Counter-Memorial of the Republic of Namibia

INTERNATIONAL COURT OF JUSTICE

CASE CONCERNING KASIKILI/SEDUDU ISLAND

(BOTSWANA/NAMIBIA)

COUNTER-MEMORIAL

OF

THE REPUBLIC OF NAMIBIA

VOLUME I

COUNTER-MEMORIAL

28 NOVEMBER 1997

Table of Contents

Introduction

CHAPTER I SCOPE OF THE QUESTION SUBMITTED TO THE COURT

CHAPTER II THE DETERMINATION OF THE MAIN CHANNEL OF THE

CHOBE RIVER WITHIN THE MEANING OF THE ANGLO-GERMAN TREATY OF
1890

A. The Scientific Evidence Establishes Unequivocally that the Southern Channel is the Main
Channel

B. In Terms of Navigability 'in relation to the needs of the regional economy,' the Southern
Channel (and not the Northern) Qualifies as the Main ChannelC. Nothing in the Language of the 1890 Treaty or the General Principles of International
Law Warrants the Rejection of the Determination that the Main Channel of the Chobe River is
the Southern Channel

1. The language of the Anglo-German Treaty of 1890

2. General principles of international law concerning river boundaries contemporaneous with
the 1890 Treaty did not embody the thalweg doctrine

3. International law sources of the relevant period demonstrate that the thalweg was to be
identified by reference to the current or flow of the channel and not merely its depth

a. Authoritative contemporaneous definitions

b. Origins and evolution of the thalweg concept and its relation to navigation

D. Conclusions as to the Identification of the Main Channel within the Meaning of the 1890
Treaty

CHAPTER III. THE SUBSEQUENT CONDUCT OF THE PARTIES TO THE
ANGLO-GERMAN TREATY OF 1890 AND THEIR SUCCESSORS IN TITLE

A. Exercise of Jurisdiction over Kasikili Island by South Africa in the 1970s

B. Alleged Interruptions in the Unbroken Record of Recognition and Acquiescence by

Botswana

1. The Eason Report

2. The Trollope-Dickinson arrangement

3. The 1984-1986 discussions between Botswana and South Africa

a. Neither South Africa nor Botswana had legal power to enter into an agreement concerning
the boundary between Namibia and Botswana

i. South Africa

ii. Botswana

iii. No agreement affecting the boundaries of Namibia could have been concluded without UN
General Assembly approval

b. The exchanges between the two countries did not purport to establish and did not in fact

eventuate in a bilateral agreement concerning the location of the boundary at Kasikili Island

i. The legal capacity of the two delegations

ii. The content of the discussionsiii. The subsequent conduct of the parties

c. The joint survey report is entitled to no weight as an expert opinion

C. Conclusion as to the Subsequent Conduct of the Parties

CHAPTER IV. MAPS AS EVIDENCE OF SUBSEQUENT CONDUCT OF THE
PARTIES

A. General Considerations as to the Map Evidence in this Case

1. The relevance and weight of cartographic evidence

2. The scale of the relevant maps

3. Off9icial as opposed to 'private' or non-official maps

B. Analysis of the Maps Relied on by Botswana

1. German maps

a. 'Kriegskarte von Deutsch-Sudwestafrika [sic] 1904'

b. 'Von Frankenberg, Karte des Caprivi Zipfels, Blatt I, 1912'

c. The German maps that Botswana does not mention

2. British maps

a. 'British War Office Map of Protectorate, 1933; Geographical Section, General Staff, No.
3915'

b. 'Bechuanaland, 1965: Directorate of Overseas Surveys (D.O.S. 847) (Z 462)'

c. 'Africa 1:2,000,000 War Office GSGS 2871 Sheet Rhodesia'

3. South Africa maps

a. 'South African Official Map 1:250,000, 1949'

b. 'South African Map Compiled by JARIC, 1:100,000 (c. 1974)

4. Third country maps

C. Conclusions as to the Map Evidence

Conclusion and Submissions of this Counter-Memorial

List of FiguresGlossary to Abbreviated Citations

Index to Annexes

Counter-Memorial of the Republic of Namibia

Introduction

1. This Counter-Memorial of Namibia is submitted pursuant to the Order of the Court of 24
June 1996, fixing 28 November 1997 as the date for filing by each of the parties of a Counter-
Memorial.

2. In its Memorial, Namibia based its claim to Kasikili Island on two separate grounds: first,
by interpretation of the language of the Anglo-German Treaty of 1890, which establishes that
the 'main channel' of the Chobe River around Kasikili Island is the southern channel; and
second, on the basis of prescription evidenced by possession, use and exercise of jurisdiction
over the Island, uninterrupted for almost a century after 1890 without objection or
remonstrance by Botswana.

3. Namibia's position is that the issue of treaty interpretation is to be decided on the basis of
objective considerations of a scientific character, and that the scientific determinant of the
main channel is the channel that carries the largest proportion of the annual flow of the river.
1
Botswana accepts this test. Namibia has shown conclusively through the expert testimony of
Professor W.J.R. Alexander in the Alexander Report that almost all of the flow of the Chobe
River goes through the southern channel and virtually none through the northern channel.
3
Botswana's Memorial does not join issue on this point. Its expert's report is devoted almost
entirely to arguing the uncontested proposition that the river in the vicinity of Kasikili Island
has not changed its general shape and configuration since 1890. Botswana has presented no

scientific evidence that the northern channel carries any part of the flow or current of the
river.

4. Despite its acknowledgement of the decisive character of scientific criteria and evidence,
Botswana devotes its Memorial primarily to two wholly different propositions: first, that the
4
main channel is to be identified on the basis of navigability; and second, that under a
contemporaneous rule of general international law, the main channel is defined by the
5
thalweg, that is to say the deepest channel.

5. On the question of navigability, Namibia maintains that it is unreasonable and therefore
incorrect to apply the criterion of navigability to a river boundary over 300 kilometres in
length, nine-tenths of which is clearly notavigable. Botswana resolutely confines its

attention only to the last 50 kilometres of the Chobe River from the confluence with the
Zambezi River. But the river boundary established by the 1890 Treaty also runs along the
Chobe River westward to its juncture with the 18th parallel of south latitude, an additional
distance of over 250 kilometres. Over all of this stretch, the Chobe River is dry for much ofthe year and in many places for years on end. The criterion of navigability is simply irrelevant
to the river as a whole.

6. Even if navigability were to be considered the touchstone for determining the main
channel, navigability is to be judged - as Botswana itself admits - 'in relation to the needs of
6
the regional economy.' The foundation of the regional economy around Kasikili Island is
tourism. The water traffic in the Chobe River in this area consists of flat-bottomed boats
carrying tourists to view the game on the Island and on the south bank of the river further

west. The overwhelming bulk of this traffic is in the southern channel. Only occasionally and
incidentally do any of these boats go through the northern channel. Thus by Botswana's own
criterion, the southern channel is the navigable channel in relation to the needs of the regional
economy and, therefore, is the main channel.

7. On the question of the thalweg, Namibia shows that, contrary to Botswana's contention, at
the time of the Treaty there was no rule of general international law establishing the thalweg

as the boundary between riverine states. An examination of the writings of publicists and the
practice of states, especially in Africa, as evidenced by the treaties they concluded during the
period, shows that, although the thalweg was frequently said or used to indicate the boundary,
this usage was by no means of the generality, uniformity and consistency necessary to
establish a rule of international law capable of controlling the meaning of the words 'the
centre of the main channel.' Again, however, even if the thalweg concept is thought to be

relevant, Namibia shows that the core element of this concept was the connection of the
thalweg with the flow or current of the river. The factor of depth was derivative and
secondary. Moreover, the thalweg concept was designed to provide each of the riparians
access to the navigational benefits of the river, and, as noted above, these benefits pertain
exclusively to the southern channel. Thus, even if the thalweg determines the main channel, in
the case of the Chobe River around Kasikili Island the thalweg is in the southern channel.

8. As to the second basis of Namibia's title, Botswana contends that under the terms of the
question submitted by the parties the Court is confined to interpreting the Anglo-German
Treaty of 1890 and cannot consider the issues of prescription, recognition and acquiescence. 7

The argument falls of its own weight. A simple inspection of the submission shows that the
Court is asked to decide 'on the basis of the Anglo-German Treaty of 1st July 1890 and the
rules and principles of international law.' (emphasis added) Moreover, the Court is asked to

determine not only the boundary around Kasikili Island, but also 'the legal status of the
island.' Such a determination can not be confined solely to the examination and
interpretation of the Treaty.

9. Botswana cites the Eason Report, the Trollope-Dickinson arrangement and the 1985 joint
survey (all fully discussed in Namibian Memorial, Part Two, Chapter II) as claims of right or
public assertions of Botswana's title. On analysis, none of them bear that construction. These
incidents do not constitute interruptions or protests sufficient to defeat Namibia's continuous
occupation, use and exercise of jurisdiction over Kasikili Island.

10. Before responding in detail to Botswana's Memorial, Namibia wishes to call the Court's
attention once again to the peculiar characteristics of the Chobe River, as described in Chapter
I of Namibia's Memorial, The Geography of the Disputed Section of the Boundary. These
unusual characteristics have a direct bearing on the resolution of the issues before the Court.The Chobe River is not a watercourse carrying water more or less continuously from its
catchment area downstream to its mouth or the junction with another river. Unlike the major
European rivers and others that spring readily to mind, the Chobe River is an ephemeral river
that is dry over most of its length for much of the year. In the area of specific concern in this
case, it is part of a complex system closely associated with the Zambezi River to the north

that, in the high flow season, carries the flood waters of the Zambezi back into that river
below the Mambova Rapids. Botswana has disregarded these decisive peculiarities of the
Chobe River. It has confined its attention to the last 50 of the more than 300 kilometres over
which the Chobe River constitutes the border. And it relies almost exclusively on visual
evidence (such as maps and aerial photographs) and documentation (like the 1985 joint
survey) that reflect the situation in the dry season of the year. During that period, the Chobe
River in the vicinity of Kasikili Island looks much like a normal river, flowing through well-

defined channels in a readily visible direction. As shown in Part One, Chapter I of Namibia's
Memorial and in Chapter II(A) of this Counter-Memorial, that appearance is profoundly
misleading. In reality during the dry season, there is no flow in the Chobe River at all above
the Mambova Rapids, including both of the channels around Kasikili Island. The Chobe River
flows only when the Zambezi is in flood. Then, Kasikili Island, including the northern
10
channel, is inundated, and, as is shown in para. 30, infra, substantially all of the flow of the
river passes through the southern channel. These features of the geography and
hydrogeomorphology of the region, which are not apparent from the evidence relied on by
Botswana, dominate the problem of the identification of the main channel of the Chobe River

11. The organisation of this Counter-Memorial is as follows:

Chapter I establishes that the question submitted to the Court comprehends Namibia's claim
based on prescription, acquiescence and recognition.

Chapter II analyses the question of treaty interpretation on the basis of the scientific evidence

(Section A) and also (assuming but not admitting that they are relevant) from the perspectives
of navigability (Section B) and the thalweg concept (Section C).

Chapter III discusses the subsequent conduct of the parties in the period since the Treaty and
refutes the contention that the sovereignty of Botswana was claimed or recognised.

Chapter IV discusses the map evidence and demonstrates, contrary to Botswana's contention,

that there is a remarkable general consistency among the official maps of Namibia produced
by Germany, Great Britain, South Africa and the United Nations, the four entities that
exercised political power in the area from 1890 to 1984, showing the boundary as being in the
southern channel and Kasikili Island as being in Namibia.

Chapter I

THE SCOPE OF THE QUESTION SUBMITTED TO THE COURT

12. This Chapter responds to Chapter I of the Botswana Memorial, entitled 'The Nature of the
Dispute.'13. By the Special Agreement between the Republic of Botswana and the Republic of
Namibia dated 15 February 1996, the Court is asked

to determine, on the basis of the Anglo-German Treaty of 1st July 1890 and the rules and
principles of international law, the boundary between Namibia and Botswana around
11
Kasikili/Sedudu Island and the legal status of the island.

Namibia bases its claim to Kasikili Island on two distinct grounds: first, that the Treaty of

1890, when properly interpreted, attributes Kasikili Island to Namibia (see Part One of
Namibia's Memorial); and second, that the rules and principles of international law
concerning acquiescence, recognition and prescription establish a title to the Island for
Namibia entirely independent of the Treaty (see Part Two of Namibia's Memorial). Botswana
asserts, however, that the question is to be limited to the interpretation of the Treaty. The
clause referring to 'the rules and principles of international law' is dismissed as 'pleonastic' by
12
which Botswana evidently means that the clause adds nothing to the Treaty. Such a
limitation on the Court's jurisdiction under the Special Agreement is unsustainable. It is
gainsaid by the plain meaning of the words, by elementary principles of treaty interpretation
and by the manifest intention of the parties to settle the entire dispute between them.

14. Botswana bases its argument for restricting the scope of the question as defined in the
Special Agreement on an entirely different document, the Memorandum of Understanding,
containing the Terms of Reference of the Joint Team of Technical Experts (JTTE) established
13
by the parties in 1992 to seek a negotiated resolution of the dispute. Indeed, the Botswana
Memorial quotes these Terms of Reference verbatim, and some of them do seem to support an
14
inference that the JTTE's task was strictly, if impracticably, limited to treaty interpretation.

15. Whether or not the Terms of Reference of the JTTE are properly construed as thus
limited, however, any such limitations cannot be imported into the Special Agreement. The
important point for the present case is that none of the restrictive language in the JTTE Terms
of Reference appears in the question submitted to the Court. The parties have asked the Court

to decide 'on the basis of the Anglo-German Treaty of 1st July 1890 and the rules and
principles of international law.'5 (emphasis added) There is no warrant whatever for giving
all the weight to one branch of this formula and none to the other. If anything, the omission of

the restrictive language in the JTTE Terms of Reference from the question submitted to the
Court gives rise to the reverse inference that no such limitation was intended. The plain
language of the question requires the Court to consider any evidence or submissions of the
parties grounded in general rules and principles of international law equally with submissions
based on the 1890 Treaty.

16. Botswana's attempt to treat the reference to the 'rules and principles of international law'
as if it were not included in the Special Agreement contravenes fundamental rules of treaty
interpretation. The Botswana Memorial suggests that the phrase refers only to the rules and
principles concerning treaty interpretation.6 But Botswana itself acknowledges that

international law rules concerning treaty interpretation are comprehended in the first clause of
the question referring to the Treaty of 1890.17 Well-known principles of interpretation,

however, forbid treating particular language in the text as redundant or repetitive when the
words can be given an independent meaning - as they can in the present case withoutdifficulty. To cite only one prominent authority, Oppenheim's International Law Ninth
Edition states:

The parties are assumed to intend the provisions of a treaty to have a certain effect, and not to
be meaningless. . . . Therefore, an interpretation is not admissible which would make a
18
provision meaningless, or ineffective.

As the Court said in construing the Question Submitted in the Case Concerning the

Continental Shelf, 'The Court must not exceed the jurisdiction conferred upon it by the
Parties, but it must also exercise that jurisdiction to its full extentSimilarly, in the Case
Concerning Maritime Delimitation and Territorial Questions between Qatar and Bahrain, the

Court said, 'Any other interpretation would encounter serious difficulties: it would deprive the
phrase of its effect and could well, moreover, lead to an unreasonable result.'20

17. The Court readily disposed of substantially the identical question in interpreting the
Special Agreement between Great Britain and Albania submitting the Corfu Channel case for
adjudication.21 The question was:

Is Albania responsible under international law for the explosions which occurred on the 22nd
October 1946 in Albanian waters . . . and is there any duty to pay compensation? 22

Great Britain asked the Court to determine the amount of compensation owed, but Albania
argued that under the Special Agreement the Court was limited to determining whether any
duty to pay compensation existed, rather than the quantum of such duty. The Albanian
position would have rendered the disputed phrase - 'is there any duty to pay compensation' -
23
'pleonastic,' since, as the Court pointed out, the duty to pay compensation followed
ineluctably from a determination that Albania was responsible under international law for the
explosions. Thus, to have adopted the Albanian position would have been to read the relevant
language out of the treaty. It would have been given no independent significance. This the

Court refused to do. It interpreted the Special Agreement to require it to determine the amount
of the compensation owed, even though the question submitted did not explicitly mention
amounts. In this case, however, the language of the question is clear. The Court is to decide
'on the basis of the Anglo-German Treaty of 1st July 1890 and the rules and principles of
international law.' (emphasis added) It is hard to imagine words more apt to express the
intention of the parties that in addition to the interpretation of the Treaty itself, the Court is to
consider all the applicable rules and principles of international law in coming to its decision.

18. Moreover, the Court is asked to determine not only the location of the boundary between
Namibia and Botswana at Kasikili Island, but also 'the legal status of the island.' In this
respect, the wording of the question here is to be contrasted with the language of the question
submitted in the Minquiers and Ecrehos case, which requested the Court 'to determine
whether the sovereignty over the islets and rocks . . . belongs to the United Kingdom or the
24
French Republic' without more. In the present case, the question, by asking generally for a
determination of 'the legal status of the island,' permits the Court to declare any legal rights in
the Island, whether or not treaty-based, as they may emerge from the evidence in the light of

the submissions of the parties.19. The parties did not intend to confine the Court to a merely technical exercise in treaty
interpretation. It was their purpose to settle the entire dispute between them concerning the
status of Kasikili Island. This purpose should not be frustrated by placing artificial limitations
on the jurisdiction of the Court that contradict the ordinary meaning of the language of the
Special Agreement.

Chapter II

THE DETERMINATION OF THE MAIN CHANNEL OF THE CHOBE RIVER
WITHIN THE MEANING OF THE ANGLO-GERMAN TREATY OF 1890

20. This chapter replies to Botswana's contentions with respect to the interpretation of the
1890 Treaty, contained primarily in Chapters V(b) and VII of the Botswana Memorial,
entitled respectively 'The Interpretation of the Anglo-German Agreement of 1890' and 'The
Issue of Fact: The Main Channel is to the North and West of Kasikili/Sedudu Island.'

21. Both parties are agreed that on the question of treaty interpretation, the task of the Court is
to determine whether the northern or the southern channel of the Chobe River around Kasikili
Island is the 'main channel.' Both are also agreed that this is a question of scientific fact to be
resolved on the basis of expertise in hydrology, geology and hydrogeomorphology.

22. In this chapter, Namibia demonstrates that, contrary to Botswana's contentions:

• first, there is no warrant in fact or law for saying that the main channel must be
equated with the navigable or deepest channel. On the contrary, as a factual and
scientific matter, the 'main channel' is the one that carries the largest proportion of the
flow of the river; and
• second, there is no warrant as a matter of fact for saying that the northern channel is
navigable and the southern channel is not.

23. Section A of this Chapter demonstrates that on the basis of scientific evidence, the
southern channel is the main channel, because it carries substantially all of the flow of the
Chobe River. Despite its recognition that the question is one of scientific fact, Botswana
adduces virtually no scientific evidence relating to the hydrology and hydrogeomorphology of
the river, the very factors its own expert says are decisive. Its argument is chiefly based on (i)
the comparative navigability of the two channels and (ii) a supposed contemporaneous rule of

general international law that for navigable rivers the deepest channel or thalweg marks the
boundary between riparian states. Although Namibia maintains that these considerations are
not relevant to determining the main channel of the Chobe River, it addresses them in this
Chapter of its Counter-Memorial.

24. Section B shows that navigability is to be judged 'in relation to the needs of the regional
economy,' which in this case is based almost exclusively on tourism. Substantially all of the

tourist traffic plies the southern channel, and practically none goes along the northern channel.
Thus, in terms of navigability, the southern channel is the main channel.

25. Section C examines the legal materials roughly contemporaneous with the 1890 Treaty
and demonstrates that no rule of general international law indicates, much less requires, thatthe northern channel be designated as the main channel. On the contrary, the clear
understanding of the thalweg concept at that time was that it corresponds to the deepest
channel with the strongest flow, in this case the southern channel.

A. The Scientific Evidence Establishes Unequivocally that the

Southern Channel is the Main Channel

26. A comparison of the two Memorials reveals that both parties are in agreement as to the
basic nature of the interpretative task before the Court. The Botswana Memorial begins its
concluding summary of its case as follows:

First:

The central question is the interpretation and application of the words 'main channel' of the
River Chobe. These words involve a reference to a question of fact and, in so far as may be
necessary, a question of scientific fact, calling for expertise in hydrology, geology and
25
hydrogeomorphology.

The Namibian Memorial endorsed this approach, quoting with approval a passage from
Botswana's submission to the JTTE that is identical to the one cited above and adducing
26
decisional authority to support the position.

27. The parties are also in close accord on the criteria to be applied to resolve this 'question of
scientific fact.' Dr. F.T.K. Sefe, Botswana's expert, states them as follows:

The word 'main' has both hydrological and geomorphological connotations relating to the
velocity of flow, hence the discharge. So it connotes the ability of the river to transport debris
in terms of particle size (i.e. its competence[Mackin, 1948]), and the maximum load
(sediment of particular grain size) a stream can carry (i.e. its capacity as defined by Gilbert
[1914]). Both the competence and capacity of a river are functions of the energy of the river.
As energy in a river reach is proportional to the product of the mass (i.e. size) and the

bedslope, 'main' is synonymous with size. Thus of two tributaries of the same river, the larger
is considered the main channel. 27

This formulation is the substantial equivalent of the criterion set forth in NM, Alexander
Report, para. 2.8: 'The main channel of a river is the channel that conveys the largest
proportion of the annual flow of the river.' (emphasis in original) In his Supplementary
Report, included in this Counter-Memorial as Vol. III, Professor Alexander repeats: 'The
basic premise is that the main channel of a river is the channel that carries the largest
28
proportion of the annual flow in the river.' (Supp. Rep., para. 2.1) (emphasis in original)

28. It will be noted that Dr. Sefe does not say that the main channel is the deepest channel.
The Botswana Memorial, however, states that 'the primary, and perhaps the only, criterion
29
would be relative depth . . .' This proposition is not supported by Dr. Sefe's scientific
analysis and is not consistent with the scientific criteria he advances. Dr. Sefe says that
'"main" is synonymous with size' and that 'of two tributaries of the same river, the larger is

considered the main channel.' He does not say, nor does it follow, that 'size' or 'largeness' of a
channel is to be measured by depth. On the contrary, in the context of the full passage fromDr. Sefe quoted above, it is clear that size is to be measured by 'the ability of the river to
transport debris' and 'the maximum [sediment] load it can carry.' In fact, as Professor
Alexander emphasises, width and depth are inappropriate measures of the main channel in an
ephemeral river like the Chobe River because these dimensions are irregular and change
rapidly from place to place along the river and from one period in time to another. (NM,
30
Alexander Report, para. 4.4)

29. The Alexander Report, analysed at length in Namibia's Memorial, establishes conclusively

that the southern channel carries by far 'the largest proportion of the annual flow of the
river.'1 Therefore, according to the scientific criterion advanced by both experts, the
southern channel is the main channel. Professor Alexander's analysis is expanded and further

elaborated in his Supplementary Report with particular reference to the contentions in the
Botswana Memorial. Namibia urges that this Supplementary Report, like the original
Alexander Report, should be studied in full. 32

30. Professor Alexander demonstrates that the flow in the lower reaches of the Chobe River is
not 'a result of the runoff from the catchment area of the system,' as Botswana asserts.3

Instead, the flow comes in two stages from the floods of the Zambezi River which occur
annually between the end of December and June. (Supp. Rep., paras. 5.6-5.9) At the
beginning of the period, the rapidly rising waters of the Zambezi back up behind the
Mambova Rapids, and as a result, the movement of the water is upstream (from east to west)
34
in the Chobe River. (Supp. Rep., paras. 5.7, 5.9(a)) In the second phase, when the Zambezi
overflows its banks along the reach between Katima Mulilo and the Mambova Rapids, the
waters advance across the floodplain, inundating almost the entire triangle bounded by
Katima Mulilo, Ngoma and the Mambova Rapids, until they reach the Chobe Ridge on the
south bank of the Chobe River. There, along the broad front from Ngoma to the Rapids, they

are turned into the channel of the Chobe River, where they flow in a south-west to north-east
direction passing through the southern channel around Kasikili Island and ultimately over the
Mambova Rapids to the confluence with the Zambezi River. (Supp. Rep., para. 5.8; NM,
Atlas, Map XIII) Substantially the entire flow of the river during this period thus goes through
the southern channel. (Supp. Rep., para. 5.15) This movement is determined by the
geomorphology of the floodplain, which slopes gently south-eastward until interrupted by the
Chobe Ridge (and further west, the Linyandi Ridge), which marks the south bank of the

Chobe River and forces the water to make a right angle turn to the north-east. (NM,
Alexander Report, para. 5.9) Although some of the Zambezi flood waters flow through two
anabranch channels in the floodplain into the northern channel east of Kasika, virtually no
waters flow longitudinally through the northern channel from the bifurcation of the river west
of Kasikili Island around the Island to the convergence of the two channels to the east of the
Island. (Supp. Rep., sec. 5)

31. The rest of the year, from the end of June through November, is the dry season and during
that period the river in the vicinity of Kasikili Island assumes the characteristic shape that
appears on maps and aerial photographs and that was observed by the inspection parties upon
35
whose conclusions the Botswana Memorial relies. But despite the appearance of a river that
flows from west to east along its channel, parting at Kasikili Island to pass to the north and
south of the Island, during this period there is virtually no flow at all in the Chobe River,
including the portions around the Island. (NM, Alexander Report, paras. 1.6, 10.15) For mostof the more than 300 kilometres over which it defines the boundary between Botswana and
Namibia, the river is dry during this period except for intermittent pools fed by groundwater.
(Supp. Rep., sec. 7) Below Serondela, in the last 20 kilometres before the Mambova Rapids,
the water appears on the maps and photographs to be a continuous stream, but it is actually
essentially stagnant, dammed behind the Rapids and receiving no replenishment from the

catchment area upstream. Thus, there is substantially no flow through the northern channel in
either season of the year.

32. Dr. Sefe concludes that 'the north channel is the main channel of the Chobe River in the
36
true hydrogeomorphic sense . . .' (emphasis in original) But Dr. Sefe does not relate this
conclusion to the 'hydrological and geomorphological connotations' he identified earlier, i.e.,
'the velocity of the flow,' 'the discharge,' 'the ability of the river to transport debris,' and 'the
maximum [sediment] load a stream can carry.' Dr. Sefe makes no comparison between the
two channels on the basis of these factors, which are not even discussed anywhere in his
report. Indeed, it is hard to determine on what basis Dr. Sefe reached his conclusion. Six of

the 12 pages of his report are devoted to showing that the position of the two channels has not
changed since 1890, a fact that Namibia does not contest. 37 The only apparent foundation for
Dr. Sefe's conclusion is his earlier statement that the results of a joint survey by South Africa

and Botswana in 1985 'clearly indicate the north channel as the main channel: its mean depth
of 5.7m exceeds the mean depth of the south channel by 2.13m.' 38 Dr. Sefe seems to rely as
well on the conclusion reached by the 1985 survey. The Alexander Report, sec. 11, analyses

the 1985 report and demonstrates that the 1985 survey team's methodology was unreliable and
its conclusion erroneous. And Dr. Sefe, on the following page, in the passage quoted above,
states that the main channel is to be determined not on the basis of depth but of 'the ability of
the river to transport debris' and 'the maximum [sediment] load . . . a stream can carry.' Since
his conclusion does not meet the criteria for the identification of the main channel that he
himself proposes, it should be disregarded by the Court.

33. The Botswana Memorial calls attention to three 'meander loops' in the river west of
Kasikili Island.9 It asserts that 'these meanders are uniformly made by the main stream of the

river,' that 'Kasikili/Sedudu Island is itself formed by such a meander loop,' and that 'like the
other loops the main channel flows to the west and north clockwise, round the island.'40 No
evidence is given and no references are cited for this series of propositions.

34. Dr. Sefe does not provide the missing evidence. He makes two or three passing references
to meander loops on pages 2 and 3 of his report. He lists 'meanders' among the characteristics
of 'low energy' rivers.1 Thereafter, he says, 'Kasikili Island was initially a sand bar deposit,

created by methods indicated above . . . '2 But since many methods are 'indicated above,'
there is no way of telling which he means. Later, after characterising the southern channel as

'the largest of the backwater channels,' he says, 'Many of the backwater channels are
associated with abandoned meander loops and lagoons resulting from spit deposition.' 43 This
does not appear to be a specific reference to Kasikili Island. Neither of these statements

supports the propositions in the Botswana Memorial quoted above.

35. Professor Alexander addresses this issue in paragraphs 9.3-9.5 of his Supplementary
Report. There he shows that the position of the meander loops Dr. Sefe refers to is controlledby a sub-parallel fault system and not by the processes described in Dr. Sefe's report. This is
apparent from inspection of Map 2 on Sheet 29 of the Alexander Report, which shows that the
leading arms of the meanders are parallel to the geological fault line at the Mambova Rapids.
It is also apparent from Photographs P11s and P12s (Supp. Rep., Sheet 10s) that the features
of the meander loops, including the nature of the soil and the topography and elevation of the

land within the loops, are fundamentally different from Kasikili Island, as shown in
Photographs P5s-P8s. (Supp. Rep., Sheets 4s-7s) Thus, Professor Alexander says that the
conclusion that the northern channel is the main channel based on its superficial similarity
with the meander loops is incorrect.

36. The 'question of scientific fact' as to the main channel of the Chobe River around Kasikili
Island must be resolved in favour of the southern channel. That channel carries substantially

all of the annual flow of the river, and the northern channel carries almost none of it. Since the
scientific criterion that defines the main channel is 'the velocity of flow, hence the discharge'
(Dr. Sefe) or 'the channel that conveys the largest proportion of the annual flow of the river'
(Professor Alexander), it follows that on the basis of the scientific evidence the main channel
is the southern channel.

B. In Terms of Navigability 'in relation to the needs of the regional economy,' the Southern

Channel (and not the Northern) Qualifies as the Main Channel

37. In paragraph 116 of its Memorial, Botswana states that 'the principal criterion on which
[the] . . . assessment [of the main channel] is based . . . [is] that of navigability.' This is
completely at odds, of course, with the position stated earlier that 'the central question'
involves 'a question of scientific fact, calling for expertise in hydrology, geology and
44
hydrogeomorphology.'

38. In Namibia's view, it would be anomalous to apply a criterion of navigability to a river

boundary that is non-navigable for most of its length. Namibia has already spoken of
Botswana's myopic view of the subject matter of this case. 45 For example, the Botswana
Memorial confines its attention to 'the Chobe River in the last 50 kilometres (30 miles) prior
46 47
to its confluence with the Zambezi River.' Dr. Sefe's report is similarly limited. But under
Article III of the 1890 Treaty, the Chobe River constitutes the boundary between Botswana
and Namibia from its intersection with the 18th parallel of south latitude for over 300

kilometres eastward until it joins the Zambezi. For most of these 300 kilometres, the Chobe is
clearly not a navigable river, and in fact it is dry for much of the time over much of its length.
Above Serondela, a portion of the river boundary from which the Botswana Memorial
resolutely averts its gaze, the concept of navigability has no application whatsoever. By
contrast, the scientific criterion, 'the channel that conveys the largest proportion of the annual
flow,' can be applied consistently over the entire length of the river because every reach of the
river has had some flow during at least some parts of some years. However, even under the

criterion of navigability rightly understood, the southern channel is the main channel.

39. Botswana argues that the negotiators of the 1890 Treaty were concerned with access to the
48
navigable rivers in the vast areas of Africa they were dividing. Be that as it may, one of the
best known geographical facts at the time about the region of concern in this case was that the
upper Zambezi was not navigable because of the presence of the Victoria Falls 80 kilometres
south of the confluence with the Chobe River. The Caprivi Strip was created to give Germanyaccess to the Zambezi River. But it does not follow that navigability rather than land access
was what was intended. The Treaty itself states that 'Germany shall have free access from her
49
Protectorate to the Zambesi by a strip of territory . . (emphasis added) Writing to the
Foreign Office in 1910 in connection with the discussions of the western sector of the
boundary, C.P. Lucas of the British Colonial Office stressed that it was the 'access by land to
50
the Zambesi . . .' to which the Germans attached importance. A few years earlier, High
Commissioner William Selborne, writing to the Secretary of State also implied that land
access to the Zambezi was what had been of primary significance to Germany:

I have always understood that the reason why His Imperial Majesty the German Emperor laid
such stress on the possession of the strip, was his desire not to be excluded from the
navigation of the Zambesi, about which, at that time, very little was known.1

40. As for the Chobe River, the Treaty Map 52 shows it proceeding through swamps for much

of the boundary area. As will be recalled, Lord Salisbury, in presenting the Treaty to
Parliament, remarked that 'during the last 300 years there has been no very eager or impetuous
torrent of trade' through the Strip. It is, he added, 'the last route in the world by which trade
54
can pass.' As Namibia has shown in its Memorial, the Chobe River was suggested as the
boundary by the British negotiators, for whom it represented a convenient geographical

marker of the northern limit of their territorial ambitions for Bechuanaland, which at the same
time satisfied the German desire for a 'strip of territory' providing access to the Zambezi.

41. No evidence has been produced of passage of commercial vessels through the northern
channel at any time on a significant scale. It appears that during the 1930s a Mr. Suzman had
a timber concession on the Chobe River front, presumably around Serondela. It is not known
how fully the concession was exploited and, if it was, by what route the timber was
transported to market. The activity apparently lapsed before 1945, however, because it was
this concession from which, a few years after World War II, William C. Ker wanted to
56
transport timber - leading to the much discussed Trollope-Dickinson arrangement. He
sought and was granted permission to use the northern channel in connection with this effort,

but it is not clear that he ex57cised it. Nor is it clear to what extent his primary motivation was
economic rather than political. The lumber mill ceased operations in the mid-1950s. This
venture remains, so far as Namibia has been able to discover, the only recorded instance of
58
the use of the northern channel for 'ordinary commercial navigation.' Thus, whatever the
negotiators may have hoped, speculated or dreamed, the Chobe River never was and never
59
will be a 'highway[] for commerce' for the transport of goods.

42. Even if navigability is relevant, however, the issue is not to be determined primarily on
60
the basis of depth, as maintained by Botswana. As early as 1861, Henry W. Halleck
recognised that

the deeper channel may be less suited, or totally unfit, for the purposes of navigation, in which
case, the dividing line would be in the middle of the [channel] which is best suited and
ordinarily used for that object. (emphasis added)43. Similarly, the Barcelona Statute on the Régime of Navigable Waterways of International
Concern (1921) applies a less mechanical and more practical standard than average depth. A
river is navigable

if now used for ordinary commercial navigation, . . . by "ordinary commercial navigation" is
to be understood navigation which, in view of the economic condition of the riparian
countries, is commercially and normally practicable. 62

This coincides with Botswana's own statement that navigability is to be judged 'in relation to
the needs of the regional economy.' 63

44. Applying this standard, Botswana asserts that 'the northern and western channel of the
64
Chobe River is the sole channel navigable in relation to the needs of the regional economy.'
This statement is flatly wrong. It is another sweeping proposition, central to Botswana's
argument, that is not supported by any references or evidence.

45. The primary basis of the regional economy, apart from subsistence agriculture, is and is
likely to remain tourism, and it is tourism that generates the commercial activity on the Chobe

River in the vicinity of Kasikili Island. (Supp. Rep., sec. 11) Nearly all boat traffic is used to
transport tourists to view the game on Kasikili Island or further west on the south bank of the
river within the Chobe National Park. (Supp. Rep., para. 11.5) The boat rides are a well-
advertised attraction for visitors to the area. (Supp. Rep., Sheet 17s, Photo P25s; Sheet 18s,
Photo P26s) The tourist boats use the southern channel almost exclusively, although a few
boats returning from Kasane use the northern channel. (Supp. Rep., para. 11.12) On Kasikili

Island itself, the game can be more easily seen from the boats in the southern channel where
the banks are very low, in contrast to those of the northern channel. (Supp. Rep., paras. 11.9,
11.11) The boats are shallow draught, so that they can be manoeuvred hard up against the
banks and into the many shallow creeks and side channels along the river. (Supp. Rep., paras.
11.5, 11.10; Sheet 14s, Photo P19s; Sheet 18s, Photo P26s) There is at present little or no
game on the northern side of the river, which is mainly agricultural country.

46. In comparing the navigability of the northern and southern channels, Botswana relies
primarily on the greater depth 'on average' of the northern channel.65 But average depth is not

determinative of navigability. Passage through a channel is controlled by the point of
minimum depth, because all craft must clear that point to traverse the channel. On the
question of minimum depth, the Botswana Memorial is a good deal less clear. It says that the
66
entry point to the southern channel 'is very shallow, 1.5 to 2.m. in depth.' And while
acknowledging obliquely that there is a 'small sand bar to the north of the National Park
67
Headquarters' that must be cleared before entering the northern channel, Botswana gives no
indication of the depth of the river at that point. Professor Alexander says that in this respect
68
the difference between the two channels is marginal. (NM, Alexander Report, para. 11.6)
At any rate, not only the entry, but the entire length of the southern channel is of ample depth
to accommodate the flat bottom boats carrying the tourist traffic at all times of the year.
(Supp. Rep., paras. 10.7, 11.10; Sheet 14s, Photo 19s)

47. Botswana's statements that the southern channel is not navigable for these purposes are

flatly wrong. • Contrary to Botswana's repeated statements, the southern channel is not blocked or
69
impaired by 'a well-established and persistent growth of Kariba Weed.' The Kariba
Weed was removed from the southern channel as well as the rest of this reach of the
Chobe River by a joint Botswana/South Africa project in the early 1970s, motivated in
70
considerable part by the need to keep the river open for tourist traffic.There are
now no Kariba Weed or any other aquatic plants obstructing the southern channel.
(Supp. Rep., paras. 10.5, 10.10-10.12, 10.16) Moreover, South Africa's participation in
this joint project with Botswana to clear the Kariba Weed from the Chobe River in
order to improve navigation, which even at that time was primarily in the southern

channel, is telling evidence that both parties considered that the international boundary
lay in the southern channel.
• Contrary to Botswana's claim, the southern channel does not dry out during drought
71
years and periods of low flow. No evidence is provided in support of Botswana's
statement. Professor Alexander has analysed the recorded water levels in the Chobe
River. His calculations show that during the entire period for which data are available,
1964 to date, neither the northern nor the southern channels were dry during the lowest
72
flow periods. (Supp. Rep., para. 10.7)
• It is not true, as Botswana states, that 'when the Chobe River is at its lowest in the
73
month of August only the northern and western channel is navigable.' No reference
is cited for this statement either. The peak tourist season runs from April to October,
so August is a major tourist month for the Chobe National Park and for the lodges in
the vicinity of Kasikili Island. Boat trips along the southern channel to view the game
on the Island operate on a frequent basis during this period. (Supp. Rep., para. 10.8) In

preparation for this case, the legal representatives of Namibia visited the area in 22
August 1996 and travelled by boat from one end of the southern channel to the
other.4 When the JTTE was there on 4 October 1993 the southern channel was full of

water.

48. If, as Botswana suggests, a major reason behind the rules about navigability is to ensure
75
that both riparian states have access to the economic benefits of the river then clearly the
southern channel is the navigable channel. Since 1991, when Botswana armed forces
unilaterally occupied the Island, boats from the resort lodges in Namibia have been prevented
from using the southern channel, which, as shown above, is the optimal channel for tourism.
The prohibition has been enforced by armed Botswana boats patrolling in both the northern

and the southern channel. As a result the economic prospects of the Namibian lodges have
been limited, and the growth of the tourist industry along the northern bank of the river has
been seriously inhibited. On the other hand, neither Namibia nor its predecessors has
prevented Botswana from using the northern channel. It is Botswana's position that deprives
Namibia of access to the benefits of the Chobe 'in relation to the needs of the regional
economy,' not the reverse.

49. It follows that, if the criterion of navigability is relevant to the determination of the main
channel, on a proper understanding of the criterion, the southern channel is the main channel.

C. Nothing in the Language of the 1890 Treaty or the General Principles of International Law
Warrants the Rejection of the Determination that the Main Channel of the Chobe River is the

Southern Channel50. Despite its protestations, the Botswana Memorial gives little more than perfunctory
treatment to navigability or to scientific evidence and criteria (except for the point that the
general configuration of the Chobe River in the vicinity of Kasikili Island has not changed
since the signing of the Treaty, a point that Namibia does not contest). The principal thrust of
Botswana's argument is that the main channel is to be identified not on the basis of these

considerations, but by the application of general principles of international law. Indeed,
Botswana devotes six pages of its Memorial to belabouring the elementary principle - with
which Namibia also agrees and of which the Court is surely perfectly well aware - that the
Treaty is to be interpreted in good faith in accordance with the ordinary meaning of its terms
76
in the light of its object and purpose. In this Section, Namibia shows that nothing in the
language of the Treaty or in the rules of general international law concerning river boundaries
dictates that the northern channel is the main channel.

1. The language of the Anglo-German Treaty of 1890

51. The words of the Treaty that are to be interpreted are 'the centre of the main channel of
[the Chobe] river.'77 In the German translation, the phrase reads 'im Thalweg des Hauptlaufes
dieses Flusses.'

52. The Botswana Memorial states 'The German text employs the term Thalweg as the
counterpart to "the main channel" in the English text.'78 This is plainly wrong. A simple

inspection of the two phrases shows that the counterpart to the English 'main channel' is the
German word 'Hauptlaufes.' The word Thalweg is the counterpart of 'the centre' in the English
text. As fully explained in the Namibian Memorial, it is evident from the circumstances in
which the Treaty was negotiated that the German text was simply a translation of the English-
79
language text. The use of the word Thalweg in the German text is therefore not equivalent
to the words 'main channel' in the English, as Botswana maintains. The term Thalweg in the
German text designates the position of the boundary within des Hauptlaufes, just as, in the
English text, the words 'the centre' designate the position of the boundary within the 'main
channel.' The word Thalweg in the German text of the Treaty cannot be used to distinguish

the main channel from another channel, any more than can the word 'centre' in the English
text.

53. Because English was the original language of the Treaty and the German a mere
translation, the interpretation of Article III depends on the English meaning of the words
'centre of the main channel.' There is no basis for the conclusion that these words, in their
ordinary English meaning, refer to the centre of the deepest channel or that the reference is
80
linked to navigability of the river as Botswana contends. Indeed, the word 'centre,' as
opposed to 'thalweg,' would seem to be determined in relation to the width of the channel,
rather than its depth. Certainly, the text of the Treaty does not contain or suggest anything that

would override the identification of the southern channel as the main channel on the basis of
scientific evidence or other applicable criteria.

2. General principles of international law concerning river boundaries contemporaneous with
the 1890 Treaty did not embody the thalweg doctrine54. The Botswana Memorial cites 17 authorities (13, if successive editions of books by a
81
single author are disregarded) for the proposition that at the time of the conclusion of the
1890 Treaty 'it was generally recognised that in the case of navigable rivers the middle of the
navigable channel (thalweg) was the boundary between the riparian States' (without, however,
82
supplying quotations from the cited authors). Although it is true that many of those authors
use some such language, the support is neither so uniform nor unqualified as Botswana
implies in its string citation. It is elementary that to establish a rule of general international
law it must be shown that the putative rule embodied a widespread, uniform and consistent
practice accompanied by opinio juris, a sense that action in conformity with the practice is a

matter of legal obligation. Namibia has examined 47 authorities who discussed the thalweg
concept between 1820 and 1930 (the most relevant period for present purposes). Quotations
from these authorities are collected in Annex 9. Namibia believes that the list comprises
substantially all of the writers of any significance who have addressed the subject in any detail
in that period, including all of the publicists cited by Botswana. The quotations advocate a
wide range of positions. Some writers argue that the thalweg should be the boundary in

navig83le rivers but not in rivers that are not navigable, where the median line should be
used. Others endorse the concept for non-navigable as well as navigable rivers, in most
84
cases without indicating how it would be applied to a river that was not navigable. Still
85
others oppose the asserted thalweg rule altogether. For many of the authors, classification is
difficult, either because they do not take a clear position or because careful analysis and
86 87
discrimination is lacking. Some even conflate the terms 'thalweg' and 'median line.' This
hardly adds up to a consensus among scholarly writers of the uniformity and solidity
necessary to establish a rule of general international law.

55. Botswana relies heavily on the resolution of the Heidelberg session of the Institut de Droit

International in 1887, which provided that 'La frontière des États sép88és par le fleuve est
marquée par le thalweg, c'est-à-dire par la ligne médiane du chenal.' F. de Martens, who
was the Rapporteur for the Institut on this occasion, did not include this provision in his

project for the Institut because, as he said in debate, 'ce principe n'est pas généralement
reconnu . . .'9 The language that was adopted as the second paragraph of Article 3 in the
90
resolution of the Institut came from Edouard Englehardt's project and was introduced on the
floor of the meeting. De Martens would appear to have accepted its inclusion de lege ferenda.
The resolution therefore provides no support for the assertion that the riverine boundary was

the middle of the navigable channel or the thalweg. On the contrary, the meeting of the
Institut exemplifies the uncertain state of general international law at about the time of the
Treaty.91

56. The treaty practice of the European colonial powers in Africa in the period of the 1890
Treaty reveals, if anything, even less uniformity than the general practice or the scholarly
writings. One of the most striking features of these treaties is the variety of language used to
describe river boundaries as well as an evident lack of concern with whether the rivers were
navigable or non-navigable. This diversity of usage is apparent within the four corners of the

1890 Treaty itself:

• Article I(2):'a line which . . . follows the course of the River Rovuma . . .'

'it ascends that [the Songwe] river . . .'

'thence it follows the river . . .'

'and thence follows that [the Kilambo] river till it enters Lake Tanganyika.'

• Article III(2) (the provision involved in the present case):

'descends the centre of the main channel of that [the Chobe] river . . .'

• Article IV(1):

'The boundary . . . ascends the mid-channel of that [the Aka] river . . .'

Thus, in the course of this single treaty, at least four different expressions are used to indicate
92
an international boundary in a river.

57. A similar diversity, often in a single instrument, is characteristic of other treaties
concluded by European powers in Africa in the period of the 1890 Treaty. A sample of the

treaties of the period cited in Professor Ian Brownlie's encyclopaedic work on African
boundaries, 93 shows the following locutions (page numbers refer to that volume, unless
otherwise indicated; emphasis added):

• 'up the course of the Limpopo River'; Great Britain/Transvaal, 1884, Art. I (p. 1093).
• 'shall follow the course of the River Kunene'; 'along the course of the [Kubango]
river'; Portugal/Germany, 1886, Art. I (p. 1028).
• 'the middle line of the Kunene River, that is to say, the line drawn equidistant from

both banks'; Portugal/South Africa, 1926, para. 2 (p. 1029).
• 'the centre of the channel of the [River] Ruo'; 'the centre of the channel of the [Shiré]
river'; 'follows the mid-channel of that [Aroangwa] river'; United Kingdom/Portugal,
1891, Art. I(2) (p. 1119).

Art. II (p. 1119): 'the centre of the main channel of the Sabi';

Art. IV (p. 1120): 'the centre of the channel of the Upper Zambesi . . .'

• 'shall follow the centre of the River Ruo up-stream'; 'shall follow the Malosa River up-
stream'; United Kingdom/Portugal, 1899, Art. 1 (pp. 1128-1129).

• 'the line of the thalweg of those [the Ruo and Shiré] rivers'; United Kingdom/Portugal,
1911 (p. 1179).
• 'the centre of the main channel of the River Loangwa'; United Kingdom/Portugal,
1911, Annex 1 (p. 1265).

Annex 2 (pp. 1268-1269): 'the centre of the channel of the Inyarumanu stream to its junction
with the centre of the main channel of the River Angwa; and thence the thalweg of the River
Angwa'; 'the centre of the bed of the Karemwe River; thence follows the thalweg of theKaremwe River . . . the thalweg of the Kazi River . . . the thalweg of the Msengezi River';
'follows the thalweg of the Mkumvaru River . . . '

• 'the centre of the channel of the River Gaeresi'; United Kingdom/Portugal, 1912 (p.
1226).
• 'follows the east channel of the Maputo'; Mozambique/South Africa, 1897 (p. 1243).
• 'shall follow the thalweg of this [Black Volta] river'; 'shall then follow the thalweg of
the western branch of this river'; 'shall then follow the thalweg of this [Nuhau] river';
United Kingdom/France, 1898, Art. I (pp. 619-620);

Art. III (p. 620): 'shall then follow the median line of the [Niger] river . . .'

94
58. There seems to be no pattern or principle governing this proliferation of terminology.
The use of 'median' or 'centre' does not correlate with non-navigable rivers, as some of the
95
authors cited by Botswana suggest. The Niger River, for instance, was clearly navigable,
but the relevant treaty stipulates 'the median line.' Some provisions, like Article I of the Niger
treaty, explicitly refer to the thalweg. But others, like Article III of the same treaty, refer just
as clearly to the median line. No inference can be drawn from this assortment of ambiguous
terms to the effect that they all refer to 'the navigable channel (thalweg),' and no presumption
96
can be derived that in case of doubt that is what was intended. The conclusion from the
African practice is, if anything, stronger than that derived from the scholarly literature: there

was no uniformity, and hence there was no rule of general international law. The position is
perhaps best characterised by Ernest Nys, who commented in 1901:

Tout ce qu'il est permis de dire c'est que dans les dernières années du XIX siècle, le thalweg a
été fréquemment adopté pour la délimitation et il faut même ajouter que les gouvernements se
97
sont continuellement servis d'une expression vague et, partant, mauvaise.

3. International law sources of the relevant period demonstrate that the thalweg was to be

identified by reference to the current or flow of the channel and not merely its depth

59. The previous section has established that there was no rule of international law according
to which 'the middle of the navigable channel (thalweg) was the boundary between theriparian
98
States.' In this section, Namibia shows that even if the notion of the thalweg is regarded as
relevant to the identification of the 'main channel,' Botswana is incorrect in saying that the
thalweg is to be defined as 'the line at the water's surface vertically above the deepest channel
99
of the river bed at low tide.' On the contrary, as will appear more fully below, the thalweg
concept has to do fundamentally with the flow or current of the river and is only secondarily
related to depth.

60. Doubtless most of the discussions of the thalweg of a river channel refer to its relation to
100
the deepest part of the channel. A careful examination of the leading authorities, however,
especially in the period leading up to the 1890 Treaty, demonstrates that the concept of
thalweg is essentially related to the current in the channel - indeed, that the flowor current of
the river is at the core of the thalweg concept. An important consequence is that the legal
criteria converge with the scientific factors that both parties agree are relevant in determining

the main channel of a river: 'the velocity of flow,' 'the discharge,' 'the ability of the river totransport debris,' 'the maximum load (sediment of particular grain size) a stream can carry' or
101
the proportion of the annual flow that it carries. As a scientific matter, it is the scouring
effect of the current that accounts for the depth of the channel. The correct understanding of
the thalweg in law - as involving the current or flow of the stream - thus permits a
determination of the main channel that satisfies at the same time both the scientific and the
legal criteria.

a. Authoritative contemporaneous definitions

61. Since the word thalweg is originally German, it is appropriate to refer to German
encyclopaedias and dictionaries of the period.

• The connection between depth and current is admirably portrayed in the Deutsches
Wörterbuch: 'the channel wherein the water flows, the deepest line of the valley floor
102
is called the thalweg.' (emphasis added) Again the thalweg of the Rhine is defined
as 'the strongest and deepest stream-linein the course of the Rhine.'103 (emphasis

added)
• Brockhaus Konversationslexikon contains the entry: '"stream-line, thalweg," the line
that connects the points of greatest surface velocity of flowing water. It usually moves
104
above the deepest channel.' (emphasis added)
• A later text, Der Grosse Herder, Nachschlagewerk für Wissen und Leben, contains
this entry: 'thalweg: the channel continuously kept open for navigation; also the centre
of the navigation channel, the line of the fastest surface current, in case of lack of
105
agreement the border-line between countries.' (emphasis added)

62. Contemporaneous French dictionaries are also in accord.

• Littré's Dictionnaire de la langue française offers the clearest definition of the
thalweg: '[La] [l]igne plus ou moins sinueuse au fond d'une vallée, suivant laquelle se
dirigent les eaux courantes. / Thalweg d'une vallée, la ligne d'intersection des plans de
pente latérale des deux berges de la vallée. / Thalweg d'un cours d'eau, la position du

filet d'eau qui se meut avec la plus grande vitesse Le thalweg d'un cours d'eau ne
correspond pas toujours au thalweg de la vallée. . . . Le thalweg du Rhin formera la
démarcation entre la France et les Etats de l'Allemagne, Traité de Paris de 1815, art.
106
2.' (emphasis added)
• The Larousse Dictionnaire universel du XIXº siècle is not as long, but just as clear in
its definition of thalweg: 'Geogr. Ligne que décrit le fond d'une vallée et suivant
107
laquelle se dirigent les eaux courantes.'

63. The point is made explicitly by many of the publicists cited by Botswana:

• C. Calvo: 'Le Thalweg n'est pas, à proprement dire, rigoreusement, le milieu exact,
absolu d'un cours d'eau; mais plutôt le milieu du courant du plus gros volume d'eau; la
position du filet d'eau qui se meut avec le plus de rapidité . . .'(emphasis added)

• G.F. de Martens: 'c'est le courant du fleuve qu'on a communément en vue, en
convenant de prendre le milieu pour limite.'109 (emphasis added) • A. Rivier: 'le thalweg d'un cours d'eau: la position du filet d'eau qui se meut avec la
110
plus grande vitesse.' And again, 'D'après cet usage, la limite est au milieu, non du
111
lit, mais du courant ou fil de l'eauqu'on appelle aujourd'hui le thalweg . . .'
(emphasis added)
112
• F. Despagnet: 'chenal formé par le courant de l'eau . . .' (emphasis added)
• J.L. Klüber speaks of: 'le chemin (variable) que prennent les bateliers quand ils
113
descendent le fleuve, ou plutôt le milieu de ce chemin,' and Bluntschli refers to the
thalweg in the same sense. 114 As will be seen in para. 70, infra, these references to the

thalweg in relation to downstream navigation are closely connected to the idea of the
current as the central element of the thalweg concept.

64. Edouard Engelhardt, the chief sponsor of Article 3 of the Heidelberg Resolution, speaks of
the thalweg formula, 'suivant laquelle la limite respective est placée au milieu du chenal ou du
115
grand courant qui dénote d'ordinaire l'endroit le plus profond.' (emphasis added) In a note
he adds, 'Le thalweg est la partie la plus basse du lit sur laquelle le courant se meut avec la
116
plus grande vitesse.' (emphasis added)

117
65. Other scholars of the period not cited by Botswana are in agreement:

• J. Westlake: 'the thalweg, a German word meaning literally the "downway," is the
course taken by boats going down stream, which again is that of the strongest current .
118
. .' (emphasis added)
• P. Fiore: '[L]a ligne médiane du fleuve [est la] . . . ligne de thalweg . . . Il ne faut pas
oublier, toutefois, que sous le nom de ligne médiane du fleuve on n'entend pas celle
qui se trouve à égale distance des deux rives, mais celle idéalement tracée au milieu de
119
la partie du lit où les eaux sont les plus profondes et les plus rapides.'(emphasis
added)
• L.F. von Neumann: 'the line that is taken by ships going downstream, more precisely
120
the center of the downward current.' (emphasis added)
• P. Orban: 'la partie la plus basse du lit sur laquelle le courant se meut avec la plus
121
grande vitesse.' (emphasis added)
• H. Bonfils: 'Un fleuve coule-t-il entre deux Etats, c'est le thalweg, le milieux du
122
courant principal qui sert de limite.' (emphasis added)
• A. Chrétien: 'On n'entend pas par là [la ligne dit le thalweg] la ligne se trouvant à égale

distance des deux rives, mais "celle idéalement tracée au milieu de la partie" du lit où
les eaux sont les plus profondes et les plus "rapides", en d'autres termes, au milieu du
chenal navigable.'123 (emphasis added)

66. A compelling example of the state of British thinking at the time is found in Resident
Commissioner Panzera's instructions to Captain Eason for his expedition to find the main
channel of the Chobe. '[T]he question under consideration [the location of the main channel]

could only be solved,' Panzera said, 'by following up the deepest channel in which there is the
strongest current?'124 (emphasis in original) Captain Eason, it will be recalled, reported that'[t]he South channel is merely a back water, what current there is goes round the North.' 125

As Professor Alexander has shown, in July, the time of the year when Captain Eason made his
observation, there is no significant flow in either channel. (NM, Alexander Report, para.
10.16) Eason's comment is uncertain on this very point. He refers to 'what current there is.' In
the light of his instructions, if he had had a full appreciation of the hydrology and
geomorphology of the river, he might have been less willing to rely on his already hesitant
observation.

b. Origins and evolution of the thalweg concept and its relation to navigation

67. From time immemorial, when people travelled on rivers - whether by dugout or birchbark
canoe, sail, paddle wheelers or modern steamships - they followed the current when going
downstream. The reason is simple efficiency. Whatever the mode of transport, it saved time
and energy to let the current do part of the work. The original idea of the thalweg as a
boundary marker arose because division at the mathematical middle of the river did not assure
126
both riparian states access to the channel - the downstream current. These origins of the
thalweg concept are recognised by the principal turn-of-the-century commentators. Westlake
states that the thalweg is 'a German word meaning literally the "downway," that is the course

taken by boats going down stream, which again is that of the strongest current, the slack
current being left for the convenience of ascending boats.'127 (emphasis added) Charles

Cheney Hyde follows Westlake closely:

The thalweg, as the derivation of the term indicates, is the downway, or the course followed
by vessels of largest tonnage in descending the river. That course frequently, if not
commonly, corresponds with the deepest channel. It may, however, for special reasons take a
different path. Wheresoever that may be, such a course necessarily indicates the principal
128
artery of commerce, and for that reason is decisive of the thalweg.

68. Englehardt makes this connection explicitly and very practically. After noting that '[l]e
thalweg est la partie la plus basse du lit sur laquelle le courant se meut avec la plus grande

vitesse,' he explains further:

. . . en pratique on ne détermine point la direction du chenal avec une précision mathématique
. . . L'on se contente d'ordinaire d'observer la course des bateaux de plus fort tonnage, et on
129
l'indique au moyen de signaux fixes ou de bouées. (emphasis in original)

69. In the present case, although no large freighters ply the Chobe, all the tourist boats travel
primarily in the southern channel, as shown in para. 45, supra. In Hyde's words, 'such a course
necessarily indicates the principal artery of commerce, and for that reason is decisive of the

thalweg.'

70. Numerous authorities in the pre-World War I period repeated the portions of these
definitions emphasising the link between downstream navigation and the current of the river:

J.L. Klüber: 'le thalweg c'est-à-dire le chemin (variable) que prennent les bateliers quand ils
130
descendent le fleuve, ou plutôt le milieu de ce chemin.'H. Bonfils: 'Quelques auteurs fixent la limite de la propriété respective des Etats riverains au
131
thalweg, le milieu du courant, du chenal suivi par les bateaux descendant le fleuve.'

C.G. de Koch and F. Sch_ll: '[L]e thalweg [est] . . . le milieu du co[u]rant que suivent les
132
bâtimen[t]s en descendant le Rhin . . .'

G. Kaeckenbeeck: 'An authoritative definition [of thalweg] was given in the "Convention de
Strasbourg," 1827. Thalweg is the proper course taken by boats going down stream at low
133
water; . . .'

In the light of the showing that the thalweg concept was based on the use of the current by

downstream shipping, these authorities should be taken 134incorporate the element of the
current in the channel into the concept of the thalweg.

135
71. By the same token, the frequent statements that the thalweg is the 'navigable channel'
cannot be read in isolation from the history and evolution of the thalweg doctrine. They
should be seen as based on the assumption that this is the channel in which the downstream

current flows. Similarly, when some authorities state that where there are two navigable
channels, the deepest is to be considered the thalweg,136 they also must be assuming a choice

between two channels in which there is a downstream current. If, as in the case of the Chobe
River around Kasikili Island, that assumption does not hold and one of the two is essentially
stagnant, it would contradict the basic purpose of the thalweg concept - access to the
downstream current - to identify the stagnant channel as the thalweg.

72. Finally, a few writers take a more mechanistic approach and identify the thalweg as the

line of the deepest soundings in the channel. The most insistent of these was Ernest Nys, who
criticised the concept of the thalweg as '. . . une expression vague et, partant, mauvaise.'
The thrust of his criticism was that the downstream current was both too broad and too variant

over time to serve the purpose of accurately demarcating a boundary. Nys's solution was to
substitute 'axe du thalweg, qui désignent la ligne déterminée par la suite non interrompue des
sondes les plus profondes.'138 (emphasis in original) There is no evidence that the problems

Nys was addressing caused any difficulties in practice or disturbed anyone but academics. At
the same time, it is clear that Nys was using 'the line of the deepest soundings' as a surrogate
for the current and not as a contradictory principle.

73. If the thalweg doctrine is to have any bearing on this case, which Namibia denies, there is

one final consideration that should be determinative by itself. It is universally agreed that the
purpose of the doctrine is to ensure both riparian states equal access to the navigational
facilities and resources of the boundary river.39 The emphasis is on practical considerations

and commercial realities, not mechanical measurements. Halleck, for example, says:

As a general rule, this line runs through the middle of the deepest channel, although it may
divide the river and its estuaries into two very unequal parts. But the deeper channel may be
less suited, or totally unfit, for the purposes of navigation, in which case, the dividing line
140
would be in the middle of the one which is best suited and ordinarily used for that object.And as noted above, Hyde insisted that:

That course, [followed by vessels of largest tonnage in descending the river] frequently, if not
commonly, corresponds with the deepest channel. It may, however, for special reasons take a
different path. Wheresoever that may be, such a course necessarily indicates the principal
141
artery of commerce, and for that reason is decisive of the thalweg.

142
Justice Benjamin Cardozo in New Jersey v. Delaware, a case much cited in international
law, captured the essence of the position. 'The underlying rationale of the doctrine of the
Thalweg,' he said, 'is one of equality and justice. . . . Unless prescription or convention has
143
intrenched another rule, we are to utilize the formula that will make equality prevail.'

74. Today, commercial traffic - tourist boats - from the Namibian side are being denied access
to the navigational resources of the southern channel by Botswana's military forces. It would

be a perverse and arbitrary application of the thalweg concept - the object and purpose of
which is to secure equal navigation rights to both riparians - to use it to perpetuate that
situation because the northern channel is mathematically deeper on the average than the
southern channel. The 'rationale of equality and justice' will not prevail unless both parties
have access to the southern channel where the actual commercial traffic takes place.
'Wheresoever that may be,' as Hyde says, 'such a course necessarily indicates the principal

artery of commerce, and for that reason is decisive of the thalweg.'

75. As Professor Alexander has shown, the overwhelming bulk of the water of the Chobe
River flows through the southern channel, though this flow is confined to the January-June
period. No current flows all the way through the northern channel at any period of the year.
144
(Supp. Rep., sec. 5) In the flood season, the waters of the Zambezi flow across the Island
and into the southern channel where they are conveyed downstream. During the dry season,
there is no flow in either channel. None of the authorities discussed above have addressed the
problem of a choice between two channels, one of greater average depth through which
substantially none of the flow of the stream travels and a shallower one carrying substantially

all of the stream flow. They assumed as a matter of course that the subject of analysis was a
'channel' in which water was flowing. 145 Engelhardt's formula, 'au milieu du chenal ou du
grand courant qui dénote d'ordinaire l'endroit le plus profond,' (emphasis added) suggests that
146
in a case of doubt the criterion of the current would prevail over the factor of depth.

76. It is apparent from a study of the history and the authorities that the idea of the 'flow' or

'current' of the river is an essential component of the concept of thalweg in the legal sense,
just as it is of the scientific definitions of the main channel. Equally, the equitable principles
underlying the thalweg doctrine require that both riparian states have access to the practical
navigation resources of the river. There is no basis for the conclusion that general principles
of international law at the turn of the century would have regarded a body of water in which
there was almost no current and almost no commercial traffic as the thalweg of a river, no

matter how deep it was.

D. Conclusions as to the Identification of the Main Channel within the Meaning of the 1890
Treaty77. Namibia continues to maintain that the identification of the main channel within the
meaning of the 1890 Treaty is a question of scientific fact to be resolved on the basis of
scientific evidence and criteria. The main distinguishing factor, as the experts of both parties
agree, is the volume of the flow of the river that passes through the channel. Since
substantially all of the flow of the Chobe River passes through the southern channel, it

follows that the southern channel is the main channel.

78. Although Botswana accepts the central importance of scientific analysis, it adduces no
evidence relating to the very scientific factors that its own expert says are decisive: hydrology,
geology and hydrogeomorphology. Rather, it rests its argument on two other entirely different
bases: first, the comparative navigability of the two channels; and second, a supposed
contemporaneous rule of general international law that for navigable rivers the deepest

channel or thalweg marks the boundary between riparian states.

79. The discussion in this Chapter has shown, and indeed Botswana admits, that navigability
must be judged 'in relation to the needs of the regional economy .' In the case of the Chobe
River in the vicinity of Kasikili Island, the regional economy is based on tourism.
Substantially all of the tourist traffic on the river uses the southern channel; only a minor and
secondary portion goes around the north. Nor is there any other significant commercial traffic

through the northern channel. Thus, on the basis of Botswana's own criterion of navigability
in relation to the needs of the regional economy, the southern channel must be chosen as the
main channel.

80. As to the asserted rule of international law, it is, in the first place, inappropriate to apply
the rules on navigable rivers to a boundary river that is admittedly not navigable for nine-

tenths of its length. But in any case, neither the scholarly writings nor the treaty practice at or
around the time of the Treaty was consistent or definite enough to give rise to a rule of
general international law that the thalweg is the boundary. Even if there were such a rule, the
concept of the thalweg includes the flow or current of the river, not only (and not even
necessarily) the depth. The northern channel has no significant current. Almost none of the
water of the Chobe River flows longitudinally along the course of and through the northern
channel to the Zambezi River. Thus, the thalweg of the river within the meaning of the

asserted rule cannot lie within the northern channel.

81. All three criteria proffered by Botswana - science, navigation and the rules of international
law - lead to the same answer to the question before the Court: the southern channel is the
main channel of the Chobe River around Kasikili Island.

Chapter III

THE SUBSEQUENT CONDUCT OF THE PARTIES TO THE ANGLO-GERMAN
TREATY OF 1890 AND THEIR SUCCESSORS IN TITLE

82. This Chapter responds principally to Chapters V and VI of Botswana's Memorial, entitled
respectively 'The Interpretation of the Anglo-German Agreement of 1890' and 'The Pretoria
Agreement of 1984 and the Joint Survey of 1985.'83. Namibia demonstrated in its Memorial that the Masubia of Caprivi had occupied and
cultivated Kasikili Island from before the conclusion of the 1890 Treaty until well into the
second half of the present century and that Namibia's predecessors in title had continuously
exercised jurisdiction over the area with the full knowledge of Botswana and its predecessors

and without any official objection or protest from them until 1984. In Namibia's view, this
record not only confirms the interpretation of the Treaty as locating the boundary in the
southern channel of the Chobe River at Kasikili Island, but also constitutes an independent
title to sovereignty over the Island by operation of the doctrines of acquiescence, recognition
147
and prescription.

84. Botswana's Memorial does not systematically address these matters. It does contain a
lengthy history of the wanderings of the peoples of the area in pre-colonial times and a
lengthy description of Botswana's role in the struggle against apartheid and for the
148
independence of Namibia. (The account does not mention, however, that Botswana took
advantage of South Africa's withdrawal after Namibia's independence to occupy the Island

unilaterally by armed force in 1991.) None of this historical material contradicts in any
substantial way the account given in Part Two of Namibia's Memorial concerning the
occupation and use of the Island and exercise of jurisdiction over it by Namibia, nor does it
have any other bearing on the issues before the Court.

85. A few points that may seem to have relevance may readily be disposed of:

• Botswana implies that the British Order-in-Council of 30 June 1890 establishing the
boundaries of the Bechuanaland Protectorate, as supplemented by Orders-in-Council
of 9 May 1891 and 20 October 1898, somehow includes Kasikili Island within those
149
boundaries. The exact language of the Order, which might easily have been quoted
by Botswana to avoid any misinterpretation, refutes this implication:

The limits of this Order are: the parts of South Africa bounded by British Bechuanaland, the
150
German Protectorate, the Rivers Chobe and Zambezi . . .

There is no indication where the boundary lies within the Chobe River.

• Botswana states that the Caprivi Strip was de facto a part of the Bechuanaland
Protectorate during the period from 1921 to 1929 when Great Britain administered the
area as the delegate of South Africa, the mandatory power. The statement is incorrect
or at least seriously incomplete. The Namibian Memorial, paras. 233-237, describes in
detail how the British administration took care to keep the two territories legally

separate. The differences were more than formal. The tax systems in the two territories
were different; the poll tax imposed on the inhabitants of Bechuanaland was not
imposed on Caprivians. The Protectorate kept separate accounts and was reimbursed
for its expenditures in the Caprivi. Furthermore, since the British authorities ruled in
the Caprivi by delegation from the mandatory power, they were subject to the
supervisory jurisdiction of the Council of the League of Nations. The relationship

imposed significant constraints on British freedom of action in the territory, as
reflected in the annual reports to the Council.
• Botswana states that in 1924 the authorities at Kasane were 'reported to have given
151
verbal permission to Kasika residents to plough on Kasikili/Sedudu Island.' No supporting authority or references are given. Many of the witnesses at the JTTE
hearings held in May 1994 were asked specifically whether the Masubia had sought or
been required to seek permission to farm on the Island. They uniformly and vigorously
152
denied that they had.
• Botswana incorrectly states, 'The island of Kasikili/Sedudu forms part of the Chobe
National Park established in 1967 and, before that, was part of the Chobe Game
153
Reserve created in 1960.' The notice proclaiming the Chobe National Park copied
154
in relevant respects the notice proclaiming the Game Reserve, which states that the
reserve boundary follows the 'northern boundary of the Bechuanaland
155
Protectorate.' No map was attached to either notice or referred to in them. In 1960,
when the Game Reserve notice was issued, the maps in current use, and indeed the
only maps available of the whole reserve, were the Bechuanaland Protectorate
1:500,000 Map of 1933 and the derived version at 1:1,250,000, 1935. On these maps
the boundary is shown in the southern channel. Moreover, the copy of the 1935 map,

specially prepared in 1959 to show the bord156 of the Crown Reserve lands, clearly
excludes Kasikili Island from the Reserve. Since it was this Crown Reserve that
was converted to the Game Reserve a year later, the boundaries must have been

identical.

86. The remainder of this Chapter deals first with materials showing that the exercise of
jurisdiction by South Africa over the Island was recognised and indeed affirmed by Botswana.
Then it addresses the three instances of asserted deviation from the consistent history of
Botswana's acquiescence and recognition of Namibia's title.

A. Exercise of Jurisdiction over Kasikili Island by South Africa in the 1970s

87. Evidence from the 1970s shows unequivocally that South Africa vigorously asserted and
defended its jurisdiction over Kasikili Island and Botswana officially recognised it.

88. In 1972, six years after Botswana's independence and five years after the establishment of
the Chobe National Park, a Botswana magistrate recognised in a criminal proceeding that
Kasikili Island was Namibian territory and that Botswana had no jurisdiction over it. On 28
September, three Caprivians were arrested on Kasikili Island by game wardens from the
Chobe National Park and were detained in Kasane for five days before being brought before
the magistrate. According to contemporaneous affidavits made by two of the men, the

magistrate dismissed the case because they were arrested outside Botswana's jurisdiction.
According to one of the affidavits:

We were kept in custody at the Kasane police Station and appeared before the Magistrate of
Kasane after five days. After the interrogation and after the defence closed its case the
Magistrate found us not guilty and said that the island was part of the Caprivi and that we had
157
been arrested illegally.

The other said the magistrate 'acquitted us when giving judgement and said that we had been
arrested illegally and criticised the game warden for arresting us on Caprivi territory.'889. Recognition of Namibian jurisdiction and disavowal of Botswana jurisdiction over
Kasikili Island by a judicial officer of the newly independent Botswana Government, residing
in the neighbourhood and having personal knowledge of the situation, is the strongest kind of
evidence of the understanding of the parties as to the location of the boundary under the
Treaty and in practice.

90. Even though the magistrate had recognised Namibian jurisdiction, South Africa was not
content to let the matter rest there. The local authorities reported the matter to Pretoria, and
the Department of Foreign Affairs immediately initiated vigorous diplomatic representations
to Botswana. In its protest note to the President of Botswana, South Africa said:

As Kasikili Island lies within Eastern Caprivi territory, and as both the South African

authorities and the Executive Council of Eastern Caprivi view the matter with concern, the
Department would be pleased to receive the comments of the Office of the President on the
allegations made in the affidavits, with particular reference to the entry of armed Botswana
officials to Eastern Caprivi territory and also to the capacity in which the officials acted when
159
making the arrests.

The note also called Botswana's attention to the disposition of the case by the magistrate.
When Botswana did not reply to this communication, South Africa sent a follow-up
160
inquiry, again without a response, as far as Namibia has been able to discover in the South
African archives. If there was any situation that called for a prompt and energetic protest from
a state claiming sovereignty over the Island, this was it. Botswana's failure to respond to this
pointed and repeated assertion of jurisdiction by South Africa must be taken as recognition of

and acquiescence in it, especially in the light of the prior determination by Botswana's own
judicial authorities.

91. The report of this event to the Secretary for Bantu Administration and Development in
161
Pretoria also contains the information that 'the Caprivians have continued to cultivate
crops and hunt on the island since 1953 [shortly after the Trollope-Dickinson arrangement]
and the Botswanas have done nothing prejudicial to the Caprivian rights.' 162

B. Alleged Interruptions in the Unbroken Record of Recognition and Acquiescence by
Botswana

92. In its Memorial, Botswana claims that three incidents evidence non-recognition of
Namibian sovereignty by Botswana and its predecessors: the Eason Report in 1912, the

Trollope-Dickinson arrangement of 1951 and the follow-on of the shooting incident in 1984.
Namibia analysed these three episodes in NM, paras. 262-286, demonstrating that they in no
way contradicted or interrupted the continuous record of Botswana recognition of and
acquiescence in Namibian jurisdiction until 1984.

93. In this connection, it must be emphasised that the report of Captain Eason, the Trollope-

Redman report and the 1985 joint survey are merely conclusions drawn from such facts as
may have been available. Whatever other significance these conclusions may have, they
cannot override the facts themselves. If subsequent observation, properly recorded and
analysed, contradicts the factual basis for conclusions drawn by earlier observers from thefacts as they then appeared, it is the subsequent factual observations that must be given
controlling weight.

94. Namibia submits that an analysis of Botswana's arguments with respect to these episodes
leaves its position unchanged: the subsequent practice of the parties confirms that the proper

interpretation of the Treaty places the boundary in the southern channel, and that until 1984
this was recognised and accepted without protest by Botswana and its predecessors in title.

1. The Eason Report

95. There is nothing in Botswana's discussion of the Eason Report to contradict Namibia's
basic position. Eason went on an expedition in July 1912 to explore 'the various channels of

the Chobe, or Linyante, River, with a view to determining which can be legitimately claimed
as the "main channel."'163 His mission was not confined to or focused on Kasikili Island, but
comprehended the whole length of the river boundary.

96. In his report Eason stated with respect to 'Kissikiri island':

Here I consider that undoubtedly the North should be claimed as the main channel. At the
western end of the island the North channel at this period of the year is over one hundred feet
wide & 8 feet deep, the South channel about forty feet wide & four feet deep. The South
164
channel is merely a back water, what current there is goes round the North.

97. This is a recommendation in an internal report that a claim should be made - no more. But

the claim was not made. As Namibia stated in its Memorial, the expression 'should be
claimed' implies that it was not then being claimed. The report elicited no response from
Eason's superiors. Britain not only made no claim, it took no action of any kind, even of an
internal nature, based on the report. Captain Eason's conclusion certainly was not made
known to Germany. The situation thus remained in status quo. Britain was not claiming that
the northern channel was the main channel, even though one of its junior officials said it
ought to.

98. One or two additional points of detail lend support to Namibia's position. Eason's
instructions from Lt. Col. F.W. Panzera, the Bechuanaland Resident Commissioner at
Mafeking, laid down the criteria to be used: 'The width of channel[s] would have little to do
with it, & the question under consideration could only be solved by following up the deepest
165
channel in which there is the strongest current?' (emphasis in original) As noted above, the
instruction coincides with Professor Alexander's criterion that the channel that conveys the
largest proportion of the annual flow in the river is the main channel. (Supp. Rep., para. 2.1)

99. Eason was travelling in July, the dry season, and at the end of an 'exceptional drought
during the last twelve months.'166 We now know that in the dry season there is no perceptible
flow in either channel at Kasikili Island. (NM, Alexander Report, para. 1.6) Eason does not

quite say that. He says 'what current there is goes round the north.' But given the difficulty in
observing the current noted in his report and given what is now known scientifically about the
flow of the Chobe River at this point, his conclusion must be disregarded. If Eason had known
the facts revealed by the Alexander Report that substantially all of the annual flow of the
Chobe goes east to the Zambezi through the southern channel, he would surely have regardedit as the main channel.167 Indeed, under his instructions from Panzera he would have been

obliged to do so.

100. Eason was unable to complete his mission

owing to the lowness of the river. . . . The result was that the main channel of the Linyanti,
where it runs through the vast reed swamps which form the junction of the . . . Sunta with the
Linyanti was merely a succession of comparatively deep pools with mud & sand banks in

between. - There are no river banks to walk on so that when I was near the old town of
Linyanti I was compelled to abandon the work for this season. 168

101. Botswana, therefore, cannot rely on the Eason Report either as a claim to Kasikili Island
by Great Britain or as a protest or objection to German occupation and control. It cannot be
cited as evidence of official British opinion at the time since as far as Namibia has been able
to determine it was shared by no one, not even Panzera who was seeking to determine the
main channel. Its force as an expert opinion is vitiated by the erroneous factual observation on
which it was based.

2. The Trollope-Dickinson arrangement

102. There is no need to repeat the extensive discussion of this incident in Namibia's
Memorial. Again, the Botswana Memorial provides little to contradict Namibia's position. A
'joint report' by L.F.W. Trollope, the South African Magistrate for the Eastern Caprivi, and
Noel V. Redman, the District Commissioner of the Bechuanaland Protectorate, expressed the

opinion that '"the main Channel" [of the Chobe River] lies in the waterway which would
include the island in question in the Bechuanaland Protectorate.'169 Equally, they 'particularly
record[ed] that we have neither arrived at, nor expressed any joint opinion on the effect of
170
those facts on the ownership of the Island.' The report was forwarded to high political
officials of both governments for action.

103. An inconclusive exchange of correspondence at the official level ensued, after which the
issue was remitted to the local administrative officials for adjustment. The result was the
arrangement of 3 September 1951, under which the two officials 'agree[d] to differ on the
legal aspect regarding Kasikili Island' and that 'the position revert to what it was de facto
before the whole question was made an issue in 1947 - i.e. that Kasikili Island continue to be
used by Caprivi tribesmen and that the Northern Waterway continue to be used as a "free for
171
all" thoroughfare.'

104. For his part, Trollope maintained the position that the Island was part of the Caprivi
172
Strip, by virtue of use and occupation if nothing else. Within the British Government and
among the officials of the Bechuanaland Protectorate, much of the discussion proceeded on
the assumption that officially accepting South Africa's position would involve a 'slight

adjustment' of the boundary, implying that the Island was really in the Bechuanaland
Protectorate. That view, however, was not uniformly held. The Legal Advisers of the
Commonwealth Relations Office stated, 'If the Island has never been treated as part of the
Bechuanaland Protectorate it would seem best to say [in an appropriate Order in Council] that
the Island shall be deemed not to be included, and never to have been included, in theProtectorate.'73 Commenting on this suggestion, Mr. G.H. Baxter of the Commonwealth

Relations Office stated that the Order might be drafted

in the form of a declaration of an existing state of affairs, "for the removal of doubts" if it can
truthfully be said that there are any; with appropriate recitals explaining the position and, in
particular, the fact that the Island has never been administered as part of the Protectorate -
assuming that to be the case. In other words, the Order would show what for practical

purposes is the true position - that it would not amount to cession at all but to confirmation for
legal purposes of existing facts.74

105. Botswana seems to attach considerable significance to what it calls 'The Opinion of the
Surveyor-General of Bechuanaland, 18 October 1965.' 175 On examination, however, it

appears that the document referred to is not an 'opinion' but an internal memorandum to the
Permanent Secretary of the Ministry of Home Affairs. The memorandum consists primarily of
a summary of the internal and external correspondence concerning the Trollope-Dickinson
arrangement, some of which is quoted in BM, para. 160. The portion quoted by Botswana
fails to include the conclusion of the memorandum:

13. It appears, therefore, that if we now wish to use the island we have no alternative but to
re-open the matter with a view to either

(a) coming to a new administrative arrangement which would allow us the use of the island
without necessarily settling the question of ownership, or

(b) once and for all settling the matter of ownership, as one feels, admittedly after the event,
ought perhaps to have been done in 1947. 176

Thus, the memorandum adds nothing to Trollope-Dickinson documentation. Indeed, it
concedes that, as of the date of the memorandum, Britain was not the sovereign. The
document was never drawn to the attention of the South African Government nor was it made
public in any other way. It therefore has no status as a claim of right by Botswana.

106. Whatever internal debates may have taken place within the British colonial bureaucracy,
the conclusion in NM, para. 278, remains undisturbed:

[The British authorities] never considered a direct protest or objection to the South African
government as regards the existing situation and never thought of putting it on notice of a
formal claim in any other way. Indeed, . . . the communications from the British side seem to

have been drafted with some care to avoid making such a claim.

3. The 1984-1986 discussions between Botswana and South Africa

107. On 24 October 1984, Botswana troops opened fire on a South African patrol in the
channel south of Kasikili Island. The incident led to a serious and extended discussion

between the two countries. Botswana asserts that these talks eventuated in an
'intergovernmental agreement' that 'constitutes "an agreement between the parties regarding . .
. the application" of the provisions of the Anglo-German Agreement' within the meaning of
177
Article 31(3)(a) of the Vienna Convention on the Law of Treaties.108. The short answer to this contention is that there was no such agreement, and if there had
been it would have been void ab initio and without any consequence under international law.

a. Neither South Africa nor Botswana had legal power to enter into an agreement concerning
the boundary between Namibia and Botswana

i. South Africa

109. In 1984, South Africa had no legal authority to make boundary agreements in respect of
Namibia, especially agreements adverse to Namibia. The United Nations General Assembly,
on 27 October 1966, terminated the mandate under which South Africa had ruled the territory
since 1921 on the ground that South Africa had failed and refused to carry out its obligations
178
thereunder. Thereafter, the General Assembly assumed direct responsibility for the
territory of Namibia and established the United Nations Council for Namibia to administer the
territory until independence.

110. The action of the General Assembly was confirmed by the United Nations Security
Council in a series of resolutions culminating in Security Council Resolution 276 (1970),
which declared 'the continued presence of the South African authorities in Namibia . . . illegal'
and called upon all states 'to refrain from any dealings with the Government of South Africa'
179
that would imply recognition of South Africa's presence in Namibia.

111. The Court in its Advisory Opinion on the Legal Consequences for States of the
Continued Presence of South Africa in Namibia (South West Africa) Notwithstanding Security

Council Resolution 276 confirmed the actions of the political organs of the United Nations
and the illegality of South Africa's continued presence in Namibia. The Court analysed the
resolutions and decisions taken by the General Assembly and the Security Council regarding
Namibia and, in particular, the legal effect of paras. 2 and 5 of Security Council Resolution
276 (1970). It concluded 'that, the continued presence of South Africa in Namibia being
illegal, South Africa is under obligation to withdraw its administration from Namibia
180
immediately and thus put an end to its occupation of the Territory; . . .'

112. Both the General Assembly and Security Council resolutions affirm that 'South Africa
has no further right to administer the Territory,' and para. 2 of Security Council Resolution

276 expressly provides that 'all acts taken by South Africa on behalf of or concerning
Namibia after the termination of the Mandate are illegal and invalid.' The newly formed UN
Council for Namibia was alone empowered to conduct the international relations of Namibia
and specifically to 'replace South Africa as the party representing Namibia in all relevant
181
bilateral and multilateral treaties.' South Africa therefore had no power to make
international agreements with respect to Namibia.

ii. Botswana

113. Botswana was equally without power to conclude any agreements with South Africa
respecting Namibia. After affirming that 'all acts taken by the Government of South Africa on
behalf of or concerning Namibia after the termination of the Mandate are illegal and invalid,'
the Security Council called upon 'all States, particularly those which have economic and other
interests in Namibia, to refrain from any dealings with the Government of South Africa whichare inconsistent with paragraph 2 of the present resolution.'2 In interpreting this provision,

the Court in its Advisory Opinion on the Legal Consequences for States of the Continued
Presence of South Africa in Namibia (South West Africa) Notwithstanding Security Council
Resolution 276 said specifically that

member States of the United Nations are under obligation to abstain from entering into treaty
relations with South Africa in all cases in which the Government of South Africa purports to
act on behalf of or concerning Namibia. With respect to existing bilateral treaties, member

States must abstain from invoking or applying those treaties or provisions of treaties
concluded by South Africa on behalf of or concerning Namibia which involve active
intergovernmental co-operation. 183 (emphasis added)

114. The Security Council responded promptly to the Court's opinion. Resolution 301 (1971)
called upon all states:

(a) To abstain from entering into treaty relations with South Africa in all cases in which the
Government of South Africa purports to act on behalf of or concerning Namibia;

(b) To abstain from invoking or applying those treaties or provisions of treaties concluded by
South Africa on behalf of or concerning Namibia which involve active intergovernmental co-
operation.184

Under the express provisions of the opinion of the Court and the implementing resolution of
the Security Council, Botswana was under an obligation not to conclude any new
international agreements with South Africa on matters relating to Namibia. A fortiori it is
precluded from invoking any purported agreement before the principal judicial organ of the
185
United Nations.

115. Botswana makes much of discussions it held in New York in November 1984 with
certain officials of the UN Council for Namibia and the South West African People's
186
Organisation (SWAPO). But these officials did not purport to authorise Botswana to enter
into negotiations with South Africa, nor were they empowered to do so, still less to authorise
Botswana to conclude a treaty with South Africa affecting Namibia's boundaries.

116. So far as we know, the Botswana delegation held three meetings in New York, in each of
which it described the shooting incident and outlined its plans for discussions with South
Africa. The first, on 27 November 1984, was with Paul Lusaka, President of the UN Council
for Namibia. After describing the situation, the head of the Botswana delegation, Attorney
General M.D. Mokama, said, 'Before meeting with the South Africans Botswana wished to
187
consult with the U.N. Council for Namibia which is the legal authority in Namibia.' After
suitable diplomatic comments Mr. Lusaka said that he would 'consult with members of the
188
Council before giving a reply.' There is no record of any reply being given.

117. Later that day, the Botswana delegation met with Andimba Toivo Ja Toivo, SWAPO
Secretary General, and other high ranking SWAPO officials. The parties again exchanged

information about the shooting incident and the proposed talks with South Africa. 'Mr. Toivo
observed that SWAPO as a political party and a liberation movement did not have jurisdictionover the border issue.' He also commented that he was new to his post and not well-informed
on such matters.189

118. The next day, the Botswana delegation met with Mr. Mishra, UN Commissioner for
Namibia. Mr. Mishra 'confirmed that the Council for Namibia could not object to the
190
proposed talks.' He proposed, however, that the UN Secretary-General should be brought
into the picture. Because the Botswana delegation had to leave New York immediately, it was
left that Ambassador Legwaila, Botswana's Permanent Representative to the United Nations,
would brief the Secretary-General. There is no record of any such conversation having taken
191
place.

119. None of the three interlocutors with whom Botswana talked purported to authorise
anything approaching treaty negotiations or the conclusion of a treaty with South Africa.
Indeed, all expressly disclaim their authority to do so. Mr. Lusaka said he would have to

consult other members of the UN Council for Namibia. Mr. Toivo opened by denying that
SWAPO had jurisdiction over boundary matters. Although Mr. Mishra said that the Council
could not object to the proposed talks, it is not clear on what basis he made the statement, and,
in any case, he thought the matter should be referred to the Secretary-General, which was not
done. If, as Botswana's Memorial states, 'SWAPO thus gave its blessing to Botswana so that
192
she could discuss with South Africa the dispute over the boundary, . . .' the benediction
was without legal relevance.

120. Furthermore, there is no indication in the minutes recording Botswana's presentation at

the three meetings that it was proposing any kind of binding international agreement with
South Africa, which, as noted above and undoubtedly understood by all the participants,
would have been inherently illegal. What seemed to be projected were political talks to clear
up a serious boundary incident involving the use of force.

121. The evidence as to what occurred at these meetings comes entirely from minutes kept by
the Botswana delegation. Namibia has made a diligent but fruitless search of the UN archives

for information on these meetings or any other interaction between Botswana and the United
Nations concerning the boundary at Kasikili Island. It hardly seems possible that the United
Nations would have authorised treaty discussions without some official record of the action
being made.

iii. No agreement affecting the boundaries of Namibia could have been concluded without UN
General Assembly approval

122. Even if Botswana and South Africa had purported to conclude an agreement with respect
to the boundary, their acts would have been legally ineffective. A boundary agreement
affecting the territory of Namibia would have required the approval of the UN General
Assembly.

123. During the period when the Mandate was in force, South Africa had treaty-making
power for the mandated territory, subject to the important exception that treaties affecting
boundaries required prior approval by the Council of the League of Nations before coming
into force. This was because, as the mandatory power, it did not have sovereignty over the
mandated territory. The requirement of prior approval by the Council derived from Article 7of the Mandate, which provided: 'The consent of the Council of the League of Nations is
193
required for any modification of the terms of the present Mandate.' In turn, Article 1
defined the territorial extent of the Mandate as comprising 'the territory which formerly
constituted the German Protectorate of South-West Africa.' 194 Thus any agreement affecting

the location of the boundary would be a modification of the terms of the Mandate, requiring
Council consent.

124. With the demise of the League of Nations in 1945, most mandates were transformed into
trusteeships under the supervision of the UN Trusteeship Council. South Africa, however,

refused to enter into a trusteeship relationship with respect to Namibia. In these circumstances
the Court held that the Mandate continued in effect, with the powers and responsibilities
formerly vested in the Council of the League of Nations to be exercised by the General
195
Assembly. With the termination of the Mandate in 1966, therefore, the legal responsibility
for administration of Namibia was assumed directly by the United Nations and exercised by
the General Assembly. 196 As noted above, from 1967 on administrative authority was vested

in the UN Council for Namibia, subject to the supervision of the General Assembly. Thus the
safeguard with respect to boundary treaties contained in the Mandate remained in force, and
any such agreement affecting the boundary would have required prior approval by the General
Assembly, the supervisory UN body.

125. In short, in 1984-1985, neither South Africa nor Botswana had any power to enter into
any bilateral international agreement whatever respecting Namibia. Moreover a bilateral
agreement affecting the boundary between Namibia and Botswana would have required the
approval of the UN General Assembly. Thus, any such agreement concluded between them
could not be recognised for any legal purpose by the United Nations or any of its organs
(including the Court) or by any member State. It would have been null and void and without

any effect either under the provisions of the Vienna Convention or for any other purpose
under international law.

b. The exchanges between the two countries did not purport to establish and did not in fact
eventuate in a bilateral agreement concerning the location of the boundary at Kasikili Island

126. On 19 December 1984, representatives of Botswana and South Africa met in Pretoria to
discuss the problems raised by the shooting incident and in the course of their discussions
decided that a joint survey should be conducted by the technical departments of the two
governments to determine the location of the main channel of the Chobe River around
Kasikili Island. The survey was conducted by officials of the Botswana Department of
Surveys and Lands and the South African Directorate of Surveys and Mapping on 2 July 1985

and concluded that '[t]he main ch197el of the Chobe River now passes Sidudu/Kasikili Island
to the west and to the north of it.'

127. Botswana implies, however, that the parties went further and agreed that the findings of
the survey would be binding as to the location of the boundary between the two countries.
Botswana's Memorial does not make this claim in so many words. The closest it comes is its
remark that 'the Pretoria Meeting was a binding intergovernmental agreement relating to a
legal subject matter and concluded with senior lawyers in attendance. The parties clearly
198
intended to achieve an effective and therefore binding outcome to their deliberations.'Botswana admits that it can point to no document or oral expression embodying the terms of
this asserted agreement.199 Instead it is fabricated from a tissue of 'a whole series of
200
documents and inferences to be drawn from the circumstances.' An examination of the
documents recording the 19 December discussions and the subsequent conduct of the parties,

however, provides no support for Botswana's contention. Botswana's minutes of the meeting
show that much of the discussion was devoted to security issues along the border in general,
which the parties agreed to handle by 'a workable agreement on the ground, between the two
Defence Forces' and 'closer liaison at a low (platoon) level . . . ' This was not the

conversation of people who were in the process of concluding a major international boundary
agreement. The decision to conduct a survey of the main channel was simply a determination
as to the next step to be taken in an ongoing diplomatic effort to resolve the problem of
recurring border incidents between the armed forces of the two countries.

i. The legal capacity of the two delegations

128. In the first place, the delegations present did not represent their states 'for the purpose . . .
of expressing the consent of the State to be bound by a treaty' within the meaning of Article 7
of the Vienna Convention on the Law of Treaties. Neither delegation exhibited full powers
(see Article 7(1)), and neither included officials who under Article 7(2) may have had
202
authority to express consent to be bound. The absence of such powers or officials on both
sides makes it clear that there was no intention to make a binding treaty or international
agreement.

ii. The content of the discussions

129. But in any case, Botswana does not and cannot point to any language in the minutes of
the meeting, either in the Botswana or the South African version, that imports any agreement
as to the consequences to be attributed to the survey. It is highly unlikely, moreover, that
either side was prepared to entrust the location of the boundary in a dangerous and disputed
area to the outcome of what all regarded as a 'technical' survey. This was clearly so with

respect to South Africa, whose delegation was under peremptory orders from the Chief of the
South African Defence Force (SADF) to 'confirm that the border lies south of Sidudu.' 203
Even under normal circumstances, a Ministry of Foreign Affairs cannot simply ignore such a

letter from the Chief of its Armed Forces. But the situation in the area at the time was
anything but normal. The SADF was waging a vicious war against the Namibian people led
by SWAPO in which the Caprivi area and Kasikili Island itself had special strategic
importance. In the light of these instructions, the middle level officials at the Ministry of
Foreign Affairs representing South Africa at the Pretoria meeting clearly could not agree to
any other location of the border without consulting the SADF.

iii. The subsequent conduct of the parties

130. Contrary to Botswana's contention, the subsequent conduct of the parties belies the
204
conclusion that any agreement was reached as to the binding character of the survey.
Botswana's first official reaction to the joint survey report was a note dated 4 November 1985,
enclosing a copy of the report and inquiring 'whether or not the South African sides [sic]
wishes to have a meeting called to adopt the report formally. Alternatively the South Africanside could simply signify its acceptance of the conclusions of the report by means of a
205
Diplomatic Note.' Thus, Botswana explicitly recognised that further political action was
necessary before the 'technical' work of the survey team could have any effect. South Africa
made no reply to this note.

131. It was not until nearly a year later that Botswana changed its position. At a meeting
between officials of the two countries on 13 October 1986 to discuss security issues,

Botswana's Secretary for External Affairs stated that 'there was no more room for negotiation
because a joint Botswana-South African team of experts had confirmed that the Island
belonged to Botswana.' 206 Botswana confirmed this position in a telex to the South African

Ministry of Foreign Affairs nine days later.07

132. South Africa, however, did not agree with Botswana's position in the 13 October
meeting. There is a difference of nuance between the minutes taken by the two sides of the
meeting, but both agree that the brief discussion of the status of the Island was initiated by the
South African representative, the Deputy Minister of External Affairs. According to the
Botswana minutes he 'suggested the maintenance of the status quo till political circumstances
208
could permit direct negotiations between Botswana and independent Namibia.' When
Botswana made its submission as to the effect of the joint survey report, the South African
209
representative '[d]ecided to go back to look at this question once again.' In the South
African version, Mr. Van Heerden, the South African representative, says that 'this finding
[that the Island was part of Botswana], if accepted, was tantamount to changing an
international border and that this could not be done without the transitional Government in
210
Windhoek.'

133. When the follow-up telex arrived, South Africa rejected it, stating:

According to International Law, such cases should be discussed between the two countries
concerned. It is therefore suggested that the Cabinet of South West Africa/Namibia should be
approached by the Botswana Government for a proper resolution of the matter under
211
consideration.

134. South Africa has maintained this position at least until the outbreak of the present
dispute. In response to Namibia's letter of inquiry in 1992, the South African Minister of

Foreign Affairs wrote:

The joint survey you refer to in your letter, a copy of which I attach, did come to certain
conclusions, but according to legal opinion in South Africa at that time, did not prove
conclusively that Sidudu island belongs to Botswana.

. . .

The matter has therefore not been resolved as South Africa has never officially recognized
212
Botswana's claim to Sidudu island.135. In summary, Botswana first took the position that further action by the foreign offices of
the two countries was necessary but later changed its mind and asserted that the joint survey
report was self-executing. South Africa, however, refused to accept Botswana's claim and
consistently maintained that further action at the political level was necessary to resolve the
dispute.

c. The joint survey report is entitled to no weight as an expert opinion

136. Botswana is finally reduced to the claim that the joint survey report constitutes expert
opinion as to the main channel of the Chobe River in the vicinity of Kasikili/Sedudu
213
Island. Like all expert opinions, however, this is to be examined in light of the factual
observations on which it was based. The report contains no analysis or reasoning in support of
its bald conclusion that '[t]he main channel of the Chobe River now passes Sidudu/Kasikili
214
Island to the west and to the north of it.' Since the principal activity of the survey team
was to take depth soundings at various cross-sections of the river, it would appear that they
regarded the comparative average depth of the channels as the principal applicable criterion.
The Alexander Report shows that depth is not the proper scientific basis on which to
determine the main channel of the river. (NM, Alexander Report, para. 4.4) Even on the

question of depth, the measurements taken do not permit an accurate comparison of the two
channels. They show only the depth at particular cross-sections, but there is no effort to
determine the shallowest point along the length of either channel, which in both cases occurs
close to the western bifurcation and is the dimension that controls navigability. Moreover, the
survey was taken in July, the low water period, as also appears from the sketch map attached
to the Report. (NM, Alexander Report, Sheet 26, Diagram 8a) At that time, as the Alexander

Report215ows, the water in both channels is stagnant, and there is no flow through either of
them. The methodology of the joint survey report was faulty and the scientific basis of its
conclusion was incorrect. It is therefore entitled to no weight as expert opinion.

C. Conclusion as to the Subsequent Conduct of the Parties

137. The foregoing review of the materials relating to the subsequent conduct of the parties in
Botswana's Memorial serves only to confirm the position that Namibia and its predecessors in
title were in possession of Kasikili Island from 1890 to 1991. During all that time it exercised
jurisdiction over Kasikili Island. All this occurred with the knowledge and acceptance of the

Botswana authorities and without protest from them until 1984. This record on the one hand
confirms the interpretation of the Treaty placing the boundary in the southern channel and, on
the other, establishes Namibia's right to the Island by operation of the doctrines of
acquiescence, recognition and prescription.

Chapter IV

MAPS AS EVIDENCE OF SUBSEQUENT CONDUCT OF THE PARTIES

138. This Chapter replies to Chapter VIII of the Botswana Memorial, entitled 'The Relevance
of Map Evidence.'

A. General Considerations as to the Map Evidence in this Case139. In its Memorial, Namibia presented a comprehensive view of the cartographic history of
the region in dispute, discussing every map it had been able to find that had been produced by
the parties to the dispute or their predecessors in interest and that was large enough in scale
216
and adequate in detail to show the boundary at Kasikili Island. What emerged from this
review was a remarkable consonance as to the location of the boundary in the southern
channel of the Chobe River at Kasikili Island. Every map, with but two exceptions (both of
which can be shown to be the result of cartographers' errors) and maps produced by Botswana

itself after 1974, agrees in this placement of the border. The Botswana Memorial seeks to
obfuscate this substantial unanimity by a mélange of inaccuracies, irrelevancies and
innuendoes. It deprecates the general significance of map evidence in boundary cases. It
erroneously implies that the maps available to assist the Court are mostly of too small scale to
show the area around Kasikili Island and the location of the boundary in adequate detail. And
it seems to suggest, again erroneously, that the principal maps are not 'official' and thus not

entitled to significant consideration by the Court. This Chapter first addresses these three
general points and then analyses in detail Botswana's comments on specific maps of interest
in the present case.

1.The relevance and weight of cartographic evidence

140. At the very outset of its discussion of the subject, the Botswana Memorial seeks to
disparage the general significance of map evidence in boundary cases. 217 The fact remains,

however, that Professor Ian Brownlie, in his justly renowned book on African boundaries,
states at the beginning that maps are important sources of evidence and provide

material for the purpose of determining the existence of an alignment and its status in terms of
218
acceptance and recognition by the states concerned . . .

A few pages later he continues:

The evidence of recognition and acquiescence by conduct may take the form of . . . the

publication of official maps, [and] reliance for official purposes on maps showing the
alignment . . .19

In his most recent pronouncement on the issue he reiterates that 'maps have been accepted as
admissions against interest and evidence of acquiescence when they were made public and
220
given official approval. . . .' And in general,

tribunals concerned with disputes as to sovereignty over territory have commonly accepted
the evidence of maps with an official provenance as evidence of the views of Governments
221
and of political figures and officials with special knowledge as to political matters of fact.

In the view of the authorities, map evidence is subject to no special vulnerability. Like all
other evidence, a map is to be weighed by the tribunal for its probative value, based on its

provenance and history, the care with which it was produced and the quality of the
information it contains.222 In this regard, Mr. W. Dennis Rushworth, former Director of the
Mapping and Charting Establishment of the British Ministry of Defence, who prepared the

comprehensive report on the cartographic history of the Caprivi Strip included in theNamibian Memorial as Annex 102, has provided a similar technical analysis of the maps
discussed in Chapter VIII of the Botswana Memorial, which is appended as Annex 1 to this
Counter-Memorial. Paragraph 22 of Annex 1 contains a list of misstatements of fact
concerning maps to be found in the Botswana Memorial.

2.The scale of the relevant maps

141. The Botswana Memorial, para. 257, states that the scale of the maps relating to the
223
Caprivi Strip 'is usually so small as to present no indication of the riverine topography.' On
the contrary, of the 16 maps included in Namibia's Atlas, 12 are of large enough scale to show
Kasikili Island, and all ten of those produced after the Treaty clearly show the boundary in the
southern channel. Indeed, it is only the Botswana Memorial that seeks to rely on maps of a
224
scale so small as not to show the Island or the boundary around it. As such, those maps
are, by Botswana's own admission, irrelevant to the issue before the Court.

142. The Botswana Memorial seems to make a point that on some of these maps the boundary
is marked on the north bank of the river. Thus, BM, paras. 270-273, discusses four 'Sketch
Maps of the Bechuanaland Protectorate' published in a series of annual reports on the
Protectorate issued by the British Colonial Office from 1912 to 1915. The scale of these maps
is not given, but it is actually 1:4,600,000, far too small to show Kasikili Island. Botswana
notes that the first three of these Sketch Maps show the boundary to the south of the Chobe,
225
but the fourth 'indicates a boundary, very clearly, to the north of the Chobe.' (emphasis in
original) The Botswana Memorial seems to imply by the emphasised phrase that the change
has some special significance. The fact is, however, that the series of annual Bechuanaland
226
Protectorate reports continued until 1965 just before the end of British colonial rule.
Nineteen of these reports included maps at the same or similar scales. Fifteen of these show

the boundary on the south side of the Chobe and four on the north side. Namibia draws no
inference from this statistic as to the position of the boundary at Kasikili Island. But it is
equally clear that the change from the south side of the river in the 1913-1914 Report to the
north side in the 1914-1915 Report has no evidentiary significance on the issue, no matter
how heavily it may be emphasised in the Botswana text. The variation in treatment results
because on small-scale monochrome maps with riverine boundaries, like those in the Colonial

Report series, the cartographic practice is to put the boundary symbol alongside the symbol
for the river to indicate that the boundary follows the centre of the river. The side of the river
on which the boundary symbol appears, however, is entirely at the draughtsman's
convenience, usually in regard to clearer presentation of features of interest near the
227
frontier. For example, the legend of GSGS 2681 of 1913 states 'where [boundaries] follow
a road or the main channel of a river, they have been shown to one side, to prevent
228
confusion.'

3. Official as opposed to 'private' or non-official maps

143. The Botswana Memorial, paras. 261-262, calls attention to the distinction made between

official and unofficial maps, citing authorities to the effect that official maps are generally
more reliable than those produced and published under other auspices. Whatever the accuracy
of this proposition as a matter of law, it has no applicability to the present case, since, without
exception, the maps relied on by Namibia are official maps. 229 The Seiner map of 1909 waspublished by a private publisher in Berlin, but it was sponsored by the German Government
and used by it during the entire time it exercised authority in the Caprivi. Streitwolf's map,
which was printed on the official presses of the Landesvermessung (Cadastral Survey), and
von Frankenberg's map were both produced by German Imperial Residents in the Caprivi on
the ground in the course of their official duties and for the purpose of assisting them in the

performance of those duties. The British maps discussed in both Memorials are the products
of government departments, in most cases of the official map-making offices, as are the maps
from South Africa and Botswana. The UN maps are self-evidently official.

144. Where the scale of these maps is large enough to show the details of the boundary in the
area of concern, the mode of indicating the international boundary is, with the exception of
the Kriegskarte, relied on by Botswana and discussed more fully below, anything but
230
'impressionistic.' The depiction of international boundaries on official maps published by
well-endowed governments is a matter of painstaking care, performed according to well-
established standard operating procedures and subject to careful vetting by political officials

in cases of doubt. That does not mean, of course, that they are necessarily free from error, and
indeed Namibia has shown that some of the official maps contain egregious mistakes. It does
mean, however, that Botswana's characterisation of the boundary placement as
'impressionistic' is wholly inapposite.

B. Analysis of the Maps Relied on by Botswana

145. The Botswana Memorial, like that of Namibia, groups the maps to be discussed in
accordance with the country of origin, and that organisation is followed here.

German maps

a. 'Kriegskarte von Deutsch-Sudwestafrika [sic] 1904' (BM, paras. 266-267)

146. After careful study in the preparation of its Memorial, Namibia came to the conclusion
that this map was simply too confusing and contained too many obvious errors to be relied on
for any purpose and therefore omitted it from the Atlas of maps Namibia presented to the
231
Court. The reasons for this decision will sufficiently appear from the following discussion.

147. In the first place, Botswana Maps 4 and 5, discussed respectively in BM, paras. 266 and
267, are not two separate sheets of the Kriegskarte, the Andara and the Linjanti sheets, as the
232
Botswana Memorial states. They are two different extracts of the same map, the Linjanti
sheet, one in monochrome and the other in colour. Neither provides the marginal information
necessary for interpretation, and neither contains all the detail of the original map, which is to
be found in the British Public Records Office and was consulted by Mr. Rushworth in

preparing his technical analysis in Annex 1, para. 12.

148. The Kriegskarte was produced in Berlin from existing maps in the short period of five
weeks. 233 An extract appears at Fig. 1, following p. 66. It was intended for military use in the

German-Herero wars. For the area of the lower Chobe River the main source was the Schulz
and Hammar map. 234 The label 'Sulumbu's I' appears north of the river, but it is not clear

what feature on the map the label refers to. This part of the river is represented on the map by
an area of blue colour bounded by black lines of varying thickness. The Botswana Memorialasserts that the map 'appears to show the northern channel by a thick black line and the
southern channel is barely visible except as the edge to the shaded area which represents the
235
island.' There is no basis for these assertions. The black lines enclose a 'shaded area' on
Map 4, but Map 5 reveals this to be a dark blue coloured area, which elsewhere on the map
and by the usual cartographic convention indicates a body of water, not an island at all. In that
case the lines do not represent channels of the river, but the banks.

149. There is no reference to boundaries in the map legend, but boundaries are shown by a

chain-dot line in some areas, such as the Western Caprivi. Along the rivers, however, there is
no such boundary symbol. The boundary is backed intermittently on the Namibian side by a
faint band of pink colour that sometimes continues along the rivers, but the pink disappears
beneath the dark blue colour of the water bodies, so that it is of no help in determining the
location of the boundary within them. In short, the map provides no warrant whatever for the
conclusion that Kasikili Island is in Botswana.

150. Other serious anomalies cast doubt on the overall reliability of the map. For example, it
shows a road going on a direct line east from Ngoma across the Chobe River to Sebuba-Fälle.
No such road is shown on any other map of the area from the explorers' maps before the
Treaty to the present. The depiction of the large areas covered by water is also mysterious. No
map has ever shown a permanent body of water of the size and configuration that appears on
the Kriegskarte in the area well to the west of the general location of Kasikili Island. Neither

can these large blue patches be explained as a representation of the river in flood. It will be
recalled that the floodplain of the Zambezi is inundated annually from east to west, beginning
at the Mambova Rapids and gradually filling the triangle between the two rivers, normally as
far as Ngoma. When the flood season ends, the waters recede from west to east, leaving the
western reaches of the floodplain exposed while much of the eastern portion is still
236
flooded. This same pattern was confirmed by the early travellers and by Streitwolf in
1909, only five years after the publication of the Kriegskarte.237 There is no record ever of a

time when the western portion of the area remained under water while the eastern reaches of
the river were in something like their usual dry season condition, as appears on the map.

151. For these and similar reasons, Namibia decided that the map was too unreliable to
present to the Court. As stated by Dr. Huber, the arbitrator in the Isle of Palmas case:

Any maps which do not precisely indicate the political distribution of territories, and in
particular the Island of Palmas (Miangas) clearly marked as such, must be rejected forthwith,
unless they contribute - supposing that they are accurate - to the location of geographical
names. Moreover, indications of such a nature are only of value when there is reason to think
that the cartographer has not merely referred to already existing maps - as seems very often to
be the case - but that he has based his decision on information carefully collected for the
238
purpose.

The language is quoted in the Botswana Memorial at para. 262. But, without changing a word

except to substitute 'Kasikili' for 'Palmas,' it tells why the Kriegskarte can be of no assistance
to the Court in this case.

b. 'Von Frankenberg, Karte des Caprivi Zipfels, Blatt I, 1912' (BM, para. 268)152. Botswana asserts that this map identifies 'the northern and western channel as the main
239
channel.' The only basis for this assertion is that the southern channel is marked 'Kassikiri
Flussarm,' which, according to Botswana, translates as 'branch or tributary of a main
river.'40 The only support for this assertion, in turn, is an 'annexure' to the map. On closer

examination, the annexure appears to be a translation prepared in 1994 for the purposes of
Botswana's submission to the JTTE. 241 It seems to be pasted on the map in Botswana's
242
atlas. The signature on the translation is illegible, and the author's credentials are not
given. It says that 'the translation also appears in standard works of reference,' but none are

cited.

153. In fact, the annexure mistranslates the word Flussarm. This appears from the legend on
von Frankenberg's map itself. Towards the left margin of the map, apparently in von
Frankenberg's own hand, appears the note:

Die politischen Grenzen sind durch den Zambesi und den Linyanti gegeben. Bei der
Mongonda Insel ist der nördliche Flussarm zwar breiter, jedoch flächer als der südliche, so
dass er wahrend der Trockenzeit grosstenteils austrocknet.

This note translates as follows:

The political boundaries are defined by the Zambezi and the Linyanti. At Mangonda Island [in
the Zambezi], the northern arm of the river is broader, but it is shallower than the southern,
and during the dry season therefore it is generally dried up.43

Thus, von Frankenberg himself used the word Flussarm to mean simply a branch of a river,
with absolutely no inference that it was 'a tributary to the main river,' as asserted by
Botswana. This is the standard definition given in many other lexicographic and hydrographic
244
sources.

154. As shown in NM, paras. 298-299, the true conclusion to be drawn from von
Frankenberg's map is the opposite: the southern channel is the boundary. Although Botswana

is correct that 'there is no boundary marked as such on the map,' von Frankenberg obviously
knew, as he said in the note, that the Linyanti River formed the political boundary between
German and British territory. Places on the north or German side of the river are named in
245
German. On the south side of the river, no place names are to be found. 'Insel Kassikiri
Flussarm' appears as a place name in German designating the Island. The inference is
inescapable that von Frankenberg thought that Kasikili Island was within his area of
246
responsibility.

c. The German maps that Botswana does not mention

155. The most important map of the German period was Seiner's map, discussed at NM,

paras. 293-295. It was the principal large scale map used by the German Government in
Berlin and in the field during the entire period of its rule in South West Africa. It was sent by
Germany to the British Foreign Office to indicate Germany's position on the southern
boundary of the Caprivi during the negotiations over the western portion of the boundarybetween 1909 and 1914. 247 This map is relegated to a footnote in the Botswana Memorial
248
which acknowledges that it is 'an impressive map.' The footnote says, however, that 'the
boundary is not shown' on the map. This is simply wrong. As can be clearly seen in the

extract in NM, following p. 121, Seiner's map shows the boundary along the Chobe River by
a fine red hatching on the Namibian side. The hatching is drawn very precisely to cover
Kasikili Island, thus placing it unequivocally in Namibia.

156. Streitwolf's map is not even mentioned in the Botswana Memorial, although it was much
more widely known than von Frankenberg's and was relied on by Captain Eason on his

exploration of the Chobe River. Like von Frankenberg's map, discussed above, Streitwolf
indicated that Kasikili was within German South West Africa by naming it, while places lying
outside his domain in English territory are not named on the map. 249

2. British maps

157. As has already been discussed, and as the Botswana Memorial admits, the sketch maps
included in the annual Colonial Reports on the Bechuanaland Protectorate are too small in
250
scale to show Kasikili Island. Moreover, the side of the river on which the boundary
appears on these maps is a matter of the cartographer's convenience and discretion and has no
significance as to the location of the boundary within the river. The same is true of the Sketch

Map of Bechuanaland Protectorate (1906), discussed in BM, para. 269, and Bechuanaland
Protectorate, D.O.S. [Misc.] 282, 1960, Directorate of Overseas Surveys, discussed in BM,
para. 276. These maps are therefore irrelevant to the issue for which Botswana cites them.
They can be of no assistance to the Court in resolving the issues before it. That leaves only
two British maps referenced in the Botswana Memorial with any bearing on the case. They
are discussed immediately below.

a. 'British War Office Map of Protectorate, 1933; Geographical Section, General Staff,
No.3915' (BM, paras. 274-275) 251

158. This map was the official British map used in London and by all Bechuanaland
administrative agencies for the entire period from 1933 to 1965, one year before
independence. As Botswana must and does acknowledge, it clearly shows Kasikili Island in
Namibia. The map is analysed at length in NM, paras. 305-309 and Annex 102, paras. 23-24,
establishing that this was the first British attempt to produce an accurate large scale map of
252
the area. It has recently been characterised as 'a significant milestone in the cartographic
253
history of Botswana.' Botswana adduces four grounds for contending that '[t]he evidential
254
value of this indication of the boundary is substantially reduced . . .'All four are without
merit.

(i) Botswana states, 'The boundary is shown as an "intercolonial" not as an "international"
255
boundary.'

Why this is supposed to be significant is not revealed. Whether 'intercolonial' or
'international,' the boundary marked the line between two territories under different

sovereignties. Even more to the point, the Court expressly held in the Frontier Dispute casethat under the principle of uti possidetis, intercolonial and even intracolonial administrative
boundaries are to be recognised as the international boundary between successor independent
0
states.

(ii) Botswana states, '[T]he map is a compiled map produced by technical experts from
1
previous maps in the absence of reference to legal questions.'

This statement is erroneous. The detail in the Chobe River area on GSGS 3915 was new,
based not on previous maps but on aerial photography taken in 1925 by the Kalahari

Reconnaissance. The boundary, of course, can only be added in the drawing office, and as
noted above, in the case of official maps like this one produced by major survey offices,
boundary placement is done by highly skilled professionals in accordance with well-
2
developed standard operating procedures subject to close technical and political supervision.

In this particular case, the production of the map was initiated by the Bechuanaland
Protectorate in order to meet its administrative needs, and the form and content were specified
by senior officers of the Protectorate acting in their official capacity. Much of the material for
3
the map was supplied by the Resident Commissioner, Sir Charles Rey, and his staff. The
High Commissioner, the Earl of Athlone, took a strong interest in the project and was
particularly concerned that the boundaries should be correctly depicted. The enterprise

consumed five years. The original compilation material has been destroyed, and therefore it is
not possible to identify the precise person who decided where the boundary symbol should be
placed, but it was very probably one of the officers of the Protectorate. In any case, the
boundary would have appeared on the proof copy of the map, which was reviewed in
Mafeking and returned to London in November 1932 with comments, but without objection
as to the location of the boundary. Thus the boundary designation was known to and accepted

by both the London and Protectorate authorities.

(iii) Botswana states, '[T]he fact that the entire alignment of the boundary along the Chobe is
indicated south of the river establishes that the map maker had no interest in a precise
4
representation of a boundary following the main channel of the river.' (emphasis in original)

The Chobe River in the relevant area of this map is shown by a double line which separates
into two double lines at Kasikili Island, one for each channel. The double lines are too close
together to put the boundary marker in the middle of the river (or of either channel when the

river separates). But there is plenty of room between the two channels to accommodate a
boundary marker to the south of the northern channel. Instead, as shown in NM, Fig.
13,following p. 125, the draughtsman deliberately chose to attach the indicator to the southern
channel.

5
Moreover, once published, GSGS 3915 (and its derivative ), became the basis for many
operational, planning and research activities of the Bechuanaland Protectorate. The Namibian
Memorial cites two maps illustrating this practice. One, which was the subject of very careful
6
vetting, showed the boundaries of the Crown Reserve lands in Botswana, and the other dealt
7
with water development schemes. Both place Kasikili Island deliberately and unmistakably
outside of Botswana. Two additional examples of the use of GSGS 3915 by administrativeagencies of the Protectorate illustrate the same understanding as to the boundary at Kasikili
Island. One, issued by the Veterinary Administration and showing zones for various animal
diseases, depicts 'Boundaries of Vet Areas' with a grey broken line. One such boundary
follows the southern channel at Kasikili indicating clearly that the Island was not within the
Veterinary Administration's jurisdiction. (See Fig. 2, following p. 73) The other, annotated

with geological information, shows potential coal bearing beds by cross-hatching in red ink.
Again the cross-hatched area stops at the southern channel of the Chobe, indicating that
Kasikili Island was not thought to be within the responsibility of the resource department
using the map. (See Fig. 3, following p. 74)

(iv) Botswana states, 'In the period 1915 to 1929 the Caprivi had been administered as part of
[the] Bechuanaland Protectorate and the maps relied on in the production of the 1933 map

relate predominantly to this period. In a period when the United Kingdom was in functional
terms the sovereign administrator on both sides of the Chobe it would be unlikely that the
precise depiction of the boundary on a map drawn on a scale of 1:500,000 would be a matter
8
of particular concern.' (emphasis in original)

The proposition is a complete non sequitur. If anything, in 1933, having recently returned
control of the Caprivi to South Africa, the authorities in London and the Bechuanaland
Protectorate would have been particularly concerned to have an accurate depiction of the
boundary, which now divided territory administered by two different sovereigns. In any case,

as fully demonstrated in NM, paras. 233-237, during the period when the area was
administered by Great Britain as delegate of South Africa, the colonial officials were
scrupulous in maintaining the legal formalities of the separate existence of the two territories
under common control.

Moreover, the source maps listed in the margin of GSGS 3915 covering the Chobe area do not
fit the Botswana Memorial's description. Von Frankenberg's map was not a product of the

period of British administration. Although, as we saw, it does not show the boundary as such,
it indicates that Kasikili Island is within German territory. Two other source maps, the Sketch
map by Stigand and the Kalahari Reconnaissance map do not show the boundary at Kasikili
and thus could not have been the source of the boundary information. The sketch map by
Forster Towne, who was on the staff of the Resident Commissioner for the Bechuanaland
Protectorate, was a manuscript specifically prepared as an input to GSGS 3915 and was the
most likely vehicle for depicting the Protectorate's view on the position of the boundary. It is
9
recorded as destroyed by the War Office along with the other compilation material.

159. Botswana's attempt to denigrate the significance of GSGS 3915 is without substance.

The map remains the most important and substantial British mapping effort of the area in the
entire period of its administration. GS 3915 was reduced in scale to 1:1,250,000 and reissued
in 1935,10 and a second edition of the original map was published in 1964. 11 In 1984, almost

two decades after Botswana's independence, the Surveyor General of Botswana recognised
the significance of this map by reprinting the 1:1,250,000 version as a historic document. The
provenance of GSGS 3915, its methodology and the uses to which it was subsequently put
leave no doubt that it reflected the British understanding and embodied the official British
position as to the location of the boundary at Kasikili Island.

b. 'Bechuanaland, 1965: Directorate of Overseas Surveys (D.O.S. 847) (Z 462)' (BM, para.
277)160. The Botswana Memorial says that the 'boundary . . . clearly follows the northern and
western channel in the vicinity of Kasikili/Sedudu.'12 This statement is not strictly true. This

map shows only one channel, not two channels in the vicinity of Kasikili Island. Thus it does
not show Kasikili Island at all. The river as depicted in that reach looks like the northern
channel on other maps, but there could have been no choice by the mapmaker of the northern
over the southern channel for the boundary, since no southern channel appears on the map.
The reasons for the obvious mistake in depicting this portion of the river are discussed at
length in NM, para. 310 and Annex 102, para. 30. The error is so egregious, however, that it

renders the map useless for the purposes of this case.

c. 'Africa 1:2,000,000 War Office GSGS 2871 Sheet Rhodesia'

161. One additional map was produced by the British War Office and published in several
editions during the period 1919 to 1958. It is a map that is part of an Anglo-French series
covering the whole of Africa, developed for military and subsequently aeronautical use. It is a

carefully drawn map, showing portions of Botswana and Namibia, including the boundary
between them in the Eastern Caprivi, in considerable detail. 13 The 1933, 1940, 1942 and
1958 editions of this map show clear depictions of Kasikili Island, slightly exaggerated in

size, with the boundary marked by a symbol along the southern side of the southern channel.
In the first three, the depiction of the Island is identical, except that on the 1942 edition, a
purple stipple band is added, internal to Botswana, the edge of which follows the southern
channel. (See Fig. 4, following p. 75) The 1958 edition has been redrawn, but Kasikili Island
appears on it too with the boundary as before, in this case emphasised by a red stipple band.
(See id.) This map was produced and published by the British War Office in London. Thus, at

regular intervals during the period of British colonial rule, the British colonial authorities
issued official maps reaffirming that Kasikili Island is in Namibia. The last of these, it should
be noted, was issued ten years after the Trollope-Redman joint inspection of the area and
seven years after the conclusion of the Trollope-Dickinson arrangement.

3. South African maps

a. 'South African Official Map 1:250,000, 1949' (BM, paras. 278-282)

162. As stated in NM, para. 312, this official South African map is the first modern map of
the area. At Kasikili Island the boundary is shown in the river between the two banks of the
southern channel. It was the principal map used by South Africa until Namibian
independence. It was reissued in 1967. A new 1:250,000 map was published in 1982, with the
same depiction of the boundary, that is still current and in use in Namibia. The Botswana

Memorial gives three grounds for disregarding the powerful evidence of this map. None can
be sustained.

163. First, Botswana claims, 'The evidence available indicates that this map reproduces,
without any further or independent verification, the boundary indicated on the War Office
14
map of 1933.' This statement is wrong in almost all aspects. The operational order for the
production of the map shows that the map was the result of a thorough new photographic and
15
ground survey and did not rely on detail compiled from earlier maps. NM, Annex 106,
which is a circular letter from the producer of the map to users, states, 'In the area covered by
Air Survey only information supplied by the field parties has been used.' The authoritiesquoted in the map, including the War Office 1933 map, were used only for '[a]dditional
locality and village names' around the periphery of the map.16 Moreover, the boundary line is

not 'copied' from the 1933 map, but is depicted differently from the line on the earlier map. It
appears between the two banks of the southern channel, rather than to the south of it as in the
1933 map, and it is shown by the symbol of an international rather than an intercolonial
boundary.

164. Second, BM, para. 280, states that the map was published in 1949, after the Trollope-

Redman interchange in 1948, and thus should be disregarded as post litam motu. This is also
incorrect. Although the official publication date was 1949, the map was drawn, as Botswana
acknowledges, 17 in 1945, three years before the Trollope-Redman correspondence. The 1945

preliminary version shows the boundary in the southern channel exactly like the published
map. 18 As discussed in NM, paras. 314-315, 329, sunprints of the map were widely
distributed to interested persons for comment. Among these were officials of the

Bechuanaland Protectorate. They returned the sunprints with numerous suggested changes,
including some in the area near Kasikili Island, but made no objection to the location of the
boundary. There is no reason to discount the evidence of this map because of the date of its
publication.

165. Third, Botswana asserts, 'The 1949 map is a compiled map and one of the authorities
19
listed is the War Office map of 1933.' As noted above, the War Office map was used only
for 'additional locality and village names' around the periphery of the map. The evidence of
the map itself shows that the boundary was independently depicted. But, as developed in CM,

Annex 1, para. 4a, the fact that a map is compiled does not necessarily derogate from its
accuracy or evidential value. Almost all maps are to some extent compiled, since some of the
information they show does not come directly from a field survey. In particular, international
boundaries are always added to the map on the basis of other source materials, including
applicable treaties and agreements, since the boundaries are very rarely marked on the ground
and therefore cannot be ascertained by a field survey.

b. 'South African Map Compiled by JARIC, 1:100,000 (c.1974)' (BM, para. 284)

166. The JARIC map, like the British D.O.S. 847, discussed in para. 160, supra, shows not
two channels in the relevant stretch of the river, but only one. There is no Kasikili Island on
the map. Therefore the depiction of the boundary on the map cannot be regarded as a choice
between the two channels. Like D.O.S. 847, this fundamental error irretrievably compromises
the reliability of the JARIC map for any purpose in this case.

167. The Botswana Memorial suggests that the JARIC map should be given special weight
because it was used by the SADF for patrolling. 20 Again the factual premise is incorrect, as it
21
appears from the Botswana Memorial itself. In describing the discussions between South
Africa and Botswana following the 1984 clash between the armed forces of the two countries

in the vicinity of the Island, Botswana states: 'At the meeting it was discovered that maps used
by the South Africans had showed Kasikili/Sedudu Island to be on the South-West Africa side
of the boundary . . .' Moreover, in their meetings at the United Nations, the Botswana

representatives stated that the maps that were carried by the South African forces in the areashowed the boundary in the southern channel. 23 Indeed, it is recorded that 'Mr. Mishra [the

UN Commissioner for Namibia] was shown the Botswana and South African maps which
differed on where the boundary should run.' 24 If that were so, the SADF could not have been

using the JARIC map for patrolling.

168. The provenance of this map is obscure. It would have been of little use to the SADF
because of its serious inaccuracies, among them the omission of the southern channel, an
important terrain feature that the SADF regularly patrolled.

4. Third country maps

169. As Namibia pointed out in its Memorial, 25 third country maps are of little value in cases

of this kind because they are almost invariably adaptations or copies of maps published by the
principal parties in interest. However, Botswana refers to two such maps, one from Zambia
and one from Zimbabwe, asserting that they show the boundary in the northern channel. 26

Both these maps commit the same error as D.O.S. 847, discussed in para. 160, supra, by
omitting the southern channel entirely. Thus, like D.O.S. 847, they cannot be taken to make
any statement whatever as between the two channels. The common error suggests that these
two maps simply copied the earlier map in this respect.

170. Without making a thorough search, Namibia has found two maps from these same
countries that place the boundary symbol in the southern channel, putting Kasikili Island in
Namibia.

• Rhodesia and Nyasaland: Livingstone 1:500,000 Director of Federal Surveys,
Salisbury, 1958 Sheet SE-35 NW 1st Edition Four colour. No disclaimer.

• Zambia South West 1:750,000 SG Ministry of Lands, Natural Resources and Tourism,
1980 Sheet 2 Coloured. Boundary disclaimer.

Namibia makes no claim on the basis of these maps, except that the practice of these
neighbouring states is not uniform and therefore cannot be cited against it. Undoubtedly these
countries followed normal cartographic practice and filled the blank space between the

boundary of their own country and the edge of the sheet with detail taken from maps
produced by the neighbouring country. 27

C. Conclusions as to the Map Evidence

171. The map evidence in this case is neither 'contradictory' nor 'confused,' despite the

effort of Botswana to make it so. The maps that are of a scale that is too small to show
Kasikili Island can provide no information on the issue before the Court, as Botswana admits,
so they cannot 'contradict' anything. Whether the boundary symbol on such maps appears to
the north or south of the river is an arbitrary cartographer's decision from which no inference
as to the position of the boundary around Kasikili Island can be drawn. These considerations

dispose of the contradictions asserted in BM, para. 296, and with respect to the British maps
mentioned in BM, para. 298. Whatever 'confusion' may exist on the matter has been
introduced by Botswana itself.172. The Botswana Memorial, para. 291, lists the four maps that, in its view, are large enough
in scale to represent accurately the configuration of the Island. It claims that 'all these maps
29
clearly show the northern and western channel as the main channel of the Chobe.' That
statement is flatly wrong. Three of the four do not show the northern channel as the boundary,
and the fourth was produced by Botswana, which is parti pris to the litigation.

• The northern channel on Bradshaw's map is a conjecture. His notes state that 'The
30
portions of the river not coloured I have not visited . . .The southern channel is
coloured and accurately depicted. The northern channel is white and marked by a

pecked course that simply joins the two ends that Bradshaw actually explored. Nor is
there any indication in terms of size or otherwise leading to the conclusion that either
channel is the main channel.
• As to the von Frankenberg map, the only basis for Botswana's assertion is its own
mistranslation of the German word Flussarm, as explained in para. 153, supra. Von
Frankenberg indicated his understanding that Kasikili Island was within his

jurisdiction and responsibility - that is within Namibia - by giving its name in German
on the map.
• The JARIC map shows only one channel. Its depiction of the area around the Island is
therefore fundamentally erroneous. In any case, it cannot be said to show the 'northern'
channel as the main channel.
• As to the 1974 Botswana map, its 'evidential significance . . . is removed' because it
31
was a unilateral act and 'self-serving' act.

173. These four maps, however, do not by any means exhaust the array of maps drawn to a

large enough scale to illuminate the question of the location of the border at Kasikili Island.
There is no threshold defining as a matter of law the scale of a map that can be used as
evidence by a tribunal in a boundary delimitation case. The Taba case, far from being an
authority for any such rigid rule, in fact reflects the true principle that the usefulness of the
scale is related to the ground distances involved in the dispute, in that case, of the order of a
32
few metres. The test is a practical one. It depends on the care and accuracy with which the
map is drawn, as determined by its provenance and by critical examination of the map itself, a
task for which the members of the Court are surely competent. In the Temple case, for

example, the key map was at a scale of 1:200,000, and the Court found no diffi33lty in using
it to define the boundary, although the critical distance was only 500 metres. And in the
Frontier Dispute case, the Court found two maps particularly useful, one at a scale of

1:500,000 and one at 1:200,000. The second of these, in a version reduced to 1:400,000 was
used to define the boundary determined by the Court, quoting distances of 0.1 kilometres and
locations to within 30 metres.34 In the present case, the critical distance is about two

kilometres, so that maps at much smaller scale than those cited above could be relevant.

174. Nor is there any warrant for the general statement: 'Except in very large scale maps the
35
drawing of a riverine boundary . . . inevitably produces inaccuracy and distortion.' Of
course, a map is a representation and like all representations does not exactly reproduce the
original. Questions of usefulness and accuracy must be resolved in the light of the functions
36
the particular map is intended to serve. One of the most important functions of official
maps issued by political authorities is to identify the political boundaries between separateterritories. In some cases the information depicted may be ambiguous and sometimes it may
even be erroneous, as we have seen in the case of D.O.S. 847 and the JARIC map, both of
which omit the Island altogether, showing only a single channel in the river. But there is no
basis for the suggestion that in general the boundaries on the maps put forward by Namibia
37
are unreliable. Again, the test of the reliability of the map and the weight the Court is
entitled to put upon it must be derived from a careful examination of the map itself and the
circumstances and history of its publication.

175. By this pragmatic test, examination of the maps put before the Court by the parties, far
from demonstrating contradictions and confusion, leads to a remarkably confident conclusion.
The most relevant maps - official maps produced and used by Germany, Great Britain, South
Africa and the United Nations during the period that they were respectively responsible
38
political authorities in the area - all are large enough in scale to show Kasikili Island and
the boundary around it, and there is both internal and external evidence that professional care
was exercised in the depiction of the boundary. They constitute evidence of an unbroken
sequence of maps emanating from all the political authorities in the area (with the exception
of Botswana after 1974) showing the boundary in the southern channel of the Chobe River
and placing Kasikili Island in Namibia.

Conclusion and Submissions of this Counter-Memorial

Conclusion

176. Consideration and analysis of the evidence and arguments advanced by Botswana show
that they are insufficient to defeat the conclusions established by Namibia's Memorial.
Kasikili Island belongs to Namibia by virtue of the Anglo-German Treaty of 1890 and by
virtue of long-continued use, occupation and exercise of jurisdiction without protest from
Botswana until 1984.

Submissions

177. In view of the facts and arguments set forth in Namibia's Memorial and this Counter-
Memorial,

May it please the Court, rejecting all claims and submissions to the contrary, to adjudge and

declare:

1. The channel that lies to the south of Kasikili/Sedudu Island is the main channel of the
Chobe River.
2. The channel that lies to the north of Kasikili/Sedudu Island is not the main channel of
the Chobe River.
3. Namibia and its predecessors have occupied and used Kasikili Island and exercised

sovereign jurisdiction over it, with the knowledge and acquiescence of Botswana and
its predecessors since at least 1890.
4. The boundary between Namibia and Botswana around Kasikili/Sedudu Island lies in
the centre of the southern channel of the Chobe River. 5. The legal status of Kasikili/Sedudu Island is that it is a part of the territory under the
sovereignty of Namibia.

______________________________

Albert Kawana
Agent of the Republic of Namibia
before the International Court of Justice

__________

1 See Botswana Memorial [hereinafter cited as 'BM'], p. 129. See also BM, Appendix to
Chapter VII, p. 5.

2 See Namibian Memorial [hereinafter cited as 'NM'], Vol. VI. All references to NM, Vol. VI
are cited hereinafter in the text as 'NM, Alexander Report.'

3 See BM, Appendix to Chapter VII.

4 See BM, paras. 116 and 129-144.

5 See BM, paras. 138-143.

6 BM, para. 35.

7 See BM, Chapter 1.

8 See NM, Annex 3, Art. I.

9 Id.

10 See NM, Alexander Report, Appendix, Sheet 25, Diagram 7, Image 25e.

11 NM, Annex 3, Art. 1.

12 See BM, para. 112.

13 NM, Annex 11.

14 See BM, pp. 5-10. For example:

para. 1(a) of the Terms of Reference requires the JTTE to 'determine where the boundary lies

in terms of the said Treaties';

para. 1(d) authorises the Team to 'carry out any act or function necessary and relevant for the
determination of the boundary . . . taking into account, and subject to, the provisions of the
Treaties referred to in (a) above'; andparas. 1(e) and 1(f) authorise the JTTE to examine relevant documents, maps, etc. and to hear
oral testimony 'without prejudice to the 1890 and 1892 Treaties .' (emphasis added)

Moreover, five of the seven preambular clauses refer to the boundary line as defined in the
Treaty.

15 NM, Annex 3, Art. 1.

16 See BM, para. 112.

17 See id.

18 R. Jennings and A. Watts, I Oppenheim's International Law (9th ed. 1992) p. 1280. In the
Anglo-Iranian Oil Co. Case (United Kingdom v. Iran), Judgment, I.C.J. Reports 1952, p. 105,
the Court accepted the principle that a legal text should be interpreted in such a way that a
reason and meaning can be given to every word in the text. The rule is an aspect of the
broader principle of 'efficacy' or 'effectiveness' in the interpretation of treaties. See also A.
McNair, The Law of Treaties (1961) pp. 385-386.

19 Case Concerning the Continental Shelf (Libyan Arab Jamahiriya/Malta), Judgment, I.C.J.
Reports 1985, p. 23.

20 Case Concerning Maritime Delimitation and Territorial Questions between Qatar and
Bahrain (Qatar v. Bahrain), Judgment, I.C.J. Reports 1995, p. 19.

21 The Corfu Channel Case, Judgment, I.C.J. Reports 1949, p. 4.

22 Id., p. 6.

23 To use Botswana's term. See BM, para. 112.

24 The Minquiers and Ecrehos Case (United Kingdom/France), I.C.J. Reports 1953, p. 49.

25 BM, p. 129.

26 See NM, paras. 121-128.

27 BM, Appendix to Chapter VII, p. 5.

28 See also Supp. Rep., para. 5.1: 'The basic premise in both Dr. Sefe's report and my two
reports is that the main channel of a river is the channel that carries the largest proportion of
the annual flow in the river.'

29 BM, para. 207.

30 See also NM, para. 131.

31 See NM, Alexander Report, paras. 1.6-1.8; NM, paras. 130-157.32 Because the Alexander Report was reviewed so extensively in the Namibian Memorial,
references in this Counter-Memorial are primarily to Professor Alexander's Supplementary
Report, Counter-Memorial of Namibia, Vol. III [cited in text as 'Supp. Rep.'].

33 BM, para. 26. See Supp. Rep., para. 5.11: 'the flow in the Chobe River at Kasikili Island

emanating from its own catchment area via the Linyandi River is sporadic and minimal.'

34 It is the height of the Mambova Rapids that controls the water gradient of the Chobe and
Zambezi rivers within the floodplain of the Zambezi, and thereby their carrying capacity, not
Victoria Falls, as asserted by Botswana in BM, para. 229 and Appendix to Chapter VII, p. 2.
See Supp. Rep., para. 5.6.

35 See BM, Chapter V(c), discussing, inter alia, the Eason report, the Trollope-Redman
inspection and the 1985 survey.

36 BM, Appendix to Chapter VII, p. 12.

37 See NM, para. 130. Table 2 of Dr. Sefe's report lists 'Data available for deciphering the

morphology of the Chobe River in the vicinity of Sidudu/Kasikili Island.' The table is placed
in the portion of the report arguing the proposition that there has been no change in the shape
of the channels, and the data listed in the table seem to relate to that proposition. See BM,
Appendix to Chapter VII, pp. 6-8.

38 BM, Appendix to Chapter VII, p. 4.

39 BM, para. 23.

40 Id.

41 BM, Appendix to Chapter VII, p. 2.

42 Id., p. 3.

43 Id.

44 BM, p. 129.

45 See para. 5, supra.

46 BM, p. 11 (section title).

47 See BM, Appendix to Chapter VII, p. 1.

48 See BM, para. 131.

49 NM, Annex 4, Art. III.

50 CM, Annex 20.

51 CM, Annex 19, para. 7.52 Map of Matabililand and the Adjoining Territories. 1:1,584,000. Intelligence Division War
Office. 1889 ID No.776 Uncorrected Proof No.2, NM, Atlas, Map II.

53 NM, Annex 31, p. 1269.

54 Id.

55 See NM, paras. 98-101.

56 See NM, para. 243.

57 See NM, Annex 71, para. 3.

58 The quotation is from Art. 1, Statute on the Régime of Navigable Waterways of
International Concern. See CM, Annex 17.

59 The phrase is from The Montello, 20 Wall. (87 U.S.) 430, 439 (1874), a decision of the
United States Supreme Court that has been very influential in the development of

international jurisprudence of navigability.

60 See BM, para. 207.

61 H.W. Halleck, International Law; or, Rules Regulating the Intercourse of States in Peace
and War (1861) p. 138 [hereinafter cited as 'Halleck, International Law']. See also J.B.
Moore, I A Digest of International Law (1906) p. 617 [hereinafter cited as 'Moore, Digest of

International Law'].

62 See CM, Annex 17.

63 BM, para. 35.

64 Id.

65 See BM, paras. 219-221.

66 BM, para. 218.

67 BM, para. 219.

68 See generally NM, Alexander Report, sec. 11.

69 BM, para. 36. See also BM, paras. 219, 220.

70 Representatives from both Botswana and South Africa met on 16 October 1972 and agreed

to develop a solution to the Kariba Weed problem. Professor Alexander was a member of
South Africa's delegation. See Supp. Rep., Appendix A. Following this meeting, Botswana
requested an additional meeting with South African officials to agree on a joint effort to
eradicate the Kariba Weed. See Supp. Rep., Appendix C.

71 See BM, para. 202.72 For further analysis, see Supp. Rep., sec. 12.

73 BM, para. 206.

74 In a diplomatic note dated 19 August 1996, Namibia informed Botswana that on 22 August

1996 the legal representatives planned to take a boat from Namibian territory and 'then circle
the disputed area and disembark on Namibian territory.' CM, Annex 30.

75 See BM, paras. 129-144.

76 See BM, paras. 118-128; NM, paras. 49-50.

77 See NM, para. 118; BM, para. 114.

78 BM, para. 115.

79 See NM, paras. 104-107. The Parties are agreed that the words 'the main channel of' were
introduced into Art III(2) of the Treaty in the final stage of the negotiations. See BM, para.

136; NM, para. 102 et seq. The insertion of these words was proposed by the British
Government, in English, and the German Government did not object. The existing German
text was then altered accordingly. The relevant part of the earlier draft text described the
border as following 'längs dem Tschobefluß' (along the Chobe River). This phrase with the
new British formula added then read 'und setzt sich dann im Thalweg des Hauptlaufes dieses
Flusses . . . fort.' This alteration of the previous German text shows two changes. The first is
that in view of the full English phrase 'centre of the main channel of that river,' the German

text had to insert a word to refer to the notion of 'centre' which the German text of 17 June
1890 did not do. This reference to the centre was expressed by the term Thalweg, which is
quite in line with general use of the term in other boundary treaties. A second change was
introduced into the German text by adding the words des Hauptlaufes to the term Thalweg -
quite unusual in German state practice. The only plausible reason for adding the words des
Hauptlaufes is that the German side, in order to follow the English-language text of the
British proposal as accurately as possible, came forward with a literal translation.

80 See BM, para. 141.

81 For example, the relevant passages in the 1890, 1895 and 1904 editions of W.E. Hall, A
Treatise on International Law (1884) repeat verbatim the passage in the original 1884 edition.
Similarly, the passage Botswana cites from the 1878 German edition of J.C. Bluntschli, Das

moderne Völkerrecht der civilisierten Staaten (1878) pp. 179-180, repeats the passage cited
from the 1874 French edition of J.C. Bluntschli, Le droit international codifié (2nd ed. 1874)
pp. 181-182 [hereinafter cited as 'Bluntschli, Le droit international'].

82 BM, para. 138.

83 See, e.g., P. Resch, Das Völkerrecht der heutigen Staatenwalt europäischer Gesittung

(2nd. Ed. 1890) pp. 70-71; A. Rivier, I Principes du droit des gens (1896) p. 168; L.
Oppenheim, 1 International Law - Peace (1905) p. 25.84 See, e.g., P. Fiore, Le droit international codifié (1890) p. 205; T.A. Walker, A Manual of
Public International Law (1895) Part II, Chap. III, p. 35; F. von Liszt, Das Völkerrecht
systematisch dargestellt (1898) p. 47.

85 See, e.g., R. Woltmann, Über das baurechtliche Verfahren bei der Verbesserung der

Flüsse. Ein Versuch zum deutschen Fluß-Rechte (1820) pp. 21-22; E. Nys, 'Rivières et fleuves
frontières. - La ligne médiane et le thalweg. Un aperçu historique,' III Revue de droit
international et de législation comparée 79-80 (1901) [hereinafter cited as 'Nys, "Rivières et
fleuves frontières"'].

86 See, e.g., Bluntschli, Le droit international, pp. 187-188; E.S. Creasy, First Platform of
International Law (1876) pp. 222-223.

87 See, e.g., T.Twiss, The Law of Nations Considered as Independent Political Communities
(1884) p. 249.

88 'Deuxième commission d'études. - Règlement organique pour la navigation des fleuves
internationaux,' IX Annuaire de l'institut de droit international 182 (1887-88). See BM, para.

139.

89 'Deuxième commission d'études. - Règlement organique pour la navigation des fleuves
internationaux,' IX Annuaire de l'institut de droit international173 (1887-88).

90 E.P. Engelhardt, Du régime conventionnel des fleuves internationaux (1879) p. 75
[hereinafter cited as 'Engelhardt, Régime conventionnel des fleuves'].

91 De Martens's general reflections on the state of the international law of river navigation are
apposite:

Ces précédents marquent les phases qu'a traversées, dans son développement, depuis 1815, le
droit international concernant la navigation fluviale. Il est impossible, en présence des faits

cités, de prétendre qu'à présent les principes de droit international sur cette matière soient
reconnus et pratiqués uniformément sur tous les fleuves internationaux. Au contraire, il faut
reconnaître qu'il existe jusqu'à ce jour une divergence assez essentielle parmi les hommes
d'État et les jurisconsultes les plus compétents, sur la portée de ces principes et le mode de
leur application. Il y a des fleuves à l'égard desquels ils n'ont reçu aucune application; il y en a
d'autres qui, étant juridiquement dans les mêmes conditions, sont néanmoins soumis à des
régimes essentiellement différents. (emphasis in original)

'Neuvième commission d'études. - Élaborer un projet de règlement organique pour la
navigation des fleuves internationaux,' VIII Annuaire de l'institut de droit international 279
(1885-1886).

92 This number does not include clauses in which the boundary is said to follow the bank of a
river.

93 See I. Brownlie, African Boundaries: A Legal and Diplomatic Encyclopaedia (1979)
[hereinafter cited as 'Brownlie, African Boundaries'].94 It is interesting that in these colonial African treaties, admittedly a random sample, the
word thalweg was first used in the treaty between the United Kingdom and France of 1899.
The term did not begin to be used with any frequency in Africa until after 1910.

95 See, e.g., W.E. Hall, International Law (1880) p. 104 ('Where [the boundary] follows a

river . . . their territories are separated by a line running down the middle, except where the
stream is navigable, in which case the centre of the deepest channel, or, as it is usually called,
the Thalweg, is taken as the boundary'). As mentioned in footnote 57, supra, the quotation
appears in all four subsequent editions as well.

96 See BM, para. 138.

97 Nys, 'Rivières et fleuves frontières,' pp. 79-80.

98 BM, para. 138.

99 See BM, para. 205.

100 Professor Alexander gives the scientific definition as 'a single continuous trace that
identifies the line of deepest water along the length of a river channel,' although he says that it
'has somewhat different legal interpretations when applied to boundaries along rivers.' NM,
Alexander Report, para. 2.11.

101 See para. 27, supra.

102 CM, Annex 12 ('die rinne, worinn das wasser flieszt, die tiefste linie der thalsohle, heiszt
thalweg').

103 Id. ('der stärkste und tiefste stromstrich in der bahn des Rheins').

104 CM, Annex 10 ('Stromstrich, Thalweg, die Linie, die Punkte größter

Oberflächengeschwindigkeit fließenden Wassers verbindet. Sie bewegt sich im allge-meinen
über der tiefsten Furche des Bettes, . . .').

105 CM, Annex 11 (Talweg: die für die Schiffahrt ständig instandgehaltene Stromrinne; auch
die Mitte der Schiffahrtsrinne, die Linie schnellster Oberflächenströmung, bei Fehlen
besonderer Abmachungen die Grenzlinie zwischen Ländern). Under a separate heading for
'Stromstritch' (stream-line), Der Große Herder gives the following definition: '(die) Linie

größter Oberflächengeschwindigkeit eines Flusses, liegt stets über dem Talweg und ist in
Kurven von der Flußmitte nach außen verlagert' ('the surface line formed by the part of the
river with the greatest surface velocity, running above the thalweg and moving with the curve
of the river').

106 E. Littré, 4 Dictionnaire de la langue française (1872) p. 2212.

107 P. Larousse, 15 Dictionnaire universel du XIXº siècle (1876) p. 42.

108 C. Calvo, 2 Dictionnaire de droit international public et privé (1885) p. 256.

109 G.F. de Martens, II Précis du droit des gens moderne de l'Europe (1864) pp. 135-136.110 A. Rivier, I Principes du droit des gens (1896) p. 169.

111 Id., p. 168.

112 F. Despagnet, Cours de droit international public (2nd ed. 1899) p. 407.

113 J.L. Klüber, Droit des gens moderne de l'Europe (1874) p. 189 [hereinafter cited as
'Klüber, Droit des gens moderne'].

114 See Bluntschli, Le droit international, p. 187, n.2.

115 Engelhardt, Régime conventionnel des fleuves, p. 73.

116 Id., p. 73, n.1.

117 In addition to the authorities cited in the text (all writing before or roughly
contemporaneously with the 1890 Treaty) a number of later authors continue to stress the
connection between the thalweg and the current of the river. See, e.g., V. Adami, National

Frontiers in Relation to International Law (1927) p. 17 ('the stream line of [the] fastest
current' (emphasis added)); C.G. Fenwick, International Law (1934) p. 275 ('strongest current
down-stream' (emphasis added)); A.S. Hershey, The Essentials of International Public Law
and Organization (1930) p. 270 ('the strongest current of the midchannel or navigable portion
of the stream' (emphasis added)); A. Möller, 1 International Law in Peace and War (1931) p.
104 ('the strongest current in the channel' (emphasis added)).

118 J. Westlake, International Law, Part I, Peace (1904) p. 141 [hereinafter cited as
'Westlake, International Law'].

119 P. Fiore, Le droit international codifié (1890) p. 205.

120 L.F. von Neumann, Grundriss des heutigen europäischen Völkerrechts (3rd ed. 1885) p.

45.

121 P. Orban, Étude de droit fluvial international (1896) p. 343.

122 H. Bonfils, Manuel de droit international public (droit des gens) (1894) p. 257
[hereinafter cited as 'Bonfils, Manuel de droit international'].

123 A. Chrétien, Principes de droit international public (1893) p. 108 (internal citation
omitted) [hereinafter cited as 'Chrétien, Principes de droit international'].

124 BM, Annex 17.

125 NM, Annex 47.

126 See, e.g., C.G. Fenwick, International Law (1934) p. 275. It is commonly agreed that the
modern development of the thalweg concept stems from the Treaty of Lunéville, of 9
February 1801, in relation to the Rhine River boundary between France and the German
Empire, although the principle had been agreed upon earlier at the Congress of Rastadt. See
Engelhardt, Régime conventionnel des fleuves, p. 73; Traité de paix entre la républiquefrançaise et Sa Majesté l'empereur le corps germanique, signé à Luneville, le 9. Février 1801,
in G.F. de Martens, VII Recueil des traités (1829) p. 298. Article Six of the treaty provided
that 'le Thalweg du Rhin soit désormais la limite entre la république française et l'empire
germanique . . .' Id. The formula with some variations was extended to other parts of the
Rhine by subsequent treaties, but always emphasising the aspect of the access to the

downstream current or 'flowing stream' of the river. See, e.g., Art. 1, Traité entre le Grand-
Duché de Bade et le Canton Helvetique d'Aargovie, signé à Aarau le 17 Sept. 1808, in G.F. de
Martens, I Nouveau recueil des traités de l'Europe (1817) p. 140 [hereinafter cited as 'de
Martens, I Nouveau recueil des traités']. Article 9 of the Treaty on the Delimitation between
France and the Grand Duchy of Baden, 30 January 1827, provided that 'the thalweg of the
Rhine is the path most suitable for shipping during normal low water.' Art. IX, Traité pour
régler les limites entre la France et le grand duché de Bâde, signé à Strasbourg le 30 Janv.

1827 (Extrait), in G.F. de Martens, VII Nouveau recueil des traités de l'Europe (1829) p. 126.
The boundary between Prussia and Westphalia was the thalweg of the Elbe, 'c'est à dire le
principal courant de l'Elbe . . .' Art. I, Convention entre S.M. le Roi de Prusse et le Roi de
Westphalie sur les limites et les droits de navigation; signée à Berlin le 14 Mai 1811, in de
Martens, I Nouveau recueil des traités, p. 383.

127 Westlake, International Law, p. 141. He remarks in a note that 'Thal in the sense of valley

enters into thalweg only indirectly. The immediate origin of the word lies in the use of berg
and thal to express the upward and downward directions on a stream, like amont and aval in
French.' Id., n.1.

128 C.C. Hyde, 'Notes on Rivers as Boundaries,' 6 Am. J. Int'l L. 902-903 (1912) [internal
citation omitted] [hereinafter cited as 'Hyde, "Notes on Rivers"']. The passage is included in

C.C. Hyde, I International Law, Chiefly as Interpreted and Applied by the United States (2nd
rev. ed. 1947) pp. 444-445 [hereinafter cited as 'Hyde, International Law'].

129 Engelhardt, Régime conventionnel des fleuves, pp. 73, n.1, 74.

130 Klüber, Droit des gens moderne, pp. 188-89.

131 Bonfils, Manuel de droit international, p. 281.

132 C.G. de Koch et F. Sch_ll, V Histoire abrégée des traités de paix (5th ed. 1817) p. 362.

133 G. Kaeckenbeeck, International Rivers (1918) p. 176.

134 Again, the reference to downstream navigation continues to be found in the writings of
more recent authors. See, e.g., P.G. de Lapradelle, La frontière (1928) p. 203; K. Strupp,
Grundzüge des positiven Völkerrechts (1921) p. 69; L.J. Bouchez, 'The Fixing of Boundaries
in International Boundary Rivers,' 12 Int'l & Comp. L.Q. 793-95 (1963).

135 See, e.g., T.A. Walker, A Manual of Public International Law (1895) p. 35; Halleck,
International Law, p. 138; Chrétien, Principes de droit international, p. 108.

136 See, e.g., E.S. Creasy, First Platform of International Law (1876) p. 223; E. Nys, I Le
droit international: les principes, les théories, les faits (1912) pp. 469-70 [hereinafter cited as
'Nys, Le droit international']; T. Twiss, I Le droit des gens ou des nations (1887) p. 231.137 Nys, 'Rivières et fleuves frontières,' p. 80.

138 Id. See also Nys, Le droit international, pp. 469-70.

139 See, e.g., H. Wheaton, Elements of International Law (1836) p. 150 ('Where a navigable

river forms the boundary of conterminous states, the middle of the channel, or Thalweg, is
generally taken as the line of separation between the two states, the presumption of law being
that the right of navigation is common to both; . . .'), cited with approval in Halleck,
International Law, p. 138 ('Where a navigable river forms the boundary of coterminous states,
the middle of the channel, - the filum aquae, - or thalweg, is generally taken at the line of their
separation, the presumption of law being, that the right of navigation is common to them
both.'); Engelhardt, Régime conventionnel des fleuves, p. 73 ('[Q]u'en général la ligne médiane

du thalweg est plus stable que la ligne médiane du fleuve, et qu'en suivant le vrai canal, c'est-
à-dire la partie la plus navigable, elle semble mieux répondre à l'idée fondamentale que
chaque riverain a un endroit égal à l'usage utile du fleuve commun'). (internal citation
omitted)

140 Halleck, International Law, p. 138, cited with approval in Moore, Digest of International

Law, p. 617.

141 Hyde, 'Notes on Rivers as Boundaries,' p. 903. The passage is included in Hyde,
International Law, pp. 444-445.

142 New Jersey v. Delaware, 291 U.S. 361 (1934).

143 Id., pp. 380, 383.

144 At certain stages of the Zambezi flood some water comes from the two anabranched
channels to the north of Kasikili Island into the final short leg of the northern channel east of
Kasika and just before it rejoins the main channel at the confluence to the west of the Island.
However, there is no flow throughout the whole length of the northern channel, beginning at
the bifurcation to the west of the Island and proceeding through the channel to the confluence

to the east.

145 The dictionary definition of 'channel' is 'A channel of running water.' 'The hollow bed of
running water.' J.A. Simpson and E.S.C. Weiner (eds.), III Oxford English Dictionary (2nd
ed. 1989) p. 19. Professor Alexander defines a channel as 'a surface conduit capable of
conveying river water . . .' NM, Alexander Report, para. 2.1. (emphasis added)

146 Engelhardt, Régime conventionnel des fleuves, p. 73 (internal citation omitted).

147 See NM, Part Two.

148 See BM, Chapters III and IV.

149 See BM, para. 67.

150 BM, Annex 13, para. 1.

151 BM, para. 75.152 See NM, para. 202.

153 BM, para. 32. (internal citations omitted)

154 See BM, Annex 37.

155 BM, Annexes 32, 33. See also NM, Annex 100 ('[T]he boundary [of the Park] runs . . .
along the northern boundary of the Bechuanaland Protectorate to its point of intersection with
the eastern boundary of the Batawana Reserve . . .').

156 See NM, paras. 307-308; NM, Atlas, Map XI; and the extract in NM, Fig. 14, following
p. 126.

157 CM, Annex 23. Although the men were released, their weapons were not returned.

158 CM, Annex 24.

159 CM, Annex 26.

160 See CM, Annex 27.

161 This was the renamed Department of Native Affairs.

162 CM, Annex 25.

163 BM, Annex 17.

164 BM, Annex 15; NM, Annex 47.

165 BM, Annex 17.

166 BM, Annex 15, para. 2.

167 In relation to Kakumba Island, at the confluence of the Chobe and Zambezi some 14
kilometres east of Kasikili, Eason takes exactly this line. Brushing aside the criteria of width
and depth of the two channels, he chooses the southern channel because 'the most important
point is that during the greater part of the year the current flows to the West along the

Northern channel and to the East along the Southern .' NM, Annex 47. (emphasis added)

168 BM, Annex 15, para. 2.

169 NM, Annex 60, para. 4. The factual basis for this conclusion was not stated. The report
states that the finding was made 'after separate examination of the terrain and the examination
of an aerial photograph . . .' Id., para. 3. What is meant by 'separate' is not clear. Although the

report is dated 19 January 1948, the date of these 'examinations' is not given. Nor does the
report state any criterion for determining which was the main channel. In any case, as the
Alexander Report demonstrates, the finding was scientifically incorrect. See NM, Alexander
Report, sec. 12.

170 NM, Annex 60, para. 8.171 NM, Annex 71, paras. 7(a), 7(c). At one point, the Bechuanaland Protectorate authorities
had issued instructions that the Masubia of Caprivi should in the future be allowed to cultivate
the Island under annual permit. See NM, para. 276. The Namibian Memorial speculated that
the instruction was never implemented. This is confirmed in an undated letter made available
from the Botswana National Archives, which recommends 'giv[ing] way over the issue of a

permit . . .' CM, Annex 21, para. 7. The recommendation was approved in a hand-written note
bearing the dates 14/11/51 and 19/11/51, but the month must be incorrect, since the
arrangement was concluded in September. See CM, Annex 22.

172 See NM, Annex 71, para. 7(c); NM, Annex 73, para. 3.

173 BM, Annex 28, para. 3(b).

174 Id., para. 4.

175 BM, para. 160.

176 BM, Annex 36.

177 BM, para. 182.

178 See G.A. Res. 2145 (XXI), 27 Oct. 1966, Official Records of the General Assembly,
Twenty-first Session, Supplement No. 16, doc. A/6316.

179 S.C. Res. 276, 30 Jan. 1970 (1529th meeting).

180 Legal Consequences for States of the Continued Presence of South Africa in Namibia
(South West Africa) Notwithstanding Security Council Resolution 276 (1970), Advisory
Opinion, I.C.J. Reports 1971, p. 58.

181 See G.A. Res. 3031 (XXVII), 19 Jan. 1973, Official Records of the General Assembly,

Twenty-seventh Session, doc. A/RES/3031. In its capacity as the entity in charge of Namibia's
international relations, the UN Council for Namibia entered into bilateral agreements with
several countries regarding the recognition of travel documents for Namibians. In addition,
the Council also acceded to multilateral conventions on behalf of Namibia and represented
Namibia at international conferences and in international organisations. See CM, Annex 31.

182 S.C. Res. 276, 30 Jan. 1970 (1529th meeting).

183 The Court made an exception for certain multilateral treaties, such as those of a
humanitarian character the non-performance of which could adversely affect the people of
Namibia. The exception is without application in the present context. Legal Consequences for
States of the Continued Presence of South Africa in Namibia (South West Africa)
Notwithstanding Security Council Resolution 276 (1970), Advisory Opinion, I.C.J. Reports
1971, p. 55.

184 S.C. Res. 301, 20 Oct. 1971 (1598th meeting).

185 See Charter of the United Nations and Statute of the International Court of Justice,
Chapter XIV, Art. 92, p. 48.186 See BM, paras. 167-169.

187 BM, Annex 42, para. 1.

188 Id., para. 2.

189 See BM, Annex 41, para. 3.

190 BM, Annex 43.

191 See id.

192 BM, para. 171.

193 CM, Annex 18.

194 Id.

195 See International Status of South-West Africa, Advisory Opinion, I.C.J. Reports 1950, p.
137.

196 See G.A. Res. 2145 (XXI), 27 Oct. 1966, Official Records of the General Assembly,
Twenty-first Session, Supplement No. 16, doc. A/6316.

197 BM, Annex 48, p. 3.

198 BM, para. 193.

199 See BM, para. 185.

200 BM, para. 184.

201 BM, Annex 44, para. 11(ii).

202 The Botswana group included G.G. Garebamono, Secretary for External Affairs, but he
was a civil servant, not the political head of the Ministry. In the authorities cited by Botswana
in BM, para. 183, on the creation of informal international agreements, officials of the highest

rank were involved. In the Nuclear Tests Case, the declaration was made by the President of
the French Republic. Nuclear Tests Case (Australia v. France), Judgment, I.C.J. Reports
1974, p. 266. In Legal Status of Eastern Greenland, the critical declaration was made by
Norway's Minister for Foreign Affairs 'on behalf of his Government in response to a request
by the diplomatic representative of a foreign Power, in regard to a question falling within his
province . . .' Legal Status of Eastern Greenland, Judgment, P.C.I.J. 1933, p. 71. Botswana's
statement that the 'delegations both included a senior legal component' is also something of an

exaggeration with regard to South Africa. Although Attorney General M.D. Mokama led the
Botswana delegation, the only lawyer listed for South Africa is 'Mr. Rautenbach (Legal
Department).'

203 NM, Annex 84, para. 9(a).204 Again, Botswana's Memorial is not very explicit on this point. The furthest it goes is to
say that 'neither Botswana nor South Africa sought to challenge the validity of the Joint
Survey Report.' BM, para. 193. See also BM, paras. 194, 199.

205 BM, Annex 50.

206 BM, Annex 51, para. 18.

207 See BM, Annex 52. Although the telex stated that Botswana had occupied the Island,
there is no evidence that Botswana troops or officials were deployed there until 1991, after
Namibia's independence.

208 BM, Annex 51, para. 17.

209 Id., para. 19.

210 CM, Annex 28, para. 5.

211 NM, Annex 86.

212 NM, Annex 88.

213 See BM, para. 196.

214 BM, Annex 48, p. 3.

215 The survey was delayed from December 1984, when the decision was taken to conduct it
'as a matter of urgency,' until July. This delay presumably occurred because between
December 1984 and July 1985 the Zambezi floodplain, including the Island and both
channels, was under water.

216 See NM, Part Two, Chapter V.

217 See BM, paras. 254-256, 262.

218 Brownlie, African Boundaries, p. 5. See also, e.g., T.S. Murty, 'Boundaries and Maps' 4
Indian J. Int'l L. 367 et seq. (1964).

219 Brownlie, African Boundaries, p. 16.

220 I. Brownlie, 'International Law at the Fiftieth Anniversary of the United Nations: General
Course on Public International Law,' in Academy of International Law (ed.), Recueil des
cours (1995) p. 161. For the convenience of the Court, Namibia provided relevant copies of
cited pages of this work in NM, Annex 116.

221 Id. (internal citation omitted)

222 See, e.g., Case Concerning the Frontier Dispute (Burkina Faso/Republic of Mali),
Judgment, I.C.J. Reports 1986, pp. 582-83.223 Botswana states that 'a map of smaller scale than 1:100,000 . . . is likely to be of little use
in representing accurately the configuration and size of the channels and island.' BM, para.
290. To support its point, Botswana cites the Taba Award (Egypt v. Israel) (1988). See BM,
para. 293. But as appears from the portion of the Taba opinion quoted by Botswana, the issues
in that case concerned points separated by a few metres where maps of 1:100,000 were clearly

too small scale. The dimensions involved in this case are of the order of two kilometres.
While it is obvious that the larger the scale of a map the more geographic detail it can provide,
it is equally obvious that many of the maps produced in the present case by Botswana and
most of the maps produced by Namibia at a smaller scale than 1:100,000 are large enough to
provide a clear and unambiguous representation of the boundary. It should be noted that
despite Botswana's insistence on the importance of the scale of a map, for the majority of the
maps cited in its Memorial and illustrated in BM, Appendix II, Dossier of Maps, no scale is

given.

224 See BM, paras. 269-273, 276.

225 BM, para. 273. (internal citation omitted)

226 See CM, Annex 1, pp. 18-19, para. 15.

227 See CM, Annex 1, pp. 9-10, para. 9. See also NM, Annex 110, where the DOS Record
Officer states with specific reference to Botswana's boundaries on its maps, '[t]he
compilations at present show several boundaries along one side or the other of a single stream
when the map material does so, but it seems clear from a study of the available sources that it
is the centre of the stream which is meant in such a case . . .'

228 NM, Atlas, Map VIII.

229 The only unofficial maps that have been called to the attention of the Court by either
party are the early explorers' maps, produced before the conclusion of the Treaty and
therefore of little bearing on the case, except as they may show the contemporary knowledge
of the topography and geographical features of the area that was available to the negotiators.

See NM, paras. 71-76 ; NM, Atlas, Maps I/1, I/2 and I/3.

230 BM, para. 257.

231 See NM, p. 124, n.285. However, the Kriegskarte is treated at some length in NM, Annex
102, para. 8.

232 The Andara sheet, which is printed on the same page as the Linjanti sheet, does not show
the area of the Chobe River at all, as might be inferred from its name.

233 See NM, Annex 111, p. i.

234 NM, Atlas, Map I/3.

235 BM, para. 266.

236 See NM, Alexander Report, para. 10.11. See also NM, Alexander Report, Sheet 25,
Diagram 7.237 See NM, Annex 141, pp. 60-61.

238 Isle of Palmas, II R.I.A.A. 831, 852 (1928), cited in BM, para. 262.

239 BM, para. 268.

240 Id.

241 See CM, Annex 29.

242 See BM, Appendix II, Map 7 (lower right hand corner).

243 The translation in the text was made by Prof. Dr. Jost Delbrück, and a certificate as to its
accuracy has been deposited with the Court. The note was translated by the British War
Office in 1923 as follows:

The Zambezi and the Limianty [sic] form political frontiers. The northern arm of the river is
broader than the southern at the island of Mangarde [sic], but it is shallower, and is therefore

generally dried up during the dry period.

The translation is attached to correspondence between the GSGS and the Surveyor General at
Windhoek requesting a copy of the map and covering the transmittal of blueprints. All of the
notations on the von Frankenberg map were translated. See CM, Annex 1, pp. 17-18, para.
14d.

244 See, e.g., the authoritative German dictionaries, J. Grimm and W. Grimm, Deutsches
Wörterbuch (1862), (1984), (1855), which offered the following definition: 'Flusz, . . . . der
Flusz teilt sich in drei Arme; . . .' ('River, . . . . The river divides into three branches; . . .'); R.
Klappenbach and W. Steinitz (eds.), Wörterbuch der deutschen Gegenwartssprache,
hersausgegeben (1977) p. 5]: 'Fluß, der; . . . . der F. nimmt Nebenflüsse auf , teilt sich in
mehrere Arme , verzweigt sich, . . .' ('The river; . . . . the river takes in affluents, divides into

several branches, branches out, . . .'); and G. Kampcke (ed.), Handwörterbuch der deutschen
Gegenwartssprache, herausgegeben von einem Autorenkollektiv unter (1984)]: 'fluß/Fluß . . . .
-arm, der, abzweigender Teil eines Flusses; . . .' ('river . . . . -branch, the branching out section
of a river; . . .')]. See also, e.g., the authoritative German hydrology text, J. Mangelsdorf and
K. Scheurmann, Flußmorphologie (1980) p. 192 et seq., where the authors use the word Arme
when they describe the branching out of various rivers and their tributaries (Haupt- und
Nebenflüsse) such as the Po River in Italy dividing into several branches ('in mehrere Arme

aufgespalten . . .'). For the convenience of the Court, Namibia has provided relevant copies of
all cited pages from these works as Annexes 12, 14, 15 and 16 to this Counter-Memorial.

245 The one exception is an 'Engl. stat.' at Kazungula on the east bank of the Zambezi near
the confluence of the two rivers.

246 See NM, paras. 296-300.

247 See NM, Part Two, Chapter IV(B).

248 BM, p. 117, n.8.249 See NM, paras. 298-299.

250 See, e.g., BM, para. 291.

251 And its derivative Bechuanaland Protectorate 1:1,250,000 Bechuanaland Survey Dept.

1935.

252 For further analysis of the map, see CM, Annex 1, p. 19, para. 16.

253 CM, Annex 8, p. 71.

254 BM, para. 275.

255 BM, para. 275(a).

0 See Case Concerning the Frontier Dispute (Burkina Faso/Republic of Mali), Judgment,
I.C.J. Reports 1986, p. 566. To illustrate its point, the Court cited South America and French
West Africa as examples where intracolonial boundaries became international boundaries

upon independence. See id.

1 BM, para. 275(b).

2 See CM, Annex 1, p. 9, para. 8.

3 See CM, Annex 8, pp. 77-78.

4 BM, para. 275(c).

5 Bechuanaland Protectorate 1:1,250,000 Bechuanaland Survey Dept. 1935, a reduction of 2.5
times of GS 3915. See NM, Annex 102, p. 14, para. 24.

6 See CM, Annex 1, p. 21, para. 16e; NM, Annex 102, p. 16, para. 27.

7 See CM, Annex 1, p. 21, para. 16e; NM, Annex 102, p. 16, para. 28.

8 BM, para. 275(d).

9 See CM, Annex 1, p. 20, para. 16c.

10 See NM, Annex 102, p. 14, para. 24.

11 See NM, Annex 102, p. 13, para. 23.

12 BM, para. 277.

13 See CM, Annex 1, p. 23, para. 17.

14 BM, para. 279.

15 See NM, Annex 105.16 NM, Annex 106, para. 1.

17 See BM, para. 278.

18 See NM, Atlas, Maps X/2, X/1.

19 BM, para. 282.

20 See BM, para. 300.

21 See BM, para. 174.

22 Id.

23 See BM, Annexes 41, 42.

24 BM, Annex 43.

25 See NM, para. 324.

26 See BM, paras. 286, 287.

27 See CM, Annex 1, p. 8, para. 7b.

28 BM, para. 297.

29 BM, para. 292.

30 NM, Annex 115, p. 209.

31 BM, para. 280.

32 See Taba Award (Egypt v. Israel) (1988) 80 I.L.R., p. 287 ('The Tribunal does not consider
these map-based indications to be conclusive since the scale of the map . . . is too small to
demonstrate a location on the ground as exactly as required in these instances where the
distances between disputed pillar locations are sometimes only of a few metres').

33 See Temple of Preah Vihear (Cambodia v. Thailand), Judgment, I.C.J. Reports 1962, pp.
26, 33-34.

34 See Case Concerning the Frontier Dispute (Burkina Faso/Republic of Mali), Judgment,
I.C.J. Reports 1986, pp. 584, 648, 649.

35 BM, para. 294.

36 In some of the maps discussed here, Kasikili Island is especially emphasised or made to
appear somewhat larger than it is so as to portray it and the boundary in relation to it more
clearly. See, e.g., Namibia 1:1,000,000 United Nations 1985 UN Map No. 3158, NM, Atlas,
Map XV/4; Africa 1:2,000,000 War Office GSGS 2871 Sheet Rhodesia, Fig. 4, following p.
75.37 See BM, Chapter VIII(g)(iii).

38 The 24 relevant maps are listed in CM, Annex 1, pp. 33-34, para. 23.