INTERNATIONALCOURTOFJUSTICE
CASE CONCERNINGKASIKILIISEDUDUISLAND
(BOTSWANAmAMIBIA)
REPLY
OF
THE REPUBLICOF NAMIBIA
VOLUME1
REPLY
27 NOVEMBER1998 INTERNATIONALCOURTOF.JUSTICE
CASECONCERNINGKASIKILIISEDUDU ISLAND
(BOTSWANAINAMIBIA)
REPLY
OF
THEREPUBLIC OF NAMIBIA
CONTENTS
Volume1 REPLY
VolumeII EXPERTREPORTS
Second SupplementaryReportof ProfessorW.J.R.Alexander
GeomorphologicalStudyby Professor KeithS.Richards
VolumeIII ANNEXES
Map-Related Annexes,Legal InstrumentsandAuthorities,
DiplomaticCorrespondence andficia1Materials,
MaterialsRelated tothe ScientificCaseandMiscellaneous Page
PARTONE: INTRODUCTION .................................................1.............
Chapter1 . INTRODUCTIONANDSUMMARY ........................1......
A . PreliminarvRemarks .............................................1...........
B. The Text of theTreaty ...........................................-6..........
1.The languageof ArticleIII(2)..............................6........
2. The questionof accessto the navigational
resourcesof the Chobe River ..............................8........
C .The SubsequentPracticeofthe PartiesEstablishinnTheir
Aaeement asto the Treaty'sIntemretation ............................
1.Themeaningof 'subsequentpractice'under
Article31(3)(b)ofthe ViennaConventiononthe
Lawof Treaties ..........................................11...........
2.Botswana's attemptsto negatethe 'subsequent
practice'ofthe parties....................................13.........
D . The EvidenceofMaps asa Guideto the Proper
IntemretationofArticleIII(2)ofthe 1890Treatv ..................15...
E. The ScientificEvidenceandthe Intemretationof the
1890Treatv .....................................................19.............
1. The legalstandardapplicableto the determination
of the 'centreofthe main channel.........................19......
2. Namibia'scase onthe identificationof the 'main
channel'of the ChobeRiveratKasikiliIsland ..................
3. Responseto Botswana's 'scientificcase'...................26....
(a) The sixpropositions..............................26.......
(b) Secondaryhydrologicalarguments .................30...
(i) Bedslope..................................31........
..
(il) Erosion...................................31........
(iii)Flow measurements ........................31....
...........
(c) The aerialphotographsandsatellitesimages 32
4. Identimg the 'centre'ofthe main channel ................33...
F. PrescriptionandAcquiescence ...................................36........
E. Conclusion .....................................................-37.............PARTTWO: THEINTERPRETATION OFTHE ANGLO-GERMAN
TREATY OF1890 ...............................................39..............
ChapterII . THETEXT OF THETREATYIN THE LIGHT
OF ITSOBJECTAND PURPOSE .........................39.......
......................................................................
A. In General -39
B. Interpretationof the Words ofthe Treaty.................................
C. Interpretationofthe 1890Treatyin ~i&t ofits Obiect
andPuniose ...................................................46................
D .The A1leg;edSelf-Executing;Characterofthe Treatv.................2..
E. Conclusion ....................................................54................
Chapter III. INTERPREATATIONOFARTICLE III(2) OF
THE 1890TREATY: REBUTTALTOBOTSWANA'S
SCIENTIFICEVIDENCE .................................56.........
A .The Proportionofthe Flowof the RiverPassinp;through
theNorthem Channel ..........................................58............
1. The ChobeRiveris not aperennialriver ..................59.....
2. The data asto waterlevelsat gaugingstations
alongthe ChobeRiver ..................................64..........
a. Thecomparisonofthe readingsat the seven
stations........................................64............
b. The gaugereadingsattheKasanestation ...........66..
3. Otherassertedindicatorsof greaterflowin the
northem channel .......................................68............
a. The 'capacity'of the channe.....................68.....
b. Slope...........................................69............
c. The sedimentologicalstudy.......................70......
d. Assertedflowmeasurements ........................4....
B. AerialPhotographvand SatelliteImages ........................-73.....
1.Generalcornments .....................................7.4..........
2 .Botswana'sphotographiccornparisons ....................76...
(a)Thecomparativewidthofthe two channels .........77.
(b) Thecomparativedepthofthe two channels ..........8
(c)The sedimentbars atthe openingofthe
main channel....................................-79........
(d)Otherindiciaof the presenceof absenceof
................................
erosionin the twochannels 80 (i)As to the stabilityofthe
configurationofthe northernchannel. ........81
(ii)As to the absenceof erosionin the
southernchannel.. ..........................8.1
(e) Striationpatterns..................................83....
(f)The so-calledsub-channeldrainingthe
southernportionofthe Island.. ......................83.
3. Discussionofparticular aerialphotographs.. ...............85
4. Conclusionsasto aerialphotogràphs.. .....................8.9
D. Determinationofthe 'BoundarvbetweenNarnibiaand
Botswana aroundKasikiliIsland'withinthe Meaning
ofArticleIII(2) ofthe 1890Treatv. ...................................
CHAPTER IV: 'SUBSEQUENT PRACTICE INTHE
APPLICATION OFTHE TREATY'WITHIN
THEMEANING OFARTICLE 31(3)@)OF
THE VIENNACONVENTION. ........................-97...
A. Botswana'sArments asto theMeaninnof Article
3 1(3)(b)..........................................................97.........
1.In general..................................................97......
2. The correctinterpretationof Article31(3)@)of
the Vienna Conventionofthe Lawof Treaties.. ..............98
B. Botswana's FactualArguments on SubsequentConduct.. .........IO2
1. Botswana's contentionthatuntil 1948theparties
to the Anglo-GermanTreatvwereof the opinion
that the northernchannelwas the main channel
andthat KasikiliIslandwasuart ofthe
BechuanalandProtectorate. ..............................10.3..
(a) LordHarcourt's Despatchof 14 January
1911 .............................................IO3...
(b)Eason'sReport.. .................................IO5...
(c) The allegedpermissionto cultivateon
the Islandgrantedin 1924.. .......................IO7.
(d) Conclusion.........................................O9..
2. Botswana's contentionthatthe exchangesthat
culminatedin the 1951Trolloue-Dickinson
arrangementconfirmedthat theparties reconnized
the northern channel astheboundarvandKasikili
Island aspart of theBechuanalandProtectorate.. ...........-110 (a)Botswana's allegationthat SouthAfnca
challengedthestatusquo and soughtto
modifLtheboundary.. .............................Il0..
(b) Botswana's allegationthatBritain,with
SouthAfiica'sacquiescence,exercised
jurisdiction overKasikiliIsland.....................Il3
(c) The allegationthat bothpartiesrecognized
the northemchannelasthe main channel
forthe purposeofthe AngloGerman
Agreementof 1890.. ...............................115..
(d) Conclusion.........................................116....
3. Botswana's Contentionthatthe 1984186discussions
betweenBotswanaand SouthAfnca resulted inan
a~eement betweenthevartiesreg -ardingthe avvlication
ofthe Annlo-Geman Treatvof 1890that resolved the
present boundarvdisvutein favourof Botswana.. .........117
Conclusion. ...........................................120.....
CHAPTER V: THE MAP EVIDENCE RELATINGTO
THE INTEPRETATIONOFARTICLEIII(2)
OF THE 1890TREATY.. ..................................22.
A. GeneralLegalandTheoreticalConsiderationsasto the
MavEvidencein this Case.. .....................................-123....
1.Therelevanceand weight ofmap.evidencein
boundary determinations. ...................................23.
2. Namibia's alleged 'quandary' .............................126
3.Botswana's four generalpropositions.. .....................127.
B. Cornmentson SvecificMaps. .....................................31..
C. Botswana's GeneralConclusionsonMavEvidence.. ..............137
D. Conclusions. ....................................................-141........
PARTTHREE: PRESCRIPTIONAND ACQUIESCENCEAS AN
INDEPENDENTTITLE OFSOVEREIGNTY.. .................142
CHAPTERVI: PRESCRIPTION ANDACQUIESCENCE.. ..............142
A. Botswana's Fallacies ...............................................42....
B. Namibia HasMet .h. ReauirementsforEstablishing
Titlebv Prescnvtion.. .............................................1...
1. Generalconsiderations. .....................................4.6
2. Botswana'sargument onthe 'critical date'. ..................146 3. Legalaspectsofthe doctrineofprescription.. ................147
(a) The facts:peacefuloccupation and use of
the Islandby Namibia.. ................................148
(b) The occupationand use ofKasikiliIsland
was inthe exerciseof stateauthorityand
'a titre desouverain'. ..................................152
(i) The occupationanduse of Kasikili
Islanddiscussedabovewereunderthe
authorityand jurisdictionof the powers
that exercisedsovereignrule overthe
Caprivi Stripatthe time.. .....................153
(ii)The occupation anduse of Kasikili
Islandby Namibia'spredecessorsin
titlewas 'atitre de souverain7.. ....... ......156
C. Acquiescence. ..................................................... ....158
D. Conclusion.. .............................................................162
PARTFOUR: OTHER MATTERS.. ..................................................165
CHAPTERVII: NAMIBIANFOREIGN POLICYAND IT'S
APPLICATIONTO THE BOUNDARY DISPUTE
BETWEENNAMIBIAANDBOTSWANA. ................165
A. Introduction. .............................................................165
B. NamibianForeim Policv. ...............................................166
C . NamibianForeim PolicvAfterIndaendence In 1990. ..............168
1. DiplomaticRelations. ...........................................168
2. Membershipin Afican andregionalorganisations.. .........168
3. Membershipin globalinternationalorganisations.. ...........169
4. Namibia'scornmitmentto internationalpeaceand
securityof al1nations.. ...........................................169
D. The RelationshivBetween NamibiaandBotswanaAfter
Namibia7sIndependence. ............................................ ..169
1. The OriginsofNamibia'sforeignpolicytoward
Botswana.. ........................:...............................169
2. First statevisit andthe signingof cooperation
agreements. .......................................... ............170
3. TheKasikiliISeduduIslanddisputeandNamibia's
conductin accordancewith theprotocolof
understandingon defenceand security ........................-171
E. Conclusion.... ......................................... .... .. ....... .176 CHAPTERVIII: MISSTATEMENTS. LEGAL FALLACIES
AND MISREPRESENTATIONSIN THE
BOTSWANACOUNTER.MEMORIAL ................178..
A .Misstatementsof Fact .............................................179.........
B Mani~ulationsof Evidence ........................................183........
C. Misre~resentationsofNamibia'sPosition ..................................
D. LegalFallacies ...................................................186...........
E. Contradictions....................................................186...........
F. UnsupportedAssertions ..........................................187..........
G .Mat,Errors .......................................................188............
1. ErroneousPropositions .........:...........................188.....
2. MisrepresentationsofNamibia'sPositionof Maps ...........189.
3. MisstatementsConcerningSpecificMaps .........................
4. Errorsof ScaleandOtherTechnicalDescriptors .............195.
Conclusionto this Reply ..........................................................196............
Submissionsof thisReply .........................................................197............
List ofFigures .........................................................
.........198..
Index to Annexes .......................................................
........199..... Part One
INTRODUCTION
Chapter 1
INTRODUCTION AND SUMMARY
A. Preliminarv Remarks
1. This Reply of Namibia is submitted pursuant to the Order of the Court of 27 February
1998.'
2. Article49 of the Court's Rules state that the Reply 'shall not merely repeat the parties'
contentions but shall be directed to bringing out the issues that still divide them.' Namibia
will, of course, adhere to this prescription. Unfortunately, however, because the errors and
misstatements in the BCM are sonumerous,it will require a Reply of somelength and it will
often be necessary to refer to Namibia's earlier argumentsto correct the serious distortionsof
Namibia's positionthat colourthe whole of Botswana's Counter-Mernorial.
3. One such distortion, which the BCM has placed at the forefiont of its case, is the
misrepresentation of the principal focus of Namibia's argument. Botswana alleges that 'the
major emphasis in the Namibian Memorial is upon an asserted title by prescription, that is, a
title operating independently of the Agreement of 1890.'~ Botswana opens its Counter-
Memorial on thisnote:
1. The case presented by Namibia consists of two inter-linked arguments
based upon a certain version of the subsequent conduct of the parties and an
alleged independent title related to prescription.ese two legal arguments
are alleged to receive confirmation fiom the scientific evidence offered on
behalf ofNamibia.
'In this Replythe followingabbreviationsare used: 'NM'is the NamibianMernorial; 'NCM'is
the Namibian Counter-Memorial; 'Reply' is the NarnibianReply; 'Main Report'is Professor
W.J.R. Alexander's FirstExpert Reportin Vol. VI of the NamibianMernorial; 'Supp.Rep.' is
Professor Alexander'Reportin Vol. III of the NarnibianCounter-Memorial;'SecondSupp. Rep.'
is Professor Alexander'sReport in VolII of the NamibianReply; and 'RichardsRep.' is the
Geomorphological Repor tf ProfessorKeithS.Richardsin Vol. II othe NamibianReply. 'BM'
isthe BotswanaMernorial;and 'BCM'istheBotswanaCounter-Memorial.
BCM,para. 133. 2. It is thus the arguments based upon subsequent conduct and prescription
which arethe essence of the Namibian legal case.3 (emphasisadded)
4. These mischaracterisations are a gross distortion of Namibia's case. The Summary of
Argument opening Namibia's Memorial States unequivocally the foundation of Namibia's
case:
1. The words of the Anglo-German Treaty of 1890, properly interpreted,
attribute Kasikili Island to Namibia because 'the main channel' of the Chobe
River in the vicinity of the Island is the southem channel and the boundary is
established by the Treatyas the centre of the main channel.
2. Moreover, by virtue of continuous and exclusive occupation and use of
Kasikili Island and exercise of sovereign jurisdiction over it from the
beginning of the century, with full knowledge, acceptance and acquiescence
by the governing authorities in Bechuanaland and Botswana, Narnibia has
prescriptive titleto the ~sland.~
The two argumentsare clearly phrased as separate and distinct claims of sovereignty.
5. The Table of Contents of Namibia's Memorial likewise leaves no doubt about the prime
importance Narnibia attaches to the interpretation of the Treaty. Part One is entitled 'The
Interpretation of the Anglo-German Treaty of 1890.' It is wholly devoted to detailed
argument about the interpretation of the Treaty in al1its 'aspects. Part Two is entitled 'The
Subsequent Conduct of the Parties to the Anglo-German Treaty of 1890 and Their Successors
in Title with Relation to Kasikili Island.' Part Two begins with the explanation of the
relevance of the subsequent conduct of theparties to the case:
In the first vlace, it corroboratesthe interpretation of the Treaty developed in
Part One. Second, it gives rise to a second and entirely independentbasis for
Namibia's claim under the doctrines conceming acquisition of temtory by
prescription, acquiescence and recognition. (ernphasisadded15
Again, the two claims are separate and distinct.
'BCM, para.2.
NM, para. 14.
5NM, para. 165. The expressions "subsequentpractice"and "subsequent conduct"are frequently
usedinterchangeably.For example,the Court, initsjudgementof 15February 1995, inQatar v.
Bahrain (Jurisdictionand Admissibiliîy),I.C.J. Reports 1995, p. 6. referred to the 'subsequent
conduct'of the parties as an elementof treaty interpretationn a contextwhere, presumably,the
words 'subsequent practice'couldhave beentechnicallymore accurate. Id.pp. 15-16, paras. 26-
28. See also the opinionof the Courtof Arbitrationin the BeagleChannelcase, 52 International
LawReportsp. 224andpara. 26, infra. Namibiawouldnot havethoughtit importantto emphasize
the distinctionwere it notfor the fact that Botswanaappearsto be anxiousto generate confusion6. Namibia's Counter-Memonalbegins:
In its Memorial, Namibia based its claim to Kasikili Island on two separate
grounds: fir syti,terpretation 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 southem channel; and second, on the basis of
prescription evidenced by possession, use and exercise ofjurisdiction over the
Island, uninterrupted for almost a century after 1890 without objection or
remonstrance by ~otswana.~(emphasis in original)
7. In the light of the repeated statements of the true Namibian position, it is difficult to see
how Botswana could have believed it was fairly representing Namibia's argument in stating
that 'the argumentsbased upon subsequent conduct and prescription . . are the essence of
the Namibian legal case." The allegation can have no odier purpose than to divert attention
fromNamibia's compelling argument that the proper interpretation of the 1890Treaty places
KasikiliIsland in Namibia.
8. The presentation of the argument on prescription as an alternativebasis of title cannot be
construed as an acknowledgement by Namibia of any weakness in its argument on the
interpretation of the Treaty. Pleading in the altemative is entirely permissible under the
Court's (and al1other modem) pleading systems. Indeed, in this case, it is required by the
terms of the parties' requestto the Court
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 KasikiliISedudu Island and the legal status of the
i~land.~
The interpretation of the 1890 Treaty must come first. If Namibia is correct in its view that
the Treaty attributes Kasikili Island to it, then,of course, the argument based upon
subsequent prescnptive conduct is unnecessary. But if the Court shouldecide otherwise as
to the meaning of the Treaty, it is required by the question submitted to consider the law and
evidence of prescription inetermining 'the legal statusof the island.'
about thenatureofNamibia'sargument.
NCM,para. 2.
BCM,para. 2.
* Art. 1, Special Agreement Between theGovernmentof the Republic of Botswanaand the
Governmentof theRepublicof Namibia(hereinafter"SpecialAgreement"),NM, Annex 3,p.3.9. The organisation of this Reply follows this same logical pattern.g After this Introduction
and Surnmary,Part Two of the Reply responds to Botswana's arguments on the interpretation
of the Anglo-German Treaty of 1890and the meaning of the critical phrase 'the main channel
of the Chobe River.' Separate Chapters address the issues O treaty interpretation that both
parties, Namibia no less than Botswana, consider to be central to the case:
Chapter II exposes the defects in Botswana's presentation of the legal aspects of the
interpretation of the 1890~reaty;"
Chapter III demonstrates the fallacies in Botswana's analysis of the scientific
evidence." (In this Chapter, Namibia will also present the results of fiesh scientific
observations on site, conducted by Professor W.J.R. Alexander in March, April and
May of 1998, and a hydro-geomorphologic analysis of the Chobe River by Professor
Keith S.Richards, Chair of the Geography Department at Cambridge University);
Chapter IV counters Botswana's denial that the "subsequent practice of the parties'
established the agreement of their predecessors in title that the main channel of the
Chobe River runs to the south of Kasikili Island and that the Island is part of
~amibia;'~and
Chapter V responds to Botswana's effort to obscure the remarkable concordance of
the maps produced by al1 the goveming authorities in the area, showing their
substantially uniform understanding that the Treaty places the boundary to the south
of the 1sland.13
10. Then, in Part Three, Namibia will take up the wholly distinct issue of prescription,
countering Botswana's claim that the conduct by Namibia's predecessors in title with the
In additionto this Volume 1containingthe substanceof the Reply, Volume II contains two
scientificreports,oneby Prof. W.J.R.Alexanderandanotherby Prof. Keith S.Richards. Volume
III, Annex 1 containsa report on Mapsby ColonelDennisRushworthand a numberof additional
documentsreferredto inthisReply as"Annexes."
'OBCM,Chapter 4.
" BCM,Chapters5 and 6.
l2BCM,Chapter 4(H) and 7.
l3BCM, Chapter 8.acquiescence (and in some cases the positive action) of Botswana's predecessors in title does
not establishprescriptivetitle to Kasikili Island in favour of Namibia.14
11. Part Four takes up two further, but separate,matters. The first is a response to the long
discussion in both the BM and the BCM of the foreign policies and diplomatic relations of
the two parties.15 This discussion is largely irrelevant to the present dispute, and Namibia has
heretofore refiained fiom responding to it. But the picture Botswana presents is so partial
that Namibia has felt compelled to reply to the allegations.(ChapterVII) The second chapter
of Part Four deals with a matter that is neither marginal nor subordinate. Botswana's
Counter-Memorial is replete with error, misstatement and misrepresentation, particularly in
the two main sections that are central to this case - the scientific material identifiing the
main channel of the River Chobe and the maps evidencing the views of the parties'
predecessors in title. The most important of these errors are dealt with in the course of the
relevant Chapters of this Reply. Nonetheless, the misstatements in the Botswana Counter-
Memorial are so numerous and serious that they cannot be treated as intermittent and
innocent oversight. Rather, their scale is such as to c&t grave doubt on the fundamental
credibility of Botswana's pleading as a whole. So that the pervasive character of these errors
can be more readily seen, Namibia has gathered a number of them together in Chapter VIII
under the heading of 'Mistakes and Misconceptions.'
12. Finally, before turning to the substance of its Reply, Namibia is compelled to cal1to the
Court's attention what it regards as serious misconduct of the Botswana Defence Force
(BDF) in interfering with Professor Alexander's observations and flow measurements in the
Chobe River. On three separate occasions fiom 10-12 April 1998, armed troops of the BDF
approached Professor Alexander's boat in '
the River near Kasikili Island in a menacing and
intimidating manner.16 Once, a BDF detachment Mer up the river in Namibian territory
fired a flare at his helicopter in flight, with evident danger to al1aboard." At one point,
Professor Alexander felt obliged to discontinue his observations and had to leave the area
without taking a number of important measurements. Notice of Professor Alexander's visit
and projected itinerary was given to the Government of Botswana by note of 18 February
l4BCM,Chapter 7 and 9.
l5BM,ChaptersIIIandIV;BCM,Chapter 2.
l6SecondSupp.Rep., paras.6.32, 6.40, 6.41.
" Id., para. 6.38.1998,which was filedwith the Registrar of the ~ourt.'~ In addition,Professor Alexander was
accompanied much of the time by H.E. Joshua Hoebeb, the High Commissioner of Namibia
to Botswana, as the Govenunent of Botswanawell knew. Namibia regards these actions as a
most serious violation of the standards that should govern the conduct of litigants before the
Court and of the requirements of fiiendlyrelations between neighbours.
13. The remainder of this Chapter sumrnarizesthe positions of the parties on the main issues
in dispute, showingthat Botswanahas failed to make its case that the northern channel is the
main channel and sustaining Namibia's submission that the main channel flows to the south
of the island. The 'boundary between Namibia and Botswana around Kasiluli/Sedudu
~sland"~is show to be in the thalweg channel of that main channel, which appears in the
aerialphotographs and maps as the southern channel."
B. The Text of the Treaty
1. The lanwane of Article III(2')
14. Namibia has consistentlypresented this case as being principally about the interpretation
and application of the 1890 Treaty. Article III(2) of the Treaty provides that the boundary
'descends the centre of the main channel of the [Chobe] river' ('und setzt sich dann im
Thalweg des Hauptlaufes dieses Flusses fort'). In accordance with Article 31(1) of the
Vienna Convention on the Law of Treaties, Namibia focuses on 'the ordinarymeaning to be
given to the terms of the treaty.' The critical words are 'the centre of the main channel,'
'Thalweg des Hauptlaufes'. It is "the main channel," or "Hauptlauf" that must first be
identified. Only when it is found, can its "centre," or "Thalweg" be sought.
15. Botswana's approach, on the other hand, is marked by complexiw and artificiality. It
insists that a controllingrole should be ascribed to the mere presence of the word "ThalwegY'
in the German version of Article III(~).~' According to Botswana, this entails that the Treaty
18See LetterinReply Annex 11.
l9Art. 1, Special Agreement,NM, Annex3, p. 3.
20Fora fullelaborationofthisargument,seeparas. 75-84,infra.
2'At the time of the conclusionof the 1890Treaty it was the practice (followedin this case) to
negotiatethetextin Frenchand thenforeachsidetoproducea version in itsownlanguage. Satow,
DiplumaticPractice (2"ded.,1922), 591, pp. 70-71. According toOppenheim,in plurilingual
treaties containingno specificationof the goveming language, where therewas a discrepancy
betweenthe meaningsof the two languages,eachparty wasboundby the text inits own language
but could not clairn the benefit of the text in the language of the other paq. Oppenheim,incorporate a so-called 'thalwegprinciple,' which in turn dictates that the boundary shouldbe
drawn in thenavigable (or deeper) ~hannel.~~Botswana is in error on both counts.
16. By focusing exclusively on the one word 'Thalweg,' Botswana would have the Court
ignorethe otherwords, 'des Hauptlaufes, 'that modiQ it &d give it meaning. Indeed, BCM,
para. 197expressly argues that
"des Hauptlaufes" in the German text of the agreement was unnecessary as
the expression "the centre of the main channel" would have been correctly
translated as"der Thalwegdieses m lus ses."^^
The law of treaty interpretation, however, does not permit the tribunal to ignore the words
that were actually used inthe text and toubstitute othersthat one of the parties to the dispute
would prefer.
17. Besides disregarding the actual words of the Treaty, Botswana's position ignores the
preparatorywork aswell. The firstversion ofthe text, drafted in French,made no mention of
the 'thalweg.' Neither did the German version at that stage, which read 'langs dem
Tsch~bifluss,'~nor the English, which read 'the centre of that [the Chobe] river.'25 The
words 'Thalweg des Hauptlaufes' only appeared in the German text after the words 'of the
main channel' had been inserted in the English text, undoubtedly to reflect that change.26
There must have been a reason for the insertion of those words at this late stage in the
negotiations, and the reason could only have been a shared intention of the parties toake
the text more precise in geographical terrns. The words that Botswana suggests should be
omitted in order to lead to the conclusion it desires, are the very ones that were specifically
Internationalaw (5" ed.1937), p. 756. Botswanais the successorof Britain; its languageis,
therefore, English. Namibia is successor of Gerrnany; its language is, therefore, German.
Consequently,it is not for Botswanato hang its caseonth & ermantext - a fortiori, not on a
distortedinterpretation thereof.
22BCM,para. 196.
23This is itselfa shift from the positionBotswanatook in its Memorial,where it arguedthat the
German'Thalweg'wasthe counterpartof the words 'main channel'in the Englishtext. BM, para.
115. Thispositionwas manifestlyuntenable,since, as Namibiapointedout in its Memorial, 'des
Hauptlaufes'wasobviously thecounterpartof 'mainchannel'and 'thalweg'wasequally clearlythe
counterpartof 'centre.' Recognizingitsistake,Botswana has nowabandoned itsfirstpositionand
optsforthenew, andequallyindefensibleinterpretationassertedinBCM,para. 197.
24SeeNM,para. 109.
''SeeNM,para. 106.
26The Germantext thenbecame:'undsetztsich dann im Thalwegdes HauptlaufesdiesesFlussesadded at the last stageof the negotiations. The Court is thereforeparticularly obliged to give
them meaning and force.
The parties are assurned to intend the provisions of a treaty to have a certain
effect, and not to be meaningless . . .[A]n interpretation is not admissible
which would make a provisionmeaningless or ineffectiveF7
18. Not only is the exclusive importance Botswana assigns to the word 'Thalweg'
unwarranted, but the so-called 'thalweg principle'28 that Botswana invokes is equally
fictitious. Examination of contemporarytreaty practice in AfXcaand elsewherereveals, first,
that a variety of other terms were also used to identiQ boundaries in rivers, and, second, that
'thalweg' was used both for navigable and non-navigable riversFg Thus, Botswana cannot
show a uniform state practice generating a rule of customary international law that the
boundary in a river must be the 'thalweg.' And what practice there is shows that the use of
the word 'thalweg7was not necessarilyassociated with navigability.
19. On the one hand is Narnibia's straightfonvard interpretation, of the English text
distinguishing between 'the main channel' and its 'centre' (and in the German text between
'Hauptlauf' and 'Thalweg') according to the ordinary meaning of the words. On the other
hand, there is the complexity and artificiality of the Botswana approach, which pitches
everything on the one word 'Thalweg' in the German text, suppresses the words 'des
Hauptlaufes' and presents a definition of 'Thalweg' that goes far beyond the meaning it
normally carries. Namibia submits that the choice for the Court is simple and can lie only
with theNamibian interpretation.
2. The auestion of access to the navigationalresources of the Chobe River
20. Botswanahas placed much emphasis on the concept of navigability as an element in the
identification of the main channel. The Botswana argument as set out in its Mernorial was
based on the assertionthat
fort.'SeeNM, para. 109.
''R. Jemings and A. Watts, Oppenheirn'sInternational Law*(9" ed. 1992)p. 1280. For a recent
applicationby the Courtof the principle thata treatymust be interpretedso as to giveeffectto al1
its words, see Qatar v. Bahrain: 'Any other interpretation . . . would deprive the phraseof its
effect.'ICJ Reports 1995, pp. 18-19,para35.
28BM,para. 138;BCM,paras. 195-196.
29 SeeNCM,paras. 56-58; para. 98, infra. [Tlhe object and purpose of the provisions of Article III(2) of the Agreement
was two-fold:
(a) To affirm the rights of navigation of the two self-appointed riparian
Statesinrespect of the Chobe; and
(b) in doing so to apply the contemporary standards of general
international law relating to boundaries ofriver^.^'
As noted above, there were no 'contemporary standards of general international law relating
to boundaries of rivers.' In support of the idea that the purpose of the 1890 Treaty was to
affirm the rights of navigation of the parties in the Chobe River, Botswana cites the General
Act ofthe Conference of Berlin (1885) and various items in the correspondence of the parties
leading up to the 1890 ~reaty.~~ The Conference of Berlin was not concerned specifically
with boundary rivers. It affirmed 'fieedom of trade in the basin of the Congo, its
embouchures and circumjacent regions.'32 Thus, it had nothing to do with the Chobe River.
Insofar as it involved the Zambezi, it was expressly limited to the reach of the river fiom the
mouth to fivemiles above its confluencewith the Shiré~iver.~~
21. The correspondencecited by Botswana reveals a good deal of discussion of navigation
rights in the waters of East Afiica, especially the region between Lake Nyassa, Lake
Tanganyika and Congo tat te.^B^y contrast, in South West Afiica, the area of concern in this
case, the referencesin the correspondenceare confined to the British insistence on control of
Lake ami^^ and the German desire for accessto the Zambezi ~iver.~~Each of these points
is expressly covered by an article in the 1890 Treaty. Furthermore, in the draft Articles of
Agreement of 21 June 1890, the following sentence appeared at the end of Article III,
immediately followingthe identification of the boundary in the Chobe:
30BM,para. 144.
31BM,paras. 131-135.
32ConferenceofBerlin,Ch. 1(heading)in BM, Vol. II, Amex 1.p. 5.
33Id., para3.
34BM,Vol.II, Amex 9, LetterNo. 47, p. 82; id., No. 48, pp. 82-84. SeealsoId., LetterNo. 84,
Inclosure2, Art. III.
35Id., LettersNo. 14,24, 27, 44,84 Inclosure1,Art. III,p. 102.
36Id., Letter No. 84, Inclosure1, Art. III. The few mentionsof the Chobe Riveras appear are
mostlyinproposed draftsof thetextof theTreatyandmakenoreferencetonavigation, It is understood that, under this arrangement, Germany shall have free access
from her Protectorate to the Zambezi bv the ~hobe.~~(emphasis added)
Only ten days later, this statement was replaced in the final draft of the Treaty by the
followingsentenceopening a new paragraph:
It is understood that under this arrangement Germany shallhave free access fromher
Protectorate to the Zambezi by a strip of territory which shall at no point be less than
20 English miles in width. (emphasis added)
The replacement of 'access . . .by the Chobe' with 'access . . .by a strip of temtory'
indicates the recognition by the parties that the route of access for Germany was not to be
fluvial but rather territorial. Therefore, the correspondence and the preparatory work give no
support whatever to the conclusion that the parties were concerned to establish rights of
navigation in the Chobe River.
22. Moreover, Botswana's emphasis on the concept of navigability disregards two other
major aspects of the matter. The first is that the Chobe River is dry for much of the time over
much of its course and is, therefore, not navigable over most of its length. It hardly seems
practicable to apply a criterion of navigability toentifi the internationalboundary in a river
thatcannotbe navigated for most ofthe distancewhere it formsthe boundary.
23. The second aspect disregarded by Botswana is that the bulk of whatevernavigationthere
is in the vicinity of Kasikili Island takes place in the southern channel. The 19" century
statesmen and international lawyers who addressed the issue of the navigability of boundary
rivers were concerned, as were earlier publicists, with the basic principle of equal access by
both riparians to the resources of the river.38 The principle of equal access remains vibrant
t~da~.~' In the Chobe River at Kasikili Island, the navigational resources of the river are
primarilyrelated to the tourist industry. Substantially al1the tourist traffic uses the channel to
the south of the island, because that is where the game is to be seen. The only evidence
Botswana produces of commercialnavigation in the northern channel is a photograph of the
Zambezi Queen. (BCM, p. 133)~' But the Zambezi Queen does not navigate the northem
37See NM,Vol. IV,Annex26, p. 122.
38SeeChapterII, paras. 108-110, infra.
39 See Case Conceming the Gabcikovo-Nagyamaros Project (Hungary/Slovakia)I.C.J. Reports
1997, paras. 78, 85;reproduced in 37 Z.L.M 162,at pp. 190-191,1998.
40SeealsoBCM,para. 347.channel. It is pemanently moored in the spur channel at the King's Den Lodge where it
provides auxiliary hotel accommodation for the gue~ts.~~Thus, Botswana's insistence on the
northem channel as the main channel on the basis of an abstract relation between navigability
and the depth of the channel would, as a practical matter, deprive Namibia of access to the
actual navigational resources of the Chobe River around Kasikili Island, thereby subverting
the principle of equal access.
C. The Subsesuent Practice of the Parties Establishinn Their Agreement
as to the Treaty's Intemretation
1. The meanine,of 'subsequent~ractice' under Article 31(3)(b) of the Vienna Convention on
the Law of Treaties
24. As noted above, Botswanahas sought to obscure the relevance of the subsequent conduct
of the parties to the interpretation of the Treaty by pretending that Namibia reliesonit only in
connection with prescription and acquiescence. The assertion, however, is directly
contradicted by Article 31(3)(b) of the Vienna Convention on the Law of Treaties, which
expressly provides that, in interpretinga treaty,
'there shallbe taken into account,togetherwith the context:
(b) any subsequent practice in the application of the treaty which
establishesthe agreement of the parties regarding its interpretation.
25. Namibia has set out the subsequentpractice of the phies at length in its Memorial and
~ounter-~emorial." and there is no need to repeat the content of that argument here. In
surn,Narnibia's argument establishes the continuousoccupation and use of Kasikili Island by
the Masubia of the Caprivi fiom the time of the 1890 Treaty through the 1960s and 1970s,
and even into the 1980s. Beginning in 1909, this occupationwas under the authority of the
German rulers, exercising their sovereignty indirectly through the chiefs and other political
bodies of the Masubia. After an interval of occupation by British forces during World War 1,
governmental authority was exercised in exactly the same way, first by Bechuanaland
officials acting for SouthAfiica under the League of Nations Mandate for South West Ahca
until 1929, and then by South Afiica itself, initially as the mandatory power, and de facto
41ReplyAmex 22, Affidavitof G.J. Visagie.
42SeeNM, Chapters 1and II;NCM,Chapter IIIthereafter. This practice was well known to the British officials, stationed in Kasane just a
few kilometres away on the south ban. of the Chobe River, and to their superiors in Cape
Town and London. Yet there is no record of a British protest in al1the time fiom the signing
of the Treaty in 1890 to the independence of Botswana in 1966,nor by Botswana authorities
before 1984. As will be seen below, the maps produced by al1the goveniing authorities,
German, British and South Afncan, before Botswana's independence confirrn and are a part
of this practice.
26. In its Counter-Mernorial, however, Botswana asserts that the practice in question does
not satis@ the requirements of Article 31(3)(b). It argues that specific agreement of both
parties is required before subsequent practice may be t&en into account in interpreting the
treaty. The argument is plainly untenable. Article 31(3)(a) deals expressly with 'subsequent
agreement[s]between the parties regarding the interpretation of the treaty.' If Botswana were
correct thatunder Article 31(3)(b) the agreement of both parties is needed for subsequent
practice to be effective, it would render Article 31(3)(b) superfluous,as it would be entirely
subsurnedunder Article 31(3)(a). Article 31(3)(b)must necessarily have intended a measure
of understanding or behaviour that falls short of being an agreement on the same level of
precision or formality required by Article 31(3)(a). This was the view of the Court of
Arbitration in theeagle Channelcase:
The Court cannot accept the contentionthat no subsequent conduct, including
acts of jurisdiction, can have probative value êsa subsidiary method of
interpretation unless representing a formally stated or acknowledged
'agreement'between the Parties. The terms of the Vienna Convention do not
speci@ the ways in which 'agreement' may be manifested. In the context of
the present case the acts ofjurisdiction were not intended to establish a source
of title independent of the terms of the Treaty; nor could they be considered as
being in contradiction of those terms as understood by Chile. The evidence
supportsthe view that they were public and well-known to Argentina, and that
they could only derive fiom the Treaty. Under these circumstancesthe silence
of Argentina perrnits the inference that the acts tended to confirm an
interpretation of the meaning of the Treaty independent of the acts of
jurisdiction them~elves.~~
27. Botswana has quoted fiom a 1957 article by Sir Gerald Fitzmaurice in an attempt to
diminish the significance of subsequent practice in the interpretation of treatie~.~~However,
43521. L. R.,p. 224.
44BCM, para.235.when the quoted passage is read with the pages that precede and followit, it becomes clear
that Fitzmauriceis, in fact, a firm exponent ofthe significanceof suchpractice. Even more
to the point is Fitzrnaurice'sstatementof the principle in his fist article in the sarneseries
devotedto the case lawofthe Court:
It is a fair inference from the attitude of the Court that, in its view the
subsequentpracticeof theparties in relationto a treaty is not only a legitimate
guide to its correct interpretation, but probably a more reliable guide than
recourse to travaux préparatoiresor the attempt to ascertain the presumed
45
intentionsofthe originalfiamers.
Moreover, as is well known, it was Fitzmaurice who, as President of the Beagle Channel
tribunal,wasresponsible forthe passagesin that award quotedin the precedingparagraph.
2. Botswana's attemvtsto negate the 'subseauent~ractice'ofthevarties
28. Botswanaadvancesthreepointsthat it says areinconsistentwith theexistenceof the
subsequentpractice shownby Namibia:
That up to 1948 the parties to the Anglo-GermanTreaty 'wereof the opinion that the
main channel in the vicinity of KasikiliISedudu Island passed to the north of the
~sland.'~~
That in the exchangesculrninatingin 1951with the Trollope-Dickinsonarrangementit
was SouthAca that soughtto changethe existingstatus
That the 1984186discussions between Botswana and South Afiica resulted in an
agreement betweenthe partiesregardingthe applicationof the Anglo-German Treaty of
1890that resolvedthepresentboundarydisputein favourof~otswana.~~
29. None of these assertions is correct. Theyare dealtwith in detail in ChapterIV, below.
Somebrief cornrnents,however,arewarrantedatthispoint.
30. As to the assertionthat up to 1948the parties to the Anglo-GermanTreaty 'wereof the
opinion thatthe main channelin the vicinity of KasikiliISeduduIslandpassed to the north of
the island':
45Fitzmaurice,TheLawandProcedureoftheInternationalCourtofJustice,Vol. 1,p.61.
46BM,para. 165.
47BCM,paras.241,729,736,732.
48BCM, para.246 Botswana's positionis directlycontradictedin a letterof 6 June 1949ftom SirEvelyn
Baring, the High Commissioner of Bechuanaland,to the Secretary of State for
CommonwealthRelations. Kasikili Island,he said 'has hithertobeenregardedaspart
of the Caprivi Zipfel,ince maps show that the main channelpasses to the south of
the i~land.'~~
31. As to the assertion that in the exchanges culrninating in 1951 with the Trollope-
DickinsonarrangementSouthAhca soughtto changethe existingstatusquo:
What was the existing status quo is shown in the preceding paragraph. Far fiom
wanting to changeit, South Ahca sought to formalize the boundary that theparties
had already recognised in practice. The British authorities were at first attracted to
this notion,but then decided againstit, not on substantivegrounds, butbecauseof the
political and legal complicationsin the United Nation(UN) regarding South Afnca
and the mandate. The administrative arrangement suggested by Trollope and
embodied in the Trollope-Dickinsonarrangementexpresslyconfirmedthe statusquo
antein whichKasikili Islandwaspart of Namibia.
32. As to the assertion that the 1984186discussions between Botswana and South Afnca
resultedin an agreementthat resolvedthe presentboundary disputein favour of Botswana:
In its Counter-Memorial, Namibia showed in considerable detail the legal
impossibilityof SouthAfÏica concludingan agreementwith Botswanain the absence
of UN approval.50Botswana'sattemptto imply thatit had the approvalof the UN for
these negotiationsis contradictedby an aidememoire fiom the UN archivesshowing
that the discussionsbetween Botswanaand SWAPO in New York were reportedto
the Secretary-Generalpurely forinformationalpurposes and that no actionof support
or approvalwas soughtby Botswana ortakenby the UN.^'
33. In short, Botswana's response,both asto thelaw andthe facts in the presentcase, cannot
obfuscatethe effectof the uniformpracticeof the parties of treatingKasikiliIsland aspart of
Namibiawithin the meaningofthe 1890Treaty.
49NM, Vol. IV,Amex 66.
'ONCM, ChapterIII(B)(3).
''SeeReplyAnnex6. D. TheEvidence of Marisas a Guide to the ProrierInterpretation of Article
III(2) ofthe 1890 Treatv
34. Of the 26 maps produced by the authorities with political responsibility in the area,
sixteen show theboundaryto the south of Kasikili Island, hine do not show the boundary, and
only one appearsto show it in the northern ~hannel.~*These maps have been fùlly described
and analyzed in the Narnibian Memonal (Chapter V) and Counter-Memorial(Chapter IV), as
well as intwo detailed annexesprepared by Col. Dennis Rushworth, fonnerly Director of the
Mapping and Charting Establishment of the British Ministry of Defence, in consultationwith
Mr.Peter Clark, formerKeeper of the Royal Geographic Society.
35. Botswana does not dispute the relevance of maps and, indeed, devotes considerable
space to them in its Memonal (Chapter VIII) and Counter-Memonal (Chapter 8). However,
in light of the weight of map authority adverse to it, Botswana tries to argue that maps should
be accorded no more than a subsidiary or corroborative role. That is a matter that hardly
needs to be debated fbrther at this point. Namibia has already cited numerous authorities
describinghow internationaltribunals, including this Court,have made extensive use of maps
tnbunal~.~~ However one usefùl reference may be added that emanates fiom the highest
authority, the Court of Arbitration in the Beagle Channelcase, a court composed entirely of
Judges or formerJudges of the InternationalCourt:
[Mlaps published afier the conclusion of the Treaty can throw light on what
the intentions of the Parties in respect of it were, and, in general, on how it
should be interpreted. But the particular value of such maps lies rather in the
evidence they may afford as to the view which the one or the other Party took
at the time, or subsequently, concerning the settlement resulting fiom the
Treaty, and the degree to which the view now being asserted by that Party as
the correct one is consistent with that which it appears formerly to have
entertained. Furthermore, the importance of a map might not lie in the map
itself,whichtheoretically might evenbe inaccurate,but in the attitude towards
52The oneapparent exceptionis South West Africa 1:50,000 Sheet 1725CC With rnilitary
overprint 1984. This is an overprintof South WestAfrica 1:50,000,Sheet 1725CCEdition3,
1982,whichdidnot show the boundary. The 1984 overprintshowsa demarcationline in red in the
northernchannel.On the basisof closeanalysisof the cartographicand rnilitary/politicalcontext,
Col.Rushworthconcludesthe overprintis not in fact a depictionof the boundarybetweenthe two
countries,but of the limitsofpatrollingagreedto betweenNamibian and Botswancaomrnanderson
theground afterthe shootingincidentof October1984. SeeReplyAnnex1,para. 27.
53NM,paras.287-292;NCM,paras. 138-140. it manifested - or action in respect ofit taken- by the Party concemed or its
officia1representatives.54
Obviouslythe cogencyofrnapevidencewill dependuponmany factors, includingthe source,
the date, the clarity of the rnap and the reaction to it (if any) of the other side. In every
respect,these factorsoperatein favourofNamibia's position.
36. In Namibia's submission, the nurnber of crucial maps in this case is small, but their
impact is overwhelrning. They demonstrate (i) the consistent presentationby Namibia, its
predecessorsin title, and the UnitedNations of the boundaryin the southem channeland (ii)
the absenceof any protest, as well indeedas the presence of affirmativeacts of concurrence,
by Botswana andits predecessorsintitle.
37. The BCM criticizes Namibia'srnap evidencein a number of respects. For example,it
says a rnap that depicts neighbouringterritories by different colours does not, in so doing,
reflect theboundary, a propositionthat is implausibleon its face.55 Al1these criticismsare
dealt with in detail in Chapter V, below, and in Annex 1, the study prepared by Col.
Rushworth. Inshort,there is no substancein Botswana'scriticisms.
38. What really matters is the positive evidence ofsix maps, the effect of whichis recalled
here in surnrnaryform and in chronologicalorder. They evidence the uniform view of the
relevantpolitical authoritiesthat the boundary nuis through the channel to the south of the
Island or thatKasikili Islandbelongsto Namibia:
39. Seiner's rnap1909 (Atlas,Map IV;NM,p. 121andFigure 10).
This rnap was produced bythe prominent Austrian cartographer,Franz Seiner,who
was an expert on Germancoloniesin Afnca. It showsthe limits of SouthWestAfnca
by finered hatchingwhich clearlycoversKasikili Islandandplaces it in that tenitory.
Publishedin 1909,it was the principal large-scalernap usedby Germanyand laterby
GreatBritain,both beforeand duringthe periodwhen British authoritiesadrninistered
the Caprivi Strip fiom 1919-1929. In 1911the British Acting ResidentMagistratein
Ngamiland, CaptainA.G. Stigand,himselfthe authorof animportantrnapof the area,
5452 1.L. R.,pp. 202-203.
55See e.g. BCM,paras. 547, 577, 626. referred to it as 'a German semi-officia1map' and in 1912 described it as 'that
excellentGerman rnap of the Caprivi~i~fel.'~~
40. Streitwolf s rnap1909 (Atlas, Map VI; NM, p. 122and Figure 11).
Although this rnap does not specifically depict the boundary, it is clearly intended to
demonstrate the understanding of Captain Streitwolf, the first German Imperia1
Resident, as to the area under his authority. In this area,he has recorded place names,
including 'Kasikiri,' while no place names are shown outside it, namely, south of the
Chobe River.
41. Von Frankenberg's map, 1912 (Atlas,Map VI,NM, p. 122andFigure 12)
Von Frankenberg was Streitwolfs successor, and his map, like Streitwolf s, shows
the place narnes in the area under his jurisdiction, while showing no place names
outside that area. Both these maps were known to and made use of by the British
authorities.
42. GSGS Map 3915,1933 (Atlas, Map IX, NM, p. 125and Figure 13)
This is the first substantialBritish rnap of the area. It is said by a Botswana author to
be a milestone in the cartographic history of ~otswana.~~ This rnap marks the
boundary along the Chobe River and places Ka~ikili Island firmly in South West
Afnca. It constitutes affirmative recognition by the governing British authorities of
that attribution. There is clearly enough space at the location of Kasikili Island to
have drawn the boundary on the south side of the northem channel if it had been
thoughtthat the island belonged to Bechuanaland.
This rnap was adopted and reproduced on a smaller scale in 1935 by the
Bechuanalandauthorities themse~ves.~~ These same authorities used this reduced rnap
in 1957 to show the limits of Crown Lands in Bechuanaland, again in 1959 to show
the districtboundaries of ~echuanaland,~~and again as late as 1963 to provide the
topographical details for a Water Development Scheme Map of Northem
56SeeAnnex1,para. 12(e).
57JeffreyC. Stone, 'The1933Mapsof "BechuanalandProtectorate"at 1:500,000: A Milestonein
theMappingofBotswana,'27BotswanaNotesandRecords,pp. 71-84 (1986),NCM, Annex8.
58Atlas,Map XI; NM, p. 126andFigure 14.
j9Atlas,Map XI; NM, p. 125. ~echuanaland." Thus the rnap appearsto have been routinelyused by the Botswana
authoritiesin carryingout their administrativeduties.
43. SouthAfricanMap,TSO400/556(1945) (Atlas,MapX, NM p. 128,Fig 16)
This rnap was prepared by the Union Defence Force in 1945 although it was not
published till949.~' It shows the boundary syrnbolon the Chobe River and very
clearly followsthe southem channel. A draft of the rnap was circulated for criticism
and comment to the Bechuanaland Resident Commissioner in Mafeking and the
District Commissionerin Maun (among others). They madea number of comments,
includingcorrectionsof place names in the area of the lowerChobe River, but made
no objection to the location of the boundary south of Kasikili Island. These
exchanges, which took place three years before Trollope wrote to Redman, belie
Botswana'sclaim that the rnap shouldbe disregardedbecause it was published after
the beginning of the Trollope-Redmancorrespondenceconcerning the locationof the
boundary.
The rnapwas revised in 1967with the same depictionof the boundary. Botswana -
by thenindependent - made no protestor comment.
44. UNMap1985 (Atlas, Map XV; BM,p. 131andFigure18)
Thisrnapwas publishedpursuantto GA Res. 35/227H of 6March 1981. The legend
states that 'this rnap represents an officia1United Nations rnap of Namibia and
supersedes any other rnap of Namibia or South West Afiica hitherto published by
SouthAfrica.' It containsno disclaimerregardingboundariesand showsthe southem
channel as the main channel and Kasikili Island as part of the temtory of Narnibia.
Botswana, although it was a member of theUN Council for Namibia under whose
auspicesthe rnapwas produced, haslodgedno protestagainstthis map.
45. Botswanahas not, and cannot,producematerialthat can effectively respondto or negate
the effect ofese maps.
SeeNM,para. 309. An extractappearsatid.followingp. 127.
61Reply Annex1, para.17. E. The ScientificEvidence and the Intemretation of the 1890 Treatv
1. The legal standardapplicable to the determination ofthe 'centre ofthemain channel'
46. In its Memonal, Botswanastatedthat:
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 hydrogeomorphology.62
Namibia maintains the same position. In its Counter-Memorial, Botswana sought to
deprecate the significance of scientific evidence as 's~~erero~ator~.'~'Citing the reports of
Eason, Trollope and Redman and the 1985joint survey, it saysreference to scientific experts
would seem 'unnecessa~y'~~and 'wholly ~u~erfluous,'~~ as though reports of so-called
'observed' fact should never be questioned on the basis of scientifically established and
objectively verified fa~t.~~But the two full chapters, Chapters 5 and 6, that the BCM devotes
to the scientificevidencebelie this pretence.
47. That the scientific evidence is decisive in determining boundaries indicated by a
geographical feature is firmly established by the decisions in the alen na a^d^Laguna del
~esierto~~arbitrations, referred to by Namibia in its Memorial, paras. 123-128. Both cases
hold that the location of the boundary is to be determined by the scientific and geographic
evidence existing at the time of the award, rather than on the basis of the state of knowledge
at the time of the goveming instnunent.
48. Botswana seeks to distinguish the Palena case on the ground that '[tlhe geographical and
political context bore no relation tna~i~ability.'~~But in the Paha case, as in the present
case, the question was one of interpreting a text that laidwn a river boundary. And in that
BM,p. 129.
63
BCM,p. 121(sub-heading),para. 324.
@ BCM,para. 323.
65BCM, para. 324.
66Namibiadeals extensivelywith thethree cited reportsin NM, paras. 263-286, NCM, paras. 95-
136.
6738 I.L.R.,p. 10.
113 I.L.R., p. 1.
69BCM,para. 228.case as in this, the question was to identie 'the major channel' of the river. The Court of
Arbitration held unequivocally that '[tlhe question whch is the major channel is a
geographical question,' and explained that 'themajor channel can be determined on historical
and scientific gr~unds.'~~Thus, although the Court took the words of the definition as its
startingpoint, it conducted its examination by reference to developments and knowledge that
arose subsequent to the formulation ofthe definition in the 1902Award, the instrument under
examination. Those who drafted that Award did not know the content of the scientific
elementsrelevant to the identification of the major channel- its length, size of drainage area
and volume of discharge. The arbitral tribunal approached the problem entirely in obiective
terms, as contrastedwith any subiectivebelief that the draftsmanof 1902may have had.
49. Botswana also dismissesthe decision in the LagunadelDesierto arbitration, stating only
that it 'does not appear to be relevant to the issues presented in the present proceedings.'71
Why is it not relevant? The case concerned the interpretation of that part of a boundary
instrument (again it was the 1902 Award between Argentina and Chile) providing that the
boundary would 'ascendthe localwater-parting to Mount Fitzroy.' At the time of that Award
the precise location of the water parting wasunknown. It was only discovered some 40 years
later. The Tribunal found that the words used in 1902Award required the boundary to follow
the water parting whatever its location might be when ultimately established as a matter of
scientific fact. To this end, the Tribunal entrusted its Expert Geographer with the task of
identieing the local water parting.72 The Tribunal expressly rejected Chile's contention that
'to apply the 1902 Award in light of geographical knowledge acquired subsequently would
be equivalent to its revision through the retrospective consideration of new facts.' The
decision says:
The 1902 Award defined in the section with which this Arbitration is
concemed, a frontier which follows a natural feature that. as such. does not
depend on accurate knowledne of the area but on its true confimation. The
ground remains as it has always been. Consequently, the local water-parting
between BP62 and Mount Fitzroy existing in 1902 is the same as that which
'O38 I.L.R., pp. 93-94.
71
BCM,para. 230.
72Award, para. 151. can be drawn at the date of the present Arbitration. Accordingly, this Tribunal
is not revising butfaithfully applying the provisions of the Award of 1902.~~
(emphasisadded)
50. Far fiom being irrelevant,both of the Palena and the Laguna del Desierto cases address
the sarne basic legal issue that arises in this case: by what critena is the Court to determine
the location of a geographical feature identified in the Treaty as the boundary between the
two litigants? Both awards give the same answer. The determination is to be made on the
basis of scientific evidence available at the time the tribunal makes its decision, not on the
basis of the suppositions the draftsmen of the instrument may have entertained. Botswana
itself recognizes this essential identity of the issues. It cites the Temple case7' for the
proposition that the determination of 'the centre of the main channel . . .is essentially a
question of fact.' TheBM continues:
There are other examples of treaty provisions referring to factual or
geographical criteria such as a crest line, or a watershed line or an escar~ment:
see the Judgment in the Temple case (Merits)I.C.J. [sic], 1962, p. 6 at p. 15.
In such cases the factual cntenon is adopted or converted into the legal
criterion. But it does not cease to be in essence a questionof fa~t.~~(emphasis
added)
Thus Botswana itself recognizes that the exact nature of the geographical feature is not
relevant to the legal standard, which requires the application of current factual and scientific
knowledge whether the boundary feature be 'a crestline, or a watershed line or an
escarpment' - or 'the centreofthe main channel' of the ChobeRiver.
2. Namibia's scientificcase on the identificationofthe 'main channel' ofthe
ChobeRiver at Kasikili Island
51. Namibia has maintained fiom the beginning that the correct scientifictest for identiQing
the main channel is the one that conveys the largest proportion of the annual flow of the
river.76 Botswana appears to sharethe sarneview. The BCM says: 'The main channel is the
73Id., paras. 157,158.
74I.C.J. Reports 1962, p. 6. This case, it wilbe recalled,dealt with a boundary markedby a
watershedand wasthus factuallymuch more remotefrom the present case than the Palenacase
whichwasprecisely abouttheidentificationof a 'majorchannel.'
75BM,para. 200.
76NM, para 157;id., Vol. VI, paras. 1.8, 2.8; seealsoNCM, Vol. III, paras.2.1, 5.1.channel which carries the greater amount of ~ater.'~~ Moreover, although the BCM
fiequently refers to the comparative depth or width of the two channels, Botswana's expert,
Dr. F.T.K. Sefe, states categorically that '[clhannel width and depth are not alternatives for
determining the main channel of a river.'78 So the issue between the two sides is clearly
joined: where does the largestproportion ofthe water of the ChobeRiver flow?
52. In the more than two years since this case was filedwith the Court, a great deal has been
learned about the Chobe River in general and in the vicinity of Kasikili Island in particular.
The chronicle of the annual cycle of the river as described by Professor Alexander in his
Main Report is now well established. The Chobe River rises with the annual rains in the
Angolan highlands (where it is known as the Rio Cuando) and makes its way southward
across the border of the Caprivi Strip,until it peters out in the Linyanti swamps, about 30
kilometres upstream of Lake Liambezi. The dryriverbed continuesto the Ngoma Bridge and
on, in the dry season, to Serondela,only 15 kilometres west of Kasikili Island. During that
penod (approximatelythe latterhalf of the year) the water in the two narrow channels,made
familiar in this case by the many reproductions of maps, aerial photographs and satellite
images,remains essentiallystagnant since there is no sourceofwater fromupstream.
53. This stagnant condition continuesuntil the Zambezi River begins its annual rise, usually
inlate February or March. The impact is felt first below Kasikili Island in the anabranched
channelsof the Zambezi River and at the Mambova Rapids. The Rapids, acting as a kind of
dam or weir, cause the nsing waters to back up in the bed of the river until Kasikili Island is
substantiallyinundated. As the Zambezi River rises further it overflows its banks along the
whole reach from Katima Mulilo to the Mambova Rapids. The overbank flow proceeds
southeast across the floodplain of the Zambezi River, until, a short time later, it reaches the
Chobe Ridge along a considerable front. The influx of these waters causes the flow in the
Chobe River to reverse direction and to proceed in a downstream direction past Kasikili
Island. This downstream flow continues at gradually decreasing rates for two or three
BCM, para. 385. Botswanasometimesargues thatthe northern channelis the main channel
becauseit hasgreatercapacitythanthe southern channelby reasonof its greaterwidth anddepth.
BCM,para. 350. However,as Botswanaitself recognizesin the sameparagraph,capacityis not
the same thing as flow, which requires in addition, among other things, information as to the
velocityinthechannel,andofthisBotswanahas providednoproof.
''BCM,Vol. II, App. 4, para. 50. The inappropriatenessof depth orwidthas stand-alonecriteria
fordeterminingvolumeof flowisdiscussed briefly at Chapter Ip IIaras. 166-168,infra;andat NM,
paras. 131, 132,157;id.Vol. VI,para. 4.4; NCM,para. 28.months, as the rate of flow in the Zambezi River decreases and the upstrearn source that feeds
the Chobe River dries up. By about July or the beginning of August, the river in the vicinity
of Kasikili Island reverts to its normal and familiar dry-season state - stagnant to al1intents
and purposes.
54. It followsthat the period for determiningthe channelthat carries the largest proportion of
the flow of the river - which both parties agree is the decisive criterion for determining the
main channel - is the period when the river is in flow. On this question, Namibia's position
has been consistent from the beginning. ProfessorAlexanderdepictedthe main channel on a
reproduction of an aerial photograph on Sheet 17, Diagram 4 in the Appendices to his Main
~e~01-t~a 'nd again in the Second Supplementary Report, Figures 16, 17 and 18. (See Fig. 1,
followingp. 23) This channelclearly flowsto the south of the i~land.~'The right bank of the
main channel is hard against the line of trees at the base of the Chobe Ridge, which is
characteristic of the 'banks of non-perennial rivers in a semi-arid region.'81 It proceeds
straightalong this tree linebeginning upstream of the bifurcation and going on past the Island
and on to the Mambova Rapids. The left bank, while perhaps less distinct, is a well-defined
line of high ground crossing the southern part of the island in a generally west-east
direction.82 When there is substantial flow in the Chobe River ernanating from upstrearn of
Kasikili Island, most of it flows between these two banks. Within this main channel is the
southem channel as it appears in most of the aerial photographs, which is in fact the thalweg
channel of the main channel.
55. Diagram 4 was based on theoretical analysis of the pre-existing geomorphologic and
hydrological data. As noted above, since then, Professor Alexander has made extensive
persona1observations and flow measurements on site, and additional aerial and helicopter
photographs have been taken. Al1of these new data are not only consistent with Professor
Alexander's predictions,but triumphantly confirm them. ,
56. Professor Alexander took flow measurements at points N1 in the northem channel and
N2 in the southem channel as located on Figure 8 of the Second Supplementary Report. (See
79NM, Vol. VI, App., Sheet 17.
SeeNM,p. 141,Submissions (1).
''NM, Vol. VI,Sheet10,Photographsa and b (caption).
82See Chapter III,para.182, infra.Fig. 2, followingp. 24) He chose these sites because they aggregatethe flow in each channel
emanating fiom upstream of Kasikili Island, while excluding the flow that enters them
downstream of the island. Therefore, measurements at these points reflect the flow of the
Chobe River around Kasikili Island. The measured flow was 247 m3/s in the southem
channel and 188 m3/s in the n~rthem.'~ Thus, the actual 1998 measurements at the site
corroborate that most of the flow (roughly 60 per cent in this case) goes to the south of the
island and only about 40 per cent goes to thenorthem channel.
57. The visual evidence supplied by photographs, particularly Botswana's photomosaic of
June 1997 andphotographs taken by Namibia in 1997 and 1998, confimis and reinforcesthis
a~count.'~These arethe only availablephotographs taken when the ChobeRiverwas in flow
around Kasikili Island. On both the 1997 and the 1998 photomosaic, the main channel
appearsin the exact form predictedby Professor Alexander at the begiiuiing of the pleadings,
with the right bank hugging the trees and the left bank defined by the line of higher ground
crossing the island in a generally west-east direction. (See Fig. 1, following p. 23, above;
Fig. 6, following p. 72).
58. Photographs 51 and 52 taken from a helicopter are particularly in~tructive.~~ (See Fig. 3,
following p. 25) Photograph 51 looks downstream from about the Chobe National Park
headquarters, at the point where the Chobe Ridge rnakes a dogleg to the north, to the
Mambova Rapids at the top of the picture. The right bank of the river running along the
ChobeRidge is sharply defined, and the left bank is clearly discernible. Photograph 52 is the
reverse direction, looking upstream fi-omthe Mambova Rapids past Kasikili Island in the top
third of the picture. Again the right bank of the river along the ridge is sharply marked, and
in this view, the leftbank is alsoquite distinct. As Professor Alexander says:
83 Second Supp. Rep. para. 7.3, Fig. 8. In his SupplementaryReport, Professor Alexander
hypothesizedthat therewouldbe substantiallyno flowin thenorthernchannel evenat timesofhigh
water, because of the fiow into the eastem leg of the northern charnel from the anabranched
channelsand the downstream control exercised by the MambovaRapids. NCM, Vol. IIIparas.
6.1-6.5. However,whenpeak flows reach thelevelthey did in1998,overbankflow reachesthe
northern charnel froma varietyof sources. In thesecircumstances,flowintothe northernchannel
at its downstream(eastern)end cannotcontrol the overbank flow,and some of the waterfrom
upstreamofKasikili Island goesthrough thenorîhern channel.See SecondSupp. Rep. paras.6.44,
7.7, 7.8.
84BCM,p. 191;SecondSupp.Rep., Fig. 17andAmex 1,p. A1126,Photographs51 and52.
85SecondSupp.Rep., App. 1,p. A 1/26. Wl Wei wjom
ma -2-JI~YOE
Bîsiuirs~Wneg pue
UseYriWwlul~Pai~
Z wnst The northem channel enters this straight reach 6f the Chobe River at right
angles in the middle of the two photographs. It is clear that the northem
channel is subsidiary to this uninterrupted reach of the Chobe River and that
the main channel can not be considered to consist of half of the straight reach
and then extend at right angles into the floodplainhalf way along its length.86
The visual appearance and the analysis of the photographs show that the straight reach is the
main channel of the Chobe River. The northem channel is not.
59. As suggested by Botswana, Professor Alexander's account of the present-day hydrology
of the Chobe River and pattern of flows around Kasikili Island is a practical and pragmatic
one.87 However, Professor Richards, working within a more theoretical geomorphologic
fiamework, reaches the sarne conclusion that the main channel is the channel to the south of
the Island.
60. Using well-established methodology, Professor Richards concludes that the meander
loops on the ChobeRiver upstream of Kasikili Island aretoo large to havebeen cut by a river
with the capacity of the present Chobe River. Professor Richards posits that they were
formed by the Zambezi River itself,'when in anearlier geologic era it flowed in a channel
furtherwest than it does today. Kasikili Islandwas a fourth in this series of meander 100~s.~~
Using the carbon dating data fkom Dr. Sefe's Sedimentological Study, Professor Richards
concludesthat about a millenniurn or so ago, the Zarnbezi River migrated east to its present
channel, leaving a fossil channel or paleo-channel that is now occupied by the Chobe ~iver.~~
Before the Zarnbezi River moved, however, the Kasikili bend
experienced a cut-off (an avulsion across the meander neck). The southem
channel in its current form thus post-dates the northem, not vice versa as
argued in para. 353 of the Botswana Counter-Memorial, Volume 1. The
formation of the southem channel as ameander cut-off is consistent with the
evidence of such phenornena over the whole of the Caprivi fan, and with the
norrnal behaviour of high-sinuosity river meanders. It is also consistent with
86SecondSupp. Rep., para. 14.25. Thepatterndisplayed inthesetwophotographsis very similar
to that in Professor Alexander's photographo sn theLinyanti reach of the Chobe River, much
further upstream.SeeSecondSupp.Rep., App. 1,p. A115,Photographs 9 and 10.
87As noted inChapterIII, para 125,Botswana'sattackon ProfessorAlexander's qualificationis s
mean-spiritedin toneandwhollywithoutmerit.
RichardsRep., paras. 22-25.
89Id.,para. 27. -;-,
mgun?3,
Contluencsofthenorihem the charactenstics of the large upstream bends at Serondela, which have
pronouncedchute channels. This cut-offis likely to have occurred very soon
aflerthe [Kasikili]bend reachedits outerlimit,at about1,000years ago ...90
61. Thenorthem 'channel'is, thus, a cut-offmeanderloop. In due course it 'willbe blocked
at its upstream end,in the same way as the spur ~hannel.'~*The only reason this will not
happen at the downstreamend is the influxof water into the eastem leg of the channel from
the anabranchedchannels east of Kasika.
62. 'Thesouthem channel,' saysProfessorRichards
is the result of an avulsion,occming well before the 1890 Treaty,in which a
chute channel beneath the Chobe Ridge across the neck of the bend was
extendedfully acrossthe neck by flood flows. Sincethis cut-off, the southem
channelhas canied most ofthe flowin the lower~hobe.'~
By virtueof this avulsion athousand yearsago,the channeltook theformshown inProfessor
Alexander's Diagram 4 in his Main Report and Figurès 16, 17, and 18 of the Second
SupplementaryReport. This channel, flowing to the south of Kasikili Island, became and
remainsthe main channelof the ChobeRiver.
3. Resvonseto Botswana's 'scientific case'
63. It remains to address the miscellany of propositions and evidence that constitute
Botswana's scientific case. At the outset of its scientific discussion, the BCM stakes its
position onsix propositions. These are discussed irnrnediately below,where it is shownthat
four are false, one is uncontested, and theremaining one is true but supports Namibia's
position. Next, the Reply refutes Botswana's secondary hydrological arguments,and then
reviews theevidencefiom aerial photographs andsatelliteimagery. Finally,Namibia returns
to the question that it reserved in its Memonal: the identificationof 'the centreof the main
channel'of the River Chobe.
(a) The sixvro~ositions
64. Under the heading 'Botswana's caserelating;to the River Chobe based on scientific
evidence,' the BCM puts fonvard six propositions that, it says, support its case.93 The
SQId.,para. 28.
91Id.,para. 30.
92Id.,para.28.
93BCM, para.327 etseq.detailed discussion of these propositions takes up the next 30 pages of the BCM and
represents the main substantivediscussion of scientificissues.The six propositions are:
'(i) The Chobe is the geographical feature in Article III of the 1890
Agreement'
'(ii)The Chobe is a river independent of the Zambezi River'
'(iii)The Chobe has a stableprofile as aperennialmature river'
'(iv)The Chobeis a perennialriverwith visible and stable banks'
'(v) Thereis an absence of zones of sedimentation inthe northern channel'
'(vi)The Chobe is a river with continuous flowyg4
The first of these propositions is uncontested. The second, third, fourth and sixth are false.
The fifth, although true, confirms that no substantial flow takes place in the northem
~hannel.'~Namibia will now discusseach of the propositions in tum:
The ChobeRiver is 'ariver independent of the Zambezi River.'
On the basis of a correlation of the river flows at Katima Mulillo and Kasane
for over 50 years, Professor Richards concludes that the consistency of the
relationship 'suggests very strongly that the water levels in the lower Chobe
are essentially controlled by the flood dischargesattainedin the ~ambezi."~
The 1945 Report of the Kalahari Reconnaissance says that 'in effect the
Linyanti [Chobe] below Kachikau is part of the Zambezi River, the land
betweenbeing really largeislands of the ~arnbezi.'~~
A Joint Report of the Namibian and Botswana Departments of Water Affairs
of 28 June 1992 concludes that 'under the current circumstances with Lake
Liambezi being dry, it was felt more relevant to consider the Chobe River in
conjunction with the Zambezi ~iver.'~~
94Thepropositionsare the headings ofBCM, paras. 328-334,respectively.
95See SecondSupp.Rep., paras. 17.1, 17.2.
% RichardsRep., para. 18.
97ReplyAmex 32,p. 5.
98SecondSupp.Rep., App.2, p. A214,para. 3.1 The Chobe does have 'a stableprofileas aperennialmatureriver.'
The Chobe is not a perennial river at all. As Botswana's expert says, 'a
perennial stream never dnes p."^ ProfessorAlexander's threereports contain
16photographsshowing various parts of the drybed of the ChobeRiver above
Ngoma ~rid~e.'"
Professor Alexander testifies that in 1997 and 1998, '1 flew by helicopter
along the whole length of the ChobeRiver as identified in the 1890Treaty. 1
observed and photographed long lengths of the Chobe River where the river
was completely drywithout visiblechannels or clearlydefined bank~."~'
In 1985, the Namibia Department of Water Affairs reported that 'There has
been no flow fiom Lake Liambezi intothe Chobe since September 1982.'lo2
In June 1992the Joint Report of the HydrologyDivisions of the Namibian and
Botswana Departments of Water Affairs states 'With Lake Liambezi having
been dry for almost a decade, there has been no contribution to the Chobe
River fiom this source.'lo3
Professor Richards states clearly that the river is not perennial.'104 He adds
that '[slince the Linyandi-Chobe passes through this lake [Liambezi], it is
difficult to sustain the argument that the Chobe is a perennial river."05
Professor Richards also shows that similarconditionshave occurred cyclically
in recent and moreremote history, andprobably in prehistorictimes as we11.'~~
99BCM,App. II,para.9.
Seealso Chaper III,paras. 131-132,infra.
'O'SecondSupp. Rep. para. 4.5. See also NCM, Vol. III,para. 7.9 ('The lower reach of the
LinyandiRiver from Lake Liambeziwestwardswas completely dry during my inspectionand no
channels werediscemablefromtheair').
'O2ReplyAnnex24.
'O3Reply Annex 25.
'O4RichardsRep., para. 16.
'O5Id. Seealso ChapterIII, paras. 131-132, infraandSecondSupp.Rep. Photographs15 and 16.
'O6RichardsRep., paras. 19-21. A 1990 report fÏom the Lonrho Namibia Caprivi Sugar Project says Lake
Liambezi 'cm by no means be classified as a perennial lake' and that in the
1940's thebed of the lake was 'largelyunder ~ultivation."~~
The Chobe is 'a perennialriverwith visible and stablebanks.'
As to the non-perennial character of the Chobe River, see the preceding
paragraph.
The visibility of the banks of the northern channel during periods of high
water is said to be shown on a series of aerial photographs, al1but one of
which were taken during periods of low water. The single exception was the
photograph taken in June 1997,a year when the peak flow in the Zarnbeziwas
abnormally low and when, as Botswana admits, the water was already
receding. 'O8
As stated above, it is true that 'there is an absenceof zones of sedimentationinthe
northem channel,' but this is an indication not of the presence but of the absence
of flow in that channel. Professors Alexander and Richards agree that the absence
of zones of sedimentation in this case indicate that the northem channel is a
meander loop in the process ofbeing closed off,not an activeriver channel.log
The Chobe is ariver with continuous flow.
In support of its proposition, the BCM adduces the records of seven gauging
stations along the Chobe fiom Upper Kwando to Kasane. These records are
records of water levels and do not contain any data fiom which can be
deduced.
It isknown fiom other evidence that there is continuous flow in the Chobe
River fiom the station at Upper Kwando, where it first enters Namibia, and for
the next 120 kilometres through to the sixth gauging station at Shaile. It is
107
ReplyAnnex 26, p.3.
'O8See SecondSupp. Rep., paras. 13.7, 13.10-13.13. The other photographon whichBotswana
particularlyreliesis the 'May 1972' photographw , hichis misdated. See Chapter III,paras. 187-
188, infra;SecondSupp.Rep.,para. 13.6.
109See SecondSupp. Rep., para. 17.2; RichardsRep., paras. 28-30. See also NCM, Vol. III,
Supp.Rep., paras.5.3-5.5, 5.17; NM, Vol. VI, MainReport,para. 8.7. equally clear fiom the photographs, eyewitness testimony and reports cited
above that in the dry season the River is completely dry for the next 40
kilometres from Shaile through to Lake Liambezi and beyond to below
Ngoma Bridge. Thus, there is no continuous flow through the Chobe River
and past Kasikili Island. There are gauging stations in this dry stretch of the
River below Shaile, but Botswana fails to include the data fiom these stations
in its tables and diagrams.
Botswana repeatedly asserts that the Chobe River maintains a 'constant level
of 925.6 metres' at the Kasane Gauging Station below Kasikili ~sland."~ Its
assertionthat there is continuous flow is pitched largely on this measurement.
In the first place, the measurement is incorrect, since it would mean that much
of Kasikili Island would be under water most of the time. The correct figure is
probably 924.6 masl."' More important, however, is that the level at Kasane
is maintained in the dry season not by flow through the Chobe River from
upstream of Kasikili Island, but by flow through the anabranched channels
from the Zambezi that enter the Chobe below the Island. The 'constant level'
at Kasane does not, therefore, constitute evidence of continuous flow in the
Chobe River at Kasikili Island.
65. Thus, al1the substantive propositions on which Botswana bases its scientific case are
shown to be utterly unfounded. The scientific case they are said to support must fa11with
them.
(b) Secondarvhvdrological arguments
66. Botswana raises a number of subsidiary scientific or quasi-scientific arguments, al1of
which are dealt with in full in Chapter III. A few of these ment brief comment in this
Introduction.
1IOE.g.,BCM,para. 334. See SecondSupp.Rep., Section 12. In fact, Botswana's terminologiys
anything butconstant. At varioustimesit refersto the 925.6 metrefigureas a 'level'(BCM,para.
284), a 'constantlevel. . . throughal1seasonsof theyear'(id.,para. 334),a 'minimumlevel' (id.,
para. 336),and a 'constantminimumlevel'(id., VolII, App.2, para. 43).
111SeeSecondSupp.Rep.,Section 12. (i) Bedslope
67. Botswana says that the northern channel carries the greater flow of water because it is
'steeper' than the southernchannel. To prove this assertion, it produces a graph, plotting the
elevations in each channel at the cross-sections measured by the 1985joint survey. The
graph implies that the total drop in elevation in the northern channel is greater than in the
southem. The graph ends, however,before the confluence of the two channels at the eastem
end of the Island. The undeniable fact is that, since the points of origin and points of
confluence of the two channels are identical, the differencesbetween the beginning and end
of each channel must also be identical. Thus, the allegation of the greater 'steepness' of the
northem channel turns on a deliberatelyincompletedata set.
(ii) Erosion
68. Ordinarily,river flow causes bank erosion. Botswana attempts to explain away the fact
that, despite its claim that the greater flow takes place through the northem channel, the
banks of that channel show no signs of erosion. To this end, the BCM, citing Dr. Sefe's
Sedimentological Study, argues that the banks are stable because they are formed of non-
erodible material. Namibia has produced photographs showing the erodible character of the
banks of the northern ~hanne1.l'~Moreover, what Botswana itself regards as "sedimentary
material" appears on the aerial photographs along the right bank of the northern channel.
Therefore, the fact that the banks have not been eroded can only be a consequence of the
absenceof flow of water in the channel.
(iii) Flow measurements
69. Botswana cites flow measurements said to be taken on three dates in March and April
1997 indicating that the flow was larger in the northem than in the southem channel.l13
Elsewhere the BCM refers to minimum flow velocities in the dry season 'on the order of 0.5
metres per ~econd.'"~ No details are given of the points in the river at which the
measurementswere taken or the methodology employed. In the absence of evidence on these
points they can stand as no more than rnere assertions to which the Court cannot properly
give anyweight.
112SeeChapter IIIpara. 155,infra.
113
Id.,para. 285.
"4 Id.,paras. 275, 283, 368.70. As will be seen below, Professor Alexander, despite harassment by the BDF,was able to
take careful flow measurements in the northem and southem channels on 30 April to 2 May
1998. The sites at which these measurements were taken and the procedures followed are
described in detail in Professor Alexander's Second Supplementary ~e~ort."' Sites N1 in
the northern channel andN2 in southern were chosen because al1the flow in the Chobe River
upstream of Kasikili Island passed one of these two points, while no flow from downstream
sources entered either channel before these points. The readings showed a flow of 247 m3/s
in the southernchannel and 188m3/sin the northem channel. Thus, almost 60 per cent of the
flow was through the southem channel and only about 40 per cent in the north. (See Fig. 2,
followingp. 24, and para. 56,above)
(c) The aerial photona~hs and satellite images
71. Botswana continues to attach great importance to aerial photographs and satellite images.
The BCM's analysis is characterized by repetitive statements that the overall configuration of
the area has not changed (a matter that is not in dispute) and that the photographs show the
northem channel to be wider and deeper than the southem. Chapter III@) below responds in
detail to Botswana's photographic analysis. For the present, it is sufficient to make a few
general observations.
72. m, al1the photographs, except the one of June 1997, were taken in the dry season,
when there is hardly any flow in the Chobe River at Kasikili Island. In the wet season, when
thereis flow in the river, the water level rises and, as Professor Alexander demonstrates, most
of the island is inundated.l16 However, it is evident fiom an inspection of the photographs
that most of Kasikili Island isabove water. It follows that these photographs were taken
during the dry season. This is only to be expected, since the photographs were taken
primarily by reconnaissance, survey and mapping expeditions interested in the terrain and
topographic features of the area.ll7 Comparisons made when there is little or no flow in
either channelcan hardly provide the basis for de~idingwhich carries the majorportion of the
flow.
115SecondSupp.Rep., paras. 6.21-6.29.
Il6SecondSupp. Rep.,para. 13.7andTable1,p. 43.
117SecondSupp.Rep., para. 13.19.73. Second, in any case, the comparisons of the channels as shown on the photographs
provide little informationrelevant to this case. Inspection of high altitude aerialphotographs
cannot show whether water is actually flowing in the channels. Likewise, it is impossible to
derive from the photographs information, in any relevant detail, about the depth of the
channels. The width of the surface of a channel cannot by itself evidence the rate of flow.
The case is essentiallythe same for satelliteimages.
74. Third, Botswana insists that the aerial photographs reveal no changes in the sediment
bars in the southem channel. This is said to prove that the southem channel is not an active
channel. Botswana is wrong for two reasons. First, the photographs do in fact reveal
changes, and these are acknowledged even by ~otswana."~ Second, the photographs also
reveal substantially identical sandbars in the Chobe River above the bifurcation. These are
created by sediment deposited in the channel by the flow df the river. Since this is true of the
sandbars above the bifurcation, it can be no less true of the sandbars actually within the
southem channel and so stands asproof of the activity of the southem channel.
4. Identifvinn the 'centre' of the main channel
75. In its Memorial, Narnibia noted that it was unnecessary at that time to discuss the
question of where the boundary lies within the main channel. The principal issue was the
identification of the main channel. Thereafter, 'the location of the centre of the main channel
would follow largely as a matter of course by reason of its dependence upon the manner in
which the principal issue is re~olved.'"~ Namibia reserved its right to return to the issue if
developments in the casemade it appropriate to do SO.'~~
76. Botswana, in its Memonal made no expressreference to the 'centre' of the main channel,
but its argument on the identification of the main channel was that 'the middle of the
navigable channel was the b~undary."~~ Presumably, therefore, Botswana equated 'the
centre' of the main channel with the middle of the navigable channel. Botswana's Counter-
Memorial also conflates its identification of 'the centre' of the main channel with its
'laSeeBCM,para. 377.
Il9NM, para. 159.
120NM, para. 160.
12'BM, para. 137.definition of the main channel in terms of the thalwegof the river.122
77. The time to address the question Namibia reserved in its Memorial has now arrived. The
main channel of the ChobeRiverhas been identified as shown originallyon Diagram 4, Sheet
17of the Appendices to ProfessorAlexander's Main Report and in Figures 16, 17, 18and 19
of his Second Supplementary Report. (See also Fig. 1, following p. 23, above) The next step
is to determinewheretheboundary lies within this main channel.
78. Article III(2) of the 1890Treaty defines the boundary as descending "the centre of the
main channel" of the River Chobe. The German text says "im Thalweg des Hauptlaufes. "
(emphasis supplied). The correct interpretation of Article III(2) is the one that fully
reconciles the English and the German texts. Othenvise, the result would be to disregard
eitherthe word 'centre' or theword 'thalweg' as superfluous, which is forbiddenby the rules
123
of treaty interpretation.
79. Theinterpretation that meetsthis requirement will construeboth words as referring to the
thalweg of the main channel in its technical and geographical sense as the line of deepest
soundings in that channel. The Court does no violence to the language by treating the words
'thalweg' and 'centre' as synonymous in this context. As noted in Chapter II, below,
although the practice was by no means uniform, the use of the English word 'centre' to
indicatethe thalweg of a channel or a river was a perfectly acceptable usage.124On the other
hand, the German word 'Thalweg' could notproperly be applied to the geometric centre of
the river.
80. Thus, the Court will give full effectto al1the words of the Treatyby determining,
-irst, that the 'main channelof the Chobe River' is the channel defined by Professor
Alexander, as indicated above, and
second,that the thalwegof that channel is the boundary.
This two-step procedurereflects the plain meaning of the text of the Treaty, and Namibia has
advocated it from the beginning of its written pleadings.
'22BCM, paras. 194-199.
'23SeeChapterII,n. 9 andn. 10,infra.
124See alsoNCM, para. 57.81. Both Professor Alexander and Professor Richards agree that the thalweg of the main
channel, as defined by Professor Alexander, liesithin what has been referred to in this case
as the southem ~hanne1.l~~They have shown that the appearance of the channel that flowsto
the south of Kasikili Island differs considerably according to the season of the year and the
rate of flow in the River. When the Chobe River is flowing, the bulk of the flow goes
through the broad conduit that Professor Alexander has identified as the main ~hanne1.I~~
Indeed, that is why it is the main channel. During thdry season, however, the river assumes
the configuration that is familiaron the rnaps and mostfstheaerialphotographs presented in
this case. The southern channel, in that configuration, is thethalweg channel of the main
channel. The main channel overlies this thalweg channel, so that during the high flow
season,when there is flow in the ChobeRiver at Kasikili Island, the flow in the main channel
submerges this thalweg channel. In the dry season, when the river assumes the configuration
familiar from the maps and aerial photographs, the water that remains in the channel to the
south ofthe Island continuesto mark the thalweg channel of the main channel.
82. This is a common phenornenon in non-perennial rivers in semi-arid regions of the world,
where the river dies up or shrinks to a narrow rivulet during the dry season and swells to
bankfull or overbank levels in annual periods of high water. The phenomenon can be seen
for the Chobe River at Ngoma Bridge by comparing the dry season photograph looking
downstream taken on 20 September 199512'with a photograph taken from the same point on
11 April 1998. (See Fig. 4, following p. 35)12' The first shows a narrow, meandering, even
discontinuous thalweg channel in the middle of the otherwise dry bed of the Chobe River,
which is marked by tree-lined banks on either side at some distance from the thalweg. In the
second, in the high flow season, the river fills alrnost the entire area of the main channel
between the two lines of trees. The same cornparison between the dry season, where a
narrow, sinuous channel marksthe thalweg over which the river flows at bankfull in the high
water season, can be seen for the Limpopo and Orange Rivers on Sheet 4 of the Appendices
125
SecondSupp.Rep., paras. 14.14, 14.19;RichardsRep., paras. 3,4, 34.
'26SecondSupp.Rep., Sec. 7, andRichardsRep., paras. 28, 32, 34,38. SeealsoNCM,Vol. III,
Supp.Rep., para.5.19; NM, Vol. VI,para. 12.1.
'" NM,Vol. VI,App., Sheetlla.
128
Second.Supp.Rep., Fig. 10.to Professor Alexander's Main ~e~0rt.l~~This is what occurs at Kasikili Island during the
high flowperiod fiom March to June.
83. On the interpretation of the Treaty that Namibia proposes here, the boundary would be
the line of deepest soundings in the southem channel, that is, the thalweg. This would
correspond to what is shown as the boundary on almost al1the maps of the area since the time
of the 1890 Treaty. It would reflect the understanding of Namibia and its predecessors in
interest fiom the time of the Treaty, and that of Botswana's predecessors in interest until at
leastthe time of Botswana's independence. And it would allowboth parties full access to the
navigationalresources of the river for purposes of the tourist indust~==y.'~~
84. In Namibia's submission, this is the interpretation of Article III(2) of the Anglo Gennan
Treaty that the Court should adopt.
F. Prescription and Acauiescence
85. A substantial portion of Namibia's Memorial is Part Two, entitled 'The Subsequent
Conduct of the Parties with relation to Kasikili Island.' This part embraces two parallel
strands - the subsequent practice of the Parties as evidence of their understanding of the
meaning of the Treaty and their subsequent conduct as constitutingprescriptive behaviour in
which the predecessors in title of Botswana acquiesced. The account of the maps of the area,
discussed in Chapter V, and sumrnarized at paras. 34-45 above, is an integral and entirely
consistentpart of this subsequentpractice and conduct. The post-1890behaviour of Namibia
and its predecessors in titletreating the southern channel as the boundary is relevant in two
ways. It serves to confirm the interpretation of the 1890Treaty that favoursthe identification
of the channel to the south of the Island as the boundary. It also serves as the foundation for
a prescnptive claim that negates such forma1 title as Botswana might conceivably have
obtained under the 1890Treaty. As noted at the beginning of this Chapter, there is absolutely
no impropriety in the presentation of arguments in support of Namibia's claim to Kasikili
Island on the basis both of Treaty interpretation and prescription. Nor is either argument
diminished by the development of the other. Each argument uses the same material in a
differentway.
12'NM, Vol. VI, Sheet4a and4b (Limpopo),4c and4d (Orange). See also SecondSupp.Rep.,
Photograph62 (MkuzeRiver),Photograph 63(MhlatuzeRiver),p. A1132.
130Namibia hasalwaysbeenwillingto allowBotswanatousethenorthernchannel.86. Once this misconception is out of the way, the argument for prescription is quite
straightforward. Chapter VI of this Reply shows, largely through testimony fiom witnesses
Botswana accepts as reliable and other undisputed evidence, that there was peaceful
occupation and use of the Island by the Masubia of Caprivi for the greater part of the period
under consideration under the authority of their traditional leadership, responsible to the
German, British (as administrators of the Mandate for South Africa) and South Afncan
authorities. There was no objection, let alone control, by the Bechuanaland authorities.
Indeed, Botswana does not seriously contest that this was the situation at least through the
1940s and into the 1950s, the so-called 'critical date' proposed by Botswana. In fact,
continuous use and control of the Island by South Afkican authorities continued until 1989,
when they finally withdrew fiom Narnibia. Although Botswana argues that there was no
exercise of jurisdiction on the Island, it is clear that the Masubia chiefs and Indunas were
actinglocally as agents of the German and laterthe South African government.
87. Conversely, there is no evidence of any attempt whatsoever at occupation of Kasikili
Island by residents of Bechuanaland or at the exercise of jurisdiction by the Bechuanaland
authorities. There was not only acquiescence, there was completepassivity. And it was not
only the rules of prescription that obligated the British authorities to take action in defence of
their assertedrights. The 1890Treaty delimited boundaries not between sovereign temtories
but between the spheres of influence of the parties. According to Jan Verzijl, '[sluch zones
or spheres did not thereby become Stateterritoryproper, but as far as they were concemed an
inchoate title was vested in the proclaiming or delimiting State(s) that could gradually
develop into full s~verei~nt~."~' Therefore, some form of positive assertion of jurisdiction
or control was necessary if Britain was to perfect its title. Under the assumptions about
acquisition of temtory prevailing at the end of thegthcentury, the silence of the British and
Bechuanaland authorities is evenmore significant,both as a matter of subsequent practice for
the interpretation of the Treaty and subsequent conduct groundinga claim of prescription.
E. Conclusion
88. Botswana proposes a disarmingly simple solution to the question presented in this case:
ArticleIII(2) of the 1890 Treaty says the boundary is in the thalwegof the river; the thalweg
is the deepest channel; the northernchannel is on averagethe deeper; ergo, the boundary runs
13'J.H.W. Verzijl,InternationaLlaw in Historical Perspectiveo,l.III, Leyden, 1970,p. 495.in the northern channel. But the Chobe is not a simple river, and this is not a simple case.
Botswana's formula is not simple,it is simplistic.
The Treaty does Say'the thalwegof the river;' it says 'the centre of the main
channel.'
Greaterdepth does g& identie the main channel;the largerproportion of the flow
in the river does.
The flow passing through a channel cannot be determined by inspection of aerial
photographs taken in the dry season; it requires expert analysis of the hydrology
and geomorphology of the Chobe River at Kasikili Island.
Moreover, Botswana's 'simple solution7would erase the centxry-long history of
occupation, use andjurisdictional control of Kkikili Island by the Masubia people
of the Caprivi and by Namibia's predecessors in political authority, al1 of it
recognized and accepted by the British and Botswana govemmentsuntil 1984.
Professors Alexander and Richards are uniquely qualified to perfonn the hydrologie and
geomorphologic analyses that are called for. They demonstrateconvincingly that when water
is flowing in the Chobe River at Kasikili Island, the largest proportion of the flow is carried
by the channel lying to the south of the Island, identified in Fig. 1, following p. 23, above.
This is 'the main channel' within the meaning of the Treaty. The thalweg channel of this
main channel is the southem channel cornmonly appearing on the maps and aenal
photographs. The thalweg - the line of deepest soundings within the thalweg channel - is
the boundary between Botswana andNamibia at Kasikili I'sland.
89. This answer to the question submitted to the Court is the only one that fully reconciles
the language of the 1890 Treaty, the scientific evidence, the subsequent practice of the
parties, the map evidence and the history of the last century. In Namibia's submission, it is
the answerthe Court should give. Part Two
THE INTERPRETATION OF THE ANGLO-GERMAN TREATY OF 1890
ChapterII
THETEXT OF THE TREATY IN THELIGHT OF ITS OBJECT AND PURPOSE
A. In General
90. As a preliminaryobservation, Namibia recalls the BCM's serious distortion of the overall
thrust of Namibia's position. Botswana is mistaken in asserting that Namibia 'discounts the
significance of the 1890 Agreement' and that 'the major emphasis' in the Namibian
Memorial 'is upon title by prescription.'1 The charge could easily be turned against
Botswana. Its Memorial contains considerablyless substantiveargument on the question of
interpretation of the 1890 Treatyhan Namibia's. Of the 45 paragraphs (roughly 15pages)
Botswana devotes to the interpretation of the 1890Treaty, only four (1 112pages) deal with
the wording of the Treaty. The rest is a lengthy recapitulation of the generals of treaty
interpretation as to which there is no disagreement at al1between the parties to this case. In
the Namibian Memorial, on the other hand, the whole of Part One deals with 'The
Interpretation of the Anglo-German Treaty of 1890,' andincludes a precise analysis of the
text of ArticleIII(2), a discussion of the object and purpose of the 1890 Treaty, and a
substantial and illuminating account of the history and preparatory work leading to the final
text of the Treaty.As Namibia stated at the beginning of thatPart 'The first ground on
which Namibia contends that Kasikili Island belongs to it rests on the terms of the 1890
Anglo-German ~reaty.'~ Thus, the charge that Namibia 'discounts the significance of the
1890Agreement' is totally unfounded.
91. However, because Botswana's Memorial and Counter-Memorial are so replete with
inconsistent and erroneousassertions regarding the literalng of the relevant passages of
the 1890 Treaty and its object and purpose, as well as with regard to the alleged 'self-
executing' nature of the 1890 Treaty, Namibia will once again address the question of the
'BCM, para.133.
*NM,para.43.correct interpretationof the Treaty. In so doing, Namibia will follow the traditional canons
of treaty interpretation as they are reflected in Article 31 of the Vienna Convention on the
Law of Treaties. Thus, this Reply addresses, first, the literal interpretation of Article III(2) of
the Treaty, and second, the interpretationof the Treaty in the light of its obiect andniose. 3
Third, Narnibia will respond to Botswana's erroneous contention that the question of the
determination of the main channel of the Chobe River is ipso iure answered by the alleged
self-executingnature of the 1890Treaty.
B. Intemretation of the Words ofthe Treatv
92. The interpretation of the 1890 Agreement centre's around the correct reading and
meaning of the relevant words used in Article III(2), which describe the course of the
boundary in the Chobe River:
[Tlhe . . .line .. . runs eastward along that parallel [18" parallel of south
latitude] till it reaches the River Chobe, and descends the centre of the main
channel of that river to its junction with the Zambezi, where it terminates.
(emphasisadded)
93. In the course of the pleadings, Botswana has substantially shifted its position on the
interpretation of Article III(2). Initially, in its Memorial, Botswana stated that '[tlhe German
text employsthe term Thalwegas the counterpart to "the main channel" in the English te~t.'~
This reading is manifestly incompatible with the German text of Article III(2). Recognizing
its mistake, Botswananow maintains that the words 'centre of the main channel of the river'
have to be read as a single whole. On this analysis, according to Botswana, the German
translation of thephrase should have been 'imThalwegdiesesFlusses '(in the thalwegof this
river): instead of.the actual German translation, which reads 'im Thalweg desHau~tlaufes
dieses Flusses '(in the thalweg of the main channel of this river ). (emphasis added) This
NO insightintothe meaningof ArticleIII(2)of the 1890 Treatycan be gained froma systematic
interpretation(Article1 paras. 1and2 of the ViennaConventionon the Lawof Treaties)because
theTreav usesthecontested phrase"centreofthemainchannel"onlyonce. Therefore,thereisno
room fora systematic or contextuailnterpretation. It appearsthat Botswanadid not put forward
anyargumentbased onthe systematicinterpretation,for the samereason. Thesubsequentpractice
of the parties within themeaningof Article 31(3)(b) of the ViennaConventionis dealt with in
ChapterIV, below.
BM,para. 115.
BCM,para. 197.amounts to saying that the words 'desHauptlaufes 'in the German text are unnecessary and
therefore meaningless.6 On both counts,Botswana is wrong.
94. As pointed out in the NCM, it is plainly wrong to maintain that 'Thalweg' is the
The English words 'of the main channel'
counterpart of the words 'the main channel."
obviously correspond to the German 'desHauptlaufes. ' Botswana apparentlybecame aware
of the error of its original approach and now acknowledges that the English text 'of the main
channel' andthe Germantext 'desHauptlaufes' may indeed havedistinct meanings from the
words 'centre' and 'thalweg.' Theparties to the Treatymay have referred to the centre or the
thalweg of the main channel because contemporary witnesses had observed that there were
islands in the~hobe,*which rneant that there could be more than one channel and, therefore,
more than one centre or thalweg along which the border could descend. If so, Namibia's
position is correct: in interpretingthe Treaty, the tasfirs to determine which of the two
channels around Kasikili Island is the main channel andthe tnfind the centre or thalweg of
that channel.
95. In an effort to avoid this straightforward result, Botswana has changed its original
position on the meaning of the words inthe Treaty. It now maintains that the formula used in
Article III(2)has to be read as a whole, and that the words 'des Hauptlauf'in the German
text must be ignored as unnecessary. Indeed, BCM, para.197expresslyarguesthat:
"des Hauptlaufes" in the German text of the agreement was unnecessary as
the expression "the centre of the main channel" would have been correctly
translatedas "der Thalweg dieses Flusses.
This approach nuis counter to the rules of treaty interpretation, which require that the
provisions of a treaty be construed as they stand,ggiving meaning to every word of the text,
SeeId.
'See NCM, para. 52.
8 See, for exarnplF. C. Selous,A Hunter's Wanderings inAfnca (1895), writtenafter extensive
travelling in 1874 in thearea relevantto the case at hand; see Reply Annex 34. Dr. B.F.
Bradshaw's 1881 articleintheProceedingsof theRoyalGeographical Societyentitled'Notesonthe
ChobeRiver,SouthCentral Africa'(cf. NM, Annex115)andhismapshowing theChobeinunique
detail. (ThemapisreproducedinNM, Fig .7 followingp. 27).
Asrecentlyas 1994in the Libya v.Chadcasethe ICJupheldthisrule in statingthat'interpretation
mustbe basedabove al1uponthetextof the treaty.'ICI Reports 1994, p. 6, at p22; seealsoSir
RobertJenningsand Sir Arthur Watts(eds.) Oppenheim's Zntemational Law, 9" edition(1992),
Vol. 1- Peace, parts 2-4,p. 1271.withoutrewriting it to fit a desiredmeaning ofthe provision in question.10
96. As a preliminarypoint, there is a question of how far Botswana is entitledto rely on the
word thalweg in the German text of the Treaty. According to Oppenheim's fifth edition,
whichmay be thoughtto reflect pre-World War 1treatypractice:
Unless the contrary is expressly provided, if a treaty is concluded in two
languages and there is a discrepancy between the meaning of the two different
texts, each party is only bound by the text of its own language. Moreover, a
party cannot claimthebenefit of the text in the language of the other
Botswana is the successor of Britain. Its language is, therefore, English. Narnibia is
successor of Germany; its language is, therefore, German. Consequently, it is not for
Botswana to hang its case on the German text - afortiori,not on a distorted interpretation
thereof.
97. In its first argument purporting to show that the words "des Hauptlaufes" in the German
text of the 1890Treaty are unnecessary, Botswanamaintainsthat the use of the German term
"Tilalweg" alone would have fully met the intentions of the parties. This is because
(Botswana claims) the establishedinternationalpractice at the time of the Treaty was to use
the term thalweg in determiningthe boundary in navigable rivers. The flaw in this argument
is that, contrary to Botswana's assertions,there was no uniform terminologydescribingriver
boundaries at the time of the Treaty. The random selection of international boundary treaties
relating to Afiica discussed in the NCM reveals that a variety of terms are used to describe
the course of river boundaries.12 Even the 1890 Treaty itself uses several different terms.13
IOSee Polish Postal Servicein Danzig (Advisory Opinion) P.C.I.J. 1925, Series B, No.11, p. 39;
Admissionofa Stateto Membershipin the UnitedNations (AdvisoryOpinion), 1.C.J. Reports 1950,
p. 8;Anglo-Zranian Oil Co Case,1.C.J. Reports(1952), p. 105,wheretheCourtemphasizedthata
treaty shouldbe interpretedinsucha waythata reasonandmeaningcan beattributedtoeveryword
in the text. See also RudolfBernhardt, 'Interpretationin InternationalLaw,' in Encyclopediaof
Public InternationaLl aw,Vol.II,pp. 1416et seq., at1420(1995). Thisprinciplegoesbackto the
writingsof HugoGrotiusandparticularlyto Emérique de Vattel,Le Droit des Gens, ou Principes
de la LoiNaturelle,livreII,8 283.
11Oppenheim, InternationalLaw (5h Edition 1937), p.756.
See also Hall,InternationalLaw (8h
edition 1924), p. 392. At the tirne of the conclusionof ,the 1890 Treaty it was the practice
(followed inthiscase)to negotiatethetextin Frenchandthenfor eachsideto producea versionin
itsownlanguage.SirE. Satow, Diplomtic Practice2"d Ed. (1922),8 91,pp. 70-71.
12
NCM, para. 57: the ten treatiescited show at leastnine differentterms used to describethe
courseof river boundaries,including theterm "thalweg".
l3NCM,para. 56: Articles1,III, IV showfour distinct formulae describinriverboundaries.Moreover, althoughBotswanaclaims that navigabilitywas the primary concern ofthe parties,
the term thalweg in the various treaties is not linked exclusively to the navigability of the
relevant rivers; likewise, terms other than thalweg are applied in relation to navigable
14
rivers. Of course, the term thalweg was more generally used in Europe. However, here
also the term was not exclusively linked to navigability. Indeed, in some thnty treaties
concludedbetween Europeanpowers during the 19thcentury, the term was used in relation to
non-navigable rivers almost the same number of times as to navigable rivers." Botswana,
therefore, cannot substantiateits contentionthat the words 'des Hauptlaufes 'in the German
text were unnecessaryby reason of an alleged establishedusage and unequivocal meaning of
the term thalweg. On the contrary, the fact that German state practice would not normally
add 'des Hauptlaufes 'to the term thalweg, but did so in this case, indicates that these words
were deliberatelyinsertedintothe German text to conforrqit to the words used in the English
text.16
98. In a second attempt to support its proposition that the words 'des Hauptlaufes' in the
German text were unnecessary, Botswana introduces two further points of a seemingly
linguistic or terminological character. On closer examination these arguments prove to be
grounded in an erroneous understanding of the scope and normative reach of the thalweg
concept. Botswana arguesthat if the words 'of the main channel' or 'desHauptlaufes ' were
given a distinctmeaning,the Ennlish text would be distorted. Botswana's explanationof this
'distortion'is stunningly confused. It states:
In English, in the context of the 1890 Agreement, the expression "the main
channel of' has no meaning of its own but is simply used to express the
thalweg principle. If one omitted the words "the'main channel of' fi-omthe
expression "the centre of the main channel of the river" it would remain "the
l4 Thus, for instance, ArticleIII of the treaty betweenthe UnitedKingdomand France of 1898
describesthe boundaryas followingthe medianline of the [Niger]river - a river thatis clearly
navigable.See NCM, para.58; Convention between Great Britainand France for the Delimitation
of their respective Possessionto the Westof the Niger, andof their respectivePossessions Easo t f
thatRiver. Signedat Paris 14 June 1898,ratified13June 1899.(emphasis added)SeeBrownlie,
Afn'canBoundaries, p. 619 et seq.
ISSeelistinReplyAM~X 5
l6 On the last minuteinsertionof the formula'centreof the main channel'and 'im Thalwegdes
Hauptlaufes diesesFlusses,'see NM, para. 102et seq., Chapter 1 para. 17,supra, and alsopara.
103,infra. centre of the river", an expression which in English does not express the
thalwegprinciple.'7
In other words,Botswana would have the Court ignore the Englishwords actuallyused and
impute tothe parties a meaning theydid notexpress. It shouldgo without sayng that sucha
procedureis incompatible with therules of treaty interpretation. The basis of this attempted
legerdemainis again Botswana'sunfoundedcontention that at the time of the conclusionof
the 1890 Treaty there was a clearlyestablishedstate practice with regardto the terminology
used in describing the courseof riverboundaries. As shownin the NCM'~andin this Reply,
statepracticeat the end of the lgthcentury - including thatof the UnitedKingdom - was not
consistent in the choice and content of the terms used for describing river boundaries, as
Botswanawouldhave the Courtbelieve. In facttherewasno suchpractice.lg
99. On closer examinationit becomes evidentthat Botswana's insistence that, regardlessof
the language used, Article III(2) should be construed to embody the thalweg concept is
motivated not by linguistic reasons, but rather because of the normative implications it
erroneouslyattaches to the thalweg concept. Botswana's seemingly linguisticargument is
developedin the contextof its repeated, erroneousassertionthat the thalwegconceptby itself
determineswhich of two channelsof a river is the main one. This assertion,in turn, is based
onthe false assumptionthat therecan onlybe one thalwegin a river. Thus,the textual claim
is inextricablylinked to the normative elementsof Botswana'sargument. In surn,Botswana
is assertingthat: (i) ArticleIII(2) in its English version describestheboundaryas descending
along the thalweg of the Chobe River - even thoughit uses the formula'centreof the main
channel'; (ii)there canbe only one thalwegin a river, and(iii) wherethe thalwegnuis, there
the main channel is to be found. Thus, in Botswana'sview there is no room forany prior
determinationof which of the two channelsaroundKasikiliIslandis the main channel. From
this it follows that,i-oma linguisticpoint of view, the words 'of themain channel' or 'des
Hauptlaufes,' though they appear in the Treaty, cannotbe given a distinct meaning of their
ownandshouldbe ign~red.~'
l7BCM, para. 196.
l8See NCM, paras. 57-58.
19
In anycase, it was surelyopento thepartiesto a treatyto departfromthe allegedprinciple if
theywishedtodo so.
20BCM, para.196.100. However,althoughBotswanarepeatedly and categoricallyasserts that 'A river has only
one thalweg,'2'in the end it admits, as it must, that "[iln a bifurcated stretch of river, such as
the Chobe River in the vicinity of Kasikili/Sedudu Island, both channels will have their
respective thalwege. '22 By this admission, Botswana has undermined its own argument. If
each channel has its own thalweg, then the question of which is the main channel must be
answered independently of the concept. Not only does Botswana retreat from its earlier
contention that there can be only one thalweg in a river, but, in another striking about-face, it
proceeds to suggest criteria by which themain channel can be dete~mined.~~
101. For its part, Namibia has consistently held that there are two channels in the Chobe
River around Kasikili Island, each possessing its own thalweg.24 Consequently, to apply the
treaty it is first necessary to determine the main channel of the Chobe River and then to
establishthe course of the boundary in that channel. This is the clear import of the words 'of
the main channel' and 'desHauptlaufes.' Thus Namibia's interpretation adheresto the plain
meaning ofthe English and German texts of Article III(2) of the 1890Treaty. It does not ask
the Court to ignorewordsclearly written in both texts asunnecessary or meaningless.
102. Finally, support for the view that the words 'the main channel' and 'des Hauptlaufes'
have a distinct meaning of their own is also provided by the drafting history of Article III(2)
of the 1890 Treaty. As the Narnibian Memorial has already pointed out, the words 'of the
main channel and 'desHauptlaufes 'were inserted into the text of Article III(2) at a very late
stage of the negotiations.25 Until25 June 1890the description of the course of the boundary
in the draft treaty did not contain any reference to the notion of 'the main channel.' Both
parties sirnplyreferred to the border asbeing 'carriedto the east alonn the centre of the River
~schobi'~~ and in the German draft text '[die] Grenze ... führt ... von da nach Osten lanas
21BCM,para.199;seealsoparas. 198, 336.
22
E.g., BCM,para. 338;seealsopara. 339.
23BCM,para. 347etseq.
24NM,Vol. VI, para. 3.9.
25NM,paras. 102et seq. Botswanaagrees essentially withthisaccountof thedraftinghistory. See
BCM,para. 193.
26QuotedfromArticle IIof the Englishtranslationof the initialleddocumentrecordingthe general
agreement between the two governrnents,see NM, Annex 21; p. 34 (emphasisadded);the original
Frenchtextreads: 'La frontière . . seportera à l'estlongeantle centredu Fleuve Tschobi. . .,'
id.p. 33.dem ~schobiflu~.'~~(emphasis added) The later insertion of 'the main channel' and 'des
Hauptlaufes' indicates that these words were meant to clariQ the original wording, not to
replace it with a whole new formula. Botswana's reading of Article III(2) would have the
Court ignorethe very words that the parties thus deliberatelyinserted at the last moment. It is
compatible neither with the plain words,nor with the clraftinghistory of that provision.
103. In summary, Namibia reaffirms that there was no established uniform terminology for
international river boundaries at the time of the 1890Treaty. Nothing can be inferred from
the terms used in the Treaty to supportBotswana's assertionthat the words 'desHauptlaufes '
in the German text of the Treaty were unnecessary and could therefore be ignored in the
interpretation of the Treaty. On the other hand, Namibia's interpretation of Article III(2)'
attributinga distinct meaning to the words 'of the main channel' or 'desHauptlaufes,' is the
only one that does justice to both the English and the German texts. Botswana now admits
that there are two channels around Kasikili Island, eachwith its own thalweg. It follows that
the parties had good reason to express their intention that the boundary should run dong the
centre or thalweg of the main channel of the Chobe River, an intention that is clearly
reflected in the language of Article III(2). Namibia, therefore, maintains its position that
Article III(2), properly interpreted, requires first, the identification of the main channel and
then the determination of the centre or the thalweg of that channel.
C. Interpretation of the 1890Treatv in Light of its Obiect and Purpose
104. Botswana states that Namibia, in interpreting the 1890 Treaty, did not concem itself
with the 'object and purpose of the Anglo-German Agreement and the significance of the
thalweg in relation to na~i~ability.'~~ This charge is another reflection of Botswana's
inaccurate reading of Namibia's Mernorial. In it and later in the Counter-Memorial, Namibia
clearly based its interpretation of the 1890 Treaty on its object and purpose.29 The main
points of Namibia's argument on object and purpose are:. Great Britain and Germany aimed
at a large-scale delimitation of their respective spheres of interest, inter alia, in South West
27Quotedfrom the officialGermantext printedin the 'Officia1Gazette'of Berlin, 17 June 1890;
seeNM, Annex 22.
28BCM, para.223.
29SeeNM, Chapters IIIand IV, NCM,paras. 50, 74.~fnca." There, the major British concem was to extend the British sphere of interest to the
north in order to secure control of Lake Ngami. The German interest centred on recognition
of a sphere of influence extending as far east as the Zambezi ~iver.~' Given the macro
dimension of the whole enterprise of the 1890 Treaty, it is hardly surprising that the
preparatory work for the Agreement as well as the actual negotiations show that the parties
did not concem themselves with particulars such as whether the main channel of the Chobe
River ran to the north orthe south of Kasikili
105. For its part, Botswana agrees that what the Parties had in mind was 'in essence a
partition of several important regions of Afnca into spheres of influence of Germany and
Great Britain respectively, with some territorial "compensation" for Germany el~ewhere.'~~
In addition, however, Botswana discovers a 'connected purpose' in the diplomatic
correspondence 'to maintain and facilitate access to the principal lakes and rivers of the
African continent.'34 It argues that although Great Britain was specifically interested in
access to Lake Ngami, while Germany's specific interest was in access to the ~ambezi,3~in
both cases, the underlying interest was in access to and fiee navigation on the major Afiican
rivers and lake~.~~Botswana continues by observing that 'reference to the Chobe in Article
III(2) occurs in close association with the Zambezi and it is clear that access to waterways
and the general question of navigation was a matter of major ~oncem.'~~ In conclusion,
Botswana states that:
[Tlhe object and purpose of the provisions of Article III(2) of the Agreement
was two-fold:
(a) to affirm the rights of navigation of the two self-appointednparian
Statesin respect ofthe Chobe; and
30NM,para. 80.
31NM,paras. 88 and90 etseq.;NCM,para. 39.
32NM,para.115.
33BM,para. 130.
34BM,para.131.
35BM,paras. 133and 134.
36BM,para. 133.
37BM, para. 137. (b) in doing so to apply the contemporary standards of general
internationallaw relating toboundaries in navigable river^.^'
As noted above, there was no contemporary standard of general international law relating to
river boundaries. In support of its invention Botswana cites the General Act of the
Conferenceof Berlin (1 885) and various items in the correspondenceofthe parties leading up
to the 1890 ~reaty.~~ The Conference of Berlin was not concerned specifically with
boundary rivers. It affirmed 'fieedom of trade in the Basin of the Congo, its embouchures
and circurnjacent regions.740 Thus, it had nothing to do &th the Chobe River. Insofar as it
involved the Zambezi River, it was expressly limited to the reach of the river from the mouth
to fivemiles above its confluencewith the ShiréRiver, in the lower Zambezi River far to the
easttowards the Indian Ocean. 41
106. The correspondencecited by Botswana reveals a good deal of discussion of navigation
rights in the waters of East Afiica, especially the region of Lake Nyassa, Lake Tanganyika
and Congo tat te.^B^y contrast, in South West Afiica, the area of concern in this case, the
references in the correspondence are confined to the British insistence on control of Lake
ami^^ and the German desire for access to the Zambezi River.44 The text of the 1890
Treaty confirms this view of the correspondence. Article IX calls generally for fieedom of
navigation in lakes,rivers and canals in East Afiica 'in'al1the portions of their respective
spheres within the limits of the free zone createdbv the Act of Berlin of 1885.'" (emphasis
added) This formula excludes the area of concem in this case. In South West Afiica, British
control of Lake Ngami and German access to the Zarnbezi are each covered by an express
38BM,para.144.
39BM,paras. 131-135.
40Conferenceof Berlin,Ch. 1,Art. 1, BM,Vol. II, Annex1,p. 4.
41Id.
42BM,Annex 9,LetterNo. 47, p. 82; id. No. 48, pp. 82-84.
43E.g. id., NO.84, Inclosure1,Art. HI.,p. 102.
Id. Thefewmentionsof the ChobeRiverthat appeararemostly inproposeddraftsof thetextof
theTreatyand rnakeno referenceto navigation.
45 A previous draft of the agreement providedthat '[tlhe two powers engage to applyin their
respectivespheresin East Africa, within the lirnitsof the free zone, the provisionsof the Act of
Berlinaccordingto which thenavigationof the lakes, rivers, canals . . .is free to bothflags.'
BCM, Annex 9, No. 84, Inclosure2, ArticleIX,p. 105.provision in Article III and no mention is made of navigation.46 Furthermore, in the drafi
Articles of Agreement of 21 June 1890,the following sentenceappeared at the end of Article
III,immediately after the identification of theboundary in the Chobe River:
It is understood that, under this arrangement, Gennany shall have fi-eeaccess
fiom her Protectorate to the Zambezi bv the ~hobe.~~(emphasisadded)
Only 10 days later, this was replaced in the Treaty as finally signed by a sentence opening a
new paragraph as follows:
It is understood thatunder this arrangement Gerrpany shall have fiee access
fiom her Protectorate to the Zambezi 7 which shall at no
point be less than 20 Englishmiles inwidth. (emphasisadded)
The replacement of 'access . . .by the Chobe' with 'access . . .by a strip of temtory'
indicates the recognition by the parties that the route of access for Gennany was not to be
fluvial but rather territorial. The correspondence and the preparatory work therefore give no
support whatever to the conclusion that the parties were concemed to establish rights of
navigation in the ChobeRiver.
107. Evenif Botswana is correct in this view of object and purpose of the Treaty, however, it
is conspicuously silent on the question of the concrete meaning of the general notions of
freedom of navigation and equal access to the rivers and lakes of Afiica. It finds the
embodiment of these notions in the abstract thalweg concept, without reference to the
practical aspects of fi-eedomof navigation on, and equal access to, the Chobe River in the
vicinity of Kasikili Island. It is therefore necessaxyto enquire into the concretemeaning of
the principle of freedom of navigation and equal access in the context of the economic needs
of theregion, includingthe vicinity of Kasikili Island.
108. The thalweg concept as an approach to drawing river boundaries represents an
evolution from earlier methodologies. For centuries riparian states have drawn boundaries
along rivers in a manner that would allow them to share equitably in the benefits of the
ordinary uses of the particular river.48 And as recently as 1997in the Gabcikovo-Nagymaros
46The secondparagraphof ArticleIII(2)quotedabove inthe textdealswith Germanaccessto the
Zambezi. The third paragraphsays, '[ilt [the spheinwhich theexerciseof influenceis reserved
toGreat Britain]includesLakeNgami.'
47SeeNM,Vol. IV, Annex26, p. 122.
48For a summaryof legaldoctrineat thetimeof theTreatysee N.Kercea,Die Staatsgrenze indencase, this Court, recalling the judgement of the Permanent Court of International Justice in
Territorial Jurisdiction of the International Commission of the River der,^' strongly
emphasized the principle of equitable and reasonable sharing of the resources of the river by
the riparian state~.~' The Court held that, even taking into account Hungary's violation of
treaty obligations vis-à-vis then Czechoslovakia, the latter's unilateral diversion of the
Danube unjustifiably deprived Hungary of its right to an equitable and reasonable share of the
natural resources of the ~anube."
109. The earlier doctrines included comrnon ownership of the river and, later, partition of the
riverbed along a middle line. The thalweg concept was introduced into state practice around
the end of the 18th centu~y.~~Aside fiom the general noiion of equality of access, a feature
of al1these methods of riverine boundary determination was their close link to the economic
interests in the uses of the river, and in particular the localneeds of the riparian ~tates.~~With
regard to the principle of equitable sharing of the navigational and other resources of a
boundary river, the thalweg concept is no different fkomthe earlier approaches to drawing
river boundaries. In essence, they al1aim to secure to the riparian states equal access to the
uses of the river.54
110. Thus, it becomesnecessary to establish what the uses of the Chobe River are in the light
of the regional needs. To start with, it is quite clear that, whatever notions the parties to the
1890 Treaty may have had, navigation on the Chobe River has never approximated that on
the large watenvays in Afnca discussed by the great powers at the Berlin Conference of
1885." Therefore, the uses of the Chobe River to which equal access is to be secured must
be established in the light ofregional needs as they have developed over time. Any other
Grenzflüssen (1916),p. 135;therelevant partsofthebookarealso reprintedinBCM,Annex9.
49JudgementNo. 16, 1929,P.I.C.J., SeriesA, No.23,p. 27
50 Case Conceming theGabcikovo-Nagymaros Project (Hungary/Slovakia),paras. 78 and 85;
reproducedin 37 I.L.M.162, at 190-191(1998). See also the cases of Wilsonet al. v. Omaha
IndianTribeet al., 442 U.S. 653 (1979), andLouisianav. Mississippi etal, 466 U.S. 96 (1984)
decided by theUnitedStatesSupremeCourt.
CaseConcemingthe Gabcikovo-NagymaroP sroject,para. 85, in 37I.L.M. 162atp. 191(1998).
52On thehistoryof theconceptseeKercea,p. 113et seq.
53SeeNCMpara. 42.
54SeeBCM,para. 224.
55BM,para.131.approach would frustrate the equal access principle thatk Botswana agrees, forms the core
element of the object and purpose of the ~reaty.'~
111. There can be no question that the needs of the riparian states with regard to the Chobe
River around Kasikili Island- aside fiom smallboat communication - are connectedwith the
tounst industry. In this respect, the NCM has already shownthat it is the southem channel of
the Chobe River that offers the main opportunities for the tourist business and therefore is the
most used channel. Professor Alexander provides additional evidence of this fact in his
Second Supplemental~e~01-t.'~It is in this channel that thebulk of boat tours for the tourists
are undertaken, departing fiom the lodges locatedin the Chobe National Park on the southem
bank of the river. There is little boat tourism in the northem channel, because it is fiom the
souththat tourists get the best view of the game on the soilthernbank of the Chobe River and
on Kasikili ~sland.'~
112. Thus, even on Botswana's view of the object and purpose of the Treaty, the conclusion
must be that the channel to the south of the Island is the main channel of the Chobe. By
insisting on the northem channel, Botswana is actually betraying the very principle of equal
rights and access forboth riparian statesthat it claims embodies the object and purpose of the
1890 Treaty. For, if the boundary were to be redrawn along the northem channel, Namibia
would be entirely shut off fiom the southem channel and thereby deniedthe use of the Chobe
River where it actually serves the needs and interests of bothiparian states. This would be
incompatible with Botswana's own view of the object and purpose of the 1890 Treaty and
with the generalprinciple of equitable and reasonable sharing of a boundary river enunciated
by the ~0u1-t.'~
113. Finally, Botswana alleges that Namibia's interpretation of Article III(2) would result in
a determination of the main channel that would 'seriously reduce the efficacy' of the 1890
Treaty and run counter to its object and purpose of establishing a readily recognizable
bo~ndar~.~~This charge cannot be supported. On the contrary, it is Botswana's candidatefor
56See BCM,para. 224.
57SecondSupp.Rep., paras. 8.1-8.2;seealsoAffidavitof G.J . isagie,ReplyAnnex 22.
SecondSupp.Rep.para. 8.3.
59See theCase ConcemingtheGabcikovo-NagymaroP sroject;n.51, supra.
BCM,para. 147et seq.the main channel that would result in an obscure boundary line. Botswana's repeated
contention that 'both banksof the northem channel are readily discemible, at al1seasons of
the year, and in particular when the island is inundated in the wet ~eason'~'is totally at odds
with reality. It is based on faulty analysis of aerial photographs, al1but one of which were
taken in the dry ~eason.~~The Island, the northem side of which forms the right bank of the
northem channel, is covered by water for a good part of the year. Indeed, the satellite
pictures of Kasikili Island and Namibia's 1998aerial photograph show that both banks of the
northern channel are under water in times of inundati~n.~~ In sharp contrast, the main
channel, as defined by Professor Alexander, has a stable southem bank under the Chobe
Ridge, clearlyvisibleat al1seasons of the year.
D. The Alleged Self-Executing Characterof the Treatv
114. Botswana bases its reading of Article III(2) on an alleged general principle of
international law, the thalweg principle. The argument 4isthat 'in the period to whch the
Anglo-German Agreement belonged, it was normal practice to establish boundaries in
navigable rivers by reference to the middle ofthe navigable channel. . . Theprovisionsofthe
Anglo-German Agreement form part of a Namibia has already dealt with this
question above, and even more extensively in its Counter-Memorial. It has establishedthat
no principle of general international law existed at the time of the Treaty that would mandate
or support a particular reading of Article III(2) of the ~~reement.~~ However, the BCM
introduces one new element in its extendedargument that is addressedhere.
115. The BCM asserts that Article III(2) of the Agreement, incorporating the thalweg
principle (as Botswana contends), determines ipso iure the main channel, because Article
III(2) is a self-executing The term "self-executing" used in relation to a treaty or a
" BCM,para. 331
62SeeChapterIII, para. 164,infra.
63SeeSecondSupp.Rep., Fig. 6; NM, Vol. VI,Sheet 25.
BCM,para. 195.
65NCM,para. 54 et seq.
66 BCM, para. 9. Later, withoutusing the express term "self-executing"Botswanain essence
argues the 'self-executingcharacterof the thalweg conceptas used inArticle III(2) of the 1890
Agreementby saying thata 'river . . has only one thalweg . . . Thus,where the thalweg may be
found,themainchannelmay be foundtoo.' Id., para. 199.treaty provision is generally understood as indicating that a treaty or one of its provisions is
directly applicable in the domestic legal order of the parties without further legislative or
other a~ts.~' Thus, Botswana's use of the term in the present context is at odds with its
ordinary meaning. This misuse of the term represents still another attempt to insinuate that
the use of thethalweg principle by itself settles the question of where the boundary runs in
the Chobe River. Inother words, the use of the term self-executingis afalsa demonstratio
for Botswana's simplistic idea that Article III(2) deteAines the main channel ipso iure.
Botswanaitself reveals the fallacy of this contention. Contradicting its position on the 'self-
executing' character of Article III(2), it puts forward various empirical criteria for
determiningthe main channel: the deepest, the widest and, correctly, that which carries the
most water or has the highest surface water velocity, and is, therefore, the main ~hannel.~~
The self-contradiction in Botswana's argument could not be more obvious. Since in this case
there are two channels and, therefore, two thalwegs, the Treaty provisions cannot be 'self-
executing.' The Court must first choose which one is the main channel and then decide
where the boundary lies within that channel- as Namibia has contended fi-omthe beginning.
Botswana has ultimately been driven to take this same position, belying its own peculiar
proposition of the self-executingcharacterofArticle III(2).
116. The hydrological and geomorphological issues raised by the criteria that Botswana
introduced are dealt with in Chapter III, which discusses the scientificaspects ofthe case and
in the scientific reports in Volume II of this Reply. Some interesting conclusions cm,
however, be drawn fi-omBotswana's discussionofthe thalweg principle.m, Botswana has
failed to show that Namibia's interpretation of Article III(2) of the Agreement contradicts any
existing, or allegedly existing,principle of general international law. Second, Botswana has
ultimately accepted that the determination of the main channel of the ChobeRiver is a matter
of fact and is noipso iure decided by the use of the word thalweg in the Agreement. Since
neitherthe Agreement nor any generalprinciple of international law defines the main channel
''For an authoritativerestatementof theconceptof self-executingtreatiesA.eBleckrnann,'Self-
executing treatyprovisions,'in R. Bernhardt etal., Encyclopediaof Public InternationalLm,
installment7, p. 414 et seq.(1984);also H. Neuhold, W. Humrner and Ch.Schreuer (eds.),
HandbookofInternationaL l aw,3rdedition, Part1,para. 342et seq.(1997). It is significantthata
standardwork like JenningsiWatt(eds.) Oppenheim'sInternational Law does not mentionmany
self-executingtreatiesat all.
BCM,para. 350;seealsoBM,Appendix toChapter VII, First OpinioofProfessorSefe,p. 5.concretely,the location of the main channel must be determined on the basis of the scientific
and other factual evidence - a position that Namibia has maintained from the beginning.
E. Conclusion
117. In this chapter, Namibia has analyzed Article III(2) of the 1890 Treaty in terms of its
wording, object and purpose and in the light of an alleged general principle of international
law mandating that Article III(2), by its use of the term thalweg and its allegedly self-
executing character, determines the main channel of the Chobe River ipso iure. ûn none of
these issues can Botswana prevail.
118. Botswana has abandoned its earlier assertion that the words 'the main channel' in the
English text of Article III(2) are the counterpart of the word 'Thalweg'in the German text of
that article. Botswana now contends instead that the words 'desHauptlaufes 'in the German
text are unnecessary because the correct German translation of the English formula 'centre of
the main channel' would have been 'im Thalweg dieses Flusses.' This contention runs
counter to the rules of treaty interpretation enunciated by this Court and its predecessor, the
Permanent Court of International Justice. According to these judicial pronouncements
'interpretation must be based above al1upon the text of the trea~y'~'and must be canied out
in such a way that reason and meaning are attributed to every word of the te~t.~' ûnly
Namibia's interpretation, attributing a distinct meaning to the words 'of the main channel' in
the English text and their counterpart,'des Hauptlaufes,' ih the German meets this standard.
119. Botswana's charge that Namibia has ignored the object and purpose of Article III(2) is
equally unfounded. In the first place, it cannot be shown that the object and purpose was to
ensure the parties' access to navigable boundary rivers, as Botswana claims. Even if, the
term 'thalweg' was intended 'to affirm the rights of navigation of the two self-appointed
riparian states,' Namibia has shown that the underlying rationale of the thalweg concept was
equal access of both riparians to the uses, navigational and otherwise, of the river. Although
Botswana itself agrees with this analysis,71it insists that the northem channel isthe main
channel, which would render the equal accessprinciple ineffective for Namibia. As has been
69Seen. 9,supra.
'OSeenote 10, supra.
71BCM, para.224.shown,it is the channelto the southof theIslandthat is the mostused channeland, therefore,
the channelof greater importanceto the regional economy. If the northem channel were to
be acceptedas the main channel, Namibia wouldbe denied equal accessto the actualuses of
the ChobeRiver.
120. Botswana's thirdargument, thatArticleIII(2) is 'self-executing'and that theuse of the
term 'thalweg' determines the mainchannel ipso iure, is wrong on two counts. First, it
alleges thatthe 'thalweg'conceptwas a binding generalpnnciple of internationallaw at the
time of the conclusion of the 1890 Treaty, from which the position of the main channel
aroundKasikili Island couldbe normatively deduced. As Namibiahas conclusivelyshown,
there was nosuch generalpnnciple of internationallaw or acceptedstate practiceat thetime.
Second, Article III(2) is not a self-executingprovision. Botswana's useof the term 'self-
executing' is at odds with the ordinary meaning of the term, which relates to the direct
applicabilityof a treaty in domesticlaw. In any case,Botswanadoes not adhere to its own
argument. It espousesa nurnberof factualcritena for detemining which of the two channels
around Kasikili Island is the main channel,belying its own peculiar proposition about the
self-executingcharacterof Article III(2). The Court is thus remitted to an evaluation ofthe
scientificevidenceidentifjmg the main channelof the Chobe River. ChapterIII
INTERPRETATIONOF ARTICLEIII (2) OFTHE 1890TREATY:
REBUTTALTO BOTSWANA'SSCIENTIFICEVIDENCE
121. The previous exchanges of pleadings reveal a considerable area of agreement between
the parties as to the significance of the scientific evidence in this case.oth Namibia and
Botswana accept that the question of the interpretation of the words 'the centre of the main
channel' in the 1890 Treaty is a question of fact to be resolved on the basis of scientific and
empirical evidence. As Botswana stated inthe 'Conclusions' to its Memorial,
First:
The central question is the interpretationand application of the words 'main
channel' of the River Chobe. These words involve a reference to a question of
fact and, in so far asmay be necessary, a questionof scientificfact, calling for
expertise in hydrology, geology andhydrogeomorphology.'
122. Both parties also accept that the critenon for determining the main channel is a factual
one, namely, which channel carries the largest proportion of the flow of the river. The BCM
says, 'The main channel is the channel which carries the greater amount of ~ater.'~
Botswana insists, further, that 'Channelwidth and depth are not alternativesfor detemining
the main channel of a river.'3 Namibia has consistently advanced the same criterion. 'The
main channel of a river is the channel that conveys the largest proportion of the annual flow
ofthe ~iver.'~
123. Botswana would now have the Court believe that the scientific evidence is
'supererogatory,' and 'contrary to comrnon sense," citing the 'unanimity of view' of prior
'BM,p. 129.
BCM,para. 385; see alsoBM,AppendixtoChapter VII, p.5
BCM,Vol. II,App. 4, para.50. The inappropriatenessof depthor width asstand-alonecriteria
fordeterminingvolumeof flowis discussedbrieflyatpara. 167,infra,andat NM,paras. 131, 132,
157;id., MainReport,para. 4.4;NCM,para. 28.
4NM,para. 157;id., MainReport, paras.1.8, 2.8; seealsoNCM, Supp.Rep., paras. 2.1, 5.1.
BCM,paras. 323, 324.observers and what is said to be plainly visible on the aerial photographs.6 As the story of
Gallileo tells us, this is the usual recourse of those who wish to divert attention fkom
uncomfortable or inconvenient scientific fact. The 'unanimity of view,' upon examination,
tums out to involve infiequent opinions of a handful of low-levelbureaucrats scatteredover a
period of almost a century. Moreover, these observers were using the criterion of depth -
which Botswana admits is inappropriate - to determine the main channel, since they made
their inspections during the dry season when the flow in both channels is insignificant.
Likewise,the aerial photographs, on which it is saidtobe obvious that the northem channel is
the main channel, were all, except one, taken in the dry season. Therefore, they can hardly
provide the basis for conclusions as to which channel carries the greater proportion of the
flow.
124. Botswana also tries to deprecate the significance of Namibia's scientific case by a
wholly unwarranted ad hominemattack on the qualifications of Professor ~lexander.~
Professor Alexander's credentials are well known. He is among the most distinguished
hydrologists in Southem Afnca. As his curriculum vitae shows, he has extensive scientific
training in matters relating to hydrology, river hydrology and fluvial geomorphology, the
relevant scientific disciplines involved in this litigation. His practical experience includes
responsibility in the field or at the headquarters level for most major river projects in South
Afnca since 1970. The extensiveness of Professor Alexander's experience is a matter of
public record and must have been known to Botswana, many of whose officials in water
affairs have taken his courses in hydrology at University of Pretoria and worked with him as
a colleague. Namibia calls the Court's attention to the extended review of his career at Reply
Annex 20 and NM, Vol. VI, sec. 17, p. 38. Under the circumstances, Botswana's effort to
denigrateProfessor Alexander's qualifications is ludicrous.
125. In addition, Namibia has asked Professor Keith S. Richards, Chair of the Geography
Department at Cambridge University and one of the world's pre-eminent geomorphologists,
to provide an analysis of the geomorphology of the Chobe River around Kasikili Island.
ProfessorRichards's cMculum vitae appears in Reply Annex 21. As willbe seenbelow, his
analysis fully supports Professor Alexander's identification of the main channel. In addition,
BCM, ChapterVI (sub-heading),p. 121; para.323.
' BCM, para. 260; id.Vol. II, App. 4,paras. 4, 5.using the data fiom Dr. Sefe's sedimentological report, Professor Richards supplies an
account of the geomorphologic origins of the main channel that corroborates Professor
Alexander's conclusions.
126. In fact,Botswana's pretence that 'thescientificevidence is supererogatory,' is belied by
its actions. Its Counter-Memorial devotes Chapters '5 and 6 - over 100 pages or
approximately one-third of the total - to discussion of the scientific case. Accordingly,
SectionA of this Chapter addresses what both parties agree is the central issue: determining
which channel carries the largest proportion of the flow in the Chobe River at Kasikili Island.
It disproves Botswana's contentionthat the northem channel is that channel.
127. Section B shows that, contrary to the BCM's contention, analysis of the aerial
photographs and satellite images does not support the propositionthat the northem channel is
the main channel, but on the contrary, confirmsNamibia's position.
128. Section C refutes Botswana's assertion that the northem channel is the navigable
channel in that it is most used for commercial traffic in the ares.* Section D returns to the
issue of the location of theoundary within the main channel.
A. The Proportion of the Flow of the River Passinp;through the Northem chahel
129. Botswana's arguments in support of its contention that the northem channel carries the
largest proportion of the flow of the Chobe River around Kasikili Island can be grouped
under threeheads:
The repeated assertions that 'the Chobe is a perennial river, independent of the
Zarnbezi River, with a stable profile, continuous downstream flow and visible
stable bank~,'~and that it 'is an independent perennial river with continuous flow
'10
at al1seasons .. . Dr. Sefe adds that '[fJlow is continuous in a downstream
direction alongthe entire course.'" Indeed,three subchapter titles or subheadings
BCM,para. 387.
BCM,para. 457(v)(b).
'OId., para. 263.
" BCM, App. II, para. 49. See also BCM, paras. 272, 331 ('TheChobe is a perennialmature
river with stable, non-erodibleand readilydiscemible banks');para. 363 ('Geomorphological and
hydrological evidence show that the ChobeRiver is an independentmature river with stableand
visiblebanks'). in theBCM contain the assertion that the Chobe is aperennial nver.12
Data fiom the water level gauging stations along the length of the river allegedly
demonstrating and recordingthis continuous downstream flow.I3
Miscellaneous hydrological and other indicators allegedly showing that the
northem channel is deeper and wider and thus carries the greater portion of the
fl0w.l4
In this section of its Reply, Namibia demonstrates that these grounds are al1 without
substance.
1. The ChobeRiver is not averennialriver
130. The BCM asserts categorically no less than 13 times that the Chobe is 'a perennial
river."' The definition adopted by Dr. Sefe is that 'a perennial stream never dries up.'I6
Other cornrnonly accepted definitions are in accord. Wilson and Moore, for example, define
aperennial stream as:
(a)A stream or reach of a stream that flows continuously throughout the year
and whose upper surface generally stands lower .than the water table in the
region adjoining the stream...,(b) a stream that flows continuously al1year
(including wet and dryyears) and has a minimum flow of 40 liters per minute
(USDOA1986,p. 17).17
Almost al1of Botswana's referencesto the Chobe as a perennialriver also stress the element
of continuous flow.18
" BCM,Chap.6, C (iv); G (iiiG; (iv)
l3BCM,pp. 103-104; 143-144i ;d., paras. 272, 284, 366, 38id., Vol. II Appendix2, pp. 29-31.
14
E.g., BCM, para. 310 and photographs, pp. 117and 107(movementof wildlife asindiciaof
depthof channel); paras. 332, 333,369, 377 (cornmentonerosionandsedhnentation);paras. 373-
375 (sinuosityof southem channel);paras383,403andPhotographp. 153(reed growth).
''BCM,paras. 251, 263,272, 275, 330, 331, 334, 351, 366 3,79, 381.457(v)(b). See also id.,
App. II, para. 48.
l6BCM, App.II, para. 9.
l7GlossaryofHydrology, p. 149,Reply Annex27. The Oxford Engli shictionary(1989)Vol. XI,
p.534specifiesthata 'perennial'stream 'flowsthroughal1seasonsoftheyear.'
'8E.g. BCM,paras. 274-275, 334,379, 381.131. The ineluctable fact, however, is that great sketches of the Chobe River over the 300
kilometres in which it constitutes the border between Botswana and Namibia are dry for
many months of the year. This condition has prevailed for at least the1st 15 years and for
long penods over the last thousand years.19 In the dry season, there is virtually no water in
the channel of the Chobe River upstream fiom Serondela, about 15 kilometres west of
Kasikili Island. When there is flow in the Chobe, that is to Saywhen there is overbank flow
fromthe Zarnbezi River,theremay be water in the channel as far upstream as Ngoma Bridge,
about 50 kilometres west of the Island. Beyond that point the river is dry throughout the
year, at least to Lake Liambezi, except for occasional pools. Lake Liambezi, 100kilometres
upstream from the Island,was completely dry this year and every yeaTfor at least the last 15
years. Thomas and Shaw, a source relied on by Botswana, refer to Lake Liambezi as 'an
ephemeral lake.'20 A report by Professor D.C. Midgely on the Lonrho Namibia Caprivi
Sugar Project States that 'it is evident that Liambezi can by no means be classified as a
perennial lake.'21 As ProfessorRichards remarks, 'Since the Linyandi-Chobepasses through
this lake, it is difficult to sustain the argumentthat the Chobe is aperennial river.'22
132. These facts are amplyprovenby direct evidence:
Directpersona1observations of Professor Alexander in April 1997and in 1998.
In April 1997and again during my three visits in 19981flew by helicopter
alongthewhole length of the Chobe River as identified in the 1890Treaty.
1observed and photographed long lengths of .the Chobe River where the
river was completely dry without visible channels or clearly defined
bank~.~~
Photomavhs taken by Professor Alexander and annexed to his Main Report,
Suvvlementarv Report and Second SuvvlementaryRevort. Sheets 9, 10, 11 and 13of the
19
RichardsRep., paras. 14,20.
20D.S.G. Thomasand P.A. Shaw, The KalahariEnvironmen p. 132; see Reply Annex 29. Dr.
Sefecharges that Professor Alexande r 'ignorant'of thiswork(BCM,Vol. II, App. 4, para.5).
But ifDr. Sefehas indeedreadit, howcanheclaimrepeatedlythattheChobeis a perennialriver?
21Reply Annex 26.
22
RichardsRep., para. 16. SeealsoSecondSupp.Rep.,App: 1,Photographs 15and 16.
23SecondSupp. Rep.para. 4.5. See also NCM, Vol.. III, para. 7.9 ('The lower reachof the
Linyandi Riverfrom LakeLiambezi westwards,was completelydry during my inspectionand no
channels werediscemablefromtheair'). Appendices to the Main Report contain 11 photographs of the dry bed of the Chobe
~iver.~~These sheets were a part of Namibia's Memorial and were thus in Botswana's
possession when it made the repeated assertions citedabove about the perennial character
and continuous flow of the Chobe River. The BCM contains not a word of comment on,
or explanationof, these photographs.
The Supplementary Report also contains a panoramicphotograph showingthe dry bed of
the Linyanti River west of Lake Liambezi and the lakebed it~elf.~Photograph P14s,
taken from a helicopteron 30 April 1997,showsthe dry bed of the ChobeRiver in a long
stretchof the border above Lake ~iambezi.~~
Photograph 13 in the Second Supplementary Report shows Professor Alexander's
helicopter landed in the dry bed of the Chobe River upstream of Lake ~iambezi.~~
Photographs 15 and 16 show a person walking on the dry lakebed of Lake Liarnbezi and
the abandoned gauging station that had earlier been used to measure water levels in the
lake.28
Records of the Namibian Department of Water Affairs. In 1985,the Department reported
that 'There has been no flow fiom Lake Liambezi. into the Chobe since September
1982.'~' In the early 1990s,there was no water in the river even further to the west. The
water supply for the village of Chinchimane, which is dependent on the river, failed and
even that of Linyanti village was threatened.30 A June 1992 Joint Report of the
Hydrology Divisions of the Namibian and BotswanaDepartmentsof Water Affairs States
'WithLake Liambezi havingbeen dry for almost a decade, therehas been no contribution
to the Chobe River from this so~rce.'~' And the Attachment prepared by the Botswana
team specifically noted that 'The whole of the Lake Liambezi area was dry at the time of
24NM,MainReport,App. Sheet9a andb; Sheet10a,b, c, d; Sheetlla; Sheet13a, b, c.
NCM, Vol. III, Illustrations,SheetIls, P13s,P14s.
26Supp.Rep.SheetIls.
27
Second Supp. Rep., para. 3.9. Photograph13,p. A117.
28SecondSupp.Rep.,para. 3.11, andalsoid., p. Al/& Photographs15 and16.
29ReplyAmex 24.
30The locationof thesevillagesis shown inFig. 5, followingp. 66, infra, and inSecondSupp.
Rep.,Fig. 2, followingp. 9.
31ReplyAnnex 25. thevisit and it would have been ver=easy to waWdrive straightthrough to ~otswana.'~~
133. The dry riverbed is not a phenomenon of the recent past. L.A. Mackenzie's 1946
Report on theKalahari Expedition states:
The eastem outlet of the Linyanti swaps becomes an insignificant streamnot
to be compared with the river .. .entering the Caprivi Strip in the north and,
during the dry season, becomes quite discontinuous opposite Kachicau. In
fact, during the dry season a trader in that area travels by car across the
Linyanti to KatimaMuliloon the Zambezi without crossingopen ~ater.~~
The 1985Annual Report of the Namibian Department of Water Affairs confirms this:
Itis imrnediately apparent that the period from 1910to 1948was an extremely
dry one when compared to that between 1948 and 1978. It seems possible
therefore that the lake witnessed by Du Toit in 1925 was already close to
drying up and that by the 1930's the majority of the area could be used for
agriculture.
It isknown that water entered [Lake Liambezi] from the Zambezi in 1948,
probablythe firsttime for aminimum of 30 years.34
Professor Richards reports long periods dating back to the end of the 1gthcentury in which
the middle reaches of the Chobe were dry.35 He suggests that similar cyclical changes have
occurred over many centuriescontinuingthroughthe present.36
134. The BCM seeks to explain these long dry sketches of the river by conjuring up an
underground flow that links the pools in the riverbed 'like beads on a string.'37 It produces
no supporting references forthis channing fan ta^ I^ .^ct, as shown in the photographs, for
32SecondSupp.Rep., App.2, p. A2123.
33SeeReplyAnnex32, p. 5.
34ReplyAnnex.24.
35Richards Rep.para. 21.
36Id. para. 13.
37BCM,para. 367. SeealsoBCM,App. 2,para. 52.
38Maidment,citedin the BCMwithouta page reference oran extract in the appendices,is a 1200
page text entitledHandbookof Hydrology. A search of the book discloses Chapter 8, entitled
'Streamflow,'by MosleyandMcKerchar,para. 8.1, whichsaysonlythatonesourceof streamflow
may be 'baseflow(returnflow fromgroundwater).' The other cited sourceis Wilson andMoore,
which was thenunpublished.Again,therewasneither an extractnor a pagecitation.long stretchesof the riverbedthere are no 'pools'; it is bonedry.39In 1993,when the river
driedup at Chinchimane,two boreholes were drilledto providewater to the village,but they
wereinadequateandfailedto solvetheproblem.40
135. The evidence conclusively showsthat the Chobe River is not a perennial river that
'neverdries up' or where water is 'presentat al1seasons.'The following statements,among
manyothersappearingin theBCM, arethus at oddswith the facts(emphasisadded):
The Chobe Riveris an independentperennial river with continuousflow at al1
seasons of the year through the northem channel around KasilcililSedudu
Island,... 41
Flow through the northem channel is continuous in a downstream direction
42
Flow in a downstreamdirection throughthe northem and westem channel is
43
continuous ...
[Tlhe Chobe River is a mature perennial river with continuousdownstream
flo44t the bifurcationof the island throughthe northernand westem channel,
.. .
Flow in a downstreamdirection through the northem and westem channel is
continuous ...45
Thetrue situationisthat duringthe dry season,when theZambeziRiveris confinedwithin its
banks, no significant arnount of water enters the Chobe River below Lake Liambezi and
upstreamofKasikili Island. Sincelittle or no watercomesin, littleorno water flowsthrough
either ofthe two channels. The water in them is as a practical matter stagnant, as Namibia
has consistently maintained. The river does not meet Botswana's or any other generally
accepteddefinitionof aperennial river.
39NM, VolVI,App.Sheet13,Photographsa, b andc; NCM, Vol III,Sheet 1ls, P13sandP14s.
40ReplyAnnex24.
41BCM, para.263.
42Id.,para. 272.
43Id,para.334.
44 Id., para351.
45Id., para.381.136. The significanceof this fact is far deeper than a simple matter of nomenclature. Al1but
one of the aerial photographs on which Botswana relies for comparison of the depth, width
andrate of flow of the two channels were taken in the dry ~eason.~~It follows that al1those
comparisons assertedlyshowing greater flow in the northern channel aremade at times when
there is little or no water flowing through the northem (or for that rnatter the southem)
channel.
2. The dataasto water levelsat aaug -ina stations alongthe Chobe River
137. Table 3 of Dr. Sefe's Second Opinion tabulates the monthly readings at seven water
level gaugingstations along the ChobeRiver over the penod forwhich datawere availableto
him, usually ffom the early 1980sto 1994-1995,but going back to the early 1970s for some
stations. Dr. Sefe's discussion of the data at paras. 47-55 of his Second is
recapitulated in BCM para. 366. Botswana argues that these readings support its contention
of continuous flow in the Chobe River in two ways: first, because the seven stations
registered positive readings in al1of the months on record; and second, because the Kasane
station had a 'minimum' reading of 925.6 metres throughout the penod.48 These two
contentions are discussed separatelybelow. It will be seen that not only does the data fail to
support either branch of Botswana's argument, but their significance is senously
misrepresented in Botswana's presentation.
a. The cornvarisonofthe readinas atthe seven stations
138. Dr. Sefe's report presents the data for each station in tabular form with a final table
entitled 'Annual Surnmary for Each ~tation.'~' The monfhly means are then plotted on two
ch art^,'w^hich are reprinted in the Botswana Counter-Mernorial at pp. 143-144. On this
basis, the BCM concludes:
46 Botswana's analysiosftheaerialphotographsisdiscussedfurther in PartB, below.
47BCM,Vol. II.App. 2.
48See BCM,para. 366.
49
BCM,App. 2,p. 26. Apparentlythefiguresin thissummarytableare themeansof theaverages
for eachmonthin al1reportedyears,but itis hardto SaybecauseDr. Sefedoesnottellus whether
the figures recorded foreach month in the underlying tablesare maxima, minima, averagesor
means. A comparisonwithSecondSupp.Rep.Table 1,p. 43 suggeststhattheyaremeans.
50BCM,App.2, p. 26, Figure7(a)and (b),pp. 30,31. Even during the persisting droughts in the last 15 years the Chobe River has
maintained its characteristics as a perennial river, as evidenced by the data
collected by the Department of Water Affairs of the Botswana Govemment at
the seven [sic] * gauging stations on the Kwando&inyanti/Chobe River south
of the Caprivi Strip at Upper Kwando down to Shaile . ... [A]t the lower 51d,
the Chobe Rivermaintainsa minimum level at Kasane of 925.6 metres; ...
139. In the first place, it is important to understandjust what the figures in Dr. Sefe's tables
represent. They record the water in terms of height above the gauge plate (or zero
point) at the gauging station. The tables contain no flow data whatsoever. Nor can the water
level readings be converted into flow data without calculations involving numerous other
factors. If we were to put gauges in two bathtubs full of water 20 metres apart, both gauges
would give positive readings, although there would be no flow at al1between the bathtubs.
Gauging stations can be and are located in lakes, reservoirs and otherstatic water bodies.
140. The figure shows lines connecting the plotted points for each station, implylng that
there is continuous flowbetween the stations. But there is nothing in the water level readings
that justifies such a conclusion. In fact, we do know fiom other evidence that there is
continuous flow over the first 120kilometres fiom the Upper Kwando station to Shaile, but
thereafter it dries up in the Linyandi (also spelled 'Linyanti') Swarnps. But we also know
fiom the extrinsic evidence reviewed above that most of the 100 kilometres of the River to
the east,between Shaile and Kasane, have been consistently dry overthe past many years and
for longperiods in the more distant past.
141. The Court should also note that in the figures at BCM pp. 143-144, al1seven stations
are represented as equidistant. This, too, isseriously misleading. The stations are shown on
the map atBCM p. 13. It reveals that the first six stationsbetweenUpper Kwando and Shaile
are located at intervals of roughly 20-30kilometres over the total distance of 120kilometres.
But the distance between Shaile and Kasane is 100 kilometres in which no gauging stations
are shown on the chart.
142. In fact, however, there are gauging stations in this reach of the river, and thus the data
provided by Dr. Sefe are seriously incomplete. Figure 2 of Professor Alexander's Second
* Actuallyonly six of the stationsare betweenUpper Kwando and Shaile. The seventhstation
shownonthechartisatKasane.
BCM,para. 366.SupplementaryReport shows two gauging stationsbetween Shaile and Lake Liambezi, one at
Chinchimane and one at Chobe Outflow. (Reproduced as Fig. 5, followingp. 66) They must
have recorded zero level foral1ormost ofthis period. As noted above, the river had dned up
at Chinchimane by 1992.'~ Lake Liambezi was dry fiom 1982on, so the gauging station at
Chobe Outflow must also have registered zero. Photograph 16 in the Second Supplementary
Report shows a water level recorder in Lake Liambezi, abandoned and in ruins because there
was no water to rnea~ure.~~There is at least one additional gauging station between Shaile
and Kasane at Ngoma ~rid~e.~~Figure 7c of the Second Supplementary Report plots the
readings at that station for March and Apnl 1998. It shows readings of zero fiom 1 to 12
March, before the overbank flow fiom the Zambezi River reached the Chobe River. In the
next week, the readings at Ngoma Bridge rose sharply to over 2.5 metres by 18 March. As
Professor Alexander has shown, this configuration is typical for the Chobe River at Ngoma
Bridge. The BCM does not include data fiom any of these stations, for what reason we
cannot say. But if these figures had been shown, they alone would have utterly refüted
Botswana'sbasic position that there is continuous flow alongthe length of the ChobeRiver.
b. The aaune readinnsatthe Kasane station.
143. To be precise, the text of BCM para. 366 does not quite Saythat the gauge station data
show continuous flow overthe entire distancebetween Upper Kwando and Kasane, but it has
to be read very carefblly to discover that. It says that the Chobe 'has maintained its
characteristics as a perennial river as evidenced by the dat. .. [fiom] . . Upper Kwando
downto Shaile.' Then, it continues in a separate sentence! 'atthe lower end, the Chobe River
maintains a minimum level at Kasane of 925.6 metre~.'~~The inference Botswana intends
the reader to draw is that the flow continues fiom Shaile down to Kasane, just as the line
linking the stations on the figures at pp. 143-144 continues fiom Shaile to Kasane. The
52Para. 132,supra.
53See SecondSupp.Rep., p. A1/7,Photographs13 and 14.
54A stationis also shownat Muyako in LakeLiambezi,but it is not on thecourseof the Chobe
River. It toomusthaveregistered zeroduringthisperiod.
55
BCM,para. 366. The gaugereading atKasanehas beenconvertedto an elevationabove sea
level. See also BCM, para. 272; BCM, Vol. II, App. 2 para. 43. The figure 925.6 masl is
evidentlya mistake,sincewith thewater at that level atKasane,mostof KasikiliIslandwould be
underwater. See SecondSupp.Rep. Sec. 12. Botswanaseemsto havemiscalculatedthe figure,
which should be ameterless,or 924.6masl.inference is negated,however,by the evidenceset forth above showing that in the dry season
there is no, or insignificant, water in the riverbed above Serondela, as far at least as Lake
Liambezi.
144. How, then, are the recorded levels at the Kasane station to be explained? Where does
the water at Kasane in the dry season come fiom, if not fi-omflow through the channels
around KasikiliIsland? There is no great mystery. It comes fiom the ZambeziRiver through
the anabranched channels, which enter the Chobe River downstream of Kasikili Island, one
of them above and one below the gauge station at Kasane. Indeed, BCM, discussing the
November 1972aerialphotograph,says:
The curving loops of the Kasai channel are filled with water and clearly
channeling floodwater fiom the ZarnbeziRiver into the ~hobe.'~
The statement must be qualified in one important respect: there were no floodwatersof the
Zambezi River in November 1972. November 1972 was the low water month in the 1972-
1973hydrologicalyear (906 million m3/month).57The sarnechannel appears on a number of
the other aenal photographs, e.g, 1925, 1947, 1981, July 1977,June 1997 and the June 1997
composite photographs. As has already been said, and as will be shown more fully below?*
al1of these photographs (with the possible exception of the June 1997 photographs) were
taken during periods of low water when there was little or no flow in the river at Kasikili
Island. Not only the November 1972photograph,but gJ of these photographs show that 'the
curving loops of the Kasai channel are filled with water and are clearly channeling... water
fiom the Zambezi River into the ~hobe."~ The conclusion is that water flows through the
anabranched channels throughout the year. It is this source, and not any water coming fi-om
above Kasikili Island, that maintainsthe positive readings at the Kasane gauge station during
the dry season. Thus the repeated references to a mininium reading of 925.6 metres at the
Kasane station (besides being inaccurate) provide no support whatsoever for Botswana's
56
BCM, para. 415. This channelisvariously calledKasai,Kassaya, Kassaia.
57SeeRichardsRep., Vol. II, App. 20.
58Seepara. 163 etseq., infra.
59BCM,para. 415.proposition that '[fllow in a downstream direction through the northem and western channels
is continuous....60
Other assertedindicators of greater flowin the northem channel
145. Throughout the BCM there arereferences to depth,width, slope, erosion and deposition
of sediment,wildlife and flow measurements al1purporting to show that the northem channel
carries the larger proportion of the flow in the river. Some are assertedly supported by
photographs or measurements. Others, like the flow measurements, do not give any hint as to
their provenance or derivation. In this section, Namibia shows that none of this miscellany
can alterthe conclusionthat the northernchannel is not the main channel.
(a) The 'capacitv'of the channels
146. The comparative width and depth of the two channels without more cannot establish
which channel carries the greater amount of the water in the river. The BCM itself says that
'[fllow in a channel is the product of cross-section area (a product of the width and depth of
the channel) and the mean velocity through the cross-~ection.'~' Botswana, however, relies
on the notion of the 'capacity' of the channel, whch does relate to width and depth, as the
indicator of the channel that actually carries the larger fiow. For example, it says, '[tlhe
northem channel carries the larger flow because it has greater ~a~acit~.'~~This statement is a
complete non sequitur. It amounts to saying that the larger of two pipes must necessarily
carry more water. But which will carry more water depends on the amount of water coming
in at the entrance to the pipe. If little or none comes in, little or none will pass through, no
matter how great the capacity of the pipe.
147. Some of Botswana's evidenceof depth borders on the fiivolous. Photographs displayed
of hippopotami and an elephant in the southem channel are designed to illustrate its
inadequate depth.63 Although it is said that the elephant is standing in the middle of the
southem ~hannel,~~it is obvious in the photograph that the animal is very close to the
BCM,para. 334. Seealso quotationsfromBCM,para. 135,supra.
BCM,para. 350;seealsoBCM, Vol. II,App.4, para. 50.
62BCM,para. 350.SeealsoBCM,paras. 314, 346, 400,457(v)(c);BM,App. toChapter VI1 p. 5.
63See BCM,para. 312; photographsare at BCM,pp. 107, 117.
64Id.,para. 279.southem bank of the channel. If the Court wishes to see elephants in mid-channel, north and
south, it should consult Professor Alexander's photographs in Fig. 13 of the Second
Supplementary ~e~01-t.~'We are told that hippopotami do not fiequent the northem channel
because of its excessive depth.66 But Sheet 16s in Professor Alexander's Supplementary
Report is a picture of a hippo happily sunning on the bank of the northem channel. We are
told that buffalo cross to the Island by the southem channel because of 'the shallowness of
the southem ~hannel.'~~But since the buffalo are coming fiom the ChobeNational Park, it is
hardto seehow else they would get to the Island but acrossthe southem channel.
(b) Slope.
148. In connection with its discussion of the thalweg,Botswana argues that the northem
channel carries the greater flow because the slope of the northem channel is steeper than the
southem ~hannel.~'This portion of the BCM recapitulatesthe argument in Dr. Sefe's Second
Opinion that the bed of the northem channel has 'a steeper gradient' than the 'relatively flat
gradient' ofthe bed of the southem ~hannel.~'The BCM reprints Dr. Sefe's Figure9 at BCM
p. 129.
149. Inthe first place, as Professor Alexander shows, the critical slope for determining water
velocities is not bedslope, but the water surface slope, that is, the difference in the height of
the water surface at the entrance and exit fiom the ~hannel.~' Ordinarily, the two figures are
in reasonable accord, but this does not hold true for caseswhere there is downstream control
of flow, such as exercised by the Mambova Rapids for the Chobe River around Kasikili
~sland.~'
150. More important, the argument, and especially the chart, egregiously misrepresent the
comparative slope of the two channels. The points plotted on the chart are the depth
measurements taken by the 1985joint survey at cross sections in the northem and southem
65SecondSupp.Rep., followingp. 39; seealsoparas. 11.2-11.3.
66BCM,paras. 310, 311.
67BCM,para. 310.
68BCM,para. 300.
69BCM,Vol. II, App. 2,para. 61.
'OSecondSupp.Rep., para. 10.4.
71Id.; seealsoRichards.Rep.para. 18.channels moving in a down stream direction beginning with a common origin at the
bifurcation to the west of the Island (cross section #1). The depth measurements taken in the
northem channel areplotted along the green line in the chart, and those taken in the southem
channel areplotted on the red line. The chart purports to show that the slope of the southem
channel is flatter than the northem channel, because the difference in elevation between the
beginning and end of the red line (about two metres) is smaller than the difference between
the beginning and end of the green line (about four metres). But the chart is incomplete. No
depth is plotted for the confluence of the two channels to the east of the Island. Since both
channels pass through this point, however, the depths for both will be equal at that point.72
As must be obvious, since the points of origin and the points of confluence for each channel
are identical, the difference in elevation between the beginning and end of each channel is
alsoidentical.
151. Finally, the bedslope of a channel, as a factor affecting velocity,pends not merely on
the difference in elevationbetween two points, but on the horizontal distance as well. A set
of rapids that drops ten metres over a distance of a few metres has a bedslope greater than a
quiet stream that drops the same amount while flowing for 20 kilometres across a plain.
Since, as Botswana admits, the length of the two channels is 'comparable'73 and the
differencein elevation is identical,the bedslope of the two channelsmust be substantially the
same.
(c) The sedimentolonical study
152. The BCM repeatedlyremarks on the stabilityofthe banks and absence of erosion in the
northem channel and cites this as evidencethat it carries the preponderance of the fl~w.~~ It
relies on Dr. Sefe's Sedimentological Study for its assertion that the banks of the northem
72
A true plot of the 1985 depth measurements,includingcross section #9,just below the
confluence,is giveninSecondSupp. Rep.,Fig. 11 following p.33. See alsoid. Sec.9.
73BCM, para. 300.
74BCM, paras. 332 (and subtitle), 305, 307, 333,368. 369.Para. 369calls attentionto a line of
trees in the photograph atp. 147 lookingtowardKasika. The trees are said to be 'onthe right
hand(Island)bankof thenorthernchannel'andthusto illustrateits 'stable non-eroidible' character.
Thetreesarenot, however,onthe island,but onthe leftbankofthe spur channelin Narnibia. Al1
theaerialphotographsshowthis lineof treesat Kasika. Nophotographfromthe air or groundhas
evershown sucha lineof treesinthispositionontheisland.channel are 'stable' and 'non-er~dible.'~~Absence of eosion, of course, cannot prove the
existence of flow in the channel. Nor does stability prove the existence of flow. The
photographs of the cut-off meanders on the Mashi reach of the Chobe River show highly
stable,well-definedchannels. But there is little or no flow goingthrough them.76
153. The only issue is whether, as Professor Alexander has stated, absence of erosion in the
northem channel is evidence of absence of flow. Botswana's insistence that the banks are
stable and non-erodible is designed to refùte this claim, for, in logic, if the banks are not in
fact erodible,then the absence of erosion would not demonstrate the absence of flow. The
issuereduces to a question of fact: are thebanks ofthe northem channelmade up of erodible
ornon-erodible material?
154. The two experts differ on this point. Professor Alexander says that the banks of the
northem channel are made up of erodible mate ria^.^^ The BCM, citing Dr. Sefe's
Sedimentological Report,says that 'the top layer of 1.5metres in depth constitutes, and has
done for thousands of years, material for stable non-erodible banks. These are plainly visible
and indicate the stability andmaturity of the Chobe system in the vicinity of the ~sland.'~~
The study,it says, showsthat:
KasikiliISedudu Island consists of two layers, a dark top layer consisting of a
clay, silt and mud admixture extending to about 1.5 metres in depth, and a
bottom layerof sand, sampled fiom 1.50to 5.00metresin depth (Table 2).79
TheBCM, however, remarkson 'the complete absence of sand in this top layer.'80
155. These assertions are mistaken. The composition of the banks of the northem channel
can be readily seen in Photograph 16s of Professor Alexander's Supplementary Report
showing a hippopotamus sunning himself on the sandy right (Island) bank of the northem
75BCM, para. 304; see alsoid., 369.
76SeeSecondSupp.Rep., p. A1/4, Photographs 7and 8.
77NM,Vol. VI,para. 9.9;seealsoNCM,Vol. IIIpara. 5.3.
78BCM,para. 304. Seealsoid. para. 305;BCM,Vol. II,App. III,para. 16.
''BCM,para. 353(i).
Id. para. 353(iv). Id. para. 353(iii)refers to a top layer of "&, clay and mud," ratherthan
"&, siltandmud" (emphasis added) as inpara. 353(i),just fivelinesearlier.Thelater reference
seemsto be a mistake, since in the nextsub-paragraph,353(iv),the BCM notes "the complete
absenceof sandinthistoplayer."channel. 'The material in the bank is underlain by readily erodible fine, white sand.'"
Photograph PlOs on Sheet 9s shows water fiowing through gaps in the right bank of the
northem channel, indicating clearlythat the bank is erodible. Photograph 17, Sheet 24 of the
Appendixto the Main Report also shows a patch of the sandy soil forming the Island bank of
the northern channel. It should be noted that Dr. Sefe took no samples within the river
channelsthemselves. Nor are there any sarnples fiom the natural levee along the right bank
of the northem channel, where the photographs and Professor Alexander's visual observation
showthat the material is fine,readily erodible and.^^
156. The BCM itself is of two minds on the issue. At this point in its pleading, perhaps for
the sake of consistency,the BCM states that '[slediment bars located at the bifurcation point
of the northem and southem channel . . . are relict bars of earlier downstream flowing
channels, not active zones of ~edimentation.'~~But 46 pages later it says, '[tlhe deposit of
sedimentin the western sector ofthe northem channel marked B indicates the zone of highest
current velocity.'" And again, 30 pages Mer on it says 'the south-western corner of the
island is a dry sandy soil .. .85 The colour version of Botswana's June 1997 photograph
testifies to the accuracy of the last two descriptions rather than the first. (See Fig. 6,
following p. 72) The southwestern corner of the Island shows up as yellow coloured sand,
and the yellow colourpersists in a narrower path alongthe right bank ofthe northern channel,
the levee referred to by ProfessorAlexander. The same configuration can be seen in the high
ground on the contour map, Figure 20 of the Second Supplrnentary~e~ort.~~As Professor
Alexander stated, there are no bore holes in this high groukd,which means that Dr. Sefe took
no coresthere. Thus, the conclusions of the SedimentologicalStudydo not apply to that area.
157. The visual evidence of the photographs both £romthe ground and fiom the air resolves
the issue in favour of ProfessorAlexander's conclusionthat the banks of the northern channel
81NCM, Supp.Rep., Sheet16s.
82Id.,para. 8.5(b).
83BCM, para.307; seealso id.,para. 377.
84Id.,para 397.
Id.,para. 430.
86SeeSecondSupp.Rep., followingp. 57.are erodible. Therefore,the absence of erosion,which Botswana repeatedly proclaims, attests
to the absence of flow in thenorthem channel.
(d) Asserted flowmeasurements
158. Dr. Sefe's Second Opinion contains a table of 'recent flow measurements' in which it
appears that the flow is larger in the northem than in the southem channel on three dates in
March and April 1997.'~ These measurementsare repeated at BCMpara. 285. The dates of
the measurements are given, but not the place on the river where they were taken nor the
methodsemployed. Without suchinformation, the measurementscannot be evaluated.
159. In three different paragraphs, the BCM refers to minimum flow velocities in the dry
season 'on the order of 0.5 metres per se~ond.'~' None of these paragraphs gives any
references or explanations whatsoever. Neither Namibia nor the Court can have any idea
where, when or how they were arrived at.89
160. Until such information is available, these purported flow measurements must be
consignedto the growingcollection of unsupported assertions by Botswana.
B. AerialPhotoa-aphv and SatelliteImages
161. Botswana attaches extraordinary importance to the aerial photography, and to a lesser
extent to satellite imagery, in supportingits contention that the northern channel is the main
channel of the ChobeRiver at Kasikili Island. It devotes 40pages of its Counter-Memorialto
this subject (pp. 155-196). Although Botswana's comrnents on the photography touch on
many of the themes discussed elsewhere in the BCM and in this Chapter, it seems desirable,
evenat the expense of somerepetition,to respondto this sectionofthe BCM as a whole.
s7BCM,App.2, Table7, p. 65.
BCM,paras.275, 283, 368.
ProfessorRichardshasmadecomputationsusingBotswana'sfiguresshowingthatthetwo setsof
measurementsare inconsistent. Applying theaverage cross-sectional area fo the northern channel
given by Botswanato the minimum flow inthechannel reportedby Dr. Sefe givesa minimumflow .'
velocityof .02m3s/l,more than an orderof magnitudelessthanthe0.5m3s/lclaimedin theBCM.
ProfessorRichardssaysthe computedfigureis 'probablynear the operational limitof the current
metres employed' (Richards Rep., para. 31). That is, slower velocitiescould not be measured
usingavailable equipment.The conclusionis, as ProfessorAlexanderStates,thatthe channelsare
essentially stagnanturingthelowflowseason,SecondSupp.Rep.para. 7.13. 1. General comments
162. Much of Botswana's concem seemsto be to show that the general configuration of the
areahas not changed from 1925,when the first photograph was taken, to the present.90 These
repetitivestatements reflect the persistence of Botswana's misconception that Namibia's case
relies upon a change in the position or functions of the ihannels since the 1890 Treaty was
signed. Although Namibia stated unequivocallyin its Memorial that it was making no such
claim:' the BCM persists in arguing the matter. Al1of this argument,however, is beside the
point, since it is addressed to an issueas to whichthere is no dispute between the parties.
163. The Court should also note that al1of the aerial photographsin the Botswana Memorial
and in the BCM (with the possible exceptionof the June 1997photograph) were made when
no substantial flow was taking place along the Chobe River.92 Table 1, p. 43 in Professor
Alexander's Second Supplementary Report demonstrates from actual water level readings
that substantialportions of the Island are under water during the annual high water periods,
the only time when substantial flow takes place in the Chobe Aerial photographs
that do not show significant inundation were not, therefofe, taken when there was substantial
flow in the Chobe River at Kasikili Island. Thus, inferences fiom them about the situation
when there is substantial flowmust be treatedwith utrnost caution.
The BCM begins its discussionof the photographswith a section entitled 'Therehasbeenno
changeoverseventyyears. 'BCM,p. 156. For commentsabout individual photographs stressing
the absenceof change in the general configuration of the area, see BCM, paras. 391,392, 393,
403, 405, 406,417, 427, 441(3).The photographs arelistedin BCM,para. 390, n.8. Sincethe
filingof the counter-memorials,Narnibia hastakenanotherset of aerial photographsand has also
taken photographs froma helicopter. These are discussed atlength in Professor Alexander's
SecondSupp. Rep., Section 6, paras. 14.24 and 14.25. See also SecondSupp. Rep.paras. 1.3,
4.5.
91NM,para. 130.
92 This is consistentwith thepurposes for which these photographswere taken. They were
primarily photographstaken by reconnaissance, surveyand mapping expeditions that were
. interestedin collecting informationaboutterrain and topographicfeatures thatwould be invisible
whenthefloodplainis inundatedduringthehigh flowseason. SecondSupp. Rep.para. 13.19.
93 Namibia has consistentlymaintainedthis position. See, e.g., NM, paras. 21, 73 (citing
Bradshaw),131, 142,220; id., Vol. VI paras. 10.2, 12.4;NCM, paras. 10, 32, 150;id., Vol. III
paras. 4.15 and 5.7-5.8; Second Supp. Rep. paras. 13.1, 13.7. Botswana attacks Professor
Alexander'sidentificationof the maincharnel on the basis of the configuration'as shownon al1
aerial photographs,'BCM,para. 255. Butwhatis shownon al1the photographsis the dry season
condition,whenthereissubstantiallyno flowineitherchannel.164. Finally, it is well to remember that the interpretation of aerial photographs is a
professional activityrequiring expertise, ski11and experience in high degree. The BCM cites
no such expert authority for its comments on the photographs. The impressions of persons
without specialtraining based on an inspection of thephotographs is likely to lead to frequent
mistakes and misinterpretati~ns.~~Only a few instancesneed be given as examples:
Frequent references to the southern channel as 'dried out' or 'completely dry' or
blocked by reeds are without sub~tance.~' Professor Alexander's Supplementary
Report demonstrated, on the basis of the recorded water levels at Kasane, that at no
time in the past 15 years has the southern channel been dry.96 This is illustrated in
Figure 14 of the Second Supplementary Report (following p. 42), showing that the
minimum water level above sea level for the period 1984to 1996was always above
the altitude of the bed of the southernchannelas shown in the 1985joint survey.
A particularly flagrant example is the assertion tbat in the 'May 1972' photograph
'the southern channel is dry for half its length.'97 This is quite impossible as can be
seen by comparing the November 1972photograph. November was the lowest flow
month in the 1971-72 hydrologie year, but the photograph shows plenty of water in
the southem channel. In fact, the 'May 1972'photographshows the southern channel
blocked by an infestation of Kariba ~eed.~~In the early 1970s,the Chobe River and
94
See R.H. Arnold, InterpretationofAirphotosandRemotelySensedImagery (PrenticeHall 1997),
Reply Annex 35. See also Manual of PhotographicInterpretation, American Societyof
Photogrammetryp. 114, Reply Annex36. Obtainingreliablemeasurementsand interpretations
from aerialphotographsand remotelysensedimagesrequiresa thoroughunderstandingof optics
(includingstereoscopy)andthe mathematics of opticaldistortion. Moreover,theAmerican Society
of Photographgrammetry cautions tha atphotographor imageis butone pieceof information,and
reliableinterpretationofspecificfeaturesalrnost alwayrequires othertypesofcorroboration.
[an almost every jobof interpretation therewill be unknownsor uncertain conclusions
which mustbe checkedin the field. Theinterpretermustaccepttheresponsibilityof field
checking wheneverit is feasible, in order to make sure his work is right, or, if it is
wrong, to fmd out why. . .. The amountof fieldwork whichwill be necessaryvaries
withthe intensivenessandaccuracy requiremeno tf the study,the complexityof the area,
thequalityofthephotographs,and theabilityof theinterpreter'.
95BM,paras. 36, 202,218, 220;BCM,paras. 279, 380, 383,396,403, 411,413.
96NCM,Vol. III,paras. 12.5, 12.6.
''BCM,para. 411. The 'May 1972' photographis erroneouslydated, thoughnot by Botswana.
Seediscussion atparas, 187-188infraandinSecond Supp. Repp .aras. 12.3-12.5.
98The Court shouldnote, that despiteBotswana'srepeated assertionsto the contrary(see BCM, others in the area were beset by this pest, and extermination of these weeds became
the object of joint action by the Botswana and South African governments.99 The
success of the program is shown by the fact that the weed growth does not appear on
any subsequentphotograph.
2. Botswana's photographie cornparisons
165. Botswana does not pretend that it is possible to directly observe flowing water in a
stream in aerial photographs. Instead its argument is based on other features that are visible
on the photographs, and that it contends provide the basis for the inference that the largest
proportion of the flow goes through the northem channel. There is no need to track the
BCM's discussion of the photographs one by one, since the cornments are for the most part
repetitive and fa11into anumber of categories, as follows:
The comparativewidth ofthe two channels
The comparative depth of the two channels
The sediment bars at the opening of the main channel as defined by Professor
Alexander
Other indicia ofthe presence or absenceof erosion in the two channels
Striationpatterns
The so-called sub-channel drainingthe southemportion of the Island
These will be discussed seriatim, followed by comments on certain especially serious errors
in the BCM treatment ofparticular photographs.
paras. 383, 403 and 412,n.9), this is the only photograph showing the channe bllocked by
vegetation.
99
See NCM, SuppRep.para. 10.5,andid., App.A, B andC. TheBCMadmitsin a footnotethat
a 'possiblealternative interpretatifor the appearanceof the souîhernchannel isinfestationby
Kariba weed.BCM, para. 412,n.9. Its reasonsfor discarding îhis explanation are insubstantial.
The general questionof whether reed growth affectsthe navigabilityof the southernchannelis
discussedatparas. 201-202,infra. (a) The comvarativewidth ofthe two channels
166. The BCM repeatedly states that the photographs show the northem channel to be the
'wider' or 'larger' of the two.loOThese judgments are quite subjective and contestable as a
rnatterof fact.lO'But extended discussion of these width comparisons is unnecessarybecause
they are essentially irrelevant. In the first place, the width of a channel without more can
provideno information aboutthe m. Botswana's own expert, Dr. Sefe, admits that channel
width and depth are not altematives for detemining the main channel of a river,'" andthe
volume of flow through a channel depends on a complicated formula involving many factors
other than width.'03Thus, no rnatterhow many times Botswanarepeats it,lo4the fact (if it be a
fact) that the northem channel appears to be wider on some of the photographs provides no
evidencethat the volume of flow in thenorthem channel iSgreater.
167. There is a second and more fundamental reason why Botswana's comparisons of the
widths ofthe two channelsas they appear in thephotographsare irrelevant. Inthe dryseason
when the photographs were taken, there is little or no flow proceeding through the channels
around Kasikili Island, and both channels are essentially stagnant.lo5 It hardly seems
plausible that comparisons of the widths of the two channels at a time when they were g&
flowing would provide much information about which of them carries the most water when
they flowing. If comparative width has any bearing on the issue, the compatison must be
made when the river is flowing. The northern channel more or less as it appears in the
photographs must be compared with the main channel as identified in Figures 16-19 of
Professor Alexander's Second Supplementary Report, which carries the main bulk of the
lrxBCM,paras. 258(ii),389, 395, 400, 410, 417,419, 423, 440, 441.1. See also BCM,paras.
447,451,454, 455 and 456 for analogouscomment5based onthesatelliteimages.
10Seediscussion in SecondSupp.Rep., para. 13.9. BCM,para. 395, statesthattheaveragewidth
of the northern channelis 200metres ascomparedto 50 metres or less forthe southern channel.
No supportor referenceis given forthesefigures,and in the absenceof an evidentiarybasisthey
mustbedisregardedas anotherBotswanaips elxit.
'OBCM,Vol. II, App. 4, para.50.
'O3BCM,App. 2,paras. 25-26. Seealsoid. paras. 27-28, discussingtheManningformula. In the
caseof theChobeRiverat Kasikili Islandt,he problemis complicatedeven furtherby a numberof
other factors,includingthe downstreamcontrol exercisedby the MambovaRapidsand theinflow
fromtheanabranchedcharnelsbelowtheIsland.
'O4BCM,paras. 389, 410,441(1), 447,451,454, 455, 456.
'O5SecondSupp. Rep.paras. 4.5, 10.2, 14.5.flow of the Chobe River. That comparison shows that the main channel is the wider of the
two,notjust marginally,but by severaltimes.
(b) The comparativedepth of thetwo channels
168. As with channel width, channel depth is not by itself a criterion for distinguishing the
main channel and provides no information about which channel carries the larger proportion
of the flow.lo6As noted above, Dr. Sefe, Botswana's expert,agrees. The fiequent references
to photographieor satelliteimagesof water 'flowing'in thenorthem channelare mistaken.lo7
Neither of these media, at the scales used, can distinguish between standing and flowing
water. For example, the spur channel looks the same as the others in the photographs, but it
does not convey any water, since it is blocked at the upstream end.''* As to the satellite
images, the BCM's principal claim is that the northem channel is 'wider and therefore likely
deeper' than the so~them,'~~an obvious non-sequitur. (emphasis added) There are many
wide and shallow nvers and many narrow and deep ones. The satellite images are at
1:100,000for the MSS images and 1:50,000 for the TM images. This is far coarser than the
image depicted in Figure 1 (following p. 23, above). As the Court will observe, the
comparative width of the two channels is strikingly different,on this Figure than on the
coarser, and less accurate,satellite images in Dr. Sefe'sFigures 14and 15."' The BCM says
'O6AShas beenfrequentlynoted, Namibiadoesnotacceptthatdepthis an appropriatecriterionfor
determiningthe maincharnel, eitheras a stand-alonedimensionor asa surrogateforproportionof
theflow. See,for example,NMparas. 131,132,157;NMVol.VI, para. 4.4; andNCMpara. 28.
'O7See,e.g., BCM,paras. 400, 411,451.
'OsThere are few direct references to depth of the channels inthe BCM's discussionof the
photographs. BCMparas. 403and411 assertthatthe southernchannel isshallower.The evidence
is notthe photographs, however,but measurementsfrom the 1985 jointsurvey. Elsewhere,the
BCM derives lack of depth fromthe assertedpresenceof reeds in the southern channel,(e.g.,
BCM,paras. 403, 414). The unreliabilityof these readingsis suggestedby Botswana'serror in
interpretingthe 'May 1972' photograph,the one unambiguousdepiction of vegetation in the
southernchannel, as a dry riverbed. (BCM,para. 411; see also para. 164, supra) Moreover,as
Professor Alexander pointsout, much of the reed growth in the vicinity of Kasikili Islandis
papyrus,whichfloatson the surfaceof the waterandthusgivesno indicationof itsdepth. (Second
Supp.Rep., para. 6.12) Thedifference, ofcourse,cannotbedistinguishedonthephotographs.
'O9BCM,para. 454.
110 BCM Vol. II, App. 2, pp. 45, 46. The contourmap in Professor Alexander'sSecond
SupplementaryReport(Fig. 20, followingp. 57), whichis alsomuchlarger in scaleandtherefore
muchmore accuratethan the satellite images,shows thesouthern channel as distinctlw yiderthan
thenorthern.the greater depth of the northem channel can be observed on Figure 15 in Dr. Sefe's
inio io necause it appears in a deeper shade of bl~e."~ Namibia has been unable to
perceive this difference and invites the Court to try for itself. Like the satellite images, the
photographs were al1taken at a scale smaller than 1:10,000.~~'The Court will appreciate the
difficulty if not impossibility of making accurate interpretations of differences in depth of at
most a metre or two on such small-scale depictions.
(c) The sedimentbars at the ovenina of themain channel
169. The BCM states that the June 1997 aenal photograph 'clearly reveals these same
sedimentbars [atpoints B and Cl in essentiallythe same size, shape and location, as on the
1925photograph and indeed on al1the interveningphotographs in the series.' This is said to
be proof that they 'are in no way active' and that the southem channel is not an active
channel.l l4
170. The assertion is incorrect. The changes are apparent on the five close-up extracts of the
area juxtaposed in Diagram 9, Sheet 27 of the Appendices to Professor Alexander's Main
Report. The changes are obvious even to Botswana. In reply, it tries to attribute these
changes to differences 'in the depth of water . . which has the effect of changing the
appearanceof the sand bars in the area of bif~rcation.'"~ but in photographs taken in the dry
season, at low water, the variation in the depth of the water can hardly have been
~i~nificantl6
"' Id.p.46.
IlBCM,para. 446.
113
BCMpara. 390 statesthat 'Al1the photographshave beenenlargedto an approximatescale of
1:10,000exceptfor the 1977photograph.' (emphasisadded)
Il4BCM,paras. 427,428.
'15BCM,para. 377.
116
SecondSupp. Rep. Table 1, p. 43, shows thatthe minimumwater levels at KasikiiiIsland
between1983 and1996 variedby onlyabout one-tenthof a metre. Liewise, an inspectionof the
tablesof daily waterlevels atsanefor the sameyearsshowsa variationduringthedry seasonof
only a fewtenthsof a metre. Botswanaclaims thatthe water is "visiblyhigher" on the1981and
1985photographs. The 1985photographwas taken in August, in which the mean daily level at
Kasanewas924.75masl,the fiftlowestoftheyear, andonly .O5metresabovethe minimum.See
Id. The 1981 photographdoes not reveal the monthin which it wastaken, but if anything,the
islandseemsdrierthaninthe 1985photograph.171. Moreover, unlike the bars at point B, the sediment bars at point C are in the
southemchannel,but some metres upstream in what is clearly the main channel of the Chobe
River before it reaches the bifurcation. Furthemore, the aerial photomosaic of June 1997,
BCM p. 191, shows another set of similar sediment bars still further up the river in the main
channel of the Chobe River at Point D. (See Fig. 7, followingp. 80) These bars like the bars
at C are unquestionably in the main channel of the Chobe River. As Professor Alexander
says, they are:
created by sediment deposited in the river channel by the flow in the river.
There can be no doubt about this as they are characteristic of al1nvers flowing
through unconsolidatedmaterial. There is no alternative e~~lanation."~
If these sediment bars are the products of the flow of the river, it must also be true that the
sediment bars at B, which are similar in every respect, are the product of sediment deposited
by the flowingriver in the southem channel. 'Thereis no altemative explanation.' Thus, the
photographs, far fiom contradicting Professor Alexander's theory, in fact provide strong
substantiation for his conclusion that the channel to the south of the Island is the 'active'
channel, and thus the main channel.
(d) Otherindicia of the presence or absence of erosion in the two channels
172. The question of erosion in the comparison of the two channels has been discussed
above in connection with Dr. Sefe's Sedirnentological ~tud~."' In relation to the
photographs also, the BCM makes fi-equentreference to erosion or sediment deposition or the
lack of it to bolster its arguments to lack of flow in the southem channel. On the one hand,
the absence of erosion in the northem channel is cited as evidencethat it is the main channel
because it is a stable ~hannel."~ On the other, the BCM often comments on the lack of
erosion in the southem channel, particularly on the sediment spit marked A, as evidence of
absence of flow in the southern ~hannel.'~~
Il7SecondSupp.Rep., para. 13.9.
118Seeparas. 152-157,supra.
Il9E.g.,BCM,paras. 304,332, 439.
BCM,paras. 277, 372, 392,403, 405, 438. Seealsoparas.410, 441(3). Despitetheserepeated
referencesto absenceof erosion as disqualiQingthe southern channel asthe main channel, the
sectionheadingat BCM,p. 114characterizesthe ideathat themainchannelmustshowerosionas a
'misrepresentation.rune19W aertphotwfaph rnrvjakChobe Riyerwitlabels
ngu~ 7
June 1897aerfaIphotdmrih mmst utthe ~hobaRlverwkh labels
[tiaseon the phko-m&nic wltli the same tlonepage191of the
t3otmna Countcr-Mernorial) i. As to the stabilitvof the confimirationof the northem channel
173. According to Professor Alexander, 'thesecharacteristics identify it as being a meander
loop that is in the process of being cut off.'12' Photographs 7 and 8 in the Second
Supplementary Report show cut-off meanders in the Mashi portion of the Chobe River. In
both photographs, the meander loop has much more sharply defined banks and even seems
wider than the channel cutting across the base of the ox bow. Yet it is clear that the channel
that runs across the neck is the main channel, and not the one that goes around the loop. In
his Main Report, Professor Alexander showed that the sediment deposits building up at the
entrance to the northem channel are similar to the changes that take place during the
development of cut off meander 100~s.'~P~rofessorRichards concurs:
Indue course, the western reach of the northem channel will be blocked at its
upstream end, in the same way as the spur channel. The northem and eastem
sections of the "northem channel" receive additional tributary flow fiom the
floodplain channels which cany Zambezi flood waters, to the north-east of
Kasika. This extra discharge . ..has maintained the depth of the northern
channel, discouraging sedimentation. Without this influence, the cut-off
meander bend would have also filled progressively with sediment at its
downstream end.123
Indeed, as discussed in Chapter1,paras. 60-62, in his view, the southem channel was created
when the river broke through the neck of the meander loop that is now the northern channel, a
millennium or more ago.'24
ii. As to the absence of erosion in the southem channel
174. (1) The spit marked A: again Botswana erroneously applies its observation to the
stagnant southern channel in the dry season rather than to the main channel as identified by
Professor Alexander. During the season when the River is in flow, the main channel, as
Professor Alexander and Professor Richards show, flows Wr the southem channel rather
than throud it. Erosionthat might be expected in a channel looping around the sediment spit
would not occur as a result of the main channel flowing over it. The spit marked A is where
the sediment bars (discussed in paras. 169-171, above) are located. There, as was shown
121
SecondSupp. Rep., paras. 14.22, 17.2.
Iz2NM, Vol.III Sheet 27.
'2RichardsRep., para. 30.
'24RichardsRep., para.29.above, the four sediment bars have in fact changed in shape, though not in their general
location and orientation,as a result of the flow over and through them of the main channel of
the ChobeRiver.
175. (2)Point B: the Botswana takes contradictorypositions as to the situationat the point of
bifurcation of the channels. BCM para. 371 states:
The Sedimentological Reportfurther confirms the absence of any fiesh
depositionof sediment on thewestern areaofthe island.
This absence of fi-eshsediment, says Botswana, 'totally contradicts Professor Alexander's
identification... of Area d on the island as the "main channel" of the southern channel and
hence a zone of active ~edimentation."~~
176. Yet only a few pages later inpara. 397the BCM contradictsitself:
The deposit of sediment in the western sector of the northern channel at the
point marked B indicatesthe zone of highest current ve10city.l~~
And againin para. 401 it says:
The largedeposit of sediment accreted to the left bank of the northern channel
at point B indicatesthe zone of the highestcurrent velocity.
The question as to which statement is correct is settled by Photograph E3 in Figure 13 of the
Second Supplementary Report (followingp. 39). It showsan elephant in the 'western areaof
the island' and not far from point B dusting itself with the fine sediment that comprises the
surface ofthis area of the Island, acrosswhichthe ChobeRiver annually flows.12'
177. If, as BCM para. 371 maintains, the absence of sedimentation would contradict
Professor Alexander's identification, the presence of sedimentationmust confirm it. What
the BCM in aras. 397 and 401 correctly calls 'the zone of highest current velocitv' is
ssuarely in the middle of themain channel as ProfessorAlexanderidentifies it.12*TheBCM
'''BCM,para. 371.
'26BCM,para. 397.
127Thephotograph also refutes Dr. Sefe'sstatementthatthe surfaceof the islandconsistsof mud,
clayandsilt.
12'SecondSupp.Rep., paras. 16.3-16.5andFig. 21; followingp. 58.has thus unwittingly substantiatedProfessorAlexander'scentral conclusionas to the location
ofthe mainchannel.
(e) StriationDattems
178. Botswana professesto be unable to see the patterns of striationor parallel groovesthat
mark the surfaceof the northem two-thirdsof the Island, in contrastto the smooth surfaceof
the mainchannelof the Chobe
179. Like Professor Alexander,ProfessorRichards, finds'the scroll-bar...patternsevidentin
the air photographs."30(emphasis added) In connection with these scroll bars, he uses the
carbon dating developed in Dr. Sefe's Sedimentological Study to explain the evolutionary
sequenceof zones 'a'to 'd'onDiagram4, Sheet 17of the Appendices tothe MainAlexander
Report. Theaccountculminateswiththe formationofKasikili Islandby the avulsionthrough
the chutechannelat thebase of themeanderabout one thousandyears ago.13'
180. No arcane issue of photographic interpretationor geomorphologyis involved onthis
particular issue, however. It requires only an examinationof the photographs by someone
with ordinary eyesightand perception. Sheets 14, 15and 16of the Appendicesto the Main
Reportjuxtapose the relevant portionsof six photographsfiom 1925to 1985. The striations
are particularly prominent in the June 1997 photograph, whichhas become available since
then.'12 But they are readily visible on al1 the others, including the first one in 1925.
Namibiainvites the Courtto see foritself.
(f) The so-calledsub-channeldraininnthe southemportion ofthe Island
181. In his Main Report, Professor Alexanderproduced a diagram of the main channel
showing the left bank running diagonally across the lower third of Kasikili Island in a
generallywest-eastdire~ti0n.l~~ TheBCM challengesthis identification:
129
E.g., BCM,paras. 394, 408,418. Thelasttwoparagraphsdealwiththe 1947photograph and
the 1981photographonbothof whichthe patternof striationon thenorthemportionof the island
appearsquiteprominently.
13'RichardsRep.,para.24.
13'Id.;seealsopara.28.
'32This photographis bestseeninBCM,Attachment,AerialPhotographs,p. 11(June1997).
'" NM,Vol. VI Part 2,Diagram4, Sheet17. In fact the line across the island to which Professor Alexander refers, far fiom
being a bank of a river, is an extension of the inlet at the eastem end of the
southem channel, and is a low lying sub-channel, dividing the western higher
part of the island fiom the lower eastem ~ect0r.l~~
Yet the only support in the BCM for this characterization is the suggestion that the August
1947 photograph shows 'a string of pools/damp areas in the mid-section of this sub-
chant~el,"~~which are also saidto appear on the Landsat images of June 1975.
182. To begin with, it is not quite clear what featureBotswana means to identiQ as the sub-
channel. It is described asa line136and in the 1925, 1947, 1981, 1985 and June 1997
photographs the label appears on or just south of a prominent feature, running in practically a
straight line across the lower third of thesland.'~~But this feature on the photographs has
none of the characteristics of a 'channel' or conduit conveying water. Onthe contrary, it has
al1the characteristics of what Professor Alexander calls it: the lefi bank of the main channel.
It has the attnbutes of a somewhat elevated ridge of higher ground, as is most clearly brought
out in the colour photograph of June 1997. There the feature appears in the light yellowish
colour that marks land showing above the water surface. (SeeFig. 6, following p. 72, above;
see also BCM Attachent, Photograph 10) Photograph 41, in the Second Supplementary
Report, p. A1/20, shows a Botswana Defence Force barracks and watchtower built on this
slight ridgeline, presurnably because it is the highest ground on the Island. (See Fig. 8,
following p. 84) That this feature is indeed a line of higher ground, and not a 'channel' is
confirmed by the contour map, Second Supplementary Report Figure 20 (following p. 57).
183. Furthermore, seven of the aerial photographs, by Botswana's own count, show water
south of the line in the relatively dry seas-n that is, within the main channel as defined by
Professor Alexander. Where did this water come fiom? Botswana's only suggestion is that
134BCM, para. 290. This so-called sub-channelis mentionedin four other paragraphsof the
BCM'scornmentaryon the aerial photographs,BCM,paras. 394, 404,408 430, and is labeledon
the 1925, 1947,'May 1972',1981,1985,July 1977andJune 1997photographsin the BCM. (The
labels appearonly in the reproductionsntheBCMitself(Vol. 1)andnot inthe appendedDossier.
On the July 1977photographthe label is 'residual pool'and on the June 1997 photographit is
calleda 'raincharnel').
13BCM,para. 290.
136BCM,para. 394.
137
In the 'May1972'photograph(BCM,p. 173)however,the labelis cornfortablyinthecentreof
themainchannelas identifiedby ProfessorAlexander.Figure8
Barracksandwatchtmr bulltontheleftbankofthe
malnchannelofhi Chabe Riverat Kasikitfislandflood waters flow into the sub-channel 'when backing ~pstream."~~ But as Professor
Alexanderhas decisively shown, floodwatersback upstrearnonly in the very earlystagesof
the Zambezi's annualrise, when it first reaches the Chobethrough the anabranchedchannels
belowKasikili Island. Thisupstreamflow is reversedwithina fewdays as the overbank flow
from the Zarnbezi reaches the Chobe River along the stretchbetween Ngoma and Kasikili
Island. Thereafier the river flows in a downstream direction through the channel along the
ChobeRidge, with themajor partof it going into the mainchannelto the south of the Island
when it reaches Kasikili Island. The dampplaces, pools, residual pools, etc. that the BCM
seesin the photographsarethe result ofthis downstreamflow.
184. A secondsourcefor this residualwaterin the bed of the main channelis overbank flow
from the Zambezi emanatingfiom the north of the Island, flowing across the northem
channel andthen across the Island itself, following the gradua1southerly slope of the land
into the main ~hanne1.l~~This is confirmed by Botswana's labelingon the June 1997
photograph, BCM p. 187. About one-third of the way fiom the right hand border of the
photographis a label in the formof an inverted'Y'. The legend reads 'floodwater' andthe
arrowsas well asthe shapeof the label showit coming fiomthe northernchannel,acrossthe
Islandtoward thesouthemchannel.
185. Professor Alexander'sthesis about the pattern of flow at Kasikili Island is confirmed,
andwith it, his identificationof the mainchannel.
3. Discussionofvarticularaenal vhotomavhs
186. As noted above, Namibia sees no need to address Botswana's comments on each
photograph individually. A few of these, however, are so seriously in error or reveal such
profond misconceptionsof theproblemthat someresponseis required.
187. The 'May 1972' vhotogravh (BCM paras. 411-413) The comment states that this
photograph was 'taken at a time when the water levers were low' (para. 411); that the
photograph depicts a periodof 'lowwater' (para. 412);that it 'indicat[es]a very lowwater
level'(para.413); and thatit representsthe ChobeRiver 'in adry season.'(Id.) Elsewherein
the BCM, however,the photographis referredto as illustratingconditions'whenthe islandis
'" BCM, para.394.
139
SecondSupp.Rep., para. 6.13.inundated in the wet season' (para. 331) or 'in flood time' (para. 380). Both of these
descriptions cannot be ~i~ht.'~'
188. Furthemore, the photograph could not have been taken in 'May 1972'. The monthly
flow in the Zambezi River at Katima Mulilo in that month was 5,906 million m3,the second
highest in that hydrologie year, and most of Kasikili Island was under water.I4' The
photograph clearly does not show the Island in that condition. It appears fiom an
examination of the records in the Surveyor General's office in Windhoek that the date of 31
May 1972,which appearsonthe flight plan, was in err~r.'~~Thus it is impossibleto Saywhat
month the photograph was taken in or what flow conditions it represents. Since there is no
way of being sure either of what Botswana thinks or of what the actual state of affairs was
when the photograph was taken, al1the comments on and references to this photograph in the
BCMshould be disregarded.
189. The November 1972 vhotoflravh(BCMparas. 414-415) This photograph is put fonvard
as a 'particularly valuable. ..record of the actual manner in which the flood waters of the
Zambezi enterthe Chobe stem.'' T^^e commentary continues:
In the five months between the May and November photographs, the rising
waters in the river north of the Island, caused by flood-waterrun-off fiom the
Zambezi, have pushed upstream into the southem channel to link up with the
western sector of the northern channel at the bifurcation zone marked B. ...
Water has alsopushed up the SpurChannel past ~abuta.'~~
The BCMrefers to this photographno lessthan seventimes. BCM, paras. 309,331, 380,396,
411,412 and441.
14See SecondSupp.Rep., para. 13.6;RichardsRep., VolII, App. 20.
14SecondSupp.Rep., para. 13.6. Professor Alexander verified thatteflightplanfor theseriesof
photographs thatincludedthis one wasindeed dated 31 May 1972. However, not onlythis
photograph,but theothersin the serieswere inconsistentwiththe flowconditions existingon that
date,since they show the Zambez iiver flowing withinits'banks, which is not possiblewith a
monthlyflow of 5,906 million m3. ProfessorAlexander concluded that the originalrecord isin
error.
'43BCM,para. 414.
144
Id. There is someconfusionof terminologyon thislocation. Botswanarefersto thehigharea
on the northern bank of the Spur channeloppositethe confluence with the northern channelas
"Kabuta" and refers to thevillagea kilometreor so to the Westas "Kasika." Narnibiarefers to
boththehighareaandthevillageas "Kasika. "This is againflagranterror. The monthly flowsin the Zambezifiom May throughNovember
1972(in million m3)were: 145
May ...................06
June ...................70
July ..................,..6
August .................58
September ............,024
October ................1...
November ..............06.
The waters of the Zambezi River were not rising during this period. They were falling,and
fallingto the lowest point of the year. There was no 'flood-water m-off fiom the Zambezi'
River fiom July onward. Thereforetherecould havebeen no 'rising watersin the rivernorth
of the Island'during this period, and no waters that'havepushed upstrearninto the southem
channel.' This is not a 'recordof the actual mannerin which the floodwatersof the Zambezi
enter the Chobe system.' It is pure fantasy. The onlyaccurate statement in the commentis
that 'There isno sign of generalfl~odin~."~~
190. TheJulv 1977~hotonrarih(BCMparas. 421-424) The BCM states: 'Thisphotograph
is part ofa series flownby the SouthAfiicanDefense authoritiesinJune 1977 ... to be used
in the production of the 1978 Joint Air ReconnaissanceIntelligence Centremap (the JARIC
map).' Col.Rushworth demonstratesthat it was n~t.'~~Instead,it was takenpursuantto the
'Working Plan' and 'Report' for the productionof the 1:50,000 1982 South West Afnca
mapping. Botswana suggests that certainfeatureson this photograph account for the failure
of theJARIC map to showthe southemchannel.I4*But the southem channelclearlyappears
inthe 1982map, whichwasthe onethatwas actuallymadefiomthis photograph.
191. June 1997 ~hoto-a~h and mosaic (BCM paras. 425-440) This is an important
photograph,because it is the onlyone beforethe Namibia Apnl 1998photographthat shows
appreciableinundationandgives an ideaofthe appearanceof the Islandandthe patternof the
14*RichardsRep., Vol. IIApp.20.
'46BCM,para.415.
147
See Reply Amex 1,para. 23.
148The reasonfor thiserror in theJARICmap isfullyexplained in NM, para. 317 and NM,Vol.
V, Annex102, para.32.waters during the high flow season.14' In its comment, the BCM asserts:
Itis also apparentthat the south-western corner of the island is dry sandy soil
of the same texture as the mainland. This indicates that this area, which falls
right acrossProfessor Alexander's "Zone d" (bed of the pseudo-Chobe River),
has already drained and therefore is higher than the rest of the isla150
The quotation is doubly inaccurate. In the first place, examination of the colour version of
the photomosaic showsthat the dry sandy soil on the southwesterncorner of the Island is not
'of the same texture as the mainland.'15' Instead it is identical with the sediment deposits
within the Chobe River to the west of the Island. This identity confirms that the channel
within which the 'dry sandy soil' lies is the main channel of the Chobe River. It lies dead in
the middle of the channel as defined by Professor Alexander. Nor is there any anomaly, as
Botswana suggests, in the fact that this area has dried out early.15* So have the similarly
coloured and textured deposits upstream in what is indubitably the main channel of the
Chobe. It is well known that in low energy gradient rivers portions of the bed may build up
to a levelhigher than the surroundingarea on either side of the river.'53
192. There is, however, a more fundamental sense in which Botswana's discussion of the
June 1997photograph is wrong. BCM para. 436 states '[tlhe photographywas taken in June
as the backed-UVfloodwaters of the Zambezi were starting to recede.' (emphasis added) This
seems to be a further instance of the fundamental misconception manifested in the BCM
comment on the November 1972 photograph (see para. 189, above), namely that the only
sourceofthe inundation of Kasikili Island is 'backedup floodwaters' emanating downstream
of the Island that are 'pushed ~~stream."~~ These formulations would exclude any
14The BCM commentary(para. 425)beginsby statingthat '[tlhisphotographywasundertakenas
thehigh floodwaterswere receding . ..,whichis correct. Butit thengoeson to Say,in apparent
contradiction,that the photographsshow 'veryhigh waterlevels7or 'highwater levels.' BCM,
paras. 429, and 440, respectively.In fact, the peak occurredmore thana monthearlier, and the
1997 peakwasitselfabnormally low. SecondSupp.Rep., paras. 13.10, 13.13, 13.16.
''OBCM,para. 430. Notethatthiscontradictsthe findingfrom Dr. Sefe's Sedimentological Stuày
thatthe toplayerof the islandconsistsof "clay, siltandmud," not dry sandysoil. BCM, App.3,
para. 35(i).
151SeeFig. 6. followingp. 73 supra. SeealsoSecondSupp.Rep., para. 13.14.
15'BCM,para. 431.
'53M. Morisawa, p. 131, ReplyAnnex30; K.S. Richards(ed.), River Channels (1997),p. 372,
Reply Annex 31.
lS4See alsoBCM,para. 394,wherethe so-calledsub-channelis describedas one 'intowhichfioodcontribution from the overbank flowof the Zambezi across the floodplainreachingthe Chobe
River upstream of Kasikili Island. This proposition, which seems to be a central premise of
Botswana's argument, flies in the face of reality. The flow across the flood plain to the
Chobe River upstream of Kasikili Island is manifest on the satellite images appended to
Professor Alexander's Main Report and in the illustrations in Figures 5 and 6 (following p.
20) of the Second Supplementary Report. As ProfessorAlexander shows,this overbank flow
fi-omthe Zambezi upstream of Kasikili Island 'is the dominant source of flow in the Chobe
River at Kasikilisland."55
4. Conclusions as to aerialvhotonravhs
193. The BCM treatrnent of the aerial photographs contains a tmly extraordinary collection
of mistakes, errors, misunderstandings, misconceptions, contradictions and inconsistencies.
They are so pervasive as to undermine the credibility and reliability of Botswana's entire
argumenton this branch of the case.156
194. Namibia's position is that navigabilityis not a criterionfor identifjmg the main channel
in this case,firs betc,use this is 'a question of scientific fact, calling for expertise in
hydrology, geology and hydrogeomorphology"57 and, second, because the Chobe is not a
navigable river over much of its length where it constitutesthe border between Namibia and
Botswana. It is dry for over half of this distance and swampy (and therefore difficult or
impossible to navigate) formuch of the rest. If the criterion of navigability is to be used,
however, itcannot be as an abstract concept,but must be judged as a practical matte- 'in
relation to the needs of the regional economy' - as Botswana ~a~s.'~~Namibia has
watersflow whenbackingupsueam.';para. 364 ('theChobeRiver is independentof the Zambezi
Basinexceptduringpersistentdroughtswhen some flow from the Zambezireachesthe Chobeby
definedandwellconfinedcharnels suchastheBokalo,Mokama,andKasaicharnels.')
Is5 SecondSupp. Rep. para. 1.9. It is true, of course, that atthe beginningof the high water
seasonthere is a period whenthe inundation proceeds upstream, freast to west. Butthe flow
reversesdirectionwithina fewdaysand the bulkof the water flooding the islancomesfromthe
west andthenorth. SeeNM, paras. 137-143;SecondSupp.Rep.paras. 6.17-6.18.
These deficienciesare not confmedto the BCM discussionof aerial photographs. A more
comprehensive listinis foundinChapterVIII.
15BM,p. 129(FirstConclusion).
ISBM,para. 35.developed these points at length in its Memorial and Counter-Mernorial and there is no need
to repeat that analysis.159Nevertheless, a nurnber of times in the discussion of its scientific
case, the BCM introduces the issue of navigability. In fact, two separate sub-sections of
Chapter 6 - E(i) and H(i) - are entitled '[dlepth is the criterion of navigability.' These
matters are addressed inthis Reply.
195. Narnibia has already demonstrated that the depth of a channel cannot serve as a stand-
alonemeasure of navigability.160It has also shownthat if depth is relevant, the determinative
value would not be average depth, but minimum depth, since it is the shallowest part of the
channel that must be cleared by boats navigating in it. The soundings of the 1985 joint
survey, upon which Botswana places such emphasis, show that the minimum depths in the
two channels differ only marginally, if at a11.161In his Second Supplementary Report,
Professor Alexander produces evidence that the southern channel is, in fact, deeper at its
shallowest point than the n~rthern.'~~He took a senes of photographs showing a herd of
elephantscrossing from the Botswana side of the river across the southem channel to Kasikili
Island and thence across the northern channel to the Namibian side. The elephants are in
deeper water in the southern than in the northem channel, and there is good reason to think
that they instinctively followthe shallowestroute onboth legs of thej~unie~.'~~
196. Professor Richards shows, using the soundings taken by the 1985joint survey, that the
western limb of the northern channel, from the bifurcation to the confluence with the Spur
channel, has a mean depth of only 2.55 metres, as compared with 2.44 in the whole of the
southernchannel. He shows that the deeper section of the-northernchannelbeyond this reach
is in reality 'a continuation of the spur channel.' Thewestern limb of the northem channel,
he continues:
is now being slowly closed at its upstream end by sedimentation, and by the
encroachment of the left-bank lateral bar ... This encourages the flood flows
to follow the channel to the southof the 1s1and.l~~
15'See,e.g., NM, paras. 208, 211;NCM,paras. 6, 24, 43,44, 45, 48, 79; seeparticularlyNCM,
VolIII,Section Il (paras. 11.1-11.18).
lm NM,paras.131, 132, 157;id.Vo1.VIpara. 4.4;NCM,para. 28.
l6'NM, Vol. VI,para. 11.6;NCM,para. 46.
16SecondSupp.Rep., Section11.
'63Id., para. 11.3. Seealso Main,ReplyAnnex33,p. 120.
RichardsRep., para. 30.197. In any case, as Botswana stated in its Memorial, navigabilitymust bejudged 'in relation
to the needs of the regional ec~nom~."~~The BCM also seems to accept that the correct test
is the 'most used' ~hanne1.l~~Despite the BCM7sprotestations to the ~ontrary,'~~ there can
be no doubt that the southem channel is the most used and the most closely associated with
the needs of the regional economy. Like the photograph on Sheet 18s in the appendices to
Professor Alexander's Supplementary Report, Photographs 46, 47 and 48 in the Second
Supplementary Report show nurnerous boats in the southern channel. There are five tourist
boats in Photograph 46 alone, and two each in Photograph 47 and 48.16*The brochures from
Chobetourist resorts likewiseevidencecontinuoususe ofthe southem ~hanne1.l~~
198. These tourist boats have no difficulty in navigating the shallowest parts of the southem
channel. Photograph 19s and 26s appended to Professor Alexander's Supplementary Report
showthe boats hard up against the bank of the southem channel in what appears to be hardly
more than a metre or so of water.170 The second photograph was taken on 22 September
1995when water level in the river was the lowest on record to that time. On Photograph 10
of the BCM Attachment, the wake of a tourist boat can be made out in the southern channel
fairly close to the bifurcation. It gives an indication of the relative size of the boats and the
channel and shows that there isplenty of room for the touristboats to navigate.
199. Mr. Girt Jehdus Visagie, the owner of the King's Den lodge at Kasika in Namibia, in an
affidavit annexed to this Reply, swears that almost al1 the tourist boats go through the
southern channel, and very few, except, obviously, those originating in Namibia, go through
the northern. In fact, he says, lack of access to the southern channel senously impairs his
ability to servethe tourists andhus harms his business prospects.171
16'BM,para. 35.
'66BCM,paras. 387, 389.
16'Id., para. 387.
SecondSupp.Rep., pp. A1123,A1124.
16'NCM,Vol.III, Illustrations,Sheet17s.
I7ONCM,Vol. III Ilustrations,Sheet14s, 18s.
17'ReplyAmex 22, Affidavitof G.J. Visagie. There is nothingmysterious about this preference
for thesouthern channel. The touristsometo the Chobeprimarilyto see garne, and the gameis on
the Botswana sideof the river. Whenanimalscrossto the island,they are visibleprirnarilyfrom
thesouthern channel becauseofthe lowerbanksthere.200. The only evidenceBotswanapresents of commercial activity in the northern channel is
aphotograph of the Zambezi~ueen.'~~The BCM states:
It is to be noted that the one sizeable ship at present navigating this section of
the Chobe River can only use the northem channel. The Zambezi Queen, 42
metres in length and three storeyshigh, is operated asa tourist shiand solely
uses the northem ~hanne1.l~~
The only accurate statement in this quotation is that the photograph depicts the Zambezi
Queen. For the rest, this account is another figment of Botswana's imagination. The
ZambeziQueen does not 'navigate' the Chobe River, nor'is it 'operated as a tourist ship.' It
was brought to its present location four years ago, dom the Zambezi River f?om Katima
Mulilo and through the anabranched channel at high water. It went through a short stretch of
the eastem branch of the northern channel at this time before tying up at its present position,
shown in the photograph, on the northem bank of the spur channel close to its confluence
with the northern channel. For the last four years, it has been permanently moored at the
King's Den lodge, where it supplies auxiliary accommodations for guests at the King's
en.'^^ Mr. Visagie, the orner of the boat, testifies that he has no intention of moving the
boat. If he did, he would have to take it back to Katima Mulilo, because it is too big to
operatein the northern ~hanne1.l~~
201. The BCM is replete with references to reeds that choke the southem channel and make
it unfit for navigation. Again, it deploys a photograph (BCM, p. 153) w,hich, it says, shows
'the growth of reeds across the western entrance of the southern ~hannel."~~Namibia has
studied this photograph with care, and invites the Court to do so as well. It is at a loss to
discover any reed growth that would prevent navigation. Infact, the photograph seems to
show a broad and rather open watenvay. Nor are there any signs of reeds obstmcting the
western entrance to the southern channel in the aerial photograph of June 1997, taken only
three months earlier.
172BCM,p. 133.
'73Id.para.347.
'74ReplyAnnex 22, Affïdavitof G.J. Visagie. See also brochures for thZambeziQueenandthe
King's DenLodge,ReplyAnnex 23.
175Id.
'76BCM,para. 383.202. It is true that the southernchannel has not been fiee of aquaticvegetation at al1times in
the past. The misdated 'May 1972' aerial photograph'77shows the channel clogged with
Kariba weeds fi-omthe bifurcation past the first large loop to the east. As mentioned in
NCM, the problem of Kariba weeds in the Chobe and other rivers of tourist interest was a
matter of concem to both Botswana and South AfÎica in the mid-1970s. They undertook a
joint and largely successful programme to eradicate the weeds, and they do not appear in later
aerialphotographs of the 1s1and.l~~
203. Upon examination, the argument on navigability, like so much else in Botswana's
scientificcase, turns out to be a gooddeal of 'sound and furysigni9ing nothing.'
D. The Determination of the 'Boundarybetween Namibia and Botswana around Kasikili
Island' withinthe Meaninrzof Article III(2)of the 1890Treaty
204. Asnoted in Chapter 1,Namibia's Memorial reserved for later consideration the question
of the exact location of the boundary within the main channel of the Chobe River.'79
Namibia foresaw at that time that 'the location of the centre of the main channel would
follow largely as a matter of course by reason of its dependence upon the manner in which
the principal issue is res~lved."~~ That has indeed proven to be true, and now, at the
conclusion of the examination of the scientific aspects of the case, it is appropriate to
recapitulatethe analysis that establishesthe location of the boundary.
205. The scientific evidence has identified the main charkel of the Chobe River as depicted
originally on Diagram 4, Sheet 17of the Appendices to Professor Alexander's Main Report
and as shown in Figures 16, 17, 18 and 19 of the Second Supplementary Report. The next
step is to determine where the boundary lies within this main channel. This two-step
procedure reflects the plain meaning of the text of the Treaty, and Namibia has advocated it
fiom the beginning of its written pleadings.
206. To determine the boundary, the Court must reconcile the English and German versions
of Article III (2) of the 1890 Treaty. The English text defines the boundary as 'the centre of
177Seepara. 164,supra.
17'See id.,andn. 99, supra.
17'NM, para. 160.
''NM, para. 159.the main channel' of the River Chobe. The German text says 'imThalwegdes Hauptlaufes. '
(emphasis supplied) In Namibia's submission, the Court should constnie both words as
referring to thethalweg of the main channel, in its technical and geographical sense as the
line of deepest soundings in that channel. The Court does no violence to the language by
treating the words 'thalweg' and 'centre' as synonyrnous in this context. Although the
practice was by no means uniform, the use of the English word 'centre' to indicate the
thalwegof a channel or a river was a perfectly acceptable usage.lgl Indeed,Article 3 of the
Heidelberg Resolution of the Institute of International Law, upon which Botswana heavily
relies,lS2provided
La frontièredes Etats séparés parle fleuve est mwquéepar le thalweg, c'est a
dire la lignemédianedu chena1.1g3
207. As the scientific analysis has shown, the appearance of the channel that flows to the
south of Kasikili Island differs considerably accordingto the season of the year. When the
Chobe River is flowing, the bulk of the flow goes through the broad conduit that Professor
Alexanderhas identified as the main channel.lg4 That is why it is the main channel. During
the dry season, however, the river assumesthe configuration that is familiar on the maps and
most of the aerial photographs presented in this case. Both .Professor Alexander and
Professor Richards agree that the southem channel, in that configuration, is the thalweg
channel of the main channel.lg5 The main channel overlies this thalweg channel, so that
during the high flow season, when there is flow in the Chobe River at Kasikili Island, the
flow in the main channel submergesthe thalweg channel. In the dry season, when the river
assumes the configuration familiar from the maps and aerial photographs, the water that
remains in the channel to the south of the Island continues to mark the thalwegchannelof the
main channel.
18'See NCM, paras. 54-58.
18*BM, paras. 139-140.
lg3Annuaire de1'Institut.1887-1888,p. 182. (emphasisadded)
184SecondSupp. Rep., Sec. 7, and RichardsRep., paras. 28, 29, 30, 31, 33, 35, 36. See also
NCM, Vol.III, Supp.Rep., para.5.19; NM, Vol. VI, para. 12.1.
ls5SecondSupp.Rep .,paras. 14.14, 14.19;=chards Rep., paras.3, 4, 34.208. This phenomenon is by no means unique to the Chobe River at Kasikili Island. It is
characteristicof non-perennial rivers in arid regions. Namibia's pleadings illustrate a number
of examples:
the Chobe River at Ngoma Bridge (see Fig. 4, following p. 35, comparing the dry
seasonphotograph looking downstream taken on 20 September 1995in the dryseason
with a photograph taken fiom the same point on 11 April 1998 in the high flow
season).
the Orange River (see Sheet 4 of the Appendices to Professor Alexander's Main
Report).
the Limpopo River (see id., comparing dry season photograph taken on 26 September
1995with high flowphotograph taken 11July 1996).
the Mkuze River (see Second Supp. Rep., Photograph 62, p. A1132,showing thalweg
channel).
the Mhlatuze River (see Second Supp. Rep., Photograph 63, p. A1132, showing
thalweg channel).
The same transformation occurs at Kasikili Island between the dry season and the season of
high flow in the Chobe River.
209. The only alternative to the construction proposed by Namibia would be to accept the
geometnc middle line of the main channel as the 'centre.' But thiswould have the effect of
disregarding entirelythe word 'thalweg 'in the Germantext.
210. The interpretation of the Treaty that Namibia proposes, placing the boundary in the
thalweg of the main channel, would mean that the boundary followed the line of deepest
soundings in the southem channel. This would correspond to what is shown as the boundary
on almost al1the maps of the area since the time of the 1890Treaty. And it would allowboth
parties full access to the navigational resources of the river for purposes of the tourist
industry.'86
IsNamibiahas alwaysbeenwillingto allow Botswana tousethenorthernchannel.211. Namibia submits that the Court should answer the question submitted as to the
boundarybetween NarnibiaandBotswanaaroundKasikili Islandby determining,
-irst, that the 'main channelof the Chobe River7is the channel to the south of the
Island definedby ProfessorAlexander,as indicatedabove,and
second,that thethalweg ofthat channelis theboundary. ChapterIV
'SUBSEQUENT PRACTICEINTHE APPLICATION OF THE TREATY'
WITHIN THE MEANTNGOF ARTICLE 31(3)(b) OF THE VIENNA
CONVENTION *
A. Botswana's Argumentsas to the me an in^of Article 3l(3Mb)
1.In neneral
212. The subsequent conduct of the parties has long been acknowledged by the Court and
other international tribunals as an important element in treaty interpretation.' It is also
recognized as such by the general rule of interpretation of the Vienna Convention on the Law
of Treaties. Subsequentconduct can be manifested in many ways: by agreementbetween the
parties on particular issues in dispute, by actions of the parties on the ground, by diplomatic
and interna1 correspondence, by statements of officiais, by maps produced or used or
accepted by the parties, and by silence orfailure to respand when response would be called
for to maintain a particular construction of the treaty involved.
213. Under the Vienna Convention, the subsequent conduct of the parties relevant to the
interpretation of a treatyan take two forms: either (1) an explicit agreement or (2) a practice
establishing the agreement of the parties regarding the interpretation of the treaty. The
distinction is embodied in Article 31(3)(a) and (b) of the Vienna Convention, which read as
follows:
There shall betaken into account, together with the context:
(a) any subsequent agreement between the parties, regarding the interpretation
of the treaty or the application of itsprovisions;
@) any subsequent practice in the application of the treaty which establishes
the agreement of the parties regarding its interpretation.
214. As the parties to the Anglo-German Treaty of 1890have not entered into any agreement
regarding itsinterpretation? the subsequent conduct relevant to the present dispute is the
NM, paras.171-174.
Botswanaargues that the discussionsheld with SouthAfrica in 1984/86 resulted in such anparties' subsequentpractice in the application of the treaty as described in Article 31(3)(b).
Thispractice, as Namibia has shown in this and inprevious pleadings, clearly 'establishesthe
agreement of the parties regarding the interpretation of the treaty' in respect of the two main
issues before the Court: first, that the boundary is in the channel that lies to the south of
Kasikili Island, and, second, that the Island is part of the territorynder the sovereignty of
Namibia.
2.The correctintemretation of Article 31(3)(b)of the Vienna Convention on
theLaw ofTreaties
215. Botswana seeks to persuade the Court that the 'practice' referred to in Article 31(3)@)
of the Vienna Convention as the basis for establishing the parties' agreement on
interpretation must be cornmon to, or jointly carried out by the parties.3 The text of the
article does not sustain this interpretation, so Botswana, in characteristicfashion, caricatures
Namibia's position and offers a partial and incorrect account of the travaux préparatoiresof
the Convention. Thus, before responding to Botswana's substantive arguments on
subsequent conduct, Namibia will examineits interpretation of Article 31(3)(b) of the Vienna
Convention.
216. Botswana describes Narnibia's interpretationof Article 31(3)(b) as an attempt to dilute
the notion of subsequentpractice in order to move away f?omthe 1890 ~reaty.~ Botswana's
allegation is based, however, on a mischaracterization of Namibia's position. According to
Botswana, Namibia's Memorial 'seeks to establish that subsequent practice may be
constituted exclusively by the "silence" or "inaction" of the parties:.. .'5(emphasis added)
This account misrepresentsNamibia's position. Namibia does not deny that Article 31(3)(b)
requires affirmative conduct. What it maintains, however, is that the provision does not
require affirmative conduct fiom &l the parties to the trea~.~ Thus, in the case of a bilateral
agreement.In its Counter-Memorial Namibiademonstratedthat therewas no suchagreementand,
evenif therehadbeen, it wouldhavebeenvoid ab initio.Newevidence thathas becomeavailable
since the filing of the Counter-Mernorial confirms Namibia' interpretation. Thisevidence is
discussedinthefinal sectionof thischapter.
BCM,paras. 237-240.
BCM,paras. 238and240.
BCM,para. 237
6 In its Memorialand Counter-Memorial,Namibiademonstrated thatsince the conclusionof thetreaty the practice of the parties regarding the interpretation of the treaty may be established
through the conduct of one oftheparties, provided that the otherparty does not di~sent.~
217. Botswana makes much of a minor change introduced into the text of Article 31(3)(b) at
the UN Conferencethat approved the Convention. Namibia's Memorial addressed this point
and concluded that this changewas merely linguistic and did not contradict its interpretation
of the convention.* Since Botswana raised this point in its Counter-Memorial, Namibia will
expand on the evidencesupportingthis conclusion.
218. The clause on subsequentconduct in the Drap Articles on the Law of Treaties prepared
by the InternationalLaw Commission, reads as follows: 'any subsequentpractice .. .which
establishes the understanding of the parties .. .." At the UN Conference on the Law of
Treaties the word 'agreement' wassubstituted for the word 'understanding.' The object of
this change, as Namibia explained in its Memorial, was to bnng the English version of the
Convention into line with the French and Spanish versions, which use the words 'accord' and
'acuerdo,' respectively. It was aminor change that did not alter the meaning of the clause,
since the words 'understanding' and 'agreement'are synonymous,in the sense of identikng
amatter on whichthe parties share a common view.
219. Botswana, however, argues that the substitution of the word 'agreement' for theword
'understanding' is a point of sub~tance.'~ In support of this contention it invokes an
amendment tabled by the Australian delegation which proposed to insert the word 'cornmon'
before the word 'understanding.' Thisamendment,though linked to the decision to substitute
the word 'agreement' for the word 'understanding,' was eventually rejected. Botswana,
however, relying on this link, concludes that the change in the wording of the subsequent
practice clause is of majorignificance.
1890 Anglo-German Treatyuntil well intothesecondhalfof thiscentury,KasikiliIslandwasunder
thecontinuousjurisdictionandcontrolof the Masubiaof the Caprivisubjectto the authorityof the
rulingpower, withfull knowledgeand no objectionsfromBotswana.Throughout this period there
was thusaffirmative conductbyNamibia andno dissentby Botswanaor any of its predecessorsin
title.
NM, paras. 175-177.
NM, para. 177.
UnitedNationsConferenceon theLaw of Treaties -First andSecondSessions:Documents of the
Conference (1968-1969) p. 37.
'OBCM,para. 239.220. An examination of the travaux préparatoires showsthat Botswana is wrong. The
Australian amendment - as acknowledged in the statement reproduced in Botswana's
~ounter-~emorial" - was prompted by the InternationalLaw Commission'scommentary
on the subsequentpractice clause. In the Commentaries,the Commission explainsthat in an
earlier draft the clause referred to 'practice which"establishes the understandingof glj the
parties."''2 (emphasis added) Eventually, however, the word 'all' was ornitted. The
Commission considered it redundant, because the phrase 'understanding of the parties'
necessarily means understanding of al1the parties.'3 The Commentaries also note that the
practiceto which that clauserefers is practice that 'showsthe common understandingof the
parties.''4 The referencein the Commentariesto the commonunderstandingof the partiesis
what prompted Australia to propose that the word 'cornmon'should be inserted before the
word 'understanding.' In view of the statementin the Corbentaries, however,it.is clearthat
the objectiveof Australia's amendmentwas toreinforcethe InternationalLaw Commission's
interpretationof that clause: that the practicemust reflectthe agreementof both parties asto
themeaningofthe treaty.
221. Understanding the objective of the Australian amendmentalso explains its rejection.
For after the Drafting Committee decided to substitutethe word 'agreement'for the word
'understanding,' the Australianamendment became superfluous. The Chairman of the
Drafiing Committee made this point succinctly at the Conference when he explained the
reason for rejecting the amendment: an agreement is always cornmon and cannot be
unilateral." Thus, although the Australianamendmenthad a bearingon the slightchangein
the English version of Article 31(3)(b), the objective of the amendment was simply to
underscorethat the Convention requiresthe parties to be in agreement. It had nothingto do
with theway inwhichthe agreement is manifested.
l'BCM,para. 239.
l2UnitedNations Conferenceon th Leaw of Treaties- FirstandSecondSessions:Documentsof the
Conference (1968-1969)p. 37.
l3Id.,p. 42
l4Id.
15
UnitedNations Conferenco en the Law of Treaties- First SessioSummaryandRecordsof the
PlenaryMeetingsandofthe Committee ofthe Whole(1968)p. 442.222. The International Law Commission explicitly refers to this point in explaining its
decisionto omit theword 'all'fromits earlierdraft. It regardedtheword 'all'as superfiuous,
but italso concluded thatincorporatingthewordwouldcreateconfusion. 'It omittedtheword
"all" merely to avoid any possible misconception that everyparty must individually have
engaged in the practice when it suffices that it should have accmted that vractice.'16
(emphasis added)This comment confirmsNamibia'sposition that Article31(3)(b) does not
requireaffirmativeconductfrombothpartiesto the treaty.
223. The decision in theBeagle ChannelArbitration also confirmsNamibia'sposition.17In
that case, Argentina argued that Chile's acts ofjurisdiction over certain disputedislands did
not qualifi as subsequent conduct under Article 31(3)(b) because they were unilateral acts
and, as such,did not express the cornmonwill of the parties. TheCourt of Arbitration flatly
rejected this argument, stressingthat Article 31(3)(b)does not speciq the way in which the
agreement was tobe manifested. The Courtthen examined Chile's acts of jurisdiction and
found that they were public, well known to Argentina and consistent with the Treaty.
'[Ulnderthese circumstances,'the Tribunalconcluded,'the silenceof Argentina permits the
inference that the acts tend to confirm an interpretation of the meaning of the Treaty
independent ofthe acts of jurisdiction them~elves."~ The record of the subsequentconduct
in the present dispute points to the same conclusion. Botswana's silence in the face of
Namibia's public and continuous exercise of jurisdiction over Kasikili Island 'tend[s] to
confirman interpretationof the meaning of the Treatyindependentof the acts of jurisdiction
themselves.'
224. In the present case, the silence of the British authonties evidences somethingbeyond
mere acquiescence.The 1890 Treaty delimitedboundariesnot between sovereign tenitories
but betweenthe spheresof influenceof the parties. Accordingto Jan Verzijl, '[sluchzones
or spheres did nottherebybecome Stateterritoryproper,butas faras theywere concerned an
l6United Nations ConferenceontheLaw of Treaties- Firstand SecondSessions:Documentsof the
Conference (1968-1969) p.42.
" BeagleChannelArbitration(Argentina v.Chile) 1977,52 I.L.R. p. 93.
18Id.,para.169,p. 224.inchoate title was vested in the proclaiming or delimiting State(s) that could gradually
develop intofull so~ereignty."~ Therefore,some form ofpositive assertionofjurisdiction or
control wasnecessary if Britain was to perfect its title. Thefailure of the British authorities
to makeany sucheffort,in the faceof the occupationanduse of the Islandby the Masubia of
Caprivi, clearlyreflects their understanding that the 1890 Treaty assigned KasikiliIsland to
the Germansphereof influence.
225. The conduct of Namibia plus Botswana's silence constitutes a 'subsequent practice
which establishes the agreement of the parties' regarding' the interpretation of the Anglo-
German Treaty of 1890: the boundary is in the channel to the south of the Island, and
KasikiliIslandis in the temtory underNamibia'ssovereignty.
B. Botswana's FactualArguments onSubseauentConduct
226. Botswana's argumenton subsequentconductis basedonthreepropositions:
That up to 1948 the parties to the Anglo-GermanI'reaty 'were of the opinion that the
main channel in the vicinity of KasikiliISedudu Island passed to the north of the
i~land.'~'
That the exchanges culrninating in 1951 with the Trollope-Dickinson arrangement
confirmed thatthe parties recognizedthe northernchannelas the boundaryand Kasikili
Island aspartof the Bechuanaland~rotectorate.~'
That the 1984186discussions between Botswana and South Afiica resulted in an
agreement between the parties, authorized by the United Nations, regarding the
application of the Anglo-German Treaty of 1890 that resolved the present boundary
disputein favourof ~otswana.~~
Inthis sectionNamibia showsthat al1of thesepropositionsareincorrect.
l9VerzijlJ.H.W. InternationalLaw in HistoricalPerspective,Leyden, 1970, p.495
20BM,para. 165.
21BCM, paras. 241, 736,732and729.
22BCM,para.264. 1. Botswana's contention that until 1948 the parties to the Annlo-German
Treaty were of the opinion that the northem channel was the main channel
and that Kasikili Islandwas part of the Bechuanaland~rotectorate~~
227. Before looking at the Botswana's evidence in detail it is well to recall the 1949
statement of Sir Evelyn Baring, the British High Cornrnissioner for the Bechuanaland
Protectorate, which directly contradicts Botswana's contention. Kasikili Island, he said, 'has
hitherto been regarded as part of the Caprivi Zipfel, since maps show that the main channel
passes to the south of the ~sland.'~~
228. Botswana presents three items of evidence in support of its contention: (1) a despatch
fiom Lord Harcourt that does not refer to Kasikili Island, (2) a peculiar interpretation of the
Eason Report and (3) an alleged permission granted by an officia1fiom the Bechuanaland
Protectorate allowingthe Chief of the Masubia to cultivate on the Island.
(a) Lord Harcourt's Despatch of 14Januarv 1911
229. Citing a despatch fiom Lord Harcourt, the British Colonial Secretary,Botswana asserts
that in 1911 senior officials fiom the British Govemment held the view that the northem
channel was the main ~hannel.~~
230. Lord Harcourt's despatch is addressed to the High Commissioner for South Afiica. In
it, he forwards correspondencerelating to the dispute withGermany over the location of the
boundary in the western sector, south of the village of ~ndara.~~As he anticipates the
possibility that the proposed arbitration with Germanymight includethe rest of the boundary
between the BechuanalandProtectorate and the Caprivi,he requests the High Commissioner
to supplyinformation about the sector of the boundary along the River Chobe.
231. Therelevantparagraphs of Lord Harcourt's despatchread as follows:
1have the honour to acknowledge the receipt of your despatch, Confidential,
No. 4 of the 30th November, relative to the boundaries of the Caprivi Zipfel,
and to forward, for your information, copies of correspondence with the
Foreign Office on this subject, fi-omwhich you will observe that it is proposed
23BM,para. 165;alsoBCM,paras. 13, 26, 34,723 and 724.
24NM, Annex 66.
25BCM, para. 41.
26NM,para. 266. to refer to arbitrationthe Article III of the Anglo German Agreementof 1890
which lays downthe southemboundaryof the territory.
1take this opportunityto observe that in the second clause of Article III the
Anglo-GermanAgreementof 1890it is statedthat the boundary"descendsthe
centre of the main channel of that river (i.e. the River Chobe) to itsjunction
with the Zambesi". As,in this sectionof its coursethe riverividesinto more
than one channel which afterwards reunite, the question as to which is the
main channel will require consideration. 1have to request, with reference to
the enclosure to Lord Selbome's desvatch, Confidential (2). of the 1lth of
Avril, 1910,that 1mav receive al1available informationfiom local sourcesin
suvvort of the view that the north channel is the main channel. Such
information should be accompanied by a map and, if possible, by
measurementsof the streams,and shouldbe in a form which can, if necessary,
be laidbefore the arbitrator aspartof the caseof Hisajesty's oven ni ment.^^
(emphasisadded)
232. It is hard to understand how Botswana can conclude that Lord Harcourt's despatch
reflects a firm view aboutthe locationof the main channel aroundKasikili Island. It is clear
that hispurpose was to request information not yet available aboutthe location of the main
channel. This is why, shortly thereafter,aptain Eason was instructedto inspect the whole
stretchof theboundary alongthe River Chobe.
233. Botswanawould like the Court tobelievethat Lord Harcourtis refemng specificallyto
the main channel around Kasikili ~sland.~~This is obviouslynot the case. Lord Harcourt
does not refer either to the main channel aroundKasikili Island or around any particular
island. If Lord Harcourt had any island in mind, it would havebeen Swampy Island on the
Chobe River, which superficially resembles Kasikili Island, but is located some 170
kilometres to the west. Swampy Island was an importanh tunting ground for the Batawana, a
tribe fiom northwestem ~otswana.'~ There had been complaints bythe Batawana who
claimed thatthe 1890 Treaty had reduced their tribal reserve. TheBritish authoritieswere
thus concemed that the eventual allocationof Swampy Islandto the Capriviwould lead to
furtherdeteriorationof their relations with the Batawana. Hence,the strong representations
by Resident Cornmissioner Panzeraabout the importancelofclairningthe Island: 'What they
''NM, Annex 44.
28BCM, paras.41 and 723.
29
SeeMM,Atlas,Map V, 'SketchMap ofBechuanalandProtectorate,1:2,000,000.'War Office,
April 1909GSGSNo.2460.[the Batawana] would bitterly feel the loss of would be the island [Swampy Island] in the
north-east corner . . . This island forms their most valuable hunting ground, containing
elephant, &c.'~'
(b)Eason's Report
234. Botswana appears to be aware that, on its face, Lord Harcourt's despatch does not
support its case. So it argues that the despatch should be read in conjunctionwith the Eason
report. The relevantparagraph of Eason's Reportreads as follows:
Two miles above the rapids lies Kissikiri Island. [Hlere 1 consider that
undoubtedly the North shouldbe 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 and eight feet deep. The South channel is merely a back
water, what current there is goes round the North. Thenativesliving at Kasika
in German territory are atpresent growing crops on it.31
235. According to Botswana, this paragraph, read together with Lord Harcourt's despatch,
make it clear 'that the British Govemment already held the view that the northem channel
was the main ~hannel.'~~The irony of this argument is that when these two documents are
read together they lead to a conclusion that directly undermines Botswana's case. For after
Eason's Report of 1912, Lord Harcourt and his colleagues in the British Govemment could
not have been in any doubt that Kasikili Island was occupied by the people fi-omthe Caprivi.
Bntain, however, failed to enter a protest. Moreover, although Eason had suggested that
there was a basis for making a claim to the northem channel, Britain did not make such a
claim.
236. Eason was not an expert on rivers. His survey of the Chobe River had been
commissioned in anticipation of litigation with Germany over the southem boundary of the
Caprivi. He was asked, in accordance with Lord Harcourt's instructions, to collect
information 'in support of the view that the northem channel is the main ~hannel.'~~
(emphasis added) Eason's report, therefore, reflects the tendentious character of Lord
Harcourt's instructionsand cannot be consideredas the opinion of an independent observer.
30NM, Annex 39.
31NM, Annex 47.
32BCM, para. 34.
33NM, para. 44.237. Eason's findingregardingthe use of Kasikili Island by the peoplefiom the Capriviis,
however,significant. The Island, together withmost of the Eastern Caprivi,is inundated for
up to six months of the year. Fertile land is a scarce resource carefully husbanded by the
Masubia authorities. Arrangements for the use ofland were thus subject to strictpolitical
controls. The following extract fkom a report on the Eastern Caprivi by C.E. Kruger,
Magistrate for the Eastern Caprivi in 1963 and Trollope's Assistant in the 1940s, briefly
explainsthe landtenure arrangementsofthe Masubia:
New villages and fields are establishedwith the permission ofthe chief after
the villagehead andhis men havesoughtout a place and approachedthe local
headman,who in tum introducesthe matter to the Kuta. The member having
junsdiction makes an inspection and demarcates if other people already
established arein close proximity. Once the area setpart for fieldshas been
fixed al1the village inmates go with the head for each to get what he wants.
Thereis no fixed limitand nopayrnent. A newcoker to thevillagewouldalso
have his field allocated by the village head without reference to higher
authority,oncepermissionhas been givenforentryto the headman's~irea.~~
238. Kruger's account is confirmedby the eyewitness testimony before the JTTE.~~ The
allocationof land betweenandwithin villageswas subjectto strictproceduresinvolvingboth
centraland local political structuresof the Masubia. Agriculturalactivitiesof the Masubia of
Kasika were thus a manifestation of the exercise of sovereign jurisdiction by the colonial
rulers through the traditional authoritiesof thesubia. Contrary to Botswana'sdistorted
interpretationof Namibia's agriculturalactivities of the Masubia are relevant to
the issue of subsequent practice not because they constituted title to Kasikili Island, but
because they presupposedit. The political significanceof the use of Kasikili Island by the
Masubia ofKasikadid not escapeEason.
239. Eason submitted his report to Panzera, the Resident Commissioner, who, on 3
September 1912, sent it to the High Commissionerin Pretoria without commenting onits
contents.37On 23 September 1912,the High CommissionerforwardedEason'sReportto the
Colonial Office in London, also without comment.38 Panzers's silence is especially
34NM, Amex 184.
35NM,AM~X1,p. 61.
36BCM, paras.19, 679, 682.
37ReplyAmex 13.
38ReplyAmex 14.significant for he had considerable experiencein boundary matters and, as such, could not fail
to be aware that land use is of fundamental importance'in decisions about the location of
boundary line~.~' He was also a zealous guardian of the Bechuanaland Protectorate's
territory, as illustrated by his despatch on SwampyIsland.
240. Thus, the Eason report and its treatrnent by British officialdom does not support
Botswana's assertion that the British authorities held the view that the northem channel was
the main channel and Kasikili Island part of Botswana.
(c) The allened riermissionto cultivate on the Island nranted in 1924
241. Botswana argues that in 1924 and 1925 British authorities exercised sovereign
jurisdiction over Kasikili ~sland.~' The evidence adduced in support of this argument is
contained in a 1948report fi-omNoel Redrnan, Assistant District Commissioner at Maun, to
the Government Secretary of the Bechuanaland Protectorate. In this report, Redrnan recounts
an incident that allegedly had taken place in the 1920s, more than 29 years earlier. The
relevant sectionsofthe report read as follows:
Since the attached report was prepared 1 have received Mer information
fiom an inhabitant of the Island that in 1924 a Caprivi Chief named
Liswaninyanaapplied to CaptainNeale, the Resident Magistrate at Kasane, for
permission for his people to plough on the Island and graze cattle there. This
was evidently granted verbally as no written agreement is known. At this time
Government Oxen were grazing on the Island but they were removed in 1925.
Before 1924 the same informant told me that there was one Caprivi family
ploughing there but they had no authorityto do so.
In defence of the Claim of ownership on the grounds of prescription evidence
could therefore be produced to show that this Governent occupied the Island
in 1925. This fact together with the acknowledgement by the people that they
applied for permission to use the Island in 1924 should cause any clairn of
prescriptiverights to be reje~ted.~'
242. Botswana interprets these paragraphs as showing 'a clear exercise of jurisdiction on
behalf of the Bechuanaland ~rotectorate.'~~This interpretation is patently wrong. In his
39 In 1896 Panzera participated in the demarcationof the tribal reservesof the Bechuanaland
Protectorate and played a prominent role in the demarcationof the tenitory assigned to the
Batawana. See Reply Annex8.
40BCM,paras. 38, 165,166.
41BM, Vol. III,AM~X22.
42BCM,para. 38.report, Redmandoes not offer any substantialevidence thatthe Island had everbeen under
the control of the Bechuanaland Protectorate. The only evidence he offers is hearsay -
indeed,doublehearsay.
243. The suggestionthat permissionwas granted'verbally'is also suspect,because it would
have been wholly inconsistent with British practiceat the time. In other cases wherepeople
from one area under British rule wished to cultivate or use land in another, forma1written
permission was required, approved by higher authority and subject to annual renewal.')
Similarfonnality was observedin extendingprivilegesto missionariesand othersto cultivate
in an area. With Portugal, the Britishgovernment concludeda forma1treaty regulatingtrans-
boundarycultivationacrossthe borderbetweenAngola andNorthern~hodesia.~~
244. Moreover,the BCM failsto recognize thatin 1924,CaptainNeale had a dual role. He
was both District Commissioner for Kasane in the Bechuanaland Protectorate and
administratorof the Eastern Caprivi under the League of Nations Mandate for South-West
Africa. Thus, even if he had authorizedLiswaninyanato cultivateon the Island, this action
would not necessarily prove that the Protectorate exercised control over the Island. Given
Captain Neale's dual functions, his alleged authorizationto use Kasikili Island cannot be
properly assessed without first determiningwhether he was acting as administratorof the
EasternCaprivi or as administratorof the BechuanalandProtectorate's DistrictofKasane. If
hewas actingin his capacity asadrninistrator of theEastern Caprivi,theallegedauthorization
to use the Island would constitute evidence of Namibia's control over the Island, not
Botswana's. Since Botswana does not state in what capacity Captain Neale was allegedly
acting,the facts recounted in Redman's reportdo not provethat British authoritiesexercised
jurisdictionin Kasikili Island.
245. Thetestimonyof eyewitnessesbeforethe JTTEconfirmsthatthe Masubianever sought
nor obtainedpermissionto ploughonthe Island:
BOTSWANA COMPONENT: Did ChiefLiswaninyanahave the permission
of the Resident Commissioner at Kasane,permission for his peopleto plough
atthe island andgaze cattle there?
43SeeNM,Annexes50, 51. ThematterisalsodiscussedatNM,para.258.
SeeNM,Annex8. WITNESS: Chief Liswaninyana, that was his land. There's no one who could
give him authority or power to go and plough there. We, his sons, we used to
go there and lough there and there's no one fiom Namibia going and plough
in Botswana. B,
246. The JTTE proceedings also confirm that people fiom Kasane never farmed nor carried
out any other activity on Kasikili ~sland.~~Indeed, as acknowledged by the High
Comrnissioner in 195 1, 'the Island has been cultivated by Caprivi tribesmen for many years
without di~pute.'~'
247. The foregoing shows that Botswana's claim that Britain exercised jurisdiction over
Kasikili Islandduring the penod 1924125lacks any substance.
(d) Conclusion
248. Botswana's contention that up to 1948the parties to the Anglo-German Treaty of 1890
were of the opinion that the northern channel is the main channel is not sustained by the
evidence. The practice of the parties throughout thatpenod can be surnmarized as follows:
Kasikili Island was occupied and regularly cultivated by the Masubia fiom Kasika.
The traditional authorities of the Masubia had jurisdiction over Kasikili Island and
controlled its use. Under the principles of indirect rule the political and legal control
of Kasikili Island by the Masubia constituted an exercise of sovereignjurisdiction by
the colonialpowers.
The occupation of Kasikili Island by the Masubia fiom Kasika was peaceful and well-
known to the British authonties in the Bechuanaland Protectorate and in London.
They failed,however, either to protest or to make a claim for the Island.
249. This practice, by the late 1940shad established the agreement of the parties that, under
the 1890Treaty, the boundary was located in the channel to the south of the Island and that
Kasikili Island belonged to Narnibia. This agreement is confirmed by Sir Evelyn Baring's
statementthat 'Kasikili Islandhas hithertobeen regarded as part ofthe Caprivi ~i~fel.'~'
45NM,Annex 2, p.25. SeealsoNM,para. 202, especially n. 63.
46NM,Annex 2, p. 25.
47NM, AM~X 69.
48NM,Annex 66. 2. Botswana's contentionthat the exchanges thatculminated in the 1951
Trollope-Dickinsonarrangement confirmedthat the parties reconnizedthe
northem channel as the boundarv and Kasikili Island as part of the
BechuanalandProtectorate
250. The BCM restates Botswana's original interpretationthat the exchangesculrninatingin
the Trollope-Dickinson arrangement reflect the parties' understanding that the northem
channelwas the main channel. Itspositionisbased on threerelated assertions:
That in 1948 South Africa challenged the status quo and sought to modiQ the
boundary.
That Britainrejected the challenge and, instead, exercisedjurisdiction over Kasikili
Island.
That,in the event, SouthAfiica backeddown andboth parties recognizedthe northem
channelas themain channelfor thepurposeof theAnglo German Agreementof 1890.
Inthis sectionNamibiashowsthat al1thesepropositionsarewithout substance.
251. The Namibian Memonal contained an extensive analysis of the Trollope-Dickinson
arrangement and the correspondence that preceded it4' The Counter-Memoial also
responded to Botswana's interpretationof these e~ents.~' Although it is not necessary to
repeatthese arguments here, it is useful to re-examineBotswana's interpretationin the light
ofthe versionof the facts containedinits Counter-Memorial.
(a)Botswana's allegationthat SouthAfiicachallennedthe status quo andsoughtto
modifvthe boundarv
252. Botswana asserts that during the exchanges of 1948 to 1951, South Afiica issued a
challengeto the status quo in proposing a modificationof the bounda~y.~'This assertionis
based on a misconception about the status quo and O; a distorted interpretation of the
diplomaticcorrespondencethat led tothe Trollope-Dickinsonarrangement. The discussion
in Section 1 above shows conclusively what the status quo was as of the end of 1947:
KasikiliIsland belongedto Narnibiaand the southem channel formed the boundary between
49NM, paras.274-279.
NCM, paras. 102-106.
51BCM, paras45,46, 736.the two territones. Thus, on the face of it, Botswana's first point is implausible. In this
section, Narnibia concentrates on Botswana's interpretation of the diplomatic exchanges of
the period 1947-195 1.
253. South Afiica's position during the negotiations is reflected in two letters addressed to
the High Commissioner, oneof 14 October 1948,andthe bther of 14February 1949.~' These
letters show that far fiom attempting to subvert the status quo, South Afiica's proposa1to
Britain was to fonnalize it through an explicit agreement. In both letters South Afiica
highlights the crucial fact that Kasikili Island had been used by the people fiom the Caprivi
since at least 1907, and their presence there had never been disputed. Indeed, South AfXca
characterizestheir occupation of the Island as a right:
From the available information it is clear that the Caprivi Tribesmen have
made use of the Island for a considerable nurnber of years and that their right
to do so has at no time been disputed either by Bechuanaland Tribesmen or
Bechuanaland ~uthorities.~~
254. After a brief period of negotiations, the British High Commissioner accepted South
Afiica's proposa1 and fonvarded it to London recommending its acceptance." Botswana
makes much of the fact that in his dispatch to London the High Commissioner referred to
South Afiica's proposa1as 'a slightadjustrnent of the boundary.' Botswana fails to mention,
however, that in the same dispatch the High Commissioner also refers to South Africa's
initiative as a proposa1 'to set the boundary on the southem channel,' thus fixing or making
firm a boundary line that the parties, in their practice, had already recognized. In any event,
it is unnecessary to quibble about isolated phrases in an othenvise extensive correspondence.
The fact is that the High Commissioner accepted South Afiica's proposa1 because, as he put
it, the Island had 'hitherto been regarded as part of the Caprivi ~i~fel.'~~Thus, by endorsing
SouthAfiica's proposal, the High Commissionerwas reaffirming the status quo.
255. As it tumed out, the British Government decided not to enter into a forma1agreement,
opting instead for an administrative arrangement. Botswana, again misinterpreting the
available evidence, claims that this decision shows Britain's detennination to resist South
52NM, Annexes 63, 65.
53NM, Annex 65.
54NM, Annex 66.
55Id.Ahca's challengeto the status
The reason why the British Govemment decided not to
enter into anforma1agreement had nothng to do with the merits of South Afiica's proposal.
The High Commissioner discloses the British motivation in a letter to the South Afiican
Govement: '1am afiaid that they [the Commonwealth Relations Office] have found this
proposa1to be beset by legal complications of an internationalnature, the solution of which
would entai1complicationsdisproportionate to the importance of the matter at issue'.57 The
'legal complications' are explained in a despatch fiom the Commonwealth Relations Office
to the High Commissioner. They arose fiom South Africa's refusa1to recognize the UN as
the supervisorybody of the Mandate for South-WestAfi-ica. Under Article 7 of the Mandate,
any decision regarding the boundaries of Namibia - even one confirming the status quo -
required approval by the body in charge of the supervision of the Mandate. South Afkica's
refusal to acknowledge UN jurisdiction generated 'considerable legal and political
~ncertaint~.'~It is not surprising, then, that the British Govemment, though agreeingwiththe
substance of South Afiica's proposal, should have opted for a low key administrative
arrangement,rather than a forma1agreement.
256. Thus, Botswana's contention that South Afnca challenged the status quo and proposed
amodification of theboundary is unsubstantiated. On the contrary,what South Afi-icasought
dunng those negotiationswas to formalizethe status quo through an agreement.
(b)Botswana's allenation that Britain. with South Afnca's acquiescence, exercised
jurisdiction overKasikili Island
257. Botswana's allegation that the British Govemment exercised jurisdiction over Kasikili
Islandis based solely onthe High Commissioner's initialresponse to the proposa1from South
Afnca to formalize the status quo.59In it, the High Commissioner suggests that the matter
could be resolved by introducing a system of annual permits to use the Island and states that
his officials in the Bechuanaland Protectorate had already instructed the District
Commissioner at Kasane to issue such permits.60 Since South Afiica had exercised
56BCM, para. 757.
''NM, Annex 69
See paras.259-262,infra.
59BCM,paras. 169, 170, 174.
NM, Annex 64.jurisdiction over the Island fiom 1929, when it assurned direct responsibility under the
Mandate, it vigorouslyrejected the High Commissioner'ssuggestionand reiteratedits initial
proposal. Ultimately, the High Commissioner acceptedthis proposal. As a result, the
instructions mentioned in his letter were never implemented and were eventually
~ithdrawn.~'
25 8. The High Commissioner'sinitial responseis not surprisingin view of his unfamiliarity
withconditionsaroundKasikiliIsland.Aftervisitingthe area, however,he changedhis mind.
As he recognizesin his letter of 10 May 1951 addressed tothe South Afiican Govement,
KasikiliIslandhad beencultivatedby the Masubiafi-omthe Caprivi 'formany years without
dispute.'62 After acquinng first-handknowledge of the area, the High Commissionerwas
able to understandthe basis of South Afiica's proposal. Accordingly, he accepted it and
referredit to Londonforde~ision.~~
259. In his letterof 10May 1951,the HighCornrnissioneragainrefersto the permits, stating
that theinstructions issuedto the Assistant DistrictCommissionerwould be rnaint~iined.A~
proper interpretationof this statement,along with other ambiguousphrases containedin the
letter,musttake accountof the delicate situationcreatedby SouthAfiica's continuingrefusal
to acceptUNjurisdiction.
260. It may be recalled that South Afiica at first sought to enlist UN support for the
annexation ofNamibia. Whenthis move failed, South Afiica declared thatit no longer had
international obligations with respectto the area,nce in its view, the Mandate had lapsed.
In response, on6 December 1949, the General Assemblyrequested an Advisory Opinion
fiomthe Court onthe statusofthe tenitory of SouthWest~fiica.~~
261. The Court confirmed that South Afiica continued to have obligations under the
Mandate. On the question whether South Afiicacould unilaterallymodifj the status of the
" NCM, AM~X22.
62NM,Amex 69.
63NM, Annex 66.
When the issuewas remittedto the local administratorsfor resolution,Trollopeadamantly
refused tocountenancethe ideathatthe Masubia shoulcdultivateontheIslandbypermissionof the
Britishauthorities.SeeNM,Vol IV, Annex 69.
65International tatusofSouth-WesAt fnca,Advisoryopinion,'I.C.J. Reports1950p. 128Territory, the unanimous opinion of the Court was negative. According to the Court, 'the
cornpetence to determine and modifv the international status of the Temtory rests with the
Union of South Africa acting with the consent of the United (emphasis added)
Nevertheless, South Ahca continued to de@the authorityof theUnitedNations.
262. The legal and political uncertaintycreatedby SouthAfiica's disputewith the UN had a
direct impact on the ongoingnegotiationsover the boundary at Kasikili Island. Although the
British govermnent had decided, in October 1949,to opt for an administrative arrangement
rather than a forma1agreement,67it did not immediately inform South Afiica of this decision.
The decision was taken only two months before the General Assembly's request for an
Advisory Opinion, and negotiations over the request were undoubtedly proceeding in New
York. As noted above, the Advisory Opinion and SouthAfrica's responsedid not resolvethe
political uncertainty,and Britain continued to postpone its officia1response to South Afiica.
It was only on 10May 1951 - nearly 18months after the initial British decision and almost
a year afier the Court's Advisory Opinion- that Bntain finally communicatedits decision to
South Ahca.
263. The delay can be readily explained. Although Britain disagreed with South Afiica's
behaviour towardsthe UN and with its racist policies at home, it could not affordto challenge
them openly. South Afiica was a founding member of the Commonwealth and played a key
economic and strategic role within the British Empire. On the other hand, because of the
pressure of domestic and internationalpublic opinion, the British government had to proceed
with extreme caution so as to avoid giving the impression that it was supporting or adding
fuel to South Africa's conturnacious behaviour towards the United ~ations.~' Hence the
arnbiguous and somewhat ellipticaltone of the High Commissioner's letter of 10 May 1951.
The letter accepts South Africa's proposa1to confirm the status quo at Kasikili Island, but it
is worded in such a way as to pre-empt possible allegations that Britain and South Afiica
66Id., p. 144.
67NM, Annex68; BM, Annex28.
The General Assembly's decisionto askthe Court foranAdvisory Opinion posed a dilemmato
Great Britain. As an important colonialpower it naturallyhadn interestin the Court'sOpinion.
Accordingly,the ColonialOfficeproposed thatBritain shouldintervenein the proceedings. The
Cabinet,however,decidedagainstit mainlybecauseof the likelihoodthat such interventioncould
be construedas implyingsupportof SouthAfrica'sposition. (ReplyAnnex 15) For an outlineof
Britain'spolicytowardsSouthAfricaseetheCabinetMemoof25 September1950. (Id.)were in any way determiningor modifjmg any part of the tenitory of Namibia without the
consent of the United Nations. The crucial point is, however, that both the officials in
Londonand the High Cornrnissioner agreedto the substanceof SouthAiïica's proposal,i.e.,
to maintainthe status quo at Kasikili Island. Theensuingnegotiations that culrninatedin the
Trollope-Dickinsonadministrativearrangementreflected the status quo: 'that KasikiliIsland
continueto be used by Caprivitribesmenandthat theNorthem Waterwaycontinueto beused
asa "free for all" thor~u~hfare.'~~
(c)The allepationthat bothpartiesrecomized the northem channel asthemain
channel forthe Durposeof the Anglo German Agreementof 1890
264. This allegation consistsof two falsepropositions. Thefirst is that 'British officialshad
at al1stagestaken thepositionthat the northem channelwas themain channel for thepurpose
of the Anglo German Agreement of1890.'~~The secondis that '[Altno stage did any South
Africanofficia1challengethe view that thenorthem channel constitutedthemain channelfor
thepurposesof the Anglo Geman ~greement.'~'
265. The inferenceBotswanawould likethe Courtto draw fiomthesepropositionsis that the
outcome of the negotiations was the recognition of the northem channel as the boundary
around Kasikili Island. The available evidenceshows, however, thatneither Britain nor
South Afiica regarded the mistaken finding of the Trollope-Redman Report about the
northem channelas concIusive.
266. As explainedabove,Britainrefi-ainedfrom enteringinto a forma1agreementwith South
Ahca becauseof thepoliticaluncertaintyregardingthe Mandate. In the event, however, the
administrative arrangement confirming the status quo, worked out by Trollope and
Dickinson,was endorsed by senior political authoritiesof the Bechuanaland ~rotectorate.~~
Thus, at no stage duringthese negotiations did Bntain recognizethe northem channelas the
main channelfor the purposesof the 1890Treaty.
69NM,Amex 71.
'OBCM,para. 737.
" BCM,para. 47.
NM,Amex 50.267. South Afiica, for its part, also did not regard the findings of the Trollope-Redman
Report on the main channelas conclusive. Indeed,its proposa1to formalizethe status
shows that for South Afiica the decisive factor ineterminingthe boundary was the use of
the Island by theMasubia. That this view was firmlyheld by SouthAfiica is establishedby
its intemal c~rres~ondence.~~Botswana relies heavily on a letter fiom Trollope to his
superiors in whichhe offers his views on thealternatives opento the Governmentregarding
the boundary around Kasikili ~sland.~~ This letter does not support Botswana's case,
however. Trollope's basic point is that a failure of South Afiica to retain Kasikili Island
withinNamibia would amountto a derelictionof duty.76 .
268. TrollopeandRedmanreportedon two matters: theirview as to the locationof the main
channel and the facts as to the use of the Island.n the first, their faultymethodology led
them to the wrong conclusion. On the second, they confirmed the long,uninterrupted and
undisputed use of the Island bythe people of the Caprivi. At the beginning of the century,
when Eason was instructedto carry out a survey of the boundary along the RiverChobehe
also reported on these two matters. Likewise, during the negotiations leading up to the
TrollopeDickinson arrangement, both Britain and South Africa regarded the long
uninterrupteduse of KasikiliIsland as crucialin the determinationof the boundary. In doing
so they weregiving effect tothe 1890Treatyby applyingits generalterms to a specific area
alongthe River Chobe.
(d) Conclusion
269. It has been demonstrated thatBotswana's interpretationof the diplomatic exchanges
that ledto the Trollope-Dickinson arrangementis wrong. The outcomeof these negotiations
was a continuationofthe statusquothat placedKasikili IslandinNamibia. Britaindid not,at
any time during this period, exercise sovereignjurisdiction over Kasikili Island. Moreover,
the exchanges confirm that Britain and South Afiica continued to regard the use and
occupation ofKasikili Island by thepeople fiom the Caprivi as central in the interpretation
andapplicationofthe 1890Treaty.
73NM, Amexes 63 and 65.
74NM, Amex 62, p. 278.
''BCM,para 745.
''NM ,Annex 61. 3. Botswana's contentionthat the 1984186discussionsbetween Botswanaand
SouthAfiica resultedin an agreementbetweenthe partiesrerrardin~the
applicationofthe An~lo-GermanTreatyof 1890that resolvedtheoresent
boundarvdispute in favour of Botswana
270. Following the shootingincidentof 25 October 1984in the area around Kasikili Island,
Botswana andSouth Africaheld a seriesof discussionsaimed atpreventing the repetitionof
suchincidents. Based upon a misinterpretationof the rules of internationallaw applicableto
the status of Namibia at that time and a distortionof the facts,Botswana contendsthat these
meetingsresultedin an internationalagreement thatresolvedthe boundaryissue in favourof
Botswana.
271. In its Memorial, Botswana characterizedthe outcome of these meetings as subsequent
conduct ofthe parties to the 1890Treaty withinthe terms of Article 31(3)(a)of the Vienna
Convention. Using the wording ofthe Vienna Convention,it argued that 'the international
agreementconcluded at Pretoria constitutes"an agreementbetweenthe parties regarding ...
the applicationof the Anglo-German~greement."'~~The BCM reaffirmsthis interpretation
ofthe 1984186discussions,asserting thattheiroutcomewas legallybindingandconcl~sive.'~
272. Namibia dealt extensively with this contention in its Counter-Memorial where it
demonstratedthat the 1984186discussionscould not haveresulted in a binding international
agreement.'5 At the time of these discussions neither South Afiica nor Botswana had the
legalcapacityto enter into treaty relations inrespectofNamibia. TheUN GeneralAssembly
had terminated the Mandatefor South West Afiica in 1966, and, accordingly, South Afiica
no longerhad anyright to administerthe temtoy or to enter into treaty obligationson behalf
of ~amibia.'~ Moreover, the SecurityCouncilhad detennined that South Afiica's presence
in Namibia was illegal and declaredthat al1actions taken by South Afiica on behalf of or
concerningNamibia afterthe terminationofthe Mandatewere illegalandin~alid.'~
73BM, para. 182and BCM,para. 154.
74BCM,paras. 73, 78, 87.
75NCM,para.107-135.
76G.A. Res. 2145(XXI),27 October,1966,OfficialRecordsof the GeneralAssemblv.Twentv-
First Session,SupplementNo. 16documentAl6316.
S. C. Res.276,30January1 , 970.273. At the same time, Botswana,as a memberof the UnitedNations, had a legal obligation
torefrainfiom dealingswith Southfica whichwould haveimpliedrecognitionof its illegal
occupationof Namibia. In particular, as the Court confirmedin its 1971 Advisory Opinion,
Botswanahad an obligationto abstainfiom enteringinto treaty relationsin al1casesin which
the Govement of South Africapurportedto act on behalf of or conceming~amibia.'~ In
short, neither South Afnca nor Botswana had the legalcapacity to enter into any bilateral
agreementor to apply anypre-existingagreementregardingNamibia. Hence,even if during
the 1984186discussions Botswanaand SouthAfîica had purportedto reach an agreementon
Narnibia'sboundaries, such agreement would have been void ab initioand without legal
effect. In fact, however,as the NCM shows, the two parties had no intention of concluding
aninternationalagreement.s3
274. It will be recalled that followingthe shooting incident of October 1984, a delegation
fiom Botswana travelled to New York for discussions with UN officials and SWAPO
representatives. Botswana,relying on minutes prepared by its own officials, impliesthat at
these meetings the UN authorized it to resolve the boundary dispute at Kasikili ~sland.~~
Namibia demonstrated in its Counter-Memorial thatBotswana neither sought nor obtained
authonzationfiom the UnitedNationsto enterinto anytype ofbinding agreementwith South
~frica.'~ After filing the Counter-Mernorial, Namibiarequested the UN Secretariatto cany
out an archiva1search for any recordor document on the shooting incidentof October 1984
or onany meeting held by Botswana with United Nations' officialsin that time fiarne. The
searchcovered theperiod 1984-1990andincluded documentsmaintainedby the officeof the
Secretary-General,the UN Councilfor Namibia,the Cornrnissionerfor Namibia,the General
LegalDivision and the Office of the Special Representative for Namibia. If there had been
any UN authorizationor approvalof the type suggestedby Botswana,it is inconceivablethat
it would not have been recordedin one or more of these archives. However, this extensive
search produced only one document, an aide memoire of a meeting between Botswana's
82Legal Consequencesfor Statesof the ContinuedPresenceof SouthAfnca in Namibia(South-West
Afnca) NomithstandingSecurityCouncilResolution276 (1970),Advisory Opinion, I.C.J. Reports
1971 , p.58.
NCM,paras. 126-135.
84BM,paras. 181-199; BCM paras.72-78.
85NCM,para. 119.ambassadorto the UN andthe Secretary~eneral.~~This aocument fully confirmsNamibia's
interpretation of the New York meeting and the character of the 1984186exchanges. It
directly contradicts Botswana'sattempt to imply that it had the approval of the UN for the
negotiationswith South Afiica. It shows thatthe New York discussionswere reported to the
Secretary General purely for informational purposes, and that no action of support or
approval was sought by Botswana or taken by the UN. The document also confirms
Namibia's analysis of the 1984186discussions as essentially political in character and
oriented tosecuritymatters. Theiraim was toprevent the repetitionof border incidentsalong
the ChobeRiver bo~ndary.~'
275. Theevidencethus disprovesBotswana'sclaimthat it was assiduousin its consultation
with the UN and had obtainedits approvalto resolve the8boundarydispute." Moreover,its
repeatedassertionthatthe 'jointsurveyreportof 1985'was 'published'89 are alsofalse.
276. Botswana is reduced to a transparently erroneous charge of inconsistency and,
incredibly, to impugning Namibia's good faith in its opposition to the South Afiican
apartheid regime.90 The charge of inconsistency is based on a willfully distorted
interpretation of two passages in Namibia's Memorial. The first states that after the
terminationof the Mandate, South Afiica couldnot take any actionto limit or derogatefiom
Namibia's 15~hts.~'The second is a transcription of a letter fiom South Afiica's Foreign
Minister stating that the South AfiicanGoverrunenthad never recognized Botswana'sclaim
to Kasikili~sland.~~The allegedinconsistencyis that Namibiarelies on SouthAfiica'sletter
to rejectBotswana's interpretationofthe 1984186discussions,while invokingthe termination
of the Mandateto invalidatetheso-calledagreementof 1984186.
86ReplyAnnex6.
87NCM,para. 127.
BCM,para. 77.
89BCM,paras. 119,122.
BCM,para.79.
91NM, para.255.
92NM,para. 256.277. There is, of course, no inconsistency. The letter quoted in Namibia's Memorial merely
Statesa fact: that South Africa never recognized Botswana's claim to Kasikili Island. This
fact, as Namibia has shown, is also established by the rest of the available evidence.
Namibia's contention that the 1984186 discussions could not lead to an international
agreement is based on a careful analysis of the law that deprives the parties of any power to
conclude such an agreement. Both passages are correct, and nothing in one contradicts
anythingin the other. Namibia was entitled to seekclarification fiom the Foreign Minister of
South Afnca, the de facto adrninistering power over Namibia at the relevant time, as to a
point raised by Botswana. In so doing, Namibia did not in any way rely on the so-called
agreementof 19December 1984,which it has consistentlymaintained is void.
278. In the end, Botswana is reduced to an unacceptable attempt to taint Namibia with the
legacy of apartheid:
It is a sad comrnentary that, today, Botswana and Namibia find themselves at
loggerheads over a boundary dispute inherited fiom the days of apartheid
South Africa, which was the defacto power in Namibia until21 March 1990.
It is not beyond the capacity of apartheid South Afkicathat, when anticipating
its own demise, it should decide to bequeath its own legacy of instability to
haunt independent Afncan countries long after it had been eradicated as an
unworkable and unsustainablesystem of govemment.93
279. Botswana's contention that the 1984186discussions with South Africa constituted
subsequent conduct of the parties to the Anglo-Geman Treaty of 1890lacks any substance.
No amount of emptyrhetoric will hide this fact.
Conclusion
280. In its Memorial and Counter-Memorial, Namibia showed an unbroken practice, dating
fiom the signing of the 1890 Treaty until at least the independence of Botswana, that
'establishes the agreement of the parties regarding [the] interpretation' of that Treaty, within
the meaning of Article 31(3)(b)of the Vienna Convention. According to this practice, which
is comprised of continuous acts by Namibia and silence and acceptance by Botswana,
Kasikili Island lay within the Caprivi Strip. This chapter has shown that the arguments on
which Botswana relies to negate this subsequent practice are not supported by the evidence.
The matenals analyzed in this Chapter further confirm that Namibia and its predecessors in
93BCM,para. 130.title exercisedexclusive sovereignjurisdiction over KasikiliIsland with fbll knowledgeand
without objection from Botswana or any of its predecessors in title. The record of the
subsequentpractice unarnbiguouslyestablishes the parties'agreementregardingthe location
of theboundaryin the channelto the southof KasikiliIsland andthe attributionof the Island
to Namibia. ChapterV
THE MAI?EVIDENCE RELATING TO THE INTEPRETATION OF ARTICLE III(2) OF
THE 1890TREATY
281. Twenty-six maps produced by the predecessors in interest of the parties to this case and
showing Kasikili Island have been discussed in the pleadings thus fa.. Of these, 16 show
Kasikili Island in Namibia,ine do not show theboundary and only one apparently showsthe
boundary in the northem channel.' As explained above, Narnibia believes that the most
important maps in this case are those published by the authorities that had political
responsibility for both Namibia and Botswana pior to independence. - Germany, Great
Britain, South Africa and the United ~ations.~ Al1 of these show the boundary in the
southem channel and must be taken as reflecting their understanding of the meaning of
Article III(2) of the 1890 Treaty. Maps published by third parties and Botswana afker
independence carry little weight.In the face of this unusual uniformity and coherence of the
map evidence, the BCM has produced some 50 pages of confusing and obfûscating
discussionofthe mapsthemselves, introducedby an effortto denigrate the significanceof the
evidence of maps in boundary disputes. This Chapter deals first with Botswana's general
legal and theoretical arguments aimed at disparaging the significance of map evidence and
second with some of the more significant of its cornrnents on specific maps. A
comprehensive response in full detail to the BCM's comrnents on specific maps is given in
Annex 1, anexpert report by Col. Dennis Rushworth, formerly Director of the Mapping and
'See Annex1,para. l(b). Themapsare listedin id. PartIV, para. 29 andthe breakdown ishown
onp. 23. Theoneapparentexceptionis SouthWestAfrica 1:50,000 Sheet1725CC Withrnilitary
overprint 1984. Thisis an overprintof SouthWest Africa 1:50,000,Sheet 1725CCEdition 3,
1982,whichdoesnot show theboundary. The 1984 overprintshowsa demarcationline in red in
the northern channel. On the basis of close analysisof thecartographieand rnilitary/political
context, Col.Rushworthconcludesthat theoverprintis not in fact a depictionof the boundary
between thetwocountries,butofthelimitsofpatrolling agreedtobetweenNamibianandBotswana
cornrnanderson the groundafter the shootingincidentof October 1984. Col. Rushworth didnot
includein his listingmapspublishedby Botswanaafter independencebecausethey are theproduct
ofoneof theparties. Namibiahasproducednomapsofthearea sinceindependence.
SeeChapter 1,para. 36,supra.Charting Establishmentat the BritishMinistryof Defence and an acknowledged authorityin
this field.3 It should be noted that BCM Appendix 1, a memorandumpurporting to answer
Col.Rushworth's first report, discloses neitherthe name nor the cartographiccredentialsof
its author. Suchcornrnentsas it containsmust thereforebe regardedas suspect.
A. GeneralLegal andTheoreticalConsiderationsasto the Map Evidence
in this Case
1. Therelevanceandweiaht ofmap evidenceinboundarvdeterminations
282. Citing ProfessorIan Brownlie, theBCM assertsthat rnapevidenceis to be used 'solely
as corr~boration.'~WhateverProfessorBrownlie'smay think,Botswana's approachdoesnot
accord with modem judicial and arbitral practice. The current view of the matter is
authoritatively set outinthe BeagleChannelaward:
Historically,rnap evidencewas originally, anduntil fairly recently, adrnitted
by international tribunals only with agood deal of hesitation: the evidenceof
a rnap could certainly never per se ovenide an attribution made, or a
boundary-line defined, by Treaty, and even where such an attribution or
definitionwas ambiguousor uncertain,rnap evidenceof what it might be was
acceptedwith caution. Latterly, certain decisionsof the International Courtof
Justice have manifested a greater dispositiontoItreat rnap evidence on its
merits. (citinginquiersandEcrehos,I.C.J.Reports 1953,p. 1;Sovereignty
over CertainFrontierLands, IC.J .eports 1959,p. 209;the TempleofPreah
Vihear,I.C.J.Reports 1962,p. 6)'
In the Templecase, it will be recalled, the Court acceptedone particularrnap as definingthe
b~undary.~ It reached this conclusion despite a major error of cartography in placing the
boundary on the rnap and a considerable amount of non-cartographic evidence that the
boundary should havebeen elsewhere.
283. Thepositiontaken in theBeagleChannelcasehasbeen continuedandconfirmedby the
Courtand othertribunals:
Col.Rushworth's qualificationsainNM,Vol. IVAnnex102, p. 53.
BCM,para. 537.
52 InternationLlawReports,atp. 202.
6I.C.J.Reports1962,p. 6. In theFrontier Case (BurkinaFaso/Mali),rnapevidence,alongwith other evidence,was
used to decide how, in general,to allocatethe disputedarea. One particular Frenchrnap
was the major piece of evidence used by the Court as the basis against which to judge
other evidenceand as the frameworkof the actualdelimitationindetail.7
In the portion of the Land, Islands and Maritime Frontier Dispute (El Salvador
ondu dura dealing with the land boundary, the Birdseye rnap was the only rnap
evidence ofconsequence,and fiom it a proposed watersheddefinitionof the westernpart
of the first sector of the boundary was derived in 1935.' The Court acceptedthis line.
Along the remainder of the first sector, and for virtually al1of the second, third, fourth
and fifth sectors, the main evidence was the records of historie land gant surveys
translated onto modem maps by the parties.'0 The cartographie expressionsof the land
grant surveys were, in al1cases, accepted as the basis of the judgement. In the sixth
sector,the disputeconcemed the main channelof theRiver Goascoranthroughits deltaat
the time of theseparation from Spain. Having in pnnciple acceptedthe course claimed
by Honduras, the Court accepted as the delimited line, without any other evidence,the
particulardetailedchannelmarkedon arnapproduced by ond duras."
In the Tabacase (Egypt/Israel),the Tribunalaccepteda particularBritishrnapof 1915as
evidence of the accepted location of the boundary at Taba during the Mandate period
(1922-1 948),which ithad decidedwas thecriticalperiodin determiningthe dispute.'*
284. Of course, the functionof maps differsdepending upon thecircumstances. Somemaps
provide importantbackground information, likeBradshaw'srnap in the present case, which
formedthe basis for the depiction of the area aroundKasikili Island in some later maps.13
Mapsafterthe 1890Treatymay reveal the understandingof one or the otherof the parties as
'I.C.J. Repom 1986, p. 99.
I.C.J. Reports,1992, p. 351.
Id.pp. 411, 421.
'OId.,p. 388.
l1Id.,p. 552.
l2R.I.A.A. Vol. 20,pp. 36, 67.
lBradshaw, itwillbe recalled,fullydepictsonlythe southemchannel,whileleavingthenorthem
channelspeculative.Thismay be someindicationofhisviewofthe comparative importanc of the
two.to the meaning ofthe relevant words of the Treaty. Thus, on the 1909Bechuanalandmap14
the words 'Main Channel' appeah righerup the Chobe,in the vicinityof SwampyIsland,in a
location where, onthe 1889map used at the time of the conclusion of the 1890Treaty,the
words 'Large islandbetweenthe two mainbranches of the Chobe' were written. This change
suggests that in the years irnmediatelyfollowing the Treaty the expression 'main channel'
had nothingto dowithnavigationornavigabilityof theriver.
285. Likewise, the date of the map's publication may affect the weight it carries or the
purposesforwhich it is used. For example,the closerthe time of publicationto the time the
Treaty was signed, as in the case of the Seiner map of 1909 showing the boundary to the
south of the Island, the more it may be thought to reflect the understandingof the parties.
Botswana,however,proposes a 'critical date'of 1948,thebeginningof theTrollope-Redrnan
correspondence,as the date on which the dispute 'crystallized.' The BCM arguesthat maps
published after thatdate should be disregarded so as 'to prevent one of the parties fiom
unilaterally improvingits position by means of some step taken afier the issue has been
definitelyoined.'I5This is a transparent attemptto discredit SouthAfi-ica1:250,000,TSO
4001556,1945149,the first large-scalemodem map of the area,which clearly showsKasikili
IslandinNamibia. The attempt mustfail.I6
First, thereis no general evidentiaryrule precludingthe considerationof maps produced
after some arbitrary date. As the Court said in the Minquiers and Ecrehos case, the
criticaldate should notexcludeconsiderationof subsequentacts of theparties 'unless . ..
taken with a view to improving the legal position of the Party ~oncerned."~The latest
edition of Oppenheim'sInternationalLaw takes a similarly cautious approach to the
matter: 'Courts have . . .been reluctant to accept critical date arguments aimed at
hampering their discretion to look at the whole of the evidence before coming to a
decision.'''
l4NM,Atlas,Map V.
l5BCM,para.538.
l6NM, AtlasMap XII; NM, Amex 102para. 26; NCM, Amex 1 para. 18;ReplyAnnex 1
para. 17.
" ICJ Repom 1953, atpp. 59-60
''R. JenningsandA. Watts,Oppenheim 'sInternationaLlaw(9" ed. 1992),Vol. 1,p. 711. Second, the date of 1948does not representthe crystallization of any dispute. This issue
is fully discussedin Chapter IV atparas. 250-263 above.
Third, the map, although published in 1949, was completed in 1945, when it was
reviewed in draft and comrnented upon and approved by the Bechuanaland a~thorities.'~
Therefore, there can be no suggestion at al1that the map was compiled for the purpose of
advancing Namibia's position in the present case. See also, paras. 303-304below.
286. In fact, Botswana itself does not adhere to the proposition it advances, as can be seen
fiom its reference to its own maps published after 1966.~'
2. Namibia's alleaed 'auandaw'
287. Botswana pretends that there is an 'inescapable' opposition between Namibia's
hydro-geomorphologic expert, Professor Alexander, and its cartographie expert, Col.
~ushworth.~' The asserted 'insoluble conflict' is said to derive fiom what Botswana
describes as 'Professor Alexander's iconoclastic the si^'t^at the main channel is located as
shown on Diagram 4, Sheet 17 of his Main Report and Col. Rushworth's conclusion on the
basis of the map evidencethat the boundary is locatedin the southem channel.
288. The contradiction that Botswanaclaims to see between these two positions is entirely a
fiction of itsown construction. Itfails to appreciate that Professor Alexander and Col.
Rushworth are engaged in two distinct exercises.Professor Alexander's task is to identifj
the main channel of the Chobe River in scientific terms, i.e. the channel that carries the
greater proportion of the annual flow of water in the river. Col. Rushworth's task is to
examinethe manner in which the situationhas been represented in the maps.
289. As to the first task, the identification of the main channel, contrary to Botswana's
assertion,ProfessorAlexander's thesis is not that 'the southem channel is wrongly shown on
al1the maps.'23 It is that 'the main channel' is not shown on the maps, because, as Col.
l9See NM,paras. 312-316 NCM,paras. 162-165.
20BCM,paras. 545-546.
21BCM, paras. 536-536.
l2BCM,para. 535.
23Id. 127
Rushworth observes, 'al1of the maps have been made in the dry season' when there is no
significant flow in either ~hannel.~~As Namibia has shown and as is confirmed by the
additional investigations carried outduring the preparation of the present Reply, the main
channel - the one that carries the largest proportion of the flow- is the one identified by
ProfessorAlexander,andthe northem channelis not. SeeChapter 1,paras. 51-62,above.
290. A furtherand separateaspectof the enquiryrequiredby ArticleIII(2) of the Treatyis to
determinewhere within the main channel the boundary lies. As shown in Chapter1,paras.
75-83 above,the southem channel that appearsin the aenal photographs and on the mapsis
the thalweg channel of the main channel, and the boundary lies within it. Col. Rushworth
shows that those who have prepared maps of the area almost uniformly agree with this
placement of the boundary, and with like concurrence they have regarded the northem
channelas being themain channel. Theasserted 'quandary'disappears. Not onlyis there
no contradictionbetweenCol.RushworthandProfessorAlexander,but thereviewof themap
evidence confirms the location of the boundary that denves from Professor Alexander's
identificationof the mainchannelon the basisofthe scientificevidence.
3. Botswana'sfour general~ro~ositions
291. Botswanaadvancesfour 'general propositions' thatit requeststhe Court tobear in mind
in consideringNamibia'smap e~idence.~~Thesepropositions rangefromthe incongruousto
thetrivial. Theyare addressedseriatimbelow.
292. 'Proposition (i): Onlya Lineplacedinthe riveritselfcanindicatetheb0~ndar-y'~~
This extraordinaryproposition ispresented as a bald assertion without a shred of authority,
either cartographie or legal, to support it. The proposition is disassembledinto two equally
unsupportedand equallyimplausiblecorollaries:
First. that a map showingthe extent of the territory of two neighbouringstates by different
colours without specifically marking the line of the boundary cannot be regarded as
24Id.quotingNM,Annex102para. 40g.
''See BCM,para. 553.
26BCM, p. 223 (heading).For Col.Rushworth's commentonthis propositions ,eeReplyAnnex1,
para.4.evidencing the location of the b~undary.~' The evident objective of ths argument is to
exclude fiom considerationSeiner's map28and the UN map,29which show Kasikili Island as
being Namibian territory by colour, without specifically marking a boundary in the southem
channel. This argument runs counter to cornmon sense. If Kasikili Island is coloured on a
map with the same colour as the territory of Namibia, and the territory of Botswana is shown
in a different colour, it can only be because the mapmakerstook the view that the Island was
in Namibia. Accordingly, the bounndary with Botswana lay to the south of Kasikili Island
and certainlynot to the north of it. This is implicit in Botswana's acknowledgementthat
For present purposes the pertinent characteristic of the boundary is its role in
the allocation of sovereignty. This role has equal operational efficacy both in
relation to land territory and within a nverine milieu. Indeed, for purposes of
legal analysis and the dispositionof sovereignty,here is no distinction. What
is involved is the territorial ambit of the State concemed. and state tenitorv
includes rivers (orparts thereof) and i~lands.'~~(emphasis added)
The secondcorollary is that 'only a boundaryrawn on one or other ofthe channels canbe of
assistance to the Court in its task of determining the "main ~hannel."'~' Again, this
unsupported 'general principle' is obviously constructed in an attempt to dispose of
Bechuanaland GSGS3915 of 1933 and its de ri vat ive This^oo is manifestly absurd. If the
scale of a map is too small to permit drawing the boundary within the river, but is large
enough to show the river parting around Kasikili Island, then the mapmaker's choice to mark
theboundary alongthe channel to the south of the Island is a clear indication that the Island is
inNamibia.
''BCM,para. 554.
NM,AtlasMapIV, NM, Annex102,para. 13;ReplyAnnex1para. 12;para.39, supra.
29Namibia1:1,000,000,UNMapNo. 31581985,NM, AtlasMapXV; NM, Annex102para. 39;
ReplyAnnex1,para. 29.
30BM,para. 7.
31BCM,para. 554.
32BechuanalandProtectorate, 1:500,000GSGS 3915 Sheet'2 1933, NM, Atlas Map M; NM,
Annex102,para.23; NCM, Annex1,para. 16;Reply Annex1,para. 16;paras.300-302,below.293. 'Proposition (ii): Mapsdrawn at too small a scale are not relevant.'33
The statement is a truism. The question is how small is 'too small?' Botswana suggests
without qualification that "no map of scale smaller than 1:100,000 .. . can provide an
accurate portrayal of the geographical features in relation to international b~undaries.'~~
Again, this is an outrageously overstated generalisation. If a river is a boundary, then the
appearanceofthat river on a map is a sufficientportrayal of the relevant geographical feature.
A problem arises only if the scale of a map is too small to show the division of the river
around an island. Kasikili Island, and consequently the division of the River Chobe, appears
on many maps of a smaller scale than 1:100,000. And al1maps that show Kasikili Island, at
whatever scale, are relevant to the argument, because the boundary line on the map can
indicate which branch of the divided river the boundary follows. This includes al1the maps
listed in footnote 4 to the Annex 1of this Reply. The InternationalCourt of Justice and other
internationaltribunals have used maps at smallerscales in similar circum~tances.~~
294. <Proposition (iii): Distortion may result from ~nlar~ement.'~~
Again, the statement is a truism. The only question in this case is whether, in the particular
instances cited by Botswana, the proposition is true. The principal example cited by
Botswana is the UN Map. It notes that 'onthe complete map at proper scale a black dash-dot
line shows the international boundary in the north-western [sic] part of Namibia up to its
junction with the Chobe River; fi-omthat point east to the Chobe's confluence with the
Zambezi, no similar line or international boundary is shown on the map.'37 From which
Botswana would have the Court infer 'that the map-makers at the United Nations did not
depict nor intend to depict any boundary along the length of the ~hobe.'~' The reason for
the dash-dot line in the north-western part ofthe boundary is obvious:thereno natural feature
33BCM,p. 223 (heading). For Col.Rushworth'scomment onthisproposition,seeReplyAmex 1,
para.5.
34BCM,para. 557.
35See NCM, Annex 1,para 6.
36BCM,p. 224 (heading). For Col.Rushworth'scommenton thisproposition,seeReply Annex 1,
para. 6.
37Id. Thereferenceto 'the north-western' paro tf Namibiaseems to be mistaken. Botswanaseems
tobe referringto thesouthwesternboundaryoftheCapriviStrip.
38BCM,para. 560.marksthe border. Furtherto the east, the borderis sufficientlyindicatedby the Chobe River
andthe lightbrown colouring depicting Narnibianterritoe. Inany case, the point Botswana
seeksto makehas nothing to do with enlargement. Theother maps Botswana refersto under
this proposition are discussed in Annex 1,para. 12(Seiner'smap) and para.26 (SouthWest
Af5ca 1:50,000 1982). In every case where an enlargementis presented in Namibia's
Mernorial or Counter-Mernorial,the completemap is available for examination inthe Map
Atlas. Enlargementdoesnothing morethanwhat amagnifjmg glassdoes.
295. 'Proposition (iv): Boundaries drawn on maps are ~nreliable.'~'
Botswana's propositions reachtheir extreme in the assertion that boundaries on maps are
generally unreliable. If it were not so ridiculous onits face, the suggestion would be an
offenceto the many cartographersal1over the worldwho attach immense importanceto the
correct delineation of boundaries on maps and exercisegreat care to achieve it. The maps
referred to by the BCM in support of this proposition are irrelevant on the question of the
location ofthe boundary aroundKasikili Island,not becausethey are inaccurate,but because
they are al1far too small in scale to show it. Moreover,they were introducedinto the case
notby Namibia,but byBotswanain an effort to obscurethe clarityofthe map evidence. Col.
Rushworth'scornments,cited by Botswana, go not to the 'accuracy'of the maps but to the
conventionsfor the representationofboundaries onsmall-scalemaps.
296. It is self-evident that one of the most important purposes of maps is to show the
boundaries between territones. Further, Botswana's proposition flies in the face of the
extensivearbitral andjudicial practice- not least in thisCourt- in makinguse of maps asan
elementin the resolutionof boundary disputes.40Why would Professor Brownliein hiswork
on African Boundaries take pnde in the fact that 'the project involved a great deal of
cartographie ~ork,'~' if he had considered it to be true asa general proposition that
boundariesdrawnon maps are unreliable? As the sameauthorsaysin the Introductionto the
samevolume:
39BCM, p. 225 (heading).For Col.Rushworth's commentonthisproposition,see ReplyAnnex1,
para. 7.
40See paras. 282-283,supra.
41Brownlie,AfricanBoundaries,(1974),p. vii. In matters of evidence, logic and commonsenseare the best guides. Thus a
rnap has probative value proportionate to its technical quality. A privately
published rnap may have as much significance as an officia1 rnap if its
technicalqualityishigh.42
297. The legal and general propositions in the BCM on the topic of rnap evidence are a
farrago of wild over-generalizations, misstatementsand affronts to common sense. They
neednot detainthe Court.
B. Comrnentson SpecificMaps
298. It is not possible in a pleading of reasonable compass to reply in detail to al1the
mistakes of fact, statement and interpretation in Chapter 8, the rnap chapter, of the BCM.
That is reserved for the memorandum 'ObservationsConceming Maps Arising from the
Counter-Mernorialof Botswana' prepared by Col. Rushworth andattached to this Reply as
Annex 1. As an acknowledgedleaderin the field,Col.Rushworth'sauthontyin suchmatters
is not open to question. His memorandum is backed by his considerable experience and
expertise, andthe Court's attention is earnestlydirectedto it. In this Chapter, onlya few of
the most egregiousof Botswana's contentionson specificmaps are addressedin the orderin
whichthey appearin the BCM.
299. The 1904 German Kriegskarte (BMAtlas,Maps4 and 5).43 Thisis an inaccurateand
confusing map, for reasons detailed in NCM paras. 146-151. These characteristics are
sharply illustrated by the BCM's assertion that thernap 'shows the northem channel by a
thick black linend thesouthernchannelis barely visibleexceptas the edge fo theshaded
areawhichrepresentsthe island. . ..[A] strongblack line indicates the thalwegin the river
and is drawnto the north of Sulumbu's Island.'(italicsin original)Almost everyword of this
comment is wrong. The 'shaded area' is shaded m, and one of the most elementary
cartographie conventionsis that blue representswater,not land. Thus the 'shadedarea' does
not 'representthe island' and the 'thick black line'cannot 'showthe northem channel.' The
'strongblack line' is notthe thalweg,as canbe seenfromits erraticcourseto the West,where
it ends after a few kilometres only to resurne further upstream and again in the vicinity of
'Linjanti.' There is no notation indicatingthat this line is the thalweg,and, of course,maps,
42
Id.p. 5.
43BCM,para. 570. SeealsoReplyAnnex1,para. 11.and certainly not Kriegskarten, do not in general depict thethalweg of a river. The label
'Sulurnbu'sIsland' does not attach to any topographieentity on the map, and the symbols
near the label do not make sense. Namibia's original conclusion that the rnap was too
unreliableto includein its own Atlasis arnplyconfirmed.
300. Bechuanaland Protectorate 1933 GSGS 3915 (NM, Atlas, Map IX).~~ The BCM,
para. 591,seeks to diminishthe importanceof this rnapby asserting that 'no original survey
or general verification work was done for this map.' This unsupported statement is
contradicted by an article entitled 'The 1933 Maps of Bechuanaland Protectorate at
1:500,000:A Milestone in the Mapping of Botswana,' byDr. Jeffrey C. Stone, publishedin
Botswana Notesand Records in 1996.~~This carefullyresearched article describesthe care
with whichthe rnapwas produced. ResidentMagistrateswere askedfor, inter alia, detailsof
rivers. The High Cornmissionerwas personally 'concemedto ensure the accuracv of the
Protectorate~undaries.'~~ (emphasisadded)
301. The rnap was widely used by the Bechuanalandauthorities for many years. It would
have been the main source for the statement of the British High Commissioner for
Bechuanaland, Sir Evelyn Baring, in 1949, that 'This [Kasikili Island] has hitherto been
regardedas part of the Caprivi,sincemaps show the main channelpasses to the Southof the
~sland.'~'The High Commissioner clearly did not regard the depiction of the boundary
alongsidethe southemchannelas amere cartographicconvention.
302. The BCM acknowledgesthat the boundaryruns alongthe southem bank of the Chobe
River throughout. It seeks to discount this as reflecting 'standard practice inrespect of
colonialboundariesalongrivers, which locatedthe boundaryline alongside the riverbank at
regular intervals of 10, 50, 100 miles accordingto the scale of the map.' Not an iota of
supportis adduced for this alleged 'standardpractice.' Thevery rnap under discussiondoes
not follow it, since, as noted above, the boundary appearson the south side of the river
throughout,without any variation 'at regular intervals.' Nor does Botswana vouchsafe any
examplesof other maps employingthis practice. It istrue that whena river is represented by
BCM,para. 591.SeealsoReply AM~X 1, para.16.
45 27Botswana Notea sndRecords 71 -84 (1996),reproduced inNCM,ReplyAnnex 8.
46Id.pp. 73-75.
47NM,Annex 66.a singleline or closely spaceddouble-line,on arnap andtheboundaryis in the river, standard
cartographicpractice is to place the boundary symbol toone side of the river. But when the
river divides around anisland, the symbolcan be placed alongside the branch in which the
boundaryis. Thus, in this case,the symbolplaced alongsidethe southem channelat Kasikili
Islandindicatesthat theboundaryis located inthat channel.
303. SouthAfrica 1949TS0/400/556 (NA4Atlas, MapXII).~*Thisis the first modemrnap
of the area, the product of a major cartographic enterpriseby the post-World War II South
Afiican government. The boundary is locatedin the southern channel,and Kasikili Island is
attributedto the Caprivi. The assertionthatthe boundarywas copiedfiom the Bechuanaland
1:500,000rnapof 1933is contraryto the evidenceabout howthernapwas made.49
304. Botswana first soughtto denigratethe importance ofthis rnap for this caseby pointing
out that it was published in 1949, after the so-called 'critical date' of the beginning of the
Trollope-Redrnan discussion^ T.h^ Namibian Memorial showed,however, that the rnap
had been substantiallycompletedin 1945,and a drafthad been circulated forcomment tothe
officiais in Bechuanaland. They made nurnerous .suggestions, but the Resident
Commissioner, to whom responsibility had been delegated, accepted the boundary as
depicted in the southem channel at Kasikili Island without ~bjection.~' In response to this
showing, the BCM produced an extensive interna1 correspondence relating to the rnap
(asserted to be fiom South African archives, but bearing the stamp of Botswana National
Archives),heretofore unavailableto Namibia. Accordingto Botswana,this correspondence
demonstrates 'the self-serving nature of the 1949 map.'52 The documents are collected in
BCM,Vol. III, Annex 17in a confusedand disorganizedway, without chronologicalor other
classification,that is mirrored in Botswana'sexposition ofthe material. Col. Rushworthhas
producedin Annex 2 a complete chronological listingof the items in BCM Annex 17. The
relevant itemsare hghlighted and copies are included in Reply Annex 2. Col. Rushworth
reviews this correspondence, letter by letter in Anne1, para. 17d, and there is no need to
48BCM,paras.601-613.See alsoReplyAnnex1,para. 17.
49See NCM,para. 163.
50BM,para.280.
5'NM,para. 315.
52BCM,para.611.repeat it here. It will sufficeto quote his conclusions:
This conespondence confirms the conclusions about this rnap in the NM:
(i) that in February 1945 the attention of the High Commissioner for
Bechuanaland was drawn to the need to define the boundary with the
Caprivi, that he delegated responsibility for doing this to the Resident
Commissioner and that later he approved the reply, al1 contrary to
BCM para. 608.
(ii) that the Resident Cornmissioner and his staff carefully studied the
request and decided that they had no firm information to contribute on
the boundary on the Chobe River at Kasikili Island, but that it could
most properly be defined by the South Af3can authorities, contrary to
BCM para. 611.
(iii) that four months later the same staff, when faced with the completed
rnap incorporating the boundary that they had just been discussing,
examined and corrected the rnap very carefully but accepted the
depiction of the boundary as correct without comment, contrary to
BCM para. 610.
(iv) that the rnap was completed and accepted in 1945, long before the
Trollope-Redmancorrespondence of 1948, 53 contrary to BCM paras.
542 and 602.
305. Bechuanaland Protectorate 1960DOS(Misc.)282 (BM Atlas, Map 16).~~This rnap
is advanced by the BCM, para. 593 as 'yet another contradiction . . . of the supposedly
"unbroken concordance in the maps of the two parties depicting Kasikili Island as within
Narnibia."' This map, however, is at a scale of 1:1,750,000 (not of 1:1,000,000 as Botswana
asserts, but nearly twice as small), far too small to be of any use. As Botswana admits, 'No
detail is visible in the vicinity of the island.' Indeed, the scale is too small for the island to
appear on the map. In these circurnstances,the fact that the boundary is depicted on the north
side of the Chobe is no more than an application of the practice by which the cartographer
marks the boundary syrnbol on one or the other side of the river line to show that the
boundary is the river. Itdoes not 'contradict' any ofthe maps Namibia relies on.
306. Bechuanaland1:500,000,1965DOS 847, Sheet 2, PM, Atlas, Map 17).55The BCM
rejects Col. Rushworth's view that the depiction of the boyndary in the northem channel was
53Annex 1,para. 17e.
54BCM,para. 593. SeealsoReplyAnnex1,para. 18.
55BCM, para. 594-596. SeealsoReplyAnnex 1,para. 19.a cartographer's error. Instead it asserts that 'the depiction of the boundary in the northem
channel . . . was deliberate in execution of the [Bechuanaland] Surveyor-General's
Opinion. The Surveyor General's Opinion, however, was intended for intemal
consurnption in the Bechuanaland govemment, and there is no evidence that DOS, which
produced the map in England, received a copy. The date of the Opinion is 18 October 1965.
According to an imprintnote on the sheet, the map was Pnnted in September 1965, and was
received in its finished form in the DOS on 20 October 1965. It is hardly likely that the map
could have reflectedthe content of an opinion written in Bechuanaland onlytwo days earlier!
307. Botswana's post-independencemaps?' These maps should be discounted as self-
serving. In any case, it is incorrect to Say that the three maps at 1:1,000,000 show the
boundary in the northem channel. They do not show any island where Kasikili should bey
and therefore they do not show two channels. They show the boundary on the northem side
of a line representing no particular channel of the river. This is in accordance with normal
practice to offset the boundary to avoid confusion, indicating only that the boundary is
somewhere in the river. These maps are thus incapable of manifesting Botswana's views as
to the location of theboundary in any meaningfül way.
308. Third party maps?* TheBCM lists four maps in the U.S.-sponsored Joint Operations
Graphics Series, produced by the United Kingdom afier Botswana's independence. Ofthe
four maps, three are versions of the same map, as the BCM adrnits, and the fourth does not
show Kasikili Island. These are third party maps and for the reasons given in NCM para.
170,shouldbe discounted.
309. The JARIC Map 1974 (?)(BM, Atlas No. 22).59 AS shown in NM para. 217 and
NCM paras. 166-168, this map depicts only one channel in the relevant stretch of the river.
There is no Kasikili Island, therefore no northem channel. The depiction of the boundary
cannot be regarded as a choice between two channels.' Narnibia attributed the error to
56Id.para. 596.
57Id.paras. 620-625.SeealsoReply Annex1,paras. 21-25.
58Id.paras.597-600. SeealsoReply Annex1,para. 20.
59BCM,para. 644. SeealsoReply Annex1,para. 22.improperuse of aerialphotographyon which thernapwas based. The BCM seeksto counter
this suggestionby the assertionthat theJARIC rnapwas 'produced by thecartographie unit
of the South Africa Defence Force, fully plotted fiom air photographytaken by the South
Ahcan Air Force in 1977in accordancewith a WorkingPlan and Report.' This is incorrect.
As Col. Rushworthshows convincinglyin Annex 1,para. 23, the Working Plan and Report
relate to the South West Afiica mapping of 1982 and have no connection with the JANC
map. Namibia's position that this is a poorly drawn rnap fiom an uncontrolled mosaic that
erroneouslyomits KasikiliIsland standsunrefuted.
310. The Military Intelligence Map1984 (BCM, Supplementary Atlas,Map 15).~' As
explained by Col. Rushworth, this rnap is an overprint of South West Africa 1:50,000,
1982.~'The 1982rnapdoes not showa boundary,but on the 1984overprint, a dividing line
is overprintedin red inthenorthernchannel. TheBCM claimsthis indicates 'thatin the eyes
of the SouthAfncan Military authorities,the Island fallswhollywithin~otswana.'~~
311. AgainBotswanais mistaken. We know fromBotswana'sown accountof the shooting
incident on24 October 1984that the SouthAfiican troopswere carryinga rnap showing the
boundary in the southem~hannel.~~That rnapwas therefore the 1984 overprint. It was
probably a rnap that hasjust been found,South West Africa, 1:100,000Director of Surveys
and Mapping Mowbray,1982.~ An extract of this rnap appearsas Fig. 9, followingp. 136,
and a copy ofthe completernaphasbeen depositedwiththe Court. It is ahighquality,large-
scalelinernapwith theboundary syrnbolin the southemchannelbackedby a pwple band of
stipple. It is an authoritative rnap that clearly expressed the South Afiican govemment's
view asto the locationof the boundary. Further,the SADFperemptorilyinstructedthe South
Afncan representatives in the Pretoriatalks in December 1984to 'liase with the Botswana
Govemment through diplomatic channelsin orderto (a) confirmthat the border lies south of
~edudu.'~~Finally, the maps used by the SADF until their departurefiom Namibia in 1989
BCM,paras.616-619. See alsoReply Amex 1,para. 27.
61ReplyAnnex1,para. 26.
62BCM,para. 613.
63SeeBM,Amex 40;MinutesoftheMeeting,afternote, id.Amex41.
See ReplyAnnex 1,para. 27.
65NM,Annex 84, para.9.are still pinned to the walls of their HQ Briefing Room in Windhoek and alsoshow the
boundaryin the southern~hannel.~~ Thusthe 1984overprintcannotrepresentthe viewof the
'SouthAfiican Militaryauthorities'in 1984asto the locationofthe boundary.
312. Col.Rushworth suggeststhat the 1984overprintwas producedin implementationof the
agreementreached on 1November 1984,after the shooting incident, betweenofficersin the
field fromthe two co~ntries.~~Under the agreement, the SADF was to stay to the north of
Kasiluli Island and the BDFwas to operate only to the south of the Island so as to avoid
furtherclashesbetweenthetwo armedforces. It is, therefore,reasonableto concludethat the
red dividingline was overprintedon the 1982rnap and usedby the SADF to indicateto the
troops thelimit of their operationsunderthis agreement.
C. Botswana's GeneralConclusionson Mar,Evidence
313. BCM, paras. 636-646 set out eleven 'general conclusions'as to the rnap evidence,
indicated by Roman numerals. Namibia agrees with none of them. Many of these
'conclusions' havebeen shownto be erroneousin the preceding paragraphsof this Chapter.
In those cases, Namibia refers below to the relevant earlier paragraphs. A few of these
'conclusions' havenotbeen dealtwith earlierandarerefutedbelow.
314. Conclusion 1. BCM para. 636. Botswana's statement that the rnap evidence is
'inaccurate'and 'inconsistent'is patently self-servingand,effectively, meaningless. On the
contrary, the rnap evidence is extraordinarily uniform. Of 26 maps produced by Britain,
Germany,SouthAfnca andthe United Nations,the entitieswith political responsibilityin the
area over the period since the 1890Treaty, 16show Kasikili Island in Namibia and nine do
not show the boundary. The remainingrnap is the one discussed in paras. 310-312,above.
The principal maps produced by each ofthese entities - the Seiner Map, the Bechuanaland
1:500,0001933, the SouthAfiica 1:250,0001945149,and the UN Map 1985 - al1place the
boundaryclearly and unmistakeablyto the south of Kasikili Island. Far frombeing 'oflittle
assistance,'the rnap evidence resoundinglyconfirmsthe analysis based on the words of the
Treaty, the subsequent practice of the parties, and the scientific evidence showing that
KasikiliIslandbelongsto Namibia.
66Affidavitof DennisRushworth, 23 October 1998, ReplyAmex 4.
67Minutesof Meetingon 19December 1984, para. 3(iii),BM,Annex 44.315. Conclusion II, BCM vara. 637. It is no doubt true that the disparity in maps was a
contributory factor to the 1984 border incident. Botswana's forces were using the self-
servingmaps producedby Botswanaafterindependence.
316. Botswana remarks that 'Maps,as the Joint Team found, wereof little assistance in the
determinationof the locationof the boundaryin the Chobe River.' Botswanasoughtthen, as
it doesnow, to diminish the roleof maps and toconfusethe overwhelminguniformityof the
rnap evidence. But if its remark is meant to imply that Namibia in some way agreed that
mapswere of little assistance, thereisothing in the Minutes or the Report of the JTTE that
couldsupportsucha suggestion,andNamibiacategoricallyrejectsit.
317. Conclusion III, BCM vara. 638. There isno conflictwhatsoeverbetween the scientific
evidencepresented by Namibia and the rnap evidence. On the contrary, they aremutually
reinforcing. Seeparas. 287-290,above.
318. ConclusionIV, BCM para. 639. Contrary toBotswana's claim, Article IIIof the 1890
Treaty itself states that 'The courseof the above boundary is traced in general accordance
witharnapofficiallypreparedfor theBritishGovernmentin 1889.' Tobe sure,the boundary
line is not drawn on that map, but it provides considerable evidence as to the parties'
contemporaneousknowledge of the area, and the Courtis surely free to consult it for that
purpose.
319. Conclusion V, BCM vara. 640. Namibia has refuted Botswana's contention that,to
establish a 'subsequent practice' withinthe meaning of Article 31(3)(b) of the Vienna
Convention onthe Law of Treaties, there mustbe express agreementamong the parties. See
Chapter IV, Section A, above. Action by one side, as by publishing an officia1map, will
constitutesubsequentpractice ifthe othersideremainssilent. Thiswaswhat happenedin the
case of the German, South Afiican and UN maps referred to at length by Namibia in this
chapter and in both the Memorial and Counter-Memorial. Moreover, Botswana's
predecessors did notmerely acquiesce. The Bechuanaland1 :500,0001933map, regarded as
a milestonein the cartographichistory of Botswana, was produced byBritish authoritiesand
used bythem for over 30 years. It is an affirmativerecognitionby them that the boundaryis
tothe southof KasikiliIsland.
320. The alleged agreementbetween the Parties on the subject of maps relating to thejoint
surveyof 1985is void ab inita indhas no legalforce.as showninNCMparas. 109-125.321. Conclusion VI, BCM Dara. 641. Botswana alleges that 'no recognition of or
acquiescence in any map placing the boundary in the southern channel was given by either
the British, Bechualaland or Botswana authorities.' The allegation does not accord with the
facts. As noted above, the Bechuanaland 1:500,000 1933map, showing the boundary to the
southof Kasikili Island, was not only acquiesced in, but produced by British authorities.
322. The Seiner map, one of the most important maps supporting Narnibia's position, was
expressly drawn to the attention of the British authorities by the German goverment." In
response, the British colonial authorities sent the German Foreign Office a copy of a map of
the Bechuanaland ~rotectorate.~' The BCM interprets this as a 'repudiation' of the boundary
as set out in the Seiner map. The tmth is quite different. Whatever the British Government
may have been repudiating, it was not Seiner's presentation of Kasikili Island. The intemal
correspondencebetween the British Colonial Office and'the Foreign Office shows that the
area of concem for them was the western part of the boundary with the Caprivi. As to that,
they said, 'Ifthe matter should come before a Court of Arbitration the present [Seiner's] map
might be quoted by the German Government in support of their case, especiallyif a copy had
been comrnunicated to His Majesty's Government and received by them without remark.'70
Thus, the Colonial Office recognized the risks of not protesting (i.e. of receiving the map
'without remark'), concluding that 'for reasons connected with the state of affairs in the
Batawana reserve . . .it does not appear desirable at the present moment to enter into a
general discussion of the question with the German Government.' So a decision was taken
to protest, but instead to send 'the German Government copies of the enclosed map of the
Bechuanaland Protectorate, which indicatesthe view which His Majesty's Government have
held with regard to the proper position of the boundary.' This 'enclosedmap' is on a scale of
1:2,000,000, far too small to show Kasikili 1sland." Thus it could not possibly have served
to put the German Government on notice of a British reservation regarding the boundary in
that location. Al1that it did - and evidently was intended at the time to do - was to reiterate
Threecopieshave been found in the PublicRecordOffice,two in theForeignOfficeand onein
the ColonialOfficerecords. SeeNM, Annex102,p. 9, para. 13.
69NM,Annexes36 and37.
'OId.
7'BM,Atlas,Map6. Thisisthe 1906predecessorof the 1909map.the British view of the location of the western part of the boundary. Far from being a
repudiationof Seiner's depiction of the boundary at Kasikili Island, the episode shows with
stunning clarity that, whereas the British Govement realised the importancegenerally of
protestingunfavourablernapsandthought thatit wouldbe desirableto react to the mannerin
which the westernpart of the Caprivi Strip had been shown, it clearlv did not considerthat
the situationin theneighbourhoodof KasikiliIsland calledfor anv reactionat all.
323. Conclusion VII. BCM 642. Here, Botswana sumrnarilyrepeats its criticism of the
1933Bechuanalandmap, which showedthe boundaryrunning alongthe southembank ofthe
southernchannel, thus placing Kasikili Islandin Narnibia. The matter is dealt with in paras.
300-302,above.
324. Conclusion VIII. BCM para. 643. The provenance of the South Africa 1:250,000
1945/49rnapis reviewed onceagainat paras.303-304,abbve,whereit is conclusivelyshown
that the Bechuanaland authoritiesup to the High Commissioner had ample opportunity to
examine a proof of the rnap in 1945 and were asked for suggestions and cornments,
particularly as to the location of the boundary. Theymade comments, which wereindeed
incorporatedin the final version of the map, but they left the boundary at Kasikili Island
undisturbedin the southemchannel.
325. ConclusionIX, BCM Dara.644. Each of the maps referredto here is discussedinparas.
305-308 above. The Bechuanaland1 :1,750,000 (not1 1,000,000as Botswana says) 1960,is
too small to show Kasikili Island. Bechuanaland1:500,000,1965,does not show the Island
either, because of a cartographic error. The Joint Operations Ground and Air maps were
produced by a third Party, the British Ministry of Defence. And the maps produced by
Botswanaitselfmustbe discarded becausethey are self-interested.
326. Conclusion X, BCM 645. The 1974JARIC rnapdoes not show the boundary in the
northern channel since it does not show Kasikili Island at allIt depicts a single unbroken
streamin the area of the Island and cannot be regardedas having made any choice asto the
boundary. The 1984 Military Intelligence rnap is discussed in detail in paras. 310-312,
above. It is shown that the overprinted lineof divisionwas probably drawn in the northem
channelto ensurethe implementationby SouthAfican DefenceForceof the November 1984
agreement that they would not patrol south of the northem channel, while the Botswana
forcesagreednot to patrolnorth ofthe southemchannel.327. ConclusionXI,BCM 646. Botswanarepeatsits assertionthat the 1985UN map shows
no boundaryin the ChobeRiver. In facttheboundaryis shownby a changeof colour and the
locationof Kasikili IslandwithinNamibiais clearlytobe seen.
D. Conclusions
328. This consideration of the maps has necessarilybeen detailed. But the necessity for
much of the detail has resulted fi-omBotswana's effortsto cloud and to complicate the
Court's task of examining the maps by misleading generalization, confusedand confusing
presentation andmuch error. This effort, however cannotobscure
that animpressiveand almost unbrokenseriesof maps clearly shows theboundaryin
the southemchannelor KasikiliIslandas fallingwithinNamibia;
that the British authorities,Botswana's predecessorsin title, relied on the Seinermap
untiltheyproducedtheirownin 1933 --both ofwhich arepart ofthis series;
that there has never been any protest against these maps by the British or
Bechuanalandauthorities;
that Botswana cannot show a single map before its independence that depicts the
boundaryinthe northern channel.
The few maps that show the boundary in a river course resembling the northem channel
represent carelessness or error on the part of the drafismen, or have been made in special
circumstancesnot calling for South West African or Namibian protest. In any event they
cannotcounter-balancethe weightand significanceof the mapsrelied onby Namibia. PartThree
PRESCRIPTION AND ACQUIESCENCEAS ANINDEPENDENTTITLE
OF SOVEREIGNTY
ChapterVI
PRESCRIPTIONANDACQUIESCENCE
329. Namibia's Memorial set out at length the legal and factual basis for its claim of
prescriptive title to Kasikili ~sland.' already noted, Botswana's Counter-Memorial
seriously distorts this argument. It confuses the issue by reasoning based on circular
arguments, quite 'eccentric' at times, to use a term recurrent in Botswana's pleadings.
Narnibiatherefore feels it necessaryto expose these distortions before restatingits case and
analyzingthe interplayofprescriptionand acquiescence.
A. Botswana's Fallacies.
330. in the first place, as already noted, Botswana considersthat Namibia is advancing a
contradictory argumentby arguing subsequentpracticeandprescriptionat the sametime. As
Botswanaputs it:
The prescriptionargument must assume thatthe original (treaty-based)title of
Botswanahas been displacedby aprocesshaving noconnectionwiththe 1890
Agreement. 'Subsequent practice' cannot claim its status if the matenal
adduced is ab initio antithetical to the concept of 'subsequent practice'2
(emphasisin original)
And itconcludes:
iii. The Argument of Namibia based uponPrescriptionis incompatiblewith
the Principleof Subsequentconduct3
331. Botswanahere ignores thedual nature of the Namibian argument. Fromthe beginning
of the pleadings, Namibia has been crystal clear regardingthe relationship of its argument
'SeeNM, PartTwo, Chapter I(B)II, III,IV.
*BCM, para.139.
BCM,p. 86(heading).based uponprescriptionto its argumentbased uponsubsequentconduct. Paragraph 14 of the
NamibianMemorial statesthe two groundsin the alternative:first, that 'the Anglo-German
Treaty of 1890,properly interpreted, attributesKasikili Islandto Namibia .. .' And, second,
that 'by virtue of continuous and exclusive occupationof Kasikili Island .. .Namibia has
prescnptivetitle to the ~sland.'~Namibiadoesnot acceptthat it is precluded fiom presenting
both grounds forits claim. Al1modern legal systems recognize this possibility, and it is
cornmonplacein this Court. As Namibia argues,eithertlie subsequent conductoperatesas a
'practice ... which establishes the agreement of theparties regarding [the]interpretation'of
the ~reaty;~or it stands as an independent root oftitle based on the doctrine of prescription
anciloracquiescence.
332. Botswana's secondfallacious argument is related to the doctrine of prescription.
Botswana argues that, 'preciselybecause this is a boundary created by treaty, prescription
was, as from 1890, ab initioandipso jure inapplicable.'6Asthe BCMpara. 660 statesit:
Thus, for Namibia,there couldbe an 'adverse possession' asfrom the date of
the conclusionofthe Anglo-German Agreement.Thisposition runs counterto
both legal principle and cornmon sense. In the first place, this position is
inimical to the very concepts of public order and stability in international
relationswhich are recognised inthe legal literatureasjustifjnng the existence
of prescription. Prescription is not intended and was not intended in
nineteenthcenturydoctrine (seeAudinetabove),tobe employedto subvertthe
legalstatusof aboundary expresslycreatedby treaty.'
333. The proposition that prescription could not affect a title established by treaty is a
surprising one, and Botswana suppIies no authority for its position. The principle of
prescription lies in the contradictionbetween the initial legal title and a contrary, peaceful
occupation,overthe years. AsSirRobertJenningsobserves:
Prescription ...comprehendsboth apossessionof which the origin is unclear
or disputed, and an adverse possession which is in origin demonstrably
unlawful.*
SeealsoNM,para. 14(2) .
ViemaConvention on the Law ofTreaties,Art. 31(3)(b).
BCM,para.668.
'BCM,para.660.
Jemings,TheAcquisition of TerritoryinInternationalaw(Manchester,1963),p. 23.334. Botswana argues that the doctrine of prescription against a treaty would 'be inimical to
the very concepts of public order and stability in internationalrelations' and would 'subvert
the legal status of aboundary expresslycreated by treat~.'~ This esoteric view of the doctrine
of prescription is in total opposition to the principle at work. Prescription is exactly about
public order and stability. Far fiom being 'subversive,' the doctrineoperates to stabilize the
defacto situationin the interest of 'publicorder and stability.'
335. AndréAudinet, the author quoted by Botswana, in no way asserts this extravagant
position. He sees no subversion of the international legal order by the operation of
prescription. He simply considers that 'La prescription aurait donc pour but de suppléerau
consentement qui fait défaut.''' In other words, he correctly states that the operation of
prescription compensatesfor the absence of consent of the parties as to the title.
336. The illogical character of the Botswana argument is unrnasked in its contrived effort to
demonstratethat the doctrine of prescription might be applicable &r 1985. It contends that
the settlement of the disputebetween the two parties that, according to Botswana, occurred in
1985 would open the possibility for an adverse possession that was not possible before the
settlement." Stated differently, peaceful and unchallenged occupation could not give rise to
prescription over a period of a century, but a canfiontational occupation afier 1985 could
have had such a result.
337. These logical fallacies only serve to confuse the issue. The fact remains: Botswana has
no effective and peaceful occupation to show and no evidence of subsequent conduct to
present to the Court. Because it cannot present evidence of its own activities with respect to
Kasikili Island, it tries to deride any resort to subsequent conduct, whether as an aid to
interpretation of the Treaty, or as the basis forthe operation of the doctrine of prescription.
B. Namibia Has Met the Requirements for Establishing;Title by Prescription
1. Generalconsiderations
338. As Namibia has continually argued, the 1890Treaty is in the present case basic to the
resolution of the dispute over title to the territory. The Treaty, properly interpreted, fixes the
BCM,para. 660.
'OBCM,para. 656.
" BCM,para. 666.boundary in the centre of the main channel of the Chobe River, that is, the channel to the
southofKasikiliIsland. But thatis not the end of thestory.
339. Boundary makingis a complexprocedure. Itcalls for delimitation andfor demarcation.
But delimitation itself may be more or less precise. Some boundary treaties go into great
detailasto the exactboundary, fix the positionsofboundarypillas, etc. Suchis not the case
with the 1890Treaty. Its temtorial compasswas considerable,delimitingthe zonesbehveen
the two parties in East Afiica, Central Afr-icaand South-West Afiica, not to mention
Heligoland. Drafted by negotiators who only had an approximate knowledge of the
geographicalfacts,it did not purportto be definitivein its conclusions. In fact, it includedin
ArticleVI a clause allowingfor rectification. For instance, as has been shown,Andara was
completely mislocated, whichcalledfor anadjustmentof ihe Treaty.
340. The 1890 Treaty gives rather general indicationsas to the exact boundaryline in the
ChobeRiver and calls for a considerabledegreeofrecourseto aidin interpretation. In such a
situation, where the interpreter is presented with of shades of grey rather than black and
white,oversimplificationwill distortthe processof interpretation.
341. Subsequent conductin such a situation is bound to have relevance to the process of
determining the exact boundary. If it amounts to 'practice,' itis taken into account to help
determine the attitude and intentions of the parties when they concluded the treaty, as has
been demonstrated in Chapter IV of this Reply. Namibia has shownthat the peaceful
occupation ofthe Island by the Masubia of Caprivi for alrnosta century, together with the
administration of theIsland by the traditional Masubia authorities exercising their power
under the authority of the German colonial rulers and of SouthAfiica as the mandatory
power, and thereafterde facto,confirmsNamibia's interpretationof the Treaty asplacing the
boundaryin the channelto the southof the Island.
342. In the present chapter, the same facts, i.e. peaceful occupation of Kasikili Island by
Namibia's predecessors in title, are exarnined in a different legal perspective. Namibia
contends that these facts - peaceful occupation by inhabitants of the Caprivi, duly
authorized by the Germanand South Afiican authorities and without any challenge by the
authoritiesresponsibleforBechuanaland - establishanindependenttitle of sovereigntyover
Kasikili Island on behalf of Namibia through operation of the doctrines of prescription and
acquiescence.343. This record of peaceful and unchallenged occupation by Narnibia's predecessorsin title
is bound to entai1 legal consequences, sometimes described as 'con~olidation."~ As the
arbitral tribunalaid in the Tabacase:
The tribunal considers that where the States concemed have, over a period of
more than fi@ years, identified a marker as a boundary pillar and acted upon
this basis, it is no longer open to one of the parties or to third States to
challengethat long-held as~um~tion.'~
2.Botswana's argument on the 'critical date'
344. Botswana places much emphasis on the alleged importance of the critical date.14 The
BCM Statesits use of the concept in the following terms:
Thepoint is well put by ProfessorBrownlie in the work already cited:
The critical date is a concept linked to the admissibility and weight of
evidence. The critical date is the point at which the dispute has
crystallized and is apparent to the parties. Evidenceemanating fiom the
partiesafter this date is presumed to be self-serving and unreliable.
However, subsequent actions may evidence consistency, and
inconsistent conduct and admissions against interest will be taken into
account. (1.Brownlie, op. citB.,CMAnnex 46).15
345. Namibia does not consider that the concept of critical date adds anything to the present
discussion. In the first place, there are many possible candidates for critical dates.1948-
1951 may be considered as a relevant critical date, but 1966 is certainly another, being the
date of Botswana independencethat triggered the application of the doctrine of utipossidetis
to the dispute, as well as the date of General Assembly resolution 2145(XXI) that terrninated
the Mandate for South West Afiica. 1984-85, the point at which the dispute crystallized
between Botswana and the South-Afiican authorities, could also be considered the critical
date.
l2D.H.N. Johnson, "Consolidation as a Rootof Titlein InternationaLaw",The CambridgeLaw
Journal (1955), pp. 223-225 ;Charles de Visscher, Problèmesde confins en droit international
public(Paris,Pedone, 1969)p.128.
l3I.L.M., 1988,Vol. 27, NO.4, pp. 1489-1490.
l4JïTE, BotswanaSupplementarySubmissions,p. 42, para. 95;BM,para. 281; BCM,para. 540.
l5BCM, para.539.346. Moreover, the effectof the critical date is both complex and somewhat uncertain. It
focuseson the issueof relativeweight ofevidence,but doesnot have a cut-off effect. It may
be interestingto have a 'photograph'of the situationin 1948,beforethe discussions Trollope
had with Redrnan and Dickinson. But Namibia's peaceful useand occupation of the Island
continued unchallenged after that date, right through the 1960s and 1970s and even later.
The undisputed occupation after 1948 clearly was not motivated by the dispute, but was a
continuation of the pre-existing state of affairs. It cannot be ignored in determining
Namibia'stitle on thebasis of prescription and acquiescence.
347. But even if 1948-1951 were to be accepted as the critical date, as Botswana argues,
Namibia contends that its independent title based on prescription and acquiescence had
already beenestablishedwithoutdoubt bythat date.
3. Legalaspectsof the doctrineof'vrescrivtion
348. Namibiahas extensivelyanalyzed thecontemporaryinternationallaw authoritieson the
doctrineof prescription in its~ernorial,'~and there is no needto repeatthat discussionhere.
Inevitably,in internationallaw sucha doctrinecannothavethe sameprecision asin domestic
law. Certain of the requirements of domestic law, such as a fixed period, cannot be
incorporatedinto international law. The versionof the principle reflected in statepractice,
and also largely accepted by courts is a rather more flexible one than that presented by
Botswana.
349. Inthe Right ofPassage case,the Courtstated:
The British did not, as successors of the Marathas, themselves claim
sovereignty, nor did they accord express recognition of the Portuguese
sovereignty, over them. The exclusive authority of the Portuguese over the
villages was never brought in question.Thus Portuguese sovereigntyoverthe
villages was recognized bythe British in fact and by implication and was
subsequently tacitly recognized by India. As a consequence the villages
comprisedin the Maratha grant acquiredthe characterof Portuguese enclaves
withinIndiantemtory."
l6NM, paras. 181-184.
" RightofPassageoverIndianTerritory,I.C.J. Reports,1960, p.39.350. Narnibia believes the present situation is comparable. The 1890Treaty gives a general
indication of the boundary, but a large degree of imprecision remains. The operation of the
doctrine of prescription must take into account these characteristics of the situation.
International law allows for the necessary flexibility to give effect to the doctrine of
prescriptionin such a case.
351. Narnibia considers that the four requirements for prescription stated by Fauchille and
reworded by Johnson have been met. The possession of the prescribing State was peaceful
and unintempted during the relevant penod. The possession was public and has been 'à titre
de souverain,' asdemonstrated in paras. 377-379,below. And the possession endured for the
necessary length of time.
(a). The facts: peaceful occupation and use ofthe Island bv Narnibia.
352. Botswana does not senously challenge the fact of occupation of Kasikili Island by
Namibia from 1890 onwards. It has tried to discredit the oral evidence of some of the
Namibian witnesses before the JTTE. But Botswana itself recognizes the importance of the
evidence given by ~hief ~oraliswani.'~ When the evidence was taken on 30 July 1994,the
Honourable Chief Munitenge J Moraliswani, 80 years old, was Chief of the Masubia people
residing in Eastern Caprivi. He traced the uninterrupted line of succession of Masubia chiefs
back to the period preceding the 1890 Treaty and German mie." He was responsible and
accountable to the Resident Magistrate and Native Commissioner, and thus to the authorities
of the Republic of South Africa, and after independence to the Government of the Republic
of Namibia. Botswana acknowledges that '[hlis evidencewas impressive .. .920
353. Chief Moraliswani's testimony clearly establishes occupation of Kasikili Island by the
Masubia of Caprivi and their economic activities on the ~sland.~~The witness is very precise
inhis statement on this point:
'*BCM, para.484.
''NM,Vol. II, pp. 207-208.
20Id.
21NM,Vol. II, pp. 204-205. JOINT TEAM: 1sthe HonourableChief speakingfiom his own knowledge or
on the basis ofwhat others have toldhim?
HONOURABLE CHIEF:What 1will Sayis that starting fiom 1916, '17, '18,
'19, '20 those are things 1saw with my own eyes and then fiom then, others
might be that they happened when 1was not born and those things 1got them
fiom my great-grandfatherandgrandfather.
CO-CHAIRMAN(NAMIBIA): (interveningto correcttranslation)And father.
INTERPRETER: Yes.
JOINT TEAM: Does the Honourable Chiefknow of anyone who owned land
on Kasikili?
HONOURABLE CHIEF: 1know them, why can 1forget them because 1was
alsoploughingthere.22
354. Such an activity is exclusive of any activity on the Botswana side during the period.
Chief Moraliswani clearly states the situation in response to a question fiom the Botswana
Component :
BOTSWANA COMPONENT:As he was part of the Royal family, does he
know whether there were some people ploughing on Kasikili from the
Botswana side?
HONOURABLE CHIEF: Therewas no ploughing there unless those people
who were working at Kasane,then he come and ask land fiom his familywho
areploughingthere andthey give hirn a place toplough.23
The Chief does not suggest theexistenceof any problemswith the Bechuanalandauthorities
or any protest on their behalf as regards thepeaceful occupation and use of the Islandby his
people.
355. The evidence given by Chief Moraliswani, given special credence by Botswana, is
generallyconsistent with the testimony of the other witnesses presented by Namibia before
the JTTE. Botswana erroneouslystates that 'he clearly contradicted someof his s~bjects.'~~
There may be slight differences of view, in particular as to occupation of the Island on a
2Id., pp.203-204.
23Id.,p.205.permanent basis after 1937. Such discrepancies tend at least to prove that, contrary to
Botswana'sallegation,25there was no collusionbetween the witnesses. But in essence, al1
witnesses confirm occupationof Kasikili Island bythe Masubiaof Caprivi, to the exclusion
ofthe inhabitantsof Botswana.
356. As to the exactdate whenthe agriculturalactivitiesdf the Masubiaof Caprivi ceasedon
the Island, the evidence is not clear. Chief Moraliswani himself is contradictory in his
statement. He answersthe Botswanacomponent'squestionon the date when people stopped
ploughingKasikili Island:
That was in 1937when nowa lot of elephantswere now entering Caprivi and
then, when people were ploughing it was found that those elephants were
destroying their fields,it's when they decidedto move and come to the other
sidehere in ~a~rivi.~~
357. But in his writtenstatement,ChiefMoraliswani declares:
Subsistence farming and fishing was the main activities practised by the
Masubia livingaroundKasikaon this Island. The activities have beenstopped
because of the shooting and killing of my subjectsby the Botswana defence
forces.27
358. The written statement thus tends to fix a date posterior to Botswana's independence.
Chief Moraliswani was certainly impressed by the importance of the 1937 fiood and
disruption ofagriculturalactivitiesthat ensued. Regular agricultural activitiesmay nothave
resumedirnmediatelyafter the flood. That they didresumeseems establishednot onlyby the
other witnesses, but also by the declarations of the local authorities concemed. Although
Botswana claims there is no photographic evidence of habitation on the 1sland,z8the 1943
photographin its Counter-Memonalshows a clearly labelled kraalin the centerof the Island
andfieldsinthe southeastemcorner.29
359. Namibian peacefül occupation and use of the Isbd continued into the 1960s and
1970s, and even the 1980's, in particular through hunting, fishing and pasturing activities
25BCM,para.471.
26NM, Vol. III,Annex2, p.209.
27NM, Vol. III,Annex2, p. 212.
28BCM para.677.
29BCM,p. 163.controlled by the authorities responsible for Eastem Caprivi. Trollope himself hred hunters
to kill elephants that were destroying crops on Kasikili ~sland.~~ The incident of 28
September 1972, discussed at paras. 388-390, below, clearly confirms South Ahcan
authorityon the ~sland.~'
360. Occupation and use of the Island by the Masubia of Caprivi is moreover clearly
confirmedby the British authorities during the entireperiad of colonial rule.
361. Captain Eason reports in 1912concerning 'KissikiriIsland,' that '[tlhe natives living at
Kasika in German temtory are atpresent growing crops on it.'32
362 The Trollope-Redman report in 1948 recognized 'that since at least 1907, use has been
made of the Island by [the] Eastem Caprivi Zipfel tribe~men.'~~ On 4 August 1951,
Trollope's letter embodying the arrangement for the Island between him and Dickinson
states:
(c) That, having regard to the foregoing, the position revert to what was de
factobefore the whole question was made an issue in 1947 - i.e.that Kasikili
Island continue to be [used] by Caprivi tribesmen and that the Northem
Watenvay continue to be used as a 'free for all' [t~horou~hfare.~~
363. In 1965, R.R. Renew, Surveyor General of Bechuanaland, reviewing the Redman-
Trollope arrangement, notes that the factual position accepted by both Govermnents on the
question of occupation of the Island was:
(b) Since the assumption of the German administration of the Caprivi strip in
1907,Caprivitribesmen have cultivated land and generally had the undisputed
use of the Island.
(c) There was no evidence of the island having been made use of, or claimed,
by Bechuanaland tribe~men.~~
30NM,Annex 1,p. 24 (Testimonyof S.M. Ntonda),p. 55 (Testimonyof F.Mayumbelo).
31NCM,paras. 87-91and Annexes23, 24,25, 26 and27.
32BCM,para. 29, quotingBM,Vol. III, pp. 234-235.
33NM AM~X 60, para. 54.
34NM, para. 277.
35BM, Vol. III, Annex36, p. 321.asinvolving'the existenceon the part of thestate exercisingauthorityof a genuinebeliefthat
ithas title.'38Neither argumentcanbe sustained.
(i) The occupationand useof KasikiliIsland discussedabove were under the
authoritv and iurisdictionof the tiowersthat exercisedsovereignrule over the
Caprivi Striuat the time
368 TheNamibianMemorialstatesvery clearlyitsunderstanding that
in order to establish sovereigntyby operation of prescription, acquiescence
and recognition, the claimant must show morethan the use of the disputed
territoryby private individualsfor their private ends. It requires, accordingto
the arbitrators opinion in the Island of Palmas case, "the continuous and
peaceful displayofterritorials~verei~nty.'~~
In the next 15 pages of its Memorial,Namibia establishedin detail that this condition had
been fully met - under German administration, under the British as delegates of the
mandatory power, andunder South Afiica until the time of Namibian i~~de~endence.~ T'he
Court is respectfullyreferredto this extended analysis
369. At the sametime, it is well established thatwhat is consideredto be the exerciseof state
authority must correspondto the nature of the territory in dispute. As arbitrator Dr. Max
Huber held in the Palmas case, 'manifestationsof territorial sovereignty assume different
forms, according to conditions of time and place.'41One cannot expectthe same kind of
exercise of authority in remote and rural areas as in densely populated places readily
accessibleto the seat ofovemmentadministration. KasikiliIslandis very far fiom the local
administrative centres of Schuckmannsberg or Katima Mullilo, and even farther f?om
Windhoek. Duringthe period under consideration,itwas evenmore difficult forthe German
and South Afncan authorities (or the British acting on South Afnca's behalf) to exercise
direct authority over Kasikili Island,given the limited resources and the remoteness of the
38Id.para.698.
39NM, para.218.
40Id.pp. 86-100.
41TheIslandofPalmas (United Statesv.TheNetherlands),HagueCf.Rep. 2d (Scott),p.93;Legal
StatusofEasternGreenland, P.C.ZJ. . ReportsSerieAIB,No.53, p. 46.place. In fact, Germany ruled Eastem Caprivi with only two German civil servants and a
handful of military personnel.42
370 The necessity, in the context of the doctrine of prescription, of calibrating the required
exercise of state authority to the practical situation of the territory involved was stressed in
the Rann of Kutch arbitrati~n.~~As ProfessorBrownlie comments:
Rann of Kutch case, 1968. The Award in this case remarked that in an
agricultural and traditional economy, the distinction between state and private
interests was not to be established with the firmness to be expected in a
modem industrial economy. In an amicultural economv gsazing.and other
economic activities by private landholders mav provide evidence of title.44
(emphasisadded)
371. The opinion of the arbitraltribunal, givenby Chairman Lagergren declares:
The rights and duties which by law and custom are inherent in, and
characteristic of, sovereignty present considerable variations in different
circurnstances according to time and place, and in the context of various
political systems. The sovereign entities relevant in this case prior to
Independencewere, on both sides of the Rann, agricultural societies ... State
and private interests coincided and were necessarily so closely assimilated
with each other that it would be improper to draw as sharp a distinction
between them asis called for in the context of amodem industrial ec~nom~.~~
372. Namibia submitsthat this is the standard that should be applied tothe prescription claim
in the present case and that it has been fully met. There canbe little doubt that the occupation
of Kasikili Island by the Masubia of Caprivi was under the control of their traditional
authorities, who were themselves responsible and accountable to the colonial ~ovemment.~~
The evidence given by Chief Moraliswani also establishes the essential elements of 'indirect
rule' as described in Namibia's ~emorial.~~German occupation in Eastern Caprivi, as we
know, was limited initially to two officials. The Imperia1 Resident, Captain Streitwolf,
carefully studied British practice in Bechuanaland and decided upon a similar method of
42NM,para. 223.
43R.I.A.A.,Vol. XVII,pp. 553-554.
44Brownlie,PrinciplesofPublic Internationa law, pp. 143-144.
45R.I.A.A.,Vol.XVII,p. 554.(emphasisadded).
46NM,para. 86 etseq,para. 218 etseq.
47NMparas. 224-232.admini~tration.~' His successor, Lieutenant Kauhann, continued the policy of indirect
ni~e.~~The chiefs were responsible and accountable to the German Imperia1 Resident.
Namibia's Memorial has arnply documented the conditions of indirect rule, in particular
duringthe Germanperiod.50
373. Material recently discovered in the South Afiican archives gives additional details as to
the operation of indirectrule in the Eastern Caprivi. For example, on 12October 1939, at the
beginning of Trollope's term as ResidentMagistrate andNative Commissioner, the Secretary
for Native Affairs in Pretoria gave him instructions as to his relationship with the Masubia
a~thorities.~'The Chiefs, secretaries and members of the Kutas (or council of indunas) were
to receive annual salaries fiom the govemment.52 In addition, the South Afncan
administration also maintained law and order in the vicinity of Kasikilli ~sland.'~ The
specific attributesof the Island- its size, the fact that it was flooded for part of the year and
that its use was limited to agricultural and pastoral activitie- obviously did not cal1for the
continuous manifestations of state authority. But whenever necessary, the competent
authoritiesdid intervene at a locallevel, for example, to issuehunting permits.54
374. Botswana, after admitting that the Masubia chiefs 'becarne in a certain sense agents of
the colonial authorities' adds an extraordinary and singular requirement for prescriptive
occupation:
there is no evidence, and no evidence is offered, to the effect that the chiefs
had authority to engage in title generating activities. Both legally and
historically, this wouldbe eccentric. . .It is an astonishingproposition that the
chiefscould engage in activitieswhich would subvertthe results of recent and
prolongednegotiationsbetween London and ~erlin.~'
48K. Streitwolf,Der Caprivizipfel(191 l),pp. 26-27,NM, para. 65; NM Annex141, pp.262-263
andtranslationof extractpp. 241-243
49NM,Vol. IV, Annex41, p. 3.
50NM,paras.222-232.
" ReplyAnnex16(a).
52See,for example,theLetterfiom theSecretaryfor Finance,Pretoria, 18September1941,Reply
Annex 17(d). Seealso notesdated2 June 1939,id., 17(a);4 March 1940,id., 17(b);14October
1941,id., 17(c);29May 1943,id., 16(c);13March 1962,id., 16(e).
53Seepara. 378, infra.
54
Seepara. 359, supra.
55BCM, para. 685. (emphasisinoriginal).375. The confusion between occupation in a sovereign capacity and in an international
capacity is bewildering. Al1that is required by the Palmas and Rannof Kutchcases is a
peaceful occupation within a framework controlled by a public authority exercising the
degree of presence called for under the prevailing circurnstances. This degree varies
according to the situation. The capacity to commit a state internationally is quite a different
thing. Nobody expects a member of govemment to show up with his plough in order to
prove peaceful occupation. But he is supposed to react formally in case of infringement of
territorial sovereignty (as Botswana and its predecessors in title consistently failed to do).
The Caprivi chiefs certainly had no authority or competence to make or unrnake international
treaties. But peacefiil occupation of territory does not mean occupation by authorities
competentto engage in internationalrelations. The Resident Comrnissionerin Kasane had no
more competence to commit Britain or Botswanainternationally than Chief Moraliswani had
to commit South Afiica orNamibia.
376. The point is that, by the authority conferred by indirect mle, the traditional structure of
the Masubia of Caprivi - the Chief, Kuta and Indunas - effectivelycontrolled the activities
of the population occupyingKasikili Island during the relevant period in accordancewith the
legal and administrative structure of the time. Whenever necessary, the authorities
intemationallyresponsible for Eastern Caprivi,includingKasikiliIsland, assertedtheir nghts,
as has been demonstrated in the Namibian Memorial. But in ordinary times, indirect mle
through traditional authority was enough to provide evidence of occupation under the
authority.ofthe state. In short, the standardsset in thelmasarbitrationwere met.
(ii) The occupation and use of Kasikili Island by Narnibia's vredecessors in
title was 'atitre de souverain.'
377. The concept of occupation 'à titre de souverain' is ambiguous. The expression is often
used to distinguish situations where occupation cannot be title generating, such as military
occupation, mandate, trusteeship, etc. Botswana does not use the concept in that sense, but
rather as involving a 'genuine belief as to the exercise of sovereignty by the authority
c~ncemed.~~
56BCM, para.698 et seq.378. Again, recent documentation fiom the South Afncan archives fùlly confirms that the
goveming authorities in the Caprivi fully understood that they were exercising sovereign
authority in the area including Kasikili ~sland.~' This is readily apparent in Magistrate
Trollope's arrangementsfor thepolicing of the area, which was a serious problem because of
hippopotamus poachers.58
379. Trollope made arrangements with the Bechuanaland authorities in Kasane for the
secondment of Sergeant Webb, of the Bechuanaland Protectorate Police, as peace officer for
the Union Govemment, in order to effectivelypatrol the Impalila and Kasika area.59 Trollope
had stationed a native constableat Kasika, but as he explains in a note of 30 March 1940, 'it
is almost certain that a number of the poachers are Europeans. It is impossible for the
constable to arrest or take other effective steps to deal with such persons.'60 The Resident
Comrnissioner in Mafeking agreed to second Sergeant Webb with no extra remuneration.
South Ahca reimbursed the extra expenditures incurred by the sergeant for these patrols.61
Given the relationships revealed in the testimony betwee'nKasika and Kasikili ~sland~~ and
the poaching problems on the Island, it is clear that Sergeant Webb's area of jurisdiction
included Kasikili Island. It is significant, in this context, that the South Afiican authorities
were scrupulous about the formalitiesofhaving Sergeant Webb cornmissioned as a constable
for the Union govemment and of reimbursing his expenses. This makes it clear that in
patrolling the area Sergeant Webb was acting in his capacity as a peace officer of the South
African govemment. The exercise of police power in an area is a fundamental attribute of
sovereignty, and certainly evidences 'a genuine belief' that the authority being exercized is
57
See,forexample,ReplyAnnex16(a)(Instructionsfor theadministrationof CapriviZipfel);16(b)
(Proposa1for settingasidetheEasternCaprivi Zipfel as a~ativeReserve);16(c)(Requestforfundsto
coverexpensesfor an officia1visit of the under-secretaryfor NativeAffairs);16(d)(Memorandurn
regardingroadrepairs).
See the Letter of 30 March 1940,ReplyAnnex 18(b)and Trollope'sReport on poachingof 2
September1943,id., 18(e). Seealso theLettersof 3 April 1941,id. 18(d);31 August1948,id.,
18(h);5 January 1957,id., 18(i)regardingthepurchaseofcanoestopatrolthesector.
59For therelevantlegislation,see id.,18(a).
@ Id., 17(b).
61For fulldocumentationseetheLetterof 24 June 1940,id., 18(c). Seealsothe Letters30March
1940, id., 18(b);2 September 1943,id., 18(e);3 April 1941, id., 18(d); 11 January 1944,id.,
18(g).
62SeeNMparas. 204-205, 208.sovereign authority. There is no reason to believe that Trollope's view of his responsibilities
was in any way a departure fi-omprevious practice under the Germans or the British, when
the British were adrninisteringthe mandate on behalf of South Açica.
C. Acauiescence.
380. Absence of opposition is an essential component of prescriptive titleto territory. In the
present case, the authorities responsible for Bechuanaland, and later Botswana, were well
aware of peaceful occupation and use of Kasikili Island by Namibia. They did not object.
Yet the situation was such that it called for a reaction if they believed the Island was within
Botswana territory.
381. Botswana's passive attitude for nearly one hundred years is the more significant in that
it may actually compromise its alleged treaty title. The Court will note that the 1890 Treaty
was not about sovereignty, but about spheres of influence. Its authors did not purport to draw
boundaries, but to delimit spheres. Delimiting spheres did not entai1 extension of
sovereignty, but committed the other partyto the Treaty not to intmde upon the conceded
sphere. Effective occupation was required to perfect sovereignty in the sphere of influence
allocated to a party. According to J.H.W.Verzijl, '[sluch zones or spheres did not thereby
become State territory proper, but as far as they were concemed an inchoate title was vested
in the proclaiming or delimiting State(s)that could gradually develop into full ~overei~nty.'~~
And Lord Salisbury remarked in 1894, '[ilt is not safe these days to establish your title to a
large territory. . and then leave it there without any effort to assert your title in a more
practical and effective fashion. The whole doctrine of paper annexation is in a very fluid and
uncertain condition.'64
382. The British authorities were quick to perfect their title in the area of Bechuanaland.
Long before Captain Streitwolf came to Schuckmannsburg, they established a post at
Kazungula, later moving it to Kasane, less than five kilometres fiom Kasikili Island. If the
British govemment believed that Kasikili Island was within its sphere of influence, it would
63J.H.W Verzij, Internationalaw inHistoricalPerspective,Vol. III, Leyden, 1970,p. 495.
G.N. Uzoigwe, 'Spheresof Influenceand the Doctrine of the Hinterlandin the Partition of
Africa,'Journalof AfncanStudies, Vol. 3 No. 2, 1976at p. 199(ReplyAnnex9). Cf. Western
Sahara, I.C.J. Reports, 1975,p. 56, para. 126; Curzonof Keddleston; Frontiers,Oxford, 1907,
pp. 42 et seq.; M.F. Lindley, The Acquisition and Govemment of Backward Territory in
InternationalLaw, London,1926,pp. 139-147.certainly have taken some steps to perfect its claim. Yet, as shown in Chapter IV, above,
when Captain Eason urged that Britain should claim the Island, his superiors took no action
either in London or in Southern Afiica. Indeed, the authorities in Bechuanaland and,
thereafier, Botswana, never took any step to occupy Kasikili Island until the unilateral
military occupation by the BDF in 1991,more than a century later. They never objected to
the effective occupation of the Island by the Caprivi Masubia under the authority of their
traditional structures, accountable and responsible to the colonial authorities.
This silence
gains added significance, since, in the light of the then prevailing doctrine concerning spheres
of influence, Britain couldnot, as it well lmew,blithelyrely onits 'paper title.'
383. Acquiescence is manifested in disputes over territory in many forms, one of which is
the failure to protest maps showing adverse attribution of the ter~itory.~~In the present case,
the silence of Botswana and its predecessors in title in the face of officia1maps showing
Kasikili Island as part of Namibia is unbroken. The map evidence has been dealt with at
length in Chapter V, above, in relation to the question of the proper interpretation of the 1890
Treaty. And as it bears on acquiescence, the map evidence has been examined at length in
NM Part Two, Chapter V and NCM, Chapter Therefore, only a brief comment is
necessary at this point.
384. The most important fact is that the principal maps produced and used by the officials
responsible forgovemment in the area show the boundary in the channel to the south of the
Island. The first of these maps is the Seiner map, produced by the Austrian cartographer in
1909. Col. Rushworth, on the basis of documentationdiscovered since the Memorials were
filed,now shows conclusively that the Seiner map wasnot only the semi-officia1map used by
the Germans during the colonial period, but that it was relied on and highly regarded by
Bntish officials,both before and during the period when they administered the Caprivi Strip
as delegate of the mandatory power between 1919 and 1929.~~ South African Map TSO
4001556,was drawn in 1945 although it was not published until 1949. It was circulated in
draft for comment to the Bntish Resident Commissioner in Mafeking and the District
65
See,e.g., Case concemingthe TempleofPreah Vihear(Cambodia v.Thailand), 1C.J. Reports,
1962, pp. 30-31.
66Seeespecially NM, paras.287, 325, 326, 333; and NCM, para. 140.
67ReplyAmex 1,paras. 12(d)and 12(e).Commissioner in Maun, among ~thers.~* Although these officials made a nurnber of
suggestions for changes in the vicinity of Kasane, they made no comment or objection with
respect to the boundary line shown in the channel to the south of Kasikili ~sland.~~The rnap
was revised in 1967, after Botswana's independence. Again, there was no comment by the
relevant authorities as to the boundary in the southem channel. The United Nations rnap was
produced in 1985 as part of the UN effort to assert responsibility for Namibia and to preserve
its territorial integrity. It showsasikili Island in Namibia. Although Botswana was, at the
time, a member of the UN Council for Namibia, under whose auspices the rnap was
produced, the rnap contains no disclaimer as to depiction of boundanes. Nor did Botswana
lodge any protest. The absence of reaction or protest fkomBotswana or its predecessors in
title in relation tothese three principal maps amounts to acquiescence to the boundary as
shown on themaps.
385. The first authoritativeBritish map, GSCS Map 3915 of 1933, also traces the boundary
in the southem channel. This rnap continued to be used throughout most of the remainder of
British colonial rule. It was adapted and reproduced on a smaller scaleby the Bechuanaland
authorities in1935." They used it to show Crown lands in 1957 and district boundaries as
late as 1959, only a few years before Botswana's independence.'l The British authorities
who produced the rnap were also the ruling authorities in Bechuanaland. Therefore, the rnap
is not merely evidence of acquiescence to the position of the boundary in the southem
channel- it constitutes a positive act of recognition by Botswana's predecessor in title that
Kasikili Island is within the territoryof Namibia.
386. The failure to protest the series of maps showingKasikili Island in Namibia is only one
aspect of the unbroken record of silence fiom Botswana and its predecessors in title.
Namibia mentions here only a few salient occasions on which some reaction or responsewas
urgently called for Botswana to protect any title it may have had.72 As has been discussed
68ReplyAnnex1, para. 17(b);seealsoNM, Vol. V,Annex106,and NM,Atlas,MapXl2.
69Reply Annex1,para. 17(d).
70NCM,Vol. II,Annex 1,para. 16(e).
71NM,paras. 259-261,307-308; Atlas,MapXI.
72In a differentfield, that of state responsibility,ProfessorAgonotedin his FourthReportto the
Law Commission,that state organshave theobligationto reactto certainsituations
createdby individuals.He remarks that'theactionof the individual can be saidto actas a catalystfrequentlyin the pleadings, as early as 1912,Captain Eason reported to the British authorities
that the Masubia fiom Kasika were cultivatingthe Island. He urged his superiors to claim the
Island on the ground that the northem channel was the main channel. His report was filed
without comment or action by the British colonial authorities, either in London or in
Bechuanaland. With respect to this 'failure to follow up on Eason's recommendation'
Botswana's only answer is that 'there was no cal1for the British Governrnent to "follow up
on" the recommendati~n.'~~ Why not? Especially in view of the contemporary legal
requirements regarding spheres of influence and paper titles, there was every reason for
Britainto protest if it wished to assert title.
387. Again in 1948, the Trollope-Redrnan report concluded that the Masubia of Caprivi had
been cultivating the Island since at least 1907 without objection. The response fiom the
British authorities was not to protest and assert their own title. Instead they accepted an
administrative arrangement recognizing the status quo, 'i.e. that Kasikili Island would
continueto be used by Caprivi tribesmen andthat the Northem Waterwaycontinue tobe used
as a "free for all" [t~horou~hfare.'~~
388. Perhaps the most significant instance of positive acquiescence occurred in 1972 after
Botswana's independence, and involved, not its predecessor in title, but Botswana itself. On
28 September of that year, three Caprivians were arrested on Kasikili Island by game
wardens from the ChobeNational Park andbrought to trial before the Botswana magistrate at
Kasane. The magistrate dismissed the case on the ground that thev were arrested outside
Botswana's iurisdiction. Thus, the casemarks an officia1recognition by the judicial authority
of Botswana that Kasikili Island was not in its temtory.
389. More was to come. The local authorities in the Caprivireported the incident to Pretoria,
and on 22 February 1973,the Department of Foreign Affairs of the Republic of South Afiica
sent a diplomatic note to the Office of the President of Botswana setting out this violation of
forthewrongfulnessof theconductof the Stateorganswhich havenot takenthe necessary steps to
preventtheoccurrenceof suchan action.'Y.I.L.C., 1972, Vol.IIp,.97, note 120.
73BCM, para. 33.
74NM Amex 67; seealsoid. Amex 58.its territorial integrity and seeking clarification. There was no response from the Botswana
govemment. On 13June 1973,the Department sent a reminder. Again therewas no reply.75
390. The incident is comparable to Prince Damrong's visit to the Temple of Preah Vihear.
As the Court said,
A clearer affirmation of title on the French Indo-Chinese side can scarcely be
imagined. It demanded a reaction. Thailand did nothing. .. . Looking at the incident
as a whole, it appears to have amounted to a tacit recognition by Siam of the
sovereignty of Cambodia (under French Protectorate) over Preah Vihear, through a
failure to react in any way, on an occasion that called for a reaction in order to affirm
or preserve title in face of an obviousrivallaim. What seems clearis that either Siam
did not in fact believe that she had any title--and this would be wholly consistent with
her attitude al1 along, and thereafter, to the Annex 1 map and line - or else she
decided not to assert it, which again means that 'sheaccepted the French claim, or
accepted the frontier at Preah Vihear as it was drawn onthe map.76
Namibia submits that, likewise, the failure by Botswana to react to this unambiguous
statement by the Ministry of Foreign Affairs of South Africa - a failure that was 'wholly
consistent with her attitude al1 along' - amounted to a tacit recognition of Namibian
sovereignty over Kasikili Island.
D. Conclusion
391. In this Reply, Namibia has not recapitulated al1the evidence and argument in its earlier
pleadings relating to prescription and acquiescence. It has limited itself to responding to
Botswana's Counter-Memonal, relying, for the most part, on witnesses and authorities quoted
by Botswana during the procedure. From this common ground emerges a clear picture of the
situation in fact and law: Peaceful occupation and use of Kasikili Island by the Masubia of
Caprivi under the control of their traditional authorities and ultimately the authority of the
sovereign state (Germany, Great Britain acting for South Afiica, and South Afiica as the
mandatory power) continued fiom the signature of the treaty in 1890well into the 1960sand
1970sand even the 1980s. Thispeaceful occupation and use of the Island has been operative
for aperiod of almost a century, meeting no protest either from the British authorities or fiom
Botswana itself after independence. It was up to the British governrnent, in the name of
75Theincidenthasbeendiscussedandsubstantiatedby thetextsof affidavitsanddiplomaticnotesin
NCM,paras. 87-91and NCM, Annexes23-27.
76Caseconceming theTempleof Preah Vihear(Cambodia v. Thailand),I.C.J. Reports,1962, pp.
30-3 1.Bechuanaland, and later to the Botswana authorities,to challenge this situation if they
believedKasikili Island waswithin their sovereignjurisdktion. Again and again they failed
to do so. Botswana can show no occupation or assertionof jurisdiction during the whole
periodunderdiscussion. Indeed,in the caseofthe GSCSMap 3915 of 1933and againin the
magistrate's dismissalof the case against the Capriviansarrested on the Island, Botswana
affirmativelyrecognizedthe validityofNamibia'sclaim.
392. The consequence is that Kasikili Island belongs to Narnibia by operation of the
doctrinesofprescriptionandacquiescence. Part Four
OTHER MATTERS
ChapterVI1
NAMIBIANFOREIGN POLICY ANDIT'S APPLICATIONTO THE BOUNDARY
DISPUTE BETWEEN NAMIBIA AND BOTSWANA
A. Introduction
393. The purpose ofthis Chapter is to respond to Chapter 2 of Botswana'sCounter-Memorial
entitled the 'DiplomaticResurgence of theundary Dispute' (BCM,paras. 103 to 130). The
contents ofBCM Chapter 2 are largely a repetition of Chapter IV of the Botswana Memorial.
Most of this material is irrelevant to the issuesbefore the Court, and for this reason Namibia
did not consider it necessary to respond in its Counter-Mernorial. However, so as not to
appear to acquiesce in an incorrect interpretation of these events in the face of Botswana's
repetition,Namibia takes this opportunity to set the record straight.
394. Before addressing Botswana's general discussion, Narnibia wishes to protest one
concrete assertion by Botswana. BCM para. 128 alleges that thernibian Deputy Minister
for Home Affairs did not follow diplomatic channels when he surnmoned the Botswana
District Cornmissioner,based at Kasane, to Kasika (Namibia) to raise the issue of the
ownerslup of Kasikili Island in 1992. Botswana's attacks on the approach adopted by the
Narnibian Deputy Minister for Home Affairs are misplaced. It was actually thevemment
of Botswana that failed to use established diplomatic channels and to follow the procedure
prescribed by internationallaw.
395. The contents of the BCM paras 103-130 should be read against the backdrop of
Botswana's militarist posturing in respect of the disputed area as evidenced by Botswana
unilaterally and illegally sending Botswana Defence Force (BDF) soldiers to occupy
KasikWSedudu Island and hoisting its flag over the Island. Indeed Botswana's present
behaviour and conduct should be put in its historicalperspective. In this connection, Namibia
wishes to bring to the attention of Court Botswana'slong-standing territorial ambitionsagainstthis part ofNamibia. On 9 September 1965 according to theRand Daily Mail, 'Dr
Q.K.J.Masire, then Deputy Prime Minister of Bechuanahd Protectorate, made a speech in
which he said,'. ..Bechuanalandwanted the whole of the stratemc Caprivi Sirir, under its
jurisdiction.(emphasisadded) Dr.Masirewas Botswana'sPresidentfiom 1980to 1998.
396. Botswana's territorial ambitionsas explained above should be related to the 1984
incident, Botswana'soccupation of the Island in 1991, and subsequent incidents along the
border between Namibia andBotswana in the Caprivi Sirip. Al1these incidents occurred
during the presidency of H.E. Sir Ketumile Quett Masire. There is no record of such
govement-sponsored incidents during the presidencyof H.E. Sir Seretse Khama, President
Masire'spredecessor.
397. Unlike the govemment of newly independent Namibia, in 1990 Botswana was
obviously aware that there had been a dispute between Botswana and the govemment of
SouthAfrica over the ownershipof Kasikili Island. Botswana knewnot only that the South
Afhcan governent had exercised defacto control over Kasikili Island (as it did over the
whole of Namibia), butalso that the Island had been regardedas part of Namibia since the
1890 Anglo-German Agreement. The ovenvhelming documentary evidence fiom the
colonial penod left by the adrninistenng colonial authoritiesof the two countries pior to
independenceconfirmsthis stateof affairsbeyond doubt. The docurnentaryevidence is fully
corroboratedby oraltestimony.
398. As long as the South African Defence Force (SADF) was in place, Botswana did
nothing to upset thestatus quo in relation to Kasikili IslanTowards the end of 1989,the
SADFwithdrew fiom Namibia (includingKasikili 1sland)as part of the United Nations -
sponsored independenceprocess. In spite of its knowledge ofthe actual status of Kasikili
Island, Botswana dispatched a contingent of BDF soldiers to occupy Kasikili Island and
hoisted its national flag over the Island in 1991. Botswana took this action without any
discussionor consultation with the govemment of newly independentNamibia. The action
therefore constituted aggression and a unilateral use of force to change the status quo, in
contravention ofinternationallaw,the UN Charter, establisheddiplomaticprocedures,and its
ownclaimed foreignpolicy.
'See Rand DailvMahSouthAfrica,9 September 1965, ReplyAnnex37. B. NamibianForeignPolicy
399. The origins of Namibian foreign policy can be traced back to thedays of the SWAPO
Party when it was still a liberationmovement. It wasSWAPO'Spolicy to maintain fnendly
relationswith countriesand international organisationsopposedto South Afnca'soccupation
ofNamibia.
400. Prior to Namibia'sindependence, foreign relationswere consideredby SWAPO to be
one of themost important policy areas. This status is reflected in Article III of SWAPO'S
Constitution,whichmaybe surnrnarizedas follows:
To CO-operate to the fullest extent with other national liberation movementsand
govermnents sympathetic to the rights of the Namibian people to national
independence, and with organisations and individualsthroughout the world who
supportedtherightsof theNamibianpeopleto self-determination.
To work in close CO-operation with al1peace-loving states towardsworld peace
and ~ecurity.~
401. SWAPO realized thatforeign relationswere an important elementin the struggle for
Namibia'sindependence and that the forum of diplomacy could not be ignored. Through
diplomacy,the organisation succeededin mobilizing fnends and sensitizingthe international
community to the justness of its cause - the right of the Namibian people to self-
detennination. SWAPO establishedoffices in many places such as Dar es Salaarn,Algiers,
Cairo,Lusaka,Lagos, Harare,Francistown(Botswana),Helsinki,Stockholm,Luanda, Bonn,
Paris, Moscow,New Delhi, Melboume, New York and London. In this early period the
organisationwas headquartered in Dar es Salaam. Following the resolutions of the Tanga
Congress(December 1969 to January 1970), SWAPO establishedits Departmentof Foreign
Affairs.
402. The mainobjectiveof SWAPO'SDepartrnent ofForeign Affairswas to forgeits foreign
relationsin a systematicandcoordinatedmanner. Thistask includedthe settingup ofParty
SWAPOConstitution. hblished by the SWAPODepartment forPublicityand Information,
ProvisionalHeadquarters: P O Box 577, Lusaka, Zambia. Printeb dy BrouwerOffset, Delft,
Holland.1976; ReplyAmex 7.offices in countnes sympathetic to SWAPO and establishing accreditation with fnendly
international organisations,includingthe Organisationof Afncan Unity (OAU), the United
Nations, the Commonwealth, the Non-Aligned MovementS , outhern Afiican Development
Community (SADC), and theWorldCouncilof Churches.
403. When SWAPO andthe then apartheidSouthAfrica agreed tothe implementationof the
United Nations Security Council Resolution 435 calling for Namibia's independenceunder
UN supervised elections, SWAPO won the elections and a Constituent Assembly was
established. A total of seven political partieswere representedin the ConstituentAssembly
(1989-1990), which was given the taskof drawing up Namibia'sConstitutionand fixing its
IndependenceDay. Mr Theo-Ben Gurirab, an Assembly member and Namibia'fsirst Foreign
AffairsMinister,inan address tothe Assembly,outlined futureNamibia'sforeignpolicy:
1look at our own situation. Just last yearwe were engaged in a war as a
people. 1do not see, and particularly on this side of the house, anybody
engineering,scheming,contemplatingwar. Weexperiencedit, it is ugly. 1
don'tsee my Namibia to beone scheming, spending millions of the tax-
payer's money toplan war. 1don'tsee that. But not only here in Namibia.
In the whole region of Southem Africa 1 see us moving away fi-omthe
destructivepolitics of confrontationand war. ~he'wordnow in currencyis
negotiations,let us talk. This is what our neighboursare trying to do now
in SouthAfiica,thisis what 1see. Intemationallyalso1see the samething.
We are moving away fi-ommilitarism, irom war. 1 see conflicts being
resolved now through negotiations, 1 see mushrooming of negotiation
meetings everywhere in the world. 1 see super-powers talking to each
other,wining and dining eachother in their capitals. 1don'tsee war on the
world scene. So, letus not be prisoners of the past, let us not be so
subjectiveaboutthe irnmediatepast thatwe seeNamibia beingin a position
to contemplatewar. 1don'tseethat.
1hope that wewill not be spending our time when we convert this House
into the National Assemblyto discuss war. 1hope we would be living in
peace, andallocatingmoney forhousing, for education,forhealth,for good
govemment.
So, 1just don'tsee the threat of war1 don'tsee mat as being part of our
political culturein thefuture.3
SeeNamibiaConstituentAssemblyDebates,21 November, 1989-31 January 1990,
Vol. 1, pp318-319, Reply Annex12. C. NamibianForeignPolicvAfierIndependenceIn 1990 .
404. Afler years of isolation,Namibiabecame a full and active memberof the international
community. Namibia'sinternationalinvolvementstems fiom the conviction,gleanedfi-omits
own history, that Namibia has a responsibilityto promote international cooperation,peace,
securityand respect for internationallaw and treaty obligations. The cardinal importanceof
this conviction is reflected in theive principles adopted by post-independenceNamibia to
govem her foreign relations. These principles are embodiedin Article 96 of the Namibian
Constitutionwhich statesas follows:
The State shall endeavourto ensurethat in its intemationalrelationsit;
(a) adoptsandmaintains apolicy ofnon-alignment;
promotesinternationalco-operation,peaceand security;
(b)
(c) creates and maintains just and mutually beneficial
relationsamongnations;
(d) fosters respect for international law and treaty
obligations;
(e) encourages the settlement of international disputes by
peacefulmeans.
1. Divlomaticrelations
405. SinceIndependence,Namibiahas forgeddiplomatictieswithmost of the memberstates
of the UN and operates a number of diplomatic missions in Afiica (one of which is in
Botswana),Europe,North America(including a missionat the UN), and Asia. A number of
foreign missions have been established in Windhoek, the capital of Namibia. Those
represented include countries fi-omLatin America, North America, Europe, Middle East,
Asia,Australia, Caribbean,andAfrica (includingBotswana).
2. Membershipin Afncan andregionalinternationalorganisations
406. Namibia is a member ofthe (OAU) and is party to the Abuja Treaty establishingthe
Pan-Afican Economic Comrnunity. Namibia plays an active role in promoting regional
integration through SADC, where Namibia is responsible for coordinating work on the
regional fisheries industry. In addition,Namibia is a memberof the Preferential TradeArea
(PTA),the CornmonMonetaryArea (CMA)and theAfricanDevelopmentBank. 3. Membershivin globalinternationalorganisations
407. On the internationaliont outsideAfnca, Narnibia is,in additionto her membershipin
the UnitedNations, a member of a number of UN agencies includingthe World Bank, the
International Monetary Fund, (IMF), the International Labour Organisation(ILO), United
Nations Educational, Scientific and Cultural Organisation (UNESCO), the Food and
AgricultureOrganisation(FAO) andthe World Trade Organisation(WTO). Namibiais also
amemberof the Non-Aligned Movement andthe Commonwealth.
4. Namibia'scornmitmentto internationalveace andsecuritvof al1nations
408. As pointed out above, Namibia is constitutionally committed to encouraging the
settlement of international disputesby peaceful means as well as to ensuring international
peace and security. The Namibian government demonstrated this in 1993 by sending a
contingent of the Namibia Defence Force (NDF) to Cambodia as part of the UN
peacekeepingforce,whereit played a vitalrole.
409. Again in 1996 the Namibian Govermnentsent a contingentof the NDF to Angola as
part of the UN peacekeeping force. In addition, Namibia was among the countries that
contributedfinancially to relieving the plight of the refugees in Rwanda with the goal of
helping to alleviate human suffering and finding a peacekl solution to the problems facing
thatcountry.
D. TheRelationshivbetween NamibiaandBotswanaAfterNamibia's Indevendence
1. TheorininsofNamibia'sforeignvolicv towardsBotswana
410. Botswana was among the first group of countries with which SWAPO sought and
maintainedfiiendly relations, both before and after the Tanga Congressreferred to in para.
401, above. Already in the early 1960's, SWAPO was unofficially represented in the
BechuanalandProtectorateby Maxton Mutongolume,who also served as the Organisation's
spokesperson. Afier Botswana's independence in 1966, Mr Mutongolume worked more
closelywith the DemocraticPart yf Botswana and the Botswana governmentto coordinate
the transit of Namibians fleeing South West Africa to SWAPO headquarters in Dar es
Salaam, Tanzania. Later, under the leadership of SWAPO representatives Uutanga and
Iipurnbu,SWAPOestablished anofficein Botswanawithits seat in Francistown.
411. It is, therefore, not surprising that the government of newly independent Namibiasought to further consolidate friendly and good-neighbourly relations with Botswana and to
make the maintenance of peace and security between Namibia and Botswana a central
objective of its foreign policy. The Namibian Government regards peace and security along
the common border of the two countries as amatter of utmost importance.
412. Since achieving its independence in 1990,Namibia has pursued a policy of cooperation
with its neighbouring states, including Botswana. In the economic realm, Namibia and
Botswana have succeeded in forging cooperation, particularly in the sphere of
communications. The two most significantresults of this cooperation to date have been the
construction of the 1,450 km Trans-Kalahari highway linking Narnibia and Botswana and
cooperation in bringing electricpower fiom Namibia to the westernpartof Botswana.
41 3. In addition to economic cooperation between the two countries, closer cooperation has
also been achieved in the social sphere. Namibia's contributions in this regard include
providing access to school facilities forhildren fiom Botswana living near the Namibian
border in regions like Omaheke (e.g., the Emest Mayer Primary School in the Gobabis area),
and Caprivi (e.g., the Divundu Junior Secondary School and Max Makushe Secondary
School in the Mukwe district).
2. First statevisit andthe simina of coo~erationagreements
414. Soon after the independence of Namibia on 21 March 1990, His Excellency President
Sir Ketumile Quett Masire of the Republic of Botswana was the first Head of State to be
invited to Namibia to pay a state visit in July 1990. President Masire was also the first
dignitary to receivethe Freedom ofthe City of Windhoek in acknowledgement of the fiiendly
and cooperativerelations between the two ~ountries.~
415. On 26 July 1990 during the State visit of Sir Keturnile Quett Masire, three agreements
were signedbetween Narnibia and Botswana: (1) the Protocol of Understanding on Defence
and Security, (2) an Agreement to create a Narnibia-Botswana Joint Commission of
Cooperation, and (3) anAgreement on Cultural and EducationalCooperation.
416. A joint communiquéissued by the two Heads of tat atethe conclusion of the visit
envisioned a comprehensive programme of cooperative relations. Among the areas of
See Timesof Namibia,Tuesday,July 24 1990,ReplyAnnex 38.cooperationidentified were thedevelopmentof transportationand communications,including
specificallythe construction of the Trans-Kalahari andTrans-Caprivi highways to link the
two countries;the establishmentof directflightsbetweenGaborone(the capitalof Botswana)
and Windhoek; the promotion of trade and commercial relations; the development and
utilisation of water resources, in particular the Okavangowater system as well as common
underground water sources; promotion of joint programmes in wildlife and nature
conservation andvetennary control;joint development oftourism; CO-operationin energy
and minera1 exploration and development; and collaboration in the fields of culture,
education, science and technology, the arts and sport by making available facilities and
scholarshipsto students fi-omeach country to study in the institutions of higher leaming,
vocationalcentresandresearch institutionsofthe othercountry.
417. President Sam Nujoma of Namibia made a return state visit to the Republic of
Botswana in September 1990. During the past seven years, numerous high-level officia1
visits have been exchanged between the two countries with the participation of Cabinet
Ministers, members of the Judiciary, and members of each country'sNational Assembly,
includingthe Honourable Speakers.
3. The KasikiliISeduduIslanddis~uteandNamibia'sconduct in accordance
with the protocolofunderstandingon defenceand security
418. Namibia has emphasized its constitutionalcommiünentto peace and security and the
resolution of disputes by peaceful means. The Defence and Security protocol referred to
aboveis a Mer reflectionof this commitment. As such,itis entirely appropriateto discuss
the most recent developments in the KasikiliISeduduIsland dispute with reference to this
agreement,towhich Botswanais a party.
419. The Protocol of Understanding underscores,inter alia, the following three main
objectives:
the need tofurtherstrengthenandconsolidatethe existing excellent relationsbetween
the twoRepublics;
the needto maintain lastingpeace andsecuritybetweenthe two countries, especially
alongtheir common border; and
the needto promote effectiveliaisonat al1levelsinthe fieldsof peace and security.420. In terms of the provisions contained in the Protocol of Understanding, the two
Govemmentsestablished theNamibia-Botswana Joint Commissionon Defence and Security.
The Commissionheld its first meeting in Windhoekon 15November 1990. The inaugural
meetingin Windhoekof theJoint CommissiononDefence and Securityfollowedthemeeting
of the Joint Technical Committeewhich met during 1 and 2 November 1990 in Kasane,
Botswana todiscussmeans for implementingthetermsof the ProtocolofUnderstanding.
421. The Commission,followingin-depthdiscussions,agreedto the following:
To give special attention to the problem of poaching and to implement an
effective jointmechanism aimed at eradicating the menace of these criminal
activities along thecommonborder;
To ensure fiee access to commonrivers insofar as they play a vitally important
role inthe daily economicandsocialactivitiesof the inhabitantsalong theriverine
portion ofthe commonborder;
To avoidat al1costscross-bordershootingincidentsthroughthe swift exchangeof
information on the activities and the movementsof criminals and other persons
threatening the stabilityandsecurityofthe two countries;
To intensity efforts aimed at promoting andprotecting the environment and
wildlife through an information campaign intendedto raise awareness about the
importanceof conservationissues;and
To urge nationals of both countries to use the designated border crossingpoints
foral1movementsbetweenthetwo countries.
422. The next plenary meeting of the Namibia-BotswanaJoint Commissionwas held on 8-
10November 1991in Maun, Botswana. By this time, various sub-cornmitteesto deal with
specializedissueshad been establishedand were fùllyoperationalin both counû-ies.
423. Bilateral agreements andconventionshave alsobeen concludedwithin the fiamework
of SADC and various other regional and national institutions with a view to further
strengthening and consolidating bilateral and regional CO-operationand unity in action.
These agreements constitute importantbuilding-blocksin the comrnon efforts to uplifi the
livingstandardsof the citizensofNamibiaand Botswana.424. The Berlin scramble for Açica, which took place in the 1880'sleft behind as its legacy
a horrendous patchwork of divided communities, arbitrary demarcations, potential political
time-bombs, and unmitigated threats to political and economic security that plague the border
regions of virtually al1Afiican states. The KasikiliISedudu Island boundary dispute, which,
unfortunately,has soured relations betweenNamibia and Botswana, is partof this legacy.
425. Since its independence,Namibiahas had only one significant boundary dispute, namely
the dispute with the then apartheid South Africa concerning the Orange River boundary.
Negotiations on the issue of the Orange River boundary commenced immediately following
Namibia's independence and the matter has since been resolved bilaterally by peaceful and
amicablemeans.
426. Unaware of the possibility of conflict with Botswana over Kasikili Island, the
govemment of Narnibiawas surpnsed when a BDF unit occupied the Island in 1991.' Prior
to these events, no mention of Kasikili Island had been made to Namibia by the Botswana
govemment, in spite of the opportunities provided by the reciprocal state visits referred to
above, which had entailed extensive discussion of issues of wide-ranging bilateral
significance. Nor did Botswana invoke the machinery of the Joint Commission on Security
and Defence.
427. The background is as follows. During contacts in New York with SWAPO
representatives on 27 November and with the then U.N. Commissioner for Namibia on 28
November 1984,Kasikili Island was mentioned by Botswana,but primarily in the context of
claimsregarding South ficals destabilizationof ~otswana and other neighbouring ~tates.~
428. The govemment of Botswana claims that the Kasikili Island issue was settled in 1985
by means of a report by a Botswana-South Afiica joint survey team. That report has been
5
Botswanamaintains thatit hoistedits national fland stationedtroops on the KasikiliIsla'in
early 1989,' a position disputedby Namibia. See the Letter dated 16 February 1996 from
Botswana's Ministerfor Foreign Affairs,the Honourable MompatiS. Merafhe, to Namibia's
Ministerof Foreign Affairs,the Honorable Theo-BenGurirab, Reply Annex10, p. 2. Even if
Botswana's occupationof the Islandtookplace in1989,as allegedby Botswana,the fact remains
that Botswanadidnotoccupy the Islanduntilafterthewithdrawalof SADFfromthearea.
See BM, Vol. III, Annex 41, Minutesof the Meetingheld in New York, 27 November 1984
between theBotswanadelegationand SWAPO.Seealso BM, Vol. III, Annex43, Minutes ofthe
meetingbetween theUN Commissionerfor Narnibia,Mr. Mishra,andthe Botswanadelegationon
28 November 1984 in NewYork.fülly discussed in Namibia's Memorial(paras 284-286), Counter-Memorial(paras. 167-137)
and in this Reply, Chapter IV, paras. 270-279, above. Its findings were inconclusive.
Correspondencebetween the govemment of Namibia andthe Govemment of South Afnca
defeats theclaim that any settlementwas reached. In response to a letter from the Namibian
ForeignMinister, The Honourable Theo-BenGurirab,the SouthAfican Minister of Foreign
Affairs, Honourable Roelof (Pik) Botha, clarified the position of the South Afican
governent in a letterdated 26 April1992:
Thejoint survey you refer to in your letter, a copy of which 1attach,
did corneto certain conclusions,but accordingto the legal opinionin
South Africa at that time, did not prove conclusively that Sedudu
Island belongs to Botswana. The South Afncan authorities have
therefore suggestedto Botswana in a telex dated 17November 1986
that thematter be taken up with the Goverrunentof an independent
Narnibia. This was not acceptableto Botswana as they considered
the joint survey'sreport to be conclusive and in fact expressed the
opinionthat "no Mer discussionon thematteris necessary."
The matter has therefore not beenresolved as SouthAfnca has never
officiallyrecognizedBotswana'sclaimto Sedudu~sland.'
429. Botswana opted instead for unilateral action. After the departure of the SADF, in
contraventionof established principles of international law and without prior consultation,
Botswana deployed BDF units on Kasikili Island and unilaterally hoisted its national flag
overtheIsland. By contrast,Namibia hasadhered tothe fundamentalprincipleunderlyingits
foreignpolicy, namely,its cornmitmentto the peacefulsettlementof disputebetweennations.
Namibia'sdedicationto this principlehas motivatedits participationin every effortto resolve
thismatternon-violentlythroughnegotiation.
430. Thegovemment of Namibia subscribesto the principle that boundariesare established
on thebasis of bilateral agreementbetweentwo neighbouringcountries, and not on the basis
of unilateralpronouncementsor acts. Moreover,as a generalprinciple of internationallaw,
when two countries establish a boundary between themselves, a pnmary objective is to
achievestabilityand finality. It is in this spiritthat Namibia acceptsthe OAUResolutionof
1964 on the status of international boundaries in Afnca, which requires Afncan States to
respect theboundaries inheritedfrom thecolonialera.
'NM, Amex 88.431. Namibia's foreign policy is for peace and against war. Namibia knows well the
devastating effects of war, having experienced over a, century of anti-colonial struggle,
including an anned strugglewaged over23 years againstthe SouthAfricanapartheidregime,
the most vicious military regimeon the Afiican continent. With the welcome demise of the
apartheidsystemin Namibia andSouthAfiica, Namibiahopes that SouthemAfiica, like the
rest of the world,is entering a penod of peace andpolitical stability. Peaceand stability are
necessary preconditions for the cornmitment of collective resources and energies to the
prosperity ofal1peoplesof SouthemAfrica. It is Namibia'sconsideredopinionthat it should
not be difficult to live byhis covenant. Namibia has upheld, and will continue to uphold,
both the letter and the spirit of its Constitution,the OAU Charter, and the Charter of the
UnitedNations.
432. It was with these considerations in mind that in 1992Namibia took the initiative of
asking His Excellency President Robert Mugabe of Zimbabwe, then Chairman of the
Frontline States, to facilitate a dialogue between Namibia and Botswana regarding the
KasikiliIslandboundary dispute. Two earlyroundsof meetings tookplace in Gaborone and
in Anisha, Tanzania, where Their Excellencies President Sir Ketumile Quett Masire of
Botswana and Sam Nujoma of Namibia held consultations in the presence of President
Mugabe.
433. Whereas Botswana continuedto occupy Kasikili Island and fly its national flag over
Namibianterritory, Namibia opted not to aggravatethe situation in any way by word or by
deed. Instead, it once again enlisted the good offices of His Excellency President Robert
Mugabe of Zimbabwe to arrange a further face-to-face meetingbetween President Nujoma
andPresidentMasire. The result was theKasanemeetingof 24 May 1992.
434. At the Kasane meeting, Namibia'spolicy was based on the position that, short of
reachinga bilateral negotiated settlement:
Namibia and Botswana should agree and declare that a border dispute does
exist.
The disputeshouldbe resolvedthroughpeacefulmethods.
Acts ofhostility shouldbe avoidedat al1costs.
Normalcommunalactivitiesshouldcontinue. No unilateral decision should be taken or implemented pending the outcome
of negotiations.
Both sides should refrain from making unilateral public pronoun-cements on
the issue.
Both sides should agreeto abideby the result of the settlement.
Both sides shouldconsult and agree on a mechanism forpeaceful settlement
ofthe dispute.
435. After extensive and frank discussions at Kasane, a Communiqué wasissued stating the
intention to create a Joint Team of Technical Experts (JTTE), three fiom each country, to
determine where the boundary lies in terms of the 1890Treaty. The JTTE reached a deadlock
and recomrnendedthat the parties shouldresolve the boundary dispute 'bypeaceful means in
accordance with international law." It is this recomrnendationthat led the parties to bring the
presentproceedingsbefore this Court.
E. Conclusion
436.
Given the good relationship that existed between Namibia and Botswana - a
relationship that stems from the days when SWAPO was still a Liberation Movement - it is
ironic that the government of Botswana has seen fit to attempt to assert its claim to Kasikili
Island through aggression and unilateral action involving BDF troops. Botswana was able to
talk to SWAPO about the details of apartheid South Afiica's destabilisation policy in the
region in general, and against Botswana in particular, as well as a host of other matters.
Given the apparent fiiendliness of this relationship, it remains unclear to Namibia why the
govemment of Botswana failed to raise the issueof the status of Kasikili Island with the post-
independence SWAPO-ledgovernment in a diplomaticsetting.
437. Between and among the citizens of Namibia and Botswana there are very strong ties of
friendship, brotherhood and historical bonds capable of withstanding threats fiom any
quarter. The govemments of the two countries have a continuing duty and responsibility to
secure,maintain and promotepeace, security and CO-operation between them. It is Namibia's
hope that peace and stabilitybetweenthe two countrieswill be preserved.
A copyofthe ReporthasbeendepositedintheLibraryof theCourtby Namibia'sAgent.438. Namibians of many generations past and present know well the horrors of war. At the
same time, we can never forget the brave martyrs who paid the ultimate pnce for Namibia's
liberty and nationhood during the war of national liberation. In this context, the Namibian
Government has chosen to exercise a maximum of restiaint and has consistently sought a
peaceful solution to the dispute by means of preventive diplomacy based upon negotiation.
In light of the consistency of Namibia'sapproach and its cornmitmentto a peaceful resolution
to the dispute, Namibia considers the fact and nature of Botswana's unilateral actions
especiallydisturbing .
439. BothNamibia and Botswana accept the fact that territorial boundaries mark the limit of
state sovereignty. Boundaries are themselves established by their acceptance by states
sharing cornmon borders. It follows that an unrecognized unilateral assertion of a territorial
claim by force is in conflict with internationally acceptable practice of friendly relations
between states, and contrary to establishedprinciples of internationallaw.
440. By claiming ownership of Kasikili Island ~otswka acted contrary to the 1964 OAU
Resolution AGWRES 16(1), which calls upon al1Member States to 'pledge themselves to
respect the frontiers existing on their achievement of national independence.' It also violated
theUN Charter, Article 2(4), which prohibits the use of force against the territorial integrity
of another state. Botswana's unilateral and illegal occupation of KasikiliISedudu Island
clearly contravenes these fundamentalnoms.
441. To surnmarize, Namibia's foreign policy since independence has been based upon the
followingprinciples:
Promotinginternational co-operation, peace and security;
Creating and maintainingjust and mutually benefi~ialrelations amongnations;
Fostering respect for international law and treaty obligations;
Encouraging settlement of internationaldisputes by peaceful means.
442. Namibia is fully comrnittedto these principles and practices and will continue to work
tirelessly tonsure that these objectivesare achieved and maintained in Africa and
universally. ChapterVI11
MISSTATEMENTS,LEGAL FALLACIES ANDMISREPRESENTATIONS IN
THE BOTSWANA COUNTER-MEMORIAL
443. In the course ofthis Reply, Namibia has refiained fiom commenting on the general tone
and language of Botswana's Counter-Memorial. As the Court will no doubt have noticed,
Botswana's tone is fiequently arrogant, dismissive and disrespectfùl. Its language accuses
Namibia, expressly and by implication, of deliberate 'misrepresentation," 'prevarication,'2
and even subomingperjured testimony of Masubia witne~ses.~These accusationsare wholly
unfounded.
444. The tone adopted by Botswana and its accusations of dishonesty go far beyond the
bounds of proper advocacy. They are inappropriate for communications behveen fiiendly
sovereigns and are unseemly and improper in pleadings before this Court. Namibia has not
replied in kind, but it feels cornpelledto record that itthe most serious exception to the
tone of Botswana's pleadings and especially to the accusations of deliberate efforts to
mislead the Court.
445. Botswana's tone and accusationsarethe moreremarkable because Botswana's Counter-
Memorial itself is tainted by a nurnber of errors and indiscretions. They are particularly
notable and numerous in the discussion of the scientific and map evidence. The most
important of Botswana's errors have been dealt with in the body of this Reply. Many others,
however, although perhaps of less significance to the merits of the case, are illustrative of a
general want of care and accuracy in Botswana's pleadings of which the Court should be
aware. This Chapter collects2 of these erroneousstatements, by no means a completelist.
446. Botswana's errors and misstatementshave been arranged according to their category of
defect. Each item begins with a citation to Botswana's Counter-Memorial and is
accompanied by an exact quotation of Botswana's Counter-Memonal in bold type.
See,e.g., BCM,paras. 259-60,280(regardingNamibia'sscientificargument).SeealsoBCM,
para. 558(regardingNamibia'smapargument).
See BCM, para. 684.
SeeBCM,paras. 471,483, 533(v).Narnibia's refutation follows in ordinarytype. For the convenience of the Court and to
illustrate the variety of errors found in Botswana's Counter-Memonal,Namibia provides a
tableof contentsto this chapter:
A. MisstatementsofFact....................................p.179
B. ManivulationsofEvidence .......................83
C. Misre~resentationsofNamibia's Position.....p. 184
D. LegalFallacies .........................:.....p...86
E. Contradictions.......................................86.......p1
F. Unsu~vortedAssertions................................... 187
G. MavErrors............................................88.......p1
1. ErroneousProvositions ..................p..188
2. MisrepresentationsofNamibia's
Positionon Maps .......................................189
3. MisstatementsConcemin-;
SvecificMavs ...........................p.... 90
4. Errorsof Scaleand OtherTechnical
Descrivtors...............................................p.194
A. Misstatementsof Fact
447. BCM, para. 347(referringto aphotographof the ZambeziQueen,p. 133):
'It is to be noted that the one sizeable ship at present navigating this
section of the Chobe River can only use the northern channel. The
Zambezi Queen, 42 metresin length and three storeys high, is operated as
a tourist ship andsolely uses the northern channel.'
The Zambezi Queen does not 'operate as a tourist ship,' itdoes not 'navigate this [orany]
section of the Chobe River' and it does not 'use the northern channel.' It is permanentl~
mooredonthe northbank of the spurchannelnext to the King'sDen lodge,whereit provides
accommodationfor guestsof the lodge. G.J.Visagie affidavit,ReplyAnnex.22.448. BCM,para. 263:
'The ChobeRiveris an independent perennial river with continuou flow
at al1 seasons of the year through the northern channel around
KasikililSeduduIsland ...'
In over thirteen places in its Counter-Memorial, Botswana asserts that the Chobe River is a
perennial river.4 The ChobeRiver is not a perennial river'with continuous flow at al1seasons
of the year. The River is dry for much of its length upstream fiom Serondela, 13 kilometres
west of Kasikili Island. Itcarries a significant flow of water in the vicinity of Kasikili Island
only seasonallyeachyear when the ZarnbeziRiver overfiows its bank~.~
449. BCM, section heading to para. 279 et seq.:
'Itis correctto describethe southern channel as a stagnant pooo lf water.
Thesouthern channel is abackwater,not a live river.'
Professor Alexander took water discharge measurements in both channels between 30 April
and 2 May 1998. Flow in the northem channel measured 188 m3/s,compared with 247 m3/s
in the southern ~harinel.~In BCM, para. 285, Botswana itself cites measurements that show
flowin the southem channel.
450. BCM, heading to paras. 266-68:
'Themisconception that the Chobe River is paro tf the flood plain of the
ZambeziRiver'
Seee.g. BCM,para. 251('TheChobeRiverisa well-recognised . ..perennialriver');headingto
para. 330('(iii)TheChobe hasa stableprofileas aperennialmatureriver');headingtopara. 331
('TheChobe isa perennial riverwithvisibleandstable banks');para. 331 ('TheChobeisa
perennial matureriver . ..')para. 351('theChobeRiver isa mature perennialriver . ..');
headingtopara. 365('(iv)The Chobe isa perennialRiver.'); para.366('theChobe Riverhas
maintained itscharacteristicsas a perennial riverevidenced,bythedata . . . [from]Upper
Kwandodownto Shaile'); para. 457(v)(b)('theChobe isa perennialriver independentof the
ZambeziRiver');para. 565('theChobe River[is] anindependentperennialriver');Appendix 2,
heading topara. 47 ('TheChobeRiver hasa stableprofileas a mature graded perenniarliver');
Appendix 2,paras. 47-48 ('TheChobeRiver isa perennialriverwitha welldefinedcourseand
withvisible stablebanks').
5
Seee.g .SecondSupp.Rep., Photograph 13 (showingProfessor Alexander's helicopterin thedry
bedofthe ChobeRiver aboveLakeLiambezi)andPhotograph15(showinga personwalkingonthe
dry bedof LakeLiambezi).
SecondSupplementaryReport,Fig . 8.Thisstatementis contradictedat least fourtimes inthe BCM itself--e.g.:
BCM,para.435:
'The MarnbovaRapids are a continuation of the upthnist Chobe Ridge along
which the Chobe River runs for some 100 kilometres . . . .This ridge
represents the southem limit of the Zambezi floodplain,the northem limit of
whichis the Zambeziescarpmentin Zambia.'(emphasisadded)
BM,para.27:
'[Kasikili Island]is part of the seasonalfloodplainwhich is normally covered
withwater fora particularseason.'
BM,para. 244:
'As a result of the floods caused by January/Marchrains al1the land lying
betweenthe upperZambesiandthe lower stretchesof the Chobeup to the first
[Mambova]rapids and beyond formsone vast lake with only here and therea
tree or smallisland appearingabovethewater level.'
BCM,Appendix 4,para. 44:
'The Chobe River occupies part of the Zambezi River floodplain which
extends from the Zambezi escarpment to the Chobe Ridge.' (emphasis
added)
See also the satellite images, NM, Vol. VI, Sheet 25, and Fig. 5 of Professor Alexander's
Second Supplementary Report, showing water from the ,ZambeziRiver flowing across the
floodplain into the Chobe River. These overbank flows are the only significant source of
flowin the ChobeRiver emanatingfiom upstreamofKasikili~sland.~
451. BCM,para. 383(refening tophotograph, p. 153):
'The growth of reeds across the southern channel is reflected in the
photograph taken in September 1997.'
There areno reeds inthe photograph.
452. BCM,para. 279 (refemng to photograph,p. 107):
'The shallownessof the southern channel is illustratedby the photograph,
taken in September 1997, of an elephant standing in mid-channel of the
southern channel ...'
7See also,e-g., ReplyAnnex 32, L.A.McKenzie, TheKalahariReconnaissance(1945) p. 5. ('[an
effecttheLinyantiRiverbelowKachikauispartofthe ZarnbezR i iver,thelandinbetweenking
reallylarge islanoftheZambezi').The elephant is standing not in 'mid-channel,'but close to the south bank of the southem
channel.
453. BCM,para. 290:
'[Tlhe line across the island to which Professor Alexander refers, far from
being a bank of a river, is an extension of the inlet at the eastern end of
the southern channel, and is a low lying sub-channel, dividing the western
higher part ofthe island from the lower eastern sector.'
Thecontourmap producedin connectionwithDr. Sefe's$edimentologicalStudy(reproduced
as Fig. 20 to the Second SupplementaryReport) shows'the lineacross the island' as higher
ground,not as a 'low-lyingsub-channel.' Even theBotswanaDefenceForce knows that this
line is higher groundbecause that is where it chose to build its watch tower and barracks as
shownin Photograph41 in the SecondSupplementaryReport.
454. BCM,para. 466:
'As a matter of fact seulement on the Island was in effect impossible
owing to the fact that the Island is always flooded between March and
May.'
Botswana'smisstatement of fact is betrayed by its own photographic evidence. The 1943
aerial photograph, BCM, p. 163, Fig. 12, labels a 'kraal' in the centre of the Island and
'fields' located near the confluence of the two channels. In addition, Masubia witnesses
testifiedthat they inhabitedKasikiliIsland. Both the Eason report in 1912and the Trollope-
Redmanreport in 1948record that the Masubiaof Capriviwere cultivatingthe Island fkom
thebeginningof the cent~ry.~
455. BCM,para. 334 (see alsoBCM, paras.272,284,366,381):
'Flow in a downstream direction through the northern and western
channel is continuous with a constant level of 925.6 metres, South African
Mean Sea Level, measured at the Kasane Gauging Station through al1
seasons ofthe year.'
Botswanaattaches uncornmonsignificanceto the figure of 925.6 metres, mentioningit five
times in its Counter-Mernorial, althoughin no case does Botswana supply a referencefor the
8
SeeNM,Annexes 1,2, 47and60.. For anexplanationofhowthe Masubia maintained
cornmunitiesinlow-lying,flood-proneareas,seeNM,Annex130. See alsoBM,Vol.III,Annex
22.figure or explain how it is derived. The figure is variously described as a 'minimum le~el,'~
'a constantlevel,'1°a 'constant minimum level'" or just 'a level."* Whatever it is, 925.6
metres cannot be any kind of a 'minimum,' for with a water level at Kasane of 925.6 metres
above sea level, Kasikili Island would be alrnost completely inundated. (See Second
Supplementary Report, Table 1, Fig. 43) Most likely, Botswana carelessly miscalculated.
The lowest level recorded at Kasane between October 1983 and February 1997, as shown in
Table 3, p. 25, of Dr. Sefe's Report, was 2.54 metres above the gauge plate in October 1996.
The gauge plate zero is 922.081 metres above sea level. Adding the two gives a total 924.62
metres (round to 924.6), exactly one metre below Botsaana's magic number. That is the
lowest recorded level at the Kasane station between those dates in terms of elevation above
sea level.
B. Manipulations of Evidence
456. BCM,para. 370:
'[Alexander] himself acknowledges thatthe northern channel, as the
wider and deeper channel, is "hydraulically more efficient and
consequently sedimentis transportedthrough(it)withoutdeposition."'
This is a deliberate misquotation of Professor Alexander. In reality, Professor Alexander
stated:
'The wider and deeper channels are hydraulically more efficient and
consequently sediment is transported through without deposition.'13
(emphasisadded)
The statement appears on a diagram where yellow lines make it clear that Professor
Alexander is referring to the parts ob& the northem and southern channelswhere they are
becoming wider and deeper due to increased inflow.14 There is no warrant for converting
Professor Alexander's 'them,' referring to both channels, into 'it' and eliminating fiom the
quotation the use of the word 'channels,' so thatProfessorAlexander's observation seems to
BCM,paras. 366, 381.
'OBCM,paras. 272, 334.
" BCM,App. 2, para. 43.
l2BCM,para. 284.
l3NM,AlexanderReport,Appendix 2, Sheet18,Diagram5.
l4Seeid.refer only tothe northern channel.
457. BCM, para.300 (see also BCM, Appendix2, para.61):
'The longitudinal slope of the bed is steeper in the northern channel than
the southern. ... Velocity is bound to be higher in the northern channel
than the southern.'
The chartproduced in BCM, p. 129,and Dr. Sefe'sReport, Fig.9, is artificiallytmcated to
eliminatethe last measurement taken in each channelat the confluenceof the channels. By
using the truncated chart, Botswana would like to show a steeper slope in the northern
channel than in the southem channel. The omitted measurement, however, is at the
confluenceof the two channels where the depth is identibalfor each channel,just as it is at
the point oforigin at the bifurcationof the two channels. Thus, the longitudinal slopeof the
bed of eachchannelis actuallythe same.
458. BCM, para. 369(referringto photograph,p. 147,lookingtowardKasika):
'The presence of trees on the right hand (island) bank of the northern
channel also indicates an established stable channel.'
Thetreesin thephotograph arenotonthe Island,but at Kasika. Al1ofthe aenal photographs
showthis lineof trees at Kasika. No photographfiomthe airor groundhas evershownsuch
a line oftrees onthe Islandinthe areacoveredby thephotograph.
C. Misrevresentationsof Namibia'sPosition
459. BCM,para. 2:
'It is thus the arguments based upon subsequent conduct and prescription
which are the essenceof the Namibian legal case.'
Botswana mischaracterizes Namibia's argument. As Namibia has stated in its Memorial,
Counter-Memorial andReply, its case rests upon the interpretation and application of the
1890 Treaty. Prescriptionis a separateand independenttitle to sovereigntyoverthe Island.
460. BCM,para. 684:
'The appearance of German administration in the Eastern Caprivi is
described in the Namibian Memorial and the description is embellished
with a photograph of the installation of Chikamatondo as chief of the
Masubia (Namibian Memorial, pp. 88-93). Not one line, not a single word,
relates to KasikiliISedudu Island. The entire section consists of
prevarication.'The German administration exerted goveming authority over the entire Caprivi Strip,
includingKasikili Island, by indirect rule through local tribal chiefs, such as chikamatondo."
The photograph is direct evidence that Chikamatondo was installed as chief under the
authority of the German administration. Furthermore, many Masubia witnesses testified
before the JTTE that Chikamatondo was their chief and ruled over the Masubia who lived
and cultivated crops on Kasikili Island.16 Therefore, through Chikamatondo and hs indunas
(including Sulurnbu),the GermanadministrationgovemedKasikiliIsland.
461. BCM, para. 255:
'Professor Alexander's identification of the course of the southern
channel [is] not ... shown on al1 aerial photographs,maps and even
satelliteimageries ...'(emphasis in original) *
The main channel as identified by Professor Alexander, with the right bank hard against the
tree lineunder the Chobe Ridge and the left bank marked by the higher ground crossing
Kasikili Island fiom west to east, is visible on al1of the aerial photographs. A selection of
seven of these photographs are juxtaposed for easy comparison in Fig. 22 of the Second
SupplementaryReport.
462. BCM,para. 378:
'[Wlater flow in a river is directionalthrough a channel,not in a wave
form on a 60 kilometrefrontas ProfessorAlexander claims ...'(emphasis
in original)
In none of his reports does Professor Alexander use the word 'wave' or 'wavefonn' or
describe the flood water fiom the Zambezi River as assurning a 'wave form.' He saysthat
the flood water 'crosses the [Zambezi River] floodplainin a southerly direction towards the
Chobe ~id~e,"' a phenomenon that is readily visible on the satelliteimages at Sheet 25, 7d
and 7e,ofthe AlexanderReport andFig. 5 of the SecondSupplementaryReport.
''SeeNM,paras.222-32.
16SeeNM,Annex 1, pp.69 (Testirnonyof M.G. Nchindo),140(Testimonyof C.L. Matondo);
Annex 2, p. 207 (Testimonyof ChiefM.J. Moraliswani).SeealsoNM,para. 228.
17
NM, para. 22 uses slightly different language todescribe the same phenomenon. It says:
'[Water]comeintotheChobenotfrom upstream reaches to thewest,butacrossthe whole width of
the Zambezifloodplain,a frontof over60 kilometres,untilit'isinterceptedby the ChobeRidge. . .
. Here again, no reasonableinterpretationof this descriptionwould indicatethe presenceof a
'wave'-likephenomenon. D. Lena1Fallacies
463. BCM, para. 139;BCM, heading to paras. 247-49:
'"Subsequent practice" cannot claim its status if the material adduced is
ab initioantithetical to the concept of "subsequent practice" ....The
Argument ofNamibia based on Prescription is incompatible with the
Principle of Subsequent Practice.'
Botswana ignores that Namibia's arguments on treaty interpretation and prescription
represent separate and independent claims to sovereignty over the Island. This form of
pleasding is accepted by al1 major legal systems. The incompatibility only exists in
Botswana's mind.
464. BCM, para. 660:
'Prescription is not intended and was not intendedin nineteenth century
doctrine, to be employed tosubvertthe legal status of a boundary
expressly created by treaty.'
Prescription is about public order and stability. It does not intend to 'subvert' alegal status,
but rather to insure stability. It contradictsthe basic idea of prescription to argue that it could
not affect a title established by treaty.
465. BCM, para. 685:
In addressingprescription and the activities of the Masubia, Botswanastates
that 'there is no evidence, and no evidenceoffered, tothe effect that the
chiefs had authority to engage in title generating activities.th legally
and historically, this would be eccentric.' (emphasisin original)
Prescription hasnever been limited to activities engaged by authorities with the capacity to
engage in 'title generating activities.' Prescriptionrests on the fact of occupation and use of
the temtory in dispute under the authority of the claiming state for a substantial period of
time, with the acquiescence of the otherarty, which are quite different matters. E. Contradictions .
466. BCM,para. 353(i):
'KasikiliJSedudu Island consists of two layers, a dark top layer consisting
of a clay, silt and mud admixture extending to about 1.5metres in depth
But compare:
BCM,para. 353(iii):
'deposition of sand, clay andmud to a depth of 1.50 metres.'
BCM, para. 353(iv): '...
'the complete absence of sand in this top laye.. .'
467. The 'May 1972'aerialphotograph
Compare:
BCM, paras. 411-13:
'this photograph taken at a time when water levels were low'; 'at low
water'; 'indicating a verv lowwater level.'(emphasis added)
BCM, paras. 331,380:
'when the island is inundated in the wet season . . . : see aerial
photographs May 1972 ...';'in flood time.' (emphasis added)
Both ofthese descriptions cannot be right.
F. Unsu~ported Assertions
468. BCM,para. 387:
'The itinerary of the ZambeziQueen and the tourist boats from the Safari
Lodges ...'
No such document nor any other evidentiary supportfor these 'itineraries'is produced. For a
brochure of the Chobe Safari Lodge, see NCM, Vol. III, Sheet 17s and Reply Annex 23.
Furthemore, Botswana is guilty of fabrication evidence when it claims that it has seen theZambezi Queenitinerary. As statedearlier,the Zambezi Queenis permanentlymoored onthe
north ban. of the spur channelanddoesnot navigatethe northemchannel.l8
469. BCM,para. 243 (see alsoBCM,para. 97):
'After1960,when agricultureon the island was prohibitedby the British
authorities, no protest ensued, either from diplornaticsources or local
sourcesin the Caprivi,until1992 ... a periodof32 years.'
There was no such prohibition. Botswana provides no officia1document or any other
evidence supportingthe statement that British authorities prohibitedcultivation on Kasikili
Island after 1960. In the absence of such an officia1action, noprotest was called for.Ig In
fact, Botswana abstainedfi-ommaking any clairn over Kasikili Island when in 1972 South
Africa filedan official protest over the arrest of three Caprivians onKasikili ~sland?' Not
only didthe local Botswanamagistratedismissthe arrestfor lackofjurisdiction overKasikili
1sland,*'but Botswana authorities remained conspicuously silent in response to South
Africa7sclear assertionof its authorityoverKasikili~sland.~~.
G. Mav Errors .
1. ErroneousProvositions .
470. BCM,headingto para. 554:
'Proposition (i): Only a Line placed in the river itself can indicate the
boundary'
Conventions of modem cartography deny support for Botswana's contention, nor can
Botswana itself provide any reference or other authorityto supportthis proposition. Thus,
Botswana's related argument - that because some of the maps of the Chobe River near
Kasikili Islanddo not draw the boundary line in the River itself,they arenot relevantto this
case- is entirelywithoutsubstance.
l8SeeAffidavitof G.J.Visagie,ReplyAmex 22.
l9SeeNCM,Amex 25.
20SeeNCM,paras. 88-91.
21See NCM, Amex 24.
22SeeNCM,Annex 26.471. BCM,headingto paras. 558-60:
'Proposition(iii): Distortionmayresultfrom Enlargement'
Enlargementwill result in distortionif thernapis enlargednot in directproportionto the
original. The fourrnapextractsinNamibia'sMemorialare enlarged accordingto standard
cartographicpractice. Theseimages areenlargedin directproportionto the originalsand
retainal1the characteristicsofthe originals. Enlargementaccordingto well-established,
accuratetechniquesis acornmonmethod forpresentingdetailsofmaps to courtsand other
bodieshavingneed for suchdetail.
472. BCM, headingto paras. 561-63:
'Proposition (iv): Boundariesdrawn onmaps areunreliable'
The depiction of boundaries is a major function of cartography, and the cartographers of
officia1political maps-as a general rule exercise great care in the placement of boundaries.
For example,the boundariesshown onmaps formingpart of ICJ Judgrnents areanythingbut
unreliable. NCM, Annex 1,paras. 8-9describehow in the case of maps that showrivers as a
single line cartographerssometimesplace the boundarysyrnbolon one side of the river and
sometimes on the other. This convention does not undermine the reliability of boundaries
drawnon maps, noris it evenrelevant tomaps where a riveris not shownas a singleline.
2. MisrevresentationsofNamibia's Positionon Mavs
473. BCM, para. 552:
'Upon appropriateanalysis,it appears that [NM, para. 3341asserts that,
of the 41 serial maps referred to by its cartographic expert Mr. D.
Rushworth, only four, the Seinerrnapof 1909, the British rnapof 1933,
the South African rnap of 1949, and the UN rnap of 1985, place
KasikililSeduduIslandin Namibianterritory.'
Inreality,Namibia stated:
'Maps of the area, produced by al1the parties in interest, with substantial
uniforrnityportray Kasikili Island as being located in Namibian tenitory.
particular,the principalmavs usedbv al1the political entities with governing
authoritv during the colonial veriod - Seinèr's rnap for the German
authorities until 1915 and the British until 1933;Bechuanaland Protectorate
GSGS 3915, used by the officials of Bechuanaland until 1965; the South
AfÎican maps beginningwith TSO 4001558in 1949; andthe UN rnap of 1985
- clearlyplaceKasikiliIslandinNarnibianterritory.' (emphasisadded)Botswana's interpretation of Namibia's statement is illogical and untrue, So far in the
pleadings, both sides have produced 26 maps showingKasikili Island that were officially
published by Germany, GreatBritain, South Afnca or the UN during the penod they were
respectively political authoritiesin thearea. Of these maps, 16 show the boundary to be
located in the southern channeland nine do not show theboundary at all. One overprints a
line ofdemarcation in the northernchannel,but Col.Rushworthbelieves thatthis is designed
to mark the limit of patrolling agreedby South Afncan and Botswana cornmandersafier the
shooting incident of October 1984. See Reply Annex 1, para. 28; id. 29 provides a list of
thesemaps.
474. BCM,para. 556 (quotingCol.Rushworth):
"'Ingeneral, maps of smaller scale than 1:1000,000(sic [footnote])do not
have sufficient detail to affectthis case. ."
Botswana Footnote: Presumablya typographicalerror. It would seem Mr. Rushworth
intendedto restrictrelevantmaps toa scal1:100,000.'
Col. Rushworth meant just what he said: in general maps with a scale smaller than
1:1,000,000do not have enough detail for this case. Thereis not the slightest warrant for
attributingto him the view espoused byBotswanathatmaps ofa scalemnallerthan 1 100,000
are too small. Namibia'sMemorialmakes it clear that many maps with scales greater than
1:100,000are relevantto this case. Botswana'sownpleadings claimthe relevanceof several
maps of scale greater than 1100,000,includingKriegskarte 1:800,000,Streitwolf 1:200,000,
DOS (Misc) 282 1:1,750,000,DOS 847 1:500,000,Botswana 1969 1:1,000,000,and Chobe
NationalPark 1980 1:250,000.
475. BCM,para. 574,n. 12:
'Whilst the Extract [NM, Fig. 10 (following page 122)] shows the red
hatching covering these areas such hatching is not apparent on the
original.'
This accusationis wholly unfounded. The Courtis invited to examinethe originalof themap
(NM,Atlas MapIV),which shows thehatchingas it appearsinNM,Fig. 10. 3. MisstatementsConcemina SvecificMavs
476. WarOfficeMap of 1891. BCM,para. 548:
'In ... the War Office Map of 1891 [ID846b] ... the boundaryis shown
withbroad red hatchings.'
Neither this map, nor the boundary it depicts,contain any red hatchings. The red hatching
appearson a totally differentmap, publishedin 1909in the third editionof TheMap ofAfrica
by Treaty, Vol.1 by Sir EdwardHertslet. Moreover,on the Herstlet map, the red hatchings
appear only on the British side of the border, while on the German side the hatchings are
brown. Thus,the boundaryis shown bythejoining of thetwo colours.
477. 1904 Kriegskurte von Deutsch Südwestafriku. BCM, para. 570 (see also BCM,
Appendix1,para. 39):
'TheAndara sheet (Botswana Atlas,Map 4) shows the northern channel
by a thick blackline, and the southern channel is barelv visibleexcept as
the edge to the shaded areawhich re~resentsthe island. On the Linjanti
Sheet (BotswanaAtlas, Map5) a strongblackline indicates thethalweg of
the river and is drawnto the north of ~ulurnbu'sisland ...'(emphasis in
original)
The Andara sheet does not appear in the BotswanaAtlas. It does not cover KasikiliIsland
and thus cannot show the northem channel. What Botswana mistakenly calls the Andara
sheet (Botswana Atlas, Map 4) is actually a monochrome version of the Linyanti sheet
(BotswanaAtlas, Map 5), an errorBotswanaalso madein its Mem~rial.*~The Linjantisheet
covers theKasikili Island area, but neither version,Map 4 or Map 5, shows the northem or
southem channel or attaches the label, Sulunibu's Island, to any recognizable geographic
feature. Furthermore, there is no justification for the assertion that the thalweg appears on
thismap.
478. VonFrankenbergMapof 1912. BCM,para. 585(see alsoBCM,paras. 35,279,588):
'[Tlhe southem channel is clearly indicated as "Kassikiri Flüss-arm"; a term
which in thecertifiedtranslationreads"as branch or tributaryof a main river"
andis so used in standardworks of reference.' VonFrankenberghirnselfuses
the word Flussarmin the legendto his map tomean 'abranch of a river,' with
no suggestionthat it implies a tributary. This accords with the usage defined
by the major Germandictionaries, which define Flussarm as 'a branch of a
23SeeBM, paras.266, 267. river,' as distinct fiom a tributary or inferior channel of a river.24 The
'certified translation'referred to in the quotation appears in an 'annexure'
pasted on the copy of the rnap in Botswana'sAtlas. This annexure is not a
certified translation. There is no 'certificate,' the signatureon the translation
is illegible, andtheuthor'squalificationsare not given. Nor are any standard
referenceworks cited.
479. 1933BechuanalandProtectorate 1,500,000,GSGS3915. BCM,para. 591 :
'No original suwey or ground verificationwork was done for this 1933
rnap ...'
Inreality,there was intensive activityonthe groundtocollectdata forthismap. Contributors
to this rnap in Bechuanaland included staff of the Resident Commissioner's Office, the
Resident Magistrates, and the Inspecter of Police and field staff and members of earlier
reconnaissance missions, including A.L. Du Toit, Captain B.E.H. Clifford and J.L.S.
Jeffares." Stonecalled it 'a significantmilestonein the cartographihistoryof ~otswana.'~~
480. SouthAfncan Map 19491:250,000TSO4001556.BCM, para. 541 :
'in disregard of specific exchanges in 1946 between the Bechuanaland
High Commissionerand the Survey Directorate,South Africa relatingto
Bechuanaland's northern boundary, South. Africa published [this
map]. ..9
First, the exchanges occurred in 1945, not 1946. Second, none referred to the northem
boundary in the vicinity of Kasikili Island. Bechuanalandofficials approved a print of the
rnap showing the boundary at Kasikili Island in the southem channel, and the rnap was
revisedby the SurveyDirectoratein accordancewith the Bechuanalandofficials'cornments.
481. South Afncan Map 1949 1:250,000TSO 4001556. BCM, para. 602 (see also BCM,
Appendix1,para. 77; BM, para.279):
'This rnapwas compiled by military cartographers, 45 Companyof the
South African Engineers ... who ... copied the boundaryline from the
1933Bechuanaland ProtectorateMap ...'
Botswanaprovidesno evidenceto supportthis statement.'~hecartographicrecordshows that
South Afnca ordered a thorough, new study using ground surveys and air photography to
" See,e.g.,NCM, Annexes 12, 14-16.
25See NCM, AM~X 8, pp. 77-80.
26Id., p71.produce this map. The boundary line on the 1949 map appears between the banks of the
southem channel and is drawn with the international boundary syrnbol, whereas the 1933
map places the boundary to the south of the southem channel and uses an inter-colonial
symbol for the boundary.
482. 1960BechuanalandProtectorate Map 1 :1,000,000DOS (Misc) 282. BCM, para. 593:
'... the sheet indicates the boundary along the northern and western
channel of the Chobe.'
This map (the scale of which is 1:1,750,000,not 1 :1,000,000)depicts the boundary along the
northern side of a single river symbol that does not show Kasikili Island. As Botswana
adrnits, 'no detail is visible in the vicinity of the i~land.'*~ According to cartographic
convention, which Botswana accepts?' this depiction indicates only that the boundary is in
the river.
483. 1965Bechuanaland Map 1:500,000DOS 847. BCM, para. 596:
'.. . the depiction of the boundary in the northern channel was not a
draughtsman's error but deliberate in execution of the Suweyor-
General's Opinion.'
The Surveyor-General's Opinion was written on 18 October 1965, whereas the map was
printed in September 1965 and copies of it were delivered in Great Britain by 20 October
1965. Moreover, the Opiniow nas never distributed outside the Bechuanaland Government,
so the DOS would not have had access to it.29Clearly, the Surveyor-General's Opiniow nas
not usedin compiling this map.
484. Chobe GarnePark Maps. BCM, para. 631 :
'A series of large scale plans in black and white were prepared by the
Suweys and Lands Departments to accompany the legal description in
thestatutory orders made under the National Parks Act 1967 in respect of
the Chobe Game Park.'
" BCM,para. 593.
28SeeBCM,para. 561.
29SeeReplyAnnex1,para. 16(a).Botswanadoes not produce these plans and adrnitsthat, 'Theearliest Plan available,No. TP
6-72-1 [was] dated 15 May 1975 .. .'30This rnap was a generalpurpose rnap and was not
preparedto support the legal definition of the boundary. Indeed, along the Chobe River, it
doesnot showthe Park boundary at all. Thefirst rnap to support the legal definition ofthe
boundary wasthe Chobe NationalPark Plan 1:250,000 BP 179, 1980,produced in 1980in
connectionwith StatutoryInstrumentNo. 126.31
485. JARICMap 1974. BCM,para. 614 (see alsoBCM, Appendix1,para. 85;p. 185;para.
421;p. 267;para. 37):
'The [JARIC] rnap [was] plotted from air photography taken by the
South African Air Force in 1977 in accordance with a Working Plan
(Botswana Supplementary Atlas Map 23, and accompanying Report:
Annex 32).'
BCMSupplementaryAtlas,Map 26:
'The21 July 1977Flight Plan was used to preparethe JARIC Map.'
Neither the 1977 air photography nor the Working Plan and its accompanyingreport was
usedin the preparationof the JARJCMap. Instead,the cartographicrecord showsthat these
sources were used forthe 1982 1:50,000SouthWest Afiica map, as shown in NR, Annex 1,
para.23.
4. Errorsof ScaleandOtherTechnicalDescrivtors
486. BCM,Appendix 1,headingto para. 64:
'Sketch Mapsof Bechuanaland Protectorate1:3801,600.'
Thecorrectscaleof these fourmaps is 1:4,600,000,not
1:3,801,600,as Botswanaasserts.
487. BCM,headingto para. 593(see alsoBCM,p. 264, no. 16;Appendix 1,heading to para.
7):
'The 1960Bechuanaland ProtectorateMap, 1:1,000,000'
Thescaleof this rnapis actually 1 1,750,000,not 1:1,000,000,as Botswanaasserts.
30BCM,para.632.
31See BCM, para.633. 195
488. BCM, heading to para. 625(see also BCM, Appendix 1,heading to para. 20):
'TouristMapof Chobe National Park,1977,1:250,000.'
The correct scale of this map is approximately 1:500,000,not 1:250,000as Botswana asserts.
The map's inset is at a scaleof:100,000.
489. BCM, Appendix 1,heading 37to para. 94:
'Southern Africa 1:50,000 Chief Director of Surveys and Mapping,
Mowbray,1982.'
The correct scale of this map is 1:500,000,not 1:50,000as statedby Botswana. CONCLUSION
490. As noted at the outset, the length of this Reply has been necessitated by the many
errors, misstatements and distortions of Namibia's position in the Botswana Counter-
Memorial. Nevertheless,Namibia believes that al1the elements of the case converge on a
single answer to thequestion submittedto the Court.
The languageof the 1890Treaty, construedin the light of the object and purpose
of the Treaty,
the subsequentpracticeof the parties,
the maps produced byal1the predecessorsin interestofbothparties and
the scientificevidence,
unite to show that, under Article III(2) of the Treaty,the main channel of the Chobe River
flows to the south of Kasikili Island. The southern channel as it appears on the maps and
aerial photographs is the thalweg channel of the main channel. And the boundary between
Namibia andBotswanais the thalweg of that channel.
491. In addition,Namibia is also entitledto sovereigntyover Kasikili Islandby prescription
by virtueof
the uninterrupteduse andoccupationof the Islandby the Masubia of the Caprivi
beginning before the signing of the Treaty and continuing thereafter under the
authorityof Germany andthen SouthAfiica (asmandatory andthereafterde facto
until 1989,whenit withdrewfiomNamibia),
the fullknowledge andacquiescenceof Botswana andits predecessorsin title and
without objectionfiom themuntil 1984. Submissions
492. In viewof the facts and arguments set forth in Namibia's Memorial and this Counter-
Memorial,
May it please the Court, rejectingl1 claims and submissions to the contrary, to
adjudgeand declare:
1. The channel thatliesto the southof KasikilUSeduduIsland is the main channel of
the ChobeRiver.
2. The channel that lies to the north of Kasikili/Sedudu Island is not the main
channelof the Chobe River.
3. Narnibia and its predecessors have occupied and used Kasikili Island and
exercisedsovereignjurisdiction over it,with the knowledge and acquiescence of
Botswanaanditspredecessorssinceat least 1890.
4. The boundarybetweenNamibia andBotswanaaroundKasikilUSeduduIslandlies
in the centre (that is to Say,the thalweg) of the southem channel of the Chobe
River.
5. The legal status of KasikiliISeduduIsland is that it is a part of the temtory under
the sovereigntyofNamibia.
ALBERK TAWANA
Agent of theRepublicofNamibia
before the InternationalCourtof Justice 198
List of Figures .
followingpage
Fig. 1..................................................-23.........
Fig. 2..................................................-24.........
Fig. 3..................................................-25.........
Fig .4....................................................5.......
Fig .5...................................................6........
Fig .6....................................................7.........
Fig. 7....................................................80..........
Fig .8....................................................84..........
Fig .9...................................................136......... Indexto Annexes
Annex No. Page
A. Map-Related Annexes
1 W.D. Rushworth,ObservationsconcerningMaps arising
from theCounter-MemorialofBotswana
2 Correspondencein 1945regarding DepictionofBoundary
on SouthAfrica1:250,000Map TSO4001556
3 Correspondenceregardingthe Colonial OfficeProposa1
for a Commissionto demarcatethe boundariesofthe Caprivi
(1911-1912)
Atfidavitof ColonelWalterDennisRushworth,23 October
1998
B. LegalInstrumentsandAuthorities
TableofBoundaryRiversinTreatiesof the 19thand 63
Beginningofthe 20th CenturyRelatedto theirnavigability
UN AideMemoireof 4 December 1984andrelated documents 68
ArticleIII ofthe Constitutionofthe South West Africa 81
People'sOrganisation (SWAPO)of Namibia
A. Sillery,TheBechuanalandProtectorate;Chapter8: 84
'TheJamesonRaid. DemarcationofBoundaries. The Tati
Area'
G.N.Uzoigwe,JournalofAfricanStudies,'Spheresof 98
Influenceandthe Doctrine ofthe Hinterlandinthe Partition
of fica'(1976) C. DiplornaticCorrespondenceand OtherOfficialMaterials
February 1996CorrespondenceBetweentheMinisterofForeign 105
Mairs ofthe RepublicofNarnibiaandthe MinsterofForeign
AfFairsofthe RepublicofBotswanaregardingthe presenceof
the Botswana DefenceForcetroops andthe Botswananational
flagon Kasikili/SeduduIsland
Letter of 18February1998,MinisterofForeignAEairsof 113
the RepublicofNamibiato the MinisterofForeignMis ofthe
RepublicofBotswana(regardmgsightvisitsbyNamibia
experts)
NamibiaConstituentAssembly,Excerpt fiomDebates
31January1990
Letter fiom theResidentCommissionerPanzerato the High
Commissioner,3 September1912
Letter fiomHighCommissionerLewisHarcourt,
14October 1912
BritishCabinet Memoranda: CAB12811 7, CM28(50)3
(4 May 1950)andCAB 129142,CP (50)214(24 September
1950);fiom TheLabour Government and theEnd of
Empire,HMSO, 1992
Correspondenceregarding 'generaaldministration'
(a) Instructionsto MagistrateandNativeCommissioner
of the EasternCapriviZipfelDistrict(Katima Mulilo)
[Trollope]fiomthe SecretaryforNative AfFis for
Administrationofthe CapriviZipfel,12October 1939
(NASA, RefNTS:file1101276)
(b)Letter,MagistrateandNativeCommissionerofthe
EasternCapriviZipfelDistrictinKatimaMulilo[Trollope]
to the SecretaryforNativeAfFairs[Proposalfor setting
asideEasternCapriviZipfelas a nativeReseme]
27 February1940(NASA,RefNTS: file 1101276
(c)Letter, SecretaryforFinanceto the SecretaryforNative
AfFairs[Expensesfor OfficialvisitofUnder-Secretaryfor
Native AfFairs]29May 1943(NASA,Ref TES 111243-
111244(d) Letter, SecretaryforFinanceto the Secretaryfor
Native Mirs [roadrepairs] 23 August 1944(NASA,
Ref TES 111243 - 111244)
(e) Letter,BantuAfFairsCommissionerinKatima Mulilo
to the SecretaryofantuAdministration andDevelopment
inPretoria [DescriptionofLocal (Tribal)Administration
inthe EasternCapriviZipfel] 13March 1962(NASA,
RefNTS: 10325)
Correspondenceregarding 'indirectrule'
(a) Letter,F.P. CourtneyClarke,Secretaryfor South
West A£iicato the Commander,S.W.A.Police,
Schuchmannsburg[Salariesof Chiefs]2 June 1938
(NASA,RefNTS: 10325)
(b)Letter,MagistrateandNativeCommissionerfor the
EasternCapriviZipfelto the SecretaryforNativeAffairs
[announcementthat chiefsmayemploysecretaries]
4 March 1940(NASA,RefNTS:10325)
(c)Letter, SecretaryforNativeMirs inPretoriato the
Magistrateforthe EasternCapriviZipfel payment of
Allowancesto chiefs,NgambelasandMembersof the
theirKutas] 14October 1941(NASA,RefNTS: 10325)
(d) Letter, SecretaryforFinanceto the SecretaryforNative
Affairs[Payrnentof Allowancesto Chiefs,Ngambelas and
Membersof theirKutas] 18 September 1941(NASA,Ref
TES 111243 -111244)
Correspondenceanddocumentsregarding 'exercise of
police powers'
(a) Union Legislation:UnionActsandProclamations
(1940)
(b) Letter,MagistrateandNativeComrnissioner[Trollope]
to the SecretaryforNativeMirs, andattached sketch
[Hippopoaching]30March 1940 (NASA, RefNTS:
7617 231329 -261329)(c) Letter, GovernrnentSecretary,ResidentCommissioner's
Officeto the Secretary for Nativeffairs[Appointmentof
SergeantWebb as Peace Officerinthe Eastern Caprivi]24
June 1940 (NASA,RefNTS: 7617 231329 -261329)
(d)Letter, Secretaryfor Native Mairs, to the Magistrateof 162
the Eastern CapriviZipfel(KatirnaMulilo), [dugout canoe
purchase for use ofSergeant-MajorWebb] 24 June 1940
(NASA,Ref NTS: 7616 231329 -261329
(e) Letter, Magistrate and Native Commissionerto the
Secretaryfor Native Mairs [PolicePatrols and Native
Constablesat Schuckmannsburg] 18January 1943
(NASA,Ref NTS: 8361311 - 8441311)
(f)ConfidentidReport byF. W. Trollopeon poaching 164
activities,2 September 1943(NASA,Ref NTS: 7617
231329 - 261329)
(g) Letter, Magistrate andNative Commissioner[Trollope] 167
to the Secretaryfor NativeAffairs[Export of hippo hidesand
elephant tusks] 11January 1944 (NASA,Ref NTS: 7617
231329 - 261329
(h) Letter, Magistrate ofthe Eastern CapriviZipfelto the
Secretaryfor Native Mairs [Purchaseof dugout canoes
for policepatrols] 31August 1948(NASA,RefNTS: 8361311 -
8441311)
(i) Letter, Naturellekommissaris,Oostliie Caprivi Zipfel
aanSekretarisvanNaturellesake (andtranslation fiom
Afïikaansto English 1948)'[Purchasedugout canoesfor
policepatrols] 5 January 1957(NASA, Ref NTS:836/311
- 8441311)
Botswana - NamibiaBoundary Dispute: Supplementary
Written Submissionsofthe Republicof Botswana to the
Joint Team of Technicd Experts, 15April 1994 D. Matenal Relatedto the ScientificCase
CurriculumVitaeofProfessorW.J.R. Alexander
CurriculumVitaeofProfessorKeithS.Richards
AiTidavitof Mr.GirtJehdusVisagie,owner of the
ZambeziQueenandKing'sDen Lodge
Brochures fromthe ZambeziQueenand King'sDen
Lodge
1985 and 1993 Reports of the Departmentof Water
Mairs SouthWest Africa/Namibia
June 1992 JointReport ofthe HydrologyDivisionsofthe
NamibiaandBotswana Departmentsof Water M'airs
D.C.Midgley,Reportof theLonrhoNamibia Caprivi
SugarProject (1990)
W.E. WilsonandJ.E. Moore (eds.),GlossaryofHydrology
(1988)
A.D.Lewis,1921KalahariReconnaissance (1926)
D. S.G.Thomasand P.A. Shaw, ïhe KalahariEnvironment
(1991)
M. Morisawa,Streams:theirdynamicsandmorphology
(1968)
K.S.Richards(ed.),RiverChannels:Environmentand Process
(1987)
L.A. MacKenzie,1945KalahariReconnaissance (1946)
MichaelMain,Zambezi:Journey of aRiver (1990)
F.C. Selous,A Hunter'sWanderingsinAfrica (1895)
R.H. Arnold,interpretationofAirphotosandRemoteZy
Sensedimagery (1997)Manual ofPhotographzcInterpretation,Amencan
SocietyofPhotogrammetry(1960)
E. Miscellaneous
'EnvoyExchangeon ConditionsBechuanasWon'tBe
Stooges',RandDailvMail, SouthAfrica,9 September
1965
'NamibiaWelcomesBotswana'sPresident',Timesof
Namibia,24 July1990
Reply of the Republic of Namibia