen-
tina in support of their claims (para 136). Traditionally, however, international
tribunals have been cautious in accepting maps as having any value beyond that
of a secondary source of corroborative evidence. As Twiss put it, maps are ‘but
pictorial representations of supposed territorial limits, the evidence of which
must be sought for elsewhere’ (Sandifer 1975:229-230). Sandifer observes that
international tribunals have ‘probably applied severer tests in evaluating maps
than almost any other evidence’ because they are secondary rather than primary
in character and frequently hearsay in nature (1975:229). Caution in giving
probative value to maps is warranted as, in the past, maps have lacked precise
geographical data, reliable cartographical techniques were not available to early
map-makers and maps can be highly self-serving in support of a state’s sovereign
interests (Hyde 1945:496). At best, the evidentiary value of a map was restricted
to the corroboration of facts or conclusions established by other means of proof
(Brownlie 1998:156). As a consequence, maps have typically been afforded only
limited weight as evidence, if any at all, by international tribunals.
The following description of how the frontier between Burkina Faso and Mali
was drawn illustrates the practical difficulties that arise in map-making (Frontier
Dispute case, para 61, quoting from a note dated 27 January 1975 and submitted
to the Chamber by IGN during the proceedings):
with the help of the texts, the cartographers tried to locate the frontier in
relation to the map base. Unfortunately, the inaccuracy of the texts made it
impossible to draw a sufficiently reliable boundary in certain areas. Some
names quoted in the texts could not be found, others referred to villages
which had disappeared or been moved, or again the actual nature of the
terrain (course of rivers, position of mountains) appeared different from
that described in the former itinerary surveys.
192 Chapter 6
that is, even official maps are not necessarily ‘rendered infallible or objectively
correct’ (para 138).
• Absolute, unqualified concordance of map evidence with the asserted claim
is not required; rather, there should be a ‘definite preponderance’ or cumula-
tive impact that ‘tell[s] the same story’ in favour of one state or another (para
139).
This chapter will consider, first, the factors that determine the probative value of
maps and, secondly, each of the identified contexts in which maps can contribute
to delimitation of a boundary between neighbouring states.
The following categories where maps have been afforded evidentiary value
have been gleaned from international judicial and arbitral decisions (Highet 1987;
Brownlie 1998). These categories will be adopted for the purposes of this chapter
as a convenient means of understanding the evolving jurisprudence.
only with the greatest caution can account be taken of maps in deciding a
question of sovereignty . . . If the Arbitrator is satisfied as to the existence
of legally relevant facts which contradict the statements of cartographers
whose sources of information are not known, he can attach no weight to
the maps, however numerous and generally appreciated they may be . . . a
map affords only an indication – and that a very indirect one – and, except
when annexed to a legal instrument, has not the value of such an instrument,
involving recognition or abandonment of rights.
A well-established rule was moreover embodied in Article 29 of the Treaty of
Versailles of 28 June 1919. This rule states that, when there is a discrepancy
concerning a frontier delimitation between the text of a treaty and maps, it is
the text and not the maps which is final.
This point was also made by the Tribunal in the Guatemala-Honduras Bound-
ary Arbitration:
A map is primarily a statement of geographical facts, designed in theory
to present visually the unvarnished truth. Its purpose is to bring home that
truth to the mind through the eye . . .
But the map-maker does not stop at this point. He commonly undertakes to
do much more – to state the political as well as geographical facts. Here
again his duty in such a case is to reveal the truth, relative to national
pretensions or accepted limits and known boundaries. The sources of his
information simply differ, however, from those concerning the purely geo-
graphical facts. The tests of his accuracy are not in the decrees of Nature,
but in those of states. When their claims clash, or are vague or loosely
defined, or only partially revealed, the map-maker, however honest, is at
best a mere guesser who is incapable of making a scientific representation
or picture of more than the scope of national pretensions (counter case of
Guatemala, 341-42, cited in Sandifer 1975:237).
Superior acts of sovereignty were, for example, given preference above maps
published by the Survey of India in the Rann of Kutch Arbitration.
Tribunals have often voiced their concerns about the flaws in maps presented
to them. The Commission in the Eritrea-Ethiopia Boundary Dispute noted (para
3.19) that
Topography is dependent upon the state of knowledge at the time the maps
were made, and particularly with older maps this may have been inadequate.
When man-made features are superimposed, such as places of habitation or
territorial limits, there is room for political factors to play a part. Particularly
in the case of maps portraying a boundary which is in the interests of the Party
responsible for the map, the possibility exists that they are self-serving.
Fear that the map is merely a political tool to advance national pretensions prob-
ably underlies many cases in which other relevant evidence of jurisdiction will
outweigh maps. There is also a concern that there is a risk of ‘multiplication of
maps copying each other’s mistakes’ (Sandifer 1975:239).
and geographical situation of Qatar and Bahrain over such a long period,
it is impossible not to accord it substantial weight in confirming the existence
of a title which Qatar possesses to the Hawar Islands (and to Zubarah as
well). It is in any event impossible to ignore that evidence completely, as
the Court’s Judgment does, without providing any explanation. (Emphasis
in original.)
For these dissenting judges, the failure to accord greater weight to the maps
adduced by the parties was compounded by the Court’s delimitation of a single
maritime boundary based on an incompatible combination of two contradictory
maps in different sectors of the maritime boundary (para 205).
by aerial and satellite photography since the 1950s. But the only result is a
more faithful rendering of nature by the map, and an increasingly accurate
match between the two. Information derived from human intervention, such
as the names of places and geographical features (the toponymy) and the
depiction of frontiers and other political boundaries, does not thereby become
more reliable. Of course, the reliability of the toponymic information has
also increased, although to a lesser degree, owing to verification on the
ground ; but in the opinion of cartographers, errors are still common in
the representation of frontiers, especially when these are shown in border
areas to which access is difficult.
Turning to the objectivity and neutrality of map-makers, the Chamber observed
(at 583):
Other considerations which determine the weight of maps as evidence relate
to the neutrality of their sources towards the dispute in question and the
parties to that dispute. Since relatively distant times, judicial decisions have
treated maps with a considerable degree of caution; less so in more recent
decisions, at least as regards the technical reliability of maps. But even where
the guarantees described above are present, maps can still have no greater
legal value than that of corroborative evidence endorsing a conclusion at
which a court has arrived by other means unconnected with the maps.
While the Chamber was able to articulate the evidentiary role of maps in bound-
ary delimitation disputes, it did not actually rely on the maps produced by the
parties when reaching its conclusions. Rather, at the close of the proceedings
the Chamber identified a paradox which has become familiar in disputes over
boundary delimitation. While a wealth of cartographic material had been pro-
duced in support of the submissions of the parties, they did not reliably identify
a frontier on which the tribunal could base its determination. As the Chamber
concluded, ‘[n]ot a single map’ available to the Chamber can reliably be said
to reflect the intentions of the colonial administration expressed in the relevant
texts concerning the disputed frontier (para 57).
While the Chamber was unable to found its conclusion on the map evidence
provided because of the historical unreliability and inaccuracies of maps (paras
110 and 154), the Chamber referred to two maps that it considered were ‘of
special overall significance’ (para 59). The first was the ‘Blondel la Rougery’
map produced by the Geographical Service of French West Africa in Dakar
(map 6.4). This map had been referred to by the colonial territorial authorities
in an administrative circular. While taking this map into account the Chamber
concluded that the maps did not ‘in themselves possess any particular authority’
(para 60).
The second map was a large-scale map which had been prepared by the
French Institut Géographique National (IGN). The Chamber observed that in
general the map had‘enjoyed the approval of both Parties in its description of
200 Chapter 6
the topography’. The IGN map was also endorsed by the parties as ‘a model of
reliability’ and as offering ‘guarantees of both technical precision and official
authority, since they were compiled by an impartial official body directly con-
nected with the administrative authorities of the period’ (para 61). The Chamber
also gave its endorsement to the map as evidence, regarding it as ‘a visual por-
trayal both of the available texts and of information obtained on the ground’ (para
62). Notwithstanding its quality as a map, the Chamber recognised the limits of
its value as a means of proof, holding that the map ‘in itself is not sufficient to
permit the Chamber to infer that the frontier line depicted . . . in the successive
editions of the IGN map, corresponds entirely with the boundary inherited from
the colonial administration’ (ibid.). The best use to which such a reliable map
could be put was in considering (ibid.):
how far the evidence offered by this or any map corroborates the other evi-
dence produced. The Chamber cannot uphold the information given by the
map where it is contradicted by other trustworthy information concerning
the intentions of the colonial power.
Ultimately, the Chamber was persuaded by the timing of the map’s production and
the neutrality of its authors to give its approval to the evidential use of the IGN
map, but only in precisely defined circumstances. In the words of the Chamber,
‘where all other evidence is lacking, or is not sufficient to show an exact line,
the probative value of the IGN map becomes decisive’ (ibid.).
In the Frontier Dispute case, the Chamber developed an influential approach to
maps as evidence of a territorial boundary. The Chamber’s approach recognises
the significant technical progress that has been made in the art and science of
cartography since the seminal decisions on territorial sovereignty and boundaries
which were made in the first half of the twentieth century. Certainly, the juris-
prudence relating to the evaluation of map evidence in international litigation
has been more robust since this decision and the ICJ has been consistent in its
use of the methodology it has developed.
The lack of official status or authority of a map may militate against its pro-
bative value. In the Frontier Dispute case, for example, Mali submitted a map
which, whilst ‘absolutely positive’ as to the frontier line (para 171), lacked any
authenticating features regarding which official body compiled it or which admin-
istrative authority may have approved this line. The Chamber was impressed by
the technical proficiency of the map’s production, but was unable to ascribe to it
the authoritative status of a document explaining the applicable colonial Order,
because it could not be shown that the map had been produced by the colonial
administration.
By contrast with maps lacking official status, international tribunals are inclined
to accord probative weight to maps drawn up by a boundary commission under
the mandate of the disputing parties as part of a peace agreement. In the Certain
Frontier Lands case between Belgium and Netherlands, the maps had been pre-
pared for the Mixed Boundary Commission established to identify the disputed
boundary. The dispute concerned sovereignty over a number of plots of land
on the territorial frontier. The states had submitted detailed survey maps and
topographical maps, as well as a ‘special map’ of the disputed region. The Court
was not only willing to consider the maps submitted by both parties but also to
attach a certain, and at that time, unprecedented, significance to them. It was of
signal importance, however, that the maps drawn up for the Mixed Boundary
Commission had been annexed to the Convention that was subsequently agreed
between the states parties (Weissberg 1963:788). The reliance placed on these
maps by the Court is therefore best understood as resting upon the integral status
of the maps as part of the treaty itself. The maps were interpreted as part of the
obligation to apply the terms of the treaty rather than as separate corroborative
evidence.
While the ICJ in the Certain Frontier Lands case found the maps to be worthy
of consideration, it stressed, nonetheless, that such maps ‘require most careful
preparation and checking’ (at 220). In addition, the Court appeared to endorse
Belgium’s argument that certain military maps which identified the disputed
plots as Belgian were evidence of its acts of jurisdictional sovereignty in the
disputed area.
Maps produced by officials of government agencies are particularly likely to
be accorded some probative value. The Boundary Commission in the Eritrea-
Ethiopia dispute observed (para 3.21) when stating the general principles upon
which their findings were to be based:
a map produced by an official government agency of a party, on a scale
sufficient to enable its portrayal of the disputed boundary area to be iden-
tifiable, which is generally available for purchase or examination, whether
in the country of origin or elsewhere, and acted upon, or not reacted to,
by the adversely affected party, can be expected to have significant legal
consequences.
202 Chapter 6
Official maps prepared by third states will also be afforded weight, in the view
of the Commission, in circumstances in which an estoppel might be raised. The
Boundary Commission remarked (para 3.21):
a State is not affected by maps produced by even the official agencies of
a third State unless the map was one so clearly bearing upon its interests
that, to the extent that it might be erroneous, it might reasonably have been
expected that the State affected would have brought the error to the attention
of the State which made the map and would have sought its rectification.
As will be discussed further below, the known or knowable existence of maps
contrary to a state’s interests requires some response from that state if it is to
avoid an implication of acquiescence made against it in any later dispute.
Disclaimers
Some unique observations in respect to the official status of maps were made
by the Boundary Commission in the Eritrea-Ethiopia dispute. The Commission
considered the impact of disclaimers that are typically appended to maps, even
to apparently ‘official maps’. Examples include the statement made on UN maps
that they are not to imply ‘official endorsement or acceptance by the UN’ (para
3.26). The Commission considered that (para 3.27):
such disclaimers do not automatically deprive a map of all evidential value.
The map still stands as an indication that, at the time and place the map
was made, a cartographer took a particular view of the features appearing
on the map.
Moreover, where a state has been adversely affected by a map, the disclaimer
could not relieve it of the need to protest against the unacceptable feature. The
Commission also opined that the disclaimer cannot ‘neutralize the fact that
that State itself published the map in question’. Rather, the map remains as a
‘geographical fact’, particularly if the state adversely affected produced and dis-
seminated the map (para 3.28). The Commission thus took a flexible approach
to maps that are accompanied by disclaimers, affording them an evidential value
in certain circumstances.
‘Signature’
It was the acceptance by Thailand of the map delineating the boundary that
rendered it a part of the treaty itself and thereby binding on both states.
As Judge Fitzmaurice recognised, the somewhat surprising effect of the finding
in the Temple of Preah Vihear case was to favour the line in the map above the
language of the treaty itself (at 65-66). It is usual interpretive practice, where
there are discrepancies between the terms of a treaty and a map which has been
incorporated in the text, that an express provision in the treaty will override the
map. But where the treaty is silent, the usual rules of treaty interpretation, under
the Vienna Convention on the Law of Treaties 1969, will be applied to the map
and the text to infer the real intention of the parties at the time of concluding
the agreement. While considerations of acquiescence and estoppel underpinned
the finding that the map had been accepted by Cambodia and Thailand, the case
illustrates how a map can be integrated into a boundary treaty and can thereby
be determinative of the location of the frontier.
that it reached this conclusion independently of the maps which were taken into
account ‘only for purposes of confirmation or corroboration’ (para 163).
was asked to consider the competing claims by Indonesia and Malaysia to two
islands in the Celebes Sea, off the northeast coast of Borneo. A dispute had arisen
in the context of discussions between them over delimitation of the continental
shelf. Crucial to the decision was interpretation of the 1891 Convention between
the Netherlands and Great Britain as the former colonial powers in the region.
The Court remained faithful to its earlier views in the Frontier Dispute case and
discounted most of the maps adduced by the parties as evidence on the grounds
that they were inconsistent and inconclusive. By contrast, significant probity
was accorded to the only map that had been appended to the subsequent 1915
Agreement which, the ICJ found, reflected the officially expressed views of both
states as to the boundary line between them (paras 72, 91).
Thailand was, in effect, estopped from denying the validity of the map, though the
decision was also based on implied recognition, acquiescence, failure to protest,
peaceful occupation, prescription and principles of treaty interpretation.
From these facts, the Court concluded that Thailand had accepted the bound-
ary as represented mistakenly on the map. The effect of the decision was that
acquiescence in Cambodia’s sovereignty over the temple and the location of the
boundary as set out on the map prevailed over the provisions of the boundary
treaty itself. The Court’s application of the principle of acquiescence is impor-
tant because, as Weissberg (1963:802) points out, the effect of the decision is
that, ‘in the interest of certainty, stability, and finality of frontiers, an unsigned
map in derogation of a treaty provision supersedes the text as a matter of treaty
interpretation’. The decision is, however, essentially one founded in equity and
is not authority for any proposition that maps can, without more, take precedence
over a treaty provision.
International tribunals have been willing to use maps as evidence of what the
state parties intended when they negotiated treaties upon which the maps were
subsequently based. Maps can thus provide evidence of the practice of states in
aid of interpretation of a treaty or other agreement settling the frontier. In the
Frontier Dispute case the ICJ gave probative weight, in particular, to a map that
had been drawn up by the IGN, a neutral body, in 1958-60. This was some years
after the frontier had been declared to be that of the former colony of Upper
Volta (Burkina Faso) in 1932. The Chamber employed the IGN map exclusively
as corroborating evidence, but conceded (para 62) that ‘where all other evidence
is lacking, or is not sufficient to show an exact line, the probative value of the
IGN map becomes decisive’.
The Chamber also confirmed (para 63) that maps can be used as evidence of
a boundary in the following terms:
the effectivités can support an existing title, whether written or cartographi-
cal, but when their probative value has to be assessed they must be system-
atically compared with the title in question ; in no circumstances can they
be substituted for the title.
The Chamber’s view of the legal principles regarding maps as evidence in the
Frontier Dispute case was endorsed by the ICJ in the Case Concerning Kasikili/
Sedudu Island between Botswana and Namibia. The Court recognised the poten-
tial role of maps as evidence relevant to interpretation of an earlier treaty that
has been drawn up to settle a boundary. The Court was required to interpret
Article III of the Anglo German Treaty of 1890 which established the boundary
212 Chapter 6
between the successor states Botswana and Namibia. The dispute concerned
whether the boundary, stated to be the ‘main channel’, was to the north or south
of Kasikili/Sedudu Island. As the maps published after the 1890 Treaty had not
purported to interpret the phrase ‘main channel’, and as local authorities did not
accept the maps as accurate, the Court rejected them as evidence of subsequent
practice (para 85). The Court concluded (para 87):
In view of the absence of any map officially reflecting the intentions of the
parties to the 1890 Treaty and of any express or tacit agreement between
them or their successors concerning the validity of the boundary depicted
in a map . . . and in the light of the uncertainty and inconsistency of the
cartographic material submitted to it, the Court considers itself unable to
draw conclusions from the map evidence produced in this case.
Yet again, while the ICJ has consistently recognised the probative potential of
maps as evidence of the subsequent interpretation of a boundary treaty, the maps
actually produced by the claimants failed the tests of official status, accuracy
and relevance.
More positively, the Boundary Commission in Eritrea-Ethiopia found that all
relevant maps supported the Eritrean claim line and that, as there was no timely
objection by Ethiopia to this line, and as other maps were consistent with it,
the maps amount to ‘subsequent conduct or practice of the Parties evidencing
their mutual acceptance of a boundary corresponding to the Eritrean claim line’
(para 5.88).
As Shaw (2007:783-784) points out, the interesting aspect of the findings of
the Commission is that, where the treaty is ambiguous, the subsequent practice
of the states parties evidenced in their response to maps based on the treaty can
be determinative of what they intended the treaty to mean.
International tribunals have long relied on maps for their evidential value in
assisting in interpretation of boundary treaties. In Cameroon v Nigeria the ICJ
employed a sketch-map to locate the course of the River Kohom, which had been
identified as the boundary between these states in a delimitation treaty known as
the Thomson-Marchand Declaration. As the text of the Declaration did not provide
a clear identification of the River Kohom, the Court referred to a sketch-map that
had been prepared by French and British officials. The map had been used when
drafting the Declaration, although it was not annexed to or an integral part of it.
Nigeria argued that the sketch-map, nonetheless, showed what the intention of
the parties was at the time when they referred to the ‘River Kohom’. The Court
compared the sketch-map – which clearly indicated the relief of the area and the
direction of the river – with other maps provided by the Parties (see maps 6.1
Maps as Evidence of Territorial Frontiers 213
and 6.2). The Court was then able to identify the course of the river and thus of
the boundary between the States (para 101).
Similarly, the Eritrea-Ethiopia Boundary Commission acknowledged the
evidentiary value of the Mai Daro map that had been used during negotiations
to identify the course of the Mai Teb river boundary (para 3.21). The similar-
ity of this map with the de Chaurand map of 1894, confirmed the view of the
Commission that the river as delineated on the maps formed an agreed basis for
negotiation of the boundary in the 1900 Treaty map.
Maps may provide evidence of certain objective facts. Where the map is relevant
evidence of that fact, and meets the standards of neutrality and objectivity, inter-
national tribunals have distinguished the Frontier Dispute case between Burkina
Faso and Mali and relied upon that map without seeking further corroborating
evidence. In the Land, Island and Maritime Frontier Dispute case between El
Salvador and Honduras (see maps 6.7 and 6.8), the Chamber attributed legal
value to an expedition report that had been combined with a map submitted by
Honduras. The map, when considered alongside a report from the expedition
which led to its preparation, left ‘little room for doubt’ as to the course of the
Goascorán river constituting the boundary (para 316). The Chamber emphasised
that this map was only ‘a visual representation of what was recorded in the
contemporary report’ and did not purport to indicate any frontiers or political
divisions. The Chamber sought to distinguish the earlier dicta of the Chamber
in the Frontier Dispute case to the effect that ‘maps can still have no greater
legal value than that of corroborative evidence endorsing a conclusion at which
a court has arrived by other means unconnected with the maps’ (Frontier Dis-
pute case, para 56).
Rather, the Chamber considered that the principle of corroboration applied
‘in the context of maps presented “as evidence of a frontier” ’ (para 316) and
would not apply where there was no intent in drawing the map to identify any
such boundary. On the basis of this distinction, the Chamber concluded that
the evidence of the expedition report, combined with the map, could be used
to reach a determination of the location of the river. Of importance in making
this distinction was the view that ‘there is no apparent possibility of toponymic
confusion, and the fact to be proved is otherwise a concrete, geographical fact’
(para 316). It seems that the Chamber of the ICJ will give probative evidential
value where a map is evidence of a fact, without the need to demonstrate that it
also corroborates other independent evidence.
In relation to numerous other antiquated maps introduced during the Land,
Island and Maritime Frontier Dispute case, the Chamber explicitly followed
the reasoning in the Frontier Dispute case between Burkina Faso and Mali by
214 Chapter 6
CONCLUSIONS
INTRODUCTION
upon whether the construction of the pulp mills presented an ‘imminent threat
of irreparable damage to the aquatic environment’ of the River Uruguay. While
the Court considered that there was no such threat at the time, and the case has
yet to be finally determined, the reasoning of the tribunal confirms the obliga-
tions to protect the ecosystems of shared river boundaries while also allowing
for sustainable economic development.
This chapter examines state practices in delineating river boundaries, typically
through the negotiation of bilateral treaties. Efforts by international bodies, such
as the International Law Commission, the Institute of International Law and
the International Law Association, to develop guidelines for the management
of river boundaries, are also considered as precedents for the conclusion of the
framework 1997 United Nations Convention on the Non-Navigational Uses of
International Watercourses.
While the principal concern of this work lies with the delimitation of territo-
rial boundaries, the particular legal issues raised by river boundaries warrant
consideration of the evolving principles of international law regulating the use
of shared water resources. This chapter provides some international case stud-
ies to illustrate the kinds of disputes that have been submitted for resolution by
international courts and tribunals. The jurisprudence arising from these cases is
examined to distil the international principles of equitable and reasonable use
of international watercourses.
State practice is that those negotiating river boundaries have selected one of three
principal means of locating the exact boundary:
• Geographic middle of the river or medium filum acquae
• Middle of the channel or thalweg
• Shore or bank of the river.
Hugo Grotius, the Dutch jurist, is credited with identifying the boundary by
using the geographic middle of the river; that is, the boundary is determined
by drawing a median line, every point of which is equidistant from the nearest
points on the opposite shores. As Grotius (1625: vol. 2, book 2, ch. 2, section
18) observed, ‘In case of any doubt, the jurisdictions on each side reach to the
middle of the river that runs betwixt them’.
River Boundaries and International Law 217
1.2 Thalweg
In some exceptional cases, the river will lie entirely within the territory of one
state, so that the boundary is formed by the shore of the other state. This will
typically arise where one state, the original holder of the territory, cedes territory
on one side of the river to another state, retaining the river within its sovereignty.
The Spanish-American Treaty of 1819, for example, provided that the Red River
and all the islands within it belonged to the United States. The boundary of Texas
is thus taken to be the south bank of the Red River (Shalowitz 1964:376). As
McCaffrey points out (2001:71), such boundaries can lead to continuing instability.
218 Chapter 7
Kasikili/Sedudu Island
The International Court of Justice was asked by Botswana and Namibia to determine,
on the basis of the Anglo/German Treaty of 1890 and of international law, the boundary
between them around the Kasikili/Sedudu Island in the Chobe River. The respective
spheres of colonial influence of Britain and Germany had been divided by the 1890
Treaty at the ‘main channel’ of the River Chobe, the English interpretation being
the ‘centre’ of the main channel and the German version being the ‘thalweg’ of that
channel. Some 110 years later, Botswana and Namibia each adopted differing means
of interpreting these terms (see map 7.1). The Court noted (para 24-25) that:
various definitions of the term ‘thalweg’ are found in treaties delimiting boundaries
and that the concepts of the thalweg of a watercourse and the centre of a watercourse
are not equivalent. The word ‘thalweg’ has variously been taken to mean ‘the most
suitable channel for navigation’ on the river, the line ‘determined by the line of deepest
soundings’, or ‘the median line of the main channel followed by boatmen travelling
downstream’. Treaties or conventions which define boundaries in watercourses nowa-
days usually refer to the thalweg as the boundary when the watercourse is navigable
and to the median line between the two banks when it is not, although it cannot be
said that practice has been fully consistent.
. . . [A]t the time of the conclusion of the 1890 Treaty, it may be that the terms ‘centre
of the [main] channel’ and ‘Thalweg’ de Hauptlaufes were used interchangeably. . . .
Indeed, the parties to the 1890 Treaty themselves used the terms ‘centre of the channel’
and ‘thalweg’ as synonyms, one being understood as the translation of the other.
On this basis, the Court proceeded to determine the meaning of the words ‘main
channel’ by reference to the commonly used criteria in international law, particularly
the rules of interpretation agreed in the 1969 Vienna Convention on the Law of Trea-
ties (Article 31). As the natural features of a river vary markedly, the Court found
that it could not rely on one criterion only. Rather it examined the mean depth and
width (based on the low water baseline) of the river, and factors such as the role of
a 50-metre high escarpment and favourable conditions for navigability. A review
of the object and purpose and travaux preparatoires of, and subsequent practices of
the states parties to, the 1890 Treaty confirmed the Court’s view that this agreement
was intended to establish a frontier line in addition to delineating spheres of colonial
influence. Relying upon evidence of navigability and on-site investigations, the Court
finally concluded, eleven votes to four, that the northern channel of the River Chobe
was to be regarded as the main channel for the purposes of the 1890 Treaty and that
the thalweg, as formed by the line of deepest soundings, marks the boundary. As the
ICJ rejected, eleven votes to four, Namibia’s argument that the island in dispute was
part of its territory under the international principles of prescription, the Kasikili/Sedudu
Island necessarily forms part of the territory of Botswana (para 351).
River Boundaries and International Law 219
In the Iran-Iraq Protocol of 1975 (14 ILM 1133), for example, the Shatt-al-Arab
waterway, which had been within the exclusive sovereignty of Iraq, was revised
to adopt the thalweg principle. Other examples of river bank boundaries are
discussed in Chapters 4 and 10.
Formal legal rules such as avulsion and accretion have been moderated in their
effect by wider considerations of equity, a point that is illustrated by the Chamizal
Arbitration case (discussed in Chapter 5). Here the Rio Grande slowly changed its
course, creating a 600-acre tract of land, known as the Chamizal Tract, formed on
the US side of the river. While the International Boundary Commission decided
that where the tract resulted from accretion, title lay with the United States and
where it resulted from a flood, title remained with Mexico, there were additional
questions as to the legal effect of the equitable considerations of acquiescence and
protest on title. The United States had, in addition to reliance upon the principle
of accretion, founded its claim to the tract on the United States’ peaceful and
uninterrupted possession of the land known as ‘prescriptive title’.
The International Boundary Commission dismissed this element of the United
States’ claim in the following terms (pp. 806-807):
220 Chapter 7
the political control exercised by the local and federal governments, have
been constantly challenged and questioned by the Republic of Mexico,
through its accredited diplomatic agents . . . From [1867] until the negotiation
of the Convention of 1884, a considerable amount of diplomatic correspon-
dence is devoted to this very question, and the Convention of 1884 was an
endeavour to fix the rights of the two nations with respect to the changes
brought about by the action of the waters of the Rio Grande.
The very existence of that Convention precludes the United States from
acquiring by prescription against the terms of their title and . . . the two
republics have ever since the signing of that convention treated it as a
source of all their rights . . .
Another characteristic of possession servicing as a foundation for pre-
scription is that it should be peaceable. . . . [H]owever much the Mexicans
may have desired to take physical possession of the district, the result of any
attempt to do so would have provoked scenes of violence and the Republic
of Mexico cannot be blamed for resorting to the milder forms of protest
contained in its diplomatic correspondence.
The Commission, it seems, was not willing to recognise the acquisition of territory
over the Chamizal tract on the sole basis of possession, where Mexico had done
all it reasonably could, short of the use of force, to protest against recognition of
United States’ title. Indeed, an international tribunal might be expected to have
favoured the state that had chosen the path of lawful and peaceful negotiations
above a forceful taking.
and resource potential become available, more precise delineations can be made
by the riparian states.
Commentators such as Caflisch suggest that the lack of consistency in such
treaty provisions demonstrates that there is not yet any rule of customary interna-
tional law on the means by which a river boundary is to be determined (McCaf-
frey 2001:72) The better view, however, is that state practice and prevailing
opinion support the adoption of either the median line or thalweg principles as
the recognised means of identifying an international boundary river.
While states have adopted notional lines in river boundaries to delineate the
extent of their respective sovereignty and jurisdiction, the practical fact remains
that exploitation of the water resource is necessarily shared. An analogy with
the rights of neighbouring states to an oil deposit which straddles an agreed
boundary may be apt to illustrate the problem of shared water resources of river
boundaries. Either state may exploit the resource, by extracting a greater share
of the oil or water than it is technically entitled to, to the significant detriment of
the other. Hardin (1968:1244) describes this as the ‘tragedy of the commons’ in
which each state has, in theory, an incentive to withdraw ever-increasing amounts
of water. There is the added risk that states will be tempted to over-fish or to
pollute the water of a boundary river even, curiously, where this may inure to
their disadvantage (McCaffrey 2001:74). Contemporary needs for irrigation and
hydroelectric power also place ever increasing pressures on the equitable use
of such shared water resources as recent international disputes between riparian
states illustrate.
Global and domestic concerns to protect the environment have prompted late
twentieth century efforts to moderate traditional and absolute approaches to ter-
ritorial sovereignty over river boundaries. Some normative rules of international
law have evolved including the obligation to exercise rights over shared water
resources in an equitable and reasonable way relative to the interests of other
states. The Institute of International Law, for example, has stated that, ‘Neither
state may, on its own territory, utilize or allow the utilization of the water in
such a way as seriously to interfere with its utilization by the other State or by
individuals, corporations, etc. thereof’ (Madrid Resolution 1974:200, article 1).
The Institute subsequently agreed, through resolutions of 1961 and 1979,
upon the core principle that water resources should be used in an equitable and
reasonable manner. Importantly, the Institute also recognised the practical value
of establishing joint management regimes as a means of protecting the interests
of all riparians. Further efforts to regulate the shared waters of a boundary river
222 Chapter 7
were made in the Convention of 31 May 1976 on the Regulation of Water Man-
agement Issues of Boundary Waters.
Also adopting the principle of equitable utilization have been the resolutions
of the International Law Association (ILA) which, in 1966, adopted a set of
articles known as the Helsinki Rules on the Uses of the Waters of International
Rivers. These Rules articulated the principle of ‘reasonable and equitable share’
as a primary obligation of international watercourse law (article 4). This seminal
work laid the foundations for the International Law Commission’s 1994 draft
articles on the non-navigational uses of international watercourses which, in
turn, led to the adoption of the Convention on the Law of the Non-Navigational
Uses of International Watercourses by the United Nations General Assembly
on 21 May 1997.
reasonable use, including social and economic needs, natural phenomena such
as climate and ecology of the region, the interests of any dependent populations,
existing and potential uses and the conservation of water resources. In meeting
the standard of equitable and reasonable use, the Convention provides that,
absent any agreement to the contrary, ‘no use of an international watercourse
enjoys inherent priority over other uses’ (article 10). Inevitably, however, there
will be conflicts between uses. The Convention stresses the need to have special
regard to the ‘requirements of vital human needs’ and to resolve any conflicts
by reference to the core principles and identified factors set out in articles 5, 6
and 7 (article 10).
Importantly, the Convention recognises and adopts the customary international
obligation not to cause significant harm by providing, in article 7, that watercourse
states are to ‘take all appropriate measures to prevent the causing of significant
harm to other watercourse States’. Where such harm, nonetheless, arises, the
state causing the harm is bound to eliminate or mitigate the damage. Reflecting
the uncertain state of international law on the issue, the Convention parties were
not able to agree upon any obligation to compensate for harm. Rather, article 7
requires the state causing harm to ‘discuss the question of compensation’ with
the affected state. Parties do, however, have an obligation to prevent and miti-
gate harmful conditions, including waterborne diseases and desertification, and
to notify other states of emergency situations (article 27).
In addition to the obligation not to cause significant harm is the positive
obligation, individually and jointly, ‘to protect and preserve the ecosystems of
international watercourses’ (article 20) More specifically, watercourse states are
bound (article 21) to:
prevent, reduce and control the pollution of an international watercourse that
may cause significant harm to other watercourse States or to their environ-
ment, including harm to human health or safety, to the use of the waters for
any beneficial purpose or to the living resources of the watercourse.
Implementation of the obligation not to cause pollution is to be achieved through
consultations, for example, to set joint water quality objectives and to list sub-
stances that are to be prohibited or monitored. The importance of preservation
of the ecosystem as a whole has also been recognised by the obligation of
watercourse states to prevent the introduction of alien species and to protect
the marine environment, taking into account international rules (articles 22 and
23). Special provisions have been agreed for regulation of the management and
flow of waters of an international watercourse and for the protection of related
installations.
224 Chapter 7
While the obligations not to pollute, to prevent significant harm, to ensure equi-
table and reasonable use and to cooperate are broadly recognised in international
state practice, it is not clear how they will relate to each other. How, for example,
will states balance their competing interests in hydroelectric power and irrigation
against environmental impacts? Is causing environmental harm, per se, wrongful
under the Convention and, if not, is compensation nonetheless payable? The Con-
vention does not attempt to articulate precise rules for resolution of such issues,
but adopts the practical approach of creating a process, including fact-finding and
dispute resolution, through which these apparently conflicting interests might be
balanced by state to state negotiations (McCaffrey 2001:309).
Integral to this process of resolution of potentially competing interests is the
Convention emphasis upon cooperation and exchange of information. Article
8 requires watercourse states to cooperate to achieve ‘optimal utilization and
adequate protection’ and to consider establishing joint mechanisms or commis-
sions to do so. Such joint management might include planning for sustainable
development and otherwise promoting the rational and optimal use, protection
and control of the watercourse (article 24).
Linked to the obligation to cooperate is the requirement that there should
be a regular exchange of data and information on the condition of the water-
course and its water quality. Moreover, states are required to employ their ‘best
efforts’ to comply with requests for data (article 9), though they may condition
compliance upon payment for the reasonable costs of collecting the informa-
tion. Watercourse states are also bound to notify states where planned measures
‘may have a significant adverse effect upon other watercourse States’ and to
provide them with available data and the results of any environmental impact
assessment (article 12). Any such notification must allow the notified state six
months within which to study and evaluate the effects of the planned measures
and to respond to the notifying state. During this period, no action may be taken
to implement the proposed measures. If the notified state responds to the effect
that the measures are inconsistent with articles 5 or 7, both the notifying and the
notified state are required to enter into good faith consultations or negotiations
to achieve an ‘equitable solution’ (article 33), paying ‘reasonable regard to the
rights and legitimate interests of the other state’ (article 17 (2)). If the notified
state does not respond, the notifying state is free to go forward with its planned
measures. Through this process of notification, the Convention has created a
dispute resolution process that depends, in the final analysis, upon the good will
of the states involved.
While the Convention establishes a regime for notification under article 12,
it wisely anticipates the possibility that a watercourse state will not comply.
Article 18 provides a procedure in the absence of notification where a watercourse
state has reasonable grounds for believing that another watercourse state is, in
River Boundaries and International Law 225
fact, planning measures that ‘may have a significant adverse effect upon it’. If
so, the potentially affected state may request the state planning such measures to
comply with the article 12 procedure for notification. Again, resolution of any
differences depends upon consultation and negotiation. Of practical value is the
provision that, until resolution of any differences is achieved, the state planning
adverse measures may not implement them for up to six months. Balancing this
moratorium on carrying out new activities, the Convention also permits a state that
is planning measures to implement them immediately if it can be demonstrated
that they are of the ‘utmost urgency in order to protect public health, public safety
or other equally important interests’ (article 19). Where this stringent test can be
satisfied, there nonetheless remains an obligation to enter into consultations and
negotiations as required under the article 17 provision for notification.
5.3 Summary
the Meuse contrary to the Netherlands’ interests. While the Court explicitly
confined itself to interpretation of the 1863 Treaty between these states govern-
ing diversions of water from the Meuse, it recognised a wider principle that the
upstream state could not divert waters within its territory if do so would diminish
downstream state rights to the normal flow of these waters.
The views of the Arbitral Tribunal in the Lac Lanoux dispute between France
and Spain in 1957 have also provided benchmarks for balancing the respective
interests of riparian states in boundary river resources. The dispute arose over
the respective rights of the states parties to the 1886 Treaty of Bayonne under
which Spain and France were accorded sovereignty over the waters within their
boundaries, particularly Lake Lanoux in the Pyrenees. The downstream riparian,
Spain, was to have a right to the ‘natural waters which flow from higher levels
without the hand of man having contributed thereto’, so long as the rights of
the upstream state, France, were not harmed. Any new proposed works were to
be subject to consultation. After the Second World War, France dammed Lake
Lanoux to increase the flow of water for hydroelectricity, thereby restricting
Spain’s use of these waters for irrigation.
The Tribunal (para 22) set out a number of guiding principles for resolution
of the dispute:
Account must be taken of all interests, of whatsoever nature, which are liable
to be affected by the works undertaken, even if they do not correspond to
a right. . . . .
The Tribunal is of the opinion that, according to the rules of good faith,
the upstream State is under the obligation to take into consideration the
various interests involved, to seek to give them every satisfaction compat-
ible with the pursuit of its own interests, and to show that in this regard it
is genuinely concerned to reconcile the interests of the other riparian State
with its own.
The Tribunal considered Spain’s argument that there is a customary law obliga-
tion not to cause serious harm, but concluded (para 129) that:
if it is admitted that there is a principle which prohibits the upstream State
from altering the waters of a river in such a fashion as seriously to preju-
dice the downstream State, such a principle would have no application to
the present case, because . . . . the French scheme will not alter the waters
[in question].
France had not infringed the treaty rights of Spain, in the Court’s view, as it
had not been ‘clearly affirmed that the proposed works would entail an abnormal
228 Chapter 7
risk in neighbourly relations or in the utilization of the waters’ (para 6). While
the views of the Tribunal are useful in stressing the importance of good faith
negotiations to reconcile conflicting interests, it should be recalled that its final
decision rested upon an interpretation of the Treaty of 1886, not explicitly upon
any principles of customary international law.
In making this decision, the Court provided valuable comments on the principles
that apply where water resources are shared between two states. The principles
recognised by the ICJ in this case are summarised below.
Proportionality: The ICJ also stressed the importance of the principle of propor-
tionality, arguing (para 85) that:
Czechoslovakia, by unilaterally assuming control of a shared resource,
and thereby depriving Hungary of its right to an equitable and reason-
able share of the natural resources of the Danube – with the continuing
effects of the diversion of these waters on the ecology of the riparian area
of Szigetköz – failed to respect the proportionality which is required by
international law.
Sustainable development: The Court emphasised the need for riparian states to
reconcile their needs for economic development with the objective of sustain-
able development and stated (para 141) that the Parties should ‘find an agreed
solution that takes account of the objectives of the Treaty. . . . as well as the
norms of international environmental law and the principles of international
watercourses.’
Citing article 5(2) of the Convention on International Watercourses, the ICJ
required the Parties to re-establish the joint regime created by the 1977 Treaty.
It urged them to ‘take all necessary measures to ensure achievement of the
objectives of the Treaty’ including the concept of common utilization of shared
water resources.
In his Separate Opinion, Vice-President Weeramantry elaborated on the view
of the majority of the Court that reconciliation of economic development with
environmental protection is to be achieved through the ‘concept’ of sustainable
development. Judge Weeramantry argued that sustainable development is a
‘principle’ of reconciliation of sometimes conflicting norms that, with world-
wide recognition, has the status of modern international law (p. 88). Whether
the notion of sustainable development is a ‘concept’ or a ‘principle’, it is an
effective tool with which to balance competing norms regulating the shared use
of water resources of a boundary river.
230 Chapter 7
Obligation to enter into good faith negotiations to resolve disputes: the ICJ in the
Gabčíkovo-Nagymaros case confirmed the evolving jurisprudence of international
courts and the Convention on International Watercourses that requires states to
enter into good-faith and meaningful negotiations to resolve disputes over access
to natural resources. The Court drew attention to the idea of ‘equitable participa-
tion’ set out in Article 5(2) of the Convention on International Watercourses:
States shall participate in the use, development and protection of an inter-
national watercourse in an equitable and reasonable manner. Such par-
ticipation includes both the right to utilize the watercourse and the duty to
cooperate in the protection and development thereof.
Summary
CONCLUSIONS
NORTH AMERICA
recorded that Britain was left in the position occupied by France in 1697, with
the American colonists playing the British role of that period.
It was natural then that the United States should press for the same bound-
aries between themselves and Canada as Britain had claimed against the
French, and the ultimate result was that Great Britain retained Quebec,
Nova Scotia and Newfoundland of 1763, but lost the Illinois country and the
lands south and west of the Great Lakes that had been included in Quebec
by the Act of 1774. (Nicholson, 1954, 18)
Paullin (1932, 52-5) has provided a well illustrated account of the negotiations
that led to the definitive Treaty of Peace on 3 September 1783. On 19 March
1779 the American Congress defined the boundary to be claimed at the end of
the war. From the Bay of Fundy the proposed boundary lay close to meridian
67º west as far as the junction of the watershed south of the St Lawrence River.
This watershed was then followed southwestwards to the Connecticut River, that
was followed to parallel 45º North. This parallel formed the boundary westwards
to the St Lawrence River and from this junction the boundary passed directly to
the south end of Lake Nipissing. From this point the boundary went directly to
the source of the Mississippi River (Paullin, 1932, 52).
By 8 October 1782, after many proposals and counter proposals by both sides,
and some contributions by French and Spanish diplomats, the American and
British representatives agreed on a line very close to that defined by the Ameri-
can Congress three years earlier. The British Government declined to accept the
new line, and one month later after some minor adjustments an agreement was
adopted by both parties that was settled formally in 1783 (Figure 8.1).
The new boundary followed the course of the earlier line as far as the St
Lawrence River. It then followed that river and passed through the middle of
Lakes Ontario, Erie, Huron and Superior to the Lake of the Woods and on to
the source of the Mississippi River. An alternative line was offered to Britain
at the same time, which would have given it parts of Wisconsin and increased
the area of Ontario. The British authorities declined the offer, preferring to keep
riparian rights in all the Great Lakes except Michigan.
The Treaty of London, signed on 19 November 1794 raised two difficulties
(Douglas, 1930, 9). The first related to the possibility of the Mississippi not
extending as far north as the parallel extending westwards from the Lake of the
Woods. The second concerned the identity of the St Croix River,that had been
a problem since 1764. Both countries nominated a commissioner to deal with
the St Croix difficulty and they then chose a third commissioner. Four years
later the three commissioners decided that the River Schoodiac and its northern
branch called Cheputnaticook were the true St Croix River. A monument was
erected at its source.
Ermen (1990, 82-3) reproduces a 1797 map, at a scale of 1:6.9 million, show-
ing the boundary from the coast to the Lake of the Woods. It was published by
William Faden of London, publisher of a series of boundary maps from 1784.
The Americas 235
Parallel 45º north had been marked in the period 1771-4 by the chief surveyors
of Quebec and New York, who were respectively called Thomas Vallentine and
John Collins. In 1802 this line was found to be as much as three-quarters of a
mile north of the true parallel. As a contribution to compromise Britain agreed
that the Vallentine-Collins line should stand (Paullin, 1932, 62).
The Treaty of Ghent signed on 24 December 1814 provided for the final adjust-
ment of the boundaries described in the 1783 Treaty by commissioners. There
were three main questions to be decided. First there was the issue of the owner-
ship of Grand Manan, an island in the Bay of Fundy, and certain islands in the
Bay of Passamaquaoddy, an inlet on the Bay of Fundy. The second uncertainty
involved the course of the boundary from the source of the St Croix River to
the St Lawrence River. The third related to the course of the boundary overland
to the Lake of the Woods.
The first commission awarded Moose, Dudley and Frederick Islands to the
United States. The remaining islands in Passamaquoddy Bay and the Island of
Grand Manan were awarded to Britain. The third commission produced a detailed
description of the boundary from the junction of parallel 45º north with the St
Lawrence River to the entrance of Lake Superior.
The second issue was referred to the King of the Netherlands in 1829 (Douglas,
1930, 17). The award was delivered in 1831.
However disposed the Government of the United States might have been to
acquiesce in the decision of the arbiter, it had not the power to change the
boundaries of a State without the consent of the State. Against that altera-
tion the State of Maine entered a solemn protest by resolution of January
19, 1832, and the Senate of the United States accordingly refused to give
its assent to the award. (Douglas, 1930, 17)
Eleven years later the Webster-Ashburton Treaty of 1842 settled the boundary
from the source of the St Croix River to the St Lawrence River. The result was
that the United States received about 1,000 square miles fewer than the award
of the Dutch King. However, Maine benefited from reimbursement for boundary
surveys and the receipt of a share of the revenue generated from timber cut in
the disputed area.
The United States regarded parallel 49º north as the boundary with British
possessions (Paullin, 1932, 60). Paullin (1923) described the origin of parallel
49º north as a boundary, and Nicholson (1954, 26) observed that it had been
so frequently marked on maps as a boundary before 1803, that its acceptance
was unsurprising. The boundary from Lake of the Woods to the Stony [Rocky]
Mountains was defined in an agreement dated 20 October 1818 (Jones, 1932).
Article II. It is agreed that a line drawn from the most northwestern point
of the Lake of the Woods, along the forty-ninth parallel of north latitude, or
if the said point shall not be the forty-ninth parallel of north latitude, then
that a line drawn from the said point north or south, as the case may be,
236 Chapter 8
until the said line shall intersect the said parallel of north latitude . . . from
the Lake of the Woods to the Stony Mountains. (Douglas, 1930, 13)
The northwestern point was located by two British surveyors and the American
authorities accepted it. It was about 27.5 miles north of the 49th parallel in a
swamp. A point was fixed about 4,600 ft to the south, where a pile of logs 12 ft
high and 7 ft square was established. An iron monument was established close to
the wooden pile. In 1912 this iron marker was recovered and set in concrete. Its
location was at latitude 49º 22’ 39.6” and longitude 95º 09’11.6”. The boundary
was carried from this location to the 49 th parallel by 13 iron markers (Douglas,
1930, 14).
The American authorities pressed for the parallel to be continued to the Strait
of Juan de Fuca on the Pacific Coast. Britain preferred a boundary that coincided
with the Columbia River that reached the coast 250 km south of the Strait. The
United States continued to offer parallel 49º north, and after four refusals, Britain
proposed its use in 1846 (Nicholson, 1954, 29). The parallel’s continuation to
the mainland coast was confirmed in the treaty of 15 June 1846 (Jones, 1937).
The boundary was continued through the Strait to the western entrance of Juan
de Fuca.
The remaining positional problem occurred in Juan de Fuca Strait. The
agreement of 15 June 1846 carried the boundary along parallel 49º north from
the Rocky Mountains to the mainland coast at the head of Juan de Fuca Strait
(Parry, 1977, 36). This agreement did not specify which of the two main chan-
nels carried the boundary to the open sea. The two countries could not settle this
problem and referred the matter to Emperor William of Germany in 1871. By
his decision of October 1872 the Emperor selected Haro Strait, flowing between
the San Juan Islands and the islands immediately adjacent to Vancouver Island
for the course of the boundary. The seaward terminus of the boundary was fixed
as an equidistance point between two nominated points that are ‘. . . nearly due
North and South true’ (Parry, 1977: 39). The Canadian point was Bonilla Point
on the south coast of Vancouver Island, 2 nm southeast of Carmanah Point. The
American point was Tatooch Lighthouse on a small island of the same name,
that is 3.6 cables northwest of Cape Flattery. A cable is 185 metres long in the
British and German Navies and 218 metres in the American Navy. The American
measure was applied in this case. The terminus appears to be located near 48º
North and 124º 40’ West, and it is from this point that the American-Canadian
maritime boundary will be delimited.
Settlement of the boundary between the United States and Mexico was more
complicated than the settlement with Canada (Figure 8.2). The boundary evolu-
tion began in 1847. The previous history of the borderland, through which the
boundary was drawn, has been described and interpreted by Bancroft (1884),
Bannon (1970), Bustamante (1979) and Spicer (1962). House (1982) provided a
useful summary of the earlier history of this region when it was either a political
or settlement frontier.
The Americas 237
A state of war legally came into existence between Mexico and the United
States on 13 May 1846. In less than a year the American forces had advanced on a
wide front and secured the Mexican Provinces of New Mexico, Upper and Lower
California, Coahuila, Tamaulipas, Nuevo Leon and Chihuahua. The American
government appointed a commissioner, who would remain with the army, and be
ready to accept any opportunity for negotiating a satisfactory peace (Miller, 1937,
261). The conditions that the United States Government would find satisfactory
were carefully laid down in a draft agreement given to the commissioner. This
discussion is only concerned with the territorial provisions.
At that time the de jure boundary between the two states was that promulgated
in 1819. It was coincident with the Sabine, Red and Arkansas Rivers and lati-
tude 42º North. Under Article IV of the draft treaty, the United States sought a
southward extension of the boundary to include all of Texas,that had joined the
Union in 1945, Mexico and Upper and Lower California.
The boundary line between the two Republics shall commence in the Gulf of
Mexico, three leagues from land opposite the mouth of the Rio Grande, from
thence up the middle of that river to a point where it strikes the Southern
boundary of New Mexico, thence westwardly along the Southern boundary
of New Mexico, to the South Western corner of the same, thence Northward
along the Western line of New Mexico until it intersects the first branch of
the river Gila, or if it should not intersect any branch of that river, then to
the point on the line nearest to such branch and thence in a direct line to the
same and down the middle of said branch of said river until it empties into
the Rio Colorado, thence down the middle of the Colorado and the Middle
of the Gulf of California to the Pacific Ocean. (Miller, 1937, 265)
In addition to this territorial gain, the United States sought to secure transit rights
for American citizens and goods across the Istmo de Tehuantepec, which is about
220 km wide. The draft agreement represents the maximum concessions America
hoped to gain. However, the government indicated that it would be satisfied with
less and outlined a series of payments that could be offered to Mexico depending
upon the territory and the rights secured. The gain of Upper and Lower California
and New Mexico and transit rights over the Istmo de Tehuantepec would be worth
$US 30 millions. The three provinces alone, or Upper and Lower California and
transit rights across the Peninsula would be worth $US 25 millions. Finally Upper
California and New Mexico would be worth $US 20 millions.
If it proved impossible to secure Lower California, the conclusion to the
boundary description would be altered.
. . . to a point directly opposite the division line between Upper and Lower
California; then, due West, along the said line which runs north of the
parallel of 32º and South of San Miguel to the Pacific Ocean. (Miller, 1937,
263)
238 Chapter 8
A map at a scale of 1:7 million by Colton, dated 1862, shows that San Miguel
was located about 10 miles south of latitude 32º North (Ermen, 1990, 108). The
final terminus on the Pacific Coast was near 32º 30’ North. While the intention
of the boundary definition was clear, Boggs (1940) exposed the problems that
the definition contained. They included identifying the middle any river of even
modest width, the imprecision of some internal Mexican boundaries, such as
the southern boundary of New Mexico and the trials of locating the point on
the western boundary of New Mexico nearest any tributary of the River Gila. In
fact the safeguard regarding the Gila River was unnecessary because the river
originated 105 km east of New Mexico’s western boundary.
Further choices were suggested to the commissioner with the army. In order to
gain the Paso del Norte and the whole of the Gila River, that had been identified
as a good route to the Pacific Ocean, it was suggested that the boundary should
follow the Rio Grande to parallel 32º and along that latitude to the middle of the
Gulf of California. This line could be extended across the Californian Peninsula
if Lower California was not available. However, it was regarded as essential
that the United States had uninterrupted access through the Gulf of California
and that San Diego was secured. This course was recommended to prevent any
dispute about the southern boundary of New Mexico, which, so far as the United
States was aware, had never been ‘authoritatively and specifically determined’
(Miller, 1937, 770).
At the first meeting between American and Mexican commissioners the latter
revealed their government’s proposals. There were probably a number of possi-
bilities but two indispensable conditions were set out that would have prevented
even the minimum American conditions forming a basis for discussion. First the
Mexicans required a neutral strip of territory adjacent to the north bank of the
Rio Grande. It was designed to provide military protection against the United
States and to restrict smuggling, which would reduce Mexico’s revenue and injure
their manufacturing industry. Second, Mexico required a land connection between
Lower California and Sonora around the head of the Gulf of California.
Instead of breaking off the negotiations, the American commissioner exceeded
his instructions and submitted to his government for consideration a line that met
the Mexican conditions. Historians have undoubtedly judged the commissioner:
a geographer can only deal with the results of his action. The recommended
boundary was defined as follows.
The boundary line between the two Republics shall commence at a point in
the Gulf of Mexico, three leagues from land, opposite to the middle of the
Southernmost inlet into Corpus Christi Bay; thence, through the middle of
said inlet, and through the middle of said bay, to the middle of the mouth
of the Rio Nueces; then up the middle of said river to the Southernmost
extremity of Yoke Lake, or Oagunda de las Yuntas, where the said river
leaves the said Lake after running through the same; thence by a line due
west to the middle of the Rio Puerco, and thence up the middle of said river
The Americas 239
to the parallel of latitude six geographical miles north of the Fort at the
Paso del Norte on the Rio Bravo; thence due west, along the said parallel
to the point where it intersects the western boundary of New Mexico; thence
northwardly along the said boundary, until it first intersects a branch of
the River Gila; (or if it should not intersect any branch of that river, then
to the point on said boundary nearest the first branch thereof, and from
that point in a direct line to such a branch) thence down the middle of said
branch and of the said River Gila, until it empties into the Rio Colorado,
and down or up the middle of the Colorado, as the case may require, to the
thirty-third parallel of latitude; and thence due west along the said parallel,
into the Pacific Ocean. And it is hereby agreed and stipulated, that the ter-
ritory comprehended between the Rio Bravo and the above defined Bound-
ary, from its commencement in the Gulf of Mexico up to the point where it
crosses the Rio Bravo, shall for ever remain a neutral ground between the
two Republics, and shall not be settled upon by the citizens of either; no
person shall be allowed hereafter to settle or establish himself within the
said territory for any purpose or under any pretext whatever; and all con-
traventions of this prohibition may be treated by the Government of either
Republic in the way prescribed by its laws respecting persons establishing
themselves in defiance of its authority, within its own proper and exclusive
territory. (Miller, 1937, 288)
The form of this description implies that the neutral strip lay within Mexico,
although the sense of the description is that it would be the responsibility of
both governments to restrict settlement there. It is not clear why, in order to give
a land connection between Lower California and Sonora, the boundary had to
be drawn along the parallel 33º north. This line would deny the United States
access to San Diego. The parallel six geographical miles north of the fort at Paso
del Norte on the Rio Bravo was coincident with the southern boundary of New
Mexico on Disturnell’s map, and this avoided any dispute about the position of
the provincial boundary.
There was no chance of the United States accepting this boundary because it
would compromise Texan sovereignty and deprive America of San Diego. The
commissioner was recalled before the resumption of hostilities that forced Mexico
to sue for peace (Miller, 1937, 289-93). The American commissioner continued
to make history by ignoring his recall and remaining in Mexico to negotiate a
treaty, although by then he lacked any authority!
Before examining further negotiations it is useful to note that before the war
recommenced, the process of boundary evolution had been normal. Both states
had proposed lines that would allocate more territory than they expected to gain.
The boundaries proposed revealed that the geographical knowledge of both sides
was imperfect. Mexico’s proposal for a neutral zone was the transparent device
of a weaker state trying to limit the territorial concessions that the stronger state
was exacting.
240 Chapter 8
The southern and western limits of New Mexico, mentioned in this Arti-
cle, are those laid down in the Map entitled ‘Map of the United Mexican
States, as organized and defined by various acts of the Congress of the said
Republic, and constructed according to the best authorities, revised edition.
Published at New York in 1847 by J. Disturnell’; of which Map a Copy is
added to this treaty, bearing the signatures and seals of the Undersigned
Plenipotentiaries. And, in order to preclude all difficulty in tracing upon the
ground the limit shall consist of a straight line, drawn from the middle of
the Rio Gila, where it unites with the Colorado, to a point on the coast of
the Pacific Ocean, distant one marine league due south of the southernmost
point of San Diego, according to the plan of said port, made in the year
1782 by Don Juan Pantoja, second sailing master of the Spanish fleet, and
published at Madrid in the year 1802, in the Atlas of the voyage of the
schooners Sutil and Mexicana. (Miller, 1937, 213-5)
There are two points to notice. First the boundary description was similar to
that originally proposed by America and it continued to reflect the generalized
topographical knowledge available about the area in question. For example,
apparently it was still unknown whether the River Gila or any of its tributaries
originated in New Mexico. It is curious that in the Gulf of Mexico the bound-
ary starts ‘three leagues from land’ and on the Pacific coast there is reference to
‘one marine league’. Before Texas joined the Union, in 1845, it had proclaimed
a maritime zone three leagues wide (Shalowitz, 1962, 132). Since these were
maritime measurements this claim was equal to 9 nautical miles. One nautical
mile is equivalent to one minute of latitude or 1,852 metres. The measurement
south of San Diego was a marine league of 3 nautical miles, but it was measured
over land. Second the definition hoped to avoid the two main points of contro-
versy by specifying the maps that were authorities for fixing the southern and
western boundaries of New Mexico and the terminal point on the Pacific Ocean.
Subsequent events showed that controversy was not avoided.
In 1849 a commission tried to determine the Pacific coast terminus of the
boundary. It found little correspondence between Pantoja’s plan of San Diego
in 1782 and the landscape 67 years later. One point near the port appeared to
coincide with the present coast. Accordingly, they measured the distance on the
map between this point and the southernmost point of the port in 1782. This
distance was then laid off on the ground and the marine league was measured
southwards from it.
In 1852 the United States Government made financial provision for the joint
commission appointed to demarcate the southern and western boundary of New
Mexico. The United States paid $15 millions for the territory gained. Then in
December 1853 a further $10 millions in gold was paid for the Gadsen Purchase.
It was named after the United States Commissioner James Gadsen. A series of
straight lines replaced the Gila River and delivered the catchment of that river
242 Chapter 8
to the United States. The new area was about 24,000 square miles in extent. The
southern boundary departed from the Rio Grande at latitude 31º 47’ North, that
was followed westwards for 100 miles before turning due south to parallel 31º
20’ north. This parallel was followed to longitude 111º West. The boundary was
continued by a straight line to the Colorado River 220 English miles, below its
confluence with the River Gila. The boundary was completed along the median
line of the Colorado River north to the straight line agreed in 1848 (Malloy,
1910, 1122). The new boundary was demarcated in 1855. However, as new set-
tlers moved into the area it became evident that the line was not satisfactorily
demarcated and in the period 1891-1898 a joint-commission discovered the line
and it was properly marked (Douglas, 1930, 39). Rebert (2000) has provided
an excellent account of the mapping of the boundary between Mexico and the
United States in the period 1849-1857.
The western boundary between Canada and the United States was based on the
Anglo-Russian boundary of 1825. Russia occupied and claimed Alaska and the
Aleutian Islands in 1803. In 1818 Britain and the United States had settled their
boundary from the Lake of the Woods to the Rocky Mountains. It was agreed
that the coastal area west of the Rocky Mountains would remain open to both
parties for a decade (Douglas, 1930, 13). The British authorities were anxious
to ensure that the territory of Canada had an outlet to the Pacific Ocean and that
this corridor should be as wide as possible. The Russian government sought a
boundary at least as far south as parallel 55º North. This parallel marked the
southern limit of the area within which the Russian-American Company had been
granted exclusive trading rights by the Russian authorities. The most southerly
Russian settlement was Sitka on Chichagof Island.
In 1823 Sir Charles Bagot, the British Ambassador in Moscow was instructed
to open negotiations with the Russian government to fix a boundary that would
provide a satisfactory outlet on the coast for British Canada. Sir Charles opened
the bidding at Cross Sound in latitude 58º North, then the Lynn Canal to its
northern extremity and longitude 135º West to the Arctic coast. Over the next
two years Russia argued for a more southerly line of latitude and Britain proposed
a more westerly longitude. In 1824 Russia and the United States agreed that
Russia would not claim territory south of 54º 49’ North and the United States
would not claim territory north of that line. This encouraged Britain to reach an
agreement with Russia in 1825.
III. Commencing from the southernmost point of the Island called Prince
of Wales Island, which point lies in the parallel 54 degrees 40 minutes
north latitude, and between the 131st and 133d degree of west longitude
(meridian of Greenwich) the said line shall ascend to the north along the
channel called Portland Channel, as far as the point on the continent where
it strikes the 56th degree of north latitude; from this last mentioned point,
the line of demarcation shall follow the summit of the mountains situated
parallel to the coast as far as the point of intersection of the 141st degree
The Americas 243
of west longitude (of the same meridian); and finally, from the said point of
intersection, the said meridian line of the 141st degree, in its prolongation
as far as the Frozen Ocean.
IV. That with reference to the line of demarcation laid down in the preced-
ing article, it is understood
1st. That the Island called Prince of Wales Island shall belong wholly to
Russia.
2nd. That whenever the summit of the mountains which extend in a direction
parallel to the coast from the 56th degree of north latitude to the point of
intersection of the 141st degree of west longitude shall prove to be at the
distance of more than ten marine leagues from the ocean, the limit between
the British possessions and the line of the coast which is to belong to Rus-
sia as above mentioned shall be formed by a line parallel to the winding
of the coast, and which shall never exceed ten marine leagues therefrom.
(Douglas, 1930, 42)
The charts that British and Russian negotiators used in delimiting this boundary
were prepared and published by Captain George Vancouver after his voyages
from 1792-4 (Vancouver, 1798). The relevant detailed charts had scales vary-
ing from 1:888,000 to 1:755,000 and the continental coast was mapped at 1:1.5
millions. The locations of major islands and some minor islands were plotted
with considerable accuracy. Davidson (1903, 55) notes that Tebenkof based his
1848 atlas on Vancouver’s charts and that American government vessels were
still using them to navigate through the Alexander Archipelago in 1867.
The charts show relief in two ways. On the islands hachures are used to portray
escarpments and low, isolated ranges. On the continent relief is shown by shaded
drawings of hills and mountains. On these charts the landward limit is marked
by a continuous range of mountains, about 9 km wide. This continuous range is
the most inaccurate feature on these charts. Along the 1200 km of this limiting
range only two peaks are named. They are Mounts Saint Elias and Fairweather.
Their respective heights were later determined to be 5490 metres and 4666 metres.
There can be no doubt that Davidson (1903, 54-5) was correct in asserting that
Vancouver sought to convey the impression that behind the continental coast
there was a range of mountains. Twenty-five years later the British and Russian
negotiators were treating the charts as an accurate representation of the landscape
and decided that the linear range was an excellent site for a boundary.
The boundary was not demarcated, but its delimitation allowed Britain to con-
centrate on completing its boundary, formed by parallel 49º North, between the
Rocky Mountains and Juan de Fuca Strait. That segment was settled in 1846.
By a Convention dated 30 March 1867 the United States purchased Russia’s
Pacific territories, consisting of Alaska and the Aleutian Islands, for $US 7.2
millions in gold. The United States also inherited the Anglo-Russian boundary
that was described in Articles III and IV of the 1825 treaty. It must have been
244 Chapter 8
evident to the American and British officials that this definition contained some
uncertainties and difficulties. First there was not a single line of mountain sum-
mits from the head of the Portland Channel to meridian 141º West. Second the
word ‘parallel’ has a precise meaning that cannot be achieved unless the coast
and the mountain range are either straight lines or arcs of circles drawn with
different radii from a common centre. Douglas (1930, 43) notes that for many
years both sides considered that the boundary should be located ten marine
leagues from the mainland coast. Ten marine leagues equals 30 nm or 55.5
km. The advance of settlers along the Stikine River valley made it necessary
for a Canadian official, called Hunter to interpret the location of the boundary
according to the Convention (Paullin, 1932, 70). He fixed the boundary about
40 km above its mouth and it was accepted by both Governments. Surveys in
the period 1892-96 were designed to provide information that would allow the
boundary to be delimited. This matter became urgent in 1896 when gold was
discovered in the Yukon valley in northwest Canada. A gold-rush started and
lasted for seven years. Prospectors sought entry to the diggings along the Lynn
Canal on the ground that Canadian authorities claimed the northern coast of that
extended Inlet, but this was refused by the United States. Eventually provisional
lines were agreed that allowed access via the Chilkoot and White Passes and the
Dyea and Skagway trails (Nicholson, 1954, 40).
Further negotiations in 1898 involved proposals by Britain that the boundary
should follow the mountains closest to the coast, ‘. . . crossing all the inlets of
the sea up to Mount St. Elias’ (Douglas, 1930, 43). This claim was not accepted
by the American commissioners, who also declined the invitation to arbitration.
When the boundary commission was ended the only temporarily defined points
on the boundary were ‘. . . the summits of Chilkoot and White Passes and a point
on the Chilkat River above Pyramid Harbor . . .’ (Douglas, 1930, 43).
During January 1903 the two governments agreed that they would each appoint
three impartial jurists of repute to the Alaskan Boundary Tribunal in the hope
that they could solve the disagreement. It might have been expected that the
American and British jurists would vote en bloc for any final arrangement. The
British delegation consisted of Baron Alverstone, Lord Chief Justice of England
and two Canadian lawyers. As it turned out Baron Alverstone sided with the three
American jurists and the United States secured most of its aims in the delimita-
tion that was announced on 20 October 1903 (Douglas, 1930, 43-4).
It [the boundary] commences at Cape Muzon. Thence it crosses in a straight
line to the mouth of Portland Channel, this entrance being west of Wales
Island, and passes up the Channel to the north of Wales and Pearse Islands
to the 56th parallel of latitude. Then the line runs from one mountain summit
to another, passing above the heads of all fiords. At the head of Lynn Canal
it traverses White and Chilkoot Passes. Then by a tortuous southwesterly
course it reaches Mount Fairweather and thence follows the highest moun-
tains around Yakutat Bay to Mount Elias. (Douglas, 1930, 44)
The Americas 245
Part of the boundary was not defined precisely but the governments exchanged
notes that enabled mutually satisfactory arrangements to be made. The boundary
section, of meridian 141º West, measured 645 miles and was defined by inter-
visible monuments by 1913. The survey of the coastal boundary, that measured
892.7 miles, was completed by 1914 (Douglas, 1930, 44).
There has been one major disagreement regarding the straight line that con-
nects Cape Muzon to the entrance of the Portland Channel. Canadian authorities
regard this straight line as being a maritime boundary. This interpretation would
mean that the United States possesses no territorial sea southeast of Cape Muzon.
The tribunal’s decision is silent on any maritime questions and there are several
international agreements where islands have been divided between two or more
countries without the lines being considered maritime boundaries.
Arbitration
The southern section of the boundary had been defined in 1881 with the
assistance of the United States Ministers in Buenos Aries and Santiago, both
called Osborn! Unfortunately the definition included an ambiguous phrase that
created serious problems for nearly a century. The phrase specified that the
frontier should ‘. . . run by the most lofty peaks of said chains which divide the
waters and to pass between the slopes which incline to one side and the other . . . ’
(Ireland, 1938, 23). Unfortunately long ago the eastward flowing rivers had cut
through the most elevated crests, so that their sources lay west of the highest
peaks. In 1896 Queen Victoria agreed to appoint a committee of British experts to
solve this difficulty. She died before the committee reported, and the award was
made by King Edward VII in May 1902. Britain was also requested to appoint
a demarcation commission that would assist officials of Argentina and Chile to
demarcate the boundary, and this was done in 1903.
Unfortunately three further disputes over alleged discrepancies in the demar-
cation have required arbitration since 1966. Two cases involved an area around
Palena and in the vicinity of Lake San Martin. The third case involved the
interpretation of Article 3 of the 1881 treaty. It specified that ‘all the islands to
the south of Beagle Channel to Cape Horn and those west of Tierra del Fuego to
belong to Chile’ (Ireland, 1938, 23). Argentina challenged the interpretation of
this clause that delivered all the islands in Beagle Channel to Chile. The Court
found in Chile’s favour in 1977 and on 25 January 1978 Argentina claimed
the judgement was a nullity and refused to accept it (de Aréchago, 1993, 720).
The dispute was resolved by the involvement of Pope John Paul II. The islands
awarded to Chile remained with that country. At the same time a maritime
boundary was settled between the two countries. This boundary discounted
Chile’s claims from the islands that would otherwise have placed Argentina at
a marked disadvantage.
In 1865 President Lopez of Paraguay, unwisely as it turned out, declared war
on Argentina. Argentina was assisted in this conflict by Brazil and Uruguay. The
war and the life of President Lopez ended in 1870. At the beginning of the war
the three allies had signed a secret treaty that would have confined Paraguay to
the area between the Paraguay and Parana Rivers. Argentina would have been the
chief beneficiary of additional territory. When the terms of this agreement were
published a number of South American states raised objections. Although Brazil
had been a signatory to the agreement it now supported Paraguay in its boundary
negotiations with Argentina. Fortunately Argentina decided that it would not assert
the rights of a conqueror and impose new boundaries (Ireland, 1938, 30-3).
An agreement between Argentina and Paraguay in 1876 identified the territory
in dispute. It was bounded by the Rio Verde to the north and the Rio Pilcomayo
to the south. The agreement further provided that if this issue could not be settled
within a year the question would be referred to the President of the United
States of America for arbitration. In the absence of agreement the matter was
referred to President Grant in January 1877. This invitation was accepted by his
The Americas 249
successor President Hayes. Each side submitted their cases in March 1878 and
the judgement was handed down eight months later. No reasons were given for
the decision that delivered all the disputed territory to Paraguay. In May 1879
Paraguay changed the name of Villa Occidental to Villa Hayes.
The Dutch and French governments agreed in 1836 that the River Maroni would
form their boundary between Surinam and Cayenne. Eventually, it was realised
that 175 km from its mouth two rivers flowed into the Maroni. The headwaters
of the Awa [Lawa] River were located to the southeast and the headwaters of
the River Tapanahoni were located in the southwest. The two governments were
unable to agree which of the tributaries should be regarded as a continuation of
the Awa River and discussions ended in 1876. Twelve years later in November
1888 the Dutch and French administrations invited Czar Alexander III to adjudi-
cate which river should mark the boundary beyond the Maroni River. The Czar
objected that he was being restricted to choosing one of two rivers when some
intermediate line might be more appropriate (Ireland, 1938, 243). After receiving
pleadings from both sides, the Czar announced, in May 1891, that the boundary
should follow the Awa River.
Further exploration presented the same problem when it was discovered that
the Awa was formed by the junction of the Itani [Litani] River and the Marouini
River 150 km above the junction of the Awa and Maroni Rivers. This time the
governments did not resort to adjudication. The more westerly Itani River was
selected, perhaps because it was almost perfectly aligned with the course of
the Awa. It was also agreed that the Dutch authorities would recognise French
properties, selected to the west of the Awa before 1888, for 40 years (Ireland,
1938, 244).
The Netherlands and Britain sought footholds on the northeast coast of South
America from 1667. At first the Dutch were successful, but British privateers
captured the Dutch provinces of Essequibo, Demerara and Berbice in April and
May 1796. In 1802, they were returned to the Netherlands by the Treaty of
Amiens (Ireland, 1938, 234). However, the three provinces were regained by
Britain eighteen months later and remained a British colony. The territory became
independent in 1966. Attempts by Britain and Venezuela to determine a com-
mon boundary were unavailing but both sides outlined their claims. For Britain
the minimum claim was the Schomburgk Line surveyed by the British botanist
in 1840-44 and advanced in 1886 (Rivière, 2007). Venezuela had no doubt that
the Essequibo River was the correct boundary between the two territories. In
the 1890s Venezuela sought the assistance of Brazil, Peru, the United States
and the Pope. Only the United States made a serious effort to assist Venezuela.
President Cleveland instructed Secretary of State Olney to issue a strong note to
Lord Salisbury, the British Prime Minister. Roberts (1999) provides an interesting
analysis of the following exchanges.
250 Chapter 8
important European maritime power for the Monroe Doctrine as it was then
interpreted and gave it an international standing it had never previously
enjoyed. After winning this point, the United States officials and most of the
public appeared to lose interest in the outcome of the fundamental territorial
question. (Ireland, 1938, 237)
In 1897 the two sides reached agreement on the formation of a five-member
tribunal to determine the boundary. It consisted of two British judges and two
American judges. Venezuela appointed the Chief Justice of the United States and
the United States Supreme Court nominated one of its members. The four judges
were then required to select a fifth member and they chose Professor Frederic de
Martens, a Russian Privy councillor (Ireland, 1938, 238). Both sides completed
their claims, counter-claims and arguments by January 1899. The Tribunal then
met in Paris from 15 June to 27 September 1899 and listened to oral arguments
from seven counsel (Ireland, 1938, 239).
The Tribunal’s decision was delivered on 3 October 1899. The new bound-
ary gave neither side all that had been claimed. Much of the line lay close to
the Schomburgk Line of 1886, but Venezuela secured important areas on the
southeast shore at the mouth of the Orinoco River. The Commission defined
the boundary to the western limit of Dutch Guiana in the mistaken belief that
Venezuelan territory ran with British territory to that point. When the Anglo-
Brazilian boundary was adjudicated in 1904 the western terminus was fixed at
Mount Yakontipu. When Brazil and Venezuela settled their eastern boundary
terminus at Mount Roraima in December 1905, Britain realised it had a much
longer boundary with Brazil than had been realised. Brazil and Britain finalised
their boundary in 1926.
It appears that Venezuela still hopes that the boundary with Guyana near the
coast might be moved eastwards. On 10 July 1968 Venezuela proclaimed a straight
baseline measuring 183 km that closes the mouths of the Orinoco River. The
western terminus is located on the Venezuelan coast, while the eastern terminus
is located 44 km east of the terminus of the land boundary at Punta Playa (The
Geographer, 1970). This line appears to close the Waini River. There is no article
in the Law of the Sea Convention that permits a country to anchor a straight
baseline on the territory of a neighbouring country without permission.
French attempts to settle the colony of French Guiana from 1635, eventually
failed in 1809, when the French Governor surrendered to a Portuguese force
(Ireland, 1938, 146). In 1817 the colony was returned to France in accordance
with treaties concluded at the end of the Napoleonic wars. The boundary with
Portuguese territory was considered to be the River Oyapock. A disagreement
arose between France and Brazil, that became independent in 1822. France
started to claim that the correct boundary coincided with the River Araguari.
that flows into the Atlantic Ocean between Cabo Norte and the Mouths of the
Amazon River.
252 Chapter 8
In 1836 France made a very ambitious claim to a boundary that followed the
River Araguari, until its course turned northwards, then proceeding westwards to
the Rio Branca, that was followed northwards to the watershed between the rivers
flowing north to the Caribbean Sea and southwards to the Amazon. This claim
would have interposed French territory between Brazil and Dutch and British
Guiana. In 1841 both countries agreed that they would avoid making new settle-
ments in the area between the Araguay and Oyapock Rivers. Negotiations then
drifted without success only enlivened when the residents of Cacouene proclaimed
an independent Republic that lasted one year, In April 1897 both countries invited
the Government of Switzerland to adjudicate the boundary between them. This
invitation was accepted in September 1898 and after both sides had delivered
initial claims and replies the judgement was given on 1 December 1900. The
Brazilian line was accepted. The boundary followed the River Oyapock to its
principal source and then coincided with the watershed between rivers flowing
north and south along the Tumuc Humuc Mountains. The total area between the
two claimed boundaries was 77,500 sq. km and Brazil secured 75,000 sq. km.
(Ireland, 1938, 151).
After a series of diplomatic skirmishes Britain and Brazil began to make deter-
mined efforts to agree on the southern boundary of British Guiana in 1891.
British offers in 1891 and 1897 would each have given Brazil more territory
than it finally secured. In 1901 both parties approached King Emmanuel III of
Italy to act as arbiter. The area, within which the boundary was to be delimited,
was bounded on the south by the Cotinga and Tacutu Rivers and on the north
by the watershed, between rivers flowing north to the Caribbean Sea and south
to the Amazon. Remarkably Brazil excluded from consideration 44,250 sq. km
north of the watershed. Apparently Brazil only wished to claim what was claimed
originally by Portugal (Ireland, 1938, 155). Each side presented their case, their
counter-case and their arguments between February 1903 and February 1904.
The decision, made in June 1904, declared that the boundary would follow the
watershed from Mount Yakontipu to the source of the Ireng River, that river to
its junction with the Tacutu River and that river to its source. From this point
the boundary continued along the watershed. The award made a clear statement
about the acquisition of sovereignty.
. . . to acquire sovereignty in new regions occupation was indispensable; there
must be effective, uninterrupted and permanent possession in the name of a
state, a simple affirmation of rights or a manifest intention to render occu-
pation effective not being sufficient; Portugal first and Brazil subsequently
had not taken effective possession of all the territory in dispute: the award
of October 3, 1899, in the arbitration between Great Britain and Venezuela
was ineffective against Brazil; the British rights, in succession to those of
Holland, were based on the exercise of rights of jurisdiction of the Dutch
West India Company, with sovereign powers from the Dutch Government,
The Americas 253
and like acts of authority and jurisdiction were afterwards continued in the
name of Great Britain, resulting in the gradual development and acquisition
of sovereignty over a part of the territory in dispute; historical and legal
claims did not fix with precision the limit of the zone during the next three
years over which the right of sovereignty of either party was established,
and the contested territory should be divided in accordance with the lines
traced by nature. (Ireland, 1938, 156)
While useful, the statement finally begs the question of choosing between dif-
ferent lines of nature.
Guatemala, part of the Viceroyalty of New Spain, proclaimed its independence
in 1821. Honduras followed this lead in 1838. Unsuccessful attempts were made
to delimit their common boundary in 1845, 1895 and 1914. In July 1930 both
countries agreed to invite a tribunal, presided over by Charles Hughes, Chief
Justice of the United States (The Geographer, 1976, 1).
The tribunal began with the Spanish boundary as it existed in 1821, and then
adjusted it in the light of major interests that might have been established, by
either party, across the 1821 line. Because the available maps were unsatisfac-
tory in some respects, the tribunal delayed consideration of the boundary until
an aerial survey was made.
With respect to the line of uti possidetis, the Tribunal decided that: (1) the
claim of Honduras to the territories adjacent to the Caribbean coast between
the Rio Motagua and British Honduras [Belize] was not sustained; (2) east
of the Rio Motagua along the Caribbean coast, available information did not
afford a sufficient foundation for the assignment of the Omoa area and the
contiguous Cuyamel area to either Guatemala or Honduras; (3) available
data did not afford an adequate basis for the assignment of the territory
from the Rio Motagua eastward to the Cordillera del Merendon to either
of the two states; and (4) in the southern part of the disputed territory,
administrative control was not continuous. (The Geographer, 1976, 2)
The award was delivered on 23 January 1933 and, in the next three years, the
boundary was delimited by 1,028 pillars.
In 1858 Costa Rica and Nicaragua agreed on a boundary stretching 320 km
between the Pacific Ocean and the Caribbean Sea. Changes to the line, including
the possible, substitution of the Rio Colorado for the lower reaches of the Rio
San Juan, were raised but no agreement was reached. In 1888 President Cleveland
of the United States was invited to adjudicate the points of disagreement. The
President confirmed the propriety of the 1858 agreement. The two neighbours
decided in 1896 to demarcate the boundary. On the recommendation of the
American authorities the survey teams of Costa Rica and Nicaragua received
the guidance of Edward Alexander to assist with the interpretation of uncertain
phrases. For example he calculated the western terminus of the boundary in
254 Chapter 8
the centre of Bahia de Salinas at 11° 03’ 47” North and 85° 43’52” West (The
Geographer, 1976a, 3).
Costa Rica and Panama finally settled their boundary, measuring 375 km,
between the Pacific Ocean and the Caribbean Sea in 1941. Attempts to delimit
a boundary continued from independence to 1896 without success, and it was
decided to turn to arbitration. President Loubet of France was invited to resolve
the problems. Unfortunately his award on 11 September 1900 satisfied only
Panama (The Geographer, 1976b, 1). This situation was reversed in 1914. The two
countries secured the services of the Chief Justice of the United States Supreme
Court in 1910. His award on 12 September 1914 satisfied only Costa Rica.
In May 1941, without foreign assistance the two countries delimited the line.
A mixed commission was appointed to demarcate the boundary. It was assisted
by a Chilean surveyor, nominated by the President of Chile.
The collapse of Gran Colombia in 1830 required Colombia and Venezuela to
define their boundary (Figure 8.5). The constitutions of both countries declared
their territories should be those bounded by the previous sub-divisions of Gran
Colombia. In short it was a declaration in favour of uti possidetis juris. A treaty
dealing with friendship, commerce, navigation and boundaries was signed in
December 1833, and it contained the proviso, insisted on by Venezuela, that
the boundary should be a line of convenience rather than strict entitlements
(Ireland, 1938, 206). The boundary was described in Article 27, and contained
both precise and general terms such as ‘. . . thence down by its waters [Rio Oro]
to the confluence of the Catatumbo, thence by the foothills of the mountains . . .’
(Ireland, 1938, 208).
Venezuela declined to ratify this treaty. Instead, in July 1842, both sides agreed
settle their boundary within four years. In April 1844 significant agreement
was reached on the boundary from the coast to the confluence of the Meta and
Orinoco Rivers. However, it proved too difficult the draw the boundary in the
upper Orinoco River and the Rio Negro. In 1845 Colombia suggested recourse
to arbitration but Venezuela refused this invitation and subsequent invitations in
1872 and 1875. At last, on 14 September 1881 the two foreign ministers signed
a treaty to arrange arbitration.
. . . for the purpose of reaching a true legal territorial delimitation, such as
existed by the decrees of the former common sovereign, and claimed by both
parties for a long period without their having come to an agreement as to
their respective rights or the uti possidetis juris of 1810, providing that:
Article 1, the parties submitted to the judgement and decision of the
government of the King of Spain as arbiter and legal judge the points of
difference on such boundary question, to obtain a final unappealable deci-
sion according to which all the territory which belonged to the jurisdiction
of the ancient captaincy-general of Caracas by royal acts of the former
sovereign till 1810 should remain jurisdictional territory of Venezuela and
The Americas 255
all which by similar acts to the same date belonged to the jurisdiction of
the viceroyalty of Santa Fe should remain territory of Colombia;
Article 2, both parties shall request the King of Spain to accept the office,
and eight months thereafter they should present cases with their claims and
supporting documents. (Ireland, 1938, 209)
Venezuela’s policy of favouring a convenient boundary had been abandoned,
apparently because that would ‘. . . involve the alienation of territory that was
prohibited by the federal constitution’ (Ireland, 1938, 209). On 19 November
1883 King Alfonso XII appointed a tribunal of five to report on the matter. The
death of the King before the tribunal’s examination was completed was finessed
by the appointment of the Queen-Regent to act on behalf of the infant King
Alfonso XIII. The award was published on 16 March 1891. The preamble noted
the imprecision and uncertainties of the royal decrees that defined the Viceroyal-
ties, and the award then proceeded to divide the boundary into six sections. The
sixth section consisted of two parts. After seven years of manoeuvring the two
countries appointed two commissions to examine the boundary sections that had
been grouped into ‘two great portions’. The first consisted of Sections 1-4 and the
second consisted of Sections 5, 6-1 and 6-2. The commissioners were appointed
in 1899 and they started work in 1900. Alas! Their work ended in 1901 because
of internal disagreements amongst the commissioners, a civil war in Colombia
and discord within Venezuela (Ireland, 1938, 215). Relations between the two
countries deteriorated and by 1914 it was clear no lasting progress had been
made. Ireland summed up the situation as follows.
The negotiations upon this boundary thus lasted for fifty years before even
arbitration could be agreed upon. It was ten years more, before the arbiter
decided the case, and the disappointed party [presumably Venezuela] post-
poned execution of the award for another twenty-five [twenty-four] years,
leaving the actual boundary on the ground nearly as far from settlement
after eighty-five years as it had been at the creation of the two Republics,
(Ireland, 1938, 215).
In November 1916 the disagreement between the two countries crystallised. Ven-
ezuela held the view that neither country could possess the territories awarded
in 1891 until the boundary had been demarcated. Conversely Colombia asserted
that it could take possession of territories granted in the award if the area was
bounded by nature or by a line determined by the commissioners in 1900. These
views were contained in a treaty signed in 1916.
In 1917 both sides agreed to invite the Swiss Federal Council to interpret
the content of the 1916 treaty and the award of 1891. A tribunal was appointed
and the cases and replies of both countries, together with additional information
requested by the tribunal, was submitted by 30 April 1921. The Swiss Federal
Council handed down the decision on 24 March 1922. It was ordered that the
256 Chapter 8
award of 1891 would be carried out in sequence. First both countries could
enter into possession of territories within natural boundaries awarded in 1891
and territories awarded by the commissioners in 1900. Second, unsettled matters
would be dealt with, by the Swiss delegates, before 31 December 1924. In fact
the adjudication was completed by 30 July 1924.
Four years later the two countries reached the following agreement.
. . . the line should be marked in a conspicuous and permanent manner, the
main channel of the Arauca and other frontier rivers should be determined
and maintained, and two sectional mixed commissions, each of one nominee
of each government, with the necessary assistants, should carry out the
technical work, with power to determine facts only and not to modify any
of the decided lines. (Ireland, 1938, 218)
This was a case where both sides tried hard to identify the Spanish boundaries
that separated the captaincies-General of Venezuela and Granada, later Colombia.
Their efforts were unsuccessful and eventually this boundary, like many others in
South and Central America, was determined with the assistance of arbitration.
There is one uncertainty about the coastal terminus of the boundary between
Colombia and Venezuela. Nweiheid (1993, 282) in an essay dealing with maritime
boundaries explains that the boundary terminates on a mound 100 metres short
of the low-water line. Therefore until the boundary is completed to the low-water
line it is impossible to define the origin of the maritime boundary. It is a fact that
Colombia and Venezuela have both been very successful in defining maritime
boundaries with neighbouring or opposite states, but they have not been able to
settle their common boundary.
from Las Juntas up the Tarija to the mouth of the Itau, and thence by the
waters of the Itau to the 22° parallel and by such parallel to the waters of
the Pilcomayo. (Ireland, 1938, 5)
The boundary was delimited on a map in 1899 but when demarcation commenced
in 1913 problems arose because the map and the landscape did not correspond
in important details. From Zapalegui there were two possible continuations of
the ridge. Argentine favoured the ridge called Azulejos and Pupascayo north of
the San Juan River, while Bolivia preferred the Esmoraca Ridge that lay to the
south of the river. The La Quiaca Ravine joined other ravines that led to the
Sococha River, that flows into the Yanapalpa River. Argentine favoured this
interpretation because it delivered the settlements of Sococha and Yanapalpa.
The Bolivian delegate argued for the boundary continuing from the junction of
the La Quiaca Ravine and the Toro-Ara Ravine to the La Raya Ravine and then
to the Yanapalpa River, that would put the two settlements in Bolivia.
Porongal Hill was interpreted by Argentina to be Mecoya Hill 40 km northwest
of Bermejo, while Bolivia argued that it was Negro Hill 50 km southwest of
Bermejo. The difficulty with the Bolivian position was that the boundary then
had to follow the Pescado River that flows into the Bermejo River south of the
town of Las Juntas de San Antonio. The boundary description clearly traced the
line southwards along the Bermejo River to the junction with the Rio Grande de
Tarija. On 9 July 1925 the two countries agreed on a reliable delimitation of the
boundary and apart from sections along rivers it was demarcated by pillars.
Most commonly when a boundary coincides with a river it is located in the
river. Often the talweg, that is the deepest continuous channel of the river, will
provide the site for the boundary. Rarely is the boundary located along one of
the banks of the river so that the river belongs entirely to one country. However
that arrangement occurs in three treaties in Central and South America.
Guatemala and Honduras declared their independence in 1838. Attempts to
define the boundary between them in 1845, 1895 and 1914 were unsuccessful.
On 16 July 1930 the disputed boundary was submitted to a Special Boundary
Tribunal whose chairman was the Chief Justice of the United States (Marchant,
1944, 317-19). The award was published on 23 January 1933 (The Geographer,
1976, 2). Under the terms of the Treaty of Arbitration the Tribunal was permitted
to modify a line based on the principle of uti possidetis juris. After an aerial survey
of the borderland had been made, the Tribunal defined the boundary between the
two states from their trijunction with El Salvador to the Gulf of Honduras.
This boundary is 265 km long but only the most easterly 41 km is relevant
to this analysis.
. . . northeasterly straight (11.2 miles) to a point at the center of the Cuyamel
Railroad bridge over the Santo Tomas river, northeasterly straight to the
southernmost point on the right bank of the Tinto river which flows out of
the Laguna Tinto, along the right bank, taken at mean high water mark, of
258 Chapter 8
the Tinto river downstream to its point of discharge into the Motagua river,
along the right bank, taken at mean high water mark, of the Motagua river
downstream to its mouth on the Gulf of Honduras. As thus described, the
boundary is established on the right banks of the Tinto and Motagua rivers
at mean high water mark, and in the event of changes in these streams in the
course of time, whether due to accretion, erosion or avulsion, the boundary
shall follow the mean high water mark upon the actual right bank of both
rivers. (The Geographer, 1976, 5)
This delimitation defines the boundary along the river-bank as the mean high
water mark.
This means that the mean high water must be determined by both countries
maintaining records at a set of stations along the boundary for an agreed period.
Where the river-bank that marks the boundary has a steep gradient, the country that
possesses the water will be favoured. When the river bank that marks the bound-
ary has a shallow gradient the country that owns the bank will be favoured.
This delimitation does not define the point where the mean high water mark of
the river terminates at the coast. Each country has a vested interest in identify-
ing the optimum intersection that will provide the origin of any future maritime
boundary. Presumably hydrographers for each country will identify this optimum
intersection, after considering the tidal range in the estuary and the configuration
of the low-water line adjoining the estuary. At the time of writing, these two
countries have not delimited their maritime boundary. However, facilitators have
been trying to arrange for a series of boundaries involving Belize, Guatemala and
Honduras. Strict lines of equidistance would constrict any maritime claims made
by Guatemala and it is hoped that Guatemala’s access to the waters beyond its
territorial seas can be assured.
Costa Rica and Nicaragua defined their land boundary on 15 April 1858 (The
Geographer, 1976a). It traverses the Panama Isthmus between the Caribbean Sea
and the Pacific Ocean and consists of two parts. In the east the boundary extends
along the right bank of the San Juan River for about 133 km from Pillars 1-2.
The remaining 186 km consists of straight lines that terminate in the centre of
the Bahia de Salinas.
The dividing line of the two Republics starting from the North [Caribbean]
Sea, shall begin at the extremity of Castilla Point, at the mouth of the San
Juan River, in Nicaragua, and shall continue along the right bank of said
River to a point 3 English miles from El Castillo Viejo, measured from the
exterior fortifications; (Marchant, 1944, 298)
The definition establishes that the boundary terminus on Castilla Point is ‘. . . in
Nicaragua . . . ’. However, no guidance is given about the water level with which
this point is identified. Presumably both countries will argue in favour of the
water-level definition that in their judgement provides the most satisfactory origin
for the maritime boundary.
The Americas 259
Following this treaty, changes were proposed by both sides over the next thirty
years, but no agreement was reached. The United States President was asked to
arbitrate and he decided that the 1858 Treaty was appropriate. Eight years later
the two countries decided to demarcate the boundary and appointed Edward
Alexander to arbitrate on any disagreements. Alexander made five awards and
only the first concerns this analysis; it was given on 30 September 1897.
. . . the following initial line of the boundary: its direction shall be northeast
and southeast . . . across the sandbank from the Caribbean Sea to the waters
of Laguna Harbor Head. At its closest point it shall pass 300 feet from the
northwest side of a small cabin now existing in that neighbourhood. On
reaching the waters of Laguna Harbor Head, the line shall turn to the left,
that is toward the southeast and shall follow the shore around the harbor
until it reaches the River properly so-called [Rio San Juan] by the first
channel that it meets. The line shall continue as provided in the Treaty,
going up this channel and up the River properly so-called [Rio San Juan].
(Marchant, 1944, 289-9)
In the absence of any map that might have been attached to the award this
boundary description is difficult to follow. For example, it is uncertain whether
the ‘sandbank’ is the submarine bar or a bank standing above high-water, on
which Castilla Point is located. On its course across the sandbank to the waters
of Laguna Harbor Head the line passes ‘. . . 300 feet from the northwest side of a
small cabin now existing in the neighbourhood’. It is remarkably unhelpful to fix
a point on the boundary by reference to a feature that might not be permanent.
The reader is then told that ‘On reaching the waters of Laguna Harbour Head, the
line shall turn to the left, that is towards the southeast and shall follow the shore
around the harbor until it reaches the river properly so-called by the first channel
it meets’. Neither in this sentence, nor in the following sentence is the phrase
‘right bank’ used. The absence of any reference to Castillo Point by Alexander
leaves the reader with at least two choices. First, relying on the Arbitrator’s
description it could be argued that the mouth of the River is only reached after
passing from the sea through waters of Laguna Harbor Head. Second, relying
on the Treaty it could be argued that Castillo Point is the most seaward point
where the river meets the sea, and therefore the waters of Laguna Harbor Head
form part of the River’s mouth or estuary.
The British Pilot refers to ‘Harbour Head lagoon’ (Hydrographer of the Navy,
1970, 99). A lagoon is defined as follows.
Lagoon. An enclosed area of salt or brackish water separated from the OPEN
SEA by some more or less effective, but not complete obstacle such as a low
sand bank. (International Hydrographic Organization, 1990, 115)
This definition of the word Laguna in Alexander’s award favours the second
interpretation.
260 Chapter 8
The British Pilot notes that the superfluous water from Lago de Nicaragua is
discharged through the San Juan River. The maximum flow occurs in October
and the lowest flow occurs in May. A principal consequence of this varied dis-
charge is that the navigable channel changes frequently (Hydrographer of the
Navy, 1970, 98-99). The bar at the entrance has a least depth of 5 feet but the
sea breaks heavily on it. Only small craft with cargo for San Juan del Norte are
allowed to enter. Vessels with cargo for Rio San Juan and Lago de Nicaragua
proceed via the Rio Colorado. This is a distributary of the San Juan River that
reaches the sea about 12 nm south of the main mouth. At the time of writing no
maritime boundary has been drawn by Costa Rica and Honduras.
Suriname became independent in 1975 and it inherited the Anglo-Dutch bound-
ary that, in part, was defined as follows.
The western boundary [of Suriname] is formed by the low-water line on the
left bank of the Corantjin [Courantyne], from origin to mouth. The bound-
ary therefore runs from a point to be further determined on the southern
boundary to the origin of the Upper-Corantjin, next from this origin along
the low-water line on the left bank of the Upper-Corantjin and the Corantjin
up to the point where the river bank changes into the coastline . . . (Republic
of Suriname, 2005, 8)
In February 2004 Guyana instituted proceedings in an Arbitration Tribunal seeking
a determination of the maritime boundary between the two countries. It appears
the whole case turns on identifying the terminus on the land that will be the
origin of the maritime boundary. In this case there are twin technical difficulties
associated with determining the point where the river’s left bank becomes the
coastline. First the Corantjin enters the Caribbean Sea via an estuary. It could
be argued that the coastline will adjoin the sea rather than an estuary. Therefore
it would be necessary to determine the extent of the estuary. Second the river’s
bank and the adjoining coastline merge in a smooth arc with a radius of 45.4
km. There is no promontory that stands out as a possible origin for a maritime
boundary. The horizontal distance between the mean low-water spring tide and
the mean high-water spring tide varies from 0.6 km to 2.4 km.
In most cases, when boundaries coincided with rivers, the river was partitioned
between the two countries. However sometimes there were complications that
required particular solutions (Figure 8.6). The 1876 boundary between Argen-
tina and Paraguay coincided with the Pilcomayo River. The section of the river
between Punto Horqueta and the waterfalls of Salto Palmar was flat and subject
to frequent changes of course.
. . . the river was one of the geographical unknowns of South America. Its
headwaters in Tarija and Chuquisaca, and the neighbourhood of its mouth,
on the Paraguay, had been known for many years, but its interior course
across the flat marshy lands of the Chaco, spreading out into wide swamps
and dividing into inconstant channels, remained a puzzle. The scattered
The Americas 261
groups of Indians were fierce and intractable and combined with the waters
to make exploration difficult. Crevaux was treacherously killed by the
Tobas while making his way downstream in 1882; Ibaretta was defeated
in his attempts at exploration in 1898, as were numerous others after him.
(Ireland, 1938, 33)
Finally in 1945, after aerial photography became available, the boundary was
defined as following the course of the waters as they exist in the dry season.
Five specific points were identified along the 132 km section between Punto
Horqueta and Salto Palmar.
The Uruguay River that separates Argentina and Uruguay for 530 km has 128
islands in the northern 400 km. The river has a width of 1 km in the north and
widens to 9 km in the south of the chain of islands. The river was effectively
selected as the boundary between the two countries in 1828. In 1916 an attempt
to allocate the islands to each side failed, but an investigation of land titles in
1935 paved the way for a final decision. On 7 April 1961 the boundary was
delimited. The boundary follows the main navigable channel. However, in the
vicinity of five islands, belonging to Uruguay, near latitude 33° S, the boundary
separates into two parts. The islands are called Filomena Grande, Filomena Chica,
Palma Chica, Bassi and Tres Cruces. The line that passes east of the islands
divides the waters between Uruguay and Argentina. The line that passes west
of the islands separates the islands, so the islands between the two segments of
boundary belong to Uruguay. The two boundary segments rejoin and continue
to the Rio de la Plata.
CONCLUSION
This review suggests five conclusions. First the indigenous populations, that had
established themselves in north, central and south America, played no significant
role in fixing the international boundaries that divide the two continents today.
Second in North America, the early involvement of French and British colonists
and administrators and Russian hunters, was overtaken by divisions between
colonists seeking independence and British authorities maintaining control over
Canada. Third in Central and much of South America, Spanish authorities and
colonists overcame the indigenous population and the difficult terrain to establish
several large colonies. Fourth in the beginning and medium term, the Portuguese
seemed more successful than any of the other European invaders by spreading
widely throughout the Amazon Basin. Finally it is clear that throughout the
Americas the European colonists played the dominant role in fixing the interna-
tional boundaries. Britain played a major role in fixing the southern and west-
ern boundary of Canada. Britain, France and the Netherlands, on the northeast
shoulder of South America, played cameo roles in the Guianas.
262 Chapter 8
REFERENCES
The Middle East is the intercontinental region bounded by Europe, Asia and
Africa. With an area of 6,504,965 sq. km, the Middle East is equivalent to 75
per cent of the Australian continent. In 2000 the population of the fourteen states
was 227,603,000.
Dewdney (1987, 7-9) notes that there are two prime physiographic zones in
the Middle East. To the north there is a region of folded mountains and pla-
teaus; to the south there is an ancient crystalline block of Arabia. The border
between the two zones corresponds closely to the southern boundaries of Iran
and Turkey. The mountains were created, during the period of the Cretaceous
to the Pliocene, from the thick sediments that underlay the ancient Tethys Sea.
The adjoining continents of Eurasia formed the anvil against which the hammer
of the Afro-Arabian plate fashioned the folded mountains that were subsequently
lowered by erosion.
The Arabian Peninsula was tilted downwards towards the north and northeast
so that the highest elevations are found in Yemen. The northern part of this stable
block was overlain by considerable thicknesses of more recent sandstones and
limestones. Thick sediments have filled the depression that previously existed
along the line of the present Tigris and Euphrates Rivers and they have recre-
ated a flat surface with low gradients. In the vicinity of the Persian Gulf and its
structural continuation along the lowlands associated with the Tigris and Euphrates
Rivers, geological conditions favoured the formation of oil and gas fields. The
thick porous strata that had been significantly deformed and tilted created traps,
within which hydrocarbons could accumulate from the rich deposits of remains
of creatures that once lived in the warm Tethys Sea.
The physiographic regions coincide with political regions. The northern region
consists of only two states, Iran and Turkey, both with populations of about 66
millions, that represent about 57 per cent of the total population in the Middle
East. The main languages spoken in the northern region are Baluchi, Kurdish,
Persian and Turkish. Both these states, at different times in their history, had been
the seats of empires. Although both negotiated boundaries, at a disadvantage,
with the powerful British and Russian governments, they secured lines that have
proved permanent and well defended.
In the southern region Arabic is the principal language for eleven of the twelve
states. The size of the populations in these states varies from 579,000 in Qatar to
266 Chapter 9
20,181,000 in Saudi Arabia. Throughout the southern region Britain and France
played influential roles in boundary construction, after the Ottoman Empire’s
domination in Arabia was ended (Figure 9.1), during and after World War I.
Before and after World War I there had been various unsuccessful attempts to
consolidate the extensive territory of Saudi Arabia into a single state. That was
accomplished by Ibn Saud in 1932. Yemen became independent in 1918. The
only other independent state in the southern region at that time was Oman, that
had dislodged the Portuguese in 1650. France governed Lebanon and Syria while
Britain played a similar role in the other territories, within boundaries that the
colonial powers decided. All these entities became independent in the period
from 1932 to 1971.
Throughout the Middle East all states are disadvantaged when claims to mari-
time zones are considered. Although no state is landlocked, Jordan and Iraq only
have coastlines 15 km long. Jordan’s only access is to the narrow Gulf of Aqaba.
Iraq’s maritime claims are constricted, by its neighbours Kuwait and Iran, at the
head of the Persian Gulf. Although Turkey can secure a significant proportion of
the Black Sea it is restricted to negligible claims in the Mediterranean by Greek
islands, close to the Turkish coast, and the presence of Cyprus and Syria. Only
Oman and Yemen can claim exclusive economic zones 200 nautical miles wide.
Yemen is favoured by possessing sovereignty over Socotra Island and the adjoin-
ing small islands called Abd-al-Kuri, Jazirat Samha and Jazirat Darsa. Socotra
Island is located about 200 nautical miles from the coast of Yemen and has an
area of 3,500 sq. km. The Persian Gulf is the only maritime area adjoining the
Middle East that has yielded great wealth. For this reason it is understandable
that claims to islands in the Gulf are made and defended vigorously.
Proceeding from east to west the boundaries defining the northern limit of the
northern region were negotiated between Britain and Persia, Russia and Persia,
Russia and Turkey, and Turkey with Bulgaria and Greece.
In the 1860s Britain was searching for a western limit to its Indian Dominions
at the same time that the Shah of Persia’s army was moving eastwards to threaten
Kalat, a British protected state. Britain had laid a submarine cable through the
Persian Gulf to improve communications, and, as a precaution, had been given
permission by the Shah to construct a land-line westwards from Gwatar Bay to
the vicinity of Bandar Abbas. When General Goldsmid began to construct this
line he was hampered by minor chieftains of Tump, Mand and Boleda. These
minor political units, of fifteen, twelve and five villages respectively, had been
raiding into Persian territory, and reprisals by the Shah’s forces were impinging
on the Kalat border (Prescott, 1975, 212-3). When the Shah suggested that a
The Middle East 267
boundary should be drawn between his territory and the territory of the British
Empress, the offer was enthusiastically accepted. The Persian delegate sent by
the Shah was unhelpful, so Goldsmid retired to Gwatar Bay while Major Lovett
was sent on a rapid reconnaissance of the border. Based on this new information
and surveys performed by Goldsmid in 1861-4, the General produced a boundary,
measuring 453 km that the Shah accepted on 4 September 1871.
In 1872 Goldsmid sent Major St. John to verify Lovett’s survey. Only one
error was found, when the Askan River was discovered lying further west of the
position recorded by Lovett. St John noted that the terrain was very complicated
in this region.
Any exact delimitation of the frontier is here, however, impossible. The
country consists of innumerable parallel ranges of inconsiderable elevation,
divided by narrow torrent beds. In these occur, at long intervals, small clus-
ters of date palms, with less frequent patches of cultivation. (India, Foreign
Department, 168/1896. No. 4, 12)
British authorities declined to take advantage of the new information about the
location of the Askan River. The Shah claimed the area of Kuhak that Gold-
smid thought was either independent or part of Kalat. The authorities in London
decided that the advantage of possessing this small territory was outweighed
by the damage that might be caused to the excellent relations with the Persian
authorities.
Delimitation and demarcation of the boundary with Persia started in Febru-
ary 1895 and was completed in March 1896. Holdich arranged the surveying to
avoid the hot season that lasted about four months. At this time Holdich was the
most eminent British military surveyor. He worked mainly in Asia and South
America. In 1916 he published ‘Political frontiers and boundary making’, that was
regarded as a surveying bible for two decades. However, when he delimited and
partly demarcated the boundary north of Kuhak, he was human and he erred. The
progress of the survey can be followed by quotations from Holdich’s report.
Fortunately, the whole area under discussion had already been surveyed by
members of my party, and the published maps were sufficiently accurate to
enable me to decide what line would constitute a sound boundary consis-
tent with the Government’s instructions, before the actual commencement
of demarcation.
There is nothing to compare with a rugged immovable line of watershed
for boundary definition. Every nomadic robber in the frontier understands
it, and is perforce obliged to respect it as it is quite beyond his powers
of interference. It was these considerations which decided me to adopt, if
possible, a line of boundary from the Malik Siah Koh to the Mashkel date
groves which would be marked by such strong natural features as would
render artificial demarcation unnecessary.
268 Chapter 9
Holdich was in such a hurry, that he declined an early proposal by the Persian
Commissioner that would have given Britain another 1300 sq. km. This would
have included the glacis that led up to the ‘craggy watershed’. Colonel McMa-
hon, who later gave his name to the boundary drawn by Britain against China,
was asked to review Holdich’s boundary. He was very critical on the grounds
that an accurate survey had not been made, that Britain gained no compensation
for the three concessions to Persia in the southern part of the line, and that the
northern section of the boundary allowed Persian forces to overlook the main
road northwards in British India. Holdich’s view that the Persian commissioner
The Middle East 269
his boundary at haphazard, and Koh-i-malik Siah being about the only fixed
point that his survey officer had been able to get a shot at in this part of
Sistan, he simply drew an imaginary line on the map from Band-i-Seistan
to Koh-i-malik Siah, without ever having been able to examine the ground
across which that line was to run. (McMahon, 1906, 344)
Goldsmid’s boundary was not demarcated but serious difficulties between Persian
farmers and Afghan herdsmen were either avoided or sensibly settled. However,
in 1902 an exceptional drought caused serious competition for the available water.
Eventually the Afghans occupied a Persian settlement. Sykes claimed that this act
was designed to made British arbitration unavoidable (Sykes, 1940, volume 2,
209). If he was correct the ploy was successful. McMahon was sent with 1,500
British soldiers and surveyors in 1903 to mark Goldsmid’s boundary. He made
a careful survey and presented the Afghan and Persian commissioners with a
written text describing each of the 90 stone pillars and two maps. The first, at
a scale of 1:253,440, showed the entire line. The second, at a scale of 63,360,
showed the line in the vicinity of the Siskar and Rud-i-Pariun Rivers.
Attention then switched to the northern sector of the Persian-Afghan boundary.
The northernmost section of the boundary coincides with the Hari Rud [River].
This river rises west of Kabul and flows along an entrenched valley for 516 km
to an Afghan settlement called Eslam Qal’eh, where it turns north. About 100 km
north of Eslam Qal’eh the Hari Rud reaches the trijunction with Russia, where it
takes the name Tedzhen River and forms part of the boundary between Iran and
Russia. The territory through which the Hari Rud flows northwards was called
Hashtadan. In April 1885 the Persian Governor ordered the local population to
clear the khanats, so that the area could support farming communities. Khanats
are underground canals that convey water many kilometres without significant
loss by evaporation. Afghan forces stopped the work and drove the labourers
away from the rivers. Both sides called for British arbitration and General C.S.
MacLean was sent to find a solution to the difficulties in 1888.
Over the next two years he compiled a history of the area and reached two
important conclusions. First it was obvious that the Hashtadan had not been used
for several decades.
But from the appearance of the ruins and abandoned fields it is quite evident
that the valley has been deserted for generations. . . . In fact there is every
reason to believe that the place has been deserted for at least one hundred
years. . . . Upon the whole, looking to the nature of my present information,
it seems to me that neither Persians nor Afghans can produce proofs of
recent possession in support of the respective claims, neither having felt
inclined to stand the brunt of collisions, in such an exposed locality, with
the Turkomans. (India, Foreign Department, 1890, 6, 25, 27)
MacLean found village ruins, faint field patterns and remnants of the system of
khanats that proved the area had been inhabited and productive. However, it was
The Middle East 271
also recorded that around 1788 an epidemic throat disease had caused deaths and
outward migration. The desolation of the region was completed by Turkoman
freebooters. MacLean implied that there now existed an unoccupied frontier
between the Afghan and Persian outposts. The Persian posts were at Kariz and
Karat while the Afghan posts were Kohsan, Eslam Qal’eh and Ghurian.
MacLean’s second major conclusion was that the waters of the khanats were
not used only by Persian settlements. At least one major khanat provided water
to Eslam Qal’eh that belonged to Afghanistan. He agreed to consider evidence
provided by both sides before he proposed his award. The evidence was certainly
imaginative, but it did not point to a single conclusion. Information provided
included a Persian tombstone dated 1426, many title deeds from both sides,
testaments, and payments of compensation for robbery and damage. MacLean
had to steer a course that would lead to a compromise that gave something to
both sides.
The case seems eminently one for compromise, in which both parties abate
their pretensions in order to render an equitable settlement possible. Such
a settlement is possible only by a division of the water-supply available for
irrigation as indicated by the old karezes or khanats. It is on this principle
that the compromise indicted on the map of Hashtadan has been based.
By accepting the above compromise Persia will secure the greater portion of
the arable land of the valley including the actual land over which the dispute
arose, and on which the work was interrupted by the Afghans, as already
described. On the other hand the Afghans will secure what they profess to
desire, viz, a supply of water for the irrigation of the Kafir Kala [Eslam
Qal’eh] and all the grazing grounds lying on the southern end of the valley.
The Afghans will also have a considerable area of arable land at Chakar
Kala, as well as a large tract near the mouth of the Shorab Pass, which can
be irrigated from the large canal, and the cultivation of Tir Kisht can be
extended to a considerable extent. (India, Foreign Department, 1890, 27)
Both parties accepted the award, although the Shah required the name Hashtadan
to be printed west of the boundary line on MacLean’s map! MacLean supervised
the location of 39 masonry pillars up to 3 metres in height. The Hari Rud defined
the boundary north of Pillar 1 but MacLean did not specify any particular line
within the river.
The gap of 400 km between MacLean’s most southerly Point 39 and McMa-
hon’s most northerly Point 90 on Koh-i-malik Siah was closed in 1935. In 1928
Afghanistan and Persia had agreed to appoint a joint-commission that would
meet to resolve territorial disputes in the borderland. It was not a perfectly sat-
isfactory arrangement and Persia suggested that a Turkish arbitrator should be
appointed to delimit a definite boundary. Turkey enjoyed good relations with
both countries and General Altai was appointed to fix the boundary. His team
worked from October 1934 to May 1935. Altai had an easier task that the three
British surveyors as part of his report recorded.
272 Chapter 9
Outside of the three farms located at Yezdan, this vast area does not contain
any inhabited place or cultivated area. There exists, however, several wells
and sources of fish and salt water. (The Geographer, 1961, 8)
The selected boundary followed a fairly direct course between the two British
termini. However, there was a small Persian salient at Yezdan and adjustments
were made to allow nomads access to traditional winter pastures and quarries that
provided mill-stones. Altai completed two unfinished matters along the MacLean
boundary. First he placed five pillars to join MacLean’s Pillar 1 northwards to the
bank of the Hari Rud. Second he defined the boundary, within the north-south
sector of the Hari Rud from the trijunction with Russia to the point, identified
by Altai, where the river’s course turned sharply southeastwards then east. Altai
selected the talweg as the Afghan-Persian boundary.
This boundary was delimited and demarcated by four foreign soldiers. Apart
from the difficulties that the Persians created for Goldsmid, the line appears to
have been accepted, with satisfaction, by Afghanistan and Iran.
Iran’s northern boundary with Russia did not involve any foreign interventions
or arbitrations. To the west of the Caspian Sea Russia and Iran share a boundary
that measures 700 km. East of the Caspian Sea the boundary extends for 990
km to the tri-junction with Afghanistan. Both west and east of the Caspian Sea
the two countries were separated by frontiers and their conversion to boundar-
ies occurred over 80 years after 1813. The frontiers west of the Caspian were
narrow while those to the east were broad. The region between the Black and
Caspian Seas was complex in terms of its geology, morphology, local climate and
agricultural potential. Invasions from north and south created the conditions for
conflict and resulted in a pattern of small states, the boundaries of which often
altered. By 1800 the Ottoman, Persian and Russian Empires had converged and
were competing for control.
In 1813 Russian forces occupied the Kura River lowlands adjacent to the
Caspian Sea, and sovereignty was conferred by the Treaty of Gulistan in the
same year. Fifteen years later Armenia and the detached fragment of Azerbaijan
around Nakhichevan also fell to Russia. Apart from some minor changes, the
present boundary was established by the Peace Treaty of Torkaman, southeast
of modern Tabriz, on 10 February 1828. It was located on the Iranian bank of
the River Aras, but in 1954 it was moved to the talweg.
East of the Caspian Sea Russian advances towards areas of Persian settle-
ment required a boundary agreement in 1869. The boundary followed the River
Attrek, which at that time flowed into the Caspian Sea. In 1881 this boundary
was extended to Babadurmaz, about 75 km east of Ashkhabad. In 1893 the
boundary was carried through to the Afghan tri-junction. The Geographer (1978,
6, 9) has noted that while the Atrek River flowed into the Caspian in 1869 it
now sinks into the sand 85 km east of the Caspian Sea. In 1954 the boundary
was redefined.
The Middle East 273
East of the Caspian Sea, the boundary leaves the seacoast [of the Caspian]
at a point 14 miles north of the Iranian Lagoon of Naftlijeh and proceeds in
a straight line east-southeast across the desert for 18 miles to join the old
boundary at Senger Tepe [Hill]. Here it turns northeast across the desert
for 13 miles to the Musa Khan, or ‘dry bed’ of the Atrek, which it follows
for 20 miles to a point near the Iranian town of Inchaburun, meeting there
the permanent channel of the Atrek. (The Geographer, 1978, 9)
In 1954, 1957 and 1970 Iran and the Soviet Union settled boundary issues
concerned with minor changes in position, Russian sovereignty over the oasis
of Firyuza, and hydro-electric plants on boundary rivers. In 1991 the Soviet
Union shattered into a number of states. West of the Caspian Sea Azerbaijan
and Armenia inherited Russia’s boundary with Iran.
The Ottoman and Russian Empires moving in opposite directions became
neighbours east of the Black Sea. Russia inflicted a major defeat on the Turks
in 1829, and by the Treaty of Adrianople of the same year extracted a number
of territorial and strategic concessions, The boundary east of the Black Sea was
delimited along the northern limits of the Turkish Sanjaks of Batumi, Ardahan
and Kars. Turkey and Russia collided again in the Crimean War of 1853-6,
but Britain and France became Turkish Allies and Russia was prevented from
obtaining further concessions from Turkey. In 1877 another war flared and this
time Turkey was left to fight alone and experience another major defeat. By the
Treaty of San Stefano of 3 March 1878 Russia exacted major concessions that
Turkey could not refuse (Roberts, 1999, 185). It appears Russia overplayed its
hand because Britain and other major powers insisted that the Treaty of San
Stefano must be changed, for reasons that Lord Salisbury set out clearly.
It is in the power of the Ottoman to close or to open the Straits, which form
the natural highway of nations between the Aegean Sea and the Euxine. Its
dominion is recognised at the head of the Persian Gulf, on the shores of the
Levant, and in the immediate neighbourhood of the Suez Canal. It cannot
be otherwise than a matter of extreme solicitude to this country that the
Government to which this jurisdiction belongs should be so closely pressed
by the political outposts of a greatly superior power that its independent
action, and even existence, is almost impossible. The result not so much
from the language of any single article in the Treaty as from the operation
of the instrument as a whole. (Roberts, 1999, 189)
At the Berlin Congress of June 1878 Russia’s demands against Turkey were
moderated. The Sanjaks of Batumi, Ardahan and Kars were ceded to Russia and
the boundary between the two countries became their southern limits.
The next boundary change was contained in the Treaty of Sevres on 10 August
1920. If it had been adopted the independent state of Armenia would have been
created. However, Soviet and Turkish forces combined. Azerbaijan became a
Soviet Republic and joined the Revolution and Turkey annexed the southern
274 Chapter 9
to the head of the Persian Gulf, a distance of about 2,000 km. Two centuries
elapsed before the Treaty of Erzerum on 31 May 1847 made provisions for a
joint committee to delimit the line. By the Treaty of Berlin, on 13 July 1878,
the northern terminus of the line was moved to a point approximately midway
between the Black and Caspian Seas. By agreements dated 21 December 1911
and 17 November 1913 the delimitation of the boundary was completed and pro-
visions were made for its demarcation by 125 pillars early in 1914. The survey
teams consisted of British, Iranian, Russian and Turkish members. A series of
maps were drawn to show the location of the pillars at a scale of 1:73,050 (The
Geographer, 1978a, 4).
After its defeat in World War I Turkey was stripped of much its Empire in
Arabia and North Africa. The British Mandate of Iraq reduced the boundary
between Turkey and Iran to about 520 km. The shortened boundary between Iran
and Turkey was demarcated by 1937. Iraq inherited the former Turkish boundary
with Iran that extends over 1,100 km.
In 1918 Turkey was divested of its Empire in the Middle East and North
Africa. The new southern boundary of Turkey consisted of two segments. The
eastern segment separated Turkey and the British Mandate of Iraq. It extended
for about 365 km from the boundary with Iran. The second, longer, section
stretched for about 820 km between Turkey and the French territory of Syria. It
extended from the Anglo-Turkish tri-junction to the Mediterranean coast. Britain
and France negotiated separate agreements with Turkey.
The British Mandate of Iraq consisted of the Vilayets [Provinces] of Al Basra,
Baghdad and Mawsil. The Turkish boundary extended from the tri-junction with
Iran to the vicinity of Fayshkhabur.
In 1923 the parties tried to negotiate a boundary but were unsuccessful. Britain
thought the line should lie between 6 km and 43 km north of the boundary of
the Vilayet, while Turkey thought that the southern boundary of Mawsil Vilayet
would be appropriate. The matter was referred to the League of Nations and no
alteration of the boundary was permitted during its deliberations. With only very
small modifications the Council announced that the existing northern boundary
of the Vilayet should become the international boundary. Both countries then
agreed to refer the matter to the International Court of Justice. The three members
of the Court agreed that the boundary should be the existing northern boundary of
the Mawsil Vilayet. By a treaty dated 5 June 1926 Turkey and Britain approved
the Vilayet boundary with one deviation. The line was diverted southwards,
in favour of Turkey in the vicinity of two villages called Aluman and Ashuta.
This was done so that the road linking them would not intersect the international
boundary (The Geographer, 1964a, 5).
In April 1920 France secured a mandate for Syria and Lebanon. The extent of
Lebanon was increased by the addition of Trablou [Tripoli], Beyrouth [Beirut],
Saida [Sidon], and Sour [Tyre]. Syria consisted of Dimashq [Damascus], Al Lad-
hiqiyah [Latakia], Halab [Aleppo], the Sanjak of Alexandretta [Iskenderon] and
276 Chapter 9
Jebel Druze. The peace treaty of 10 August 1920 outlined the boundary between
Syria and Turkey. Agreement between France and Turkey on 20 October 1921
defined some changes in the line from the Mediterranean to the tri-junction with
Iraq on the Tigris River. The Sanjak of Alexandretta was subjected to a special
status within the French Mandate.
The boundary was defined in four sectors. They were the Mediterranean to
Maydan Ikbiz, from Maydan Ikbiz to Cobanbey, Cobanbey to Nusaybin and
Nusaybin to the Tigris River. The second and third sectors were defined in the
Franco-Turkish Convention of 30 May 1926. The fourth sector was set out in
the Franco-Turkish Protocol of 22 June 1929. Some small changes were made
in these sectors by an agreement dated 23 June 1939 (The Geographer, 1978b,
14-15).
In 1936 the future of the Sanjak of Alexandretta became an important issue
in Franco-Turkish relations.
When France commenced negotiations with Syria regarding its independence
Turkey raised the question of the future of the Sanjak of Alexandretta. Con-
ditional recognition given by Turkey in 1921 to French rule in Antioch and
Alexandretta had not implied the inclusion of these districts in the mandated
territories, and that if France intended to reduce her responsibilities in
the Levant, she should concede independence to the Sanjak as a separate
unit, on the same terms as to Syria and Lebanon. In December 1936 the
Turkish Government, with the assent of the French, referred this question
to the League of Nations. With assistance from that quarter, French and
Turkish representatives agreed on the separation of the Sanjak from Syria,
qualified by Syrian control of its foreign policy, and by tariff and monetary
union with Syria. The Statute and Fundamental Law of this territory were
adopted at Geneva on 29 May 1937, and were to come into operation six
months later. The preparation of electoral registers for the first session of
the Sanjak’s Legislative Assembly gave rise to serious friction between the
Turkish and the various non-Turkish communities and led ultimately to the
replacement of the League’s Electoral Commission by a body consisting
of three Turks and one Frenchman. The Turkish community secured 22 of
the 40 seats on the Assembly; during its first session in September 1938 it
elected as head of the new State a Sanjak-born Deputy of the Turkish Great
National Assembly, and adopted the Turkish name for the Sanjak, which
henceforward was to be known as the Republic of the Hatay. The new
Republic, nominally in close relations with Syria but in reality controlled
by Turkey, was absorbed by the latter within twelve months of its effective
establishment. (Naval Intelligence Division, 1942, 329-30)
Two months before the start of World War II in Europe, on 23 June 1939, the
eastern and southern boundary of Hatay became the boundary between Syria
and Turkey.
The Middle East 277
The southern region can be considered in two sections. First there are the bound-
aries that developed within and between the mandates awarded to Britain and
France after World War I. Second there are the boundaries generated between
Saudi Arabia and its seven neighbours.
The Conference of San Remo on 24 April 1920 delivered the mandates of
Lebanon and Syria to France, and those of Palestine and Trans-Jordan to Britain.
The British mandate over Iraq was confirmed nine days later. Britain and France
were able to delimit, unilaterally, boundaries between their mandates, and in
concert they could delimit boundaries between their adjoining mandates.
Once in control of the Levant, France lost no time in achieving a goal that,
General Gourand noted, had been identified in 1862 (Temperley, 1924, 167-8).
France issued an arête on 31 August 1920, creating a Greater Lebanon State.
To the original area of Lebanon were added Beirut, Tripoli and the valley called
El Beqa’a. Toynbee (1927, 355-6) drew attention to the large Arab population
included in the new territory, and hinted at the problems that developed in the
1970s.
Britain determined the boundary between Jordan and Iraq in 1932, when the
latter became independent. It was achieved by an exchange of letters amongst
Nuri as-Sa’id, Prime Minister of Iraq, King Abdullah of Saudi Arabia and C.H.F.
Cox the British Resident in Amman. This short boundary joined the tri-junction
with Syria to the tri-junction with Saudi Arabia.; located at 33° 22’ 29” N and
38° 47’ 33” E and 32° 13’ 51” and 39° 18’ 09” N (The Geographer, 1970, 3).
In 1984 Iraq and Jordan exchanged small areas of land measuring about 220
sq. km in the most southern 40 km of this boundary, giving it a rectangular
configuration.
Britains unilateral boundary separating the mandates of Jordan and Palestine
was published on 2 September 1922 (Naval Intelligence Division, 1943, 2). It
was based on the north-south drainage pattern of the Jordan River, the Dead Sea,
Wadi el Jeib and Wadi Araba. The sector between Wadi al Yabis and a point
near the middle of the Dead Sea, is shown by the Times Atlas (2000, map 35)
as a disputed boundary between the West Bank and Jordan.
The boundary between the British and French Mandates was defined from the
Tigris River to the Mediterranean coast on 23 December 1920. This Convention
fixed the principal alignment of the boundary that was only modified slightly in
subsequent years.
The boundary between British Palestine and French Lebanon and Syria was
settled by the Anglo-French Agreement of 7 March 1923 (The Geographer, 1967,
9). This Agreement contained a report on the demarcation of the boundary from
the Mediterranean coast to the lower valley of the Yarmuk River about 1 km
above El Hamma. This demarcation was approved by the League of Nations in
1934.
278 Chapter 9
Eight years later the boundary between the Mandates of Syria and Jordan
was defined in the Anglo-French Protocol of 31 October 1931 (The Geogra-
pher, 1969, 10). Only small changes were made to the line defined in the 1920
Convention. The definition of the boundary includes references to the railway
from El Hamma to Dar’a.
. . . the marking of the boundary will be made south of the [railway] track
and parallel to it in such a way as to leave to Syria, in addition to the
railway itself and its permanent structures, the “emprisea” [undertakings]
borrow-pits, stations, outbuildings and ground necessary both for the tech-
nical protection of the railway track and its structures as well as for its
exploitation. (The Geographer, 1969, 8)
This railway is shown on Map 35 of the 1967 The Times Atlas of the World
(1967). It does not appear in subsequent editions.
The boundary description provides evidence that attempts were made to avoid
disrupting local territories close to the boundary.
It is understood that, if the village of Kirbet Aouad or any part of this village
were found to be south of the frontier line marked out as stated above, the
frontier would bend to a point situated at sixty metres south and round the
last group of houses at present existing: the junction with the general line
being made east and west of the village by lines forming an angle of about
90 degrees with the point above indicated (60 metres south of the village)
in such a way as to include in Druze territory all the inhabited part of this
village as well as the territories situated in this angle reconnecting with the
general line. (The Geographer, 1969, 9)
From a point 3.2 km north of the highest summit of Tell [Hill] Romah, the
boundary proceeds ‘. . . as a straight line . . .’ in the direction of Abu Kemal on
the Euphrates River. The Syria-Jordan boundary terminated about 15 km south
of Jabal at Tanf located at 33° 22’ 29” North and 38° 47’ 33” East.
The boundary between the Mandates of Iraq and Syria was delimited in a
Report issued in Geneva on 10 September 1932, twenty-three days before Iraq
became independent (The Geographer, 1970a, 7). The northern origin of the
boundary was the confluence of the Tigris and Khabur Rivers and the talweg of
the Tigris River was followed for 1 km downstream from Fayshkhabur. Leaving
the Tigris the boundary followed a southwesterly then southerly course for about
390 km to the Euphrates River. The twenty straight segments, measuring from 6
km to 66 km, were drawn between trigonometrical points. The terminus of this
section lay close to the northeast point of Baghuz Island. The boundary followed
the talweg of the Euphrates northwards for a short distance to the boundary
separating the territories belonging to the villages of Dulaym al Husaybah and
Heri, before turning westward to the Leachman Stone about 1 kilometre from the
river. The origin of this stone is uncertain. During World War I no Iraqi force
The Middle East 279
rose to assist the British against the Turks. However, the Beduin assisted the
Allied cause. The Anaiza,, the great Beduin Confederation of the Syrian Desert
were an exception.
They were under the influence of Colonel Leachman, the remarkable Politi-
cal Officer for the desert, whose power with the beduin throughout the war
not only rivalled that of Lawrence in the Hejaz but was built up simply
on force of character unaided by the ample subsidies which Lawrence so
liberally dispensed. The Anaiza never undertook any major military opera-
tions, but they helped to blockade the desert. (Naval Intelligence Handbook,
1944, 270)
As a result of Iraqi dissatisfaction with progress to independence in the first half
of 1920 a revolt started in July and gained pace. Fortunately the insurrection was
restricted to two tribal zones extending from Al Miqdadiyah to Ba’qubah and
Al Musayyib to As Samawali.
But the Tigris tribes from Baghdad to Basra and the tribes of the lower
Euphrates around [An] Nasiriya gave no serious trouble, nor did the desert
beduin. This localization of the rebellion was due to the outstanding courage
and force of character of the divisional Political Officers who, with no support
save their Arab policemen or levies, remained firm in their offices or toured
their districts under the shadow of assassination. Some were murdered, as
Colonel Leachman, who by his personal prestige and courage kept the peace
for two months above Falluja, a feat as great as any done during the war by
the better known Lawrence. (Naval Intelligence Division, 1944, 290)
It is unclear whether the Leachman Stone was erected by the Colonel or in his
memory.
The next point is located at the intersection of an arc drawn with a radius
of 30 km from the minaret of Abu Kemal, and a straight line drawn from the
minaret to a point 3.2 km from the highest peak of Tell Romah. This intersection
is about 29 km west of the Leachman Stone. From the intersection the boundary
continues in a straight line to the trijunction with Jordan.
Richard Schofield (1993) produced an excellent, detailed account of the evolu-
tion of the boundary between Iraq and Kuwait (Figure 9). Until 1896, Britain had
accepted that Turkish jurisdiction extended along the south coast of the Persian
Gulf from Basra to Qatif, just north of modern Bahrain. In the next three years
Britain’s view was reversed.
Chiefly responsible for this transformation was Shaikh Mubarak, who, as
unofficial Commander of the Al-Sabah armed forces and with the support of
many Kuwaiti townspeople and Bedouin tribes, murdered his brothers, Shaikh
Muhammad (the ruler) and Jarrah (effectively the financial secretary), to
assume rule in May 1896. Mubarak, effectively carved out a niche for him-
self and a prominent position for Kuwait by playing off the Ottoman Empire
280 Chapter 9
and Britain against one another. Before his death in 1915, the shaikhdom
of Kuwait had been promised independence under British protection and
possessed territorial limits, although to the south these were to be consider-
ably foreshortened after the First World War. For this Mubarak owed much
to good fortune, particularly his backing of Ibn Saud against Ibn Rashid
for the supremacy of Najd [ultimately Saudi Arabia], and the demise of the
Ottoman Empire during the First World War. (Schofield, 1993, 14)
By 1908 Lorimer, on the staff of the Government of India, had produced a
gazetteer of the Persian Gulf, Oman and central Arabia. It included a map on
which were marked the known limits of Kuwait. The territory was shaped like
a diamond and extended 160 km west of the coast and the same distance along
the coast south of Kuwait to Manifah. This British definition of Kuwait’s terri-
tory had been modified by 1912.
. . . while the town, harbour and immediate surroundings [of Kuwait] should
be completely autonomous the remainder of Koweit territory to the extent
attributed to it in Lorimer’s Gazetteer should be specifically recognized
as being under the administrative influence of the Shaikh of Koweit and
the Porte should agree neither to place military posts nor take any action
within it without the previous consent both of the Sheikh and ourselves.
(Schofield, 1993, 42)
Schofield notes that this modification was the origin of the inner ‘Red Line’ and
the outer ‘Green Line’. By 29 July 1913 the British and Ottoman Empires had
agreed on the extent of Kuwait influence within the Red and Green Lines, but,
as Schofield observes, the onset of the First World War prevented the agree-
ment from being tested. After the war ended Britain had a mandate in Iraq and
controlled Kuwait, and it moved quickly to create boundaries, between these
territories and between both territories and Saudi Arabia. The Green Line of the
1913 Anglo-Turkish agreement provided the fulcrum from which these three
boundaries were drawn.
In the north the Green Line started about 5 nm above the mouth of the Khor
[Arm of the sea] Zobeir, at 30° 1.5’ N and proceeded northwest then west south
of Safwan to the Wadi [Watercourse] al Batin.
From Zubair, 10 miles south-west of Basra, the ground rises gently to the
crest of the desert steppe known as Ar Raha, north of Kuwait territory, to
the west of which is the mouth of the Wadi Batin, marked by two bluffs, Ar
Ratak Ash Shamali and Ar Ratak al Janubi. There is excellent grazing in
Ar Raha during March. Where the Wadi Batin forms the boundary between
Iraq and Kuwait it has steep sloping banks, with a width from 1 to 4 miles
between crests on either side. The banks are cut by numerous small dry
tributaries (shaib). A clayey loam covers the bottom of the depression and
affords a fertile soil for grass and bushes. (Naval Intelligence Division,
1944, 116)
The Middle East 281
The eventual tri-junction between the boundaries of Iraq, Kuwait and Saudi
Arabia was located about 7 km north of Ar Ruq’i.
Cox, the British High Commissioner in Baghdad was the primary architect
of Saudi Arabia’s boundaries with Iraq and Kuwait. In May 1922 Cox and Ibn
Saud signed the Treaty of Mohammerah [Khorramshahr] and then delimited the
boundaries in December 1922.
Meanwhile Sir Percy Cox, who had already settled the disputed boundary
between Nejd and Kuweit, arranged the Treaty of Mohammerah (1922),
which assigned the Muntafiq, Dhafir and Amarat tribes to Iraq and the
southern Shammar to Nejd, and also forbade tribal aggression. Ibn Saud met
Cox at ‘Oqeir [Uqayr] in the winter of 1922-3, when a protocol defined the
boundary separating the traditional wells and lands of the tribes concerned,
and formed a neutral zone. (Naval Intelligence Division, 1946, 297)
In fact Cox had indicated to Ibn Saud in early 1921 that Britain was only prepared
to guarantee the Red Line of the 1913 Convention as the southern boundary of
Kuwait (Schofield, 1993, 57).
On 2 December 1922 the Protocol of Uqayr defined the boundary that was
mainly determined by Cox. The line extended for about 690 km from the tri-junc-
tion with Kuwait, near Ar Ruq’I to the Jabel [Mountain] Unayzah, the tri-junction
with Jordan. Adjacent to the tri-junction with Kuwait a neutral zone was defined.
It had an area of about 13,800 sq. km and was available for the peaceful use of
both governments. The Protocol contained two other provisions.
Whereas many of the wells fall within the Iraq boundaries and the Najd side
is deprived of them, the Iraq Government pledges itself not to interfere with
those Najd tribes living in the vicinity of the border should it be necessary
for them to resort to the neighbouring Iraq wells for water, provided these
wells are nearer to them than those within the Najd boundaries.
The two governments mutually agree not to use the watering places and
wells situated in the vicinity of the border for any military purpose, such
as building forts on them, and not to concentrate troops in their vicinity.
(The Geographer, 1971, 9)
Schofield (1994, 27) describes the boundary changes made by Iraq and Saudi
Arabia in the period from 1975 to 1981 (Figure 9.4). A line, joining its west
and east corners, divided the Neutral Zone, and the boundary westwards to the
Jordan tri-junction was smoothed. The new boundary measures about 830 km.
Features named Nisab, Al Lifiyah, Judayyayit ‘Ar’ar and Maqar an Na’am were
on the original boundary and remain on the 1981 line.
Wilkinson commented on Cox’s role in deciding the northern boundaries of
Saudi Arabia.
The result was Cox’s famous line-drawing act. He carefully marked out
the Iraqi frontier at Ibn Saud’s expense, but to make up for this, ruthlessly
282 Chapter 9
1992. Although the report on the demarcation reports only that he resigned for
‘personal reasons’ there is no mention of the reservations he expressed about the
issue of delimiting the maritime boundary through the Khor Abdulla (Alexander
and Pietrowski, 1998, 2401).
Though there is no question that the former Indonesian Foreign Minister
probably wanted to move on to other things, it was no secret that the
UNIKKBDC chairman was troubled by what he regarded as the inadequate
mandate possessed by the border demarcation commission to demarcate, or
more correctly delimit, the boundary along the Khor Abdulla. The problem
as he saw it was that no delimitation for the Khor existed in UNIKBDC’s
‘delimitation formula’ (the boundary so vaguely described by the 1932
exchange of notes) which could then be demarcated, This he made clear in a
letter addressed to the Legal Counsel of the United Nations on 6 November
1992. (Schofield, 1993, 180)
Alexander and Pietrowski (1993) do not refer to this development. Within a
month the new Chairman had decided that ‘there was sufficient basis to proceed
with demarcation of the Khor Abdullah’ (Schofield, 1993, 181).
By the Khwor Abd Allah section, the Commission refers to the maritime, or
offshore boundary from the junction of the Khor Zhobier and the Khowr Abd
Allah to the eastern end of the Khowr Abd Allah. The Commission felt that
the closing statement of the delimitation formula mentioning the islands of
Warbah, Bubiyan, etc. as appertaining to Kuwait, gave an indication that
the existing frontier in that section lay in the Khowr Adb Allah. (Schofield,
1993, 188)
Schofield, rather gently calls the basis for this decision tenuous! In fact, while
there is no doubt that eventually a boundary would be drawn within the Khor Abd
Allah, no such line had been delimited in the Khor by the 1932 Exchange of Let-
ters, that constituted the Commission’s delimitation formula for the demarcation
of the Iraq-Kuwait boundary. The Agreed Minutes consisted of two sentences.
The first delimited a boundary which UNIKBDC could demarcate. The second
sentence simply names the islands that appertain to Kuwait and does not delimit
any boundary.
The Secretary-General made a simple and appropriate assertion in his letter
covering the report on the boundary demarcation.
In accordance with its mandate and terms of reference, the Commission was
called to perform a technical and not a political task and as it is stressed in
the Final Report, the Commission has made every effort to strictly confine
itself to this objective. In the statement of the President of the Security Coun-
cil dated 17 June 1992, issued on behalf of its members and in the Security
Council resolution 773 (1992) of 26 August 1992, related to the work of
the Commission, it was pointed out that through the demarcation process
The Middle East 285
the Commission was not reallocating territory between Kuwait and Iraq,
but was simply carrying out the technical task necessary to demarcate for
the first time the precise coordinates of the boundary set out in the Agreed
Minutes referred to above. (Alexander and Pietrowski, 1998, 2395)
The Secretary-General’s observation does not apply to the offshore section. The
Agreed Minutes did not delimit a boundary through the Khor Abd Allah. The
reference to Warbah and Bubiyan belonging to Kuwait did not delimit a bound-
ary through the Khor. The Commission did that on the basis of weak arguments.
It will be remarkable if Iraq, at some time in the future, does not argue that the
definition of the maritime boundary through the Khor Abd Allah needs to be
revisited.
In the 1920s Britain tried to negotiate a boundary between its Mandate of
Trans-Jordan and Saudi Arabia, that gave Jordan access to the Gulf of Aqaba
and the Red Sea. These efforts were unsuccessful, but in 1927, Ibn Saud agreed
to respect the British boundary pending a final solution. Jordan became inde-
pendent in 1946 and 19 years later a boundary was defined with Saudi Arabia
(Figure 9.5). The major change provided Jordan with a longer coastline on the
Gulf of Aqaba, south of the port of Aqabah. The boundary changes in the inte-
rior transferred 6000 sq. km to Jordan and 7,000 sq. km to Saudi Arabia (The
Geographer, 1965a, 3).
In 1892, when Turkey allowed Egyptian authorities to administer Sinai,
Egypt became effectively a British protectorate, although still technically part
of the Ottoman Empire. The British then proposed a northern boundary for
Egypt consisting of a straight line drawn from a point just east of El Arish and
terminating on the Gulf of Aqaba west of Elat. A few years later the Ottoman
Empire proposed two alternative boundaries. The first joined a point east of El
Arish to the southern end of the Suez Canal and then on to Elat, The second
started from the same northern point and went directly to Ras Muhammad, the
southern tip of the Sinai Peninsula. Control of the western shore of the Gulf of
Aqaba was a major aim of British policy, and that aim was secured when, in
October 1906, the two countries drew a boundary from the vicinity of Rafah on
the Mediterranean coast to the head of the Gulf of Aqaba just west of Elat (The
Geographer, 1965b).
In the First World War and after the Second World War the tides of battle
flowed across this boundary in both directions. Since 1978 the Anglo-Turkish
boundary has separated Israel, including Gaza, and the United Arab Republic.
Schofield has drawn attention to the private nature of boundary agreements of
the states that fringe Saudi Arabia from Qatar to Yemen.
Ever since the early 1920s the Saudi state has shown a marked preference
for coming to private territorial arrangements with its neighbours in the
Gulf and the southern Arabian Peninsula (precise details of which have
often remained elusive) rather than negotiating explicit boundary agreements
286 Chapter 9
and registering their texts with the United Nations or other appropriate
international institutions. (Schofield, 1994, 52-3)
Schofield (1994, 58) provides a clear map showing the boundaries drawn between
Oman and the United Arab Emirates in the 1950s and 1960s, the Saudi-Qatar
boundary of 1965, the Saudi-United Emirates boundary in 1974, the Saudi-Oman
boundary of 1990, and the Oman-Yemeni boundary of 1992.
The only border in Arabia not governed by an agreement of any kind is
that between Saudi Arabia and the newly constituted Republic of Yemen,
that is, except for its westernmost stretch delimited by the Taif line of 1934.
(Schofield, 1994, 60)
Following the defeat of Turkey Ibn Saud began to establish the limits of Saudi
Arabia towards the southwest. By 1920 the mountainous Asir region had sub-
mitted. Following a civil-war Hussein accepted protection from Ibn Saud on
the understanding that the territory would be incorporated into Saudi Arabia
on Husein’s death (Naval Intelligence Division, 1946, 301-2). Husein raised a
rebellion in 1932 and Saudi Arabia was victorious in 1934 when the Treaty of
Taif was signed.
With remarkable moderation Ibn Saud imposed only the settlement he had
offered before the outbreak of war; that is, a strip of disputed territory
in the south of Asir was definitely incorporated in the Saudi kingdom, an
arrangement which shifted the northern frontier of Yemen some distance to
the south of that shown on most maps. The frontier now touches the coast
only just north of the Yemeni port of Meidi [Midi], whence it runs steeply
northeastward to the watershed, then bends southeast and finally takes a
generally easterly direction nearly to the edge of the Great Desert. (Naval
Intelligence Division, 1946, 302)
The terminus lay about 30km from Najran. On 12 June 2000 Saudi Arabia and
Yemen affirmed the binding nature and legitimacy of the treaty. The boundary
identified in the treaty was re-affirmed and demarcated by 17 columns. The ter-
minus of the 1934 boundary is located at 17° 26’ N and 44° 21’ 58” E. It was
agreed that the continuation of the boundary eastwards would terminate at 19°
N and 52° E that is the trijunction with Oman (Al-Enazy, 2005). It was reported,
on 31 May 2006, that Saudi Arabia and Yemen had delimited their common
boundary from the terminus of the Taif line to the terminus of the Oman-Yemen
boundary. In 2006 Saudi authorities published a fine map of the Kingdom at a
scale of about 1:2.03 millions. The new boundary with Yemen started from the
terminus of the 1934 Line at about 44° 35’ East and 17° 21’ North. The new
boundary follows an easterly course for about 240 km to the Wuday Ah salient.
This is about 45 km wide and 40 km deep. The boundary then continues east-
northeast by four straight lines for 455 km to the tri-junction with Oman. The
map shows Saudi Arabia’s boundaries with its neighbours, including Qatar, the
The Middle East 287
United Arab Emirates and Oman. The southern sector of the boundary between
Oman and the United Arab Emirates is also shown as far as Al Ayn. It does not
show Oman’s territory on the southern shore of the Strait of Hormus, nor the
Omani enclave shown in the Times Atlas (2003, 33), a few kilometres west of
Khawr Fakkan, on the shore of the Gulf of Oman.
Roberts published a useful analysis of final boundary agreement.
Assessing who won and who lost is an almost impossible task, for this was
a dispute in which it is still not clear what the Saudis and the Yemenis were
claiming in the first place. Neither side ever went public with its assessments
of where the border should be, although it appears that the Yemenis were
largely relying on putative boundary lines proposed (but not formalised)
by the former British colonial authorities ruling Aden, while Saudi Arabia
appeared to be pushing for a boundary line much further south, in places
along the 17° N parallel. The closest hints that outsiders could ever get came
from commercial maps published in the two countries and which, it might
reasonably be assumed, reflected at least the aspirations of the respective
governments concerning the locations of the borders.
In practice, both sides seem to have won – but to have won somewhat
different conflicts. For the Yemenis their success was territorial and eco-
nomic, for the Saudis it was a major diplomatic and security achievement.
(Roberts 2000, 70)
CONCLUSIONS
In terms of boundary evolution the northern and southern regions provide sharp
contrasts. Iran and Turkey were sometimes at a disadvantage in negotiating their
northern boundaries with Russia, but Iranian claims to the east, where Britain held
sway, were not severely compromised. After World War I Turkey was stripped of
its Empire, but, in terms of national development, that was a benefit. The resur-
gence of Turkey preserved its territory in Europe and produced a fair boundary
with Syria and Iraq, that was followed, just before the onset of World War II, by
the return of Hatay from the French Mandate. Meanwhile, Iran’s long-standing
boundary with Turkey, had been inherited by its new neighbour Iraq.
The principal powers that dominated boundary development in the southern
Region were, in order of importance, Britain, Saudi Arabia and France. By April
1946 both France’s Mandates had been discharged within definite boundaries.
While Britain was able to achieve the same success with Iraq and Jordan in 1932
and 1946 respectively, events in Palestine in 1947-8 were an inglorious chapter
in British history, following closely on the disastrous partition of India in August
1947. Britain did manage to withdraw gracefully from Kuwait, Bahrain, Qatar
and the United Arab Emirates by 1971, in contrast with its 1967 abdication in
288 Chapter 9
REFERENCES
Al-Enazy, A.H., 2005. The long road from Taif to Jeddah: resolution of a Saudi-
Yemeni boundary dispute, Abu Dhabi.
Alexander, L.M. and R.F. Pietrowski, 1998. ‘Iraq-Kuwait’, in International
Maritime Boundaries, Volume III, edited by J.I. Charney and L.M. Alexan-
der, 2387-2432.
Dewdney, J., 1987. ‘Structure and relief’, in The Cambridge Atlas of the Middle
East and North Africa, G. Blake, J. Dewdney and J. Mitchell, editors, Cam-
bridge University Press: Cambridge, 7-9.
Goldsmid, Sir F.J., 1876. Eastern Persia, 2 volumes, London.
Holdich, Sir T.H., 1916. Political frontiers and boundary making, Macmillan:
London.
IBRU, 2000. ‘The international boundary treaty between the Republic of Yemen
and the Kingdom of Saudi Arabia’, Boundary and Security Bulletin, 8 (2),
63-7.
India Foreign Department, 1890. No. 49 Enclosures of a letter to Her Majesty’s
Secretary of State for India, 5 May 1890, Simla.
India, Foreign Department, 168/1896. Demarcation of the boundary between
Persian Baluchistan and Kalat, Simla.
McMahon, Sir H., 1906. ‘Recent survey and exploration in Seistan’, Geographi-
cal Journal, 28, 209-28, 333-51.
Naval Intelligence Division, 1942. Turkey, Volume I, HMSO: London.
———, 1943. Palestine and Trans-Jordan, HMSO: London.
———, 1944. Iraq and the Persian Gulf, HMSO: London.
———, 1946. Western Arabia and the Red Sea, HMSO: London.
Prescott, J.R.V., 1975. Map of mainland Asia by treaty, Melbourne University
Press: Melbourne.
Roberts, A., 1999. Salisbury: Victorian Titan, Weidenfeld and Nicolson:
London.
Roberts, J., 2000. ‘The Saudi-Yemeni boundary treaty’, Boundary and Security
Bulletin, 8 (2), 70-3.
Schofield, R., 1993. Kuwait and Iraq: historical claims and territorial disputes,
The Royal Institute of International Affairs: London.
The Middle East 289
———, 1994. ‘Borders and territoriality in the Gulf and the Arabian peninsula,
during the twentieth century’, in Territorial foundations of the Gulf States,
edited by R. Schofield, University College London Press: London.
Sykes, Sir P., 1940. A history of Afghanistan, 2 volumes, London.
Temperley, H.W.V., 1924. A history of the Peace Conference of Paris, Volume
VI, Oxford University Press: London.
The Geographer, 1961. ‘Afghanistan-Iran boundary’, International Boundary
Study, No. 6, State Department: Washington.
———, 1964. ‘Turkey-USSR boundary’, International Boundary Study, No. 29,
State Department; Washington DC.
———, 1964a. ‘Iraq-Turkey Boundary’, International Boundary Study, No. 27,
State Department: Washington DC.
———, 1965. ‘Bulgaria-Turkey Boundary’, International Boundary Study, No.
49, State Department: Washington DC.
———, 1965a. ‘Jordan-Saudi Arabia Boundary’, International Boundary Study,
No. 60, State Department: Washington DC.
———, 1965b. ‘The Israel-United Arab Republic Boundary’. International
Boundary Study, No. 46, State Department: Washington DC.
———, 1967. ‘Israel-Lebanon Boundary’, International Boundary Study, No.
75, State Department: Washington DC.
———, 1969. Jordan-Syria Boundary, International Boundary Study, No. 94,
State Department: Washington DC.
———, 1970. ‘Iraq-Jordan Boundary’, International Boundary Study, No. 98,
State Department: Washington DC.
———, 1970a. ‘Iraq-Syria Boundary’, International Boundary Study, No. 100,
State Department: Washington DC.
———, 1970b. The Kuwait-Saudi Arabia Boundary’, International Boundary
Study, No. 103, State Department: Washington DC.
———, 1971. ‘Iraq-Saudi Arabia Boundary’, International Boundary Study,
No. 111, State Department: Washington DC.
———, 1978. ‘Iran-USSR Boundary’, International Boundary Study, No. 25,
State Department; Washington DC.
———, 1978a. ‘Iran-Iraq’, International Boundary Study, No. 164, State Depart-
ment: Washington DC.
———, 1978b. ‘Syria-Turkey’, International Boundary Study, No. 163, State
Department: Washington DC.
The Times, 1967, The Times Atlas of the World, Times Newspapers: London.
Times Books, 2000. The Times Atlas of the World, Times Books: London.
———, 2003. The Times Atlas of the World, Times Books: London.
Toynbee, A.J., 1927. Survey of international affairs 1925, Volume I, Oxford
University Press: London.
Wilkinson, J.C., 1993. Arab frontiers: The story of Britain’s boundary drawing
in the desert, Tauris: London.
10. AFRICA
Africa, the second largest continent, has nearly twice as many states as Asia.
Most of Africa’s international boundaries were drawn by European governments
before 1914. Since that date only minor changes have occurred. ‘The Scramble
for Africa’ is a familiar phrase generally associated with the conference that
produced the General Act of Berlin in the period 15 November 1884 to 27
February 1885.
The signing of the General Act of Berlin on 26 February 1885 must have
brought a general sigh of relief from the assembled delegates. Bismarck
gave the concluding speech. They were in ‘complete accord’ about ‘all
points of the programme’. They had secured free access to the interior of
Africa for all nations, freedom of trade in the whole basin of the Congo (and
in a region beyond the Congo named the ‘conventional basin’). They had
also shown ‘much careful solicitude’ for the moral and physical welfare of
the native races, and the Chancellor cherished the hope that this principle
would ‘bear fruit’ and help introduce the populations to the advantages of
civilization.
At least the conference was over – and amicably so. But what exactly
had been achieved? There were thirty-eight clauses to the General Act,
all as hollow as the pillars in the great saloon. In the years ahead people
would come to believe that this Act had a decisive effect. It was Berlin that
precipitated the Scramble. It was Berlin that set the rules of the game. It
was Berlin that carved up Africa. So the myths would run.
It was really the other way round. The Scramble had precipitated Berlin.
The race to grab a slice of the African cake had started long before the first
day of the conference. And none of the thirty-eight clauses of the General
Act had any teeth. It had set no rules for dividing, let alone eating, the cake.
(Pakenham, 1991, 254)
The Scramble was not a sprint. It was the final stages of a marathon.
292 Chapter 10
The early invasions of Africa occurred along the Mediterranean coast from
Arabia. First the nomadic Hyskos invaded Egypt and held sway from 1730 BC
to 1570 BC. Assyrians invaded in 671 BC and were replaced by the Persians
from 525 BC to 404 BC. The Romans held sway from 146 BC until 428 AD.
The Phoenicians, dislodged from the Syrian coast by the Assyrians established
themselves in modern Tunisia in 814 BC. Carthage was founded and the new
state was founded on a powerful navy and a fertile hinterland that was farmed by
Berbers. Mercenary regiments were employed to defend the state, but the state
was exposed if its fleet was defeated or evaded by enemies.
This weakness was discovered in the fourth century BC by Agathocles, the
Greek ruler of Sicilian Syracuse, who when hard pressed by the Carthag-
inians in Sicily, evaded their fleet, threw an army into Tunisia, and nearly
succeeded in overthrowing the Carthaginian power. After Agathocles the
Carthaginians found a new enemy, the Romans, who, after uniting all Italy
into a confederation, were extending their power as protectors into Sicily.
The Carthaginians fought two great wars in the third century BC, the first
to save and the second to restore their position in the central Mediterra-
nean. The first (264-241 BC) was mainly a naval war and ended with the
294 Chapter 10
destruction of their navy and the loss of their bases in Sicily and Sardinia.
The second war, like the first, ended with complete defeat, yet Carthage was
able to recover again because she never lost the basis of her power, her
maritime commerce. The third war, deliberately provoked by the Romans and
fought solely in Africa ended in 146 BC with the obliteration of Carthage
and the massacre of a great part of its inhabitants, although many fled into
the interior. (Naval Intelligence Handbook, 1945, 128)
Kasule (1998, 36-7) provides an excellent map showing Rome’s total control
of the coast of North Africa from the Strait of Gibraltar to the Red Sea. In 428
Vandals overran the Roman territories, attacking from Ceuta in the west and
stopping only at Tripoli. In turn the Byzantine Empire had cleared the Vandals
from North Africa by 534 and held the entire coast except that part between
Oran and Ceuta. After 639 the successful Arab invasion of Egypt pressed on to
modern Algeria. Finally, having captured Egypt in 1517 the Ottoman Empire
extended its reach along the coast to dominate the Mediterranean coast as far as
Oran. The Empire’s final foothold of Libya was taken by Italy in 1912.
Fernandez-Armesto (1991) edited a fine illustrated account of the history of
world exploration on which the following account, of the European discovery
of the west and east coasts, is based.
West Africa and the Sahara . . . was an area which had nourished indigenous
civilizations and long-range trade for centuries before Europeans arrived. It
also played a vital role in the early history of European maritime expansion,
since it was precisely the wealth of the area which attracted explorers and
the inaccessibility of the desert routes that encouraged seafarers to find a
way round it in their ships. (Fernandez-Armesto, 1991, 59)
Mali was the most remote and largest source of gold in the first half of the 14th
century. Attempts to sail down the northwest coast of Africa started at the end
of the 13th century but yielded no lasting cartographic results. The prospect of
success increased when the Portuguese developed a caravel that could sail close to
the wind. By 1450 the mouths of the Rivers Senegal and Gambia had been reached
and partially explored. However, by the time Mali was reached the Empire had
been in decline for a century. There were two other discoveries before the route
to the Indian Ocean was unlocked. First the Arquipelago dos Açores, 1,600 km
west of Lisbon, had been discovered, claimed and settled by the 1430s. Second
the coast beyond the Gambia was explored southwards to Punta de Campo by
1471. This point, located at 2° 19’ N, subsequently became the terminus of the
boundary between French Cameroun and Spanish Rio Muni. This remarkable
advance was achieved during the six-year lease of exclusive trading rights, granted
by Portuguese authorities in 1469 (Fernando-Armesto, 1991, 64).
In 1485 Diego Cao continued beyond the equator, found the mouth of the
Congo River, and sailed on to 22° S near Walvis Bay
Africa 295
In the summer of 1487, Cao was followed by Bartolomeu Dias. Dias left
Lisbon with three ships and a commission to find the ocean route around
Africa. At first retracing Cao’s route, he subsequently, and with great dar-
ing, turned away from the coast, perhaps in around 27° or 28° S, in search
of a favourable wind. His success in encountering westerlies, which carried
him to a landfall some 300 miles east of the Cape of Good Hope, made a
major contribution to knowledge of the wind-system of the South Atlantic.
This expedition seems to have been exceptionally well-provisioned, suggest-
ing that the detour into the open sea was planned in advance. (Fernando-
Armesto, 1991, 64)
In 1498 Vasco da Gama rounded the Cape of Good Hope and sailed on, to the
coast of Mozambique. Cabral followed two years later. By 1501 the Portuguese
had established a presence at the Arquipelago dos Bijagos, the islands of Sao
Tome and Principe near the Niger Delta, the Angolan coast south of the mouth of
the River Congo and the Mozambique coast, especially at Sofala and Kilwa.
The Slave Trade, made possible by the European discoveries of the west
and southeast coasts of Africa and of the Central Americas by 1500, flourished
for two centuries after 1650. The numbers transported during that time have
been calculated as about 12 millions (Oliver and Crowder, 1981, 145). Kasule
(1998, 70) reports a figure of 40 millions, but the numbers on his map showing
the routes of vessels carrying slaves across the Atlantic Ocean only total 9.39
millions. Kasule (1998, 68-9) shows the trading forts established by Europeans
by 1700, along the African coast from Cape Vert to Whydah on the Dahomey
coast. There were 40 forts built and occupied by Europeans from Denmark,
England, France, the Netherlands and Portugal. There were no forts along the
coast from Sherbro Island in the west to Assini in the east along the coasts of
modern Liberia and Côte d’Ivoire.
The triangular pattern of trade carried slaves from the west coast of Africa to
work on plantations in the Americas.
Of the several explanations for the trans-Atlantic slave trade, the most
important concerns resistance to disease. The sparse local American Indian
population was decimated, by imported European diseases such as small-pox;
European mortality rates were high due to a host of tropical ailments to
which they had little immunity; Africans, accustomed to the tropical disease
environment and also to the diseases of the Euro-African landmass, had the
lowest mortality. (P.M. Martin in Oliver and Crowder, 1981, 146)
The plantations produced coffee, cotton, indigo, rice and tobacco that were sent
to Europe. European manufactured commodities were then taken to Africa to
be exchanged for slaves.
Buying and selling operations became highly conventionalised at many
points along the coast. African currencies such as iron bars in upper Guinea,
296 Chapter 10
ounces of gold-dust on the Gold Coast, cloth and cowries in the Bight of
Benin, brass manillas in the Niger delta-Cameroon region, and lengths of
cloth on the Loango coast, were adopted as the standard units of value.
European goods and African slaves were valued according to such units of
account, leaving the selection of European goods to be included in the pay-
ment for a slave as the main subject for negotiation. Items most demanded
by African businessmen were guns and powder, cloth, spirits and tobacco
and hardware. Local rulers often required the payment of fees and customs
duties, which might be paid in more exotic prestige goods. (P.M. Martin in
Oliver and Crowder, 1981, 147)
Manillas are rings of metal once worn by Africans on their arms and wrists;
Loango lies just north of Pointe-Noire on the coast of Congo.
At the height of the Slave Trade the European enterprises were concentrat-
ing on the coast rather than exploration of the interior of the continent. Bovill
explained how this situation arose.
The slave trade largely accounted for the failure of Europeans to penetrate
the interior from the coast of Guinea. Not only did it engender the bitter-
est hatred of the white man among the suffering tribes of the interior, but
it so enriched the coast chiefs that they offered determined opposition to
any move which threatened their privileged positions as middlemen. When
England eventually proceeded to abolish the trade the hostility of these
wealthy coast chiefs was at once excited, greatly to the prejudice of her
legitimate commercial interests.
Added to the hostility of the natives were the serious natural obstacles
to any attempt by Europeans to explore the interior. Almost everywhere
along the coast the dense forests of the tropical rain belt pressed hard
upon the shore. Not only did the forest present a serious impediment to
human movement but conditions of life within it were, before the discovery
of the prophylactic use of quinine, highly inimical to the well-being of the
European. (Naval Intelligence Division, 1943, 355)
Opposition to the cruelty of the Slave Trade mounted in Europe at the end of the
18th century. In 1792 Britain established Freetown in Sierra Leone as a refuge
for freed slaves. Between 1805 and 1818 Denmark, Britain, Holland and France
made the Slave Trade illegal and Portugal and Spain made it illegal south of the
equator (Oliver and Atmore, 1967, 34). Slaves freed by the French navy were
settled in Libreville, eventual capital of French Gabon. The Slave Trade did not
cease abruptly, but the last vessel to evade the Royal Navy’s blockade at Ouidah,
was the steamer Ciceron in 1863 (Naval Intelligence Handbook, 1944, 45). It
landed 1,100 slaves at Cuba.
There was a profound transformation in the relations between Africa
and Europe during the 19th century. Fundamental to the change was the
Africa 297
In 1918 the victorious powers were allocated mandates over Germany’s African
colonies. France secured the eastern parts of Kamerun and Togoland. Britain
occupied the western parts of Kamerun and Togoland and Tangayika. Belgium
was rewarded with mandates over Burundi and Rwanda adjacent to the Belgian
Congo. They had previously been part of German Tanganyika. South Africa
obtained a mandate over German Southwest Africa. By the Treaty of London
of 26 April 1915, Britain and France had assured Italy that if their territories in
Africa were augmented after the war, Italy could receive some equitable com-
pensation. In 1924 of the Italian Somaliland boundary with British Kenya was
moved about 125 km westwards of the Juba River that had previously been the
international boundary (Figure 10.4).
In 1935 Italian forces successfully invaded Ethiopia. They were driven out
in 1941 and Britain governed Eritrea as a mandate until 1952, when Eritrea
federated with Ethiopia. Ten years later Ethiopia became a unitary state and
the struggle for Eritrean independence started. It succeeded in 1993 within the
pre-1939 boundaries.
In 1960 the French mandate over the former German Cameroons ended and the
territory became independent. The next year a referendum was held in the British
area of the former German colony. It resulted in the southern part joining the
independent Cameroons and the northern part joining the independent Nigeria.
In 1969 Spain returned the coastal foothold of Ifni to Morocco. Spain then
withdrew from Western Sahara and it was partitioned between Morocco and
Mauritania in 1976.Three years later Mauritania renounced its claim to any
part of Western Sahara and Morocco claimed the whole area within the former
Spanish boundaries.
The final boundary change occurred in 1994 when South Africa generously ceded
Walvis Bay to Namibia and the international boundary of 1878 disappeared.
GEOMETRIC BOUNDARIES
Geometric boundaries consist of straight lines, arcs of circles and median lines.
Maling (1989, 542) has observed that since a chart or map is based on a pro-
jection, the nature of a straight line measured on it depends on the projection’s
mathematical properties. The shortest distance between two points on the earth’s
surface is the geodesic that passes through them.
Geodesic line. The shortest line on a mathematically derived surface, between
two points on that surface. A geodesic line on a reference spheroid is called
a geodetic line. Also termed a geodesic. (International Hydrographic Orga-
nization, 1990, 90).
300 Chapter 10
The arc of a circle is a curved line, every point of which is equidistance from
the centre of the circle. A median line is normally calculated in a water body
such as the sea, a bay, a lake or a river. It consists of a series of straight lines.
The termini of each line will be equidistant from at least one point on one side
and two points on the other side.
The boundary between Malawi and Mozambique involved the use of straight
lines, median lines and arcs of circles under the terms of various treaties dated
from 1891 to 1954 (The Geographer, 1971, 4).
From Pillar 1 at the eastern bend in the Malosa River at 15° 56’ 06.77”
S and 15° 49’ 36.74” E the boundary is carried by 27 straight segments
to Pillar 17 on the shore of Lake Nyassa at 13° 28’ 57.99” S and 34° 56’
27.01” E (The Geographer, 1971, 18-21).
1. In execution of the preliminary agreement concluded between the Gov-
ernment of the United Kingdom and the Portuguese Government by an
Exchange of Notes dated the 21 st January 1953, the frontier on Lake
Nyasa shall run due west from the point where the frontier of Mozam-
bique and Tanganyiika [11° 34’ S] meets on the shore of the Lake to
the median line of the waters of the same Lake and shall then follow the
[segments of the] median line to its point of intersection with the geo-
graphical parallel of Beacon 17 [13° 28’ 57.99” S] as described in the
Exchange of Notes of the 6th of May 1920, which shall then constitute
the boundary. (The Geographer, 1971, 4)
The two parallels mentioned total 56 km of the boundary in the lake. These two
parallels are then joined by the median line, that consists of a number of segments.
Each successive segment will follow a different bearing from its predecessor.
When the direction changes, it will do so when the point is equidistant from two
points on one side of the lake and one point on the other side of the lake. Figure
10.5 shows an example of the construction of a median line.
2. The Government of the United Kingdom shall retain sovereignty over
the islands of Chisamulo and Likoma, together with the exercise of all
rights flowing from such sovereignty, including full, unrestricted and
unconditional rights of access. The Government of the United Kingdom
shall also retain sovereignty over a belt of water two sea miles in width
surrounding each of these islands, except that where the distance between
Likoma and the mainland is less than 4 miles the waters shall be equally
divided between the two Governments. (Brownlie, 1979, 1194-5)
This provision means that except to the east of Likoma Britain is entitled to sur-
rounding waters to a distance of 2 sea miles. Presumably a sea mile is a nautical
mile. It is the length of one minute of longitude that equals 1,852 metres. An
English mile measures 1600 metres. This means that around Chisamulo Island
the area of water belonging to Malawi will be enclosed by arcs of circles, with
Africa 301
a radius of 2 nm, drawn from the most prominent points of the coastline. The
same arcs of circles will surround Likoma, except where the distance between the
coast and the Mozambique shore is less than four nautical miles. In that section
a median line will replace arcs of circles. Figure 10.6 shows the situation.
The 1926 agreement between Egypt and Italy used straight lines and arcs of
circles in the northern sector (Figure 10.7).
From the junction of Masrab el Ajram with the northern border of the Melfa
Oasis, the boundary shall be marked out as follows.
The line shall extend in an exact south-southeasterly direction (157° 30’
Greenwich east) to a point on Mount Guegab 10 kilometres north of
Manasseb Pass (Naqb al Munassib). Then it shall follow an arc of a circle
having as its centre Manasseb Pass and a radius of 10 kilometres. This
arc shall pass through Masrab Jalo and continue until its intersection with
another arc of a circle that shall have as its centre Williams Pass and a
radius of 10 kilometres. The line shall follow this second arc of a circle
to a point west-southwest ten kilometres from Williams Pass; then it shall
continue in an exact south-southeasterly direction until its intersection with
the 25th Greenwich Meridian. It shall then follow that Meridian until its
intersection with the 22nd North Parallel. (The Geographer, 1966, 5-6)
A comparison of the political and climatic maps of Africa indicates that the
majority of the continent’s geometric boundaries are located in the arid and
semi-arid regions. There appear to be two reasons for this correspondence. First,
when neighbouring states in arid regions negotiated international boundaries, they
were more concerned to secure established settlements. They were not insistent
to define precisely possibly convoluted tribal or community limits. This was
certainly true of Britain and Germany in the Kalahari Desert and France and
Spain in the Western Sahara.
The second explanation concerns the administration by colonial powers of huge
areas of desert and semi-desert terrain. When the French Colonial office drew the
administrative boundaries of Algeria, Mali, Mauritania, Niger and Chad, there
was no thought that they were disrupting the traditional circulation of groups in
the neighbourhood of these lines. The lines had not been drawn after extensive
research into the ancient political and social structures of the societies being
enclosed. From time to time the alignments of boundaries were changed. The
boundary between Algeria and Mauritania was revised on 7 June 1905 and during
the Niamey Conventions of 20 June and 26 August 1909. By a Memorandum of
18 March 1931, the Chad-Niger boundary through the Tibesti region was shifted
westwards to place the entire region in Chad (The Geographer, 1966a, 7).
When straight boundaries are drawn through areas where mixed farming, involv-
ing cattle and the production of beans, bananas, cassava, maize and sorghum,
has encouraged tribes to create small kingdoms, the use of straight boundaries
by European states, almost guarantees that the boundary will not coincide with
302 Chapter 10
tribal limits. That is exactly what happened in the borderland between southern
Uganda claimed by Britain and northern Tanganyika claimed by Germany. In
1886 Britain and Germany drew straight lines to carry their common boundary
from the Indian Ocean to the east coast of Lake Victoria. In 1890 this boundary
was continued dues west across Lake Victoria and on to the boundary with the
Congo Free State.
. . . thence from the parallel of 1 degree of south latitude with the eastern
shore of Lake Victoria, crossing the Lake on that parallel, it follows the
parallel to the frontier of the Congo Free State, where it terminates. (The
Geographer, 1965, 2)
This line was perfectly satisfactory when crossing the lake, but where it crossed
land it ignored the frontiers that existed between neighbouring kingdoms. The
frontier between the lacustrine kingdoms west of Lake Victoria coincided with
the Kagera River. The line of latitude severed two small areas of the southern
Kingdom and a large area of the northern Kingdom. In addition the line lay
across the northern tip of Ruboba Point that projected from the coastline of the
southern Kingdom.
WATER BOUNDARIES
The boundary between the Congo and the Democratic Republic of Congo
extends for about 1620 km. In the south about 500 km coincides with the water-
shed between the Congo and Nairi Rivers, and then links the watershed to the
Congo River. The remaining 1120 km of boundary coincides with the Congo
and Ubangi Rivers. Only a short section of the river boundary is defined. It is
located in Stanley Pool.
The Convention of 5 February 1885 between France and the International
Association of the Congo defined the boundary in Stanley Pool (Figure 10.8).
The median line of Stanley Pool to the contact point of that line with Bamu,
the southern shore of that island to its eastern extremity, and then the median
line of Stanley Pool.
Bamu Island, and the water and islets from Bamu Island to the north-
ern shore of Stanley pool shall belong to France; the water and islands
from Bamu island to the southern shore of Stanley Pool shall belong to
Belgium.
The territory of Bamu Island shall be permanently under a neutral regime.
No military establishment may be set up there, and it is understood that
the territory so neutralized shall also be under the regime specified in the
final provision of Article XI of the General Act of Berlin. (The Geographer,
1972, 5)
Africa 303
The Anglo-Portuguese agreement of 11 June 1891 that defined the first boundary
between the areas that subsequently became Malawi and Mozambique, followed
the eastern coastline of Lake Nyasa.
To the west [of Mozambique] by a line which, starting from the above men-
tioned frontier [German Tanganyika] on Lake Nyasa, follows the eastern
shore of the lake southwards as far as the parallel 13° 30’ south . . . (The
Geographer, 1971, 2)
The Anglo-Portuguese agreement of 18 November 1954 moved the boundary
from the eastern shoreline to the median line through the Lake.
In two other cases boundaries coincided with the right bank of major rivers.
In 1890 the Anglo-German boundary separated South Africa and Southwest
Africa.
. . . a line commencing at the mouth of the Orange River, and ascending
the north bank of that river to its intersection by the 20th degree of east
longitude. (Brownlie, 1979, 1276)
Because the flow throughout the year varied, before dams were constructed in
the upper reaches of the Orange, the boundary moved with the fall and rise of
water levels. When South Africa generously decided to cede Walvis Bay and the
Guano Islands to Namibia in 1994, it also offered to renegotiate the boundary
within the Orange River rather than along the right bank.
The boundary between Mauritania and Senegal commences on the Langue de
Barbarie, a narrow sand spit that separates the Senegal River from the Atlantic
Ocean, a few kilometres above Saint-Louis. The boundary crosses the spit and
then follows two marigots [distributaries] before reaching the right bank of the
Senegal River, that is followed for 800 km (The Geographer, 1967). The only
exception occurs near the mouth where the boundary follows the south coast of
Ile au Bois. This island has a length of about 2 km.
Stanley Pool is a natural feature in contrast with Lake Kariba that was created
when the Zambezi River was dammed in 1961. The boundary separating Zambia
and Zimbabwe, when they were British colonies, was defined in identical terms
in 1895 and 1923.
. . . thence by that [Hunter’s] Road to the River Zambesi, and by that river
to the Portuguese boundary. (The Geographer, 1964, 4)
In 1963 a British Order in Council defined the present boundary in three sections.
Proceeding from the Mozambique tri-junction the boundary followed the medium
filum [median line] to a brass stud numbered NRT/T153 set into the wall of the
Kariba Dam. The second section consists of 14 straight lines from the wall to
the vicinity of the Victoria Falls. Each straight line is defined by a true bearing
measured from due North and a distance measured in feet. The longest line has
a true bearing of 247° 11’ and measures 145,300 feet [44,030 metres]. The third
304 Chapter 10
section follows the median line from below the Victoria Falls, to the quadripoint
with Botswana and Namibia. The median line is drawn to allocate islands named
after Livingstone, six Princesses and King George VI to the appropriate country.
The boundary crosses the Falls west of Livingstone Island.
Belgium and Germany defined their territories in central Africa on 11 August
1910. They fixed their common boundary in Lake Tanganyika as a median line
measuring 456 km (The Geographer, 1965a, 5). After 1918 Britain secured a
mandate over most of German East Africa but the northwest corner of that
colony was reserved for a Belgian Mandate. It adjoined the Belgian Congo and
the Belgian-German boundary was preserved.
From Lake Tanganyika to Lake Kivu:
The boundary leaving, the median line of Lake Tanganyika, curves in
order to follow the talweg of the main western branch of the Russisi delta
as far as the northern tip of the delta.
It then takes the talweg of that river to the point where it flows out of
Lake Kivu. (The Geographer, 1965b, 4)
The surface of Lake Kivu stands 697 metres above Lake Tanganyika and the
Russisi River connects the two. This boundary definition introduces the concept
of the talweg.
Thalweg. The line of deepest soundings along the course of a river. In
international law the term is judicially construed to denote the main navi-
gable channel of a waterway which constitutes a boundary line between
two nations or states.
Fr: Thalweg; Ger: Talweg; Stromstrich (Kerchove, 1961, 829)
The talweg is a convenient site for a boundary on a river when both countries
seek access to the deepest continuous channel.
The southern section of the boundary between Kenya and Uganda traverses
Lake Victoria for 137 km. The selection of the line appears to be related to the
fortuitous arrangement of Kenyan islands. Pyramid, Ilemba, Kiringit, Mageta and
Sumba are located up to 17 km off the Kenyan mainland. However, with minor
variations, they are arranged almost along a meridian. Administrators determin-
ing the boundary between the two Colonies in 1926 decided that the boundary
would connect each of the Kenyan islands.
Commencing in the waters of Lake Victoria on parallel 1° south latitude,
at a point due south of the westernmost point of Pyramid Island; thence the
boundary follows a straight line due north to that point; thence continuing
by a straight line to the most western point of Ilemba Island; thence by a
straight line, still northerly, to the most westerly point of Kiringiti Island;
thence in a straight line, still northerly, to the most westerly point of Mageta
Island; thence by a straight line north-westerly to the most southern point
of Sumba Island; thence by the south-western and western shores of that
Africa 305
ADJUDICATED BOUNDARIES
During the colonial period there were five boundary adjudications and Britain
was involved in all of them.
In 1616 Portugal established a garrison at Cacheo, in what is now Guinea Bis-
sau, to control The Slave Trade. Britain acquired Boloma Island in 1826 about
110 km east of Cacheo. During the 1860s Portugal disputed Britain’s claim
to the Island and the two sides agreed to refer the matter to President Ulysses
Grant. The President found in favour of Portugal and Britain withdrew (Naval
Intelligence Division, 1944, 125).
Portugal was again the beneficiary of a judgement in respect of Delagoa Bay
on the coast of Mozambique. Britain had challenged Portugal’s sovereignty over
Delagoa Bay now called Baia de Laurenço Marques, because of its potential
value as an outlet for northern South Africa. The matter was referred to President
Marshal MacMahon of France who decided that Portugal had the best claim on
24 July 1875 (Brownlie, 1979, 1239).
Mr Paul Vigliani, formerly Chief President of the Court of Cassation in Flor-
ence, Minister of State and Senator of the Kingdom of Italy, was requested by
Britain and Portugal to interpret a phrase in the Anglo-Portuguese boundary
treaty of 11 June 1891.
. . . thence running eastward direct to the point where the River Mazoe is
intersected by the 33rd degree of longitude east of Greenwich; it follows
Africa 307
that degree southward to its intersection by the 18° 30’ parallel of south
latitude; thence it follows the upper part of the of the eastern slope of the
Manica Plateau southwards to the centre of the main channel of the Sabi,
follows that channel to its confluence with the Lundi . . . (The Geographer,
1971a, 3. Emphasis added.)
Vigliani delivered his judgement on 30 January 1897 (Figure 10.10). He delimited
a boundary over 310 km between latitude 18° 30’ South and the confluence of
the Sabi and Lundi Rivers. It consisted of sections coincident with the watershed
connected by straight lines.
Britain and Portugal delimited a boundary between their spheres of influence
in the areas known today as Angola and Zambia, on 11 June 1891. Article IV
of the agreement referred to the Barotse Kingdom.
It is agreed that the western line of division separating the British from the
Portuguese sphere of influence in Central Africa shall follow the centre of
the channel of the Upper Zambezi, starting from the Katima Rapids [Katima
Molili Rapids] up to the point where it reaches the territory of the Barotse
Kingdom.
That territory shall remain within the British sphere; its limits to the
westward, which will constitute the boundary between the British and Por-
tuguese spheres of influence, being decided by a Joint Anglo-Portuguese
Commission, which shall have power, in case of difference of opinion, to
appoint an Umpire. (The Geographer, 1972a, 1-2)
On 5 June 1893 the governments decided to adopt a modus vivendi until an
agreed boundary could be demarcated.
Pending the delimitation of the boundary line as laid down in Article IV of
the treaty of 11 June 1891, the line formed by the course of the Zambezi
from the cataracts at Katima up to the confluence of the Kabompo River,
and then by the course of the Kabompo, shall be the provisional boundary
between the respective spheres of influence in that region and the provisions
of Article VIII of the treaty above referred to shall be applicable to the terri-
tories separated by the said provisional boundary until a definitive boundary
shall have been substituted in its stead. (The Geographer, 1972a, 2)
No progress was made in determining the extent of the Barotse Kingdom and
Victor Emmanuel III, King of Italy, was invited to define the boundary of the
area, over which the Barotse King was Paramount Ruler on 11 June 1891.
Brownlie (1979, 1064-70) has reproduced the King’s Award in the original
French and in an English translation. The King’s starting point was to ascer-
tain the tribes that were in a position of real dependence on King Lewanika on
11 June 1891. In such circumstances the King would appoint subordinate Chiefs,
resolve disputes between them and depose those that misbehave (Brownlie,
1979, 1068).
308 Chapter 10
Whereas, such powers had beyond doubt already been exercised by the King
of the Barotse in the Province of Nalolo, to the west of the Zambezi; and they
had also been exercised over the tribes of the Mabuenyi and the Mamboe,
so that their territory formed an integral part of the Barotse Kingdom.
Whereas, as regards the Balovale, although they have paid tribute, they
were on 11 June 1890, in a state of independence, having in fact their Para-
mount Ruler, who appointed subordinate Chiefs, and the King of Barotse
had up till then performed no act of jurisdiction or government over the
Balovale. . . .
Whereas, however, King Lewanika exercised some rights of lordship over
the zone which, bordering his real dominions, lies between the Zambezi
and the Lungubungu, and is inhabited by Balovale, so that, in view of such
rights of lordship, it may be admitted that this zone formed an integral part
of the Barotse Kingdom.
Whereas, as far as the region of the Balunda is concerned, a part was
inhabited by Balekwakwa, who are ethnologically Barotse, and whereas the
southern zone had been more directly under the influence of the King of the
Barotse until its actual subjection, so that the territory comprised between
the lower course of the Kapombo [sic. Kabompo], the Zambezi and the 13th
parallel must be considered part of the Barotse Kingdom.
Whereas the Bampukush, the Bamarshi, the Mambunda, and the Bama-
koma were absolutely independent tribes, and consequently, could not be
considered as an integral part of the Barotse Kingdom. (Brownlie, 1979,
1068)
The King then delimited the boundary he had been asked to define.
The straight line between the Katima Rapids on the Zambezi, and the village
of Andara, on the Okavango, as far as the point where it meets the River
Kwando; The eastern side of the bed of the upper waters of the Kwando, as
far as the point of intersection with 22nd meridian east of Greenwich; The
22nd meridian east of Greenwich as far as the point of intersection with
the 13th parallel; The 13th parallel as far as the point of intersection with
the 24th meridian east of Greenwich; The 24th meridian east of Greenwich
as far as the frontier of the Independent State of the Congo. (Brownlie,
1979, 1068, 1070)
This admirable definition had only one weakness and predictably it caused further
delay in settling this boundary. The phrase ‘The eastern side of the bed of the
upper waters of the Kwando’ is capable of more than one interpretation. The
matter was finally settled on 18 November 1954.
And whereas difficulties were still encountered in the delimitation of the
boundary owing to a difference of opinion as regards the interpretation of
the expression “le bord oriental du lit des hautes eaux du Kwando” in the
Africa 309
In December 1986 the Court delimited a line that the two parties agreed to
demarcate. The new boundary divided the disputed zone almost equally between
the two countries between Mahou and Douna, settlements in Mali. The Judges’
decision emphasized the principle of uti possidetis juris, a principle frequently
evident in South America, during the period of boundary construction in the 19
th century. The difficulty in this case was that the countries relied on different
French boundaries! The parties provided the Court with texts of French admin-
istrative acts that created Upper Volta, now Burkina Faso, in 1919, abolished it
in 1932 and recreated it, within the same boundaries in 1947. Other texts and
many maps were also provided as evidence. The Court was unable to establish
any single description of the boundary or any map that could be relied on for
the whole sector in dispute. Accordingly the Court engaged in an exercise of
detection that involved, for example, whether the word ‘village’ included all the
farming hamlets used at certain seasons by cultivators, who normally lived in
the principal village. This was an important question because some boundary
definitions allocated villages to one side or the other.
The Court determined 13 points along the boundary and the description used
phrases such as ‘. . . in a northerly direction’, ‘. . . approximately 7.5 km’ and ‘. . . a
generally east-west direction’. The Court offered the assistance of three experts
to assist in the demarcation. The judgement was received well by both sides.
Mali National Radio reported that the result was ‘. . . a joint brilliant victory for
the two peoples in history’. The Burkina Faso official newspaper Sidwaya was
also supportive, noting that a sombre chapter in the history of relations between
Mali and Burkina Faso has thus come to an end and must not be re-opened.
The success of this judgement owed much to the low level of inter-territorial
movements of people, the lack of major refugee movements across the boundary,
and the policies of laissez-faire adopted by French administrators who adjusted
boundaries when difficulties arose, without consulting Paris.
In June 1973 Libya annexed the Aozou Strip that lay along its entire southern
boundary with Chad. The Strip had an area of about 11,000 sq. km. It is desert
with more variation in topography than in climate, vegetation or economic use.
The Tibesti Highlands are a volcanic mass in the shape of a heart, part of which
extends into the Aozou Strip. This zone called the Tarso Emissi rises to 3,376
metres and its grim, unvegetated peaks are surrounded by rock-strewn platforms,
deep gorges and precipices. Intermittent rivers called the Bardage, Yebigé, Sanaka,
Tidedi, Borou and Korossom radiate from the massif after unreliable rains in
July and August. They nourish small deltas and oases in level areas, where
dates, millet and vegetables are grown. These include Yebbi-Souma, Guezenti,
Tirenno, Aozou and Ouri. Eastwards from Rason Emissi, lies the Jef-Jef Pla-
teau. This is a barren, waterless desert with a varied appearance as blackened
sandstone yardangs rise above a sea of dunes that are interspersed with gravel
plains. Sometimes, this area was used by nomadic pastoralists raising goats and
sheep. Eastwards the land rises to the Erdi Plateau with elevations about 1,000
Africa 311
metres. This undulating surface covered with dunes and rocks offers the poorest
prospects for pastoral activity in the entire Strip.
When Libya annexed the Strip, Chad was beset by a civil war that lasted for
14 years. Libya played favourites with whichever party in Chad suited its politi-
cal interests. After President Habre had consolidated control over most of Chad
in 1987 negotiations with Libya were opened and the dispute was referred to
the International Court of Justice in 1989. Although the preparatory stages were
protracted and the judgement was not delivered until 1994 the issue before the
Court was not difficult (Figure 10.11). Chad relied on the boundary that was
defined in the Franco-Italian treaty of Friendship and Good Neighbourliness
of 10 August 1955. That agreement defined the boundary between Libya and
French territories according to various international treaties and agreements
listed in an annexe. They were the Anglo-French Convention of 14 June 1898
and the Additional Declaration of 21 March 1899, the Franco-Italian Accords of
1 November 1902, the Franco-Turkish Convention of 12 May 1910, the Anglo-
French Convention of 8 September 1919 and the Franco-Italian Arrangement
of 12 September 1919 (Brownlie, 1979, 121-3). In contrast Libya relied on the
Franco-Italian Agreement signed on 7 January 1935 by Pierre Laval and Benito
Mussolini. The agreement was not ratified by Italy, and both countries repudiated
it in 1938. It derived from the Treaty of London, devised by Britain, France,
Italy and Russia and signed on 26 April 1915. The treaty was designed to per-
suade Italy to enter the First World War on the side of the Allies and included
a provision that if Britain and France increased their territories in Africa, at the
expense of Germany, then Italy could claim some equitable compensation. The
1935 Agreement secured the Aozou Strip for Italy. The benefit for France was
that Italy abandoned its claims against eastern Tunisia.
The Court found the boundary claimed by Chad was the correct line and cited
the fact that Libya had recognized this by helping to define the eastern terminus
of the boundary after 10 August 1955. The Court also made the important find-
ing that although the Treaty of Friendship and Neighbourliness only lasted for
20 years, which had elapsed, the boundary remained valid and would only be
superseded by a new agreement.
On 13 December 1999 the International Court of Justice announced its decision
in the territorial dispute between Botswana and Namibia. Both countries claimed
sovereignty over an island in the Chobe River that forms the boundary between
them. The island is called Kasili by Namibia and Sedudu by Botswana. It has an
area of about 3.5 sq. km and is submerged during the wet season. During the dry
season the island re-appears and it is used to graze cattle and as a tourist destina-
tion. The Chobe River is famous for its subsistence and recreational fishing.
The case turned on discovering whether the main channel of the Chobe lay
north or south of the Island.
In its Judgement, the Court finds, by eleven votes to four, that ‘the boundary
between the Republic of Botswana and the Republic of Namibia follows the
312 Chapter 10
line of the deepest soundings in the northern channel of the Chobe River
around Kasili/Sedudu Island and, by eleven votes to four again, that Kasili/
Sedudu Island forms part of the Republic of Botswana’.
The Court adds unanimously that ‘in the two channels around Kasili/
Sedudu Island, the nationals of, and vessels flying the flags of, the Republic
of Botswana and the Republic of Namibia shall enjoy equal national treat-
ment’. (Anonymous, 1999-2000, 26)
On 10 October 2002 the International Court of Justice announced its decision
in the boundary dispute between Cameroon and Nigeria. The boundary between
the two countries consists of three sections. Starting in Lake Chad the Anglo-
French boundary, delimited by Thomson and Marchand, respective Governors
of Nigeria and Cameroon, extends south to Mount Kombon. That boundary was
defined in 1929-30 (Brownlie, 1979, 569-78). The second section consists of the
internal British boundary between Northern and Southern Cameroon delimited
in the Second Schedule of an Order in Council dated 2 August 1946 (Brownlie,
1979, 584). It connects Pillar 64 and Mount Kombon. The third section, leading
to the sea, is the Anglo-German boundary. The northern sector from Pillar 64
to the Cross River was demarcated by Nugent (1914) and Detzner, who started
their survey at Yola. Their report was dated 12 April 1913 (Brownlie, 1979,
561-4). The southern sector between the Cross River and the Sea was delimited
and demarcated by Grey and Lichnowsky, with the results recorded on 11 March
1913 (Brownlie, 1979, 557-60). One major part of the judgement confirmed that
the Bakassi Peninsula belonged to Cameroon. Many Nigerians had settled on
the Peninsula.
In a contribution to a discussion group, organized by the International Boundar-
ies Research Unit, Odutan has raised questions about the final determination of
the boundary and drawn attention to the difficulties that might be experienced.
Furthermore, there are still many hydra-headed and challenging demarcation
issues left in relation to the 1800 km boundary covered by the ICJ judgement.
These of course include the familiar issues of straddling villages, inaccurate
maps, ambiguous portions of court judgements, dried-up rivers, incorrectly
identified features (e.g. watershed that allegedly contain streams), and per-
haps more usual features as allegedly disappeared villages or villages that
allegedly exist in two separate places at the same time. Interesting stuff.
(Personal communication, 23 August 2006)
CONCLUSION
It has been noted that the Scramble for Africa started as early as 1600, and Britain,
France, Portugal, Spain and Turkey had established coastal footholds by 1830,
Africa 313
55 years before the General Act of Berlin was signed. However, it is the case,
that the delimitation and demarcation of boundaries between colonies proceeded
apace after 1885. Brownlie’s monumental table of 273 documents contains only
12 dated before 26 February1885.
Ethiopia, Liberia and Morocco were the only African states that engaged with
European governments in delimiting all or some of their boundaries. Through-
out the remainder of the continent the boundaries were delimited by European
powers. Several sections of the boundaries consisted of straight lines and many
coincided with rivers or watersheds. The Ethiopian lake T’ana Hayk, with an
area of 3,600 sq. km, is the only major lake not intersected by at least one inter-
national boundary. Only occasionally were boundaries deliberately designed to
avoid the partition of tribal areas. In northeast Africa, for example, boundaries
divided the extensive Somali and Masai homelands.
The colonial powers were diligent in delimiting and often demarcating agreed
boundaries and, as colonies became independent states, most had a clear under-
standing of their territorial extent. There have been some boundary disputes,
but only a few, considering that there are 102 bilateral boundaries. This situ-
ation has been greatly assisted by all members of the Organization of African
Unity, making a pledge to respect the boundaries existing on the achievement
of national independence.
In 1983 President Chadly of Algeria described Africa’s boundary problems as
‘. . . delayed action bombs left by colonialism’. That has not turned out to be an
accurate prediction except in the vicinity of Eritrea. Indeed it can be argued, that
the colonial powers acted much more responsibly in delimiting the territories of
colonies than they did in managing their progress to independence. The scramble
of European powers to divest themselves of expensive and troublesome colonies
was not well managed in the period after 1956. In 1975 the dereliction of duty
by the Portuguese authorities, in the decolonisation process in Africa and Timor,
can be judged disgraceful.
Post-colonial African history has been marked and marred by civil wars, tribal
massacres, political dictatorships and financial corruption on a grand scale. The
most recent example involves Zimbabwe. This is a country that had a strong
economy based on mineral and agricultural production when it became inde-
pendent in 1980. In March 2007 the World Bank predicted that by the end of
the year inflation would reach 4000 per cent, assisted by the Reserve Bank of
Zimbabwe printing billions of banknotes to keep the economy afloat. Fortunately
there are a few countries, such as Namibia, that have avoided the excesses that
have plagued the majority of states. On 21 March 1990, when Namibia became
independent, European territory on the continent of Africa was reduced to five
tiny coastal features. Spain, that had not managed to secure any significant part
of the continent, and that had abandoned Western Sahara in 1975, clings to five
outposts on the north coast of Morocco. They are Melilla acquired in 1497, Ceuta
in 1538, Penon de Velez de la Gomera in 1564, Islas Chaferinas in 1848 and
Alhucemas in 1928 (Naval Intelligence Division, 1942a, 111-19).
314 Chapter 10
REFERENCES
Allcock, J.B. et alia, 1992. Border and territorial disputes, 3rd edition, Longman
Current Affairs: London.
Anonymous, 1999-2000. ‘Botswana-Namibia’, Boundary and Security Bulletin,
7 (4), 26.
Brownlie, I., 1979. African boundaries: a legal and diplomatic encyclopaedia,
Hurst: London.
Fernando-Armesto, F., 1991. The Times atlas of world exploration, Times Books:
London.
International Hydrographic Office, 1990. Hydrographic Dictionary, 4th edition,
Special Publication No. 32: Monaco.
Kasule, S., 1998. The history atlas of Africa, Macmillan: New York.
Kerchove, R. de, 1961. International maritime dictionary, Van Nostrand Rein-
hold: New York.
Maling, D.H., 1989. Measurements from maps, Pergamon: Oxford.
McCaffrey, 2003. The law of international watercourses: non navigational uses,
Oxford University Press: Oxford.
Naval Intelligence Division, 1942. French Equatorial Africa and Cameroons,
HMSO: London.
———, 1942a. Morocco, Vol. II, HMSO: London.
———, 1943. French West Africa, Vol. I, HMSO: London.
———, 1944. French West Africa, Vol. II, HMSO: London.
Naval Intelligence Handbook, 1945, Tunisia, HMSO: London.
Nugent, W.V., 1914. ‘The geographical results of the Nigeria-Kamerun Boundary
Demarcation Commission 1912-13’, Geographical Journal, 43, 639-48.
Oliver, R., and Atmore, A., 1967. Africa since 1800, Cambridge University
Press: Cambridge.
Oliver, R., and Crowder, M., 1981. The Cambridge Encyclopedia of Africa,
Cambridge University Press: Cambridge.
Pakenham, T., 1991. The Scramble for Africa: White man’s conquest of the Dark
Continent from 1876-1912, Avon Books: New York.
The Geographer, 1964. ‘The Zambia-Zimbabwe Boundary’, International Bound-
ary Study, No. 30, State Department: Washington DC.
———, 1965. ‘Tanzania-Uganda Boundary’, International Boundary Study,
No. 55, State Department: Washington DC.
———, 1965a. ‘Democratic Republic of the Congo-Tanzania’, International
Boundary Study, No. 51, State Department: Washington DC.
———, 1965b. ‘Burundi-Democratic Republic of the Congo’, International
Boundary Study, No. 48, State Department: Washington DC.
———, 1966. ‘Libya-Egypt Boundary’, International Boundary Study, No. 61,
State Department, Washington DC.
Africa 315
This chapter is concerned with that part of mainland Asia that lies east of meridian
60° east. This line lies close to the boundary separating Iran from Afghanistan
and Pakistan. The evolution of political boundaries throughout this extensive
region finds a unity in the relationships between China and the colonial powers
of Russia, Britain, France and Japan.
China was the hub about which other states manoeuvred. Russia occupied
the sector from Vladivostok to Afghanistan; the British theatre extended from
Afghanistan to Thailand and included the Malay Peninsula.; France dominated
Indo-China and Japan held sway in the Korean Peninsula. Britain, France, Ger-
many and Portugal secured small, temporarily important footholds on China’s
mainland coast. Even the duration of Britain’s occupation of Hong Kong was
but a moment in the perspective of Chinese history.
Asia’s political boundaries evolved in three phases. Before 1914 the colonial
powers carved out their Asian Empires and set the limits of those Asian states
such as Afghanistan, Bhutan, China, Nepal and Thailand that were not directly
controlled. Apart from the Sino-Russian treaties of 1689 and 1727, these Asian
states invariably negotiated from a position of weakness. Indeed, with the excep-
tion of China, it seems likely the other states were not totally annexed because
of Britain’s pre-occupation to avoid boundaries with other colonial powers and
China. That British policy of self-denial did not stop Russia from acquiring large
areas of northern Afghanistan or preventing France from compelling Thailand to
disgorge extensive tracts in the Mekong Valley. Britain’s ineffective attempt to
create a buffer state of Tibet from a Chinese Province was successfully duplicated
by Russia, that detached Outer Mongolia in 1910 and subsequently prevented
any resumption of that area by China.
The second phase lasted until the end of World War II, and it was character-
ised by the preservation of the status quo. There were minor boundary altera-
tions involving Afghanistan and Thailand, but Japan’s major efforts to redraw
boundaries in Indo-China and Manchuria failed entirely. During this period
the internal boundaries of the British and French Empires were maintained.
These unilateral domestic limits provided the lines of cleavage along which the
Empires split in the third phase. The third stage was characterised by a desire
for independence on the part of the indigenous population, and an anxiety on
the part of Britain and later France to escape from colonial responsibility. The
318 Chapter 11
With one exception, the international boundaries in this region were either estab-
lished before 1917, or were foreshadowed by arrangements reached before that
Asia 319
date (Figure 11.1). The solitary exception is provided by the sector of boundary
between Russia and Mongolia from Tengis Gol to Huyten Orgil. Russia obtained
a salient, along a front of 440 km for a depth of 120 km, between Uvs Nuur
[Lake] and Hovsgul Nuur. Today this region is Respublica Tyva in the Russian
Federation.
The Sino Russian line was defined in three sectors. In each sector the nego-
tiations were prompted by contact or the threat of contact between the Russian
frontier advancing eastwards and southeastwards, coming into contact with Chi-
nese and associated peoples subject to the Emperor. Unbeknown to the Chinese
authorities the negotiations resulting in the Treaty of Nerchinsk in 1689 was the
highwater mark of Chinese ambitions. The Russian exchequer was depleted, it
was desperate for trade with China and it wanted a quick settlement. The sub-
sequent treaties of 1727-68, 1858-1860 and 1864-1915 chronicled a succession
of Chinese retreats.
The Russian principal interest was trade, while the Chinese wanted a treaty
that would prevent Russian interference in border communities with a propen-
sity for rebellion. The boundary was fixed along the watershed between the
rivers flowing south to the Amur and rivers flowing north to the Arctic and the
northeast coast of Asia. The eastern terminus was between the Rivers Uda and
Amur that reached the coast 450 km apart west of the north entrance to Tatarskiy
Proliv [Strait], that separates Sakhalin from the mainland. Most seriously China
unwisely agreed that the territory between the watershed and the River Amur
would remain neutral pending the final delimitation.
In the period 1727-1768 China and Russia negotiated a boundary 2719 km long
between Shabina Dabaga [Pass], that overlooked the Abakan River, to Abagaytu
the origin of the 1689 boundary on the Argun River. The boundary consisted
of two sections east and west of Kayakhta, lying south of Ozero [Lake] Baykal.
Once again the same national motives prevailed. China was prepared to offer
trading opportunities providing Russian forces and traders avoided contact with
China’s nomadic tribes. The boundary was surveyed and partially demarcated.
The 1673 km of boundary westwards from Khyakhta to the Shabina Dabaga
was defined by twenty-three named features, mainly passes and peaks. The
instructions were simple.
They will adhere to the tops of those mountain chains which will be divided
by the middle and be considered as the frontier. If any mountain chains
cross between them and rivers adjoin, the mountain chains and the rivers
will be cut in two and divided equally. (Prescott 1975, 20)
In this sector there was no attempt to draw a boundary between existing areas of
authority over indigenous peoples, probably because the people were nomadic
and because neither China nor Russia had exercised any significant degree of
authority. Mancall, reports the view of the Russian delegation.
320 Chapter 11
. . . much land was delimited which had never before been in Russian pos-
session, namely; from the Khan Tengeri River a distance of approximately
eight days horseback ride in length and in width three days to the Akaban
River, and these places have never been under the domination of the Rus-
sian Empire. (Mancall 1971, 301)
The ability of the commissioners demarcating the eastern sector to finish their
work two weeks before the western team, is explained by the facts that the
boundary was 600 km shorter, the topography was easier to negotiate and a
well-marked Chinese boundary existed.
In 1855 The Chinese authorities became aware of Russia’s expectations in
respect of the Amur River in the contents of a memorandum provided by Gov-
ernor-General Muraviev, who governed Eastern Siberia.
In 1689 the two Empires entered into the Treaty of Nerchinsk which stipu-
lated that the maritime region by the Eastern Sea should be neutral territory.
The Eastern Sea is within the domains of Russia. The Hielungkiang and
the Sungari, which are known as the Amur in Russian, have their sources
from within Russian territorial limits. Up to the present, the Region from
the Hielungkiang to the Eastern Sea has remained as yet undelimited. The
Amur constitutes an important area of defence against foreign aggression.
Moreover, since the summer of this year [1855] Russian troops have been
stationed at the mouth of the Sungari and have erected fortresses and
established stations on both banks of the river. Throughout the summer
ships have plied on its water. In the winter they will continue to traverse
the ice on horseback.
Since the Heilungkiang has its source from within Russian territory, it
would be conducive to peace to recognize Russia’s title to the left bank of
that river. Up to the present, however, the region covered by the Heilung-
kiang, the Udi and both banks of the mouth of the Sungari has remained to
be delimited. Once the left banks to the Heilungkiang and Sungari should be
ceded to Russia, it would be up to China to decide whether or not to remove
the Orochon, Heche, Fiyakha, and other tribesmen inhabiting thereon. As
to the Zeya, the Silimji and the Niman, although they are within China’s
domains, these rivers should also be ceded to Russia, since the country by
the mouth of the Sungari is muddy and difficult to journey by land, in sum-
mer as well as winter. (Yi Wu Shih Mo, 1930, 12, 18b-19b)
The weaknesses of the Russian case are evident. The Sungari River [Songhua
Jiang] is a south bank tributary of the River Amur and is within China. It was
well known that some tribes living on the north bank of the Amur owed tribute
to China. When noting that the Amur constitutes an important area of defence
against foreign aggression it is implicit that this includes defence of China against
Russian threats. One of the confusions in this region is that the Russians called
the Amur Heilungkiang above the confluence with the Sungari, and the Sungari
Asia 321
below the confluence! In the 1689 negotiations China had made it clear that the
Russians could ascend the Sungari [Songhua Jiang] as far as the important city
of Tsitsikar [Qiqihar].
Alas. The Russian arguments might have been weak, but the state of the Chi-
nese government was weaker. Copies of the Treaty were made and signed on 6
May 1858. The treaty was ratified in March 1859. Then the Chinese immediately
received a request to delimit the boundary with Russia west of Shabin Dabaga.
The second Article of the additional Treaty of Peking dated 2 November
1860 defined the Sino-Russian boundary from Shabina Dabaga to the territory
of Kokand, a straight-line distance of about 2000 km. The boundary commission
met in 1864 and quickly produced the Protocol of Chuguchak. Two consider-
ations encouraged the speedy agreement. First a Moslem revolt developed at
Kucha in June 1864. While there had been twelve previous uprisings in China’s
administration of Singkiang none had spread so rapidly (Hsu, 1965, 22). By the
end of July, Kucha, Manass, Yarkand, Yangihissar and Kashgar were controlled
by rebels, and Urumqi and Kuldja were besieged. Second in May 1864 Russian
forces had launched a pincer attack against Kokand, from Kzyl-Orda and Alma
Ata. By 22 September, these forces had linked up after capturing Dzambul,
Turkestan and Chimkent. China had a vested interest in securing a firm claim to
territory over which it had previously exercised authority, before an independent
Moslem state emerged to sign treaties with Russia.
The second Article of the Treaty of Peking was sparing in its description of
the boundary from Shabina Dabaga to Kokand.
Article 2
The border line to the west which has previously been undefined, must
henceforth be brought into being following the direction of mountains, the
flow of significant rivers and the line of recently established pickets, from
the last beacon called Shabindabaga, set up in 1728 at the conclusion of the
Treaty of Kiakhta, southeast to Lake Tsaysan and thence to the mountains
overlooking the more southerly Lake Isskul and the so-called Tegeni-Shan
or Khirgez Alatai, otherwise known as the Tien’ Shan’-nan-la and along
these mountains to the Kokand territories. (Prescott, 1975, 55)
Despite the rather brief description the Commissioners were instructed in the
principles that should be observed in selecting a particular line. The bound-
ary must be related to the direction of mountains, the flow of large rivers and
the line of recently established Chinese pickets. The boundary had to follow a
north-northeast alignment across a landscape where the majority of mountains
ranges and rivers lay nearly due east-west. The resulting boundary zig-zagged
with east-west segments coincident with ranges such as the Altai, Tarbagatay,
Alatau and Tien Shan, connected by north-south segments across plains following
Chinese markers. Rivers were rarely used. However, the boundary did follow
the Black Irtysh, near lake Zaysan and short sections of the Daratu and Naryn
Nalga tributaries of the Tekes River.
322 Chapter 11
The Protocol of Novyi Margelan dated 22 May 1884 carried the Sino-Russian
boundary as far as the pass called Uz Bel. That might have been the terminus
in 1884 but in fact the boundary continues south for another 307 km to the
trijunction with Afghanistan. This sector does not seem to have been defined
in any treaty.
China and Russia began negotiations to define their boundary, measuring
4,300 km, more precisely in the 1950s, and in 1991 and 1994 the eastern and
western sections were delimited. The remaining undelimited section involved the
territory in the vicinity of the Amur and Ussuri confluence. The treaty of the 2
November 1860 was imprecise about ownership of the islands at this junction.
On 14 October 2004 the two countries signed a supplementary agreement dealing
with this section that divided the islands at the confluence.
in the 17 th century, but its authority was nominal until the end of 19 th century.
Friters (1951, 156) notes that apart from quelling occasional rebellions China
seemed content to divide the area into regions within which Chinese officials
carried out very limited functions. Colonization of the region by Chinese was
forbidden, officers were compelled to leave their families in China, and no inter-
marriage was allowed. There were only a few trade routes into Outer Mongolia
and Chinese traders were forbidden to give credit to Mongolians.
These policies were reversed at the beginning of the 20 th century as Japan
and Russia exerted pressure against China in Korea and Outer Mongolia. In
November 1911 Outer Mongolia was declared independent as disorder spread
throughout China. China re-occupied Outer Mongolia in 1919, but independence,
with Russian assistance, was proclaimed on 13 March 1921. Three years later,
with the assistance of Soviet troops, Mongolia became the Mongolian People’s
Republic. Japanese incursions into eastern Inner and Outer Mongolia in the 1930s,
were reversed by the Soviet army in 1940-42. On 14 August 1945 an exchange
of notes between China and the Soviet Union agreed that Outer Mongolia should
be independent if that was confirmed by a plebiscite of the population. It was
held on 20 October 1945 and China recognized Outer Mongolia’s independence
on 5 January 1946 (Friters, 1951, 210-15).
On 26 December 1962 the boundary treaty was signed and the line delim-
ited in twenty-six sections. The treaty noted that this boundary had never been
delimited previously, and denied any recognition of earlier Japanese-Mongolian
agreements. The boundary was demarcated in 18 months and the final boundary
protocol was signed on 30 June 1964. The longest segment is 496 km and the
shortest is 43 km. The distances between pillars vary widely. In the Altai Range
two pillars are 129 km apart. In other sections pillars are only a few metres apart.
Each point is described by its immediate locality, distance and bearing from the
previous pillar and bearings from prominent features nearby. Cultural features
are rarely used, the only exceptions include roads, the Sino-Mongolian railway
and two animal enclosures. This boundary is now clearly established and various
provisions in the protocol ensure that markers are maintained and guidelines are
provided if there are any questions of interpretation.
delimited from the Hari Rud in the west to Lake Zorkul in the east by 1893.
The short eastern segment from the Lake to the Chinese tri-junction was settled
in March 1895.
In the final analysis British authorities were able to persuade the Emir of
Afghanistan that the boundary was appropriate. In fact it was plain that Russia had
succeeded in driving the boundary southwards and eastwards securing important
extensions along the valleys of the Hari Rud, Murghab and Amudar’ya Rivers
that provided the main lines of communication.
Despite the early fear of Yates (1888, 179) that the arbitrarily defined bound-
ary ‘. . . cannot be expected to be permanent . . . ’, this boundary has lasted without
creating serious problems between Afghanistan and the Russian Republic. This
happy result may follow from the supposition that each side secured its main
aims in the final line. Britain secured a practical boundary for the Emir that
linked the Zulfikar Pass, Maruchak on the Murghab River and Adkhavoy, and
that excluded the Turkoman tribes that might have given trouble on their own
account or provided a cause for Russian interference in the future. For their part
the Russians had secured well-watered territory south of the desert that gave a
forward line of defence. The Russians also acquired dominion over the Turkoman
tribes and were able to plan comprehensive policies to end the tribal fighting that
had made this an unstable area.
Although the final position of the line was argued in great detail Yate put the
matter into its proper context.
The fact that by this last settlement [St Petersburg, 1887] the Russian
frontier has been advanced 10 or 15 miles [16-24 km] nearer Herat, as I
have seen mentioned in the newspapers, does not appear to me to be worth
discussion. Once the old frontier from Sher Tepe to Sari Yazi proposed by
Sir Peter Lumsden, was given up and Pul-i-Khatun and Pandjeh, the only
two points of any strategical importance, were surrendered to Russia, the
question of 10 miles [16 km] here or there on the sterile down of Badghis
became of little moment. (Yates, 1888, 382)
The boundary between Afghanistan and Pakistan extends for 2430 km from the
peaks of the Hindu Kush to the baked deserts of Balochistan 450 km from
the coast. When the boundary was drawn the borderland had two pronounced
qualities. First there were limited opportunities for economic development by
the indigenous communities who lacked capital and technical skills. Soils are
thin and the narrow valleys restricted opportunities for irrigation. Second the
distribution of indigenous social and political communities produced a complex
pattern. This borderland straddled the main historical invasion routes from the arid
west towards the more fertile lowlands of the Indus and the Vale of Peshawar.
Davies (1932, chapter 4) provided a graphic account of the ethnic divisions in the
borderland that were underlain by the nomadic movements, while the boundary
was being demarcated.
Asia 325
By the middle of the 19th century British authority, searching for a western
boundary, had reached beyond the Indus River. Fraser-Tytler expressed the Brit-
ish dilemma neatly.
In both cases the fundamental, underlying cause was the juxtaposition
of stability and instability, of ordered government and misrule: the Empire
pushing on in search of a frontier and finding no halting place, no physical
or man-made barrier on which its outposts could be aligned and behind
which its nationals could move in freedom and safety. (Fraser-Tytler, 1967,
122)
The British sought friendly relations with Afghanistan, without military entangle-
ments, and a stable boundary behind which British subjects would be secure.
But there was no such line and from time to time British forces felt obliged to
take action against hill tribes beyond the self-imposed boundary. Then when the
British forces retired, sooner or later cross-border raids would follow.
There were difficulties about the alignment of any boundary between Afghani-
stan and British India. There was no single dominant physical feature that might
have been followed for better or worse. In the serried crests of the Toba and
Kakar ranges there was no reason to choose one watershed rather than another.
A few convenient rivers flowed north-south, but these rivers made poor bound-
aries since they offered no barrier during the dry season, and were settled by
homogeneous groups on both banks when they were in flood. In any case the
tribes did not recognize fixed boundaries. At any time the political boundaries
were simply a reflection of the strength and determination of the group and its
ability to defend a particular area.
British policy was not consistent for long periods, a characteristic blamed by
Davies on domestic party-politics.
The truth is that the baneful effect of party politics in this country [Britain]
has prevented the adoption of any consistent or settled frontier policy. With
shame be it confessed India has been the sport of English political factions.
In a country where more than anything else, continuity and firmness are
essential, on an Asiatic frontier where vacillation spells loss of prestige, our
administration has been marked by sudden advances and ill-timed retreats.
(Davies, 1932, 182)
At various times strong arguments were advanced for different boundaries. The
proponents of the scientific frontier wanted a line from Kandahar to Kabul.
Enthusiasts for a physical boundary sought a retreat to the Indus River. Sande-
man, who did excellent work in Baluchistan, urged a boundary through the Zhob
valley. Still others preferred the line of administered districts, under a policy of
masterly inactivity!
The limit of the administered districts lay west of the Indus River and followed
its alignment. Commencing at the mouth of the Hab River, west of Karachi, the
326 Chapter 11
boundary was never more than 105 km from the Indus as it passed just west of
the towns of Mehar, Jacobabad, Taunsa, Bannu. Kohat and Peshawar.
The opportunity to settle a definite boundary occurred in 1893 when Sir
Mortimer Durand led a British delegation to Kabul. Its aim was to persuade the
Emir of Afghanistan to give up his claims to the trans-Oxus area of Roshan and
Shignan, claimed by Russia under the 1872-3 Anglo-Russian agreement. In return
the Emir would receive the Wakhan Strip that would separate Russian and British
territory. The Emir agreed and then, on 12 November 1893, Durand persuaded
the Emir to accept a boundary between Afghanistan and British India. The agree-
ment consisted of seven short articles and a small-scale map and was far from
a model boundary agreement! The line was demarcated by British and Afghan
surveyors, between April 1894 and May 1896. That might seem a reasonable
rate of progress in view of the nature of the country and its climate, but all the
members of the demarcation teams, who published comments, refer to the fact
that the work would have been completed much sooner had it not been for the
nature of the terrain and adverse weather, and the map attached to the treaty.
The map on which the line was marked, and to which they had ‘. . . to adhere
with the utmost exactness’, was on such a small scale and contained so many
topographical errors that the surveyors were continually forced to make inter-
pretations on important points. They might reasonably have expected that their
responsibility was to transfer the line on the map onto the ground.
The map was deficient because no surveyor had been sent with Durand. This
omission had been made to allay any fears that the Emir might have that such an
officer was spying out the land between the boundary and Kabul. The lack of a
competent officer meant that a map was hastily patched together to illustrate the
alignment of the boundary. Seven sections of boundary were precisely demar-
cated by the surveyors, who had a facility for reaching reasonable compromises
(Figure 11.3).
King, who surveyed from Kwaja Khidr to Domandi did not mince words.
It may be noted that as regards this part of the boundary the map is hope-
lessly wrong. The line as shown on the map takes a turn to the west at a
distance of about three miles from the Khand Kotal (about 32° 14’ north)
and crosses over to another range to the west of the Spera, which is repre-
sented to contain the Nazan Kotal. As a matter of fact, however, the Spera
is continued without a break to the Nazan Kotal and the boundary has been
drawn accordingly in a straight line along the crest of this range to within
four miles of the Nazan. (King, 1895, 3)
The boundary in the vicinity of the Khyber Pass between Nawa Pass and
Sikaram Peak remained unmarked until 1919. The section north of Charkhao
Pass to the Chinese trijunction, along the Hindu Kush was considered to need
no demarcation.
Hayat Khan (2000) provides some useful information about the Durand Line
put into the context of the period between the delimitation of the Afghan-Russian
Asia 327
boundary, and the Pushtunistan Question when the sub-continent achieved inde-
pendence. The volume would have benefited from the inclusion of more and
better-drawn maps. The best, entitled ‘Demarcation of a buffer state’ appears to
be based on a figure by Lamb (1968, 87).
At the end of these labours two sections of the line remained undemarcated.
The boundary in the neighbourhood of of the Khyber Pass, between Nawa Pass
and Sikaram Peak remained unmarked until 1919. The section north of the
Charkao Pass to the Chinese tri-junction, along the Hindu Kush is satisfactorily
defined by nature.
Anglo-Nepalese contact was first established in the 1790s after China had rebuffed
Nepalese northward advances, and the Nepalese turned their attention east, west
and south. British interests were mainly commercial and a treaty was signed in
1792. At that time the Nepalese Kingdom extended from Kangra in the west to
the River Tista in the east, a distance of about 1415 km. In 1801 a further treaty
was agreed to control bandits along the frontier (Aitchison, 1909, volume 2, 92).
Aggressive southward advances by Nepalese in 1812 led to a British ultimatum
in 1814 and a brief war ended with the Peace Treaty of Segowlee on 2 Decem-
ber 1815. The rivers Kali and Mechi marked the west and east boundaries and
reduced the east-west extent of Nepal to 880 km. Britain also secured a forested
tract called the Terai, an unhealthy zone where malaria was endemic. In the
following year Nepalese officials questioned the interpretation of the phrase the
lowlands of the Terai. Did it mean all the Terai or only the marshy grassland
section? Britain reconsidered the situation and realised that the area had little
commercial value, and that by restoring it to Nepal Britain would be saved pay-
ing pensions of 200,000 rupees to displaced chiefs!
As a result of Nepalese forces assisting the British army during the Indian
Mutiny Britain returned the Terai located between the Kali and Rapti Rivers
and a small triangular area between the Rapti River and the British territory of
Gorakhpur. The new boundary was surveyed according to the boundary treaty
of 1 November 1860. An undelimited section of boundary, 80 km long, was
completed in January 1875, when the eastern terminus of the 1860 line was
connected to the western terminus of the 1816 line.
Shrestha (2003, 209-15) noted that although the final demarcation of the
boundary between India and Nepal started in 1980 about 3 per cent is incomplete.
Most of this area involves the area of 372 sq. km. called Kalapani-Limpiyadhura
where the territories of China, India and Nepal meet.
Throughout the period from 1864 to 1894 the British authorities in India were
greatly assisted by Pundits.
Within the last thirty or thirty-five years the term has acquired in India a
peculiar application to the natives trained in the use of instruments, who
have been employed across the British India frontier in surveying regions
inaccessible to Europeans. This application originated in the fact that two of
328 Chapter 11
Stewart’s otherwise excellent book has only one elementary defect. The three
maps do not have a scale, nor indications of latitude and longitude and the names
used are those current in the nineteenth-century. This makes it difficult to match
the 19th century names to modern names. As noted above, today Yarkand is
called So-ch’e. The lake Lob Nur [Lop Nur] is shown as lying within Mongolia
when it lies 500 km south of the Mongolian boundary. Each map does have a
totally unnecessary elaborately decorated north point!
Nain Singh was awarded a pension, granted a village, made a Companion of
the Indian Empire and received the Patron’s medal of the Royal geographical
Society.
The boundary between Sikkim and China is the oldest of the Himalayan
international boundaries. Britain and China agreed on the alignment of this line
in 1890 and it has survived to the present. It follows the watershed of the Tista
River from Bhutan in the east to Nepal in the west. The general elevation of
the watershed varies from 5185 m to 7625 m, and it is intersected by 14 passes.
Lamb (1960) has provided a detailed account of British contacts with Tibet after
Sikkim became a Protectorate. The Anglo-Chinese Convention on 17 March
1890 defined the boundary.
. . . the crest of the mountain range separating the waters flowing into the
Sikkim Teetsta and its affluents from the waters flowing into the Tibetan
Mochu and northwards into other Rivers of Tibet. The line commences at
Mount Gipmochi on the Bhutan frontier, and follows the above-mentioned
water-parting to the point where it meets Nipal territory. (Prescott, 1975,
263-4)
This line has existed for 117 years and it continues the general alignment of the
1961 Sino-Nepalese boundary alignment.
Shrestha (2006) has provided a useful account of the delimitation and demar-
cation of the boundary between China and Nepal. It measures about 1111 km,
between the Zanskar Range in the west to Janak Himal Range in the east. Contacts
began in the 7th century and although relations were not always friendly a series
of treaties from the Khasa Treaty of September 1775 to the Treaty of Thapathali
in March 1856 established a customary line that was eventually translated into a
clear boundary. By a treaty of 5 October 1961 it was agreed that the traditional
boundary would be demarcated.
In connection to execute the treaty, boundary delimitation was made on the
basis of existing traditional customary line with the technical principles of
determining the water-parting line to connect snow-capped high altitude
mountain peaks, passing and crossing through mountain passes and spurs,
saddles and cols, rivers and rivulets, pastureland and river basin or valleys.
The most important aspect adopted was to maintain certain adjustments in
accordance with the principles of mutual accommodation on trans-frontier
330 Chapter 11
Thailand’s boundary with Laos and Cambodia was constructed through negotia-
tions with the French from 1867-1925. Apart from a short section crossing the
valley west of Boeng Tonle Sab the boundary is coincident with watershed and
332 Chapter 11
rivers. The Cardamones and Dangrek Ranges carry the boundary from the sea to
the confluence of the Mun and Mekong Rivers. France secured its foothold at the
mouth of the Mekong when Annam ceded the Provinces of Bein Hoa, Gia Dinh
and My Tho in 1862. France considered it had inherited Annam’s rights respect-
ing Cambodia. By July 1867 France and Thailand had resolved the situation by
Thailand recognizing France’s protection of Cambodia and France recognizing
the provinces of Batdambang and Siemreab as part of Thailand.
In 1884 France’s ambitions shifted northwards and by 1893 France had
advanced to the Mekong River. A quarrel was forced upon Thailand and France
secured a boundary along the Mekong, north of 13° 14’ north, by a treaty dated
3 October 1893. A French ultimatum was issued in April 1893 and the British
Ambassador in Paris described the French diplomatic technique.
The Siamese Government were now in possession of an ultimatum, a pen-
ultimatum and an ante-penultimatum. In fact the word ‘ultimatum’ had
completely lost its meaning, for each new one seemed to procreate a suc-
cessor. (Prescott, 1975, 432)
By treaties dated 13 February and 20 June France secured land west of the
Mekong dominated by Louangphrabang and the area south of the Dangrek Ranges
opposite Stoeng Treng on the Mekong River. Small areas were also secured at
the northern end of the boundary, called Western Kop and Dan Sai, and at the
southern terminus on the Gulf of Thailand in the vicinity of Trat. By a treaty
dated 23 March 1907 Thailand ceded large areas south of the Dangrek Ranges
and west of the Cardamone uplands. In return Thailand regained the two small
areas of Trat and Dan Sai.
China and France settled the northern boundaries of Vietnam and Laos in
the decade 1885-95. The Sino-Vietnam boundary measures 1287 km while the
boundary of Laos and China is 418 km long. The boundaries traverse a border-
land that consists of a number of ranges that are prolongations of the Yunnan
Plateau. China negotiated from a position of strength and these boundaries were
never regarded as unequal by China. China and France concluded a preliminary
Convention of Peace on 11 May 1884 and 13 months later settled a Treaty of
Peace, Friendship and Commerce on 9 June 1885. This Treaty dealt with the
maintenance of order, identification of the boundary, cross-boundary trade and the
construction of railways. On 26 June 1887 the two countries signed a Convention
that dealt with the boundary as far as the Black River. The Gulf of Tonkin was
divided by the meridian 105° 43’ East of Paris [108° 3’ East of Greenwich], that
passed through the eastern point of the Vietnamese island Tra-co. This provision
was involved in the maritime boundary delimitation through the Gulf agreed on
25 December 2000.
When France secured the Mekong River as the boundary with Thailand it
became necessary to extend the boundary with China west of the Black River. A
Supplementary Convention dated 20 June 1895 overlapped the previous bound-
Asia 333
ary between the Red and Black Rivers and extended the line to the confluence
of the Mekong and Nam-la Rivers.
The boundary between China and North Korea extends for 1416 km and
apart from 32 km it coincides with the Yalu and Tumen Rivers. These rivers
flow respectively west and east from the watershed formed by Pai-t’ou Shan
and Namp’ot’aesan, and form an obvious divide between the continent and the
peninsula. Hulbert (1962, chapters 3 and 4) reports that some of the earliest
political boundaries of Korea coincided with these rivers. Apart from their estu-
aries the rivers are bounded by steep cliffs that lead to sharp crests or plateaus
at about 915-1525 m, with some summits reaching 2135 m. The tributaries have
also carved deep valleys into the crystalline granites and gneisses, so that the
landscape from the air resembles gale-swept seas. Coniferous forests of larch,
spruce and pine cover much of the terrain; there are only limited opportunities
for agriculture on the narrow valley floors and the wider estuaries.
All the treaties governing this boundary were concluded by China and Japan.
In July 1894 Japan forced a war on China seeking to gain ascendancy in the
affairs of Korea. The events leading to the war and the conduct of subsequent
campaigns, have been carefully described by Hulbert (1962), Conroy (1960) and
Kim and Kim (1967). Japan was victorious and by the Treaty of Shimonoseki
China recognized Korea’s independence and ceded to Japan the southern area
of the Liaotung Peninsula occupied by the Japanese army. The southern bound-
ary of the ceded area was the northern boundary of Korea defined as the Yalu
River to its tributary the An-ping, south of Fenguang. This occupied region
was retroceded to China in November 1895 when Japan had been given a clear
indication of disapproval by the major powers. On 4 September 1909 China and
Japan reached an agreement dealing with the Yen-Chi [Chien Tao] District and
the Tumen River.
The Governments of Japan and China declare that the river Tumen is
recognized as forming the boundary between China and Korea and that in
the region of the source of that river the boundary line shall start from the
boundary monument and thence follow the course of the stream. Shihyishwei
(Prescott, 1975, 505-6)
While France fought to maintain its Empire the Atlee Government of Britain
decided to divest itself of the former Jewel in the British Crown. Unfortunately
it turned out to be government by panic. It was announced Britain would grant
independence to India by June 1948. Threats of communal violence between
Hindus and Muslims resulted in Lord Mountbatten persuading the Prime Minister
334 Chapter 11
The boundary through the Punjab was delimited in 24 days by the Commission
and Lord Radcliffe, in accordance with the following instructions.
. . . demarcate the boundary of the two parts on the basis of ascertaining
the contiguous areas of Muslim and non-Muslim. In doing so it will take
account of other factors. (Whyte, 2004, 64)
The Indian case rested on allocating large administrative units according to the
majority of Muslims or non-Muslims. The Pakistan case was based on a line
separating the smallest administrative units, called tahsils, according to the ethnic
majority. The difficulty arose in drawing the boundary in the vicinity of the
Bari Doab lying between the Sutlej and Ravi Rivers. Finally four tahsils called
Ferozepore, Zira, Nakodar and Jullundur, with significant Muslim populations,
that lay east of the Sutlej River were awarded to India. This was justified by
the view, that if they had they been awarded to Pakistan, the Indian rail system
would have been seriously disrupted.
The boundary north of the Punjab separates Kashmir between India and
Pakistan. Perhaps unwisely the British allowed the rulers of Princely States to
join with either India or Pakistan. The Hindu Maharajah of Jammu and Kashmir
opted for union with India even though nearly 80 per cent of his population
was Muslim. Gupta (1966) and Lamb (1966) have provided full accounts of the
historical and political backgrounds to the dispute.
On 16 December 1971 Bangladesh became independent and inherited Pakistan’s
boundary with India. This boundary was defined by the Radcliffe Commission
for Bengal. The Indian and Pakistan members of the Commission had reversed
the arguments used by their counterparts in the Punjab. The Indian view was
that the division should be based on the smallest administrative units called tha-
nas. The Pakistan position was that the religious majorities should be calculated
in the largest administrative areas called districts. In the end Radcliffe had to
make a judgement. The final boundary coincided with 1302 km of the Indian
proposal and 116 km of the Pakistan proposal. There were only three sections
of boundary that did not coincide with existing boundaries of thanas or districts.
The northern extension of Bangladesh was curtailed to ensure that India had a
corridor to a cul-de-sac formed by the Provinces of Assam, Arunchal Pradesh,
Nagaland, Tripura, Manipur and Mizoram adjacent to Burma and China.
Whyte has established that the most complex pattern of enclaves exists in the
Bangladesh-India borderland.
The enclaves themselves number 198 in total. 106 enclaves of India in
Bangladesh and 92 of Bangladesh in India. These totals include 3 Indian and
21 Bangladeshi counter enclaves inside the enclaves of the other country, and
one Indian counter-counter-enclave inside a Bangladeshi counter-enclave.
West Bengal and Assam possess a further four enclaves between them, the
final survivors of 61 enclaves of Cooch Behar and British India that became
internal Indian enclaves after 1949. Indian enclaves total nearly 70 sq. km,
336 Chapter 11
In 1891 the Governor of Cochin China published a map showing the bound-
ary south of Giang Than following the Mandarin’s Way, a more direct line to
Hon Ta. This boundary transferred 21 sq. km of marshland to Cochin China.
The area between the former boundary and new boundary was used by about
fifty people to produce 111 hectares of rice. These farmers were mainly wealthy
Annamites, living in Cochin China who had to pay taxes to Cambodia. The
Governor justified this cartographic aggression by asserting that the boundary
had been mistakenly placed along the river road. In fact the telegraph line had
been along Mandarin’s Way, but it had been moved to the Giang Thanh to allow
easier defence against interference. The Governor-General instituted a commis-
sion of enquiry and it was found that the telegraph line had always been along
the river bank since 1870-1. To have the last word the Governor, in 1914, used
the loss of the land between the Mandarin’s Way and the telegraph line to claim
extension of the boundary around the outer fortifications of Ha Tian. However,
this extension to Cochin China was offset by transferring the Kompong Tasang
Pedicle to Cambodia (Prescott, Collier and Prescott, 1977, 66-7).
The northern section of this boundary was fixed in the period 1893-1929. This
final section extended from the headwaters of the Dam River to the Cambodian,
Laos tri-junction. At first part of this borderland was placed within Laos but
plainly access from Vietnam was much easier. There was no obvious physical
feature to carry the boundary northwards and finally the line followed an arbitrary
course across the dissected, forested landscape.
The boundary between North and South Korea is a military demarcation line. It
stretches for 241 km across the Peninsula trending northeast-southwest across
parallel 38º North, that is intersected close to the west coast. Except in the west
the boundary traverses rough mountain country of the Taihaku Sammyaku, with
isolated peaks to 1700 m. A severe winter lasts four months and the natural veg-
etation is coniferous forest of larch and spruce with some deciduous oaks and
maples. Cleared slopes often become eroded and remain barren and cultivation
is usually confined to the floors of the valleys. The western borderland possesses
low foothills and alluvial plains, that together with the shorter winter and hot to
warm summer, provide better opportunities for cultivation. Population densities
are higher in the west than in the east.
The boundary resulted from the arrangements made at the end of World War
II and then from the armistice at the end of the Korean War. The Soviet Union
declared war on Japan on 8 August 1945 a week before the end of World War
II. Russian troops invaded Korea at Unggi and Najin, close to Vladivostok on
12 August. Then there was a landing at Wonsan. near parallel 39º North and it
338 Chapter 11
was clear that the Russians must play some role in accepting the Japanese sur-
render. It was agreed that the Russians and Americans would accept the Japanese
surrender north and south of parallel 38º North respectively.
McCune (1949) and Grey (1951) have described the selection of this line and
both agreed that it was chosen in haste and was not considered to be a permanent
division. However, as relations between the two super-powers deteriorated the
parallel became an inflexible international boundary.
McCune observes that parallel 38º North had an earlier significance in Korean
history. In 1896 Marshal Yamagata, the Japanese representative in Moscow,
proposed that the Peninsula should be divided between Russia and Japan by
parallel 38º North. In 1903 the Russian Government suggested that parallel
39º North should mark the southern limit of a neutral zone drawn across the
Peninsula. In 1904, just before the onset of the Russo-Japanese War, Russian
commanders were instructed to meet with force any Japanese advance north of
parallel 38º North.
At the end of the Korean War an Armistice Agreement, dated 27 July 1953,
selected parallel 38º North as the boundary. The boundary lay through a demili-
tarised zone 4 km wide and it was marked by 1292 concrete pillars. The waters
of the Han Estuary, that extend for 61 km west of the final pillar, are open to
the vessels of both countries. Offshore islands were allocated and South Korea
secured the Paengnyong-do close to Chansang Gol, a North Korean headland
just north of parallel 38º North.
The Union of Soviet Socialist republics legally terminated when the Union
Republics of Belarus, the Russian Federation and the Ukraine signed an
agreement in Minsk on December 8 1991, which not only set up a new
regional organization, namely the Commonwealth of Independent States
(CIS), but which also declared the end of the USSR. A separate declaration
issued by the heads of the three Republics on the same day revealed ‘. . . that
the talks on the drafting of the new Soviet treaty have become deadlocked
and the de facto process of withdrawal of Republics from the Union of Soviet
Socialist Republics and the formation of independent states has become a
reality.’ The Preamble of the Agreement Establishing the Commonwealth of
Independent States pronounced correspondingly “that the Union of Soviet
Socialist Republics as a subject of international law and a geopolitical
reality no longer exists”. Polat (2002, ix)
In central Asia, lying north of Iran and Afghanistan and between the Caspian Sea
and China, there are five Muslim Republics that achieved independence in the
period August to October 1991. They became members of the United Nations
Asia 339
on 2 March 1992. The five Republics occupy 3.99 million square kilometres
and support a population of 57.13 millions. Turkmenistan inherited Russia’s
boundary with Iran that had been settled in 1881 and 1893. It also shares the
northern boundary of Afghanistan, delimited in 1872 and 1895 with Uzbekistan
and Tajikistan. The former Sino-Russian boundary now marks the eastern limits
of Tajikistan, Kyrgyzstan and Kazakhstan with China.
Polat (2002) published an excellent review of the boundary issues in central
Asia that has informed the following discussion. After independence had been
achieved the five new Republics and China and Russia discussed and negotiated
arrangements to improve boundary stability and accord. The so-called Shanghai-
Five, China, Russia, Kazakhstan, Kyrgyzstan and Tajikistan signed an Agree-
ment on Confidence Building in the Border Areas on 26 April 1996. In June
2000 Uzbekistan joined the Agreement that became the Shanghai Cooperation
Organization.
China’s boundary with its three new neighbours extends for 2,805 km. China’s
boundary with Kazakhstan measures 1,533 km. Two areas totalling 944 sq. km
were in dispute in the Chagan-Obo valley and the Bay-Murza Passes. By an
agreement dated 4 July 1998 China secured 57 per cent of the disputed zones.
The boundary was demarcated by November 1999.
Kyrgyzstan’s boundary with China extends for 858 km. By 4 July 1996 most
of the boundary had been agreed and it was demarcated by 28 October 1998.
The outstanding section was settled on 26 August 1999.
Although Tajikstan’s boundary with China measures only 414 km there is a
dispute over 28,000 sq. km. Progress in solving this dispute was delayed because
of a civil war in Tajikistan that ended in 1997. Working southwards the two
countries defined a part of the boundary by an agreement dated 13 August 1999.
The remaining sector between the Uch’bel Pass and the trijunction of Afghanistan,
China and Tajikistan, on the northern edge of the Wakhan Strip, measures about
300 km. This sector has never been defined by any treaty.
A survey of The Times Atlas of the World from 1966 to 2005 shows only two
small changes in the boundaries of the five Muslim sub-divisions that eventually
became independent states. Between 1967 and 1988 a narrow lake 200 km long,
from east to west, formed south of Toshkent. The lake is called Ozero Aydarkul.
The boundary crosses the lake from north to south leaving the larger area of the
lake to Uzbekistan. However, a rectangular Kazakhstan salient lies on the south
bank of the lake attached to Kazakhstan. This salient has an area of about 1500
sq. km. in which there are settlements called Yenbekshi, Atakent, Zhetysay and
Step. Before the lake was formed and the salient created, the boundary followed
the Syr Dar’ya along the northern shore of the salient.
The other change involves a new boundary. In 1966 The Times Atlas of the
World showed two short pecked lines, in the Aral Sea, to indicate Ostrov [Island]
Vozrozhdenya and Ostrov Kendyrh were part of Uzbekistan and Ostrov Uyayly
was part of Kazhakstan. By 1994 Ostov Vozrozhdenya had been enlarged.
340 Chapter 11
A straight boundary was drawn across the island on the alignment of the termini
of the land boundaries on the northwest and southeast shores of the Aral Sea.
Eleven years later the Times Atlas revealed that Ostrov Vozroz had become a
peninsula projecting from the south shore of the Aral Sea, that is part of Uzbeki-
stan. A solid straight line is shown crossing the island and the Aral Sea from the
southeast to the northwest termini of the existing mainland boundaries.
The boundaries of the former federated republics (Soviet Socialist Republics,
SSR) in Soviet Central Asia were drawn, as often indicated, according to the
security and economic objectives of the center of the Union. Consequently,
the ethnic linguistic and historical factors were largely ignored in delineating
the territories of the republics. Despite the greatly arbitrary character of
the boundaries for the people who populated the region, however, the fact
remains that the boundaries therefore established were treated throughout
the life of the Union as “particularly inviolable”. (Polat, 2002, 46)
Turkmenistan and Uzbekistan were created as Socialist Soviet Republics in 1927;
Tajikistan followed in 1929 and Kakakhstan and Kyrgyzstan having previously
been united were separated as Republics in 1936. When the central Republics
became independent the inviolability of the existing boundaries was confirmed in
the Minsk Agreement of 8 December 1991, the Almaty Declaration of 21 Decem-
ber 1991 and the Commonwealth of Independent States on 22 June 1993.
During the period after 1936 the Republics were permitted to lease territory and
facilities from adjoining Republics (International Crisis Group, 2002). Uzbekistan
leased gas fields and a reservoir within Kyrgyzstan in exchange for pasture within
Uzbekistan. Further complications developed when some enclaves were created.
The two largest are both in Kyrgyzstan. Vorukh is a small area occupies by a
mainly Tajik population while Sokh is occupied by Uzbeks. Both enclaves lie
about 10 km from the nearest international boundary.
The boundaries that the new states inherited appear to have been maintained,
but modifications have been proposed in some cases. Kazakhstan and Rus-
sia settled their boundary, measuring 6,846 km in the period October 1998 to
2003. Kazakhstan’s boundary with Kyrgyzstan, that extends for 1051 km was
delimited between July 1998 and 15 December 2001. This is an uncomplicated
line following the northern edge of the Khirgiz Range and the Kungei Alatau.
Kazakhstan’s boundary with Uzbekistan measures 2,203 km. Negotiations began
in November 1999. The short boundary between Kazakhstan and Turkmenistan
was the subject of talks held from March to June 2000 and the delimitation
was completed by August 2006. The boundary, measuring 426 km, followed
exactly the line laid down by Russia in 1924. The eastern shore of the Caspian
Sea contains very large gas and oil reserves. Kazakhstan completed a pipeline
to China in December 2005. Turkmentistan is planning to connect with China
by January 2009.
Uzbekistan and Kyrgyzstan commenced negotiations in 2000 to delimit the
boundary that extends for 1,099 km. This boundary is complicated by historical
Asia 341
and geographical factors. The historical difficulties are concerned with the leas-
ing to Uzbekistan of up to 140 areas within the Kyrgyzstan border in the 1950s.
In 1955 a draft settlement was prepared but the Supreme Council of the Two
Republics refused to approve the arrangements (Polat, 2002, 51). The geographi-
cal problem relates mainly to the Ferg-ona Basin, a salient of Uzbekistan lying
between the Alai and Chatkal Ranges. Some progress had been made. About
250 km of the boundary had been defined by April 2001, and the trijunction
with Kazakhstan was defined on 15 June 2001. In August 2003 it was reported
that about 900 km of boundary had been delimited. Berghanel (2004) reported
that although the Basin is fertile the collapse of state-owned farms and industrial
establishments has contributed to widespread unemployment.
Relations between Uzbekistan and Tajikistan are bedevilled by two situations.
First Tajikistan is accused by Uzbekistan of failing to curb the activities of ter-
rorists raiding across the boundary. Second Tajikistan complains that Uzbekistan
has illegally occupied areas across the boundary in Tajikistan.
There had been some concerns over the Turkmenistan-Uzbekistan boundary
that extends for 1,621 km, because of Uzbek settlements south of the Amudar’ya.
However, an understanding was reached in January 1996 and the boundary was
fixed by a treaty signed at Ashbagat on 21 September 2000.
The land boundaries of mainland Asia seem to be fixed definitely, or, in the
Himalayas, located in ways that do not promote discord. Along the coast of
eastern Asia there are seven territorial disputes involving islands. The islands
are not large but their existence allows the country that controls them to claim
extended areas of sea and seabed.
Malaysia and Singapore are engaged in a case before the International Court
of Justice dealing with sovereignty over three features called Pulau Batu Puteh,
Middle Rocks and South Ledge.
The Spratly Islands are located in the South China Sea. They include small
islands and reefs that stand above high tide, low-tide elevations that are covered
at high tide and banks and shoals that are covered permanently. Hancox and
Prescott (1995) have listed 48 features and 40 of them are occupied by military
posts belonging to China. Malaysia, the Philippines, Taiwan and Vietnam. From
these features can be claimed areas of seas and seabed totalling about 617,000 sq.
km. Accurate charts of this area, known as The Dangerous Ground, were finally
completed by American, British, French and Japanese hydrographers in the late
1930s, just in time for the Pacific War (Hancox and Prescott, 1997).
The largest island in the Spratly Islands is Itu Aba on Tizard Bank and Reefs.
It measures 1400 metres in length and its maximum width is 370 m. The highest
point on the islands is 2.4 metres above high-tide. Spratly Island that gives the
342 Chapter 11
archipelago its name has the shape of an iscoceles triangle. The base aligned
northeast-southwest measures 750 metres and its apex is 350 metres distant.
Most of the features that stand above low-water are perched on columns rising
1800-2000 m from the seabed. This means that they are not surrounded by shal-
low continental shelves, such as those that border the mainland coasts of China,
Malaysia and Vietnam.
In 1947 the Chinese government issued a map showing a line composed of
eleven segments starting in the Gulf of Tonkin and proceeding around the Luconia
Shoals, between the Spratly Islands and Borneo and The Philippines and continu-
ing east of Taiwan. The defeated Chinese government retreated to Taiwan and
maintained troops on Itu Aba until 1950. The troops returned in 1956 and con-
tinue to occupy the feature. In 1956 South Vietnam claimed the Spratly Islands
and this claim was maintained when Vietnam was unified. Vietnam occupies
19 islands and reefs. In 1979 The Philippines claimed and occupied nine reefs.
In the same year Malaysia claimed five features in the extreme southeast of the
Spratly Islands. China started to make claims and occupied six features in 1988.
These reefs seem to have been selected throughout the archipelago to provide a
basis for China’s claim to the entire archipelago.
Since 1988, when China secured its footholds in the Spratly Islands peace
has reigned. Schofield and Storey (2006) have described confidence-building
measures supported by ASEAN and China in a Declaration on the Conduct of
Parties in the South China Sea.
Samuels (1982) has provided an excellent account of the contest between
China and Vietnam over the Paracel Islands located about 120 nm southeast
of Hainan Island. There are two groups of islands called the Amphitrite and
Crescent Groups. France considered that these islands belonged to the Colony
of Vietnam and Vietnam maintained this view after independence. However,
in the first quarter of 1974 Chinese forces occupied the Paracel Islands that are
now part of Kwangtung Province. The Chinese name for the group is Hsi-sha
Chun-Tao. Its resources include phosphates, guano and fish. Vietnam maintains
its claim to the Paracel Islands, but this territorial dispute did not prevent China
and Vietnam drawing a maritime boundary through the Gulf of Tonkin on 25
December 2000 (Colson and Smith, 2005, 3754-58). The maritime boundary was
drawn to discount Vietnam’s entitlement from the small isolated island of Dao
Bach Long Vi located near the middle of the Gulf.
Between the Hsi-sha Chun-tao and The Philippines is located Scarborough Reef.
It lies about 120 nm from the coast of Luzon and it is claimed by both countries.
There is no island on Scarborough Reef but there are rocks above high water
and the large South Rock stands about 3 m above high water. Comparison of
modern charts with the chart produced by Edward Miles, Master of HMS Swal-
low in 1866, at a scale of 1:48,900 indicates that the topography of the reef has
not changed in 141 years. The reef has attracted fishermen from both countries
and it can be assumed that the Reef can sustain an economic life of its own in
Asia 343
accordance with Article 121 (3) of the Convention on the Law of the Sea. This
would allow either of the countries to claim the feature as a basepoint for deter-
mining any maritime boundary between them. The area of seas and seabed that
is closer to Scarborough Reef than any other land is about 54,000 sq. nm.
There is a dispute between China and Japan over the ownership of the Diaoyu/
Senkaku Islands that lie north of Taiwan. The Group consists of five islands and
three rocks in three shoal areas. The Islands Diaoyu Dao [Uotsuri Shima], Bei
Xiaodao [Kitako Shima] and Nan Xiaodao [Minamiko Shima] and the rocks
Feilai [Tobi Se], Peihsiao [Okinikita Iwa] and Nanhsiao Okinominami Iwa] are
situated on a triangular shoal measuring about 10 sq. nm, with depths less than
100 m. Huangwei Yu lies 11 nm northeast of Pseihsiao and is the summit of an
extinct volcano standing 116 m above high water. It has an area of 79 hectares
and is covered with palm trees and undergrowth. Chiwei Yu, the most isolated
of the Diaoyu Islands lies 48 nm east of Huangwei Yu. It is steep-to and rises
to a peak at 83 m.
This brief description indicates that the features have little value as territory that
might be mined or cultivated. However, their value lies in the claim that can be
made from them to the seas and seabed measuring 20,400 sq. nm. About three-
quarters of this area is located on the continental shelf adjacent to the Chinese
mainland and the remainder covers the continental slope, that leads to the deep
seabed. There is a chain of treaties from April 1895 to 1951 that stretch from
China’s loss of Formosa to Japan in 1895 to its return in 1952. In 1895 Japan
defeated China and imposed the Treaty of Shimonoseki, that delivered to Japan
the island of Formosa together with all islands ‘appertaining to or belonging to
the said Island of Formosa’ (Prescott, 1997, 49). In September 1951 the Multi-
lateral Peace Treaty included the following statement in Article 2 of Chapter II
dealing with territory.
(b) Japan renounces all right, title and claim to Formosa and the Pescadores.
(Bureau of Public Affairs, 1977, 1120)
It would be reasonable to conclude that Formosa included islands that appertain
to Formosa.
It is possible that this dispute is more important to Japan than to China. China
has extensive areas of continental shelf that can be explored for oil, gas and min-
erals. In contrast Japan’s principal islands are surrounded by narrow continental
shelves that will offer fewer opportunities for mining.
Japan and South Korea face each other in the East China Sea, the Korea Strait
and the Sea of Japan. In the Sea of Japan there is a group of Islands called Lian-
court Rocks consisting of two islets and some rocks. The Korean name is Tok
Do and the Japanese name is Takeshima. If lines of equidistance are drawn on
the basis that first Japan and then South Korea owns the Liancourt Rocks, they
enclose an area of sea and seabed of 15,000 sq. nm. At present the islands are
occupied by Korea.
344 Chapter 11
The Kuril Islands are aligned like stepping-stones between Hokkaido and
the Kamchatka Peninsula. They extend about 630 nm and the southern tip of
Kunashir, that is the southernmost island in the chain, is only 6 nm from Hok-
kaido. Lying 30 nm from the south coast of Kunashir is group of islands called
Habomai and an island called Shikotan.
Russia occupies the Kuril Islands, Shikotan and the Habomai Group. Japan
claims that it should possess Shikotan Island and the Habomai Group, as well
as the two most southerly Kuril Islands called Iturup and Kunashir.
In any précis of the origin of this dispute, the following facts are important.
Working from the north and south respectively, Russian and Japanese explor-
ers and hunters charted the islands and began to wrest a living from them. By
the Treaty of Shimoda of 7 February 1855, the two countries divided the Kuril
Islands at the Strait passing between Iturup and Urup Islands. Today this Strait is
called Proliv Friza in Russian and Etorofu Kaikyo in Japanese. This treaty thus
gave Japan the islands known today as the Northern Territories. Stephan (1974)
who has written a full account of the history and politics of the Kuril Islands,
records that Russia claimed all the Kuril Islands in 1855 but agreed to withdraw
its claim to the two most southern islands when its strategies were focussed on
the Crimean Peninsula.
By the Treaty of St. Petersburg signed on 7 May 1895, Japan and Russia agreed
to an exchange of territory. Japan ceded its territories in southern Sakhalin to
Russia and received the Kuril Islands lying between Kamchatka and Proliv Priza
(Stephan, 1974, 237-8). However, by the Treaty of Portsmouth of 5 September
1905, Japan regained southern Sakhalin [Karafuto] and some adjacent islands.
At the end of World War II Japan forfeited all the Kuril Islands and Shikotan
and Habomai Shoto. According to Article 2 of the San Francisco Treaty of 8
September 1951, Japan renounced all rights, titles and claims to the Kuril Islands
and to southern Sakhalin and some adjacent islands gained in 1905.
Successive Japanese governments have maintained their claim that Russia
should cede all islands lying west of Etorofu Kaikyo. Such a claim, if successful,
would deliver 57,000 sq. nm of sea and seabed. Although John Foster Dulles, at
the San Fransisco Conference charged the Soviet Union with illegal occupation
of Shikotan and Habomai Shoto, there was never any suggestion that Kunashir
and Iturup should be Japanese territory (Stephan, 1974, 200).
CONCLUSIONS
The main lines of political cleavage in Asia were established by European pow-
ers in two stages. First before World War I they carved out Asian empires by
negotiating boundaries with China and with those Asian states such as Afghani-
stan, Nepal and Thailand that were not annexed. Second, before World War II
Asia 345
REFERENCES
Colson, D.A. and Smith, R.W., 2005. ‘People’s Republic of China and Vietnam’,
International Maritime Boundaries, volume V, Martinus Nijhoff: Leiden.
Conroy, H., 1960. The Japanese seizure of Korea: 1868-1910, Philadelphia.
Davies, C.C., 1932. The problem of the Northwest Frontier 1890-1908, University
Press: Cambridge.
Fraser-Tytler, W.K., 1967. Afghanistan: a study of political developments in
central and southern Asia, 3rd edition, revised by M.C. Gillett: London.
Friters, G.M., 1951. Outer Mongolia and its international position, London.
Grey, A.L., 1951. ‘The thirty-eighth parallel’, Foreign Affairs, 29, 482-7.
Gupta, S., 1966. Kashmir: a study in India-Pakistan relations, Bombay.
Hancox, D. and Prescott, J.R.V., 1995. ‘A geographical description of the Spratly
Islands and an account of hydrographic surveys amongst those islands’,
Maritime Briefing, 1(6) International Boundaries Research Unit.
———, 1997. Secret hydrographic surveys in the Spratly Islands, Maritime
Institute of Malaysia: Kuala Lumpur.
Hayat-Khan, A., 2000. The Durand Line: its geo-strategic importance, Area-
Study Centre: Peshawar.
Hulbert, H.B., 1962. The history of Korea (ed. C.N. Weems, 2 volumes:
London.
Hsu, I.C.Y., 1965. The Ili crisis: a study of Sino-Russian diplomacy 1871-81,
Oxford University Press: Oxford.
International Boundary Research Unit, 8.1. 2007. Bhutan and India complete
latest demarcation, Boundary News.
International Crisis Group, 2002. ‘Central Asia: border disputes and crisis reso-
lution’, Asia Report 3.
Karaeve, Z., 2005. Border disputes and regional integration in central Asia, IX
(4), 6pp.
Kim, C.I.E. and Kim, H., 1967. Korea and the politics of imperialism, Los
Angeles.
King, L.W., 1895. Afghan Boundary Delimitation, 391F, Report on demarca-
tion operations on the southern section of the Waziristan-Afghan Boundary,
8 Arch 1895, India Office Library, London.
Lagergren, G., 1968. ‘Judicial decisions: The Indo-Pakistan Boundary Case
Tribunal’, Indian Journal of International Law, 8, 247-65.
Lamb, A., 1960. Britain and Chinese Central Asia, London.
———, 1966. Crisis in Kashmir, 1947-1966, London.
———, 1968. Asian frontiers: studies in a continuing problem, Cheshire:
Melbourne.
Mancall, M., 1971. Russia and China: their diplomatic relations to 1728, Harvard
University Press: Cambridge: Massachusetts.
McCune, S., 1949. ‘The thirty-eighth parallel in Korea’, World Politics, 1, 223-
32.
Montgomerie, T.G., 1871. Report on the Mirza’s exploration from Kabul to
Kashgar’, Journal of the Royal Geographical Society, 41, 150-5.
Asia 347
Alsace-Lorraine was acquired by the French through the methods which have
led to the consolidation of most modern States, namely conquest, trickery
and cession. (Temperley, 1920, 159)
The brutal honesty of this statement would be preserved if names such as Arica,
Burundi and Sikkim were substituted for Alsace-Lorraine and the charge was
made against Chile, Belgium and Britain. However, it would be an error to
assume that because European states had similar territorial ambitions in Europe
and other continents, and used similar policies to achieve them, that there are
close similarities between the evolution of international boundaries in Europe and
other continents. There are more important differences in respect of boundaries
between Europe and the rest of the world than there are between any other two
continents.
into small parcels, that were sometimes aggregated into larger administrative
blocks. These local divisions provided a series of building blocks from which
national territories could be fashioned. Thus the international bargaining centred
on pieces of territory, and once their disposition had been decided there was a
ready-made boundary, that might, in different sections, coincide with known
boundaries between parishes, bailiwicks or bishoprics.
This advantage was not available to the same extent in the Balkans where
feudalism as known and practised in Western Europe did not exist (Pounds, 1947,
129). Nor did it apply after World War I, when strategic considerations were a
powerful factor in determining the alignment of boundaries.
Just as the physical and political geography of Europe was known clearly to
negotiators, so the economic geography of disputed regions was understood.
Examples of states pursuing clear economic objectives are the determination of
German representatives to secure as many of the iron-ore fields of Alsace-Lorraine
as possible in 1871; the competition for coal-fields in Silesia by Czechoslovakia,
Germany and Poland in 1919; Yugoslavia’s enthusiasm for securing the port
of Rijeka by the Treaty of Rapallo; and Rumania’s insistence on control over
the railway from Timisoara to Arad. Although these are comparatively recent
examples, there is no reason to suppose that nobles, in earlier periods, were not
completely aware of the economic advantages of securing the transfer of specific
feudal rights from a defeated or ruined neighbour.
In the broad swathe of central Europe, stretching from the coasts of Germany
and Poland to the shores of the Adriatic, Aegean and Black Seas, present national
boundaries were fashioned in a series of major wars. To a much greater extent
than on any other continent, boundaries in this region were created during the
peace conferences that followed widespread conflict. Judged by the speed with
which agreement was reached at the Congress of Vienna in 1815, the Congress
of Berlin in 1878, the London Conference in 1913 and the Peace Conferences
that ended World Wars I and II, it is possible to conclude that wars simplify the
drawing of boundaries. The successful powers usually have clear ideas about
the territorial arrangements they will demand before victory is achieved, and
the defeated states are rarely able to resist most of the territorial adjustments.
In the majority of cases there is probably an overwhelming desire by all parties
involved to secure an agreement that will allow the abnormality of war to be
ended. Successful governments want to demonstrate to the tired populace the
gains bought by military sacrifices. They also wish to avoid the need to maintain
large armies and navies, and to reduce to a minimum the funds that have to be
expended on administering the territory of defeated states. Defeated countries
generally seek to regain a level of independence necessary for the reconstruction
of national morale and wealth.
352 Chapter 12
In 1828 and 1834 Prussia created customs unions, called Zollvereine, with
neighbours.
These unions prospered as railways were built and manufacturing industries
were developed. Bismark became Chancellor of Prussia in 1862.
The political unification of Germany involved wars against Denmark in
1864, against Austria in 1866 and against France in 1870. Austria was
excluded from the North German Confederation in 1867 that comprised
the German states north of the River Main. Any influence that remained to
Austria in Germany, as a relic of the dualism that had obtained before 1866,
was finally destroyed in 1871, when the German southern states joined the
German Empire. (Barraclough, 1978, 216)
The unification of Germany and Italy in central Europe was matched by the
disintegration of the Austrian, Russian and Ottoman Empires. It is this latter
process that was important for the evolution of European boundaries and it began
in southeastern Europe on 21 July 1832 and spread southwestwards.
In 1832 a treaty signed in Constantinople by Britain, Russia, Turkey and
France established the limits of continental Greece (Hertslet, 1975, 903-8). The
successful Greek war of independence occurred two years after Belgium seceded
from the Netherlands.
by the major powers and arrangements were made to transfer the two zones. By
accident or design the Turkish withdrawal from some areas occurred before the
Montenegro forces were ready to move, and the military vacuum was filled by
Albanian irregulars. They proceeded to establish themselves in force in fortified
positions astride the road between Podgoritza and Scutari.
Earl Granville described the situation in this region in the following terms.
Your Excellency [G.J. Goschen] is aware from the reports, which have
been received, that the state of the country in North-East Albania is little
short of anarchy. The Turkish officials are powerless to execute justice;
murder and violence and forced exactions are prevalent, and the peace-
able population is at the mercy of armed Committees, who, under the state
of the Albanian League, have been allowed to assume absolute authority.
(Hertslet, 1891, 3000)
Turkey was either unwilling or unable to dislodge the Albanians from the eastern
part of the area ceded to Montenegro. Negotiations then began to discover if there
was additional territory that could be ceded to Montenegro in exchange for the
area around Dinosi and Metagbush occupied by the Albanians.
The British and Austrian consuls in Scutari, Messrs Green and Lippich, pro-
posed that Turkey should cede the coast between Vieux Kruci and the Bojana
River, including the town of Dulcigno. Turkey eventually accepted this arrange-
ment providing Dinosi was excluded from the area occupied by Montenegro.
This was the last serious difficulty and the way was now open to delimit and
demarcate the boundary.
The line from the coast at the mouth of the Bojana River to Lake Scutari was
fixed by the European Commission meeting in January and February 1881. There
was some debate about whether the boundary should follow the west bank of the
Bojana River, preferred by Turkey, or its talweg preferred by Montenegro. The
matter was resolved in favour of Montenegro (Hertslet, 1891, 3029).
The remainder of the boundary from the northern shore of Lake Scutari to
Secular was delimited in three sections by a joint Turkish-Montenegran Commis-
sion between January 1883 and December 1884. The reports of this Commission
indicate that in the vicinity of the Lim River the Commissioners were able to
use old village boundaries to define the international limit (Hertslet, 1891, 3193).
In June 1887 a decision was made by both countries to eliminate the Turkish
pedicle about Kjanitza, and this area was transferred to Montenegro.
The Turkish corridor from the Black and Aegean seas to the Adriatic Sea
was eliminated west of the Maritsa River during the Balkan Wars of 1912-
13. Albania emerged as a separate country and Bulgaria, Greece and Serbia
annexed the remaining areas (Carrié-Albrecht, 1958, 281). The scene was now
set for the impact of disintegration to be experienced by the Austro-Hungarian
Empire.
Europe 359
At the conclusion of World War I the victorious powers signed separate peace
treaties with their five opponents. The Treaty of Versailles was signed with Ger-
many on 28 June 1919, and this was quickly followed by treaties with Austria
and Bulgaria, signed respectively at St Germain-en-Laye on 10 September 1919
and at Neuilly on 27 November 1919. The treaty with Hungary was signed on
4 June 1920 at Trianon, and finally, on 10 August 1920 the Treaty of Sevres
was signed by Turkey. The Carnegie Endowment for International Peace (1924)
published a very useful collection of these treaties, together with a commentary
and maps prepared by Martin.
Toynbee (1935) distinguished two main aims of the territorial provisions of
these treaties.
In the territorial chapter of the peace treaties, as in the Disarmament
chapter, one of the things which the winners of the war were aiming at
was to strengthen themselves and weaken their late enemies comparatively,
as much as possible. The second and more ideal purpose in the territorial
resettlement of Europe was to redraw the political map is such a way that
as many Europeans as possible would be living under national governments
of their own choosing. This was called the principle of self-determination.
(Toynbee, 1935, 72)
Germany was weakened by the loss of Alsace-Lorraine to France, of northern
Schleswig to Denmark and of western Prussia and part of Silesia to Poland. The
loss of Western Prussia weakened Germany in two ways. First it forfeited the
population and resources of the region; second East Prussia was detached from
the rest of the country. Germany also forfeited to Belgium four small territories
called Eupen, Malmedy, Prussian Moresnet and Neutral Moresnet. Bulgaria was
weakened by the loss of its Aegean coastline to Greece and the loss of three
salients on its western boundary that had been used to attack Serbian communi-
cations (Temperley, 1921, 455).
A chain of new or modified existing states was created from the Arctic Ocean
to the Adriatic Sea. The chain consisted of Finland, Estonia, Latvia, Lithu-
ania, Poland, Czechoslovakia, Austria, Hungary and Yugoslavia (Figure 12.4).
Hupchick and Cox describe the complexities of the reconstruction of Eastern
Europe
At Versailles, the old anational political structures were dismantled, but,
unfortunately, the new states were constructed more to ensure both the
prolonged punishment of the defeated Central Powers and the political and
economic interests of the Western Great Powers in Eastern Europe than to
uphold Wilsonian ideals of national self-determination. Most of the new states
artificially encompassed disparate nationalities: Czechoslovakia (Czechs,
Slovaks, Germans, Hungarians and Ruthenians); Yugoslavia (Serbs, Croats,
360 Chapter 12
Roumania claimed the entire Banat. This was a condition contained in a secret
treaty of 17 August 1916 when it entered the war on the side of the Allies. But
neither the western limit of the Banat, nor the line to which the Serbians could
advance became the boundary. A commission of the Peace Conference drew up
proposals to allocate territory according to ethnic majorities. When these plans
were almost complete France intervened and insisted that Yugoslavia should
receive two salients into the Banat that contained the towns of Vrsac and Bela
Crkva (Naval Intelligence Division, 1944, 152). When the Peace Commission
announced the final allocation of territory, on 13 June 1919, it was evident that
Yugoslavia had received a wide belt of territory to the north and east of Belgrade.
It also received most of the Vojvodina that contained a remarkable intermin-
gling of Serbs, Croats, Slovaks, Czechs, Krassovans, Roumanians Germans and
Magyars. The Naval Intelligence Division (1944, 74, 75) produces three maps
drawn by British, Hungarian and Serb cartographers using the 1910 ethnographic
statistics. The maps show little correspondence.
Bulgaria’s boundary with Yugoslavia was modified in favour of Yugoslavia
on 27 November 1919 (Figure 12.7). For strategic reasons the Strumica salient
was transferred to avoid any threat to the Vardar railway. Yugoslavia also gained
two adjacent districts called Bosiljgrad and Caribrod. A very small area, at the
head of the Timok valley was also detached from Serbia.
It was noted earlier that Yugoslavia’s boundary with Greece needed no further
delimitation. The boundary with Albania was also settled without difficulty. The
Council of Ambassadors had drawn the boundary between Albania and Austria
in 1913 (Figure 12.8). Albanians had been migrating across the 1913 line and
by 1921 there were about 500,000 Albanians in the centres of the Yugoslav
areas Pec, Mitrovica, Prizren and Tetoyo. The 1913 boundary, with a few small
adjustments was confirmed as the boundary between Albania and Yugoslavia.
In an effort to find boundaries that accorded with national settlements, plebi-
scites were held in the following areas: northern Schleswig, Eupen and Malmedy,
upper Silesia, the Klagenfurt Basin on the borders of Austria and Yugoslavia,
Oldenburg in the borderland of Austria and Hungary, and the southern and
western parts of East Prussia.
Russia started the war on the side of the Allies, but after its transformation
into the Soviet Union it was defeated by Germany and forced to sign the Treaty
of Brest-Litovsk in 1918. This treaty detached from the Soviet Union a strip of
territory occupied mainly by minorities who sought national self-determination.
The German plan to dominate this region was thwarted by the Allies’ victory.
The Allies then presided over the emergence of Finland, Estonia, Latvia, Lithu-
ania and Poland. The Ukraine that had been a German target was left with the
Soviet Union. However, it could not retain Bessarabia, that once again changed
hands and became part of Romania. Toynbee (1935, 75-7) perceptively noted
that the principle of national self-determination would not have been carried out
to such a marked extent in Europe, if the Soviet Union had remained attached
to the Allies during the war.
Europe 363
The Supreme Council established by the Allies to oversee the postwar territo-
rial arrangements played a major role in allocating territory to the various new
countries. The work of this Council has been described by Baker (1923), Hankey
(1963) and Mantoux (1955).
Ogilvie (1922) provided an excellent survey of the processes by which
boundaries were established after World War I. He identified four main stages
that began when the Supreme Council instructed one of its several Territorial
Commissions to start work. These Commissions dealt with the boundaries of
Albania, Czechoslovakia, Greece, Romania and Yugoslavia. They consisted of
persons skilled in solving demographic, economic and strategic problems within
the borderlands under consideration. Through consultations the Commissions
produced a report recommending, in general terms, the description of a boundary
on a map at a scale of 1:1 million. The second stage involved a Sub-Commission
that worked on maps at a scale of 1:200,000 to produce a precise line suitable
for demarcation.
In the Sub-Commission stage such questions as the military value of hill
ranges and the military and economic importance of railway junctions were
considered, as were the possibilities of maintaining, intact, geographical
units, such as intermontane basins and valleys. (Ogilvie, 1922, 6)
This refined boundary was then submitted to the Central Geographical Committee
that produced the final description of the boundary and defined the latitude that
should be granted to the Demarcation Commissioners. The Committee worked
on cadastral maps and sought to ensure that the new international boundaries
did not partition individual properties. The use of survey points to simplify the
process of demarcation and specific details about the boundary applying to rivers
were also the responsibility of the Central Geographical Committee. When the
boundary had passed through these stages it was incorporated into the relevant
treaty. After the treaty was ratified demarcation could proceed.
Although Hinks (1919a, b, c and 1920a, b) published helpful notes and articles
on the new European boundaries, members of the delimitation and demarcation
Commissions did not publish articles with the frequency of surveyors in Africa
and Asia. Cree (1925) provided an article dealing with the work of the Yugoslav-
Hungarian Boundary Commission. He described how the Commission had a
covering letter and detailed instructions, provided by the Central Geographical
Committee, as well as the description of the treaty. Before the fieldwork com-
menced the delegates representing Hungary and Yugoslavia were required to
present their interpretation of the treaty’s description, and identify any section
where an appeal to the League of Nations might be considered. Then after a pre-
liminary reconnaissance, meetings were held on the same day to hear the views
of interested communities in the borderland. Mayors, Presidents of Chambers
of Commerce, large landowners, factory owners and farmers would attend these
meetings and present their case for a particular line and answer any questions the
Commissioners might raise. The President of the Commission then consulted the
364 Chapter 12
Hungarian and Yugoslav delegates to secure total agreement. Once agreement had
been reached one member of the team would be asked to define the boundary.
Then in turn other members would be invited to propose any modification to
the line, and each proposed change would be debated until there was agreement.
If no agreement was possible a vote would be taken. The boundary was then
marked on a map at a scale of 1:75,000 and this was signed by the members. The
Hungarian and Yugoslav Assistant Commissioners then had the responsibility of
demarcating the boundary. In a few cases, involving the railway station at Gola
and the water supply for the railway station at Gyekenyes, the Commissioners
prepared protocols to govern particular transboundary functions.
During the 1930s Germany, Italy, Hungary and Poland secured additional ter-
ritories from their neighbours. The Rhineland and the Saar were reoccupied
by Germany in 1935-6. Germany also received three blocks of territory along
Czechoslovakia’s western boundary in 1938. In the same year Germany annexed
Poland’s Memel March from Lithuania. In 1939 Hungary annexed parts of
southeast Czechoslovakia and the province of Ruthenia. In 1938 Poland seized
a small area of Czechoslovakia in the upper valley of the Odra River. On Good
Friday that fell on the 7 April 1939, Italy invaded Albania. The territory not only
possessed useful quantities of oil and bauxite but would provide also a platform
for the invasion of Greece.
After World War II the main boundary changes were in northern and eastern
Europe (Figure 12.9). The Soviet Union regained most of the territory lost by the
Treaty of Brest-Litovsk. Although Finland remained independent, the boundary
was adjusted in three regions in favour of the Soviet Union and Finland lost its
coastline on the Arctic Ocean. Poland continued its eastern march across the
Baltic plains and secured those areas of Germany east of the Oder and Neisse
Rivers in exchange for eastern areas of Poland that were lost to the Soviet Union.
The Soviet Union also acquired sub-Carpathian Ruthenia, that formerly had
been part of Czechoslovakia, and for the fifth time since 1812, Bessarabia was
transferred from one country to another! There were two other minor territorial
alterations. Romania, having lost Bessarabia to the Soviet Union, also forfeited
southern Dobrogea to Bulgaria. Most of Italy’s gains along the Adriatic coast
in 1920 were lost to Yugoslavia. The Soviet Union encouraged the establish-
ment of communist governments in the band of countries from the Baltic Sea
to the Adriatic Sea, consisting of Poland, Czechoslovakia, Hungary, Romania,
Yugoslavia, Bulgaria and Albania.
Albrecht-Carré pointed to the marked difference between the construction of
boundaries in 1919 and 1945.
Europe 365
. . . at all events the peacemakers of 1919 had adhered to the humane prin-
ciple that the land belongs to the people and that frontiers therefore must
be adapted to their existing distribution. Things were now [1945] otherwise
and the ethnic map of Europe was brought into much closer agreement with
its political divisions through the simple, if crude and inhumane, device of
making the people fit the frontiers. (Albrecht-Carré, 1958, 605)
Starting in the north 200,000 refugees fled from Estonia, Latvia and Lithuania.
About 2.3 million Russians swarmed into the new western territories annexed by
the Soviet Union. Polish refugees totalling about 4.5 millions moved into western
Poland. The largest flight involved 12.35 million Germans heading towards those
parts of Germany controlled by the Western Allies. The International Refugee
Organization settled about one million refugees overseas.
Judged by the published studies, it was the boundary between Italy and Yugosla-
via that provoked the most interest amongst political geographers. Moodie (1945),
Kiss (1947) and Le Lannou (1947) provided perceptive and detailed studies of the
area that Moodie called the Julian March. Alexander (1953) published a useful
account of the small boundary adjustments made in 1949 along the boundary of
Germany with the Netherlands, Belgium and Luxembourg.
Italy forfeited the Peninsula of Istria, that was divided between the two Republics.
Italy retained a narrow section of coast containing Trieste.
Bosnia and Herzegovina possesses only a narrow corridor 10 km wide giv-
ing access to a narrow gulf called Zaliv [Gulf ] Klek Neum formed by the Klek
Peninsula. Zaliv Klek Neum is an excellent harbour for vessels of deep draught.
Blake and Topalovic (1996) described the history of the corridors to the sea that
Austria and later Bosnia sought. This narrow Bosnian corridor was created in
1700 according to the Treaty of Karlowitz and the treaty of Passarowitz of 1718,
however, the seas adjoining Zaliv Klek Neum were controlled by Venice. Austria
gained outlets at Klek Neum and at Sutorina that led to the Bay of Kotor. After
World War II Bosnia-Herzegovina retained the Klek Neum outlet but Sutorina
was delivered to Montenegro.
Bosnia Herzegovina looked for an outlet to the Adriatic on the Croatian coast.
The building of Ploce started in 1940 but did not make significant progress until
after 1949. Ploce is now the most important port for Bosnia.
Macedonia is the only former member of Yugoslavia that is landlocked. Its
boundaries with Albania, Greece and Serbia were the international boundaries of
Yugoslavia from the end of World War I until the dissolution of the federation in
1991. These international boundaries were settled at the Bucharest Peace Agree-
ment of 28 July 1913. The boundary with Bulgaria was demarcated in 1920-22,
the Albanian demarcation followed in 1922-25, and finally the boundary with
Greece was completed in 1930. The boundary with Greece was resurveyed in
1981-3.
In total there were 2,577 border markers along the border between Greece
and Macedonia. On land there are 177 major and about 2,366 subordinate
border markers, three light signals and 30 buoys. The deviation at the rail-
road Bitola-Lerin (Florina) is marked with border signs. Two light signals
are positioned on Lake Dojran where the border proceeds from land to
water. One signal light is positioned on Lake Prespa. There are four types
of border markers: border stones 110-180 cm high; stones 50 cm high;
and border markers in the form of cylinders made of asbestos, 70-80 cm
high. There are 18 buoys in the shape of a double cone (1.5 m high) on
Lake Dojran and 12 on Lake Prespa, spaced 600 m part. (Milenkowski
and Talevski, 2001, 82)
Macedonia’s boundary with Serbia is based on the northern limits of the smallest
administrative limits of Macedonia that coincide with topographic features.
The boundary with Yugoslavia can be considered to be mainly ‘natural’
because it extends along the high parts of the mountains. It consists of dif-
ferent relief features: mountains, canyons and plains. The mountain part of
the border extends across the mountains: German, Kozjak, Skopska Crna
Gora and Sar Planina. The border passes along the Pcinja and Kacanik
Europe 367
CONCLUSIONS
The boundary readjustments of the past have been worked out through a
series of gruesome wars; those of the future contain elements of still greater
difficulty. Altogether the problems of just and proper boundaries is hopeless
and growing worse on the basis of old statecraft. (GilFilan, 1924, 484)
In his very interesting but gloomy paper GilFilan rightly stressed the roles of wars
in producing new boundaries in Europe and of territorial disputes in promoting
wars. This characteristic and the fact that there was no colonial discontinuity
distinguishes boundary evolution in Europe from the formation of boundaries
in other continents.
Political geographers of the 1920s and 1930s should not have been surprised
by the onset of World War II. The boundary adjustments of 1920 had sown the
seeds of conflict that were harvested in 1939. But the developments after World
War II eventually produced three trends that have contributed to the avoidance
of World War III. First all the major powers appear to be content with the terri-
tory they possess and do not seek to expand into neighbouring areas. Second in
the early 1990s Russia liberated nine former territories throughout its southern
borderland from the Baltic Sea to the Caspian Sea. At the same time Czecho-
slovakia peacefully bifurcated and the threat of major wars in Yugoslavia was
apparently avoided by each of the Republics finding independence alone or in
association with a neighbour. Third the European Economic Community reduced
the divisive nature of international boundaries making it easier for people, goods
and ideas to circulate throughout the continent. For the present the previous cycle
of territorial dispute, war, boundary adjustment and new territorial dispute seems
to have been broken.
370 Chapter 12
REFERENCES
Ogilvie, A.G., 1922. ‘Some aspects of boundary settlement at the Peace Confer-
ence’, Help for Students of History, No. 49, Society for promoting Christian
knowledge. Macmillan: London.
Pounds, N.J.G., 1947. An historical and political geography of Europe, London
University Press: London.
Roberts, A., 1999. Lord Salisbury: Victorian Titan, Weidenfeld and Nicolson:
London.
Shepherd, W.R., 1922. Historical atlas, London University Press: London.
Smith, W. and G. Grove, 1874. An atlas of ancient geography: biblical and
classical, Murray: London.
Strayer, J.R., 1965. ‘Feudalism in Western Europe’, in Feudalism in history,
R. Coulburn (editor), 26-48, Archon Books: Connecticut.
Temperley, H.W.V., 1920 and 1921. A history of the Peace Conference, volumes
2 and 4. Oxford University Press: London.
The Geographer, 1978. ‘Iran-USSR’, International Boundary Study, 25, Depart-
ment of State: Washington DC.
Toynbee, A.J., 1935. ‘Territorial arrangement’, in The Treaty of Versailles and
after, Lord Riddell (editor), 70-93, Allen and Unwin: London.
Treharne, R.F. and H. Fullard, 1965. Muir’s historical atlas: ancient, medieval
and modern, George Philip: London.
Zamoyski, A., 2007. Rites of Peace, the fall of Napoleon and the Congress of
Vienna, Harper Press: London.
13. LAND BOUNDARIES ON ISLANDS OFF
SOUTHEAST ASIA
March 1889 a violent storm struck Apia harbour. There were seven ships
there at the time, three American, three German and one British. The only
one not driven ashore was HMS Calliope, a new ship with more powerful
engines than the others. By dint of good seamanship, she was able to put
to sea. (Naval Intelligence Division, 1943, 597)
Britain, Germany and the United States held talks in Berlin in June and created
a cumbersome administration that lasted until 1899. The parties then agreed that
the United States would secure Tutuila leaving Upolu and Savai’I to Britain and
Germany. Britain agreed to withdraw from Samoa in favour of Germany and
was rewarded with Choiseul, Ysabel, Ontong Java and some other small islands
in Bougainville Strait from Germany. In 1914 Australian troops occupied the
German Solomon Islands.
BOUNDARIES ON TIMOR
The Portuguese established settlements on Timor in 1520. Some years later Dutch
colonists began to arrive and by 1613 they had occupied the larger, western part of
the island. Both groups signed treaties with the indigenous population in the first
half of the 19 th century and the governments signed a treaty on 20 April 1859
to end uncertainty about the territory over which each held jurisdiction (Deeley,
2001, 37-9). According to the first Article of this agreement, the Netherlands
received the territories of Djenilo, Naitimu, Fiarlarang, Mandeo and Lakecune;
Portugal secured Cova, Balibo, Lamakitu Tahakay and Suai. Although the island
was divided into an eastern Portuguese region and a western Dutch sector, each
country possessed small areas of territory in the other half of the island. The
Dutch retained Maucatar in the Portuguese region and Portugal retained Ocussi,
Tahakay, Tamiru Ailala and Noe Muti in Dutch territory. This agreement also
resolved disputes over the ownership of islands lying north of Timor. Portugal
obtained Poulo Kambing, now called Atauro, lying north of Dili, while the
Netherlands secured Flores, Adenara and Solor Islands.
In 1893 the two countries agreed to define the boundary and to eliminate
the enclaves in the other state’s territory. During the period 1898 to 1899 a
joint commission surveyed the boundary across the island and a strip on either
side, and this enabled both sides to make specific proposals for the exchange
of territory. Attempts to complete the delimitation of the Ocussi boundary were
thwarted by hostile tribesmen. Negotiations in 1902 tried to reconcile the various
conflicting proposals and they lasted two years. Portugal sought to eliminate the
Dutch salient of Fialarang, but the Dutch stood firm and threatened to raise the
question of whether Ocussi was an enclave that should be eliminated. Because it
has a coastline Ocussi cannot be considered to be an enclave, but the Portuguese
decided to avoid any threat to Ocussi and Fialarang remained Dutch. The large
Land Boundaries on Islands Off Southeast Asia 375
Dutch enclave of Maucatar was exchanged for the small Portuguese enclave of
Noe Muti. As a concession to the disparity in the size of these two territories
the Portuguese also ceded Tahakay and Tamiru Ailala.
When a joint commission began to demarcate these boundaries in 1909 a
positional dispute developed along the eastern limit of Ocussi. The two coun-
tries referred the matter for decision by the Permanent Court of Arbitration. The
Netherlands sought a segment of boundary extending from a point on the Noel
Bilomi to the source of the Noel Meto, a distance of 17 km. Portugal claimed
that the boundary should start from a point 7 km east of the Dutch origin and
proceed parallel to the Dutch claim to the coast about 7 km east of the mouth of
the Noel Meto. On 25 June 1914 the Court found in favour of the Netherlands
and two straight lines meeting at Kilali, traversed the gap between the Noel
Bllomi and the source of the Noel Meto.
In 1975 Indonesia invaded Portuguese Timor and occupied it until 1999 when
the United Nations supervised a plebiscite in the territory that resulted in a deci-
sion in favour of independence. East Timor became independent in 2002.
The following account of subsequent boundary negotiations is based on
information provided by Arif Havas Oegroseno of Indonesia’s Foreign Affairs
Department (Figures 13.1 and 2). On April 8, 2005 in Dili, Timor Leste, the
Minister for Foreign Affairs of the Republic of Indonesia and The Minister for
Foreign Affairs of the Democratic Republic of Timor Leste, on behalf of their
respective governments, signed a Provisional Agreement on the land boundary
between the two Republics.
The Provisional Agreement contains items that were agreed by both countries
through negotiations namely, 907 border point coordinates that will serve as the
permanent border between the two countries. The provisional character of the
Agreement takes into account the facts that more border coordinate points are
still needed; and three segments namely Dilumil-Memo, Bijael-Sunan Oben and
Noel Besi-Citrana remain unresolved. However, the Agreement is final in the
sense that there will be no modifications of the 907 agreed border coordinates.
The signing of the Agreement signifies the importance of a definite boundary
between the two countries and ensures the supremacy of law over their respec-
tive territory that will result in improving the security condition in the border
regions. The interactions among the people living in the border area should result
in increased economic activity, especially with the opening of markets around
border area. It is hoped this will decrease the level of smuggling activities in the
area. Both countries have agreed to start the delimitation of maritime boundaries
after the completion of the unresolved territorial segments.
Land boundary negotiations between Indonesia and Timor-Leste were initiated
in 2001. They produced a Memorandum of Understanding between Indonesia and
Timor-Leste by Coordinating Minister For Political and Security Affairs, Susilo
Bambang Yudhoyono, acting as Minister of Foreign Affairs and Sergio Vieira
De Mello, Administrator of UNTAET on 14 September 2000.
376 Chapter 13
Institutions called Joint Border and Border Liaison Committees were estab-
lished under the Memorandum to discuss the management of border coopera-
tion between the two countries. The Joint Border Committee is led by Director
General for Public Governance, Department of Home Affairs and the Border
Liaison Committee is led by the Government of Nusa Tenggara Timur Province.
The leading role for boundary negotiation is taken by the Director for Political,
Security and Territorial Affairs of the Department of Foreign Affairs, assisted
by the Head of the Technical Delineation and Demarcation Team Indonesia-
Timor-Leste, Mr. Sobar Sutisna from National Coordinating Agency for Survey
and Mapping.
Both Governments took this opportunity to sign a Memorandum of Understand-
ing between the Government of the Republic of Indonesia and the Government
of the Democratic Republic of Timor-Leste concerning access to Pulau Batek
or Gala Bata Island. With the signing of the Memorandum, Indonesia provides
access for the people of Timor-Leste living in the border area of Ocussi, to
pay traditional and social-cultural visit to Pulau Batek or Gala Bata Island. In
turn Timor-Leste recognizes Indonesia’s sovereignty over the island. Technical
arrangement for this matter will be further discussed by the two countries.
The map that is Annex B to the Provisional Agreement is drawn at a scale of
1:125.000. The boundary is marked to show where the boundary follows rivers
and terrain features. Along some rivers, such as Motta (River) Massin and Motta
Taloe in the east and Noel (River) Bilomi in the west the boundary follows the
median line. The three unresolved boundary segments can now be considered
in descending order of extent. The largest area in dispute is bounded by two
distributaries of the Noel Besi. It was created by the following definition of the
land boundary’s coastal terminus.
Proceeding from the mouth of the Noel [River] Besi, from where the sum-
mit of Pulu [Island] Batek can be sighted on a 30° 47’ NW astronomical
azimuth . . . (Deeley, 2000, 70)
Defining a point in two ways is usually fatal to the satisfactory delimitation of a
boundary. In this case there are two channels discharging into the sea but only
one has the bearing to the Indonesian island recorded in the Agreement. The two
channels differ markedly in width and discharge and it is the smallest sometimes
intermittent channel that reaches the coast at a bearing close to that nominated in
the 1904 Treaty. Timor Leste regards the western, minor channel as the proper
boundary while Indonesia regards the eastern major channel as the boundary.
The area bounded by the two channels and the sea is about 9.5 sq. km.
The disputed area on the Oecussi boundary at Bijael-Sunan Oben measures
about 2 sq. km. It is understood that during the 1960s there was some border
conflict in this region. The unresolved area on the eastern boundary near Memo
is the smallest of the three territories. The disputed area seems to involve the
headwaters of the Motta Malaibaka; it has a length of about 1.9 km and a maxi-
mum width of 250 metres.
Land Boundaries on Islands Off Southeast Asia 377
The boundary separating Indonesia and Papua New Guinea lies close to the limit
of Dutch annexations in the first half of the 19th century. Van der Veur (1966)
published a useful account of the early history of this line. In 1828 the Dutch
having made vague claims to New Guinea, on the strength of the jurisdiction of
the Sultan of the Moluccas, decided to take direct action. An expedition was sent
to the island and a small settlement was established at Triton Bay after the coast
had been explored from Tandjung Jamursba to the Digul River. These actions
were followed by the proclamation of Dutch territory on 24 August 1828.
That part of New Guinea and its interior, beginning at the 141st meridian
east of Greenwich on the south coast, and from there west-northwest and
northward to the Cape of Good Hope [Tandjung Jamursba], situated on
the north coast. (Van der Veur, 1966, 10)
After Britain requested a clear statement of the extent of territory occupied by
tribes subject to Dutch authority, a secret definition was made in 1848.
From Cape Saprop Maneh [Tanjung Djar] 140° 47’ meridian east of
Greenwich on the north coast, along that coast, the Bay of Wanammen
[Teluk Sarera] to Cape Kain Kain Beba [Tanjung Jamursba] and further
west, south and southeast to the, by Proclamation of 24 August 1828, pro-
visionally adopted boundary at 141° E.L. on the south coast; including the
interior, so far as this . . . will appear to belong to Netherlands territory.
(Van der Veur, 1966, 12)
The secret declaration was made public in 1865 and then in 1875 the eastern
boundary was described as a straight line from Tanjung Djar on the north coast
to the intersection of meridian 141° east on the south coast. Thus the alignment
of the present boundary was fixed fairly closely by a unilateral Dutch declaration
before Britain and Germany acquired their respective colonies in the east of New
Guinea. They both claimed colonies in 1884 and they were separated in April
1885 by a boundary consisting of straight lines from Mitre Rock on the east
coast of New Guinea, 0.6 nm north of Cape Ward Hunt, that is at 8° 03’ south
latitude, to the intersection of meridian 141° east and parallel 5° south.
The Anglo-Dutch negotiations commenced in 1893 after Britain had complained
about raids of headhunters from Dutch territory east of Longitude 141° east.
Officials visited the coast and selected the mouth of the Bensbach River as the
starting point. The mouth was located at 141° 01’ 47.9’ east, that meant Britain
was conceding a strip of territory about 3 km wide to the Anglo-Dutch-German
tri-junction. To offset this concession Britain suggested that the meridian should
be intersected by a section of the Fly River that lay west of the meridian. This
was recommended on the grounds that explorers and gold-seekers would use
the Fly River as a route inland and it would be inconvenient if part of the route
378 Chapter 13
fell entirely within Dutch territory. This justification was accepted by the Dutch
negotiators and the boundary was settled on 20 July 1895.
Germany was in no hurry to settle the boundary with the Dutch north of
parallel 5° south and it was 1910 before negotiations began. German surveyors
believed that since the coastal terminus had been decided and the boundary was
the meridian through that point there was no need to mark the boundary inland
because white settlement was improbable. It is likely that German surveyors were
working hard around the perimeters of Cameroun, Tanganyika and Southwest
Africa in the first decade of the 20th century. When work did start the Dutch
wanted a detailed survey of a wide strip to allow the selection of a natural bound-
ary. The Germans argued successfully for the survey of a narrow strip within
which a meridian could be selected. The surveys produced useful results but war
intervened and by 1918 Australia was governing the former German colony.
Cook, Macartney and Stott (1968) have produced an excellent account of the
evolution of this boundary after 1919. In 1928 A.G. Harrison, an Australian sur-
veyor placed a marker on the north coast near Wutong on meridian 141° 0’ 13.5”
east, and recorded that the Dutch boundary lay 400 m to the west. In 1933 sur-
veyors from both countries discovered there was a gap of 398 m between the
Dutch and Australian determinations of the terminus. It was decided to split the
difference but the selected point was unsuitable for the placement of a marker.
Accordingly the marker was placed 168 m from the Dutch determination and
230 m from the Australian determination. In 1936 the two governments agreed
the boundary should be the meridian through that marker.
In 1960 the Australian and Dutch governments agreed that the boundary
should be the great circle course passing through the obelisk on the north coast
and the most northerly point where the great circle course intersected the Fly
River. South of the river the boundary would be the meridian that passed through
the mouth of the Bensbach River. A new determination of that mouth had been
made in 1958 and produced a meridian 141° 01’ 07” east. The two intersections
of meridians on the Fly River were completed in 1962.
In 1964, after Indonesia succeeded the Netherlands in Papua, Australia and
Indonesia had to agree on the boundary. It was decided that north of the Fly
River the boundary would be 141° east and south of the river it would be merid-
ian 141° 01’ 10” east, that intersected the mouth of the Bensbach River. The
boundary was demarcated in two years starting in 1966. Ten beacons were placed
north of the Fly River and four marked the line south of the Fly River. The final
agreement was signed on 26 January 1973.
It is evident from large-scale maps of the area, where the straight boundaries
intersect the Fly River, that the river contains many meanders that have some-
times have been cut off to form ox-bow lakes. It is possible that the Fly River
will move and intersect the straight boundaries north or south of the two limiting
monuments erected in the 1960s. However, there have not been any reports of
such events creating serious problems.
Land Boundaries on Islands Off Southeast Asia 379
Borneo is the world’s third largest island with an area of 484,330 sq. km and it
is shared by three states. Indonesia, with an area of 281,019 sq. km, occupies
the largest area in the south and east. The Malaysian territories of Sabah and
Sarawak occupy 203,311 sq. km of the northernwestern and northeastern shores.
Brunei has an area of 2,226 sq. km on Sarawak’s north coast. It is the only state
located entirely on Borneo and the only one divided into two parts. The island
straddles the equator and the borderland between the Indonesian and Malaysian
territory coincides with folded mountains of sandstone and granite with the
highest peaks of 5,125 m.
Irwin (1955) opened his brilliant account of diplomatic rivalry in 19th century
Borneo with a quotation from the Englishman John Hunt.
The island of Borneo extends from 7° 7’ north to 4° 12’ south latitude and
from 108° 45’ to 119° 25’ east longitude; measuring its extreme length 900
miles, at its greatest breadth seven hundred and in circumference three
thousand. . . . Situated in the track of the most extensive and valuable com-
merce, intersected on all sides with deep and navigable rivers, indented
with safe and capacious harbours, possessing one of the richest soils of the
globe, abounding with all the necessaries of human life, and boasting com-
mercial products that have in all ages excited the avarice and stimulated
the desires of mankind – with the exception of New Holland, it is the largest
island known. (Hunt, 1812)
Irwin then proceeds to show that in the next eighty years Hunt’s predictions
were totally unjustified. The rivers had only shallow mouths; only in the north
were there good harbours; the local population was savage and the soil becomes
leached when the forest is cleared. After the Napoleonic wars ended Britain and
the Netherlands looked at Borneo from different perspectives. Britain’s strategic
view was that northern Borneo protected the eastern flank of maritime trade
routes to China. The Netherlands thought of Borneo in relation to the Dutch
Empire based on Java.
Dutch colonies had been established on the south coast of Borneo at the begin-
ning of the 17th century. Britain’s interest developed in the 1840s when James
Brook arrived in Sarawak, and in return for assistance to the Sultan of Brunei
he was given a grant of land. At that time the Sultan of Brunei was nominally
sovereign over the whole territory that now comprises Brunei, Sarawak and
Sabah. However, he had great difficulty in collecting taxes from his subjects
beyond the immediate neighbourhood of Brunei, and his kingdom had been
decaying since the middle of the 17th century. This process accelerated after
the arrival of Brooke and became a headlong decline after 1881, when the Brit-
ish North Borneo Company was given a Royal Charter and began to extend its
influence throughout Sabah. Brunei’s territory was whittled away as Sarawak
380 Chapter 13
and the Company competed with each other and advanced towards a common
boundary as they secured the cession of valley after valley. Sarawak enjoyed
greater territorial success and its purchase of the Terusan valley in 1884 and the
Limbang valley in 1890 ensured that Brunei was divided into two parts. Tarling
(1971) and Wright (1970) provided detailed accounts, of the British occupation
of northern Borneo.
Britain’s authority in north Borneo was confirmed when protectorates were
established over Sarawak, Brunei and North Borneo in 1888. Britain then turned
to the Netherlands, that claimed the rest of Borneo, to delimit boundaries between
the British and Dutch territories. The first Anglo-Dutch boundary was settled on
20 June 1891 (Figure 13.3). The original cession, secured by Overbeck and Dent
from the Sultan of Brunei in 1878, named the River Sebuku as the southern limit
on the east coast. However, the British officials heading south towards the River
were frustrated by pirates in Darvel Bay, and this allowed the Dutch to wait
for them at 4° 10’ North on the island called Sebatik. The island was divided
by the parallel and on the mainland the boundary proceeded northwest, so that
the Simengaris River was left to Dutch Borneo. At the intersection of parallel
4° 20’ North and Meridian 117° East the boundary swung westwards towards
the main range separating rivers flowing to the coasts of Sabah and Sarawak
from those flowing to the south and west coasts of Borneo. The definition of
the section of the boundary leading to the main watershed along parallel 4° 20’
North was confusing.
. . . in the event of the Simengaris River or any other river flowing into the sea
below 4° 10’ North, being found on survey to cross the proposed boundary
within a radius of 5 geographical miles [8 km], the line shall be diverted so
as to include such small portions or bends of rivers within Dutch territory:
a similar concession being made by the Netherland Government . . . (British
and Foreign State Papers, 1891-2, 42)
It is unclear whether this condition refers to rivers flowing across the boundary
that have their source within 8 km of the crossing, or to rivers that cross the
boundary more than once where those crossings are less than 8 km apart. The
western terminus of the Anglo-Dutch boundary was located at Point Datu, a
well-known coastal landmark. The boundary along the watershed was based on
geographical faith, because the courses of the rivers originating in the mountains
had not been established by survey.
In 1905 there was a disagreement between the Dutch and North Borneo offi-
cials about the course of the boundary close to the east coast. The Company
officers believed the the boundary followed parallel 4° 10’ North due west before
swinging northwestwards towards the intersection of parallel 4° 20’ North and
meridian 117° east. The Dutch objected that such a boundary would intersect
the Simengaris River close to the coast, and insisted that this difficulty had been
foreseen and avoided by the negotiations in 1891. The British Government agreed
with the Dutch view and the Company was advised accordingly.
Land Boundaries on Islands Off Southeast Asia 381
The section of the boundary from the east coast to Moeloek Mountain was
demarcated in 1912-13 and the description was embodied in an agreement signed
on 28 September 1915. The boundary was shown on an attached map at a scale
of 1:500,000. The map also records two lines 5 km north and south of the bound-
ary noted in the 1891 treaty. This indicated that the surveyors were concerned
with the sources of rivers that crossed the boundary. Their interpretation of the
boundary, where major rivers crossed parallel 4° 20’ North, but had a source
more than 8 km from the parallel, was not consistent. It must be presumed that
negotiators made mutual concessions. Pillars were erected on parallel 4° 20’
North on the banks of the Pensiangan, Agisan and Sebuda Rivers.
The third and final Anglo-Dutch boundary treaty was signed on 26 March
1928. It concerned a short section of boundary measuring about 30 km between
the Api and Raya peaks. This section of boundary was trending northwards
towards its western terminus, and was cutting across the grain of the low ranges,
that still have a northeast-southwest axis. The watersheds between some rivers
flowing to the coast east and west of Point Datu are low and it is probable that
settlement extending westwards along the valleys had spilled over them, This
appears to be the explanation for the cession, by the Dutch, of about 100 sq. km
in the upper Separan and Berunas valleys. This boundary segment was marked
by fifteen wooden pillars and four cement pillars. Their positions were marked
on a map at a scale of 1:50,000.
The largest section of Brunei is bounded by watershed boundaries. On the
west the watershed is formed by Malaysia’s Balang Baram and Brunei’s Sungai
Belait. The smaller eastern portion of Brunei is bounded on the west by the Sungai
Pendaruan and on the east by the watershed between the Sungai Tamburong, that
belongs to Brunei, and the Batang Trusan that flows through Sabah.
In 1958 the United Kingdom was responsible for Sarawak and North Bor-
neo and had a special relationship with Brunei. On 11 September 1958 Britain
issued The Sarawak (Definition of Boundaries) Order in Council (Charney and
Alexander, 1993, 924-7). The Order in Council defined maritime boundaries
between Brunei and Sarawak that divided the continental margin out to the
100 fathom isobath between Brunei and Sarawak. Three maritime boundaries
were necessary because Brunei consists of two parts separated by the Limbang
Valley that belongs to Sarawak. The central and eastern boundaries were lines
of equidistance. Sarawak’s coastline along the Limbang Valley is shelf-locked.
The western boundary followed a line of equidistance close to the coast but then
diverged along a line perpendicular to the general direction of the coast. This
was done to diminish the influence of the protruding Tanjong [Point] Baram.
This adjustment was to Brunei’s advantage. A Malaysian map entitled ‘Map
showing territorial waters and continental shelf boundaries of Malaysia’, was
published by the Director of National Mapping in 1979. It appears that Malaysia
has accepted the boundaries drawn by Britain, However, Malaysia is likely to
insist that any projection of the western boundary cannot ignore the existence
of Tanjong Baram.
382 Chapter 13
CONCLUSIONS
REFERENCES
BRITISH CLAIMS
On 23 June 1843 British authorities issued details of the arrangements for the
government of the Falkland Islands and their Dependencies (International Court of
Justice, 1956, 41). This proclamation did not define the extent of the Dependen-
cies, nor did subsequent proclamations, governors’ commissions or laws clarify
this question in the period before 1907. However, from 1887 the Colonial Office
Yearbook referred to South Georgia as part of the Dependencies.
In the last decade of the 19th century there was a marked increase in voyages
to Antarctic seas for purposes of scientific research and the capture of whales
and seals. This activity encouraged the British authorities to set out the limits
of the Dependencies in Letters Patent dated 21 July 1908.
Whereas the group of islands known as South Georgia, the South Orkneys,
the South Shetlands. The Sandwich Islands and the territory known as
Graham’s Land, situated in the South Atlantic Ocean to the south of the 50th
parallel of south latitude, and lying between the 20th and 80th degrees of
west longitude, are part of our Dominions, and it is expedient that provision
should be made for their government as Dependencies of our said Colony
of the Falkland Islands . . . (Polar Record, 1948, 241)
According to British sources, South Georgia was discovered, or rediscovered
by Captain James Cook on 17 January 1775. Having named them in honour of
the King and claimed them for Britain, he sailed eastwards and two weeks later
discovered the South Sandwich islands that were named for the First Lord of
the Admiralty and also claimed for Britain. The South Shetland Islands were
discovered by Captain William Smith on 18 February 1819. He claimed them
for Britain in October of that year when he called them New South Britain.
Graham Land was discovered by Captain Edward Bransfield, in company with
Smith, on 30 January 1820. Some American scholars, including Hobbs (1939,
1941) and Martin (1938, 1940), assert that Captain Nathaniel Palmer first sighted
the mainland of Antarctica, but their evidence is shown to be dubious by Gould
(1941) and Hinks (1939, 1940, 1941). According to British sources Captain
George Powell discovered the South Orkney Islands on 6 December 1821, and
he claimed them the following day when he landed on the island to which he
gave the name Coronation.
The British authorities must have been embarrassed to discover that the limits
set out in the 1908 Proclamation included the southern tip of South America. There
had been no intention to claim that area. To remove uncertainty the boundaries
of the British claim were amended on 28 March 1917.
The Dependencies of Our said Colony shall be deemed to include and
to have included all islands and territories whatsoever between the 20th
degree of west longitude and the 50th degree of west longitude which are
Antarctica 385
situated south of the 50th parallel of south latitude; and all the islands and
territories whatsoever between the 50th degree of west longitude and the
80th degree of west longitude which are situated south of the 58th parallel
of south latitude. (Polar Record, 1948, 242)
This definition of the Dependencies was confirmed in The Falkland Islands
(Legislative Council) Order in Council dated 26 November 1948 (Great Britain,
1948, 59). On 26 February 1962 Britain detached part of the Dependencies to
create the colony called British Antarctic Territory.
. . . the British Antarctic Territory means all islands and territories whatso-
ever between the 20th degree of west longitude and the 80th degree of west
longitude which are situated south of the 60th parallel of south latitude . . .
(Great Britain, 1962, 356)
The Dependencies were defined as the remaining area of the 1917 definition.
The boundaries of the Ross Dependency were proclaimed on 30 July 1923 (New
Zealand Gazette, 1923, 1). For some years British authorities had been searching
for a way to extend control over additional areas of Antarctica without display-
ing excessive greed. The Ross Sea was a region of particular concern because
Norwegians were seeking rights to catch whales there (O’Connell and Riordan,
1971, 312, 314). Eventually it was decided to issue an Order in Council under
the British Settlements Act of 1887.
From and after the publication of this Order in the Government Gazette of
the Dominion of New Zealand that part of his Majesty’s Dominions in the
Antarctic Seas, which comprises all the islands and territories between the
160th degree of east longitude and the 150th degree of west longitude, which
are situated south of the 60th parallel of south latitude shall be named the
Ross Dependency. (New Zealand Gazette, 1923, 1)
The boundaries of the Ross Dependency have not been altered since that date.
That part of His Majesty’s Dominions in the Antarctic Seas which comprise
all the islands and territories other than Adélie Land which are situated
south of the 60th degree of south latitude and lying between the 160th
degree of east longitude and the 45th degree of east longitude is hereby
placed under the authority of the Commonwealth of Australia. (Bush, 1982,
Vol. 2, 143)
This announcement fulfilled a plan that had been set out in the report of the
Imperial Conference of 1926 (Bush, 1982, Vol. 2, 100-4). That report identified
six areas of coast suitable for Australian control. The six areas were Enderby
Land (45° to 52° 30’ east), Kemp Land (58° 30’ to 60° east), Queen Mary Land
(86° 30’ to 101° east), an unnamed section (131° to 135° 30’ east), King George
V Land (142° to 153° east) and Oates Land (157° to 159° east). Discounting
Adélie Land, that Britain assumed was reserved for France, and which separated
the unnamed section and King George V Land, there were four other wedges
of territory separating the nominated areas. The largest extended for 30° east of
Queen Mary Land. These four wedges were claimed for Australia by Douglas
Mawson in 1930 and 1931 (Ayres, 1999, 192-206). When the limiting meridians
were announced the western boundary coincided with the eastern limit of New
Zealand’s Ross Dependency.
In 1933 the exact extent of Terre Adélie had not been declared by France. Bush
(1982, Vol. 2, 481 et seq.) has prepared an indispensable collection of documents
connected with national claims to Antarctica and the international arrangements
made for the continent. On 20 December 1911 Britain enquired whether France
claimed any part of Antarctica and was told firmly that a claim to Terre Adélie
was maintained. The French reply of 16 April explained that Captain Dumont
d’Urville’s voyage and claims on behalf of France were reported in various
newspapers, including the Sydney Morning Herald of 13 March 1840.
D’Urville and members of his crews landed on an island off the coast of
Antarctica at about 9.0 pm on 21 January 1840.
Following the ancient and lovingly preserved English custom, we took pos-
session of it in the name of France, as well as the adjacent coast where
thrice had prevented a landing. . . . The ceremony ended, as is mandatory,
with a libation. To the glory of France, of which we were very mindful
at that moment, we emptied a bottle of the most generous of her wines
that one of our companions had had the good sense to bring along with
him. Never was Bordeaux wine called to play a nobler role; never was a
bottle emptied more appropriately. Surrounded as we were on all sides by
eternal ice and snow, the cold was intense. The generous liquor reacted
to our advantage against the harshness of this temperature. All that took
less time than it takes to write about it. Then we all set to work to collect
everything of interest for natural history that this inhospitable land had to
offer. (Rosenman, 1987, Vol. II, 474)
Antarctica 387
On 29 March 1913 Britain advised France that it intended naming part of the
coast of Antarctica King George V Land. The coordinates of the sector were
provided and it was noted that, as Britain understood Terre Adélie extended from
136° east to 147° east. The British claim would not impinge on French territory.
No answer was the stern reply!
By 1933 there were three Anglo-Australian versions of the limits of Terre
Adélie. The 1913 letter specified meridians 136° east and 147° east. The report
to the Imperial Conference in 1926 left an area between 135° 30’ east and 142°
east. The gap left by Mawson’s claims on 5 January 1931 and 13 February 1931
was bounded by meridians 138° east and 142° east.
When France declared the limits of Terre Adélie in a letter dated 24 Octo-
ber 1933, the British authorities were dismayed to discover the the French had
nominated 136° east and 147° east (Bush, 1982, Vol. 2, 498). This claim was
rejected by the British Government and it was suggested that 136° 30’ east and
142 east were the proper limits of Terre Adélie because they were shown on a
chart based on d’Urville’s work published in 1840. In a reply dated 5 October
1936 France trumped the French chart with the British letter of 1913. It was also
remarked that the 1840 chart had been available for more than 60 years when
Sir Francis Bertie wrote to M. Pinchon, and that it was impossible to understand
why the charts were now being interpreted in a different way. Although France
held stronger diplomatic cards it offered a compromise that would leave the west-
ern boundary at 136° east, while the eastern limit would be settled somewhere
between 142° east and 147° east.
It took the British Foreign Office a year to reply, an unsurprising delay at that
period in European history. On 13 October 1937 Britain rejected the French claim
for a boundary east of 142° east and suggested that the maximum extent of Terre
Adélie was 136° east to 142° east. This letter also pointed out that in 1913 Lord
Bertie had relied on the Sydney Morning Herald of 13 March 1840 for a definition
of Terre Adélie, and that the paper had printed 147° east in mistake for 142° east.
The correct coordinates 136° east and 142° east were contained in the French
report, published originally in a Hobart journal of 3 March 1840. Remarkably the
French accepted this laughable explanation, and agreed to the British definition
of Terre Adélie on 5 March 1938. Appropriately the decree giving effect to this
agreement was published quickly by both parties on 1 April.
The islands and territories situated south of parallel 60° south latitude and
between meridians 136° and 142° east of Greenwich are within the jurisdic-
tion of French sovereignty. (Bush, 1982, Vol. 2, 505)
These limits have persisted since that proclamation.
The British authorities explanation of the change in their opinion between 1913
and 1933 was not soundly based. The account in the Sydney Morning Herald
appeared on 20 March 1840; it was based on a newspaper published in Hobart
and it did cite 147° east. The Hobart newspaper was the Austral-Asiatic Review:
388 Chapter 14
Tasmanian and Australian Advertiser of 3 March 1840. When this journal was
consulted it was found to include a reference to 147° east, and so no misprint
was involved.
NORWAY’S CLAIM
Norway’s first claim to Antarctic territory was made on 23 January 1928, when
a Royal decree asserted sovereignty over Bouvet Island, located at 54° 26’ south
and 3° 24’ east. It lies about 1600 km north of Antarctica. Britain objected on the
grounds that the Island had been discovered by the Frenchman Pierre Bouvet in
1739 and re-discovered by Captain Norris, a British captain in December 1825.
Eventually British withdrew its objection and Norway’s claim was reiterated in
a law on 27 February 1930.
The Bouvet Island is placed under Norwegian sovereignty. (United States
Naval War College, 1950, 238)
A year later on 1 May 1931, a Royal Proclamation announced that Peter I Island
had been claimed by Norway.
We, Haakon, King of Norway, make known: Peter I Island is placed under
Norwegian Sovereignty. (United States Naval War College, 1950, 239)
Peter I Island off Ellsworth Land in latitude 68° 55’ south and longitude 90° 50’
west was discovered by a 1929 Norwegian expedition.
Although Norwegian explorers, such as Christensen and Mikkleson, discovered
about 2080 nautical miles of the Antarctic coast and mapped from the air about
80,000 sq. km of the continent, in the period from 1926 to 1937, no formal claim
was made by Norway to the sector between the British and Australian claims.
In 1938 and 1939 German authorities sent the Schwabenland, commanded by
Alfred Richster, to stake a claim to the continent in the vicinity of the Greenwich
meridian. Planes were used to photograph 350,000 sq. km of territory during
flights totalling 12,000 km. The German flag was dropped every 25 km to sup-
port the claim that was planned. When the Norwegian authorities learned of this
activity a formal claim was quickly drafted. After receiving a recommendation
from the Ministry of Foreign Affairs on 14 January 1939, the King issued a
proclamation on the same day.
That part of the mainland coast in the Antarctic extending from the limits of
the Falkland Islands Dependency in the west [the boundary of Coats Land]
to the limits of the Australian Antarctic Dependency in the east [45° east
on Enderby Land] with the land lying within this coast and the environing
sea, shall be brought under Norwegian sovereignty. (United States Naval
War College, 1950, 239)
Antarctica 389
Norway is the only claimant state in Antarctica that has not claimed an entire
sector bounded by converging meridians that meet at the South Pole. Norway
eschewed such a claim because sector claims by its neighbours in the Arctic
Ocean would put it at a marked disadvantage.
A decade earlier it had seemed possible that Norway would claim a larger area
than was defined in the 1939 Proclamation. In 1929, when Commander Byrd
was making the first of three expeditions to Antarctica, the Norwegian authorities
informed the United States of those parts of the continent that were excluded
from possible American claims because of Norway’s prior discoveries.
. . . the territory immediately circumjacent to the South Pole, which, as will
be known, was taken possession of in the name of the King of Norway by
Captain Roald Amundsen in December 1911, under the name of Haakon
VII’s Plateau, nor to compromise the territories on both sides of Captain
Amundsen’s route to the South Pole, south of Edward VII’s Land and includ-
ing Queen Maud’s Range. (United States Department of State, 1929, 717)
The Norwegian note to the American representatives observed that though it
was not intended to claim sovereignty at that time, Norwegian authorities were
convinced that all requirements to justify such a claim had been satisfied.
Finally the overlapping claims by Argentina, Britain and Chile to the Antarctic
Peninsula can be considered together. These claims centre on the Antarctic Penini-
sula, It is a narrow sinuous, mountain range with peaks to 3,050 m that reaches
closer to another continent, South America, than any other part of Antarctica.
The northern tip of the Peninsula lies 430 nm from Cape Horn, the northern edge
of Drake Passage, and is shielded by the South Shetland Islands.
On 21 July 1908 Britain issued Letters patent appointing the Governor of
the Colony of the Falkland Islands to be Governor of South Georgia, the South
Orkneys, the South Shetlands, the Sandwich Island and Graham’s Land (Christie,
1951, 301-2). Graham’s Land is now known as the Antarctica Peninsula. The
claim was defined by parallel 50° S and meridians 20° W and 80° W. Unfor-
tunately when these lines were drawn on a map they included the southern
tip of South America! A new set of Letters Patent were issued on 28 March
1917 defining the territories as lying south of parallel 50° S between meridians
20° W and 50° W, and south of parallel 58° S between 50° W and 80° W (Polar
Record, 1948, 242). This definition of the British Dependencies was confirmed
by the Falklands Islands Order in Council of 26 November 1948 (Great Britain,
1948). Finally on 26 February 1962 Britain detached part of the Dependencies
to create the Colony called British Antarctic Territory.
. . . The British Antarctic Territory means all islands and territories between
the 20th degree of West longitude and the 80th degree of west longitude
which are situated south of the 60th parallel of south latitude . . . (Great
Britain, 1962, 356)
390 Chapter 14
Although Chile claims that its title to Antarctica stems from the award of Hoz
in 1539, it seems the next piece of evidence relates to the concession awarded
to the Chilean Company Fabry-de Toro Herrerra on 27 February 1906 (Christie,
1951, 279-80). The company was authorised to exploit the animal and mineral
wealth of the area including Shetland and lands situated further south. Importance
was also attached to the activities of the whaling company Sociedad Ballenera
de Mallaganes formed five months later. However this Company operated under
a licence granted by the Governor of the Falkland Islands (Polar Record, 1956,
148). In 1916, Shackleton made a succession of unsuccessful attempts to rescue
his stranded crew on Elephant Island, Eventually, the daring rescue was achieved
by Lieutenant Pardo of the Chilean Navy, commanding the tug Yelcho on 30
August (Shackleton, 1991, chapter XV).
In 1906 Chile launched a preliminary investigation into the reasonable Antarctic
claims it might pursue. In 1939 Professor Julio Escudero, who played an important
role in developing the philosophy of the Antarctic Treaty, completed the Chilean
investigation. The Chilean claim was announced on 9 November 1940.
All lands, islets, reefs of rock, glaciers (pack-ice), already known or to be
discovered, and their respective territorial waters, in the sector between
longitudes 53° and 90° west, constitute the Chilean Antarctic or Chilean
Antarctic territory. (Bush, 1982, Vol. 2, 311)
The western limit of 90° west corresponds to no Chilean territory. The island
called Juan Fernandez lies about 80° west and Easter Island is in the vicinity
of 110° west. Possibly 90° west was selected because it was as close as Chile
could claim to Peter I Island, that had already been claimed by Norway. The
island lies 10 nm west of meridian 90° west. The eastern limit of Chile’s claim
does not correspond to any Chilean possession outside Antarctica, but it lies
25 nm east of Clarence Island, the easternmost feature of the British South
Shetland Islands. This group has been claimed by Chile.
Argentina responded to Chile’s claim in three days! Among the points made
was the belief that Argentina could have justly issued a declaration of the same
class ‘. . . did it not think that because such a declaration would be unilateral it
would not have improved such rights and titles in any way’ (Bush, 1982, Vol. 1,
609). This lofty position was soon abandoned and Argentina’s claim in Antarc-
tica was made clear in a map produced by the Instituto Geographico Militar
(Bush, 1982, Vol. 1, 610). This map showed the limits of Argentina’s claim to
be meridians 25° and 75° west and parallel 60° south. The Chilean authorities
must have been surprised by the western boundary, since in 1906 Argentina had
asserted that Chile was not entitled to claim any territory east of meridian 67°
west which passes ‘through Cape Horn’.
In a note to the British authorities dated 15 February 1943 Argentina announced
a new western boundary along meridian 68° 34’ west. This meridian was first
defined as a boundary dividing Tierra del Fuego in the agreement between
Antarctica 391
Argentina and Chile on 23 July 1881. This claim was reiterated in a further note
to the British representatives on 3 June 1946.
To this is added its [Argentina’s] indisputable rights to the lands situated
south of the 60th parallel between the meridians 25° and 68° 34’ west
longitude. (United States Naval War College, 1950, 222)
Five months later a new western boundary was published on a map produced
by the Instituto Geographico Militar. It showed the boundary along meridian
74° west, that lies just west of the Mountain Cerro Bertrand, the most westerly
point of Argentina. On 28 February 1957 Argentina re-established the national
territory of Tierra del Fuego, Antarctica, the Islands of the South Atlantic, and
defined the Antarctic sector as follows.
. . . The Argentine Antarctic Sector contained within the meridians 25° west
and 74° west and the parallel 60° south. (Bush, 1982, Vol. 2, 26)
That claim has not been changed.
Chile did not set a northern limit to its territorial claim in its original dec-
laration in 1940. There have been subsequent clarifications. On 11 June 1961
Chile defined he boundaries of its departmentos, comunas subdelegaciones and
districts. In the Departmento de Terra Magallanes the Comuna Subdelegaciones
de la Antarctica lay south of the Bellingshausen Sea and Drakes passage. It was
divided into districts. Piloto Pardo is bounded by Drake’s Passage to the north,
Bransfield Strait to the south, and meridians 53° and 64° west. Tierra de O’Higgins
occupies the remainder of the sector. On some Chilean maps the location of the
northern boundary of districts through Drake’s Passage and the Bellingshausen
Sea coincides with parallel 60°.
The Antarctic Treaty was drafted in December 1959 following the successful
period of international cooperation during the International Geophysical Year.
By the middle of 1961 the 12 countries named in the preamble to the treaty had
ratified it.
These states pledged themselves to preserve Antarctica for peaceful purposes,
to foster unhindered scientific activity in the continent, and to preserve and con-
serve the living resources of the continent. The provisions of this succinct Treaty
apply to the area south of 60° south, including all ice shelves, but the rights of
states in respect of high seas south of 60° south were not affected.
In effect these twelve states appointed themselves trustees of Antarctica for
the international community, and they specifically stated in the Treaty their con-
viction that it furthered the purposes and principles embodied in the Charter of
392 Chapter 14
the United Nations. Articles 9 and 13 made arrangements for other countries to
accede to the Treaty and this has been done by a number of countries.
The seven sector claims to territorial sovereignty in Antarctica have been
described in chapter 5. This part of the chapter considers the legal questions
prompted by the boundaries delineated under these claims at public international
law:
• What is the legal validity of the territorial and, hence, the boundary claims
in Antarctica?
• What is the impact of the Antarctic Treaty 1958 and its interlinked conven-
tions on the current and future validity of these boundaries?
• Does the concept of the ‘common heritage of mankind’ have any contem-
porary legal effect on the claimed boundaries in Antarctica?
• What is the legal impact on Antarctic boundaries of the work of the UNCLOS
Commission on the Limits of the Continental Shelf (CLCS)?
The legal validity of Antarctica has not been tested by an international tribunal;
the attempt by the UK to resolve its overlapping claims with Argentina and
Chile before the ICJ failing when the matter was struck out of the Court’s docket
for lack of jurisdiction (Antarctica UK v Argentina, ICJ Reports 1956). In the
unlikely event that the validity of these sovereignty claims was to be amenable
to international judicial determination, a tribunal would need to examine each of
the roots and evidences of title that have been discussed in Chapter 5.
DISCOVERY
CESSION
The Antarctic claims of Australia and New Zealand rest in the first analysis
upon cession of title by the United Kingdom. The UK passed title to Australia
by Imperial Order in Council of 7 February 1933 and to New Zealand in 1923
It is axiomatic, as Judge Huber pointed out in the Island of Palmas case, that a
state cannot ‘transfer more rights than she herself possessed’ (RIAA 842). The
validity of the titles received by New Zealand and Australia in 1933 is thus
dependent upon whether the United Kingdom had a valid title to the Antarctic
territories at the time of the transfers. A definitive answer to this question is not
necessary, however, as both New Zealand and Australia found their claims today
on acts of effective occupation.
As has been described above, each of the claimant states, with the exception
of Norway, has delimited its boundary by reference to lateral boundaries that
converge at the South Pole; a technique described as the ‘sector principle’. This
Antarctica 395
Russia has adopted the sector principle as a root of title in the Arctic, stating
its legal position in the Sturman Declaration of 20 September 1916 in which it
annexed certain islands to the north of Siberia on the ground that they formed
‘a northern extension of the Siberian continental upland’ (Triggs, 92). In April
1926, the Soviet Union consolidated its position by declaring that ‘all lands,
either known or unknown, within a triangle formed by two meridians of longitude
starting from the eastern and western boundaries of the territory already held and
continuing until they met at the North Pole, was now Soviet territory’. (Triggs,
396 Chapter 14
1986, 92) Recent and dramatic acts asserting sovereignty over the North Pole
appear to be based on the extension of Russia’s continental shelf to the pole, as
contained within the described sector. (The Times, 28 July 2007, at 39; Financial
Times, Monday 20 August 2007, at Analysis 9)
The concepts of hinterland, regions of attraction and contiguity lend some credence
to the sector principle as a root of title in the Arctic as Russia, Canada, Denmark
and Norway are geographically proximate to their Arctic island territories. By
contrast, the sector principle as a root of title has only the most tenuous role in
the Antarctic where the seven claimant states lie significant distances from the
South Pole. Australia is for example 2,000 miles from the nearest point on the
Antarctic mainland. Moreover, the current Antarctic sector lines to the South
Pole do not reflect the current territory of Australia, United Kingdom, France,
Chile or Argentina in the Southern Hemisphere. Norway has sovereignty over
Bouvet Island in the South Atlantic and Peter I Island in the South Pacific,
just off Antartica’s continental margin. Neither island is sufficiently substantial
to support a sector claim founded upon contiguity. Any argument founded in
socio-economic needs of a contiguous territory or peoples, along the lines of the
Anglo-Norwegian Fisheries case must fail. (Triggs, 1986, 93)
The sector principle as a root of title, as distinct from a means of identifica-
tion of the limits of sovereignty, appears to have been first applied to Antarctica
in a Dispatch of 23 December 1929 by the British Minister for Foreign Affairs
to the Norwegian Minister of Foreign Affairs in which he stated that Great
Britain had:
. . . unimpeachable rights to the whole of these sectors, including all land down
to the South Pole, an extension of which was looked upon as the inseparable
hinterland of the coastal territory in each sector. (Triggs, DATE 89)
No response to this extensive claim is recorded.
Most Antarctic claimants have rejected the sector principle as a root of title.
The United Kingdom employed sectors to define the Falkland Islands sector in
1917 and the Ross Dependency in 1923 but subsequently appears not to have
supported Argentina’s attempts to support its claim on the basis of geographical
considerations. Similarly, France and Australia delimit their claims by sectors
but do so as a means of delimitation only. While the Soviet Union, and now
presumably Russia, Norway and Germany have rejected the sector principle as a
root of title, the United States, Denmark, Argentina, Chile and France are either
equivocal or support their claims in part on the grounds of geographical unity.
Antarctica 397
RECOGNITION
Five of the claimant states mutually recognize their territorial claims, Australia,
Britain, France, New Zealand, and Norway. (Bush, 1982) However, the claims of
Argentina, Britain and Chile overlap in the Antarctic Peninsula and the boundar-
ies of their claims are disputed amongst them. Of greater legal significance are
the positions adopted by USA and Russia, which have strong potential claims
to sovereignty in Antarctica, but have chosen not to assert any territorial rights
and reject those claims that have been made. Article 4 (2) exists to protect their
right to make a claim if they chose to do so in the future. Other members of the
ATS also do not recognize the validity of claims. Argentina, for example, does
not recognize any claim other than its own.
Moreover, most states outside the Antarctic Treaty regime do not recognize
any Antarctic claims to sovereignty, the legal consequence being that they treat
the maritime zones in the Antarctic as part of the high seas and the seabed as
within the area as defined by UNCLOS. (Prescott and Schofield, 2005, 536) While
recognition by international organizations such as the specialized agencies of the
UN can also be of evidentiary value in supporting title, the General Assembly
and the Security Council have taken no position on the question of Antarctic
sovereignty. Antarctica was, moreover, excluded from the ambit of the Law of
the Sea negotiations as a potential impediment to reaching a consensus.
398 Chapter 14
PROTESTS
Third States
There are 44 ratified parties to the Antarctic Treaty which represent over 80 per
cent of the world’s population. The inescapable fact remains that the overwhelm-
ing majority of the nations in the world, 148 to be precise, are not parties to the
Antarctic Treaty and its interlinked conventions and recommendations. The risks
posed to the stability of the Antarctic Treaty regime is potentially that it could
be ignored by the rest of the international community.
The effectiveness of the Antarctic Treaty system rests on the creative, though
deliberately obscure Article IV (1). This provision was intended to preserve the
Antarctica 399
and the need for nuclear waste disposal sites, become more pressing. While it is
unlikely in current circumstances that there will be any serious challenge to the
Antarctic Treaty system, it remains valuable to explore the future implications
of termination of the Treaty for claimant and non-claimant states alike.
International law recognizes that a treaty cannot create rights or obligations for
non-party states without their consent (VCLT, Art. 34). On this basis, it is legally
possible that a Party to the Antarctic Treaty could use their activities during the
life of the Treaty to consolidate a claim to title against third states, albeit that to
do so would be in breach of its obligations to other Parties.
It is also possible, that the Antarctic Treaty has created a form of ‘objective’
regime that is capable of creating rights and obligations for third parties erga
omnes, including the benefit of protection by Article IV. Examples of treaties that
have successfully created rights for all states include the Berlin Act of 1885, which
created a right to free navigation through the Congo and Niger rivers, the Suez
Canal Convention of 1888, establishing a right of free passage for non-signatories
and the Treaty of Versailles that asserted the Kiel Canal was ‘free and open’
to all vessels. Similarly, treaties of neutralization or de-militarization appear to
have created regimes for all, the international status of the Aaland Islands under
the Convention of 1856 being a prime example. International status has been
recognized by the ICJ in the mandate established by the League of Nations in
the Southwest Africa Case. The UN Charter was also considered by the ICJ in
the Reparations Case to have created an international legal personality that can
affect the rights of third states.
The ILC was, nonetheless, doubtful about the concept of an objective regime
and no provision to this effect was included in the VCLT. Waldock, in a minority
on this issue, proclaimed that it was clear to him that the Antarctic Treaty was
intended to create an objective regime. Certainly, there are several provisions of
the Treaty that can be adduced to support Waldock’s view. The better position
remains, however, that the Antarctic Treaty was not intended to create legal rights
or obligations for third states. Article IV provides potent support for this view
and the subsequent practice of states parties confirms it (Triggs, 1986, 146).
402 Chapter 14
CUSTOMARY LAW
More probable than the creation of an objective regime by the Antarctic Treaty
system itself, is the possibility that it might have, over time, come to reflect
customary law. Indeed, it is well recognized that a treaty can have a law-making
effect. While the ILC rejected the idea that a treaty can create an objective regime
in the sense of creating rights and obligations for non-parties, they preferred to
consider that a treaty can stimulate a wider consensus of states, thereby creating
a customary norm. It is thus arguable that the practices of the Antarctic Treaty
Parties and of the international community have created a ‘habit of cooperation’
that has crystallized into a binding norm. Several practices including non-military
activity, free access for scientific activities and regulation of tourism, suggest that
they have become accepted customary norms. Article IV remains, nonetheless, a
powerful impediment to any assertion that territorial claims have been prejudiced
in any way by the Treaty. Rather, Article IV has been steadfastly retained by
the parties throughout all their interlinked agreements and recommendations, in
order to protect their respective juridical perspectives.
The idea that normative rules have evolved through the Antarctic Treaty regime
has been subsumed to some extent by the wider concept of the common heritage of
mankind, described by some as a common space, world park or internationalized
territory, or an area beyond the limits of national jurisdiction and thus within the
mandate of UNCLOS. In essence, the notion of a common heritage imports the
principle that resources cannot be exploited unilaterally, but are subject to joint
management, communal ownership and sharing of any product on an equitable
basis. A core aspect of common heritage as it might apply to Antarctica is that
the continent is not terra nullius and cannot be subject to exclusive claims to
territorial sovereignty by any one state. Common heritage thus strikes at the
heart of national claims and denies legal validity to the boundaries that have
been delineated along sector lines.
Article 136 of the Law of the Sea Convention declares the deep seabed to be
the ‘common heritage of mankind’ and the outer-space is similarly not subject
to national appropriation . . . The notion of common heritage does not have any
clear legal meaning beyond the intention of the parties to those special treaties
that have employed it.
In a speech to the UN General Assembly by Dr Mahithir, Malaysia’s Prime
Minister, on 29 September 1982, it was suggested that the concept of common
heritage should also apply to Antarctica. That Antarctica and its mineral resources
should be subject to a similar regime as had been negotiated for the deep sea-
bed, was an idea whose time appeared to have arrived. Many developing states,
Antarctica 403
particularly the Group of 77, were concerned that Antarctica was controlled by
a few self-interested states with Consultative Party status under the Antarctic
Treaty (Rothwell, 1996, 106). Such fears were compounded by the negotiation
of the Minerals Convention in 1989. The UN General Assembly requested the
Secretary-General to conduct a study of the Antarctic Treaty System between
1983-5 and the Question of Antarctica remained on the UN agenda for many
years. The intensity of the argument that Antarctica was rightfully the common
heritage of mankind has, however, diminished as negotiation of the Madrid Pro-
tocol has ensured that minerals exploitation is prohibited, at least for 50 years,
and that a comprehensive environmental regime is now in place. The continued
relative success of the management regime in Antarctica provides the best guard
against adoption of the common heritage a concept.
As a matter of legal analysis, it is doubtful that Antarctica has the legal status
of common heritage. Unlike the Treaty on Principles Governing the Activities
in Outer Space 1967 and the Moon Treaty 1979, the Antarctic Treaty and its
related conventions do not explicitly adopt this concept. It is also unlikely that
the continent has acquired the status of common heritage through custom as the
international community has not, in fact, applied the concept to Antarctica.
SUMMARY
The better view is that international law does not yet recognize the right of treaty
parties directly to create legal obligations for third states, but that treaties can in
time be recognized by the international community as articulating a new rule of
customary law that is binding on all states.
While this work does not include analysis of maritime boundaries, as the findings
of international courts and tribunals on maritime delimitation are fully considered
elsewhere, some attention might usefully be given to the work of the United
Nations Commission on the Limits of the Continental Shelf established under Art.
76 (8) of UNCLOS. In effect, the work of the Commission provides a procedure
through which a coastal state can achieve legal recognition for its claims to an
extended continental shelf beyond 200 nautical miles, thereby establishing the
outer boundary of its shelf.
It is well recognized, both by UNCLOS and customary law, that a coastal
state ‘exercises over the continental shelf sovereign rights for the purpose of
404 Chapter 14
exploring it and exploiting its natural resources’. (Art. 77) The right exclusively
to exercise sovereign rights over the continental shelf is not a territorial right in
the usual sense and does not depend upon effective or notional occupation or
any express proclamation. Nonetheless, the sovereign rights of a coastal state
over the continental shelf arise as a consequence of the ‘natural prolongation of
its land territory to the outer edge of the continental margin.’ (Art. 76 (1)) The
outer limits of the continental shelf may not exceed 350 nautical miles, measured
from the baselines from which the breadth of the territorial sea is measured or,
alternatively, 100 nautical miles from the 2,500 meter isobath. Those negotiat-
ing UNCLOS were alert to the fact that many states claim extensive continental
shelves beyond 200 nautical miles; claims that are by no means accepted by their
neighbours and that often overlap with competing claims. Some mechanism was
therefore necessary to provide certainty to inconsistent or unacceptably extensive
claims.
The functions of the Commission are set out in Annex 11 (3) to the Conven-
tion and they include:
• receiving submissions from coastal states, within 10 years of the coming
into force of the Convention for that state, concerning the outer limits of
the continental shelf in areas beyond 200 nautical miles
• making recommendations in accordance with Article 76 and the Statement
of Understanding adopted on 29 August 1980 by the Third UN Conference
on the Law of the Sea.
Where the limits of a state’s continental shelf have been established by a coastal
state on the basis of recommendations by the Commission, they will be final
and binding (Art. 76 (8)). As most members of the international community
have ratified UNCLOS, the boundary will be binding on a significant number of
states, though it may be doubted that it will be binding on non-party states in the
absence of custom to this effect. Through this procedure of coastal state submis-
sion and adoption of the Commission’s recommendations, UNCLOS provides
the possibility of finality of the limits to continental shelf rights.
In practice, however, such finality may be hard to achieve. It is notable that
the Commission has the power to make recommendations only, reflecting the
jurisprudence of the ICJ in the North Sea Continental Shelf case, adopted by
UNCLOS, that the continental shelf is the natural prolongation of the land terri-
tory. As the coastal state has sovereign rights over the continental shelf, defined
by Article 76 according to its geomorphological features and distance criterion,
no treaty-based body could affect those rights without the consent of the state
concerned. The special interests of coastal states are particularly recognized by
UNCLOS as they have the right to establish the outer edge of the continental
margin, so long as it does not exceed 350 nautical miles from the baseline from
which the breadth of the territorial sea is measured or 100 nautical miles from
the 2,500 meter isobath. Subject to these limitations, the coastal state has the
Antarctica 405
right to determine for itself the outer edge of its continental shelf. In these ways,
UNCLOS acknowledges the profound sensitivities of coastal states with respect
to their sovereign rights over their continental shelves and their determination
to protect them.
If a coastal state rejects, or fails to implement, the recommendations of the
Commission in response to a submission, the claimed outer limits will not
attract binding legal status. The possibility that there might be a disagreement
by a coastal state with the recommendation of the Commission is recognized by
Article 8, of Annex 11 by allowing the coastal state to submit a revised or new
submission that might commend itself to the Commission. The coastal state is,
nonetheless, not required to comply with the Commission’s recommendations.
Rather, while failing to secure legal certainty from the Commission, the coastal
state remains free to maintain its claim to extended continental shelf limits and
to seek diplomatic recognition from its neighbours.
It is also notable that the actions of the Commission are not to prejudice
the delimitation of boundaries between states with opposite or adjacent coasts.
(Art. 9, Annex 11). It is not easy to imagine how a recommendation will do
otherwise than prejudice the claim of one or another of such states where their
claims overlap and are disputed. Hypothetically, were the Commission to make
a recommendation on the acceptability of Russia’s claim to the full extent of its
continental shelf to the North Pole, a claim that is disputed by others, the rec-
ommendation would have a critical effect on the political and legal viability of
Russia’s claim. The prejudicial effect of any such recommendation is moderated by
the requirement that all meetings of the Commission and its sub-commission and
subsidiary bodies are to be held in private (Rule 23 of the Rules of Procedure of
the Commission). Additionally, all recommendations in response to submissions
are to be confidential. Finally, and most importantly, the Commission has adopted
the rule that it will not consider a submission in respect of an area of extended
continental shelf where more than one state claims sovereign rights, unless the
consent of all parties to the dispute has been obtained. The United Kingdom,
Ireland, France and Spain have for example, lodged a joint submission with the
Commission on 2 June 2006 and agreed to include within it certain unresolved
claims to an extended continental shelf.
Submissions have been made thus far by Russia (2001), Brazil (2004), Aus-
tralia (2004), Ireland (2005), New Zealand (2006), France, Ireland, Spain, and
the United Kingdom ( jointly, 2006), Norway (2007) and France (2007). It will
be recalled that the deadline for submissions is 10 years from the date on which
the Law of the Sea Convention came into force for each state. Accordingly the
deadlines vary, so that, for example, Canada, has until 2013, and Denmark until
2014, to make their respective submissions. Vitally, no claims to extended con-
tinental shelf limits will be recognized under this procedure once the deadlines
have passed. Each of the submissions is being considered by sub-commissions that
then report to the Commission for final recommendations. The process has proved
406 Chapter 14
activity in support of sovereignty claims in the Arctic reflects the projected oil
wealth and strategic importance of the region and, as is so often the case in
international law, provide the impetus for clarification and development of legal
rules. It can be expected that the recommendations of the UN Commission,
particularly in respect to ownership of the Lomonosov Ridge at the North Pole,
will have a galvanizing effect on efforts to secure the most extensive state rights
possible in the region.
REFERENCES
Martin, L., 1938. ‘An American discovered Antarctica’, Proceedings of the 15th
International Geographical Congress 1938, Amsterdam, Vol. 2, 215-18,
International geographical Congress: Leiden.
———, 1940, ‘Antarctica discovered by a Connecticut Yankee: Captain Nathaniel
Brown Palmer’, Geographical Review, 30, 529-52.
New Zealand Gazette, 1923, 2815, Government Printer: Wellington.
O’Connell, D.P., International Law (vol. 1), Stevens and Sons, London 1970,
450-3,
O’Connell, D.P. and A. Riordan, 1971, Opinions on imperial constitutional law,
Law Book Company: Sydney.
Polar Record, 1948, ‘Antarctic claims – recent diplomatic exchanges between
Great Britain, Argentina and Chile’, 5, 228-44.
Reeves, J.S., Antarctic Sectors (1939) AJIL 519-21.
Richardson, I.L.M., ‘New Zealand’s Claims in the Antarctic’ (1957) 33 NZ LJ
38-42.
Rosenman, H., 1987. An account in two volumes of two voyages to the South
Seas by Rear-Admiral of the French Navy to Australia, New Zealand, Oceania
1826-1829 and to the Straits of Magellan, Chile, Oceania, Southeast Asia,
Australia, Antarctica, New Zealand, and Torres Strait 1837-40 in the corvettes
Astrolabe and Zélée, Melbourne University Press: Melbourne.
Rothwell, D., The Polar Regions and the Development of International Law 1996,
CUP.
Sahurie, E.J., The International Law of Antarctica (New Haven, CT, 1992) 193-
350.
Shackleton, E., 1991. South: the story of Shackleton’s last expedition, edited by
P. King, Century: London.
Svarlien, O., ‘The Sector Principle in Law and Practice’ (1960) 10 Polar Record
248-63.
Triggs, G.D. International Law and Australian Sovereignty in Antarctica 1986,
Legal Books, Sydney.
——— International Law: Contemporary Principles and Practices, 2006,
Butterworths/LexisNexis, Australia.
United States Naval War College, 1950, International Law Documents, 1948-9,
Government Printer, Washington, DC.
ANNEXES
ILLUSTRATIONS
BULGARIA
YUGOSLAVIA
Istanbul
ALBANIA Salonika
Kerkira
TURKEY
Ionia
n
Izmir
Athens
I sla
Illustrations
nds
D
1832
ode
1864
ca
Megisti
ne
1881 se
Kithira
1913 Islands
1922
1947
0 100 km
Kriti
413
5
5
4
5 3
3
4
3
3 2
4
4 2 1
4 3
Illustrations
5 1 2 2 3
2 2
1
1 Land held by non-Indians
2 Lands gained 1820-34 2
3 Lands gained 1835-49
4 Lands gained 1850-64
5 Land held by Indians
0 400 km
2.1 Advance of the American frontier. Based on the Historical atlas of the United States, National Geographic Society, 1988
Baltic Sea
Nor th Sea
BILLUNG MARCH
FRIESLAND
SAXONY
WEST
FRANCONIA
UPPER EAST BOHEMIA
Illustrations
LORRAINE FRANCONIA
EAST
SWABIA BAVARIA MARCH
MARCH OF
CARINTHIA
Duchy of
CARINTHIA
MARCH
MARCH OF OF
CARNIOLA STYRIA
0 100 km
Adriatic MARCH
sea OF ISTRIA
415
r
A Gando
ve
Kano
Ri
N O
D Zaria K
Yelwa Bauchi ◘
Kontagora O ◘◘
BORGU ◘
O
◘◘
◘ T Be nue
Jebba Muri
Bida O ○
Ni
Ilorin Nassarawa ◘◘
ge
r
r
ive
Illustrations
R ◘
Rive
r
Ketu Ibadan
Egba TIV Do
ng ○
Dahomey a
Iiebu
Porto Novo Ado Ode ○
Benin
FRONTIER OF CONTACT
Little Tribal mixing
Considerabletribal mixing
Bi g h t o f B e n in
Frontier of contact
March
◘Pagan enclave
0 200 km
○Fulani enclave
Gre
atF
ish
Great K
ei
Gam
Riv
er
Su
toos
nd
ay's
Rive
r
Grahamstown
Gou
Ri
ve
r
ritz
Zuurveld
Illustrations
Ri
ve
Cape Town r
Rive
r
0 10 km
2° 6° 10° 14°
Bouza
NIGER CHAD
Barrua 14°
14° Birnin n`Konni Gouré
Zinder Bosso
u
Maradi go Lake Chad
ou
Say Mainé Soroa m ad bé
Sokoto Katsina Ko Yo
U
N D Kuka
A
G W Gwandu
Birnin
Ri
Kebbi BORNU
ve
r
TO NIGERIA Kano
O KO
Illustrations
Nig S
er
10° 10°
2° 6° 10° 14°
Ri
o
Ri
Matala
o
Cu 15°
15°S an
Moçâmedes Ri
o
do
ZAMBIA
ANGOLA
Cuito
Cu
ban
Za
go
m
be
zi
C un e ne Vila Pereira de Eca
Chitado
Illustrations
Rio K un e ne
Fóz do Cunene Cuangar Luiana
Ruacana Mucusso
Falls Ondangua nt i
ya
n
Dirico Okavango
Li
Ohopoho
NAMIBIA
B OTS WA N A
0 100 km 15° 20°
9°E
Cr
os
s
Ajessi Ri
ve
r
Nkenaya
4
5°30´
5°30´
Nkpanaya
Iyama
Ekoroman
Ekundukundu
5°N Ekonako
5°
Abasi Ndian
Etim
1 1885 Agreement
2 Commission’s proposal
Moor’s proposal 1
Moor’s proposal 2
Moor’s Neutral Zone
1 River Akpayate
2 River Ndian
3 River Min
Rio
4 River Awa
de
0 20 km
lR
y
e
9°
3.3 The Anglo-French boundary in the vicinity of the Rio del Rey and the Cross
River rapids
Illustrations 421
Amanab
4° Se
pik
Biake River
St
ar
Mt
ns
.
INDONESI A
PAPUA
NEW
Namas G UI NEA
Digul
6°
Angamarut
Mabduam
Lake
Fly
ei
Murray
ng
Kuem
Su
Merauke
Bia
Riv
er
i
Wa
8°S
i
n ge
Su
R.
Merauke
Weam
Morehead
sbach
0 50 km n
Be
Ribe Fyn
Christiansfeld
Romo Li
lle
St
ra
it
Als
Tonder
Sild
Flensburg
Schleswig
Husum
Eid
Danevirke c.1060
er
R ive r
1864 0 20 km
1920
0 100 km
A
Me k o n g
M South
(M
18°N
ae
g Nong Khai China
an
Na
mK
Hu Sea
Nam
ho
E
n g) FRENCH
Mukdahan
I ND O-C HI NA
THAILAND
Mae Na
m
Mun
15° Champasak
Nakhon
ngrak
Ratchasima (Bassac)
T h i u K h a o P h a n om o
rek
D g
C h ai ne des n
Do
Chong
Ken
Krung
(Bangkok) Huai Nam S isŏphŏ G B
Sai
n
S. Siemréab
G 13°14´48.1˝N
D FRENCH
I ND O-C HI NA
C
12°
Phnum Pénh
Gulf of (Phnom Penh)
Thailand
Laem Samit Sài Gòn
102°E 105°E
R
io
COLOMBIA
Ri
E C UA D O R
o
Pa
ya ayo
Rio
Guayaquil gu Putum
2ºS
as
Cura
ray
R io
Na U
po
Rio
R Rio
za
M A Y N A S Iquitos Am
azon
Illustrations
Ma
P a st a
pon
a
ari
Ya v
R io
P
n on
ra
Ma
Rio
BRAZIL
Jaen
Rio
6º 78º 74º
0º 0º
Put
u
ma
COLOMBIA
yo
Ri
ve
r
E C UA D O R
River
Na
Riv po
er Cur
Ri
Guayaquil
v
itra
y
Ri
er
ve
r
S
Illustrations
A
N River
Co
r
San
Y Ti
g re Am
A
rie
azo
M ntes
Iquitos n
ti a go
Tumbes
4º
n i
ano v ar
Mar Ya
River
er
Riv
PERU
Jaén BRAZIL
0 100 km
425
4.2 The 1910 Spanish Award and the Rio Protocol Line 1942
Village
426
5 km Pt.
Boundary post Ribado
* n ue Filingo
Be
R.
Junjure
Yola Junsengo
Bapaio R.
Gurin
For
o
600m
900m 6
*
Toja Mts.
*7
Illustrations
ti
*
R. In
Be *
i
Ka
17 R.
* ** Mt rin
s.
**Mt. Sa
mi
R. lo
So
ns
*16 hi
15
* Mt. Baraji
*
0
300
k a
0 10 km 2000 a n t i
A l
K F
Nxonidas E
H
Rooi Kop D
L
Rooibank C
Sw
Scheppmansdorf
akop
River
M
W
Illustrations
al
vi
0 5 km s
Fredericks
German claim Dam
Worfel
(B - M) Boundary Dam Dorop
Ba
Dam Surveyors
surveyed by Wrey Pelican Dam
y
(M - B) Wrey’s closure Point
A
Areas of water and B
vegetation
16 National Capital 16
Town
0 25 50 75 100 125 km
Keren
0 25 50 75 m
Akordat Massawa
This map has been prepared only for convenience of
reference from the series of 45 maps on a scale 1:25,000
which are attached to the Statement of the Eritrea-Ethiopia
Boundary Commission dated 27 November 2006 and which
are the sole authoritative illustration of the locations for
pillar emplacement set out in the Annex to that Statement.
Barentu Asmara
15 15
Tserona
Zalambessa
Adigrat
Aksum
14 14
D
Illustrations
a
n
a
Tserona
k
Mekele
i
Be
l
l es
aB
Be
le
ETHIOPIA
sa
Assab 13
13
A
Zalambessa
M un a Bure
November 2006
DJIBOUTI
37 38 39 40 41 42 43
5.1 Overview map of the international boundary between Eritrea and Ethiopia (source: Permanent Court of Arbitration
website, www.pca-cpa.org)
International Boundary between
the State of Eritrea and
the Federal Democratic Republic of Ethiopia
Z A I CENTRAL SECTOR
G U Mercator Projection
L E Datum: WGS-84
E ERITREA Scale: 1:360,000
H Base map is taken from the 1:100,000 Soviet Union topographic mapping series,
with supplemental date from satellite imagery acquired from SPOT and ASTER/
C TERRA Place names are compiled by the Commission based on various sources
(see “Technical Note Relating to Maps”).
C This map is produced for illustrative purposes only.
14°45’ A Reference Point
4 0 4 8 12 16km
eb
Senafe A gh
ir Eritrea-Ethiopia Boundary Commission
Mar
SHIMEZANA
En
d eli
11 12
Tserona
Belesa
Barachit 26
b
13
Be
25
are
le
M Bel
esa C
IROB
sa B
24 Ragali
B Guna Guna 29 30
23
e le
Sebao Alitena
MonoxeitoE i
sa A
Zalambessa n da 22 27 al
From the confluence of the Mareb Kelloberda Massolae ag BADA
Da R
and the Mai Ambessa to Point 11, 20 shi
m
the boundary follows the Mareb. 14 21
15 18 Mun do Muna / Berbero Gado
a/ Berbero Ga Rendacoma
16 En
Illustrations
Fort Cadoma de
li
28
17
GULOMAKHEDA
31
Salt
Debra Damo Lake
E
14
M
G A
15 Adigrat
Be
A
les
Belesa
a
16
A
B
14°15’
17
ETHIOPIA
39°00’ 39°00’ 39°30’ 39°45’ 40°00’ 40°15’
429
5.2 Map of the Central Sector of the international boundary between Eritrea and Ethiopia (source: Shaw 2007)
430
N
ig
Me u
er
Mekro
BURKINA FASO NIGERIA
Niger’s claim
[line of deepest soundings]
NIGER
( )
Nig
er
BENIN
TOGO
12°00°N
Illustrations
SKETCH-MAP No. 3
Claims of the Parties in the
sector of the River Niger
N.B. This sketch-map has been prepared
for illustrative purposes only
BENIN Left bank
Right bank
3°00°E
3°30°E
5.3 Claims of the Parties in the sector of the Niger River, in the Benin/Niger case (source: [2005] ICJ Rep 123)
SKETCH -MAP N O. 1. G ENERAL G EOGRAPHICAL S ETTING
(N.B. This sketch-map has been prepared for illustrative purposes only.)
116˚ 120˚ 124˚ 128˚ 132˚ 136˚ 140˚
16˚ 16˚
Luzon
12˚ 12˚
Philippine Sea
Palawan
Sulu Sea
8˚ Mindanao 8˚
Illustrations
Ligitan
4˚10’ Sipadan 4˚10’
4˚ Sebatik 4˚
Nanukan Republic of Indonesia
Tarakan
Celebes Sea
Malaysia
5.4 Map of the general geographical setting in the Pulau Ligitan/Pulau Sipadan case (source: [2002] ICJ Rep 635)
432 Illustrations
B A
E
D
C
Map 3 Carte 3
(See paragraphs 5-9) (Voir paragraphes 5-9)
The maps in the present Judgment were pre- Les cartes jointes au présent arrêt ont été
pared on the basis of documents submitted to établies d’après les documents soumis à la
the Court by the Parties, and their sole pur- Cour par les Parties et ont pour seul objet
pose is to provide a visual illustration of the d’illustrer graphiquement les paragraphes de
paragraphs of the Judgment which refer to l’arrêt qui s’y réfèrent.
them.
5.5 Map from ICJ proceedings in the North Sea Continental Shelf cases (source:
[1969] ICJ Rep 15)
S KETCH -M AP No. 2
the suppression of
UPPER-VOLTA 16°
Hombori
ted frontier
pu X 15°
dis
Y
Mondoro
Cercle de Dori
W Z
Yoro ARIBINDA Tillabéry
Dori
14°
N IGE
NIGER
Cercle de Ouahigouya R
Ouahigourya
Ouahigouya
Illustrations
13°
0 100 200 km
4° 3° 2° 1° 0° 1° 2° 3°
433
5.6 Map from ICJ proceedings in the Frontier Dispute case between Burkina Faso and Mali (source: [1986] ICJ Rep)
434 Illustrations
1 Limani
2 Keraua River
3 Kohom River Sector of the land boundary delimited by the
Thomson-Marchand Declaration
4 Ngosi to Humsiki Sector of the land boundary delimited by the
British Order in Council of 2 August 1946 12˚
5 Mt. Kuli to Bourha Sector of the land boundary delimited by the
6 Kotcha Anglo-German Agreements of 11 March and 12
April 1913
7 Source of the Tsikakiri River 1
From Beacon No. 6 2
11˚
8 3
to Wamni Budungo
Maio Senche 4
9
5
10 Jimbare and Sapeo
Noumberou - Banglang 7 6 10˚
11
12 Tipsan
13 Crossing of Maio Yin
14 The Hambere Range area
8 9˚
From the Hambere Range
15 9
to the Mburi River
16 Bissaula - Tosso 11 10
17 Sama River 8˚
12
13
NIGERIA 17 7˚
7˚ CAMEROON
16
14-15
6˚ 6˚
SKETCH-MAP No. 3
5˚ 5˚
Points in Dispute between
Lake Chad and Bakassi
Bakassi
Peninsula
Peninsula
N.B. This sketch-map has been prepared
for illustrative purposes only 4˚
6.1 Sketch map depicting the points in dispute between Lake Chad and the Bakassi
Peninsula, in Cameroon v Nigeria (source: [2002] ICJ Rep 361)
Illustrations 435
CAMEROON
4°50’
4°50’NN NIGERIA r
ve 4°50’ N
Ri
e
y af
wa
Ak
Cro
ss
Riv
er
4°40’NN
4°40’ King Pt. 1
4°40’ N
2 Bakassi Pt.
4 3
Rio D
5 Akwabana
6 Island
el Re
Tom Shot Pt.
7
8
y
9 Sandy Pt.
10
West Pt. A 11 Bakassi Peninsula
4°30’ N A1 12 4°30’ N
East Pt.
A
C 12
A1
6.2 Sketch map depicting the maritime boundary in Cameroon v Nigeria (source: [2002]
ICJ Rep 449)
BADIN TALUKA T H A R P A R K A R D I S T R I C T
(Hyderabad District)
436
ET
p
Diplo
Rahim Ki Bazar O
Loo
Loo
Surai
S I N D N
tern
B Indian Claim Line E K L
tern
A W
Kanjarkot Point 84
Eas
D A
Dhara Banni
Wes
Shakur P V K M Nagar Parkar V
JATI TALUKA Ding C Vigokot Chhad Bet H
(Karachi District) Sardar Sindri F
Karimshahi Biar G
Nara
Bet Bet
JATI
K U T C H Dharamsala
Gainda
24º Longitude Bet 24º Longitude
WT
G R E A T R A N N O F K U T C H
i
S
ek
re r
C
A
Line SCHEMATIC MAP: NO SCALE
R aim
’s Cl
Illustrations
an A = Trijunction of Kutch state, and Karachi and Hyderabad districts of province of Sind
A P akist
B
WT = Western Terminus (Longitude 68º48’E Latitude 23º58’N)
i ET = Eastern Terminus
I Knor reek = Pillars on ground
C
A K U T C H = Pakistan’s Claim Line
N = Indian Claim Line
GULF = 24 Longitude
S OF = Awarded Boundary
E KUTCH = District Boundary
A
PVK = Pirol Valo Kun
KATHIAWAR L I T T L E R A N N O F K U T C H
OR
SAURASHTRA
6.3 Overview map in the Rann of Kutch dispute (source: Untawale, ‘The Kutch-Sind Dispute: A Case Study in International
Arbitration’ (1974) 23(4) International and Comparative Law Quarterly 818-839)
SKETCH-MAP No. 3
MT. TABAKARACH
ré MARE
é MI TIN EOULT OJOOEL
ab
TO
R
I KA
FITILI
MI OUACOU NT
RO
MARE
T. T
TIN TABORE IN-ABAQ M
15° N MARE MARE TIN KACHAN
AFERERE a
TIN TABAKOT urm
TIN AKOF Nco
Mt.
TIN
TANGOUN TANMARA
FADAR-
CUE COLLIN
FADAR
KABIA
Illustrations
OURSI MARE
6.4 Sketch map drawn from the ‘Blondel la Rougery’ map in the Frontier Dispute case between Burkina Faso and Mali (source: [1986]
437
Sh
IRAQ at
ia
l ‘A
ra
b
30°N
KUWAIT
29°N
Bushire (Bushehr)
IRAN
28°N
27°N
Strait of Hormuz
Illustrations
Ha
sa
BAHRAIN 26°N
Co
QATAR
as
t
Dah
Doha
wat
Wakrah
25°N
Saiw
a
Khor Al Udeid
24°N
6.5 Map depicting general geographic context in the Qatar v Bahrain case (source: [2001] ICJ Rep 53)
Illustrations 439
Sketch-map No. 3. Equidistance Line Taking into Consideration All the Islands
and Those Low-tide Elevations Located in the Territorial Sea of One State only
50°E 15’ 30’ 45’ 51°E 15’ 30’ 45’
20’ 20’
equidistance line
Line if Fasht al Azm is regarded as a low-tide elevation
Line if Fasht al Azm is regarded as part of the island of Sitrah
10’ 10’
Iran
27°N 27°N
50’ 50’
40’ 40’
30’ 30’
20’ 20’
10’ 10’
Bahrain
26°N 26°N
50’ 50’
40’
Qatar 40’
Saudi Arabia
30’ 10 kilometres
30’
Mercator projection
This sketch-map, on which maritime features are shown in simplified form, has been prepared for illustrative
purposes only.
It is without prejudice to the nature of certain of these features.
The framed zones are enlarged in SKETCH–MAP Nos. 4, 5 and 6
6.6 Map showing the equidistance line taking into consideration all islands and those
low-tide elevations located in the territorial sea of one state only, in the Qatar v Bahrain
case (source: [2001] ICJ Rep 105)
89°20’ 89°15’W
440
SKETCH-MAP No. A-1
N First Sector — Disputed Area
B Agreed boundary
El Salvador claimed boundary
Honduras claimed boundary
A
Hacienda Other claims mentioned in
14°25’ N Montecristo mola 14°25’ N
B' Judgment
Rio Po
C Endpoint of disputed sector
J D
H'
Polcho
Rio de
a
E
H C' Q.
mp
de Gualcho
Le
Illustrations
Q.
de
Citala Poy
Ingenio
Rio
Miguel
G'
R. Jupula
n
Sa F
o
Ri G 0 5 KM
14°20’ 14°20’
89°20’ 89°15’W
6.7 Map depicting the disputed area in the first sector in the Land, Island and Maritime Dispute case between El Salvador and
Honduras (source: [1992] ICJ Rep)
88°55’ 87°50’ 87°45’ W
N
R
L
13°55’
13°55’ N
N no
e itu l
Ac ena
Ar
el e
A
.d d
Q Q.
Rio
Q
Toro
la P
0 5 10 KM
Illustrations
Boundary in Judgment
441
6.8 Map depicting the judgment boundary in the Land, Island and Maritime Dispute case between El Salvador and Honduras
(source: [1992] ICJ Rep)
25°07’E 25°08’E 25°09’E
er
442
Riv
17°48’S 17°48’S
be
Cho
N A M I B I A Kasahe
p
Stri
Air Kabuta
(Source:
Game
nel
Scout
l
(Source :Botswana
an
Camp
l
Kasika
ne
Ch
n
nne
ur
a
Sedudu Island
ha
BotswanaJudges’
Sp
C
Judges’
th
17°49’S 17°49’S
Folder
uth
or
Folder
So
N
Illustrations
——
ACCORDING TO B OTSWANA
Section
Section
4) 4)
Kasane
Forest
be River Reserve
Cho Chobe
MAP SHOWING THE TWO CHANNELS AROUND KASIKILI/SEDUDU ISLAND
National Park
Headquarters
B O T S W A N A
25°07’E 25°08’E 25°09’E
7.1 Map depicting the two channels around Kasikili/Sedudu Island according to Botswana, in the Kasikili/Sedudu
Island case (source: [1999] ICJ Rep 1068)
Bratislava
M
al
y
D
an
ub
Žitný Ostrov e Slovakia
Čunovo
Dunakiliti
dam
Headrace canal
Gabčíkovo
Tailrace canal
Sap
Nagymaros
Mo
soni
Da Visegrád
nub Komárom Nagymaros
e Györ
Illustrations
barrage Szentendre
Island
SKETCH-MAP No. 2
Hungary
The Original Project
N.B.: This sketch-map has
been prepared for illustrative
purposes only
Budapest
7.2 Map depicting the original project in the Gabčíkovo-Nagymaros case (source: [1997] ICJ Rep 21)
444
OCEAN
Gulf of St. Lawrence
en c
Vancouver
Illustrations
aw r
Calgary
L
Island Vancouver
3
S t.
M
S
Arkansas T
Ri A
ve T
r
er
Rio
E
E
Riv
S
do
Riv
ra
Puerco
lo
er
Red
Gra
River
Co
33°N Ri
San Diego Gila Sa b
nde
ve
r
in
e
33°N
(Pecos) R
iv
Paso del Yoke
er
I
Lake
R iv e r
Norte
Pacific
Nue
(Ri
o
ces
Illustrations
O cean C Br
av
o)
Gu
G u lf o f
lf
Ca
0 300 km
of
lif
o
rn
Mexico
ia
Adams-Onis Treaty, 1819
Original American proposal, 15-4-1847
First amendment, 15-4-1847
Second amendment, 19-7-1847
Line suggested by Trist. 2-9-1847
445
49°45´W
(Ribeiro 1529)
42°30´W
N e g ro
(Cantino 1502)
Japur 0°
a Rive er
River r Riv
nas
Amazo Belem
Manaus Sao Luis
r
Rive
Iquitos Leticia
r
ve
Ri
r i
Yava
ira
de
Ma
10°
Gua
por
e
Lima R.
Cuzco
Cuiaba
La Paz
Corumba
er
Riv
Rio de 20°
Janeiro
Asuncion
Pacific
a
P aran R.
Ocean
Atlantic
Uru gu ay
MEXICO
EMPIRE OF
PERU BRAZIL
BOLIVIA
PARAGUAY
ARGENTINA
CHILE
URUGUAY
0 500 km
ela
zu of
12°N l
f
Gune
Ve
10°
R.
o
u mb
tat
Ca
Riv
8° er r
ur
e Rive
Ap Rive
r
Sar
arl
co
a
uc
ino
a
River Ar
Or
6°
Rive
r VENEZUELA
M e tr
Apostadero
COLOMBIA
4°
Or
ino
co
Riv
er
2°
1833 boundary
Venezuelan claim
Colombian claim Cucuhy Rock
Brazil - Venezuela
Neg o
0 50 km
BRAZIL
r
0°
75° 73° 71° 69° Riv 67° 65°W
er
Ri
o
Pi
lco
ma
yo
Esteros PA R A G U AY
24° 24°
marshy area
ná
ara
Espinill
Ri Villa Háyes
Illustrations
o
Rio P
Be Clorinda ASUNCIÓN
rm Puerto
ej Presidente
A R GENT INA o
Stroessner
Eldorado
Pedro Isla
González Yacyretã
y
Resistencia gua
Corrientes Uru
0 100 km Isla
Apipé Grande Apóstoles
449
Ca
Black Sea
sp
ia
n
RUSSIA
Baku
Ankara
Erzerum
A N A T O L I A
Se a
Ti Tabriz
g
ris
Alexandretta
Mosul
River
ea
Eu
n S
ph
ra
PERSIA
tes
nea
rra
dite
Damascus
Baghdad
Me
PALESTINE
Rive
Jerusalem Amman
r
EGY
PT
Aqaba
A R A B I A
Pe Gul
Kuwait
rs f
ia
n
60° 62°
36°
RUSSIAN
FEDERATION
Maclean
Award
1888-91
Hērat
34°
34°
Turkish
delimitation and
demarcation
15 May 1935
AFGHANISTAN
IRAN
32° 32°
Hāmūn-e Sāberl
Daryācheh
Sisiān
McMahon
Award
1903-05
30° 30°
PAKISTAN
60° 62° 64°
IRAQ
ShaArab
Safwãn I RAN
tt a
l
30ºN
BŪBIYĀN
ţin
Bā
K U WA I T
al
dĩ
Wā
KUWAIT
Al Jahrah
Wādĩ
Persian
al ‘Awjā’
Gulf
Minã’
al Ahmadi
29º + Ar Ruq‘l
KUWAIT
SAUDI ARABIA
SAUDI A R A B IA
Ra’s al
Khafji
Ra’s al
1913 Anglo-Turkish boundary Mish‘ab
1913 Anglo-Turkish Green Line
28º
1922 Kuwait-Saudi line
1922 Neutral Zone
1969 Neutral Zone
0 25 km
32° 32°
+Muqr an Na‘am
TRA
NS- Judayyadit ‘Ar‘ar Eup
hrat
PIP ARAB es
ELI I IRAQ
NE AN
‘Ar‘ar
+ Al ma‘aniyah Hawr
Al Lifiyah al Hammar
+
Al ‘Uwayqliah
30°
Illustrations
Rafha
KUWAIT
SAUDI ARABIA
in
B at
1922 boundary
al
Dead Sea
Al Hadithah
Hadda Agreement
ISRAEL 1925
1965 Boundary
0 40km
J OR D A N
Illustrations
SAUDI ARABIA
30°
Al ´Unāb
Al ´Aqabah
Gulf of
Aqaba
Halat ´Ammār
Tichit
Tidjikja
Oualata
Timbuktu
S e n e gal
GHANA GOA
Hombori
Nioro Goumbou
C. Verde Bakel
Dori Tillabéri
G a m b ia
Djenne Ouahigouya
MALI
Say
Koulikoro Ségou
Ouagadougou
Kangaba
Fada N’Gourma
Bl ac k
Illustrations
Volta
Nig
er
Vo
lta
Approximate limits of
the Empire of Ghana
Approximate limits of
the Mali Empire
Approximate limits of
Gao Empire
0 200 km C. Palmas
455
Nige
r
Benue er Ibi
Riv
Lokoja 86
8 5-
18
RY
River
A
U ND
Owo
BO
N I G E R I A AN
M
ER
-G
LO KAMERUN
Benin G
AN
Illustrations
Mahin
. Warri
Ben
in R
R. COMMERCIAL TREATIES
a d os Pre 1885
F o rc
Calabar Post 1885
CONSULAR TREATIES
Pre 1885
R. Post 1885
l e t o nAkassa
R.
d d 0 100 km
Opobo
Mi
R.
B ra ss R.
Nun
10.2 British treaties in the Niger Delta and along the Niger and Benue Rivers
Illustrations 457
Lake Chad
Ft. Lamy
NIGERIA
Douala
gi
Uban
Rio Muni er
Riv
Riv
Libreville
er
FRENCH
EQUATORIAL BELGIAN
AFRICA CONGO
Brazzaville
Anglo-German boundary
g o
C on Original Franco-German
boundary
0 250 km 1911 Franco-German
boundary
FRENCH
Gu l f o f A den
SOMALILAND Djibouti
ETHIOPIA HA
UD
B E I D YA H A N
Riv
er
A
LL
MAREHAN
OGADEN
A
G
Riv
SOMALI
er
Sheb
Dolo
elie
Indian
Ocean
BORNA
Juba
Mogadishu
K E N YA
International boundary
Former Anglo-Italian boundary
Western limit of Somali tribes
G BORAN Tribal name
A
LL 0 200 km
A
F g
E
Lake Nyasa e
D
d
Illustrations
C c
Malawi b
B Mozambique
A
0 5
km a
459
Chisumulu
Lake Nyasa
Likoma
Mozambique
2 nm overlapping arcs
Median Line
0 5 nm
A
A Junction of Masrab
SSE
el Ajrane on north
EGYPT border of Melfa Oasis
LIBYA
B Intersection of 10km
arc from Manasseb
Pass and SSE line
B
10km C Intersection of 10km
arcs around Manasseb
Manasseb Pass Pass and Williams Pass
D Intersection of Williams
Pass arc and bearing
C WSW from Williams
10km Pass
C Intersection of meridian
25°E with parallel 22°N
E
0 10 km
25°E
F 22°N
SUDAN
0 5 km
U
Illustrations
M
A
B
s
DEMOC RATI C
id
a p
K i nt a mo R REP U B LI C
OF CONG O
Lake Chad
limit 1943 14°
NIGER
CHAD
e
Yob
g ou
ou
ad
m
Illustrations
Ko Ri
ve
r
R iver
Dikwa Ndjamena
NIGERIA 12°
CAMEROON
0 50 km
Ghari
33ºE
Rio Caure
Sallsbury
SOUTHERN RHODESIA
si
Inyanga
18º30´S Rio
Pún18º30´S
goè
Edge of plateau
Boundary award Umtali Vila de Manica
Agreed boundary
Vila Pery
0 50 km
Sa
bi
Lucite
Fort Victoria Rio
Chipinga
Búzi
o
Ri
Lu MOZAMBIQUE
nd
i
Nu
an
Mbizi ave
els
Rio S
i
33ºE
Tu mm o
EGYPT
Fr a Jef - Jef 21
n co Plateau . 3.
18
- It
a 99
20˚ 7. 1 l i an A 20˚
.1 9 g r
35 e e m
en
t
*
CHAD Ou ni anga Erdi
Ke bi r Plateau
SUDAN
0 100 km
16˚ 20˚ 24˚
Yenise
C D 1689
y
60°
D E 1727 U S S R
River
E F 1945
na
F G 1864
Riv
Le
Am
er
G H ? ur B Khabarovsk
C
Omsk
Lake
River
Baykal Sea of °
Irkutsk D 40
E Tsitsihar Japan
Kyakhta Vladivostok
50°N A
Semipalatinsk
Karaganda g un
F Ulan Bator Ar
NORTH
Illustrations
Lake
KOREA JAPAN
Balkhash
MONGOLIA
Chuguchak
lli Riv
Alma-Ata er Peking
Wulumuchi
40°
SOUTH
KOREA
Kashgar °
G C H I N A 30
AFGHANISTAN
Yellow Sea
H 0 400 km
PAKISTAN Lanchow
Am
ū
Bukhara Samarkand
C H IN A
Da
ry
ã
RUSSIAN F E D E R AT I O N AN
(O SH
38°N xu O
AZ R
s)
R W AN
Mary B DA I GN Oz
SH Zorkul
(Merv) Kwaja Salar
CA N *D
H
K
A
Sarakhs V
IRAN
Pass
er
orghab
Illustrations
Riv
M
Rū AFGHANI STAN
d
To
Khost valley
To
Khost valley 22 Laram
23
To Khost valley
Tsatai
Chela 21 Ping
20 G
or e
Kuchan s ht a R. To
.
Kurram
R
A Narexi
B 18 C valley
xi
ha
Wri Psarlai Chappara 19 rk
13
ha
i
12 17
R.
2 11 16
3 7 14 15
4 Do
6 8 9 10 w gh
5 a r R. Kanibogh Ziarai
1 Ka
Illustrations
ni b
S o gh
R.
ar
to
i
To Tochi valley
R
. y
le
val
chi
To To To
c hi vall To
ey
Area A ceded to Afghanistan because Area B ceded to Afghanistan because
Durand Line Durand Line was supposed to Gurbuz living there are subjects
Demarcated boundary follow the watershed between of Amir, because demarcated
10 Boundary pillar the Tochi and Khost valleys line followed more prominent
features and because grazing is
0 4 km inferior
us
A K RAJASTHAN
Ind
P HYDERABAD DIVISION
EASTERN
Islamkot TERMINUS
S TAT E
Bedin
THAR PARKAR DISTRICT
SINDAL BET
68°41´E
23°58´N ch
WESTERN Kut
TERMINUS
of
ek
Rann
e
Lakhpat Pachham Khadir Radhanpur
Kori
Sir Cr
k
ee
r
ri C
Illustrations
G U J A R AT S TAT E
Ko
A
I
N D
I
Bhuj
International boundary
Boundary segment
Anjar claimed by India
Ara bi an Boundary segment
Se a claimed by Pakistan
ut ch Disputed area
of K
Gulf 0 40 km
469
River
CAMBODIA
AY
d
W Roa
anh
GI
R iv e r
H
Th
AN
’S
IN
TH
H
the
AR
N
ND
ng
A
MA
fo llows Gia
H
T
mpart
ry
Outer ra
da
SAKY A
un
H
Bo
KAN
CHANLOT par
t VIETNAM
er ram
Hon Ta Inn
SWITZERLAND
AUSTRIAN EMPIRE
VENETIA
LOMBARDY 1866
1859
PIEDMONT
PARMA
1860 ROMAGNA
MODENA
1860 OTTOMAN
EMPIRE
KINGDOM
TUSCANY MARCHES
UMBRIA
1860
CORSICA
(FRANCE) PAPAL
OF
STATE
1870
SA
S ICI L IE S
RD
1860
IN
IA
O
TW
E
H
T
0 100 km
Dn
ies
ter RUSSIA
Da AUSTRIA - HUNGARY
nub
e
Vienna
Riv
Budapest
er
River
Sa
va
River
ROMANIA
BOSNIA Belgrade Bucharest
HERZEGOVINA
Sarajevo SERBIA
BULGARIA
MONTE Black
NEGRO Sofia Sea
EASTERN ROUMELIA
Adriatic
Sea
OTTOMAN EMPIRE Constantinople
ITALY Solonika
THESSALY
Athens
GREECE
Boundary of Ottoman
Empire, 1870
Boundaries after 1881
Areas lost by the
Ottoman Empire
0 200 km
Secular
G RO
E NE
Mt. Planinica
T Murina
N
O
M
Plava
Gusinje
Podgoritza
IR E
E MP
N
A
M
O
T
T
Lake
O
Scutari
Scutari
July 1878, Treaty of Berlin
January 1881, Delimitation
Lake January 1883, Delimitation
Sas August 1883, Delimitation
December 1884, Delimitation
Adriatic
Sea 0 10 km
FINLAND
SWEDEN
NORWAY
ESTONIA
LATVIA
DENMARK
EAST LITHUANIA
PRUSSIA
SOVIET
U.K. UNION
NETHERLANDS
POLAND
BELGIUM GERMANY
LUXEMBOURG
CZECHOSLOVAKIA
AUSTRIA
FRANCE SWITZERLAND
HUNGARY
RUMANIA
YUGOSLAVIA
ITALY BULGARIA
12.4 The chain of new and modified states after World War I
AUSTRIA
HUNGARY
Slovenia
ITALY Ljubljana Subotica
Zagreb ROMANIA
Osijek
Croatia
Beograd
Bihac
Bosnia - Herzegovina
Knin Sarajevo
A
Serbia
dr
i at
ic
Illustrations
Se
a
Neum Montenegro BULGARIA
Igalo
International boundaries
Republic boundaries
Based on international and Skopje
administrative boundaries first
delimited before 1900
Macedonia
Based on boundaries delimited
after the Balkan Wars, 1913
ALBANIA
Delimited after 1945, and not
based on previously delimited
boundaries 0 200 km GREECE
475
L ava n t
Carinthia in Austria St Veit
Gurk
before 1921
0 20 km
Dr
av
a
er s ee Klagenfurt Lavamünd
r th
Wo
Villach
Rosegg
Illustrations
J U G O S L A V I A
R O
U
M
A
N
I
A
M o ra v a
Vidin
Danu
A
be
V
I
A
A
I
R
L
Caribrod A
S
Sofia
G
O
Bosiljgrad
L
G
U
U
B
J
Stru
Ne
ma
sto
s
Strumica
C E
E
E
R
Territory ceded to
Jugoslavia, 1919 G
0 50 km Salonica
Mitrovica
Pec
A
Scutari
L
Skoplje
B
A
Adriatic
Sea Debar
N
I
In 1913
A
Up to 1941
In 1943
Albanians in Yugoslavia
1921
0 50 km
FINLAND
NORWAY
SWEDEN ESTONIA
LATVIA
DENMARK
LITHUANIA
SOVIET
UNITED NETHERLANDS UNION
KINGDOM
Russian
zone POLAND
BELGIUM GERMANY
LUXEMBOURG
Anglo - French CZECHOSLOVAKIA
United States
zone
AUSTRIA
FRANCE SWITZERLAND
HUNGARY
RUMANIA
ITALY
YUGOSLAVIA
SPAIN
BULGARIA
Areas gained by Soviet Union 1940-45
ALBANIA
Communist states
Yugoslavian gain from Italy
0 200 km GREECE
12.9 Boundary changes in Eastern Europe during and after World War II
480 Illustrations
Savu Sea
Batugade
INDONESIA
fa ra
Ta
M.
EAST TIMOR
Takis
Disputed boundary
River boundary
Terrain boundary
M. Masin)
INDONESIA 0 10 km
(Mas
sin
T i m o r Se a
124˚30´E
13.1 The boundary between East Timor and Indonesia in the east of the Island
9˚30´S
to
el Me
Savu Sea
No
Keli
EAST TIMOR
(OECUSSI)
i
Citrana lom
9˚30´S l Bi
Noe
N
oe
l
Illustrations
Be
si
an
l Ek
Noe
Disputed boundary
River boundary
i
Terrain boundary
Bon
Sunan
INDONESIA 0 10 km
no
No
Bidjsel
9˚30´S 124˚10´E 124˚20´ 124˚30´
481
G. Raya
H
A A
SE B
5° BRUNEI A 5°
Bandar Seri Begawan S
BRUNEI
A G. Moeloek
Labang B
P.
an ge
NA Sebatik
at R
G. Brunei HI
Du
C
o
Ap
Area ceded by G. Api
Dutch in1928 0 4 km
Convention Celebes
Bintulu
Sea
H K
T A
Illustrations
S OU W
A
n
A
ra
I
P. I
T. Datu
A S 2°
S
E
N
G. Raya O
Kap ua s H u l u R ang
e
SEE G. Api
INSET D A B Boundary defined
N by Agreement
I
28 September 1915
0 100 km
111°
I
RR
IT
and Chile 53º T I IC
TE S
BR
E N CT R
G R O
A
CT º 60º
AN R
60º 70
SE TA
QUEEN MAUD LAND
AU
Enderby
R
Land
STR
WEDDELL SEA Kemp Land
RY C T I C
A LI
AN
I TO
74º º Mac. Robertson
80 Land
TER R AN TA
N
80º
A
Princess
Elizabeth Land
C H I LE
90ºW Wilhelm II Land 90º
ELLSWORTH LAND
Queen Mary
ANTARCTIC
Land
BYRD LAND
Illustrations
Wilkes
Ross Ice Shelf
Land
TE
RR
Ad
i a La n d
eli
Ross Sea
ITO
e
RY
t or George V 120º
La
120º
nd
N
NC
IA
E
AL I C 136º
TR
AUS ARCT RY
T 142º
AN RITO
R
TE
150º RO S S D
E P EN D E NC Y
150º
(N E W Z
E A L A ND ) 160º
180º
483
Allee, T.L. and P.K. Huth, 2006, ‘Legitimizing dispute settlement: International
legal rulings as domestic political cover’, American Political Science Review,
vol. 100, no. 2, pp. 219-234
Allen, S., 2006, ‘Case Concerning the Frontier Dispute (Benin/Niger)’, Interna-
tional and Comparative Law Quarterly, vol. 55, no. 3, pp. 729-742
Allott, P., 1990, Eunomia: New Order for a New World
Anene, J.C., 1970, The International Boundaries of Nigeria 1885-1960: The
Framework of an Emergent African Nation
Auburn, F., 1982, Antarctic Law and Politics
Bernhardt, J.P.A., 1975, ‘Sovereignty in Antarctica’, California Western Inter-
national Law Journal 297
Brownlie, I., 1979, African Boundaries: A Legal and Diplomatic Encyclopae-
dia
———, 1990, Principles of Public International Law (4th edition)
———, 1998, The Rule of Law in International Affairs: International Law at
the Fiftieth Anniversary of the United Nations
Bush, W., 1982, Antarctica and International Law: A Collection of Inter-state
and National Documents
Castles, A.C., 1996, ‘The international legal status of the Australian Antarctic
Territory’, in D.P. O’Connell (ed.) International Law in Australia
Grotius, H., 1625, De Jure Belli Et Pacis
Hardin, G.H., 1968, ‘The tragedy of the commons’, Science, vol. 162, pp. 1243-
1248
Hayton, R.D., 1956, ‘The “American” Antarctic’, American Journal of Interna-
tional Law, vol. 50, no. 3, pp. 538-610
Highet, K., 1987, ‘Evidence, the court and the Nicaragua case’, American Journal
of International Law, vol. 81, no. 1, pp. 1-56
Horlin, E.J., 1994, ‘Submerged boundaries’, The Australian Surveyor, vol. 39,
no. 1, pp. 40-45
Hyde, C.C., 1945, International Law, Chiefly as Interpreted and Applied by the
United States
Jennings, R. and A. Watts (eds) 1992, Oppenheim’s International Law (9th
edition)
486 Bibliography
Jessup, P.C., 1928, ‘The Palmas Island Arbitration’, American Journal of Inter-
national Law, vol. 22, no. 4, pp. 735-752
Johnson, D.H.N., 1951, ‘Acquisitive prescription in international law’, British
Year Book of International Law, vol. 27, p. 332
Lauterpacht, H., (1958) Oppenheim’s International Law, Vol. 1, 8th ed.
Lee, H.K., 2005, ‘Mapping the law of legalizing maps: the implications of the
emerging rule on map evidence in international law’, Pacific Rim Law and
Policy Journal, vol. 14
Marston, G., 1991, ‘United Kingdom materials on international law 1991’, British
Year Book of International Law, vol. 62, pp. 535-722
McCaffrey, S.C., 2001, The Law of International Watercourses: Non-Naviga-
tional Uses
McCorquodale, M. and R. Pangalangan, 2001, ‘Pushing back the limitations
of territorial boundaries’, European Journal of International Law, vol. 12,
no. 5, pp. 867-888
McNair, A., 1961, The Law of Treaties
Nicholson, H., 1933, Peacemaking
Oppenheim’s International Law, 9th Ed 1996 (eds. RY Jennings and AD Watts)
Oxman, B.H., 2006, ‘The territorial temptation: A siren song at sea’, American
Journal of International Law, vol. 100, no. 4, pp. 830-851
Prescott, V. and C. Schofield, 2005, The Maritime Political Boundaries of the
World (2nd edition)
Ratner, S.R., 1996, ‘Drawing a better line: Uti possidetis and the borders of new
states’, American Journal of International law, vol. 90, no. 4, pp. 593-624
Reisman, M., 1999, ‘International decisions: The Government of the State of
Eritrea and the Government of the Republic of Yemen’, American Journal
of International Law, vol. 93, no. 3, pp. 668-682
Richardson, I.M., 1957, ‘New Zealand’s claims in the Antarctic’, New Zealand
Law Journal 38-42
Rothwell, D., 1996, The Polar Regions and the Development of International
Law
Rushworth, D., 1998, ‘Mapping in support of frontier arbitration: maps as evi-
dence’, International Boundaries and Research Unit (IBRU) Boundary and
Security Bulletin, vol. 5, no. 4
Sahurie, E.J., 1992, The International Law of Antarctica
Sandifer, D.V., 1975, Evidence Before International Tribunals (2nd ed.)
Shalowitz, A.L., 1964, Shore and Sea Boundaries, US Department of Commerce:
Publication 10-1, Vol. I
Shaw, M.N., 1986, Title to Territory in Africa: International Legal Issues
———, 2007, ‘Title, control and closure? The experience of the Eritrea-Ethio-
pia Boundary Commission’, International and Comparative Law Quarterly,
vol. 56, no. 4, pp. 755-796
Bibliography 487
Svarlien, O., 1960, ‘The sector principle in international law and practice’, Polar
Record, vol. 10, pp. 248-263
Triggs, G.D., 1986, International Law and Australian Sovereignty in Antarctica
(Legal Books Sydney)
———, 2006, International Law: Contemporary Principles and Practices.
Waldock, C.M.H., 1948, Disputed Sovereignty in the Falkland Island Dependen-
cies, BYBIL 311-353
Watts, A.D., 1993 International Law and the Antarctic Treaty System
Weissberg, G., 1963, ‘Maps as evidence in international boundary disputes:
a reappraisal’, American Journal of International Law, vol. 75, no. 4, pp.
781–803
Westlake, I., 1910, International Law (2nd edition)
LEGAL MATERIALS
TABLE OF CASES
Aegean Sea Continental Shelf Case (Greece v Turkey) [1978] ICJ Rep 3 (‘Aegean
Sea Continental Shelf case’)
Antarctica Case (UK v Argentina) [1956] ICJ Rep 12
Argentine-Chile Frontier Case (1967) 38 ILR 10
Award of the Alaska Boundary Tribunal (Great Britain v USA) (1903) XV RIAA
481 (‘Alaska Boundary Award’)
Beagle Channel arbitration between the Republic of Argentina and the Republic
of Chile (1977) 52 ILR 999 (‘Beagle Channel arbitration’)
Behring Sea Arbitration (1833) US Parl. Pap. No. 1
British Guiana v Venezuela Boundary Arbitration (1899–1900) 92 BFSP 160
(‘British Guiana Boundary arbitration’)
Case Concerning East Timor (Portugal v Australia) [1995] ICJ Rep 90 (‘East
Timor case’)
Case Concerning Kasikili/Sedudu Island (Botswana v Namibia) [1999] ICJ Rep
1045 (‘Kasikiki/Sedudu Island case’)
Case Concerning Maritime Delimitation and Territorial Questions between Qatar
and Bahrain (Merits) [2001] ICJ Rep 40
Case Concerning Right of Passage Over Indian Territory (Portugal v India)
(Merits) [1960] ICJ Rep 6 (‘Rights of Passage case’)
Case Concerning Sovereignty over Pulau Ligitan and Pulau Sipadan (Indonesia/
Malaysia) (Merits) [2002] ICJ Rep 625
Case Concerning the Frontier Dispute (Benin/Niger) [2005] ICJ Rep 90 (‘Benin/
Niger’)
Case Concerning the Frontier Dispute (Burkina Faso v Republic of Mali) [1986]
ICJ Rep 554 (‘Frontier Dispute case’)
Case Concerning the Frontier Dispute (Burkina Faso v Republic of Mali)
(Requests for the Indication of Provisional Measures) [1986] ICJ Rep 3
(‘Frontier Dispute, provisional measures’)
Case Concerning the Gabčíkovo-Nagymaros Project (Hungary v Slovakia) [1997]
ICJ Rep 7 (‘Gabčíkovo-Nagymaros case’)
490 Legal Materials
Case Concerning the Land and Maritime Boundary between Cameroon and
Nigeria (Cameroon v Nigeria) [2002] ICJ Rep 303 (‘Cameroon v Nigeria’)
Case Concerning the Land, Island and Maritime Frontier Dispute (El Salvador/
Honduras) [1992] ICJ Rep 351 (‘Land, Island and Maritime Dispute case’)
Case Concerning the Legal Status of Eastern Greenland (1933) PCIJ Rep, Series
A/B, No. 53 (‘Eastern Greenland case’)
Case Concerning the Temple of Preah Vihear (Cambodia v Thailand) (Merits)
[1962] ICJ Rep 6 (‘Temple of Preah Vihear case’)
Case Concerning the Territorial Dispute (Libyan Arab Jamahiriya/Chad) [1994]
ICJ Rep 6 (‘Territorial Dispute case’)
Case Relating to the Territorial Jurisdiction of the International Commission of
the River Oder (1929) PCIJ, Ser. A, No. 23 (‘River Oder case’)
Chamizal Arbitration 1911, American Journal of International Law, vol. 5,
no. 3, pp. 782-833
Colombia-Venezuela Arbitration (1922) 1 RIAA 223
Delimitation of the Continental Shelf between the United Kingdom of Great
Britain and Northern Island and the French Republic (1977) 18 RIAA 3
(‘Anglo-French Continental Shelf case’)
Diversion of Water from the River Meuse case (Netherlands v Belgium) (1937)
PCIJ, Ser. A/B, no. 70
Eritrea v Yemen, Award of the Arbitral Tribunal in the First Stage of the Pro-
ceedings (Territorial Sovereignty and Scope of the Dispute) (1998) 114 ILR 1
(‘Eritrea v Yemen (Phase One)’)
Eritrea-Ethiopia Boundary Commission Award (Delimitation Stage), April 2002
(‘Eritrea-Ethiopia dispute’)
European Community Arbitration Commission on Yugoslavia (Badinter Com-
mission) Opinion No. 2 [1992] 92 ILR 167
European Community Arbitration Commission on Yugoslavia (Badinter Com-
mission) Opinion No. 3 [1992] 92 ILR 170
Fennings v Lord Grenville (1808) 1 Taunt [Taunton’s Reports] 241 at 248
Fisheries Case (United Kingdom v Norway) [1951] ICJ Rep 116 (‘Anglo-Nor-
wegian Fisheries case’)
Frontier Land case (Belgium v Netherlands) 1959
Grisbådarna Case (Norway v Sweden) (1909) Hague Court Reports 121
Honduras Borders Arbitration (Honduras v Guatemala) (1933) 2 RIAA 1307
In the Matter of the Boundary between the Dominion of Canada and the Colony
of Newfoundland in the Labrador Peninsula [1927] 2 DLR 401 (‘Canada-
Newfoundland Boundary Dispute’)
Iowa v Illinois 147 US 1 (1893)
Island of Palmas Case (US v Netherlands) (1928) 2 RIAA 829
Lake Lanoux Arbitration (France v Spain), Arbitral Tribunal (1957) 24 ILR 101
(‘Lac Lanoux arbitration’)
Legal Materials 491
1945 Charter of the United Nations, 59 Stat. 1031, 26 June (‘UN Charter’)
1945 Statute of the International Court of Justice (‘ICJ Statute’)
1959 Antarctic Treaty
1960 Declaration on the Granting of Independence to Colonial Countries,
Adopted by General Assembly resolution 1514 (XV) of 14 December
1961 Resolution Concerning the Utilisation of Non-Maritime Waters for Pur-
poses Other than Navigation, International Law Institute
1963 Charter of the Organisation of African Unity
1964 Resolution 16(1) of the OAU Conference of Heads of State and Govern-
ment, Cairo
1966 International Covenant on Civil and Political Rights, G.A. res. 2200A
(XXI), 21 U.N. GAOR Supp. (No. 16) at 52, U.N. Doc. A/6316 (1966),
999 U.N.T.S. 171, entered into force Mar. 23, 1976
1966 Helsinki Rules on the Uses of the Waters of International Rivers, Inter-
national Law Association
1967 Treaty on Principles Governing the Activities of Outer Space
1969 Vienna Convention on the Law of Treaties, entered into force on 27
January 1980, United Nations Treaty Series, vol. 1155, p. 331
1976 Convention on the Regulation of Water Management Issues of Boundary
Waters, 31 May
1979 United Nations Moon Treaty 1979
1980 General Treaty of Peace (El Salvador and Honduras)
1982 United Nations Convention on the Law of the Sea (‘UNCLOS’)
1988 Convention on the Regulation of Antarctic Mineral Resources, 27 ILM
895 (‘Minerals Convention’)
1990 Security Council Resolution 662, S/RES/0662 (1990), 2934th meeting,
9 August
1991 European Community Guidelines on the Recognition of New States in
Eastern Europe and the Soviet Union
1994 United Nations Draft Declaration on the Rights of Indigenous Peoples,
UN Doc E/CN.4/Sub.2/1994/2/Add.1 (1994)
1997 United Nations Convention on the Law of the Non-navigational Uses of
International Watercourses (not in force)
Name Index 493
NAME INDEX
Cook, D., 67, 86, 378, 302 Galeotti, M., 367, 368, 370
Cornish, V., 39, 47 Garrett, W.E., 4, 20, 24, 28
Cox, H.E., 360, 361, 371 Gay, B., 91, 133
Cree, D., 363, 370 Geelan, P.J.M., 102, 133
Crowder, M., 21, 295, 297, 314 Genicot, L., 52, 53, 86, 350, 370
Curzon of Kedleston, 5, 20, 30, 35, 47, GilFilan, S.C., 369, 370
59, 86 Goldsmid, F.J., 269, 270, 288
Gould, L.M., 384, 407
David, E., 156 Great Britain, 385, 389, 407
David, S., 46, 48 Grey, A.L., 338, 346
Davidson, G., 243, 262 Grotius, H., 216
Davies, C.C., 32, 48, 324, 325, 346 Grove, G., 52, 88, 349, 372
Davis, S.L., 60, 86 Gupta, S., 335, 346
De Arechago, E.J., 248, 262 Gyorgy, A., 8, 20
Deeley, N., 69, 70, 86, 121, 133, 374, 376,
382 Hancox, D., 341, 346
Dennis, W.J., 30, 48, 97, 111, 133 Hankey, Lord, 363, 370
Department of Commerce, 68, 86 Hardin, G.H., 221
Dewdney, J., 265, 288 Harisse, H., 245, 262
Doherty, E.W., 49 Harrison, J.A., 29, 48
Dorion, H., 10, 20 Hartshorne, R., 12, 20, 107, 133
Douglas, E.M., 234, 235, 236, 242, 243, Haushofer, K., 7, 8, 9, 10, 20
244, 245, 262 Hayat-khan, A., 326, 346
Duby, G., 86, 370 Hayton, R.D., 392, 407
D’Urville, D., 102, 133 Hertslet, E., 4, 7, 20, 38, 48, 53, 63, 64,
76, 86, 87, 117, 127, 133, 350, 355, 356,
East, W.G., 30, 35, 48, 107, 133 357, 358, 371
Ekert, R.D., 13, 20 Highet, K., 194
Eloff, G.G., 92, 133 Hildesheimer, E., 86, 370
Engel, J., 42, 86, 349, 370 Hill, N.L., 6, 20, 62, 87, 118, 120,
Engelfield, G., 365, 370 133
Ermen, E. van, 234, 238, 262 Hinks, A.R., 10, 20, 62, 87, 118, 120,
Escalada, V., 17, 18, 20 133, 363, 370, 371, 384, 408
Hobbs, W.H., 384
Fairlie, L.D., 368, 370 Hogben, S.J., 40, 48
Fernando-Armesto, F., 294, 295, 298, Hohnel, L. von, 33, 47
314 Holdich, T.H., 38, 48, 69, 87, 267, 288
Fernstrom, J.R., 49 Horlin, E.J., 219
Fifer, J.V., 35, 48 House, J.W., 236, 262
Fischer, E., 34, 49, 110, 133 Hoyle, P., 91, 133
Fisher, H.A.L., 133 Hsu, I.C.Y., 321, 346
Flood, E.T., 83, 86 Huber, J., 150, 152, 153, 158, 161, 173,
Fraser-Tytler, W.K., 325, 346 178, 181, 184, 185, 187, 194
Freeberne, M., 58, 86 Hulbert, H.B., 333, 346
Friters, G.M., 323, 346 Hunt, J., 379
Fullard, H., 52, 88, 349, 372 Huntingford, G.W.B., 34, 48
Name Index 495
Hupchick, D.P., 360, 361, 371 Lagergren, G., 158, 210, 334, 346
Huth, P.K., 138 Lamar, H., 43, 48
Hyde, C.C., 191, 220, 407 Lamb, A., 11, 21, 87, 327, 329, 335, 346
Hydrographer of the Navy, 103, 133, Lambert, A.F., 70, 87
262 Lapradelle, P. de, 9, 12, 21, 60, 87
Lattimore, O., 38, 48
Indian Foreign Department, 107, 133, Lauterpacht, E., 169
268, 270, 271, 288 Lauterpacht, H., 151, 152
International Boundary Research Unit, 4, Laws, J.B., 66, 87
330, 346 Leathes, S., 52, 89, 95, 106, 135
International Court of Justice, 139, 143, Le Lannou, M., 365, 371
144, 145, 146, 147, 148-50, 156, Lovering, J.F., 383, 408
158-162, 164-166, 169, 171, 173, Luguenin, P., 77, 87
175-178, 180, 183, 185, 187, 188, 192,
195-198, 201-204, 206, 211, 218, 220, Macartney, J.C., 67, 86, 378, 382
228, 230, 384 Mackinder, H.J., 51, 87
International Crisis Group, 330, 346 Maclean C.S., 270, 271
Ireland, G., 4, 20, 21, 76, 87, 103, 104, Maling, D.H., 299, 314
113, 131, 133, 246 247, 248, 249, 250, Malloy, W.M., 242, 263
262 Mancall, M., 320, 346
Irwin, G., 379, 382 Mantoux, P., 363, 371
Ivekovic, R., 75 Marchant, A., 118, 134, 263
Margo, R.D., 21
Jennings, R., 139, 156, 184 Marston, E. de M., 21, 144
Jessup, P.C., 185 Martin, E. de M., 15, 18
Johnson, D.H.N., 149, 163 Martin, L., 384, 408
Johnson, R.W., 11, 21 Martin, P., 15, 17, 18, 21
Jones, B., 20 Matte, H., 14, 15, 21
Jones, S.B., 9, 10, 21, 60, 66, 107, 133, McCaffrey, S.C., 122, 134, 215, 217, 221,
235, 236, 262 305, 314
McCarthy, J., 77, 87
Karaeve, Z., 346 McClean, J.D., 21
Kasule, S., 293, 295, 314 McCorquodale, M., 139, 140
Kerchove, R. de, 304, 314 McCune, S., 338, 346
Kerr, D.G.G., 48 McEwen, A.C., 9, 21
Kim, C.I.E., 333, 346 McMahon, H., 270, 288
Kim, H., 333, 346 McNair, A., 400
Kinahan, J., 21, 123, 127, 133 Menke, T., 52, 87, 349, 371
King, L.W., 69, 87, 326, 346 Milenkowski, M., 365, 366, 367, 371
Kirk, D., 108, 133 Miller, H., 237, 238, 239, 240, 241,
Kirk, W., 48 263
Kiss, G., 365, 371 Ministère des Colonies, 31, 48
Klemencic, M., 11, 21, 365, 367, 371 Mitchell, R.D., 27, 48
Konrad, V.A., 26, 48 Mitteis, H., 350, 371
Kristof, L.A.D., 24, 48 Mostert, N., 43, 44, 45, 46, 48, 76,
Kroell, J., 15, 21 87
496 Name Index
Stephan, J.J., 344, 347 United Kingdom, 77, 78, 79, 88, 382
Stewart, J., 328, 347 United Nations, 5, 14, 16, 17, 22
St John, R.B., 11, 22, 91, 111, 113, 114, United States, Department of State, 22,
115, 116, 135 389
Stone, K.H., 29, 49 United States Naval War College, 388,
Storey, I., 342, 347 391, 408
Stott, P.M., 67, 86, 378, 382
Sykes, P., 270, 289 Vancouver, G., 243, 264
Straight, S., 27, 49 Van der Veur, P., 377, 382
Strayer, J.R., 52, 88, 350, 353, 372 Varela, L.V., 68, 88
Vatell, E. de, 56, 88
Taberd, J.L., 102, 135 Vietnam Foreign Ministry, 102, 135
Talevski, J., 365, 366, 367, 371
Tarling, N., 380, 382 Wagner de Reyna, A., 114, 135
Tate, J.J.M., 82, 88 Waldron, A., 12, 22
Temperley, H.W.V., 51, 88, 106, 109, 110, Ward, A.W., 52, 89, 95, 106, 108, 135
123, 277, 289, 359, 372 Ward, F.K., 135
The Geographer, 4, 22, 62, 88, 104, 118, Watts, A.D., 139, 156, 397
119, 135, 247, 263, 264, 272, 273, 274, Weigend, G.G., 93, 135
275, 276, 277, 278, 281, 282, 285, 289, Weigert, H.W., 34, 49
300, 301, 302, 303, 304, 305, 306, 307, Weissberg, G., 201, 210, 211
314, 315, 368, 372 Westlake, I., 217
The Grolier Society of Victoria, 28, 49 White, T., 20
The Observer, 9, 22 Whittlesey, D., 241, 249
The Times, 396 Whyte, B.R., 40, 49, 74, 75, 76, 89, 334,
Thompson, L., 43, 48 335, 336, 347
Times Books, 29, 49, 74, 88, 289 Wilkinson, H.R., 39, 49, 282
Topalovic, D., 366, 370 Wilkinson, J.C., 289, 298, 315
Toynbee, A.J., 277, 289 Wiskemann, E., 73, 89
Treharne, R.F., 52, 88 Wright, L.R., 380, 382
Triggs, G.D., 100, 135, 143, 148, 150,
151, 152, 153, 155, 156, 157, 160, Yates, A.C., 324, 347
162, 164, 169, 174, 178, 179, 180, Yesner, D.R., 31, 49
181, 183, 184, 186, 392, 393, 395, Yi, W.S.M., 320, 347
396, 400, 401 Yule, H., 328, 347
Trotter, J.K., 59, 88
Turner, F.J., 24, 25, 26, 49 Zamoyski, A., 7, 22, 95,135, 352, 353,
Twitchett, D.C., 102 372
SUBJECT INDEX
Accretion, erosion and avulsion, 174, 219 Anglo-Norwegian Fisheries case, 156
Acquiescence, 176-8 Angola, 39, 61, 106
Acquisition by prescription, 162-6 Antarctica, 51, 102
Acquisition of territorial sovereignty, Antarctic Treaty Regime, 1959, 398-400
148-9 Antecedent boundary, 12
Acquisitive possession, 163 Arabian peninsula, 265-6
Activities must be referable to Arbitrations, 92
sovereignty, 156-7 Argentina, 68
Acts of effective occupation in Antarctica, Argentina-Bolivia boundary, 256-7
394 Argentina-Brazil boundary, 247
Afghanistan, 34, 62, 63, 66, 69, 98, 100, Argentina-Chile boundary, 247-8
324 Argentina-Paraguay boundary, 248-9,
Afghanistan-Iran boundary, 269-70 260-1
Afghanistan-Pakistan boundary, 69, 324-7 Arguments supporting territorial claims,
Afghanistan-Russia boundary, 53, 58, 66, 100-111
323-4 Armenia, 74, 368
African adjudicated boundaries, 306-12 Asian boundaries in the British sector
African adjudications in the 19th century, from Russia to Thailand, 323-31
306-9 Asian boundaries in the division of
African adjudications 1900-2002, 309-12 British and French Colonies after 1945,
African boundary changes after 1918, 299 333-7
African geometric boundaries, 299-302 Asian boundaries in the
African water boundaries, 306-12 Franco-Chinese-Japanese sector,
Africa’s indigenous states, 292-3 331-3
Aims in territorial disputes, 99-100 Asian boundaries in the
Air space, 15-19 Sino-Russian sector from Vladivostok
Alaska, 24, 31 to Afghanistan, 316-23
Alaska-Canada boundary, 233 Asia’s offshore territorial disputes,
Albania, 54, 358, 364 341-4
Alsace-Lorraine, 51, 95, 359 Australia, 24, 28
Anglo-Dutch boundaries on Borneo, Australian claims to Antarctica, 385-8
379-81 Austria, 7, 359, 361
Anglo-Dutch division of New Guinea, Azerbaijan, 5, 74, 368
377-8
Anglo-German division of the Solomon Balkans, 53, 73
Islands, 373 Bangladesh, 5, 24, 75
Subject Index 499
Pakistan, 8 Rhodesia, 3, 4, 62
Papua New Guinea, 67 River boundaries, 216-7
Persian Gulf, 265 Bank, 217
Peru, 97, 101 Medium filum aquae, 216
Plebiscite, 111 Thalweg, 217
Poland, 53, 74, 359, 362, 364 River boundaries and international law,
Polish and Czechoslovakian case, 194 215-31
Political frontiers, 30-40 River Niger, 71
political frontiers of contact and Role of acts subsequent to the date of
separation, 30-3 title, 172
political frontier markers, 33-4 Romania, 53, 54, 356, 364
political frontiers and the evolution of Rome’s control of North Africa, 293-4
landscapes, 39-40 Russia, 2, 5, 8, 52, 54, 57, 362, 364, 365,
political organization within frontiers, 367
35-39 Russia-Turkey boundary, 273-4
reduction of frontiers, 34-5 Russian Federation, 74
Portugal, 3, 4, 9, 52, 56, 57, 61, 62, 65,
105 Sanjak of Alexandretta, 225, 276
Positional boundary disputes, 116-23 Saudi Arabia, 27, 38
Preah-Vihear case, 204-5, 206, 208-10 Saudi Arabia’s boundaries from Qatar to
Principles developed by international Yemen, 285-7
courts and tribunals, 226-31 Saxony, 352
Principles of delineation of a river Schleswig, 359
boundary, 216-9 Scramble for Africa, 8-9
Principles of equity and delineation of Secession, 75
river boundaries, 219-20 Sector principle and territorial boundaries
Probative weight given by international in Antarctica, 394-6
tribunals to maps, 194-203 Self-determination of people, 181-2
Procedures for implementation of Serbia, 2, 54, 356, 358, 367
obligations, 224-5 Settlement frontiers, 23-30
Protests against Antarctic sovereign primary settlement frontiers, 24-7
claims, 398 secondary settlement frontiers, 27-30
Pulau Ligitan-Pulau Sipadan case, 207-8 Shleswig, 359
Shomburgk Line, 249
Qatar-Bahrain case, 197-8, 205-6 Slave trade, 30-1, 296
Slovenia, 365
Rann of Kutch, 157-8, 196, 208, 210 South Africa, 8, 25, 28, 65
Recognition, 175-6 South America, 52
Recognition, aquiescence and estoppel, South America: arbitration, 247-56
174-5 South American geographical boundary
Recognition in Antarctica, 397 problems, 256-61
Relationships between title, the principles South Tyrol, 93
of uti possidetis and effectivités, 171-2 Soviet Union, 74
Relativity of title, 150 Spain, 5, 52
Relict boundary, 12 Special vigilance, 195
Reluctance of international tribunals to Spheres of influence and interest, 38-9
accord probative value to maps, 196-8 Spratly Islands, 5, 341-3
504 Subject Index