12487
INTERNATIONAL COURT OF JUSTICE
CASE CONCERNING THE FRONTIER DISPUTE
(BURKINA FASO/NIGER)
COUNTER-MEMORIAL OF BURKINA FASO
20 JANUARY 2012
[Translation by the Registry] TABLE OF C ONTENTS
INTRODUCTION........................................................................
...................................................... 1
Section 1 The points of agreement between the Parties................................................................ 1
Section 2 The points of disagreement between the Parties ........................................................... 5
1. The pre-eminence of the 1927 Arrêté and its Erratum......................................................... 5
2. The exhaustive list of documents to be taken into consideration should the Erratum
not suffice ........................................................................
................................................... 7
Section 3 The lacunae in Niger’s argument ........................................................................
........ 11
Section 4 Structure of the Counter-Memorial........................................................................
..... 11
CHAPTER I THE METHODOLOGICAL SHORTCOMINGS OF NIGER’S MEMORIAL ....... 13
Section 1 A flawed analysis of the content and scope of the Erratum of 5 October 1927.......... 13
1. The Erratum fixes the territorial boundaries of Upper Volta and Niger ............................ 14
2. The confirmation of the delimitation effected by the Erratum .......................................... 23
Section 2 Approximations and errors in the implementation of Article 2 of the Agreement
of 28 March 1987........................................................................
........................................... 29
1. “Should the Arrêté and Erratum not suffice”..................................................................... 30
2. “... the course shall be that shown on the [IGN map], and/or any other relevant
document accepted by joint agreement of the Parties”..................................................... 33
CHAPTER II THE DISCUSSIONS ON THE DEMARCATION OF THE FRONTIER
AND THE INCONSISTENCY OF NIGER’S CLAIMS............................................................ 38
Section 1 Niger’s one-sided and partial presentation of the facts............................................... 38
1. The consensual work........................................................................
.................................. 38
2. The crystallization of the dispute........................................................................
............... 41
Section 2 The inconsistency of Niger’s claims........................................................................
... 41
CHAPTER III THE COURSE OF THE FRONTIER IN THE “TÉRA SECTOR”........................ 45
Section 1 Events prior to the amended Arrêté of 1927 ............................................................... 55
Section 2 Events subsequent to the 1927 Erratum ...................................................................... 61
1. The Vibourié marker is not a frontier point ....................................................................... 63
2. The claimed village effectivités do not exist...................................................................... 64
A. “[from the Tao astronomic marker] the IGN line passes to the west of Petelkolé
(the village’s co-ordinates are 14° 00' 35.7" N, 00° 24' 52.6" E), which it leaves
to Niger . . . The frontier line follows the IGN line as far as the outskirts of
Petelkolé. It then deviates slightly to the west so as to meet the endpoint of the
upgraded stretch of the Téra-Dori road constructed by Niger (co-ordinates:
14°00'04.2"N, 00°24'16.3"E). It then rejoins the IGN line at the point
having co-ordinates 13° 59' 39" N, 00° 25' 12" E.” (Memorial of Niger, p. 94.) ...... 72
B. “The frontier then follows the IGN line, leaving Fetokarkale (Burkina Faso) to
the west. It then passes through a frontier point known as Baobab - ii -
(13° 58' 38.9" N, 00° 26' 03.5" E), and through Tindiki (13° 57' 15.4.9" N,
00°26'23.6"E), as far as the break in the line of crosses in the vicinity of
Ihouchaltane (Oulsalta on the 1960IGN map, Sebba sheet).” (Memorial of
Niger, p. 94.)........................................................................
....................................... 73
C. “The frontier passes through a point situated on the river to the west of the
encampment, whose co-ordinates are 13° 55' 36.4" N, 00° 27' 07.2" E . . . The
boundary passes through the point having co-ordinates 13°531 ' 2.8"N,
00°28'13.5"E located on the Kalsatouma-Sidibébé road. It then rejoins the
IGN line at the point having co-ordinates 13°53'24"N, 00°29'58"E.”
(Memorial of Niger, p. 95.) ........................................................................
................ 74
D. “From that point the boundary follows the 1960IGN line as far as the point
having co-ordinates 13°52'04"N, 00°31'00"E, where the area of Komanti
encampments (Kamanti or Comanti on certain documents) commences...
The frontier marked on the 1960IGN sheets is drawn with many gaps, to
indicate that its course is particularly problematic in this sector... From the
point having co-ordinates 13° 52' 04" N, 0° 31' 00" E, where there is a break in
the line of crosses on the 1960 IGN map, the boundary passes through the point
having co-ordinates 13° 48' 55" N, 0° 30'23"E, then reaches the point with
co-ordinates 13°46'31"N, 0°30'27"E. It then runs to the point with
co-ordinates 13° 46' 18" N, 0° 32' 47" E located to the north of Ouro Sabou on
the tributary arm of the River Tyekol Dyongoltol. The frontier then follows
that tributary until its confluence with the Tyekol Dyongoltol at the point with
co-ordinates 13° 46' 51" N, 00° 35' 53" E; from there, it follows the IGN line
as far as the point with co-ordinates 13°46'22.5"N, 0°37'25.9"E, located at
the level of Bangaré on the River Folko . . .” (Memorial of Niger, pp. 95-97.) ........ 75
E. “At this point, the frontier line takes a clear south-west orientation. The
co-ordinates of the point where the frontier line changes direction are the
following: 13°46'22.5"N, 00°37'25.9"E. To the south of Bangaré, the
boundary returns to the IGN line.” (Memorial of Niger, p. 98.) ............................... 76
F. “Following the watercourses, where there are no crosses, it passes between
Kolangoldagabé, in Burkina Faso (co-ordinates 13 453'N",
00°361' 4.5"E) and Lolnan[g]o, in Niger (co-ordinates 13°435 ' 0.3"N,
00° 36' 49.0" E).” (Memorial of Niger, p. 98.).......................................................... 77
G. “The frontier then passes through the locality of Sénobellabé (geographical
co-ordinates: 13° 36' 52.6" N, 00° 50' 00.8" E).” (Memorial of Niger, p. 98.) ........ 77
H. “The IGN line meets the line which at the time constituted the boundary of Say
(tripoint for the cantons of Tillabéry, Dori and Say) at the point with
co-ordinates 13° 29' 08" N, 01° 01' 00" E.” (Memorial of Niger, p. 99.).................. 79
CHAPTER IV THE COURSE OF THE FRONTIER IN THE “SAY SECTOR”.......................... 81
Section 1 The frontier reaches the River Sirba at Bossébangou ................................................. 83
1. There is no mistake in the Erratum on the point where the frontier reaches the River
Sirba at Bossébangou........................................................................
................................ 86
2. The documents invoked by Niger to contradict the terms of the Arrêté of
31 August 1927 as amended by its Erratum do not support the course it claims ............. 88
A. Documents prior to the Erratum ........................................................................
.......... 90 - iii -
B. Documents subsequent to the Erratum........................................................................
. 91
Section 2 The line claimed by Niger in the “salient” sector is not justified................................ 92
Section 3 The frontier between the intersection of the River Sirba and the Say parallel and
the beginning of the Botou bend is made up of a single straight line .................................... 95
⎯ The village of Dissi........................................................................
.............................. 98
⎯ The villages of Fombon, Latti and Tabaré................................................................. 100
SUBMISSIONS ........................................................................
..................................................... 103
Summary of sketch-maps........................................................................
.................................. 104
List of annexes ........................................................................
.................................................. 105 INTRODUCTION
0.1. The present Counter-Memorial is submitted in accordance with the Court’s Order
of 14 September 2010 fixing procedural time-limits in this case. It replies to the Memorial of
the Republic of Niger dated 20 April 2011.
0.2. Burkina Faso must firstly point out that the Memorial of Niger places it in a
difficult position, as it is based on a series of assertions and vague comments which are
inconsistently argued and lack legal foundation and therefore only call for a cursory response.
That is why the present Counter-Memorial is relatively brief, it being understood that Burkina
Faso upholds all the arguments that it put forward in its Memorial, but does not consider it
necessary to repeat them in full here.
0.3. Having made this point, it can be noted that the Memorial of Niger reveals:
⎯ points of agreement between the Parties (Section 1);
⎯ points of disagreement (Section 2); and
⎯ lacunae (Section 3).
S ECTION 1
T HE POINTS OF AGREEMENT BETWEEN THE PARTIES
0.4. There would actually appear to be fe w areas of disagreement in principle between
Burkina and Niger. They agree that:
⎯ the Court is called upon exclusively to confer the force of res judicata on the Parties’
agreement in respect of the following sector s of the frontier that have already been
demarcated:
“(a) the sector from the heights of N’Gouma to the astronomic marker of Tong-
Tong;
1
(b) the sector from the beginning of the Botou bend to the River Mékrou” ;
⎯ the 1987 Agreement cited in the Special Agreement indicates exhaustively the documents
that are to be taken into consideration for the purposes of demarcating the frontier;
⎯ it being understood that the boundary is delimited by the Arrêté of 31August 1927 as
clarified by its Erratum of 5 October 1927.
1Article 2, paragraph 2, of the Special Agreement of 24 February 2009. - 2 -
0.5. The table below sets out the Parties’ agreement in principle on each of these points.
Memorial of Niger Memorial of Burkina
1. The Court is called upon “It is on [Article2 of the The Parties “only agreed [on
to confer the force of res Special Agreement] that the the demarcated sectors, by
judicata on the Parties’ Court is called upon to focus the exchange of letters in
agreement on the demarcated its attention in relation to the2009] to define their
sectors of the frontier. merits, in order . . . to place ‘agreement’, which they have
on record the Parties’ requested the Court to place
agreement on the results of on record ” (MBF, p9 .0,
the work of the Joint para. 3.33).
Technical Commission on
Demarcation of the Burkina This is “an ‘agreement’
Faso-Niger Frontier and, in which the Parties would like
so doing, to confer on that to acquire the force of res
bilateral agreement between judicata” (MBF,9p1.,
the two States the force of para. 3.36).
res judicata ” (MN, p4 . 8,
para. 3.22).
2. The boundary is delimited “The boundary between the “[T]he boundary between the
by the Arrêté of 31August two Colonies was fixed by Parties was fully defined in
1927 as clarified by its the Erratum No.2602/APA Arrêté général No.2336 of
Erratum of 5 October 1927. of 5 October 1927, rectifying 31 August 1927 which was
Arrêté 233.6 of superseded by the Erratum of
31 August 1927. The 5 October 1927, and it has
boundary established by never been modified since ”
those two instruments was (MBF, p. 57, para. 2.8).
never changed until the
accession of the two Colonies
to independence” (MN, p. 24,
para. 1.32).
“Neither of the two Parties “[T]he Parties have always
contends that there was any considered their common
change to the legal situation frontier to be that which
existing [between 1947 and existed at the time of their
1960]. It follows that we accession to independence
have to go back to see what and that the frontier line in
was the instrument which, on question could be determined
5 September 1932, governed by reference to the line
the boundaries of the two described in the 1927 Arrêté
Colonies. That instrument and its Erratum” (MBF,
was in fact the Erratum of p. 58, para. 2.9).
5 October 1927 to the Arrêté
of 31 August 1927 fixing the “[T]he description of the line
boundaries of the Colonies of in the Arrêté and that in its
Upper Volta and Niger” Erratum differ in parts …
(MN, p. 61, para. 5.3). consequently, preference
should be given to the text of
“In order to determine what 5 October 1927, which is
were the boundaries of the both subsequent to and more
territory of the two States as precise than that of
at5ugu1st960, it is 31 August” (MBF, p. 69, - 3 -
necessary to seek out the para. 2.41).
most recent legislative or
regulatory acts of the “[A] document issued by the
colonial power having Governor General of FWA
determined those boundaries. describes in full the course of
the boundary between the
The only ones to have been
found are the colonies of Upper Volta and
Erratum 2602/APA of Niger. It has neither been
5 October 1927 correcting modified nor called into
question since it was
Arrêté 2336 of
31 August 1927, which has, adopted. Furthermore, in the
moreover, been consistently Agreement of 28 March 1987
cited by the parties ” (MN, cited in the Special
p. 61, para. 5.3). Agreement, both Parties
officially considered that the
“This text was, however, present frontier, in its
never the subject of any entirety, was described in the
addition, amendment or 1927 Arrêté as clarified by its
correction during the Erratum” (MBF,5p8.,
colonial period. It remained, para. 2.10).
at the time when the two
States became independent,
the only reference text for the
determination of their
common frontier” (MN,
p. 104, para. 7.12).
Regarding the “Téra sector”: “[T]he Parties have
“[t]he only text from the determined by mutual
colonial period determining agreement that the 1927
the boundaries of the two Arrêté is the only title
Colonies in this area is the they can rely on and,
Erratum No2 .602/APA of secondly, that same
5 October 1927, which instrument . . . describes the
corrected Arrêté No.2336 of entire common frontier ”
the Governor-General of (MBF, p. 69, para. 2.40).
French West Africa of
31August of the same year”
(MN, p. 83, para. 6.9).
For the “Say sector”, “the
only text from the colonial
period determining the
boundaries of the two
Colonies in the Say sector is
the Erratum …, correcting
[the] Arrêté” (MN, p.103,
para. 7.9).
3. The 1987 Agreement cited “[Articles1 and 2 of the “This shows the extreme
in the Special Agreement 1987 Agreement] state very importance of the Agreement
indicates exhaustively the precisely what is to be of 28March1987 and of the
documents that are to be understood in this case by the instruments to which it refers
taken into consideration for application of the principle of for the purposes of settling
the purposes of demarcating ‘the intangibility of the dispute submitted by the - 4 -
the frontier. boundaries’” (MN, p 6.0, Parties to the Court” (MBF,
para. 5.2). p. 9, para. 0.19).
“For purposes of the practical “[I]n the Agreement of
application of that principle, 28 March 1987 on the
the text of the demarcation of the frontier,
1987Agreement, as referred they expressly enumerated
to in the Special Agreement, the sources of law applicable
relies on three criteria [the for those purposes” (MBF,
pieces of legislation from pp. 61-62, para. 2.20).
1927(Section1); the 1960
IGN map (Section2); any
other relevant document
accepted by joint agreement
of the Parties (Section3)]”
(MN, p. 61, para. 5.2).
“[T]he Agreement signed in
Ouagadougou on
28 March 1987 . . . provided
that the 1927 texts were to
remain the bases for
determining the frontier
between the territories of
Upper Volta and Niger”
(MN, p. 24, para. 1.32).
“In accordance with the “[By the 1987 Agreement,
general approach of the the Parties] established the
Republic of Niger regarding pre-eminence of the frontier
the principles applicable title constituted by the Arrêté
through the determination of of 31 August 1927, as
the frontier in the present clarified by its Erratum, over
dispute ⎯ and in accordance any other evidence of the
with the terms of the 2009 frontier line” (MBF,
Special Agreement and of the pp. 61-62, para. 2.20).
1987 Agreement between the
two States ⎯ it is thus the “The Agreement of
text of the 192Erratum 28March1987 … does not
which will constitute the place on the same footing the
1927 Arrêté and its Erratum,
primary basis for
determination of the course on the one hand, and the
of the frontier between the 1960 IGN France map and
two States in this second any other document accepted
sector” (MN, pp1.04-105, by joint agreement of the
para. 7.12). Parties, on the other” (MBF,
p. 66, para. 2.35).
“[T]he two States, aware of “[O]nly the 1960 IGN France
the limitations of the colonial map can be used to clarify
texts, provided in the the course of the frontier
Agreement of 28 March 1987 between the Parties.
for recourse to two Nevertheless, it follows from
subsidiary criteria [1960 the very text of the 1987
IGN map, on the one hand; Agreement that it may only - 5 -
any document accepted by be used on a subsidiary
joint agreement of the Parties basis” (MBF7 ,1,.
on the other hand]” (MN, para. 2.47).
p. 75, para. 5.13).
“Conscious of the limitations
of the colonial texts, Burkina
Faso and Niger provided in
the Agreement of
28 March 1987 for recourse
to subsidiary criteria, among
which the 1:200,000 map of
the Institut géographique
national[e], 1960 edition,
plays a pivotal role” (MN,
p. 91, para. 6.16).
0.6. A priori, the Parties therefore agree:
⎯ on the subject of the dispute: it is to determine the course of those parts of the frontier
that have not been demarcated and to place on record the Parties’ agreement on the two
demarcated sectors; and
⎯ on the law to be applied by the Court in the present case, as defined by Article6 of the
Special Agreement and the Agreement of 28 March 1987 to which that provision refers:
in accordance with the letter and spirit of Article 2 of the 1987 Agreement, they agree that
the frontier between the two countries is fixed by the Arrêté of the Governor-General of
FWA as clarified by its Erratum of 5October 1927, which represents the legal situation
existing at the time of the Parties’ accession to independence and applicable in this case
according to the principle of the intangibilit y of boundaries inherited from colonization,
also explicitly mentioned in Article 6 of the Special Agreement.
S ECTION 2
T HE POINTS OF DISAGREEMENT BETWEEN THE P ARTIES
0.7. Nevertheless, when it comes to implementation, Niger seeks to neutralize the very
principles that it admits are applicable, whether in respect of the pre-eminence of the Erratum
of 5October 1927 over any other document or the exhaustive nature of the list of other
documents to which Article2 of the Agre ement of 28March 1987 refers, should the Arrêté
and its Erratum not suffice.
1. The pre-eminence of the 1927 Arrêté and its Erratum
0.8. Despite professing to agree with the pr inciple established in Article 1 of the 1987
2
Agreement , Niger refuses to acknowledge that, given that the frontier has been delimited by
the Erratum, the Court’s task in the present case is ⎯ solely ⎯ to clarify its course when ⎯
and only when ⎯ that document does “not suffice” ; otherwise, it should confirm the course
2
See the table in para. 0.5 above.
On the notion of “not sufficing”, see paras. 1.42-1.45 below. - 6 -
described therein. In this respect, Niger’s Memo rial appears to be have been designed to call
into question the Arrêté and its Erratum, despite the fact that these have been recognized by
the Parties as the title on which the Court should rely in order to determine the course of the
disputed frontier.
0.9. It seeks to present the Erratum as:
4 5 6
⎯ “particularly rudimentary” , “particularly succinct” , “particularly lapidary” or
“summary and imprecise” ; 7
⎯ erroneous , “hav[ing] no basis in the situation prior to [its adoption]” 9 and “never
10
confirmed in the subsequent practice” ;
⎯ contested and “criticized from the outset by the colonial officials and authorities of the
two Colonies” . 11
0.10. Moreover, Niger does not hesitate to assert that “the text of the Erratum should
12
not be read too literally...” and, more explicitly still, “that there are well-established
reasons for not following it in certain respects” 13. This is contrary to the Parties’ mutual
understanding, as recorded in their 1987 Agreemen t and reiterated in the Special Agreement.
Moreover, this would be the case even in th e absence of a formal agreement between the
Parties stating as such: as Niger recognizes in the same paragraph of its Memorial, the
Erratum “remained, at the time when the two States became independent, the only reference
14
text for the determination of their common frontier” ; it therefore constitutes the legal
title ⎯ a notion that Niger is very wary of mentioning 15 ⎯ on which the two Parties and the
Court must rely in order to determine the fron tier. And, as the Court firmly recalled in the
case of Cameroon v. Nigeria, “where there is a conflict between title and effectivités,
16
preference will be given to the former” .
MN, p. 65, para. 5.6; see also p. 84, para. 6.10.
MN, p. 83, para. 6.9; see also p. 104, para. 7.11.
MN, p. 116, para. 7.34.
7
MN, p. 66, para. 5.6; see also p. 86, para. 6.11.
8
Cf.SubsectionA.“There was no jus tification for continuing the inter-co lonial boundary to the village of
Bossébangou” (MN, pp. 105-111, paras. 7.14-7.24); see also p. 65, para. 5.5; p. 83, para 6.9.
9
MN, p. 116, para. 7.35; see also p. 120, para 7.39.
10
Ibid., see also p. 108, para. 7.18.
11
MN, p. 66, para. 5.6; see also pp. 26-28, para. 2.3.
12
MN, p. 115, para. 7.32.
1MN, p. 105, para. 7.12 (emphasis added).
1Ibid.
1The word “title” does not appear at all in the Memorial of Niger with this meaning.
1Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial
Guinea intervening), Judgm ent, I.C.J. Reports 2002, p. 15, par. 23; see also Frontier Dispute
(Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, pp. 586-587, para. 63. - 7 -
2. The exhaustive list of documents to be taken into consideration
should the Erratum not suffice
0.11. As Niger recognizes in Chapter V of its Memorial, which is devoted to “the legal
bases for determination of the frontier”, th e Agreement of 28March 1987, to which the
Special Agreement refers, states “very precisely wh at is to be understood in this case by the
application of the principle of ‘the intangibility of boundaries’, that is to say the uti possidetis
at the date of independence of the two States in 1960”; and, as it goes on to explain, “[f]or
purposes of the practical application of that principle, the text of the 1987Agreement, as
referred to in the Special Agreement, relies on th ree criteria”; “criteria” might not be the best
choice of word, but it means the following documents or series of documents:
⎯ the pieces of legislation from 1927 ⎯ the Arrêté of 31August and its Erratum of
5 October;
⎯ the 1:200,000-scale IGN map of 1960; and
⎯ “any other relevant document accepted by joint agreement of the Parties”.
0.12. On this last point, Niger agrees that “[i]n the course of the Joint Commission’s
work, no document was accepted on this basis” 17. Unless the exchange of letters of
29 October 2009 and 2 November 2009 qualifies as such, there is actually no other document
accepted by joint agreement of the Parties: consequently, only the 1960 IGN map can be
taken into account and even then only if the modified Arrêté does not suffice. In fact, having
conceded this point, Niger does not hesitate to give precedence to:
18
⎯ the 1960 IGN map over the Erratum, including when the latter is perfectly clear ;
⎯ maps from 1915 and 1927 over the Erratum and the 1960 map ; 19
⎯ various documents, which it believes prove the effectivité of Niger’s presence in certain
20
disputed territories, over both the map and the Erratum .
0.13. The argument deployed in Niger’s Memorial for the sector between the point
where the frontier “leaves the salient and enters the Botou Loop” 21 illustrates this tendency to
reinvent a frontier line on the basis of various documents whose relevance is ruled 22t by the
1987 Agreement. Claiming the wording of the Erratum to be lapidary , when it is in fact
perfectly clear, Niger dismisses the Arrêté in favour of the IGN map 23, only to challenge it in
turn and maintain its claim to “a frontier in two straight-line sections, as it ap
pears on those
maps and sketch-maps of the colonial period” 24. It therefore only has recourse to the map to
17
MN, p. 76, para. 5.15; similarly, see MBF, p. 71, para. 2.46.
18
MN, p. 93, para. 6.21 or p. 97, paras. 6.24-6.25.
19MN, p. 114, para. 7.30.
20MN, p.93, para.6.20; p.94, para.6.22; pp. 95-96, para.6.23; p.110, para.7.21; p.114,
paras. 7.30-7.31; or p. 120, para. 7.40.
21MN, pp. 116-120, paras. 7.34-7.40.
22MN, p. 116, para. 7.34.
23
MN, p. 116, para. 7.35.
24
MN, p.120, para.7.40; see also, for the sector from Tong-Tong to Tao, pp.91-92, para.6.18 and
pp. 92-93, para. 6.20 or, for Petelkolé, p. 94, para. 6.22. - 8 -
the extent that it backs up its claims. Moreove r, Niger is clear about the status of the 1960
map: in its view, “unless we find abnormal devi ations in relation to the texts or manifest
lacunae in the information on the canton boundaries, and subject to the necessary caution
where the hesitation of the map’s drafters is re flected in gaps in the line of crosses, these
results should in principle serve as a guide to determine the course of the inter-colonial
boundary in 1960” . 25
0.14. As a result of these disagreements a bout how the applicable rules are to be
implemented, differences of opinion about the delimitation of the frontier have arisen and it is
these that lie at the heart of the present disput e. Along the course of the contested line the
Parties’ claims only coincide on three points: Tong-Tong, Tao ⎯ although its co-ordinates
26
are incorrect in Niger’s version ⎯ and the point marking the beginning of the Botou bend
(the point known as Tyenkilibi) . 27
0.15. The points of disagreement between the Parties can be summarized as follows:
⎯ in the Téra sector : 28
• According to Burkina, two straight lines connect the three frontier points in this sector
(the Tong-Tong marker, the Tao marker and Bossébangou) . 29
• According to Niger, from Tong-Tong to Tao the line follows not one but two
straight-line sections, since it passes through the Vibourié marker before reaching the
Tao marker; from there, the frontier lin e “basically” follows the line on the IGN
map 30 ⎯ to the extent permitted by the liberties taken by Niger and “justified” by the
effectivités that it claims ⎯ as far as Bangaré, before following the IGN line and
stopping at “the tripoint of the former boundaries of the cercles of Say, Tillabéry and
31
Dori” rather than descending, as the Erratum nevertheless stipulates, to the River
Sirba at Bossébangou.
32
⎯ In the Say sector :
• According to Burkina, from the point wher e the frontier reaches the River Sirba at
Bossébangou (a point known as “P” 33), the line forms a salient consisting of three
25MN, p. 91, para. 6.16; see also MN, p. 93, para. 6.20 or p. 120, para. 7.40.
26Co-ordinates: 14° 03' 02.2" N, 00° 22' 52.1" E (MN, p.94, para.6.22); the co -ordinates, measured by
GPS by Burkina, originate from the Nevière data sheet of 1927, Ann. MBF41, and are as follows:
14º 03' 04.7" N; 0º 22' 51.8" E (see MBF, p. 104, para. 4.16).
27The first and last of these points are fixed by the agreement between the Parties (record [procès-verbal]
of 3July 2009) embodied by the exchange of letters of 29October 2009 and 2Novemb er 2009. In actual fact,
therefore, only one single point on the disputed line, the Tao marker, is the subject of an agreement between the
Parties.
28Burkina uses the term “Téra sector” for convenience, but believes it to be a simplification that should be
used with great caution (see paras. 3.14-3.17 below); a more correct term would be: “the Dori/Téra sector”.
29MBF, p. 132, para. 4.82.
30MN, “From the Tao astronomic marker to Bangaré”, pp. 93-97, paras. 6.21-6.23.
31MN, p. 100, para. 6.26.
32
Like “Téra sector” (see fn.28 above),this term should be used with caution; again, Burkina uses it
purely for convenience.
33
MBF, p. 133, para. 4.83. - 9 -
sections, the first turning back up the River Sirba to a point known as “P1”, from
where the second runs in a north-westerly direction as far as a point “P2” and joins
the third, which descends in a straight lin e to the south until it meets the intersection
of the River Sirba with the Say parallel, a point which marks the end of the salient 34.
Niger’s description is radically different. Moreover, it does not envisage an actual
salient , since it has a single straight line, running in a south-westerly direction,
connecting two points that are, furthermore, different from those in the Erratum: the
line claimed by Niger therefore departs from the “tripoint of the former boundaries of
the cercles of Say, Tillabéry and Dori” (and not from Bossébangou), reaching the
36
Sirba approximately at the level of (and not at its intersection with) the Say parallel .
• For its part, the last section of the frontier between the end of the salient and the
beginning of the Botou bend (Tyenkilibi) 37 is described by Bu rkina, in accordance
with the Erratum, as consisting of a single straight line, whereas Niger draws it as two
straight-line sections, the line changing direction at a frontier marker on the road from
Niamey to Ouagadougou . 38
34
For an illustration, see MBF, p. 153, sketch-map No. 14.
3Niger only uses the word salient in inverted commas : given that its reasoning is based on the previous
section terminating at “the tripoi nt of the former boundaries of the cercles of Say, Tillabéry and Dori”, it regards
the word salient as problematic, making “no sense in relation to the inter-colonial bounda ry”: “the frontier cannot
create a salient in this area. It simply turns in a s outh-westerly direction from th at ‘tripoint’” (MN, p.112,
para. 7.26).
36
MN, pp. 115-116, paras. 7.32-7.33.
37
The co-ordinates given by Niger for this last point di ffer very slightly from those accepted by Burkina:
the difference is one latitudinal second ⎯ see fn. 27 above.
38
MN, p. 118, para. 7.38. - 11 -
SECTION 3
T HE LACUNAE IN N IGER ’S ARGUMENT
0.16. Reference should be made to one final aspect of Niger’s Memorial: its lacunae.
It actually fails to address, or barely touches on, a number of points in the present case, some
of which are significant.
0.17. The following are of particular note:
39
⎯ Niger’s rather limited exposition of the origins of the 1987 Agreement , its analysis of
which is practically non-existent ;
⎯ the extremely brief reference to the negotiations to which the dispute has given rise ;1
42
⎯ the cursory treatment reserved for th e marked sections of the frontier (moreover, the
submissions in Niger’s Memorial only contain a description of the disputed portion of the
line and not the whole of the line, including the marked sections, to which Niger does not
refer, even though the Special Agreement in cludes them in the subject of the case
submitted to the Court ); and
⎯ Niger’s almost total silence 44on the consensual line of 1988 4, which nevertheless shows
that its representatives believed at the time th at it was perfectly possible to determine the
frontier between the two countries on the basis of the instruments referred to in the 1987
Agreement.
0.18. Given that these various points are not developed in Niger’s Memorial, Burkina
sees no point in returning to them in the presen t Counter-Memorial. However, it wishes to
state categorically that it does not intend to renounce any of the arguments that it put forward
in its own Memorial.
SECTION 4
STRUCTURE OF THE C OUNTER -M EMORIAL
0.19. The complete absence of methodology that characterizes Niger’s Memorial is so
blatant and leads so systematically to deep flaws in its submissions that it seemed necessary to
devote the whole of the first chapter of the present Counter-Memorial to this issue.
Furthermore, it would seem worthwhile to point out the contradictions between the positions
adopted by Niger in negotiations and those adopted in the Memorial, given that they are so
39
MN, p. 40, para. 3.3 ⎯ comp. MBF, pp. 43-44, paras. 1.61-1.65.
40
MN, p. 75, para. 5.13 ⎯ comp. MBF, pp. 71-73, paras. 2.45-2.50; or MN, pp. 60-61, para. 5.2 ⎯ comp.
MBF, pp. 56-61, paras. 2.4-2.19.
41
MN, pp. 39-44, paras. 3.1-3.11 ⎯ comp. MBF, pp. 32-54, paras. 1.34-1.88.
42MN, p. 46, para. 3.17 and p. 48, para. 3.22 ⎯ comp. MBF, pp. 78-89, paras. 3.11-3.30.
43MN, pp. 122-123 ⎯ comp. MBF, pp. 160-162.
44See, however, MN, p. 40, para. 3.4, where the meeting of technical experts is mentioned in passing.
45
MBF, pp. 45-46, paras. 1.67-1.69. - 12 -
glaring (Chapter II). Having provided these cl arifications, Burkina will return, in two
separate chapters, to the determination of the two unmarked sections of the frontier, in the
Téra sector (Chapter III) and the Say sector (Chapter IV), in accordance with the provisions
of the 1987 Agreement as cited in the Special Agreement. C HAPTER I
T HE METHODOLOGICAL SHORTCOMINGS OF N IGER ’SM EMORIAL
1.1. One might have expected the Memorial of the Republic of Niger, which includes
five general chapters before turning to the i ssue of the determination of the frontier in the
“Téra sector” (ChapterVI) and the “Say sector” (ChapterVII), to be based on a clearly
explained and rigorously implemented method. Far from it: Chapter V 46boasts of presenting
“[t]he legal bases for determination of the frontier”, but the “methodo47gy adopted” ⎯ which
includes a clear description of the course of the frontier in the Téra s, de facto repeated
for the Say sector48⎯ disregards these “legal bases” in favour of a muddled and selective
approach, whose only discernible “methodology” seems to be to seek a line that is favourable
to Niger, however arbitrary and devoid of legal justification that might be.
1.2. The methodological shortcomings of Niger’s argument are reflected in particular
in:
⎯ a flawed analysis of the content and scope of the 1927 Erratum (Section 1);
⎯ approximations and errors in the implementa tion of Article2 of the Agreement between
the Parties of 28 March 1987 (Section 2).
SECTION 1
A FLAWED ANALYSIS OF THE CONTENT AND SCOPE OF THE
E RRATUM OF 5 O CTOBER 1927
1.3. As was pointed out in the introduction to the present Counter-Memorial 4, Niger
indicates that it accepts that the Erratum of 5October 1927 to the Arrêté of the
Governor-General of FWA of 31August of the same year, fixing the boundaries of the
Colonies of Upper Volta and Niger, is the fundamental document delimiting the frontier
between the two countries. Nevertheless, it makes every possible effort to rule out its
application. In particular:
⎯ it refuses to accept that it fixes the new boundaries between the two Colonies (1); and
⎯ it wrongly attempts to interpret the protests of certain local officials as proof that it was
not observed during the colonial period, whereas, like many of the incidents that took
place after the adoption of the Arrêté and the Erratum, they are actually evidence that this
delimitation existed (2).
4MN, pp. 60-78.
47
“A. Methodology adopted” (MN, pp. 82-91).
4MN, pp. 104-105, paras. 7.9-7.13.
4See paras. 0.4-0.6. - 14 -
1. The Erratum fixes the territorial boundaries
of Upper Volta and Niger
1.4. One of the postulates on which Niger’s thesis rests is that the purpose of the Arrêté
and its Erratum was not to effect a delimitation between Upper Volta and Niger but to effect a
transfer of cantons from one Colony to the other: certain cantons of Dori cercle and all those
of Say cercle (with the exception of Gourmantché Botou canton), all of which were
previously part of Upper Volta, had apparent ly been transferred to Niger with their
50
boundaries unchanged . The reasoning is curious: according to Niger, the 1927 Arrêté was
purely an implementing text of the Decree of 28 December 1926, which effected a transfer of
cantons:
“It will be recalled that the justification for the Arrêté of 31August1927
lay in the Decree of the President of the French Republic of 28 December 1926,
‘transferring the administrative centre of the Colony of Niger and providing for
territorial changes in French West Afri ca’. [Quotation from Article2 of that
decree.] It was on the basis of this Decree that, a few months later, the Arrêté of
31August1927 and its Erratum of 5Oc tober1927 were adopted. The purpose
of these texts could only have been to transfer the above-mentioned cantons.” 51
1.5. It is this link between the two texts, supported by the use of preparatory works, that
leads Niger to assert that the Arrêté also serves purely and simply to effect a transfer of
territories, adding that “the 1927 Arrêté . . . could not conceivably have been in contradiction
with that Decree, since it was from it that it derived its legitimacy” : 52
“Thus we should not lose sight of the essential point: the purpose of the
1927 Arrêté and its Erratum was to transfer from one Colony to the other a
cercle composed of cantons. The texts confirm this. The 1927 Ar53té and its
Erratum were adopted pursuant to the Decree of 28 December 1926.”
1.6. Therefore, still according to Niger,
“[t]he new boundary was defined as a series of juxtaposed cantonal boundaries ,
themselves composed of a series of village and/or hamlet boundaries. In
sparsely populated areas, the canton boundaries were quite vague: for example
54
on rocky hillsides and infertile plateaux, and in open pastureland.”
It adds:
“There was thus no question of drawing geometric lines but of
incorporating cantons into the territory of each Colony. Where the boundaries of
those cantons reflected occupation on the ground by the local people (in
villages), they did not follow straight li nes. That was the case in particular for
50
This “logic” of canton transfers probably explains why Niger us es the term “tripoint” to denote the
meeting-point of the cercles of Dori, Say and Tillabéry, through which it believes the frontier to pass (MN, p. 108,
para. 7.17; p. 110, para. 7.20; or p. 111, paras. 7.22 and 7.24).
51
MN, p. 84, para. 6.11.
52
MN, p. 111, para. 7.22.
53MN, p. 70, para. 5.10 (underlining in the original text).
54MN, p. 80, para. 6.6 (emphasis added). - 15 -
Tillabéry cercle, contrary to Say cercle, which was largely uninhabited at the
time.” 55
T1h.i7s. ⎯ erroneous ⎯ view is the pretext for introducing a bogus methodology:
“we know the names of the cantons which were transferred. This can give two
valuable indications. The first concerns the content of those cantons . . . , where
these can be found on the administrative documents of the colonial era. As will
be seen later, indications of this kind, although few in number, can supplement
the summary description in the Arrêté and Erratum of 1927. The second
indication is a presumption that the areas composing these cantons . . . did not in
principle follow abstract lines (whether curved or straight), but were based on
56
land occupation and followed the configuration or nature of the ground.”
1.8. There are numerous objections to this analysis.
1.9. As Niger recognizes, the canton boundaries themselves were often quite vague 57
and, in this part of the French colonial empi re at least, were never the subject of a formal
delimitation text, to the extent that Niger itsel f has to concede that “the...possibilities
58
offered by this approach” are “modest” . The reply from the FWA Geographical
Department to a request made in 1938 by the Director of Political and Administrative Affairs
for a sketch-map of the region 59 shows that such an approach only leads to deadlock:
“I have the honour to inform you that it is not possible at present for the
Geographical Department to produce an accurate sketch-map as requested
(showing the division into cantons of Koutiaia, Gao, Fada N’Gouma, Say,
Tillabéry, Zinder and Gouré cercles and Dosso, Gaya and Filingue subdivisions),
due to a lack of information.
The Atlas of Cercles is currently being revised, but this is a very lengthy
and painstaking task that will require the participation of the local administrative
authorities, which at present are the only ones able to define ⎯ at least
approximately ⎯ the canton boundaries.
In most cases, these are de fa cto boundaries which have never been
defined by texts.” 60
Moreover, despite insistently hammering home this thesis of a simple transfer of cantons
from one Colony to the other, Niger is forced to admit that “notwithstanding the wish
frequently expressed by officials of the two Colonies, the course of the boundary was never
55
MN, p. 72, para. 5.10.
56MN, p. 86, para. 6.11; see also MN, pp. 90-91, para. 6.15.
57MN, p. 80, para. 6.6.
58MN, p. 91, para. 6.15.
59
Letter No.18 AP/2 from the Director of Political and Administrative Affairs of the
Government-General to the Military Chef du Cabinet regarding a sketch-map of cantons in Fada N’Gouma, Say
and Tillabéry cercles, amongst others, 7 June 1938, Ann. CMBF 5.
60
Note No.521 CM2 from the FWA Geographical Depa rtment to the Director of Political and
Administrative Affairs of the Government-General regarding a sketch-map of cantons in Fada N’Gouma, Say and
Tillabéry cercles, amongst others, 25 June 1938, Ann. CMBF 6 [emphasis added]. - 16 -
clarified by a new text so as to correspond more closely with the actual boundaries of the
61
cantons in practice” .
1.10. This refusal to regard the Arrêté and its Erratum as fixing the boundary between
the two Colonies has a remarkable consequence: Niger completely refrains from using the
word “title” to refer to these fundamental instruments, although it admits that they established
the boundary between the two Colonies and that this was never changed until independence . 62
They are referred to as “the most recent leg63lative or regulatory acts of the colonial power
having determined those boundaries” ; the Erratum is the “instrument which...governed
the boundaries of the two Colonies” 64, the “only reference text for the determination of their
65
common frontier” and the “only text from the colonial period determining the boundaries of
the two Colonies” . 66
1.11. Such an instrument very precisely constitutes a territorial title, “that is, a
document endowed by international law with intrinsic legal force for the purpose of
establishing territorial rights” . As the Parties and the Chamber of the Court acknowledged
in the Burkina/Mali case, “the title which is accorded pre-eminence in the [French] colonial
system is the legislative and regulative title” 68, which is precisely the status of the
1927Erratum. Moreover, in the same case, the Chamber of the Court pointed out “that the
1927 Order does not directly concern the boundary between Sudan and Upper Volta, but only
the boundary between Upper Volta and Niger, a nd that for the purposes of [that] case, the
Chamber [was] consulting the Order solely as evidence which may shed some light on the
intentions of the colonial power concerning the course of the boundary between French Sudan
69
and Upper Volta” . By contrast, in the present case the Arrêté and its Erratum would appear
to constitute the very basis for identifying the Pa rties’ respective territorial limits and, as the
Chamber also recalled in 1986: “The purpose 70 the 1927Order was to fix the boundaries
between the colonies of Upper Volta and Niger” .
1.12. This is also perfectly clear from the very title of the Arrêté and its Erratum, both
of which seek to fix the boundaries of the Colonies of Upper Volta and Niger ⎯ refuting
Niger’s inconsistent allegation that “[t]he purpose of these te xts could only have been to
transfer the above-mentioned cantons” . Their purpose was both different from and
complementary to that of the Decree of the President of the French Republic of
28December1926, “transferring the administrative centre of the Colony of Niger and
providing for territorial changes in French W est Africa”. In fact, once this transfer ⎯ the
61
MN, p. 91, para. 6.16.
62
MN, p. 24, para. 1.32.
63MN, p. 61, para. 5.3.
64MN, p. 62, para. 5.3.
65MN, p. 104, para. 7.12.
66MN, p. 83, para. 6.9 and p. 103, para. 7.9.
67
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986 , p. 582, para. 54; see
also Land, Island and Maritime Frontier Dispute (El Sa lvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992, pp. 388-389, para. 45.
68Ibid., p. 582, para. 53.
69
Ibid., p. 590, para. 69.
70
Ibid., p. 642, para. 167.
71
MN, p. 84, para. 6.11 ⎯ see para. 1.4 above. - 17 -
purpose of the Decree of 28December 1926 ⎯ had been decided, it was necessary to
determine the course of the new inter-colonial boundary; this is precisely the purpose of the
Arrêté and its Erratum.
1.13. Indeed, this was explicitly provided for in Article 2, paragraph 2, of the Decree of
the President of the Republic:
“An Arrêté of the Governor-General in Standing Committee of the
Government Council shall72etermine the course of the boundary of the two
Colonies in this area.”
This is what is done by the Arrêté of the Governor-General of FWA of 31 August 1927, the
third citation of which refers explicitly to the Decree of 28 December 1926.
1.14. Moreover, this was also the understanding of the Chamber of the Court in its
Judgment of 12 July 2005 in the case concerning the Frontier Dispute (Benin/Niger):
“[T92e7 arrêté] was adopted by the Governor-General following, and
as a consequence of, the decree of 28 December 1926 incorporating the cercle of
Say into the colony of Niger (created some years earlier). It was thus for the
Governor-General to define the boundaries between the colonies of Haute-Volta
and Niger, in the exercise of his power to define the boundaries of the cercles:
that was the purpose of the arrêté of 31 August 1927.” 73
1.15. It is therefore true that “[i]t was on the basis of this Decree that, a few months
74
later, the Arrêté of 31 August 1927 and its Erratum of 5 October 1927 were adopted” .
1.16. The preparatory work for the Arrêté also leaves no doubt that this was the case:
⎯ in a letter dated 2April 1927 to the Governor of Niger regarding the incorporation of
certain territories of the Colony of Upper Volta into Niger, the Governor-General of FWA
acknowledges receipt of the letter of 19Fe bruary 1927 from the Governor of Niger
together with the appended Records of Agreement and adds:
“Once the Nevières mission has been able to establish the course of the
new boundary on the ground in the Botou region, I would be grateful if you
could send me a draft arrêté established in accordance with the procedure
75
provided for in the above Decree of 28 November [sic] 1926” ;
⎯ furthermore, the two Records of Agreement that were appended to the Governor of
Niger’s letter have just one citation: “Having regard to the Decree dated the
twenty-eighth of December, one thousand nine hundred and twenty-six”; and
72
Ann. MBF 26 (emphasis added).
7Frontier Dispute (Benin/Niger),Judgment, I.C.J.Reports2005, p.146, para.135; see also p.113,
para.[35]: “An arrêté général of 31August 1927 and the erratum thereto of 5October of the same year
determined the boundary between the co lonies of Haute-Volta and Niger” (same wording p.116, para.39 or
p.145, para.131 of the Judgment);or again p.147, para.136: “an arrêté whose purpose, as was clear from its
title, was to fix the boundary between Niger and Haute-Volta”.
7MN, p. 84, para. 6.11.
7Ann. CMBF 1 (emphasis added). - 18 -
⎯ a letter dated July 1927 from the Director of Political and Administrative Affairs of the
Standing Committee of the Government Council regarding the boundaries between Upper
Volta and Niger states:
“Report No.40, which was presented to the Standing Committee of the
Government Council at its meeting of 22 January 1927, announced that a draft
arrêté would subsequently be submitted to the committee to determine, as
proposed by the Lieutenant-Governors concerned and in accordance with the
terms of the Decree of 28 December 19 26, the precise boundary of the Colonies
of Niger and Upper Volta, in the area that has thus been reorganized.
I therefore have the honour of sending you this draft arrêté, which has
been established in accordance with the following Records of Agreement: the
first dated 2February 1927, determining the boundaries of the new Tillabéry
cercle with Upper Volta; the second dated 10February 1927, fixing the
boundaries of Say cercle and Upper Volta; and the third dated 9May 1927,
indicating the boundaries of Botou canton with that same Colony.
Should these provisions meet with your approval, I would be very grateful,
Governor-General, if you could sign the enclosed draft arrêté in Standing
76
Committee of the Government Council.”
1.17. However, it is not true that “[77he purpose of these texts could only have been to
transfer the above-mentioned cantons” : the transfer having been effected pursuant to the
Decree, the Arrêté and its Erratum draw the appropria te conclusions and determine the
resulting new inter-colonial bound78ies, as expressly stipulated by the Decree. In fact, as
Burkina noted in its Memorial , the Arrêté itself states that, “[t]he boundaries of the Colonies
of Niger and Upper Volta shall . . . be determined as follows” ; its intention is not therefore
to describe boundaries between cercles or cantons but rather those between the two Colonies.
Moreover, according to Niger, the Erratum w as adopted precisely to clarify the difference
between inter-colonial boundaries and boundaries between cercles, which had been confused
80
in the Arrêté of 31 August .
1.18. Moreover, the local officials expressly note, contrary to what Niger is now
claiming, that the Arrêté and its Erratum do not speak in terms of cantons but of boundaries.
1.19. Thus, in a letter dated 9August 1929 to his counterpart in Tillabéry, the
Commander of Dori cercle, Taillebourg, who was complain ing about the disadvantages
76
Ann.CMBF2 (emphasis added); Burkina has not found either report No.40 or the draft arrêté that is
described as being enclosed with the letter (the wordi ng of which was probably identi cal to the version that was
finally adopted on 31 August of the same year).
77
MN, p. 84, para. 6.11.
78MBF, p. 137, para. 4.95.
79Article 1 of the Arrêté of 31 August 1927.
80
See MN, p.20, para.1.26 and p.64, para.5.5; se e also MBF, p.137, para.4.95. For a different
explanation, which nevertheless compleme nts the one put forward by Niger, see Frontier Dispute (Benin/Niger),
Judgment, I.C.J. Reports 2005, p. 147, para. 136. - 19 -
81
resulting from “the 1927 delimitation” and proposing adjustments to it, made the following
very precise clarification:
c“otrsee, Arrêté and the Erratum signed by the Governor-General
nolonger refer to cantons but only to boundaries ; and that, I acknowledge, is
82
crucial.”
1.20. Similarly, in a letter dated 14 August 1929, the same Commander of Dori cercle
sent a copy of his letter of 9 August to the Governor of Upper Volta in which he notes (with
83
regret) “that the list of cantons [drawn up in the Brévié-Lef illiatre Record of Agreement ]
has not been recorded in the Arrêté and the Erratum of delimitation” while considering that “it
is logical to think that Volta has transferred cantons to Niger and that it was the boundaries of
84
those cantons that were adopted as the boundaries of the two Colonies” ; however, this is
not the case, a fact that he also laments in a letter of the same day again to the Commander of
Tillabéry cercle, in which he notes that the decision on delimitation has been made;
85
therefore, he adds: “I realize th at my request has a weak foundation” . The same is true of
Niger’s argument, which appropriates th e position of Commander Taillebourg, without
indicating that 86is was not a description of the situation resulting from the Arrêté but a
criticism of it .
1.21. Moreover, and this is another of the numerous inconsistencies in Niger’s position,
Niger acknowledges that “[i]t was . . . solely on the basis of the three Records of Agreement
87
of 2 February, 10 February and 9May 1927[ ] that the new boundaries of the two Colonies
resulting from these territorial changes were subsequently described in Arrêté No.2336 of
88
31 August 1927” . In so doing, it recognizes:
⎯ that Commander Taillebourg was not heard;
⎯ that the 1927 Arrêté is the consequence of the territorial changes resulting from the
transfer of certain cantons to Niger; and
⎯ that it describes the new boundaries of the two Colonies resulting from this.
1.22. Neither Burkina nor Niger has anne xed to its Memorial documents reacting
directly to the original Arrêté and none is in Burkina’s possession. Nevertheless, Niger relies
81Thus clearly showing that he considered the officials on the ground to be bound by ⎯ see para.1.32
above. Moreover, the Commander of Dori cercle adds: “The resident of Téra, bound by the texts, I
acknowledge . . .” (MN, Anns., Series C, No. 24). The version of the annex provided by Niger is truncated.
82Ibid. (emphasis added).
83
Commander Taillebourg is referring here to the list of cantons included in the Brévié-Lefilliatre Record
of Agreement of 2 February 1927 (Ann. MBF 30), which he mentioned earlier in the letter.
84
MN, Anns., Series C, No. 25.
85
MN, Anns., Series C, No. 27.
86On this point see Subsection 2 below.
87Anns. MBF 30, 31 and 33 respectively.
88
MN, p. 19, para. 1.26 (emphasis added). - 20 -
heavily on what it calls “the Delbos/Prudon Agreement of 1927” 89, thus suggesting that this
90
was an alternative to the Erratum . However:
91
1. it states that it is unable to produce the text of that “agreement” ;
2. the documents said to have formed cert ain parts of it and which Niger produces are
92
themselves incomplete ;
3. these documents do not contain any criticism of the original Arrêté of 31 August;
4. Niger maintains that these documents were not taken into account in the preparation of
93
the Erratum , which is not in doubt, even though it is actually unlikely that they did not
reach Dakar ⎯ if indeed they were ever sent there ⎯ before the latter was adopted on
5October1927, given that the documents on which Niger relies date variously
from 3 , 4 95and 27August 1927 96. But this only serves to make a more significant
point: these documents, of which Niger makes so much, were deliberately not taken into
account in the preparation of the Erratum;
5. the documents are actually internal to each Colony, being prepared by the Commanders of
Dori and Tillabéry cercles in response to requests from the Governors of their respective
97 98
Colonies ⎯ Upper Volta and Niger ⎯ and addressed to them ; and it is highly likely
89
See in particular MN, p. 28, para. 2.4; p. 72, para. 5.11.
90
See in particular MN, pp. 28-30, para. 2.4 and p. 72, para. 5.11.
9MN, p. 19, para. 1.25: “We do not have the report of Administrator Delbos on the route followed on that
occasion together with Administrator Prudon”; and p. 88, para.6.12: “The report from Delbos on the joint
reconnaissance carried out in June, sent on 3 August 1927 to the Governor of Upper Volta under cover of a Note
bearing the No. 438 has not been found.”
92
See in particular “Extract No.25 from the Tour Report of AdministratorPrudon dated 4August1927”
(MN, Anns., Series C, No.15). Burkina Faso formally prot ests against the withholding of information by Niger:
the latter has reproduced this “Extract No.25” in isolation and has neither appended, nor filed with the Registry,
the full text of the document from which it is taken. The Ag ent of Burkina Faso has asked for the full text of that
document to be provided through the Registrar of the C ourt (letter dated 25November2011); at the time of
printing of this Counter-Memorial, no response had been given to that request.
93
MN, p. 19, para. 1.25, “[t]hese documents did not, however, reach Dakar in time to be taken into account
in the preparation of the Arrêté of 31August1927”; p.72, para.5.11, “t heir views were not taken into account,
because they arrived after publication of the Arrêté”; p. 88, para. 6.12, “[i]n any event, the proposals from the two
officials reached Dakar too late, after the Arrêté of 31 August 1927 had been publis hed, and could have no effect
on its text, or on the text of the Erratum”.
9The Delbos report, of which Niger has only appended th e sketch-maps that were originally annexed to it
(see MN, Anns., Series C, No. 20) was sent to the Governor of Upper Volta on 3 August 1927 (see ibid.).
9Extract No.25 from the Tour Report of Ad ministratorPrudon, Commander of Tillabéry cercle (Niger),
dated 4August1927 (MN, Anns., Series C, No.15), to whic h a sketch-map is attached (MN, Anns., Series D,
No. 3).
9Letter from Administrator Delbos, Commander of Dori cercle (Upper Volta), to the Governor of Upper
Volta dated 27August 1927, containi ng a draft delimitation of the cercles of Dori and Tillabéry (MN, Anns.,
Series C, No. 16).
97
Only the request from the Governor of Upper Volta has been included in the dossier (telegram/letter
No.1166/AG from the Lieutenant-Governor of Upper Volta, Hesling, to the Commanders of Dori and Fada
cercles, dated 27April1927 - MN, Anns., Series C, No. 11): “Request send me soon as possible precise
information to enable preparation Arrêté général fixing new boundaries between Colonies Niger and Upper
Volta”. There is every reason to think that the Gover nor of Niger sent a similar request to the Commander of
Tillabéry cercle. - 21 -
that the Governors did not consider it worthwhile to transmit them to the central
government of FWA;
6. consequently, whatever the situation was, these documents did not in any way influence
the delimitation adopted by the Erratum, as also emerges very clearly from the rough
99
superposition of the two sketch-maps on which Niger relies over the line in the Erratum;
and
7. this same diagram shows that the two mi ssion sketch-maps of Delbos and Prudon do not
entirely coincide and that, in any case, there was actually a disagreement about the alleged
“actual” boundaries “in practice”. This is show n in particular by discrepancies between
the Delbos and Prudon documents on the one ha nd and, on the other hand, between those
documents and the line described in the Record of Agreement between the Governor of
Niger and the representative of the Governor of Upper Volta, known as
“Brévié-Lefilliâtre”, recording the incorporation into the Colony of Niger of the
Territories on the right bank of the river, pursuant to the Decree of 28 December 1926 100,
101
which essentially follows the Coquibus line .
1.23. In other words, the reports of Delbos (as virtual as they remain for the purposes of
the present case) and Prudon
⎯ do not constitute an inter-colonial “agreement” in the sense that Niger understands it;
⎯ were not taken into consideration at all for the purposes of preparing the 1927 Arrêté and
its Erratum;
⎯ do not coincide with the delimitation agreed on 2February 1927 by the two Governors,
which did form the basis for the line definitively adopted on 5 October.
98Cf. Extract No.25 from the Tour Report of Administrator Prudon, Commander of Tillabéry cercle
(Niger), dated 4 August 1927 (MN, Anns., Se ries C, No. 15), p. 3: “I would ask you kindly to support the wishes
expressed by these Chiefs with the Governor of Upper Volta, since all these villages or groupements who wish to
transfer to Niger have for many years been settled on theterritories recently attached to that colony”. Delbos’s
letter has not been included in the dossier.
99
Sketch-map (“reduction to a scale of 1:1,000,000 of the 1:500,000 map sent by letter No. 438 of 3 August
1927 based on routes surveyed by the Commander of Dori cercle, Delbos”) of the boundary as surveyed in June
1927 by Chief Administrator Prudon, MN, Anns., Series C, No. 20 and Series D, [No. 2 and] No. 3.
100
Ann. MBF 30.
101Administrator Delbos points out th at the boundaries agreed in the Brévié-Lefilliatre Record of
Agreement of 2February 1927 “had been established on the basis of the map prepared by Captain Coquibus”
(letter No.731 dated 17December 1927, MN, Anns., Series C, No.20, p. 1). The line drawn jointly by
Administrators Delbos and Prudon only follows it in part (cf. Prudon’s Tour Report, MN, Anns., Series C, No. 15,
p. 1: “Apart from this slight modification [regarding the chain of hills to the north of Nababori], following natural
frontiers, the delimitation of the cercle made by Lieutenant Coquibus is indeed the line that we followed and the
line recognized by the various chiefs of the frontier cantons in the two colonies concerned”). - 23 -
2. The confirmation of the delimitation effected by the Erratum
1.24. Niger makes much of the disadvantages resulting from the 1927 boundaries and
the criticisms to which they gave rise. It devotes an entire chapter to this issue 102, which it
summarizes as follows:
“Thus, as has already been explained, this text was criticized from the
outset by the colonial officials and authorities of the two Colonies. From all
sides there was a chorus of complaints over the lack of precision in the
boundaries and the constant disputes to which those shortcomings gave rise on
the ground. The text was full of the ki nd of errors to be avoided in the
description of a frontier, as was pointed out, in general terms, by the Head of the
French West Africa Geographical Department in a letter of 8 May 1942 . . .
It follows from the summary and imprecise nature of the description of the
boundary in several sectors that the practical scope of the Arrêté and its Erratum
103
remains extremely limited.”
1.25. Although Niger prudently deduces that “[ i]t is therefore necessary to consider the
possibilities for interpreting these texts by having recourse to cartographic or textual criteria,
104
preparatory work or the practice” , in making these observations it is clearly calling into
question the frontier described by the Erratum: for Niger it is not a question of “consider[ing]
the possibilities for interpreting these texts” but of actually correcting what it regards as
“errors”. And indeed this is what it seeks to do in its Memorial when it rein105ts a frontier
line that departs considerably from the one described by the 1927 Erratum .
1.26. In fact, while it is true that some co lonial officials in the field did criticize the
delimitation effected by the Arrêté and its Erratum, sometimes in rather harsh terms, far from
strengthening Niger’s argument, these attacks confirm the reality of the disputed delimitation.
1.27. According to Niger, “[t]hroughout this period [from 1927 to independence], the
conclusions of the Delbos/Prudon Agreement of 1927 continued to serve as a reference basis.
They were often cited or recommended” 106. However, all of the “examples” (no doubt the
107
fruit of extensive research by Niger) that are given in support of this statement essentially
lead to one conclusion ⎯ which Niger refrains from drawing: the authors of these documents
seemed, in certain cases, to have a preference for this supposed agreement, but
1. they contrasted it with the Arrêté and its Erratum; and
2. often they recognized, directly or a contrario, that, even if they regretted the fact, it was
the Erratum and not this supposed “agreement” that had delimited the frontier between the
two Colonies; and
102Chap. II ⎯ The difficulties and incidents in the disputed area (MN, pp. 25-38).
103
MN, p. 66, para. 5.6.
104
Ibid.
105See Section 2 of the present chapter and Chaps. III and IV of the present Counter-Memorial below.
106MN, p. 28, para. 2.4.
107Ibid., pp. 28-29. - 24 -
3. in other cases, they referred not to the deli mitation attributed to Delbos and Prudon but to
the arrangements for its implementation.
1T2u8.:
⎯ the telegram/letter No. 815 from the Commander of Tillabéry cercle to Dori cercle dated
10 October 1929 recommends to “maintain status108o, namely tolerance zone accepted in
1927 without encroachment or spoliation” ; the issue here is not delimitation;
⎯ the report from the Commander of Dori cercle dated 7July1930 notes that the Arrêté
général of 31 August 1927 “took no account of the delimitation carried out on the ground
by the two cercles Commanders of Dori and Tillabéry” 109; given that an agreement
between officials in the field could not take precedence over an arrêté général (and a
subsequent one at that), this amounts to a recognition that the Arrêté and its Erratum had
established the frontier (even if that frontier gave rise to incidents);
⎯ the letter sent on 10April 1932 to the Gover nor of Upper Volta by the Commander of
Dori cercle (in which he reports on an agreement reached with his Tillabéry counterpart,
the “Roser/Boyer” Agreement 11) lists two possibilities for identifying the “exact course
of the boundary”, the first of which is to consider that the Arrêté and the Erratum “were
intended to endorse the work of Ad ministrators Delbos and Prud’hon [sic], and officially
establish the boundary they proposed after their inspection tour”, which Commander
111
Roser regards as “the only logical one” ; nevertheless, the author remarks that Delbos’s
protests as voiced in his letter of 17 December 1927 112 did not receive “any response” and
that “no new Erratum was provided to correct the errors in question” 113. In other words,
only a new text could “repair” the errors attributed to the modified Arrêté and simply
applying the “Delbos/Prudon Agreement” would not be sufficient;
114
⎯ the Record of Agreement between Garnier and Lichtenberger of 15[?] April 1935
resolving a territorial dispute at Sinibellabé does not refer directl
y to the supposed
“Delbos/Prudon Agreement”, but states that “in principle, this boundary [between Dori
and Téra] shall be determined in accordance w ith the indications given in letter No.438
from the 115mander of116ori] cercle to the Governor of Upper Volta of
3 April [ ] 1927” ; the signatories are very careful to respect the orographic tradition
and to resolve the dispute in accordance with it, not necessarily by relying on texts: it is
worth noting that a few days previously, on 13April1935, the same Administrators,
Garnier and Lichtenberger, had resolved a dis pute at Ouiboriels or Vibourié, believing
that they were applying the Erratum:
10MN, Anns., Series C, No. 31.
10MN, Anns., Series C, No. 38 (emphasis added).
11MN, p. 74, para. 5.12 and p. 90, para. 6.14; and MN, Anns., Series C, No. 45.
111
MN, Anns., Series C, No. 45, p. 5.
112
MN, Anns., Series C, No. 20.
11MN, Anns., Series C, No. 45, p. 5.
11This is probably a typing error: the annex concerne d (MN, Anns., SeriesC, No.57) is actually dated
25 April 1935.
11According to Niger, this should read “August” (MN, p. 29, para. 2.4).
116
MN, Anns., Series C, No. 57, p. 5. - 25 -
“Referring to the delimitation determined by the Arrêté of 31 August 1927
(Erratum) between Dori and Téra, we decided to visit the site in order to observe
the placement of said land [Ouiboriels, disputed] in relation to the
117
above-mentioned boundary.”
⎯ To resolve the problem definitively, they establish a marker (thereby committing a
geodesic error) on “a notional straight line starting from the Tong-Tong astronomic
marker and running to the Tao astronomic marker” 118, which ⎯ with the exception of the
119
error ⎯ is a pure and simple application of the 1927 texts ;
⎯ the letter of 9May 1935 from the Commander of Dori cercle to the Governor of Niger
refers to the Garnier/Lichtenberger Record of Agreement, after being careful to note ⎯
which Niger forgets to point out ⎯ that the “[b]oundaries of Dori cercle with Téra
Subdivision have been determined by the Arrêté of 31 August 1927 fixing the boundaries
of the Colonies of Upper Volta and Niger followed by an Erratum dated
5 October 1927” 12;
⎯ in his letter of 10May 1935 to the Commander of Tillabéry cercle , the Head of Téra
Subdivision confines himself to indicating: “To ensure that our data about the frontier
coincides as far as possible with Dori’s, I have taken a copy of Administrator Delbos’s
letter of 27 August 1927; I am quite happy to accept this boundary, but I think that, for
121
the sake of the local people, it needs to be marked out” ;
⎯ on 19 May 1943, the Commander of Dori writes to say that he has found in his archives
an “undated and unsigned sketch-map … most probably by Mr.Roser and likely to date
from 1932”. The boundary drawn on the sketch-m ap, which seems to conform with the
Roser report of 1932 is a “ de facto boundary , tacitly confirmed by the
Garnier/Lichtemberger Record of Agreement of 1935, but not submitted for approval by a
higher authority, which we will have to st udy again and, if necessary, submit to the
Governor for approval” 122. He writes that he has also found another rough drawing
which “seems to be by Administrator Delbos and therefore to date from 1927
(Delbos/Prudhon [sic] settlement, following which proposals were made to the Governors
of Niger and Upper Volta to modify the 1927 Arrêté and the subsequent Erratum,
123
proposals which were not acted upon)” ; that says it all . . .
⎯ the letter of 11July 1951 shows that in the view of the Head of Téra Subdivision, the
boundary might be poor, but it is given by the 1927 texts:
“[T]he Commander of Dori cercle stated again that he believes it is
important to demarcate the boundary on the basis of the Erratum to the Arrêté of
the Governor-General of 1927, by conn ecting the Tao boundary marker directly
with Bossébangou.
117
MN, Anns., Series C, No. 56, p. 2.
11Ibid., pp. 2-3.
11See Chap. III, Section 2, 1, below.
120
MN, Anns., Series C, No. 58, p. 1.
121
MN, Anns., Series C, No. 59. Again the document is barely legible.
12MN, Anns., Series C, No. 67 ⎯ underlining in the text.
12Ibid. (emphasis added). - 26 -
Apparently, he does not understand so me of the consequences of that
position. The attached reproduction of a sketch-map prepared by Mr.Delbos
noted that this action would cut off the Yagha from a salient delimited by Iga,
Tingou and Nabambori.
The inaccuracy and imprecision of the Erratum have moreover been
pointed out numerous times. As an exam ple, I refer to the solution proposed by
Mr. Roser . . . in 1932” 124;
⎯ the Record of Agreement of 17 May 1953 125, which Niger also invokes in support of its
argument that “the Delbos/Prudon Agreement of 1927” served as a “reference basis”,
settles a dispute between two villages without making any reference to that “agreement”
or, moreover, to any other text;
⎯ finally, the Lacroix report of 24December 195[ 3] explains that “[t]he basic document
relevant to these issues [Téra-Dori delimitation] is the Arrêté général of 31 August 1927,
126
as amended by an Erratum of 5October of the same year” , and it is only in an
historical context that it adds:
“Mr.Delbos and Mr.Prud’hon [sic] . . . had travelled this boundary
previously . . . Their conclusions may ha ve been taken into consideration in the
provisions of the text cited above, a lthough the Delbos report was not sent to
Ouagadougou until 27August. If that was the case, it is unfortunate that the
relevant departments of the Government-General did not adopt the draft Arrêté
as proposed by Mr.Delbos , because, although almost identical with the 127
provisions of the Arrêté and its Erratum, it provided these additional details.”
None of the documents invoked by Niger for this purpose suggests that the officials in the
cercles concerned confused the proposals made in 1927 by Delbos and Prudon (which do not
appear to coincide 128and which are often judged to be preferable to the Arrêté) with the
delimitation in force, as established by the Arrêté and its Erratum.
1.29. In fact, there is no doubt that, as soon as it was published, the Erratum of
5October 1927 attracted criticism, sometimes st rong criticism, from certain administrators.
However, although they protested against the delimitation and criticized it, they recognized,
implicitly or explicitly, albeit with regret, that it was established.
1.30. The beginning of the controversy over the boundaries fixed by the Arrêté was
marked by the letter from the Commander of Dori cercle, Delbos, dated 17 December 1927,
which lists the errors that he believes it to contain and asks that “the boundaries indicated in
129 130
[his] letter 438 be maintained” .
12MN, Anns., Series C, No. 73, p. 1 (emphasis added).
12MN, Anns., Series C, No. 76.
126
MN, Anns., Series C, No. 79, p. 1.
127
Ibid., p. 2 [emphasis added].
12See sketch-map No. 2 above.
12Probably his letter to the Governor of Upper Volta dated 27 August 1927 (MN, Anns., Series C, No. 16).
13MN, Anns., Series C, No. 20. - 27 -
1.31. His successor, Taillebourg, also contest s the very principle on which the new
frontier between the two Colonies was based (it was defined de novo instead of being based
on the pre-existing cantonal boundaries ⎯ in so far as they had been defined, which in fact
they had not been) and points out the resulting disadvantages relating to the re-allocation and
re-occupation of land ⎯ causing various conflicts over ownership ⎯ and to the carrying out
of censuses of the population, which caused particular difficulties with raising taxes 131.
1.32. Subsequently, it is true that the colonial period is punctuated by complaints from
certain territorial officials. However, these co mplaints are directed against the existing
delimitation that is being experienced “in prac tice” as such. For example, the Commander of
Dori cercle complains to the Governor of Upper Volta that:
“the 1927 delimitation seems to have been designed to create difficulties for
Volta”; “I could, with your permission, prepare a report on the difficulties that
the 1927 delimitation has caused, a report requesting a new delimitation for the
end of 1930” 132.
In the same spirit, for their part, all of Taille bourg’s letters of July-August 1929 point out the
excessive rigours of the delimitation 133.
1.33. It is also necessary to take into acc ount extrinsic factors. For example, it was
common practice amongst colonial officials to denounce so-called artificial frontiers.
Moreover, the arbitrary nature of the delimitation was due to fiscal considerations or for
reasons of administrative efficiency, which had scant regard for the physical and ethnic nature
of territories with which the colonizer was unfamiliar: the priority was to deploy a direct,
centralized administration that was capable of assimilating the populations 134; sometimes this
detachment from reality, and in particular the ethnographic reality, was even intentional, to
prevent groupings of related populations and to bring together very different ethnic groups,
thereby quelling attempts to resist the occupation 135.
1.34. Niger also cites a number of agreements between officials seeking to clarify the
136 137
boundaries of their administrative divisions . None of them was approved by a higher
131
See paras. 1.19-1.20 above and MN, Anns., Series C, Nos. 24, 25 and 27.
13Letter from the Commander of Dori cercle to the Governor of Upper Volta dated 26February 1930,
MN, Anns., Series C, No. 32 (emphasis added).
13See paras. 1.19-1.20 above.
13See I. Brownlie, African Boundaries , London, Hurst, 1979, pp.6-7, or J. de Pinho Campinos,
“L’actualité de l’uti possidetis”, in SFDI, La frontière, Paris, Pedone, 1980, pp. 96-97.
13See Y. Person, “L’Afrique noire et ses frontières”,Revue française d’études politiques africaines,
No. 80, August 1972, pp. 31-32. There are even instances of colonial boundaries being modified depending on the
need for labour on either side of the frontier: see L. Joos, “Des frontières trop souvent établies au gré des
colonisateurs”, Le Monde diplomatique, February 1965, p. 11.
13MN, pp. 33-34, para. 2.8.
137
According to Niger, an exception should be made for the 1935 agreement on the establishment of a
marker at Vibourié (“In reality, this was the only agreement between cercles, subsequent to 1927, which was
approved by a higher authority” ⎯ MN, p.93, para.6.20). It was i ndeed approved by a higher authority;
however, it was not approved by the Governor-General of FWA (author of the 1927 Arrêté) but by the Governor
of Niger, which is appropriate as, since Upper Volta had been dissolved, it concerned an intra-colonial boundary (a
delimitation between cercles belonging to the same Colony, Niger). Inci dentally, as Niger itself points out: “this
agreement dates from after the disappearance of Upper Volta and hence its retention following the reconstitution
of the Colony could be regarded as debatable” (ibid.). - 28 -
authority and none is therefore binding. In any case, these “agreements” are not recognized
by the Agreement of 28 March 1987 as documents that enable the course of the frontier to be
determined and the Parties have made no declaration agreeing to take them into consideration
for that purpose.
1.35. In the same spirit, it can be noted th at Niger devotes all of Chapter II, Section 2,
of its Memorial to “[t]he difficulties encountered during the period subsequent to
138
independence” . This account of incidents that can have no influence on the delimitation of
frontiers inherited from colonization is also of no legal consequence: since they occurred
after the “‘photograph of the territory’ at the critical date” 139, these facts cannot in any way
140
shift or weaken the colonial title constituted by the Arrêté and its Erratum . Moreover, even
if they could do so under general international law ⎯ quod non ⎯ then the 1987 Agreement
between the Parties would prevent them from being taken into consideration.
1.36. Furthermore, Niger’s insistence on highlighting the criticism of the 1927 texts
backfires: the fact that certain colonial officials regarded the boundary as arbitrary shows that
they were aware both of its existen ce and of the line that it took. A fortiori, their repeated
requests for that delimitation to be modified suggest that they regarded it as established.
However, that boundary was never modified.
1.37. Moreover, although these criticisms might have provided an opportunity to adopt
a new delimitation arrêté, that opportunity was never taken by the competent authority. For
example, in a letter dated 19 March 1930, the Governor of Upper Volta asked the Commander
of Dori cercle to draw up a report on the difficulties created by the 1927 delimitation and to
attach “any proposals that you see fit” 141; the report that he produced in response to this
142
request clearly did not lead to any new delimitation even though it concluded: “After this
tour, if appropriate, I will submit proposals to rectify the boundary between Dori cercle and
143
Tillabéry cercle.” The officials’ complaints were heard, but no action was taken as a
result. As the Court noted in the Burkina/Mali case, it cannot take account of a modification
proposed by a cercle administrator that contradicts the frontier title having force of law
144
between the Parties if it has not been approved by the competent higher authorities .
1.38. Incidentally, although it emphasizes the shortcomings attributed to the Arrêté and
its Erratum, Niger nevertheless recognizes that the boundary was never modified in the light
of the officials’ complaints. Upon asking what effect should be given to local agreements
made between the Commanders of different cercles, it recalls that they had “no power to
replace the competent colonial authorities in order to modify or clarify the boundaries
138
MN, pp. 35-38, paras. 2.9-2.11.
139
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 568, para. 30. See
also MBF, p. 26, para. 1.21 and p. 57, paras. 2.6-2.7.
140Land, Island and Maritime Frontier Dispute (E l Salvador/Honduras: Nicaragua intervening),
Judgment, I.C.J. Reports 1992, p. 398, para. 61; or Land and Maritime Boundary between Cameroon and Nigeria
(Cameroon v. Nigeria: Equatorial Guinea intervening), Judgment, I.C.J. Reports 2002, pp. 351-355, paras. 64-70.
141MN, Anns., Series C, No. 33.
142Report of 7 July 1930 on the difficulties created the delimitation estab lished in 1927 between the
Colonies of Niger and Upper Volta ( Arrêté of 31 August 1927) regarding the boundaries between Dori cercle and
Tillabéry cercle, MN, Anns., Series C, No. 38.
143Ibid., p. 12 (emphasis added).
144Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 627, para. 137. - 29 -
145
between Colonies” and goes on to list the specific cases in which officials recalled that
only the higher colonial authorities (Governor s and the Governor-General) were empowered
to effect inter-colonial delimitations46.
1.39. Niger points out that “[d]espite the wish frequently expressed by officials of the
two Colonies for the course of the boundary to be clarified by a new text so as to accord 147e
closely with the true boundaries of the cantons in practice, this was never done ” . It
therefore recognizes that to modify the line of the boundary determined by the Erratum a
further erratum was essential. None was ever adopted.
SECTION 2
A PPROXIMATIONS AND ERRORS IN THE IMPLEMENTATION OF
A RTICLE 2 OF THE A GREEMENT OF 28 M ARCH 1987
1.40. Having sought to discredit the 1927 Arrêté and its Erratum, Niger endeavours to
empty Article2 of the Agreement of 28Marc h 1987 between the Parties of all practical
meaning. Article 2 reads as follows:
“The frontier shall be demarcated by boundary markers following the
course described by Arrêté 2336 of 31August 1927, as clarified by
Erratum2602/APA of 5October 1927. Should the Arrêté and Erratum not
suffice, the course shall be that shown on the 1:200,000-scale map of the Institut
Géographique National de France, 1960edition, and/or any other relevant
document accepted by joint agreement of the Parties.”
The second subparagraph of the preamble to th e Special Agreement of 24 February 2009, by
which the Parties referred this case to the Court, specifically cites this provision.
1.41. Niger’s Memorial displays a singular , particularly lax, subjective and uncertain
notion of the expression “[s]hould the Arrêté and Erratum not suffice”, which appears in
Article 2 of both the Agreement of 28 March 1987 and its Protocol (1), and gives no attention
to the provision limiting the documents to which the Parties have agreed to have recourse in
such a case (2).
14MN, p. 73, para. 5.12.
146
MN, pp.73-74, para.5.12; cited on these pages are: a telegram of 10October 1929 from the
Commander of Tillabéry (MN, Anns., Series C, No. 31); the agreement reached at Ossolo Pool on 12 March 1931,
which had to be approved by the Governors of the two Co lonies (MN, Anns., Series C, No. 41) but which, though
it was approved by the Governor of Niger by letter No.1361 A.G.I. of 13April1931 (mentioned in
Ann.CMBF3), did not receive the a pproval of the Governor of Upper Volta (see the annual report of Tillabéry
cercle for 1931, Ann. CMBF 3, p. 3), despite calls by Niger’s officials for it to be ratified (see letter No. 40 A.G.I.
from Tellier, the Chef de cabinet of the Lieutenant-Governor of Upper Vo lta, dated 6 February 1932, MN, Anns.,
Series C, No. 44, p. 2) ⎯ neither did it receive the approval of the Governor-General of FWA; the Roser/Boyer
Agreement of April 1932 (MN, Anns., Seri esC, No.45); a circular of sent by the Governor-General of
FWA to his Lieutenant-Governors recalling that “[a]ny boundary of a cercle or a subdivision merely deriving from
a practice, not yet endorsed by an official text, should be confirmed as soon as possible by a local arrêté in the case
of subdivision boundaries, and by a draft Arrêté général in the case of cercle boundaries” (MN, Anns., SeriesC,
No.48, Niger’s italics) ⎯ which clearly shows that only a new dArrêté général was able to modify the
1927Erratum in law; finally, Nige r acknowledges that “[t]he only agreement from the colonial period which
appears to have been regarded as determining the boundary of Tillabéry cercle was that adopted by the Record of
Agreement of 13April1935 (concerning the Ouiboriels [V ibourié] marker)”, and again that Agreement was only
approved by the Governor of Niger, st ates Niger in its Memorial (p.74, para.5.12); on the Agreement and its
approval, see fn. 137 above. See also MBF, pp. 66-73, paras. 2.35-2.50.
14MN, p. 75, para. 5.13 (emphasis added). - 30 -
1. “Should the Arrêté and Erratum not suffice”
148
1.42. On presenting the Erratum, Niger describes it as “particularly rudimentary” , in
a bid to justify an argument whose main objective is to avoid the Erratum’s application 149,
without ever taking the trouble to define what might be denoted by “not suffice” within the
meaning of Article 2 of the 1987 Agreement.
1.43. The text of the Erratum is described as “particularly” or “extremely succinct”, but
this characteristic is stated without ever being demonstrated. In general, Niger confines itself
to referring to the length of the section of the frontier described:
⎯ in respect of the Téra sector, the Erratum is said to be “particularly succinct” on the
grounds that it only mentions two frontie r points over a distance of 150km: the
150
Tong-Tong astronomic marker and the Tao astronomic marker ⎯ which is actually
incorrect: the Erratum specifies that th e frontier reaches “the River Sirba at
151
Bossebangou”; but Niger declares peremptorily that this “remains problematic” ;
⎯ in the Say sector, the argument is based th is time on the number of lines in the Erratum
152
compared with the distance described : five lines to describe almost 160km ; “[t]his
description of the boundary was thus extremely succinct” 153.
1.44. Twice, Niger does not hesitate to dismiss outright the very words of the Erratum.
This is the case firstly with the expression “at th e level of the Say parallel”, which is used in
the Erratum to designate the place where “turning back to the south, [the frontier line] again
cuts the Sirb154 Niger asserts that “the text of the Erratum should not be read too literally on
this point” . Similarly, Niger does not hesitate to dismiss the term “salient”, which it says
“makes no sense in relation to the inter-colonial boundary” 155, despite subsequently seeking
to “interpret” this part of the Erratum in its favour.
1.45. For the rest, Niger appears to be more cautious, but it in fact dismisses provisions
of the Erratum without saying so explicitly:
⎯ either it invents points that are not mentioned by the Erratum (for example the Vibourié
156
marker, presented as an interpret157on of the Arrêté ; or the “frontier marker” on the
Niamey-Ouagadougou road );
14MN, p. 65, para. 5.6.
149
According to Niger, “indications [concerning the cantons]... can supplement the summary description
in the Arrêté and Erratum of 1927” (MN, p. 86, para. 6.11).
150
MN, p. 83, para. 6.9.
151
Ibid.
15MN, p. 104, para. 7.11.
15Ibid.
15MN, p.115, para.7.32. It is rather ironic to see that here Niger cautions against reading the amended
Arrêté “too literally”, while elsewhere it continually criticizes it for being excessively succinct.
15MN, p. 112, para. 7.26.
156
MN, p. 93, para. 6.20.
157
MN, p. 120, para. 7.40. - 31 -
⎯ or, conversely, it conjures away points that are explicitly mentioned by the Erratum ⎯
158
Bossébangou, for example ⎯, on the pretext that they are a mistake , the disappearance
of that point in itself having an impact on the subsequent course of the line, as it also
leads to the disappearance of the salient, which no longer makes any sense in the line
159
claimed by Niger ;
⎯ or it wrongly invokes an alleged silence on the part of the Erratum, for example when it
asserts that: “[f]rom the Tao astronomic marker... the official text gives no further
indication until the point where the inter-colonial boundary rejoins the boundary of Say
160
cercle” ; but this is simply a statement of fact: the text is no more or less articulate
here than it is with regard to the previous section (from the Tong-Tong marker to the Tao
161
marker), with which Niger is perfectly content ;
⎯ or again, it recognizes that the text of the Erratum is perfectly clear before going on to
wave it aside: this is the case with the sector “end of the salient/start of the Botou Loop”,
where Niger acknowledges that “[t]his description[ 162] appears to be of great simplicity”,
before adding the non sequitur “[h]owever the straight-line boundary which it establishes
appears to have no basis in the situation prior to the adoption of the Erratum and was
never confirmed in the subsequent practice” 163.
1.46. Niger commits two errors here: firstly, it (wrongly) takes as its starting point that
the Arrêté and its Erratum intended to establish the status quo ante and were referring
implicitly to that, when in fact they sought to “fix the boundaries of the Colonies of Upper
164
Volta and Niger” ; secondly, it seeks to give precedence to so-called subsequent practic
e
over the clear text of the Erratum, something whic h is not in conformity either with Article 2
of the 1987 Agreement between the Parties or with the general principles for the delimitation
165
of land frontiers .
1.47. In reality, Niger’s “judicial strategy ” is not based on any legal principle. It
favours the pick-and-choose technique and only accepts the few points in the Erratum that
166
seem likely to serve its own interests best . For the rest, it is all a pretext to criticize the
Arrêté and its Erratum so as to rule out their app lication in favour of a range of instruments
(“agreements” allegedly concluded between local officials, exchanges of letters between
158
MN, p. 105, para. 7.14.
15MN, p. 112, para. 7.26.
16MN, p. 93, para. 6.21.
16In the Téra sector, which is referred to here, the Erratum reads as follows:
“the Tong-Tong astronomic marker; this line th en turns towards the south-east, cutting the
Téra-Dori motor road at the Tao astronomic marker located to the west of the Ossolo Pool, and
reaching the River Sirba at Bossebangou. It al most immediately turns back up towards the
north-west, leaving to Niger, on the left bank of th at river, a salient which includes the villages of
Alfassi, Kouro, Tokalan, and Tankouro; then, turni ng back to the south, it again cuts the Sirba at
the level of the Say parallel.”
16“From that point the frontier, follow ing an east-south-east direction, cont inues in a straight line up to a
point located 1,200 m to the west of the village of Tchenguiliba.”
16MN, p. 116, para. 7.35.
16See paras. 1.10-1.12 above.
165
See paras. 1.49-1.53 and 4.3-4.8 below.
166
Essentially as far as the sector with the four villages is concerned; see MN, pp.112-116,
paras. 7.25-7.33. - 32 -
colonial authorities, mission reports) or vari ous kinds of behaviour (alleged colonial or
post-colonial effectivités) which neither Article2 of the 1987 Agreement between Burkina
and Niger nor the general principles of international law that are applicable in this regard vest
with the slightest legal authority.
1.48. In this regard it is suf ficient to recall, in the famous words of the Chamber of the
Court in the Burkina/Mali case, which have been repeated many times, that:
“Where the act corresponds exactly to law, where effective administration
is additional to the uti possidetis juris, the only role of effectivité is to confirm
the exercise of the right derived from a legal title. Where the act does not
correspond to the law, where the territory which is the subject of the dispute is
effectively administered by a State ot her than the one possessing the legal title,
preference should be given to the holder of the title. In the event that the
effectivité does not co-exist with any legal title, it must invariably be taken into
consideration. Finally, there are cases where the legal title is not capable of
showing exactly the territorial expanse to which it relates. The effectivités can
167
then play an essential role in showing how the title is interpreted in practice.”
In accordance with these rules,
“To determine the course of the intercolonial boundary at the critical date
it is necessary to examine first the legal titles relied on by the Parties, with any
168
effectivités being considered only on a confirmatory or subsidiary basis” .
1.49. Furthermore, where there is a solid legal title ⎯ as is clearly the case with the
1927 Erratum 169 ⎯ the Court refuses to disregard its text for any reason whatsoever and even
merely to examine the additional arguments that the Parties have discussed during the
proceedings 170. Thus, in the case concerning Cameroon v. Nigeria, the Court noted “that the
text of paragraph 25 of the Thomson-Marchand Declaration provides quite expressly that the
boundary is to follow ‘the incorrect line of the watershed shown by Moisel on his map’” and
held that “[s]ince the authors of the Declar ation prescribed a clear course for the boundary,
171
the Court cannot deviate from that course” . In the same Judgment, such unconditional
respect for the text also appears in connection with the interpretation of paragraphs 26 and 27
of the Thomson-Marchand Declaration 17: although it recognizes that the Nigerian village of
167
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, pp. 586-587,
para.63; confirmed by Land, Island and Maritime Frontier Dis pute (ElSalvador/Honduras: Nicaragua
intervening), Judgment, I.C.J. Reports 1992, p.398, para. 61; Land and Maritime Boundary between Cameroon
and Nigeria (Cameroon v. Nigeria: Equatorial Guinea interven ing), Judgment, I.C.J. Reports 2002, p. 353,
para. 68 or p. 415, para. 223; Sovereignty over Pulau Ligitan and Pulau Sipadan (Indonesia/Malaysia), Judgment,
I.C.J. Reports 2002, p. 678, para. 126; or Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 120,
para. 47. See also MBF, pp. 59-61, paras. 2.13-2.19.
168Frontier Dispute (Benin/Niger), Judgment, I.C.J.Reports2005, p.143, para.128; see also p.149,
para. 141.
169See paras. 1.3 and 1.11 above.
170
Cf. Territorial Dispute (Libyan Arab Jamahiri ya/Chad), Judgment, I.C.J.Reports1994, pp. 39-40,
paras. 75-76.
171
Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial
Guinea intervening), Judgment, I.C.J. Reports 2002, p. 372, para. 118.
172
“26. Thence the boundary runs through Mount Mulikia... 27. Thence from the top of Mount Mulikia
to the source of the Tsikakiri, leaving Kotcha to Britain and Dumo to France and following a line marked by four
provisional landmarks erected in September 1920 by Messrs. Vereker and Pition.” - 33 -
Kotcha has expanded to either side of the lin e into Cameroonian territory, the Court recalls
that “it has no power to modify a delimited boundary line, even in a case where a village
previously situated on one side of the boundary h as spread beyond it. It is instead up to the
Parties to find a solution to any resultant probl ems, with a view to respecting the rights and
173
interests of the local population” .
1.50. These principles, which Niger’s Memorial completely disregards, should find full
application in the present case.
2. “. . . the course shall be that shown on the [IGN map], and/or any other
relevant document accepted by joint agreement of the Parties”
1.51. Article2 of the Agreement of 28March 1987 does not confine itself to
recognizing the pre-eminence of the course under the frontier title constituted by the
1927 Arrêté and its Erratum; should these acts not suffice, it also limits the other documents
which may be used to establish the course of the frontier to, firstly, “the 1:200,000-scale map
of the Institut Géographique National de France , 1960edition” and/or, if necessary, “any
other relevant document accepted by joint agre ement of the Parties”. Notwithstanding this
perfectly unambiguous provision, and despite recognizing that these two criteria are
subsidiary 174, and that no other document has been accepted by joint agreement of the
Parties 175, Niger, when it believes that it could benefit, does not hesitate to:
⎯ give precedence to the line shown on the IGN map over the text of the Erratum; and to
⎯ dismiss both that text and the map in favour of a random line supposedly established by
various instruments and by both colonial and post-colonial effectivités on which it relies.
1.52. As indicated above, Niger, confusi ng concision with impr ecision, wrongly takes
as its starting point that, even though it is pre-eminent, the Erratum is imprecise, “particularly
rudimentary” 176and excessively succinct 177. And it claims that this is proven for the two
178
sectors of the disputed line: both the Téra and the Say sectors . It therefore believes that it
is not practicable: “It follows from the summary and imprecise nature of the description of
the boundary in several179ctors that the practical scope of the Arrêté and its Erratum remains
extremely limited.”
1.53. Niger claims that this rudimentary natu re is justification for disregarding both the
Arrêté and the 1960 map and for “look[ing] elsewhere in order to identify” the frontier 18. To
this end it suggests having recourse to “administrative documents of the colonial era”, so as to
173Ibid., p. 374, para. 123.
174
MN, p. 75, para. 5.13.
175
Burkina also agrees, except that it does not regard the agreement on the frontier problems constituted by
the exchange of letters of 29 October and 2 November 2009 as such a document (MN, p. 77, para. 5.16).
176
MN, p. 65, para. 5.6; see also p. 84, para. 6.10.
177MN, p. 83, para. 6.9, or p. 104, para. 7.11.
178See paras. 1.42-1.43 above.
179MN, p. 66, para. 5.6.
180
MN, p. 84, para. 6.11. - 34 -
“supplement the summary description in the Arrêté and Erratum of 1927” 181 and relying on
effectivités, without too much concern as to the cont radictions between them and the text of
the Erratum 182. This goes well beyond what is envisaged by the 1987 Agreement and what is
permitted by the generally recognized principles fo r interpretation: to interpret is not to
supplement ⎯ above all when the Parties have explic itly agreed on an exhaustive list of the
instruments to which it is possible to have recourse, should the title not suffice.
1.54. This process of neutralizing the Arrêté and its Erratum is at work all along the
line which is the subject of the dispute submitted to the Court. In the “Téra sector”, the text is
described as “particularly succinct” 183 and Niger sees obscurity in its simplicity. It is a
question of looking “elsewhere in order to identify this stretch of the boundary between the
two territories” and it is to “[t]he history of its origins” that Niger turns 184. It believes that the
preparatory work provides indications (about the cantons) that can supplement the wording of
the official text 185. Furthermore, Niger does not hesita te to rely on documents, such as the
186
reports from 187bos and Prudon, which were not taken in to account in the preparation of
the Arrêté but which it believes nevertheless “both have the merit of showing that the
boundary was a sinuous one” 188.
1.55. In the second sector, the Say sector, Niger openly goes on the offensive with
regard to the text: it claims to “show that there are well-established reasons for not following
it in certain respects” 189. It believes that these reasons can be found in colonial practice,
which runs counter to the wording of the Erratum, in particular where Bossébangou is
concerned. It believes that both the cartographic material of the period 190 and the attitude of
191
the colonial authorities both before and after the adoption of the Arrêté demonstrate that,
contrary to the text of the latter, the inter-c olonial boundary did not reach the River Sirba at
Bossébango192ut stopped at “a point close to the hamlet of Nabambori, not far from
Alfassi” . Niger therefore believes that the modified Arrêté is wrong and that “[t]here was
18MN, p. 86, para. 6.11.
18These are based on the complaints of officials (see in general MN, pp.25-34, and especially the letters
from the Commander of Dori cercle, p. 26, para. 2.3). Reference is made in particular to the Delbos/Prudon line,
which, according to Niger, enabled the colonial authorities to deal with the lack of precision that it attributes to the
Erratum (see especially MN, pp.28-30, para.2.4), and Niger highlights the Records of Agreement concluded
between officials to clarify their bounda ries, in particular the Roser/Boye r Agreement of 1932 and the Ossolo
Agreement of 1931, although Niger itself remarks that they were not given the necessary approval (MN, pp. 33-34,
para. 2.8).
18MN, p. 83, para. 6.9; see para. 1.43 above.
184
MN, p. 84, para. 6.11.
185
MN, p. 86, para. 6.11 and p. 90, para. 6.15.
186
Which “has not been found” (MN, p. 88, para. 6.12); see also paras. 1.22-1.23 above.
18MN, p. 88, para. 6.12 and para. 1.22 above.
18MN, p. 87, para. 6.12.
18MN, p. 105, para. 7.12 (emphasis added).
190
MN, pp. 107-108, paras. 7.17-7.18.
191
MN, pp. 108-110, paras. 7.19-7.20.
192
MN, p. 110, para. 7.20. - 35 -
no justification for continuing the inte r-colonial boundary to the village of
193
Bossébangou” ⎯ to its great advantage.
1.56. In the “sector of the four villages”, Niger affects to return to an approach of
interpreting the text 194, but it ignores its terms (the word “salient” in particular 195) and, where
it believes that the Erratum does not suffice, it has recourse, not to the 1960 IGN map, as the196
1987Agreement would have it, but to cartographic material of the period and to a
telegram/letter from the Head of Say Subdivision to Dori cercle 197. The same type of
sources ⎯ colonial cartographic material, a preparatory document and an isolated tour
report ⎯ lead Niger to say that the Erratum should not “be read too literally” 198and, in
reality, to “interpret” it in a way that is completely incompatible with that text as regards the
point marking the end of the salient 199. It goes without saying that, once again, this
“interpretation”, if it can be termed such, is very favourable to Niger.
1.57. Primacy is again given to certain aspects of alleged colonial practice over the
Erratum, as interpreted in this very lax manner, in the portion “end of the salient/start of the
Botou Loop”. Believing that “[t]he final section of boundary in200e Say sector is described in
the 1927Erratum in a particularly lapidary manner” , Niger asserts that “the straight-line
boundary which it [the Erratum] establishes appear s to have no basis in the situation prior to
201
the adoption of the Erratum and was never confirmed in the subsequent practice” .
1.58. This mode of reasoning does not have any foundation in the 1987Agreement,
under the terms of which, in th e absence of an agreement between the Parties on any other
202
document whatsoever , only the 1960 IGN map may be consulted, should the amended
Arrêté not suffice.
1.59. However, when Niger alleges that it does not suffice, it is not necessarily the map
that it turns to. Apart from the examples give n above, the argument deployed for the sector
“end of the salient/start of the Botou Loop” illust rates the capricious use that Niger makes of
the IGN map 203.
1.60. Niger’s line of argument regarding th e Tao-Bangaré section also establishes that
the map is only used when it matches Niger’s carefully selected examples of colonial practice.
19MN, p.105, SectionA. According to the Erra tum, the boundary reaches “the River Sirba at
Bossebangou”.
19MN, p. 112, para. 7.27.
19See para. 0.15 above.
196
MN, p. 114, para. 7.30.
197
MN, p. 114, para. 7.31 (see MN, Anns., Series C, No. 61).
19MN, p. 115, para. 7.32.
19MN, pp. 115-116, para. 7.33.
20MN, p. 116, para. 7.34.
201
MN, p. 116, para. 7.35; see also pp. 118-120, para. 7. 39, where Niger claims to find confirmation of its
contra textum “interpretation” in colonial and post-colonial practice.
202
See para. 1.51 above.
203
See the remarks made in this regard in the introduction to the present Counter-Memorial, para. 0.13. - 36 -
Having disregarded the Arrêté on the pretext that it gives “no further indication until the point
where the inter-colonial boundary rejoins the boundary of Say cercle”, Niger considers that it
is “reasonable to rely for this section, subject to any justified exception , on the 1960IGN
line” 204. However, once again the map is only used to the extent that it respects the supposed
lessons from colonial practice as analysed by Niger. This is particularly flagrant in the case
of Petelkolé: “the IGN line passes to the west of Petelkolé . . . which it leaves to Niger. This
205
is in accordance with the a206nistrative information from the colonial period” ⎯ which is
in fact highly debatable .
1.61. The same applies in the salient sector: the Erratum having been declared
problematic, the IGNmap, whose immense virtues Niger nonetheless extols elsewhere 207, is
dismissed in favour of cartographic material that is both older and more limited: a
sketch-map of Say cercle prepared in1915 by AdministratorTruchard, and the map entitled
“new frontier between Upper Volta and Niger” allegedly “prepared following the adoption of
the 1927 Arrêté and its Erratum”. Confirming Niger’s interpretation of the Erratum, this
map ⎯ which it is worthwhile recalling was “proposed by Niger, was not accepted by
Burkina and thus was not retained as a ‘relevant document accepted by joint Agreement of the
Parties’” 208 ⎯ is preferred to the IGNmap, which “makes the frontier in this area run
significantly further to the east than that shown on the previous maps. Here again, this line
does not correspond to the traditional shape of Say cercle, as it was consistently represented
209
during the colonial period” .
1.62. Moreover, according to Niger, this old cartographic material takes precedence
over the IGNmap even if the latter confirms the Erratum. Since, therefore, in its view, the
Erratum is wrong to make the boundary descend so that it reaches the River Sirba at
Bossébangou, the IGNmap, which also makes the line descend to that point, should be
disqualified because it reproduces that alleged mistake 210.
1.63. Thus Niger argues that the course shown on the 1960 map takes second place to
211 212
maps or (alleged) colonial and post-colonial effectivités which do not confirm it,
contrary to the provisions of Article 2 of the Agreement of 28 March 1987.
1.64. Nevertheless, it should be noted that colonial practice only takes precedence to
the extent that it suits Niger. The case of Bangaré is a striking example of this: in the section
213
from Bangaré to the boundary of Say cercle , Niger intends unreservedly to follow the
IGNline, which gives it the village of Banga ré, despite mentioning at the same time a
204MN, p. 93, para. 6.21 (emphasis added).
205MN, p.94, para.6.22. Niger also adds that “[ Petelkolé] has remained under Niger authority since
independence” (ibid.).
206See Chap. III, Section 2, 2 A, below.
207MN, p. 75, para. 5.14.
208
MN, p. 76, para. 5.15.
209
MN, p. 114, para. 7.30.
210MN, p. 110, para. 7.21.
211MN, pp. 98-99, para. 6.25.
212MN, pp. 93-97, paras. 6.22 and 6.23.
213
MN, pp. 97-99, paras. 6.24-6.25. - 37 -
colonial practice which to a significant extent fails to place Bangaré in Niger; this is the case
in particular with the supposed “Delbos/Pr udon Agreement”, on which Niger nevertheless
relies so heavily elsewhere 214, which places Bangaré in Upper Volta 215.
1.65. The only consistent aspect of Niger’s Memorial is its inconsistency: it does not
follow any clear method; its thesis is not based on any particular principle; it “picks” from
the vaguely possible arguments depending on the solution that is most advantageous to it,
without any regard for the preliminary agre ement between the Parties on the applicable
sources of law (despite this being one of the striking features of the present case ⎯ the other
being the existence of a clear and indisputable title, which is moreover given precedence by
the 1987 Agreement between the Parties), a point which will be developed in Chapter II of the
present Counter-Memorial. ChaptersIII andIV will then highlight in greater detail the
inconsistencies in Niger’s line of argument, firstly concerning the “Téra sector” and secondly
concerning the “Say sector”.
214See paras. 1.22-1.26 above.
215MN, p. 97, para. 6.24. C HAPTER II
T HE DISCUSSIONS ON THE DEMARCATION OF THE FRONTIER
AND THE INCONSISTENCY OF N IGER ’S CLAIMS
2.1. Niger’s Memorial offers a version of the discussions on the demarcation of the
frontier that is both inaccurate and incomplete , and which fails to explain either how the
dispute between the Parties regarding their common frontier came about or where they
disagree over its course. It will therefore be n ecessary to correct the most obvious errors in
the account that is reproduced in the ten page216f Chapter III of the Memorial that are devoted
to the “attempts to settle the frontier dispute” and to complete it (Section1). There will
then be an opportunity to highlight something on which Niger remains silent, namely the total
inconsistency of its successive claims (Section 2).
SECTION 1
N IGER ’S ONE -SIDED AND PARTIAL PRESENTATION OF THE FACTS
2.2. The account produced by Niger claims to present “[t]he attempts to settle the
frontier dispute peacefully”. However, this title is completely biased, since the discussions
between the Parties about demarcating the frontie r were perfectly consensual until the end of
the 1980s; no dispute existed between the Parties at that time (1). The dispute only arose
from the moment that Niger decided to invent new ways of interpreting the Erratum (2).
1. The consensual work
2.3. There was no dispute between Burkina and Niger between 1964 and 1990. During
this period, as Burkina duly reports in its Memorial 217, the Parties on the contrary worked
together on good terms with a view simply to demarcating their common frontier, which they
agreed without difficulty to recognize as being fixed by the Erratum. Given that Niger’s
account of this period suggests the opposite, it therefore needs to be c
orrected, in particular
with regard to the following four points.
2.4. Firstly, the Protocol of Agreemen t of 23June 1964 is not the reflection of a
218
dispute, contrary to what Niger suggests ; rather, it defines the Parties’ perfect common
understanding of the “basic documents for the determination of the frontier”. The wording of
the Protocol is clear in this regard:
“By agreement between the Parties it was decided to take as basic
documents for the determination of the frontier Arrêté général 2336 of
31August1927, as clarified by Erratu m 2602 APA of 5 October 1927, and the
1:200,000-scale map of the Paris Institut Géographique National.
A Joint Commission of not more than ten members, which shall include
the heads of the administrative divisions concerned, shall carry out the work of
216
MN, p. 39.
217
MBF, pp. 34-48, paras. 1.38-1.75.
21MN, p. 39, para. 3.1. - 39 -
demarcation, starting in mid-November 1964 and beginning with the disputed
219
sectors, in particular the stretch of the frontier lying between Téra and Dori.”
2.5. Admittedly, the work of the Join t Commission that was provided for by this
Agreement was not accomplished immediately, as Niger indicates, but it is incorrect to say
220
that the initiatives to proceed with its work resumed “only some 20years later” . This is
forgetting that:
⎯ the Topographic Service and Land Registry of Niger contacted the IGN Annex in Dakar
as early as 20July 1964 in order to obtain a 1:1,000,000 map showing the main
221
astronomic points along the frontier ;
⎯ on 25July 1964, the equivalent service in Upper Volta asked the same IGN Annex in
Dakar how much it would cost to mark out the frontier by placing a marker
222
approximately every 10 kms ;
⎯ on 6 March 1967, the President of Niger suggested to his counterpart in Upper Volta that
223
an effort should be made to implement the Protocol of Agreement ;
⎯ on 16March1967, the Minister for the Interior and Security of Upper Volta asked the
Commanders of the cercles bordering Niger to send him all the relevant documents and
information, with a view to a meet224 of the Joint Commission on Demarcation of the
Frontier the following month ;
⎯ a ministerial meeting took place on 9 and 10 Ja nuary 1968 in Niamey, which decided to
“entrust [the Institut Géographique National de Paris ] with the task of demarcating the
225
frontier” , clearly demonstrating that the frontier issue was of a purely technical nature
and did not follow from any “dispute”;
⎯ the principle of establishing a joint commission to demarcate the frontier was reiterated
on 16 September 1982 by a ministerial meeting 226.
227
2.6. Secondly, while it is true, as Niger indicates , that in February 1985 the ministers
recommended demarcating the frontier on the basis of the Protocol of Agreement of
23June1964 and establishing the “joint commi ssion responsible for the demarcation of the
228
frontier between the two States” , the Agreement and Protocol of Agreement of
28March1987 do not seek to “provide the [dem arcation] works with a current framework”
229
on the basis of the Protocol of Agreement of 23June 1964 . The Agreement marks the
219Ann. MBF 45; MN, Anns., Series A, No. 1.
220MN, p. 39, para. 3.1.
221
Ann. MBF 4; see MBF, p. 37, para. 1.47.
222
Ann. MBF 47; see MBF, p. 37, para. 1.47.
223
Ann. MBF 49; see MBF, pp. 37-38, para. 1.48.
224Ann. MBF 50; see MBF, p. 38, para. 1.48.
225Ann. MBF 54; see MBF, pp. 38-39, para. 1.51.
226Ann. MBF 69; see MBF, p. 40, para. 1.55.
227
MN, p. 40, para. 3.3; see also MBF, p. 40, para. 1.56.
228
Ann. MBF 63; MN, Anns., Series A, No. 2.
229
MN, p. 40, para. 3.3. - 40 -
definitive agreement between the Parties both on the texts determining the frontier and on the
method to be followed for its demarcation ⎯ an exclusive and more precise method than that
suggested by the 1964Protocol of Agreement ⎯ while the Protocol establishes the Joint
Technical Commission on Demarcation and tasks it with carrying out that demarcation 230.
2.7. Thirdly, the discussions within the Joint Technical Commission on Demarcation
were not “negotiation[s] . . . between the two States over the course of the common
frontier” 231. The Commission could not be the place for such negotiations since, in
accordance with Article5 of the 1987Protocol of Agreement establishing it, it only enjoyed
full autonomy in “executing the works of demarcation” 232. Up until 1990, the members of the
Commission therefore confined themselves to pursuing its demarcation, given that the
delimitation had already been carried out.
2.8. Fourthly, it is incorrect to sugg est, as Niger does, that the work done
by the Commission between1988 and1990 amounted to placing “23markers out of the45
233
envisaged” . The Commission did a great deal more than this, as it managed to carry out a
full survey of the frontier line on the ground 234.
2.9. It concluded this work at the meeting of 26, 27 and 28 September 1988 in Niamey,
the purpose of which was to plot on the 1960map “the line resulting from the field survey
conducted by the Technical Sub-Committee and to submit the results of said work to both
Governments for final selection of the frontier line” 235. This was a complete success. The
report of the meeting states that:
“The experts are... unanimous as to the map interpretation and the field
survey of the boundary line defined in the basic documents cited in the
Agreement and Protocol of Agreement, signed in Ouagadougou on
28March1987. Only the Tokébangou point, which could not be identified
despite numerous investigations, was subj ect to interpretation by the Technical
Sub-Committee.” 236
2.10. The resulting “consensual line” illustr ates the p237ect common understanding of
the members of the Commission in September1988 . Moreover, on this basis, it only
remained for the Commission actually to mark out the frontier. On 18 May 1989, with this in
mind, it was also able to draw up a list of the co-ordinates of 32 defining points in preparation
for demarcation 238, and subsequently to place 23 markers 239.
230MBF, pp. 43-44, paras. 1.61-1.65 and pp. 62-65, paras. 2.21-2.28.
231MN, p. 40, para. 3.3.
232
Anns. MBF 72 and 73 and MN, Anns., Series A, No. 4.
233
MN, p. 41, para. 3.4.
234MBF, pp. 44-46, paras. 1.66-1.69.
235Ann. MBF 81.
236Ibid.
237
MBF, p.46, para.1.69; the consensual line is shown in Cartographic Annex MBF15; it is also
reproduced on page 164 of Burkina’s Memorial, sketch-map No. 16.
238
Ann. MBF 83; see MBF, p. 47, para. 1.70. - 41 -
2. The crystallization of the dispute
2.11. The Parties began to disagree over the course of the frontier from the moment
when, in February 1990, Niger decided un ilaterally to renounce the consensual line
of 1988 240. From that date, Niger sought to assert new readings of the amended Arrêté
of 1927, while Burkina remained firmly convinced that the consensual line was the only valid
line to result from a determination of the fr ontier pursuant to the 1987Agreement. The
dispute was born at that moment.
2.12. In this context, the decision made at the ministerial meeting held in May 1991 in
241
Ouagadougou appears to be the first of the “attempts to settle the frontier dispute
peacefully at diplomatic level” that ChapterIII, Section2, of Niger’s Memorial purports to
describe 242, although it is not specifically mentioned as such. Incidentally, during that
meeting of May1991, Ministers did not “[find] that there were lacunae in relation to the
implementation of the Arrêté... and its Erratum”, contrary to what Niger claims 243; they
merely took note of the deadlock within th e Joint Technical Commi ssion on Demarcation,
which had turned to them in accordance with Article5 of the 1987Protocol of Agreement,
and decided to settle the matter by applying th e Erratum to the letter for the section of the
frontier that reaches the River Sirba at Bossébangou, deciding by way o244ompromise to have
recourse to the line shown on the 1960 IGN map for the remainder .
2.13. Furthermore, while it is true, as it recognizes, that it was Niger that rejected the
compromise solution of May1991 245, its reason for doing so cannot be that “the solution
proposed did not comply with the conditions la id down by Articles 1 and 2 of the Agreement
of 28 March 1987” 246. This is clearly just a pretext, for the simple reason that the sole
purpose of the 1991 decision was to establish a compromise solution which acknowledged,
precisely, that in the light of Niger’s position it would be impossible to achieve a
straightforward application of the Erratum, in accordance with the 1987 Agreement. In fact,
in 1991 Niger and Burkina were able, through the exercise of their sovereign rights, to adopt a
different line from the one arising from Articles 1 and 2 of the 1987 Agreement, but only by
agreeing at the same time to modify their common frontier by agreement ⎯ an agreement
which was, in turn, rejected by Niger.
S ECTION 2
T HE INCONSISTENCY OF N IGER ’S CLAIMS
2.14. There are obvious variations in the positions adopted by Niger on the frontier line
that it has claimed over the years, illustrating th e inconsistency of the new thesis presented in
its Memorial, which now seeks to rely on boundaries that are either alleged to be
239
Ann. MBF 87; see MBF, pp. 47-48, paras. 1.71-1.73.
24Ann. MBF 88; see MBF, p. 48, para. 1.74.
24Ann. MBF 49 and MN, Anns., Series A, No. 6.
242
MN, p. 44.
243
MN, p. 41, para. 3.5.
24MBF, pp. 50-51, para. 1.80.
24MN, p. 42, para. 3.6.
24Ibid. - 42 -
long-standing 247or correspond to what it contends was former practice 248, new arguments that
it clearly “discovered” when it was preparing its Memorial.
2.15. With regard to the section of the line from the Tong-Tong astronomic marker to
the River Sirba at Bossébangou, passing thr ough the Tao astronomic marker, Niger has
changed its mind no fewer than five times, maintaining:
⎯ in September 1988, that the frontier is formed by two straight-line sections connecting
these three points (this results from the consensual line adopted by the Niger and Burkina
experts in 1988) 249;
⎯ in July 1990, that the section connecting the Tong-Tong astronomic marker to the River
250
Sirba at Bossébangou describes a curve ;
251
⎯ in May 1991, that the frontier at this level consists of two straight-line sections ;
⎯ in July 2001, that it consists of a curved line 252.
2.16. In April 2011, in its Memorial, Nige r has completely changed its position and is 253
now maintaining that its claim of a curved line, “[a]t all events”, “[is] debatable” , and that
the line is formed by a miscellaneous combinati on of sections, some of which are alleged to
correspond to the indications in the Erratum, ot hers to the 1960IGNmap, and yet others to
the former boundaries of cantons and other administrative subdivisions, the reality of which
is, moreover, far from being established.
2.17. The same goes for the section that m eets the River Sirba at Bossébangou. Niger
has maintained:
⎯ in September 1988, that the frontier reaches the River Sirba at Bossébangou (this results
254
from the consensual line adopted by the Niger and Burkina experts in 1988) ;
⎯ in May 1990, that the frontier does not r each the Sirba at Bossébangou because the map
“French West Africa, new frontier between Upper Volta and Niger based on the Erratum
of 5October1927 to the Arrêté of 31August1927” suggests that it stops further to the
255
north ;
256
⎯ in July 1990, that the frontier reaches the River Sirba at Bossébangou ;
24See Chap. IV below.
248
See Chap. III below.
249
MBF, pp. 46-48, paras. 1.69-1.73.
250
Ann. MBF 87; see MBF, p. 49, para. 1.77 and MN, pp. 68-69, para. 5.8.
25Ann. MBF 89; see MBF, pp. 50-51, paras. 1.79-1.81.
25Ann. MBF 94; see MBF, p. 52, para. 1.84.
25MN, p. 70, para. 5.9.
254
MBF, pp. 46-48, paras. 1.69-1.73.
255
Ann. MBF 85; see MBF, pp. 48-49, paras. 1.75-1.76 and p. 137, paras. 4.93-4.94.
256
Ann. MBF 87; see MBF, pp. 138-139, paras. 4.97-4.98. - 43 -
⎯ in May 1991, that the frontier reaches the River Sirba at Bossébangou 257;
⎯ in 1994, that the 1991 line is not entirely consis tent with the terms of Articles 1 and 2 of
258
the1987 Protocol of Agreement ⎯ which suggests that the frontier still reaches the
River Sirba at Bossébangou because that Protocol refers to the Erratum, which expressly
stipulates that this is so.
2.18. In its Memorial of April2011, Nige r returns to its position of May1990 and,
relying on arguments that are as novel as they are incompatible with the provisions of
the1987 Agreement, is now claiming that “[t] here was no justification for continuing the
259
inter-colonial boundary to the village of Bossébangou” .
2.19. In the area of the salient, Niger has claimed:
⎯ in September 1988, a line consistent in every respect with the line shown on the
1960 IGN map, as indicated on the consensual line 260;
⎯ in May 1990, a line not composed of a “true” salient, invoking the map “French West
Africa, new frontier between Upper Volta and Niger based on the Erratum of
261
5 October 1927 to the Arrêté of 31 August 1927” ;
⎯ in July 1990, a line forming a salient, but beginning immediately as it leaves
Bossébangou ⎯ contrary to the description given by the Erratum ⎯ “turn[ing] back on
itself at an angle o262ome kind, but only on the condition that it does not cut the loop of
the River Sirba” ; Niger also asserts that the village of Takalan “corresponds to the
exact location of the current village of Takatami” 263;
⎯ in May 1991, a line consistent in every respect with the line shown on the
264
1960 IGN map ; and,
⎯ in July 2001, Niger recognizes that it is impo ssible to identify the villages referred to in
265
the Erratum .
2.20. In its Memorial of April 2011, Niger returns to its position of May1990 and
claims, amongst other things, firstly that there is no salient 266 and, secondly, that Takalan is
267
located to the east of Takatami and is very close to the site of Tangangari .
257
Ann. MBF 89; see MBF, pp. 50-51, paras. 1.79-1.81.
258
Ann. MBF 91; see MBF, p. 51, para. 1.81.
25MN, pp. 105-111, paras. 7.14-7.24.
26MBF, pp. 46-48, paras. 1.69-1.73.
26Ann. MBF 85; see MBF, pp. 48-49, paras. 1.75-1.76 and p. 137, paras. 4.93-4.94.
262
Ann. MBF 87; see MBF, p. 139, paras. 4.97-4.98.
263
Ann. MBF 87; see MBF, p. 144, para. 4.115.
264
Ann. MBF 89; see MBF, pp. 50-51, paras. 1.79-1.81.
26The report of the fourth ordinary session of the Jo int Technical Commission on Demarcation, held in
Ouagadougou from 18 to [21] July 2001, notes “the failure to identify the villages referred to in the Erratum” and
recommends a further survey mission in the field, which never took place; Ann. MBF 94. - 44 -
2.21. Finally, as regards the line between th e intersection of the River Sirba with the
Say parallel and the beginning of the Botou bend, Niger has successively declared itself
convinced:
⎯ in Sept268er 1988, that it is formed by a st raight line, as indicated on the consensual
line ;
⎯ in May 1990, that the frontier follows a strai ght line at this point, as this is what is
indicated on the map “French West Africa, new frontier between Upper Volta and Niger
based on the Erratum of 5October1927 to the Arrêté of 31August1927”, which Niger
269
invoked at that time to reject the consensual line ;
270
⎯ in May 1991, that it follows the line of the 1960 IGN map .
2.22. In April 2011, in its Memorial, Niger argues for the first time that the line consists
of a combination of two straight lines forming an angle pointing towards the south.
2.23. This inconsistency on the part of Niger cannot be justified on the pretext that “the
proposals put forward by the experts... were merely provisional positions, in view or hope
of reaching a negotiated settlement of the dispute which the parties have committed
themselves to seeking” 271. The reality is quite different: the experts have not negotiated
anything and have kept to the extremely precise “road map” that had been fixed for them by
the 1987Protocol of Agreement. Burkina has taken due note and has consistently stood by
the consensual line of 1988. It is Niger, and Niger alone, that has persisted since 1990 ⎯ and
continues to persist in its Memorial ⎯ in claiming frontier lines that have no basis in the law
which, as the Parties have agreed many times, is the only law applicable. This conduct
further re272orces the position of Burkina, which, as Niger’s Memorial reports, this time
correctly , having had its confidence shaken by Niger’s successive about-turns, insisted
in 2006 that the Parties bring the resulting dispute before the International Court of Justice 273.
266MN, p. 112, para. 7.26.
267MN, p. 115, para. 7.31.
268MBF, pp. 46-48, paras. 1.69-1.73.
269
MBF, pp. 48-49, paras. 1.75-1.76 and p. 137, paras. 4.93-4.94.
270
Ann. MBF 89; see MBF, pp. 50-51, paras. 1.79-1.81.
271MN, p. 44, para. 3.11.
272MN, p. 44, para. 3.13.
273
MN, Anns., Series A, Nos. 9-11. C HAPTER III
T HE COURSE OF THE FRONTIER IN THE “T ÉRA SECTOR ” 274
3.1. On reading the chapter in Niger’s Memo rial that is devoted to what Niger terms
“[t]he first section of boundary concerned by the present dispute” 275, it is apparent that in this
sector some of the aspec276of the dispute between the two Parties have declined in importance
since the negotiations , while others have come to the fore.
3.2. In its Memorial, Burkina explained why there is no doubt that the corrected Arrêté
of 1927 retained a frontier in this sector cons isting of two straight-line sections connecting
three frontier points in turn. Indeed, the letter of the Erratum, according to which
“[from] the Tong-Tong astronomic marker[, ] this line then turns towards the
south-east, cutting the Téra-Dori motor road at the Tao astronomic marker
located to the west of the Ossolo Pool, and reaching the River Sirba at
Bossebangou”
leaves no doubt about the fact that this delimitation document
⎯ designates three frontier points (the Tong-Tong astronomic marker, the Tao astronomic
marker and the point where the inter-colonial boundary “reach[es] the River Sirba at
Bossebangou”),
⎯ and that they are connected by two successive straight lines, in accordance with generally
observed colonial and boundary practice and w ith the consistent interpretation of the
Erratum by the colonial authorities 277;
⎯ as confirmed, incidentally, by the documents annexed by Niger to its own Memorial,
which in turn establish that, in the eyes of the colonial authorities: the boundaries fixed
by the Erratum in this sector were “estab lished on the basis of the map prepared by
CaptainCoquibus, which only showed theoretical lines and points” 27; the boundary
followed “a notional straight line starting 279m the Tong-Tong astronomic marker and
running to the Tao astronomic marker” ; instead of “the lines” (in the plural and also,
incidentally, “theoretical”) proposed by Administrator Delbos, the Erratum opted for “the
274
The expression “Téra sector” is used in the title of Chapter VI of Niger’s Memorial. Burkina believes
that it is inappropriate, for the reasons set out below (s ee paras.3.14-3.17 below). Th is expression is only used
here for practical purposes, to show that the present ch apter will respond to the factual and legal arguments put
forward in the chapter of Niger’s Memorial that bears this title.
27MN, p. 79, para. 6.1.
276
See Chap. II above. Obviously Burkina prepared its own Memorial on the basis of the positions adopted
by Niger during the negotiations.
277
See MBF, pp.101-132, paras.4.6-4.81 and p.132, pa ra.4.82 for the exact co-o rdinates of these three
frontier points. MN, Anns., Series C, No. 62 (p. 7) (a le tter from the Governor of Niger from 1936) confirms that
during the relevant period the village of Bossébangou was located on the bank of the River Sirba (“People go to
collect water from the Sirba, which is less than 500 m from the village . . .”).
278
See para.3.31 below and MN, Anns., SeriesC, No .20 (letter from Admini strator Delbos, Commander
of Dori cercle, to the Governor of Upper Volta dated 17 December 1927).
279
See para.3.47 below and MN, Anns., SeriesC, No.56 (certified copy of Record of Agreement of
13 April 1935 between Administrator Garnier (Dori cercle) and Deputy Lichtenberger (Téra Subdivision)). - 46 -
Tao-Sirba line [in the singular]” 280; the boundary in this sector formed “a theoretical and
artificial frontier” 281, “by connecting the Tao boundary marker directly with
282
Bossébangou” .
3.3. Moreover, in its Memorial Niger conced es that in several respects this description
of the line is justified.
3.4. It acknowledges firstly that the Tong-Tong and Tao astronomic markers are
283
frontier points. Niger gives the same co-o rdinates for the first marker as Burkina .
However, the Parties do not accept the same co -ordinates for the second marker, although the
284
difference is minimal . Both Parties take as a basi s the co-ordinates established by
Captain Nevière in 1927, but the co-ordinates communicated in Annex No. 105, Series C, of
Niger’s Memorial do not correspond to those that appear in the data sheet on the astronomic
markers which was drawn up by that mission and which Burkina has appended to its
Memorial as Annex41. Given that it is more precise, the data sheet should, in Burkina’s
view, take precedence. Moreover, Niger seems to agree, as, after citing its own version of the
co-ordinates from the Nevière mission, it explains that “[h]owever the frontier marker is
situated slightly further south and east, at the following co-ordinates: 14°03'02"N,
285
00°22'52"E. It is this latter point which should be taken as a frontier point” . These
co-ordinates ⎯ whose method of identification Niger does not specify 286 ⎯ only differ by a
few seconds from the more precise co-ord inates measured by GPS by Burkina:
14º 03' 04.7" N; 0º 22' 51.8" E. Burkina stands by these, given their greater precision and the
reliability of the methodology used to obtain them.
3.5. Niger also acknowledges in its Memorial, but this time more awkwardly, that there
is a third frontier point in the sector defined in the relevant part of the Erratum. Niger in fact
refers to “two or three points designated by the Arrêté of 31 August 1927 as amended by the
287
Erratum of 5October1927” . The use of the conjunction “o r” reflects a certain hesitation
that is removed by the text of the Erratum, which refers indisputably to the frontier passing
through three successive points in the present sector.
3.6. Moreover, breaking with the interpreta tion that it believed it could attribute to the
288
Erratum in 1990 ⎯ which was itself at odds with the consensual interpretation backed by
the two Parties in 1988, the one that Burkina still defends today 289 ⎯ Niger no longer
believes that these frontier points are connected by one or more curved lines, an interpretation
280See para.3.55 below and MN, Anns., SeriesC, No.79 (report of a tour conducted from 16 to
23 November 1953 by Deputy-Administrator Lacroix (Tillabéry cercle), dated 24 December 1953).
281See para. 3.60 below and MN, Anns., Se ries C, No. 30 (letter from the Lieutenant-Governor of Niger to
the Lieutenant-Governor of Upper Volta dated 27 September 1929).
282See para. 1.28, 8 indent, above and MN, Anns., Series C, No. 73 (official telegram/letter from the Head
of Téra Subdivision to Tillabéry cercle dated 11 July 1951).
283MN, p. 92, para. 6.19; MBF, pp. 102-103, paras. 4.11-4.13.
284
MN, pp. 93-94, para. 6.22; MBF, pp. 103-104, paras. 4.14-4.16.
285
Ibid.
286And to which it adds three extra [tenths of] seconds in its submissions (see MN, Submissions, p. 122).
287MN, p. 86, para. 6.11.
288MBF, p. 49, para. 1.77. See also MBF, p. 52, para. 1.84.
289
MBF, pp. 118-123, paras. 4.46-4.57. - 47 -
that was clearly indefensible. “At all events”, it writes in its Memorial, rather casually, the
validity of the curves thesis was “debatable” 290. Admittedly, references to the line being
291
curved still surface here and there in Niger’s Memorial . The thesis defended in that
document is, however, radically new, compar ed with both the curves thesis put forward
in 1990 and the consensual line of 1988, which had established that the line was composed of
two straight-line sections. In Niger’s view, the line either follows the one shown on the
1960map or, in the other cases, consists of “straight lines” 292. This last point constitutes a
recognition that when the frontier passes through two points, save indications to the contrary,
it follows straight lines. This is also the thesis defended by Burkina. Niger has also endorsed
it once again in paragraph7.40 of its Memorial 293. This interpretation is all the more valid
given that, as Niger points out, “the sector of the frontier involved i294he present dispute”, of
which the Téra sector is just one part, is “relatively restricted” .
3.7. This is where the points of agreem ent between the two Parties end and the new
points of disagreement appear. They have one thing in common: Niger’s legally erroneous
conception, not to say its actual misrepresentation, of both the delimitation method that the
Parties have agreed to employ and the very text of the legal title that is applicable in the
present case, the 1927 Erratum 295. A simple glance at the line claimed in this sector by Niger
in its Memorial makes the misrepresentation very striking: instead of the two straight-line
sections connecting the Tong-Tong astronomic marker to the Tao astronomic marker and then
that marker to the River Sirba at Bossébangou, and instead of the line shown on the 1960 map
that Niger nevertheless claims “essentially” to fo llow, Niger calls for the following line to be
adopted, claiming that:
⎯ the two astronomic markers referred to in the Arrêté are not connected by one straight
line, but by two straight lines passing through an intermediate point, the Vibourié marker,
which is not, however, mentioned in the Erratum of 1927 (see sketch-map No. 3 below ⎯
Course of the boundary between the Tong Tong and Tao markers);
⎯ furthermore, the boundary does not then go on to connect the Tao astronomic marker to
the River Sirba at Bossébangou with a straight line; nor does it follow the line shown on
the 1960 map: according to Niger, here the boundary sporadically follows the line shown
on the map, but departs from it considerably several times so as broadly to enclave a
number of villages in Niger’s terr itory (see sketch-map No.4 below ⎯ Course of the
boundary after the Tao marker);
290MN, p. 70, para. 5.9.
291See, for example, MN, pp.91-92, para.6.18 in fine, asserting that the line shown on the 1960map
“adopts a shape broadly incurvated to the west”. Accordi ng to Niger, “[t]hat incurvation is new. We will now
examine step-by-step whether it is justified”. See also MN, pp.83-84, para.6.10, regarding the 1:1,000,000 map
of 1927 (MN, Anns., SeriesD, No.13): “[t]he shape of the line connecting these three points is slightly curved”
(this statement is debatable because the map actually reproduces two straight lines, even if the line, which is drawn
freehand, is rather approximate).
292MN, Chap. VI, B, Section 2, a), p. 91 and p. 93, para. 6.20 in fine.
293MN, p. 120, para. 7.40:
“Here again, nothing in the practice of the colonial authorities, or in the representations of
this part of the frontier on the maps and sketch-map s of the colonial period appears to justify this
deviation [the one found on the line shown on th e 1960map]. Niger accordingly maintains its
claim here to a frontier in two straight-line sect ions, as it appears on thos e maps and sketch-maps
of the colonial period.”
294MN, p. 49, para. 4.1.
295See Chap. I, Section 1, paras. 1.3et seq. - 48 -
⎯ finally, the boundary does not reach the River Sirba at Bossébangou, but touches on the
salient referred to in the Erratum (a salient that Niger would have disappear 296) more than
30km to the north-west of that point (see sketch-map No.5 below ⎯ Course of the
boundary in the area of the salient).
3.8. This line is completely unfounded. Firstly, as regards the method for determining
the frontier, and without repeating the points already made about this issue in the present
Counter-Memorial 29, it is important to note the inconsis tent use that Niger believes it can
make of the 1960 map.
3.9. On the one hand, Niger rightly chooses not to follow the line shown on the
1960 map when this has no basis in the text of the Erratum (in favour, it is true, and without
any justification, of a line which is also not described in that text). This is what it does in the
sector between the two astronomic markers of Tong-Tong and Tao, where it f298s that the line
shown on the map adopts “a shape broadly incurvated to the west” , which is not described
by the Erratum. Moreover, this presentation is very far removed from reality, as the line
shown on the map in this sector does not take the form of a single curved line connecting the
two markers, but follows a complex route connecting a whole series of intermediate points to
which the Erratum makes no reference whatsoever.
3.10. On the other hand, however, and without providing any explanation, Niger adopts
the opposite course of action in the subsequent sector of the frontier (see sketch-map No.4
below ⎯ Course of the boundary after the Tao marker), where it indicates that it prefers the
line shown on the 1960map to the one that results from the clear terms of the amended
Arrêté, even though, as in the previous case, th299ine shown on the map is extremely tortuous
and has no basis in the text of the Erratum . The double standard adopted by Niger is not
explained. Moreover, it is not quite true to say that in the second case Niger prefers the line
shown on the map to the one resulting from the amended Arrêté, since the line that it is
claiming in the Téra sector actually only follows the line shown on the 1960map very
sporadically, while departing radically from the line defined in the amended Arrêté of 1927.
29See paras. 4.40-4.53 below.
297
See paras. 1.40-1.64 above.
29MN, pp. 91-93, paras. 6.18-6.20.
29MN, pp. 96-101, paras. 6.21 et seq. - 52 -
3.11. The inconsistency is intensified by the fact that Niger refuses to follow the line
shown on the 1960map as far as the third frontier point, even though both the Erratum and
the line on the map fix its location in the same place: the point where the frontier reaches the
River Sirba at Bossébangou. Niger’s line avoids this crossing point, in contravention of both
the text of the Erratum and the line shown on the 1960 map, which coincide here.
3.12. Niger considers that it can depart s till further from the subsidiary method adopted
by the Parties by establishing a unilateral ex ception to reliance on the line shown on the
1960 map: it indicates that it relies on it “subject to any justified exception” 300. None of this
is in conformity with the method approved by the Parties 30.
3.13. As for the text of the amended Arrêté, Niger reorganizes it unilaterally without
even attempting to justify an approach that is, nevertheless, legally unacceptable. This
reorganization takes several forms.
302
3.14. As indicated above , the expression “Téra sector” may be convenient, but it is
merely a simplification that should be underst ood subject to the following three important
considerations.
3.15. Firstly, Niger itself points out that this sector is the one “where the frontier
separates the current département of Téra (Gorouol, Téra, Diagourou, and Dargol cantons),
on the Niger side, from the provinces of Oudalan, Seno (Dori) and Ya gha (Sebba), on the
303
Burkina Faso side” . It would therefore have been more appropriate to designate this sector
without referring exclusively to the Niger territorial subdivision. It is best described as the
Téra/Oudalan, Seno and Yagha sector, precisely because it is a frontier sector.
3.16. Secondly, in defining this sector with referen ce exclusively to a territorial
subdivision of Niger and, moreover, in making the delimitat304 in this sector end at what it
calls “the junction of the boundary with Say cercle” , Niger perpetuates the error committed
by the author of the Arrêté in its version of August 1927. That text did not seek to delimit
subdivisions that were internal to one colony, an y more than its effect today is to delimit a
territorial subdivision of Niger: it defines an inter-colonial boundary, which has become an
international frontier. The author of the first version of the Arrêté had admittedly committed
an error on this point, by including elements of intra-colonial delimitation ⎯ in this instance
certain cercle boundaries on either side of the inter-colonial boundary ⎯ in the inter-colonial
305
delimitation. However, that is precisely why the Erratum was adopted . In keeping with its
purpose, the latter no longer refers to cercle boundaries except, by reference , in its last
300MN, p. 93, para. 6.21.
301See paras. 1.40-1.64 above.
302
See the introduction to the present Counter-Memorial, para. 0.15, fn. 28, above.
303MN, p. 79, para. 6.1.
304Ibid.
305See MN, pp. 19-21, para. 1.26 and pp. 63-65, para. 5.5 and MBF, p. 137, para. 4.95. See also para. 1.17
above. - 53 -
subparagraph, regarding the final section of the frontier between Upper Volta and Niger 306.
By contrast, no reference is made in the Erratum’s other subparagraphs to “the boundary with
307
Say cercle”, as Niger wrongly contends .
3.17. Thirdly, in restricting the first disput ed sector to the “Téra sector”, Niger betrays
the text of the amended Arrêté of1927. Niger claims that this “Téra sector” “runs from the
308
Tong-Tong astronomic marker to the junction of the boundary with Say cercle” . In fact the
1927Erratum states something completely different, in the clearest possible terms. The
boundary adopted by that text does not pass through this junction, which it does not mention
at any point. On the contrary, it indicates that the boundary “cut[s] the Téra-Dori motor road
at the Tao astronomic marker located to the west of the Ossolo Pool, and reach[es] the River
309
Sirba at Bossebangou” . Moreover, it is this that Niger is obliged to acknowledge when it
recognizes implicitly that in the light of the text of the Erratum , the “Téra sector” necessarily
extends to the point where the frontier reaches the River Sirba at Bossébangou. Indeed, after
citing the text of the Erratum and putting the passages relating to the sector in question in
italics, in particular the extract reproduced above, Niger blandly asserts that “the only stretch
still in dispute for the Téra sector is that indicated in italics in the two preceding
quotations” 310. However, the line claimed by Niger ne ver either reaches this point or passes
through it, in contravention of the letter of the Erratum.
3.18. This leads to some very confusing st atements. For example, Niger alleges that
from the Tao astronomic marker “the official text gives no further indication until the point
where the inter-colonial boundary rejoins the boundary of Say cercle” 311, as if the text
referred to that latter point. Similarly, Nige r defends the idea that the 1960 map “reproduces
the mistake contained in the Erratum of 1927” in making “the boundary between Upper Volta
312
and Niger descend as far as Bossébangou” ⎯ in other words, Niger comes to acknowledge
that both the Erratum and the line shown on the 1960 map make the point where the boundary
reaches the River Sirba at Bossébangou a frontier point, while nevertheless excluding this
313
point from the line that it claims in its Memorial .
3.19. Niger does not merely commit a sin of omission. It not only takes the liberty of
leaving out a frontier point expressly referred to in the Erratum, but also has no hesitation in
adding new points not mentioned in the Erratum. This is the case for the Vibourié marker
which, according to Niger, and despite the fact that unlike the Tong-Tong and Tao markers it
is not mentioned in the text of the Erratum, h as been “since the colonial era, the next point on
314
the boundary”, located between the two markers designated in the Erratum . The same goes
306
“It then follows the course of the Tapoa upstream until it meetthe former boundary of the Fada and
Say cercles, which it follows as far as the point where it intersects with the course of the Mekrou.” [Emphasis
added.]
30MN, p. 79, para. 6.1.
308
Ibid. (emphasis added). See also MN, pp. 97-100.
309
Ann. MBF 35 (emphasis added).
31MN, pp. 82-83, para. 6.9 (emphasis added).
31MN, p. 93, para. 6.21.
31MN, pp. 110-111, para. 7.21.
313
On the argument that the explicit referen ce to the Sirba at Bossébangou in the amended Arrêté of 1927
was a mistake, see paras. 4.20-4.29 below.
314
MN, pp. 92-93, para. 6.20. On the Vibourié marker, see Section 2.1, paras. 3.44-3.52 below. - 54 -
for the numerous frontier points that Niger i nvents between the Tao marker and the endpoint
of the frontier in this sector 315.
3.20. The result of these inventions is that the line claimed by Niger departs completely
from the text of the Erratum. According to th e clear terms of the latter, which can be split
into two stages, the frontier points being indicated in italics:
“⎯ [from]the Tong-Tong astronomic marker [,] this line then turns towards the
south-east, cutting the Téra-Dori motor road at the Tao astronomic marker
located to the west of the Ossolo Pool,
⎯ and reaching the River Sirba at Bossebangou.”
This is what becomes of the line in Niger’s Memorial:
“the course of the frontier in the Téra sector will be examined by being
subdivided into three sections: from Tong-Tong to Tao (a)316rom Tao to
Bangaré (b), and from Bangaré to the boundary of Say cercle (c)” .
3.21 Above and beyond these errors, Niger is retreating into a wholly anachronistic
strategy. Visibly dissatisfied with the delimitation made by the colonizer in the1927
Erratum, to which it nonetheless agreed to give full effect, as was required by customary
317
international law, by concluding the19 87 Agreement and Protocol of Agreement , Niger
endeavours to circumvent the amended Arrêté of 1927 in two complementary ways:
(i) by seeking firstly to rewrite its terms on the basis of certain previous proposals for
delimitation which have not, however, been confirmed by the competent authority
(Section 1); and
(ii) by then seeking to rely on factors which emerged subsequent to the 1927 Erratum in
order to redefine its terms, even though the Parties have never ceased recalling that
this legal act constituted the only applicable title, and without in any case providing
any factual evidence in this regard (Section 2).
315See paras. 3.65 et seq. below.
316
MN, p. 46, para. 6.17.
317Moreover, it may be useful to recall that those nvolved at the time apparently considered that the
Colony of Niger had emerged with a distinct advantage from the delimitation exercise carried out in 1927, to the
detriment of Upper Volta. This is reflected in the ecriminations of Administrator Delbos, which Niger itself
makes much of. See paras3.22 et seq . below, and also MN, Anns., SeriesC, No.21 (LetterNo.96 from the
Commander of Dori cercle to the Governor of Upper Volta date d 23April1929, second page: “For the
delimitation established between Niger and Upper Volta was generous enough (this is a euphemism . . . ) to leave
Niger with the major watering places . . .”). - 55 -
SECTION 1
E VENTS PRIOR TO THE AMENDED A RRÊTÉ OF 1927
3.22. The “methodology adopted” by Niger in its Memorial h as the sole aim of
318
circumventing the clear text of the amended Arrêté of 1927 . Considering the text of this
Arrêté to be “particularly succinct”, containing only “rudimentary indications” 319, Niger
contends,
“[i]t is therefore necessary to look elsewher e in order to identify this stretch of
the boundary between the two territories. Th e 320tory of its origins offers such a
possibility, which should . . . be explored.”
3.23. It is strange to see Niger looking elsewhere even before it has explored the
possibilities offered by the 1927 Erratum, which is sufficient in itself (the description of a line
in large sections does not preclude it from bei ng precise, as long as the orientation of the
sections in question and the identification of their respective endpoints are clear and
“operational”). The line defined in the Erratum being perfectly c321r, all that is required is to
follow its course, as Burkina has explained in its Memorial .
3.24. Niger’s reasoning is characterized, furt hermore, by an intrinsic contradiction.
Exploring “the history of its origins” actually leads it to the finding that the proposals for
delimitation submitted before the adoption of the Arrêté by Administrators Delbos and
Prudon, which were based on an attemp t to establish the “boundaries of the cantons of their
respective cercles” 322, differed from the solutions finally chosen by the competent authority.
This was the reason why Administrator Delbos later “protested vigorously” against the Arrêté
as adopted, but, as Niger agrees, his urgent plea “had no effect, and no change was made to
323
the legislative text up to the time of independence” . It is to be deduced from this that the
amended Arrêté of 1927 did not intend to confirm “the boundaries of the cantons of their
respective cercles” that the two Administrators had attempted to establish ⎯ the latter
324
furthermore disagreeing with each other on this point .
3.25. Niger however relies definitively in the present sector on the proposals of Delbos
and Prudon in order to interpret (or rather, in actual fact, to revise) the 1927 Erratum. Niger
indeed claims that,
“notwithstanding the wish frequently expressed by officials of the two Colonies,
the course of the boundary was never clarified by a new text so as to corr325ond
more closely with the actual boundaries of the cantons in practice” .
31See Chap. I above.
31MN, p. 83, para. 6.9 in fine, and p. 84, para 6.10 in fine. See also paras. 1.42-1.43 above.
320
MN, pp. 84-85, para. 6.11.
321
See MBF, pp. 98 et seq., Chap. IV, Section 1.
32MN, p. 87, para. 6.12.
32MN, p. 90. para. 6.14.
32See sketch-map No.2 above, and MN, p.87, para.6. 12: the reports of the two administrators “are
similar, even though they do not totally coincide”.
325
MN, p. 91, para. 6.16 [emphasis added]. - 56 -
3.26. This strategy should not mislead the C ourt. By implying that it would have been
possible to clarify the text of the Arrêté to make it correspond more closely with the claimed
“actual boundaries... in practice” of the cant ons, Niger is suggesting that actual possession
of territory (if accepted as probative and proven) might constitute a title complementary to the
corrected Arrêté of1927. This is indeed the strategy that it deploys further on in its
Memorial, when it considers that the line on the 1960 map could be added to and even replace
the delimitation definitively la id down in the 1927 Erratum, on 326 basis of findings on
possession subsequent to the date of adoption of the latter . The idea that the proposals of
Administrators Delbos and Prudon could thus be used to “clarify” the Erratum is however
contrary to “the history of [the] origins” of the Arrêté on which Niger relies: the
1927 Arrêté ⎯ as is attested by Administ rator Delbos’s “urgent plea” ⎯ actually diverged
from the boundaries of the cantons which the two Administrators had sought to establish ⎯
327
but without arriving at an identical definition of them . They cannot therefore be used, after
the event, to clarify the Erratum, since the latter did not have the intention of confirming
them. In order to make the delimitation a dopted in 1927 “correspond more closely” to the
claimed “actual boundaries of the cantons in practice”, as Niger itself admits, a “new text”
would have been necessary. Such a text would not have clarified the delimitation decided
upon in 1927; on the contrary, it would have amended it. Nonetheless, such a text was never
forthcoming. Niger’s strategy is t328 in fact to circumvent the legal title applicable in the
present case, the 1927 Erratum .
3.27. Furthermore, upon careful examinati on, Niger’s thesis is characterized by
absurdity in its very wording. As has previously been recalled 329, the proposals submitted by
Administrators Delbos and Prudon were simultan eously: (i) different from the delimitation
established in the Record of Agreement of 2 February 1927 between the Governors of the
Colonies concerned, which was adopted befo re the work of the Administrators was
undertaken; (ii) different from the line in th e 1927 Erratum which was adopted after that
work had been submitted and which diverg ed from it by going back, subject to a few
amendments, to the line laid down in the 1927 R ecord of Agreement; and (iii) different one
from the other. In such ci rcumstances, arguing that alleged pre-existing boundaries might
“clarify” the text of the Arrêté , i.e., help to interpret its terms, on the grounds that, according
to Niger, it is supposed to have establishe d a delimitation in consideration of those
boundaries, does not make sense, since at the critical date of 1927, the Administrators’
proposals were specifically not taken into account in the Erratum. Nor is it possible to
consider that the Erratum would have referre d to the “actual” boundaries which might have
32Id., and Section 2, paras. 3.41 et seq. below.
327
See fn. 324 above.
328
See Chap. 1, Section 1 above.
32Ibid. - 57 -
been “in practice” subsequent to its adoption: firstly, it in no way provides for this and,
secondly, such a way of proceeding would ha ve been quite nonsensical in an act of
delimitation 330.
3.28. On another level, it is equally diffic ult to defend the thesis advanced by Niger.
According to Niger, “the new boundary” established by the amended Arrêté of 1927 “was
defined as a series of juxtaposed cantonal boundaries, themselves composed of a series of
village and/or hamlet boundaries” 331. This would appear to give rise to a
“presumption that the areas composing these cantons, occupied by indigenous
peoples, and the villages, fields or pastures and nomad routes, did not in
principle follow abstract lines (whether curved or straight), but were based on 332
land occupation and followed the configuration or nature of the ground” .
3.29. It is strange to apply a “presumption” to a text. A delimitation of this kind, based
on “land occupation”, does not emerge in any way from the text of the amended Arrêté.
Moreover, Niger itself accepts a system of straight lines along various sections of the border,
of up to around 100 kms in length 333. It also acknowledges that “in sparsely populated areas,
the canton boundaries were quite vague: fo334xample on rocky hillsides and infertile
plateaux, and in open pastureland” . The contradiction is blatant: the canton boundaries
were vague in open pastureland, yet it is on such factors that Niger claims to base the colonial
delimitation.
3.30. The indications provided by Administrators Delbos and Prudon relating to the
alleged “actual boundaries... in practice” on the eve of the adoption of the 1927 Arrêté do
not in any case provide any support whatsoever for the line claimed by Niger. As the Delbos
report has not been found, it has not been possibl e to establish directly the factual bases for
33The dicta of the International Court of Justice concerning conventional border delimitations are entirely
valid in the present case since the obj ect of these delimitations does not f undamentally differ from a delimitation
between colonies. In the case concerning the Temple of Preah Vihear , the Court pointed out that “in general
terms, when two countries establish a frontier between them, one of the pr imary objects is to achieve stability and
finality” (Judgment of15 June1962, I.C.J. Reports 1962, p.34). Similarly, in the Libya/Chad case in1994, the
Court recalled its 1959 dictum in the Frontier Land case in these terms:
“Similarly, in 1959 in the case concerning Sovereignty over Ce rtain Frontier Land , the
Court took note of the Preamble to a Boundary C onvention as recording the common intention of
the parties to ‘fix and regulate all that relates to the demarcation of the frontier’ and held that ‘Any
interpretation under which the Boundary Conventi on is regarded as leaving in suspense and
abandoning for a subsequent appreciation of the status quo the determination of the right of one
State or the other to the disputed plots would be incompatible with that common intention’
(I.C.J. Reports 1959, pp. 221-222).” (Judgment of 3 February 1994, I.CJ. Reports1994 , p.24,
para. 47.)
In the present case, the intention of the author the 1927 Erratum is no different: the aim was to lay
down once and for all, on the day of adoption, the boundary between the Colonies of Upper Volta and Niger, as
was explicitly requested by the Decree of the President of the French Republic of 28December1926 (see
paras. 1.12-1.15 above).
331
MN, pp.80-81, para. 6.6; see also paras. 1.4 et seq. above.
33MN, pp. 84-86, para. 6.11.
33See in particular paras. 4.54 et seq. below.
334
MN, pp. 80-81, para. 6.6. - 58 -
335
the purely geometric boundary line he put forward . Administrator Prudon’s Report of
4August1927 has for its part been appended by Niger to its Memorial. Reading it is an
instructive exercise in several respects 336:
⎯ in the relevant sector, Administrator Prudon has listed on the sketch-map attached to his
report only about ten villages between the Ta o marker and the River Sirba, which are
nonetheless around a hundred kilometres apart; that goes to show that the region was
sparsely populated;
⎯ this is confirmed by the fact that the Admi nistrator notes that in this same sector from
“Doulgou to Tao” then from “Tao to Diamafoundé” he has recorded “no dispute”;
⎯ in this same Report which, according to Niger, establishes the so-called “actual
boundaries . . . in practice” at the time when the 1927 Arrêté was adopted, the village of
Bangaré (spelled “Bengaré” on Administrator Delbos’s sketch-map), which today Niger
claims to be on its side of the frontier line 33, is not marked as one of the “villages and
groupements asking to be transferred to Niger and located in Tillabéry cercle”, according
to the wording used by the map’s legend. In fact, it is placed by Administrator Prudon on
the Upper Volta side of the frontier that he proposes. The same solution is reached by the
line adopted by the amended Arrêté of 1927: the straight line linking the Tao astronomic
marker to the point where it reaches the River Sirba at Bossébangou also places the
village of Bangaré within Burkina Faso’s territory;
⎯ there are no findings, on the other hand, wh ich would provide support for the frontier
Niger is claiming today, and in any case Niger does not invoke any such findings.
3.31. Although Administrator Delbos’s Re port has not been found, an idea of its
content is provided indirectly by a letter h338en t to his administrative authority, the Governor
of Upper Volta, on 17 December 1927 . This letter confirms Burkina’s reading of the 1927
Erratum: the latter in no way set out to confir m so-called “actual boundaries . . . in practice”,
nor to resort to the use of natural frontiers (to the great regret of the Commander of Dori
cercle, who felt that the Colony of Upper Volta ha d thereby lost considerable territory in
335
The description of the course of this boundary is known through a letter from Delbos, Commander of
Dori cercle, to the Governor of Upper Volta, dated 27 August 1927 (MN, Anns., Series C, No. 16):
“From this point the boundary, descending on a bearing of 156°, crosses the Téra-Dori
road5.75km from Tao (Soum Pool); on reac hing Tao it descends on a bearing of 135°for
27.5km, then for 26.5km on a bearing of 147°, until it reaches a point 5km to the north of the
Iga Pool. It then turns back up in a north-easterly direction on a bearing of 79° for 31.5 km, before
redescending on a bearing of 127°for a distance of 13.5km, and then on a bearing of 190°for
25.5km, before finally following a bearing of 170°until it reaches the boundary of Say cercle to
the west of Alfassi on the River Cirba.”
It is clear that this proposal does not name the pl aces where the boundary changes direction, which makes
it more difficult to visualize the line suggested. It should be noted, furtherm ore, that the Administrator confuses
the Soum Pool, which is much further north (and which wa s therefore the object of debate in the case concerning
the Frontier Dispute (Burkina/Mali) ⎯ see I.C.JJ.udgment of 12December1986, I.C.J. Reports 1986,
pp. 627-628, paras. 138-139) with the Solo Pool. Such confusion also limits the probative value of this report.
336MN, Anns. SeriesC, No.15 (Extract No.25 from the Tour Report of Administrator Prudon, dated
4August1927) and MN Anns., Se ries D, No.3 (Tillabéry cercle, 1:200,000-scale sketch-map drawn up by
Administrator Prudon in June 1927).
337See paras. 3.80-3.83 below.
338MN Anns., Series C, No. 20 (Letter No. 731 from Administrator Delbos, Commander of Dori cercle, to
the Governor of Upper Volta dated 17 December 1927). - 59 -
339
comparison with the delimitation that he himself had proposed ). According to
Administrator Delbos, who in his letter pr ovides an expert interpretation of the Arrêté with
particular regard to “the history of its origins”, in which he had been one of the protagonists:
340
“The boundaries as described in Official Journal No. 1021 are an exact
copy from the Report signed in my presence at Téra by Governor Brévié and
Inspector Lefilliatre. They had been established on the basis of the map prepared
by Captain Coquibus, which only showed theoretical lines and points . . .”
3.32. That goes to confirm that the author of the amended Arrêté chose to use
“theoretical lines and points” to delimit the territory of the two Colonies, as Burkina
maintains.
3.33. Niger finally cites in its Memorial “the agreement between CommandersRoser
(Dori) and Boyer (Tillabéry) of 21March1932”, the two officials, according to Niger,
wishing “better to reflect the true situation on the ground”, having referred to the work of
Administrators Delbos and Prudon 341. This document, which given its date sheds useful
retrospective light on this work, is for its part particularly instructive when one contrasts its
terms with the line claimed by Niger. Firstly, it shows that the canton boundaries which,
according to Niger, the 1927 Arrêté had confirmed, were actually very ill-defined at that date;
secondly, it confirms that the line adopted in the Erratum diverges from the boundaries
proposed by Administrators Delbos and Prudon; finally, it contradicts Niger’s claims relating
to certain villages and as a result the conclusions drawn by Niger when establishing its line.
3.34. On the first point, the document points out that the canton of Diagourou (Niger),
bordering Dori cercle (Upper Volta), was never delimited. Commander Roser’s observations
on the matter, written five years after the adoption of the 1927 Arrêté, the purpose of which,
according to Niger, was to confirm the alleged pre-existing canton boundaries, are
enlightening:
“In 1919 or 1920, he [the Chief of the Diagourou] was given a territory,
without precise boundaries, that forms the current canton of the Diagourou. He
himself acknowledges that he does not know the boundaries of his canton.”
3.35. On the second point, Commander Roser’s letter is unambiguous: the author of
the 1927 Arrêté should have adopted as a boundary the natural frontier proposed by
Administrators Delbos and Prudon, whose “ideal nature... is obvious”; it is because this
was not done that Commander Roser considers th at “common sense and reality require that
this boundary be modified” 342 by means of a “further erratum”. The proposal pre-drafted by
Commander Roser for this purpose provides for the adoption of a partly natural frontier which
339
Ibid.:
“While we can accept the Kabia-Iga section . . . it seems to me that it would be difficult to
abandon the area which I have marked in red, sin ce this area, which is surrounded by hills forming
natural boundaries, has always belonged to the Yagha canton without ever being challenged by any
of the neighbouring peoples.”
340
This is actually O.J. No. 1201 (see Ann. MBF 34).
34MN, p.90, para6.14, and MN, Anns., SeriesC, No.45 (Letter No.112 of 10April1932 and
Tour Report from Civil Service Deputy Roser, Acting Commander of Dori cercle, to the Governor of Upper Volta
(Political Office)).
34Emphasis added. - 60 -
follows various watersheds. A contrario, this is not what the amended Arrêté does in this
sector. This confirms that its author, rather than selecting a boundary considered closer to the
“reality of the situation”, which as a result woul d have been more complex, opted for a more
straightforward frontier made up of artificial lines.
3.36. With regard to the third point, Niger ma intains in it343emorial that the village of
Bangaré “has always been located in the territory of Niger” . However, this is in no way
what emerges from Commander Roser’s letter of 1932. Commenting on the validity of the
line adopted in 1927, the latter writes:
“However, if this boundary is accepte d as correct and definitive, then
Upper Volta must immediately annex the large village of Bangaré, which has
always belonged to Téra canton, but which is located to the west, on the Volta
side, of the famous ‘line’ . This example clearly shows that common sense and
344
reality require that this boundary be modified.”
As Niger points out in its Memorial, this boundary was never modified and the amended
Arrêté of 1927 has been held by the Parties to this dispute to constitute the one and only legal
title to be relied upon. Incidentally, it will be recalled that in 1927 the village of Bangaré had
been placed on the Upper Volta side of the inter-colonial boundary by Administrator Prudon
himself 345. Therefore Commander Roser is mistaken when he contrasts the work of
Administrators Delbos and Prudon with the amended Arrêté of 1927 on that point: they all
place Bangaré on the Upper Volta side of the frontier.
3.37. Other documents appended by Nige r to its Memorial serve to confirm,
retrospectively, that the author of the 1927 Arrêté meant to adopt in this sector a route
following an abstract line and not one ratifying allegedly pre-existing canton boundaries,
which would have followed an extremely comple x succession of natural frontiers. This is a
solution which may have been regretted by certain colonial authorities; it is nonetheless the
delimitation which was finally adopted and which has legal authority in the present case.
3.38. In a letter dated 9August 1929 to the Commander of Tillabéry cercle, the
Commander of Dori cercle writes, for example:
“These local people, as you pointed out to me, live and grow crops in Téra
territory, which is territory established by the Arrêté and Erratum on
delimitation. They have been there for a long time; and, unaware of our
346
[???]fic347ons of administrative geography , believe themselves to be at
home.”
Considering that the mismatch between “ad ministrative geography” and the situation
perceived by the populations caused some difficulties, the Commander of Dori cercle
proposed to his counterpart that they introduce a policy of tolerance vis-à-vis these local
populations by “slightly modifying our current boundaries”, in order to avoid any “annexation
34MN, pp. 97-98, para. 6.24.
344
MN, Anns., Series C, No. 45 [emphasis added].
345
See para. 3.30 above.
34The beginning of the word is missing. This word appears to be “modifications”.
34MN Anns., Series C, No. 24 (Letter No. 399 from the Commander of Dori cercle to the Commander of
Tillabéry cercle dated 9 August 1929) (emphasis added). - 61 -
of neighbouring cantons allocated to Dori”. The Commander immediately added: “Of course,
the Arrêté and Erratum signed by the Governor-General no longer refer to cantons but only to
boundaries; and that, I acknowledge, is crucial.” 348 The Commander also referred to the
“rigours of the 1927 delimitation” which his proposal to modify the boundaries would have
set out to “mitigate”349.
3.39. On 7 July 1930, the Commander of Dori cercle once again wrote in his “report on
the difficulties created by the delimitation established in 1927”:
“The Arrêté général of 31August1927 fixed the boundaries of the
Colonies of Upper Volta and of Niger. This Arrêté reproduced the Record of
Agreement signed in Téra on 2February1927 and took no account of the
delimitation carried out on the ground by the two cercles Commanders of Dori
and Tillabéry . It should also be noted that this Arrêté is established at the
instigation of the Lieutenant-Governor of Niger. An Erratum to this Arrêté
makes almost no changes to the boundary, except that the frontier line is to reach
350
the River Sirba at Bossébangou instead of Boulkebo.”
3.40. Once and for all, it is not possible for Niger to circumvent the perfectly clear text
of the Erratum by trying to revise it on the basis of previous proposals which were not
adopted in 1927.
SECTION 2
EVENTS SUBSEQUENT TO THE 1927 E RRATUM
3.41. Considering it necessary to clarify the 1927 Erratum to make it “correspond more
closely with the [supposed] actual boundaries of the cantons in practice”, and unable to take
as a basis the lines proposed by Administrators Delbos and Prudon which the amended Arrêté
did not confirm, Niger ends up basing its lin e on actual possession afte r 1927, by proposing
the following “method”:
“the drafters of the 1960 map based them selves on a body of relevant data in
order to represent the probable boundaries of the cantons as these were applied
in practice at the critical date. In consequence, unless we find abnormal
deviations in relation to the texts or ma nifest lacunae in the information on the
canton boundaries, and subject to the necessary caution where the hesitation of
the map’s drafters is reflected in gaps in the line of crosses, these results should
in principle serve as a guide to dete rmine the course of the inter-colonial
351
boundary in 1960.”
348
Ibid. (emphasis added).
34Ibid., also para. 1.32 above.
35MN, Anns., SeriesC, No.38 (2 ndpage) (Report No.416 of the Commander of Dori cercle on the
difficulties created by the delimitation established in 1927 between the Colonies of Niger and Upper Volta (Arrêté
of 31August1927) regarding the bounda ries between Dori and Tillabéry cercles, dated7 July1930) (emphasis
added).
35MN, p. 91, para. 6.16. - 62 -
3.42. Th352way of proceeding does not comply with the applicable law, as has already
been recalled . Moreover, it gives rise to the following comments:
(i) the subjectivity which flaws and discredits it finds expression in the course of the
boundary put forward by Niger itself, which at one point prefers straight lines to the
line on the 1960 map, then prefers this latter line to straight lines, and then again
substitutes its own line for the one laid down in the amended Arrêté of 1927 or the 353
one featured on the 1960 map, for reasons which are impossible to understand ;
(ii) Niger claims to uphold ⎯ though it actually does this only in a very partial way ⎯
the line shown on the map between the Tao astronomic marker and the endpoint of
the sector covered by this Chapter, even though this line is characterized by
“abnormal deviations” compared to the text of the Erratum: indeed, nothing in the
text of the latter could result in the line given on the map between these two points.
If Niger had followed its own methodology, it should not therefore have upheld the
line shown on the map in this sector ⎯ in accordance, moreover, with what it does
between the Tong-Tong and Tao markers:
(iii) besides, Niger implicitly admits as much in its Memorial:
“It is clear that, in the absence of reliable information from the local
authorities, the drafters of the map followed the rivers, marigots and ridgelines,
which together represent [close to] 50percent of the boundaries for the Téra
sector... All of this implies that, far from relying on the old sketch-maps,
which showed straight or curved lines connecting isolated points, the drafters of
the 1960 map based themselves on a whole body of pertinent data in order to
represent the probable boundaries of the cantons as they were applied in practice
354
at the critical date.”
The drafters of the map are not, however, the author of the Arrêté and the purpose of the latter
(to define a new inter-colonial boundary) is not the purpose of the former, as described by
Niger (“to represent the probable boundaries of the cantons as they were applied in practice at
the critical date” 355). In fact, the 1927 Erratum does not refer to any natural frontier in this
sector, contrary to what it does in other sectors. The decision not to refer to such a natural
frontier therefore rules out any addition to the text of the Erratum along the lines advocated
by Niger using the 1960 map: it is simply not possible to add to the text of the Erratum
natural boundaries amounting to “[close to] 50percent” of this section which are entirely
unmentioned by that text, unless the text itself is replaced with a new definition of the course
of the frontier.
3.43. This strategy followed by Niger, that is, re-writing the Erratum by replacing it
with a different line on the basis of findings subsequent to its adoption, takes a two-pronged
approach: firstly, Niger invents a new frontie r point, the Vibourié marker, put in place
352
See paras. 1.40-1.64 above.
35See paras. 3.8 et seq. above.
35MN, pp. 75-76, para. 5.14.
355
The “critical date” mentioned by Niger (1960) is moreover not the correct critical date. If it is accepted
that the so-called “actual boundaries of the cantons” might have constituted a factor enabling interpretation of the
amended Arrêté of 1927, they should have been examined on the day of its adoption, in 1927. However, as is well
known, on that date the author of the Arrêtéspecifically disregarded these alleged boundaries (see Section1
above). Niger’s argument cannot find a way out of the vicious circle in which it is trapped. - 63 -
in1935; and secondly, it invokes supposed village effectivités in order to depart from the
course of the boundary laid down in the Erratum.
1. The Vibourié marker is not a frontier point
3.44. Between the two astronomic marker s of Tong-Tong and Tao, mentioned in
the 1927 Erratum, Niger contends that a new frontier point, not provided for in the Erratum,
should be inserted. According to Niger, the Vibourié marker, put in place in 1935, should be
356
granted “the status of a frontier point” . This frontier point causes Niger to replace the
straight line linking the Tong-Tong and Tao mark ers with “two straight lines” whose effect is
to shift the frontier line eastwards (see sket ch-map No.3 above, Course of the boundary
between the Tong Tong and Tao markers).
3.45. Niger’s position is totally unjustified. Niger relies on a Record of Agreement of
13Apri1l935 between Administrator Garnier (Dori cercle) and Assistant Deputy
Lichtenberger (Téra cercle), who established this marker “in order to prevent any . . . further
territorial disputes in this area” 35. Yet, contrary to Niger’s assertions, the establishment of
this marker could not have had the effect of moving the line laid down by the 1927 Erratum.
3.46. For one thing, as Niger admits, this agreement of 1935 and the approval given to
it by the Governor of Niger that same year “date[s] from after the disappearance of Upper
Volta and hence [their] retention following the reconstitution of the Colony could be regarded
358
as debatable” . In fact the boundaries of Upper Volta were fixed in 1947, when it was
reconstituted, by reference to those which were in force in 1932, when it was dismembered.
What happened between these two dates therefore has no effect on the delimitation.
3.47. Secondly, and in any case, even accepting that the Vibourié marker was
established at the place where Niger situates it in its Memorial, according to the co-ordinates
given, it would appear obvious that a mistake was made in1935. The intention of those
drafting the 1935 Record of Agreement is not at issue. On this point, Burkina unreservedly
endorses Niger’s359servation that this document constitutes an “interpretation of the
1927 Erratum” . The authors of the agreement in fact decided,
“[r]eferring to the delimitation determined by the Arrêté of 31 August 1927
(Erratum) between Dori and Téra,... to visit the site in order to observe the
placement of said land in relation to the above-mentioned boundary.
[...] in order to prevent any similar fu rther territorial disputes in this area, we
have established a marker designed to fix the boundary between Dori and Téra,
the boundary in principle following a notional straight line starting from the
Tong-Tong astronomic marker and running to the Tao astronomic marker.
356
MN, pp. 92-93, para. 6.20.
357
MN Anns., SeriesC, No.56 (Certified copy of 14 April of Record of Agreement of 13April1935
between Administrator Garnier (Dori cercle) and Deputy Lichtenberger (Téra Subdivision)).
35MN, pp. 92-93, para. 6.20.
35Ibid. - 64 -
The Ouiborie360[Vibourié according to Niger] marker being located on this
notional line . . .”
3.48. It can be seen from this Record of Ag reement that the two Administrators agreed
on the interpretation to be given to the 1927 E rratum: the boundary it defines is “a notional
straight line starting from the Tong-Tong astronomic marker and running to the Tao
astronomic marker”. This is the interpreta tion which BurkinaFaso has never ceased to
defend, and its position is thus fully validated . Niger furthermore confirms that this
interpretation is well-founded when it notes that
“[t]he sketch-maps prepared in 1927 by Delbos and Prudon, as well as the map,
‘New frontier of Upper Volta and Niger’, published the same year, connect these
361
two points [the Tao and Tong-Tong markers] with a straight line” .
3.49. In accordance with this interpretation, it was decided to establish a new marker ⎯
the Vibourié marker ⎯ “on this notional line”. This shows that this marker was not intended
to become a genuine frontier point. This marker was established on the boundary defined in
the amended Arrêté of 1927 not because the boundary had to pass through this point , but,
quite the reverse, because this point was located on the boundary.
3.50. According to Niger, it would nonethel ess appear in retrospect that the Vibourié
marker was not established on this boundary, which explains the fact that Niger’s line, instead
of following a single straight line between the Tong-Tong and Tao markers, moves
considerably further eastwards so as to li nk the Tong-Tong, Vibourié and Tao markers via
two straight-line sections.
3.51. However, as the Vibourié marker is not mentioned in the 1927 Erratum (and with
good reason, since it was established after the la tter was adopted), and as, furthermore, the
purpose of establishing it was not to amend the line laid down in the Erratum (in fact quite the
reverse, since the marker was supposed to be placed on that line), it is quite obvious that the
spot where the marker was apparently establishe d cannot be turned into a new frontier point
altering the course of the boundary adopted in 1927. Once again, this was not the intention of
the authors of the Record of Agreement of 1935 (their intention in fact was quite the
opposite); and in any case they would not have been empowered to do such a thing.
3.52. Far from justifying the line claimed by Niger, the Record of Agreement of 1935
thus in fact confirms that between the Tong- Tong marker and the Tao marker, the course of
the boundary laid down in the Erratum of 1927 follows one straight line, and only one.
2. The claimed village effectivités do not exist
3.53. In the section which continues from the Tao marker, Niger adopts a course which,
in its own words, “basically follows the IGN line[ 362]”363. In other words, Niger is claiming a
36MN, Anns., Series C, No. 56 (emphasis added).
36MN, pp. 91-92, para. 6.18 (emphasis added). This contradicts what Niger states in para.6.10 of its
Memorial (pp.43-44), where it considers that between the Tao and Tong-Tong markers, the “New frontier” map
shows a “slightly curved” line. - 65 -
course which is neither that of the Erratum, nor that shown on the 1960 map, but a third one.
In fact, it is wrong to claim that Niger’s course “basically” follows the line shown on the
1960 map. It actually diverges from it substantially (see sketch-map No. 4 above, Course of
the boundary after the Tao marker).
3.54. Niger bases its claim on tortuous reasoning involving a succession of interlinked
premises; it contends that:
⎯ the Arrêté of 1927 and its Erratum are imprecise in this sector;
⎯ they should therefore be aligned more clo sely with the so-called “actual” boundaries of
the cantons “in practice”;
⎯ the 1960 map is (on that date, and not in 1927) the reflection of those boundaries;
⎯ therefore it is appropriate to follow the course it shows;
⎯ at least, if it corresponds to the boundaries (tha t is, according to Niger, if the line on the
map corresponds to possession of the villages located in the frontier zone);
⎯ if this is not the case, then actual possessi on becomes the determining factor and takes
priority over the 1960 map.
3.55. Each one of these premises is erroneous:
⎯ firstly, the amended Arrêté of 1927 is not imprecise: the Governor-General of FWA
made the choice of linking points by means of theoretical lines, which are of necessity
364
straight lines ; the artificial delimitation he thereby establishes is sufficient in itself and
was fully understood as such by the colonial officials ⎯ not least those who complained
365
about it ;
⎯ the fact that the implementation of this theoretical delimitation on the ground might have
posed a problem relates to an entirely separate debate. The few documents noting a “lack
of precision” in the boundaries are not using this term to refer to the lack of precision of
the delimitation made by the Erratum, but to the difficulties involved in actually applying
it, which is a different matter entirely: thus when, for example, it was recalled in 1929
that
“⎯ the incidents occurring over the last few months were due to...lack of
precision when indicating the demarcation of certain areas to neighbouring
366
local chiefs” ;
362
The term “IGN line” is incorrect. In conformity with the terms of the 1987 Protocol of Agreement,
reference should be made to the line shown on the 1960 IGN map.
363
MN, pp. 93-97, paras. 6.21 et seq.
364See MBF, pp. 94 et seq., Chap. IV, Section 1.
365See above, in particular paras. 3.2, 3.31, 3.35, 3.38 and 3.39.
366MN, Anns., SeriesC, No.22 (Letter No.E/251AP from Fousset, Chief Colonial Administrator, to the
Lieutenant-Governor of Niger dated 31 July 1929) [emphasis added]. - 66 -
or when in 1953 Deputy-Administrator Lacroix (Tillabéry cercle), comparing the delimitation
made in the 1927 Erratum to that proposed by Administrator Delbos, noted with reference to
the latter:
“This shows that the least clearly defined portion of the boundary, from
Tao to Iga, was determined in meas ured segments along compass bearings:
Again, this clearly involved theoretical lines ⎯ which still did not mean much to
the local people concerned ⎯ but which were nonetheless easier to identify in
the field than the Tao-Sirba line in the Arrêté.” 367
This confirms that the colonial authorities had no doubt that the delimitation established by
the amended Arrêté of 1927 followed a single line (“the Tao-Sirba line” as opposed to the
“lines” or “sections” proposed by Delbos), which was artificial (“theoretical”) in nature;
⎯ the amended Arrêté of 1927 moreover never set out to confirm the canton boundaries as
368
they would have existed in 1927 ⎯ in fact, it did quite the opposite ; nothing therefore
permits these boundaries to be relied upon;
⎯ if the author of the amended Arrêté had, as Niger claims, intended to define the
inter-colonial boundary in this sector in such a way as to allocate a particular village to a
particular colony, he would have expressly indicated this; the same would apply if he had
intended to confirm a natural frontier;
⎯ it is not permissible to rely on effectivités subsequent to the adoption of the Erratum, as
title has priority over actual possession subsequent to the title;
⎯ neither can the use of the 1960 map as a reflection of actual possession existing at the
date of independence constitute an argument allowing the delimitation established in the
1927 Erratum to be validly disregarded or even merely adapted.
3.56. Even accepting that it could legally have been followed, which is not the case, the
approach consisting of relying on colonial effectivités subsequent to the frontier title is ruled
out in factual terms in the present case, for the following reasons.
3.57. First of all, it is ha rd to imagine that genuinely effective possession could have
become established on a lasting basis subsequent to the 1927 Erratum. Indeed, just five years
after that act was adopted, the Colony of U pper Volta was dismembered. In the following
fifteen years, the territories bordering on the current frontier between Burkina and Niger
belonged to one and the same colony, Niger. Any possession ⎯ which would necessarily
have been held by Niger ⎯ between these two dates cannot therefore have the slightest effect
as far as the present case is concerned, sin ce from 1932 to 1947 the inter-colonial boundary
had disappeared. This therefore disqualifies from the outset the effectivités claimed by Niger
369
when they date from this period .
367
MN, Anns., SeriesC, No.79 (Report of a t our conducted from 16 to 23November1953 by Deputy
Administrator Lacroix (Tillabéry cercle), dated 24 December 1953) (emphasis added).
368
See Section 1, paras. 3.24-3.29 above.
36See the documents mentioned by Niger in fns. 268, 287, 288, 289 and 290 in its Memorial. - 67 -
3.58. Incidentally, Upper Volta was to be re-established in 1947 within its 1932
borders, hence within its 1927 borders, as they had not been changed in the meantime. Niger
itself recalls in its Memorial that
“neither of the two Parties contends that there was any change to the legal
situation existing between 4 September 1947 and 5 August 1960 . . . It follows
that we have to go back to see what was the instrument which, on
370
5 September 1932, governed the boundaries of the two Colonies.”
3.59. The indications of possession invoked by Niger are without relevance to an even
greater extent, given that it becomes obvious, upon reading the Annexes appended by the
Parties to their Memorials, that the delimitation adopted in 1927 in this sector, precisely
because it took the form of an artificial boundary, met with various difficulties or protests
when it was implemented on the ground, and this was the case right up until the
dismemberment of Upper Volta in 1932. In particular, this led the Commander of Dori cercle
to draft a long report on 7July1930 on the “difficulties created by the delimitation
established in 1927 between the Colonies of Niger and Upper Volta ( Arrêté of371
31August1927) regarding the boundaries between Dori cercle and Tillabéry cercle” .
What happened on the ground should not un der these circumstances be granted any
significance for the purposes of settling the present case.
3.60. In actual fact, the colonial author ities were fully aware that the “artificial”
colonial boundary which had been adopted coul d not reflect the complex situations on the
ground, far removed from any ideas of frontier division. In a letter dated 27 September 1929,
for example, this is what the Governor of Niger wrote to his counterpart in Upper Volta:
“With regard to the individuals or families who are in dispute, their
situations will have to be examined individually by common agreement between
the heads of neighbouring districts. It appears difficult to adopt an invariable
criterion for attaching them to one Colony or the other. However, if domicile
cannot be used, perhaps one useful determining factor to consider might be the
location of croplands of the parties concerned. Nonetheless, given that these
may interlock and overlap the frontier , one cannot stick rigidly to this criterion
and everything will have to be settled on a case-by-case basis. In any case, there
can be no question of systematically and forcibly returning natives from one side
or the other of the frontier, depriving them of their annual croplands, and neither
should they be prevented from grazing their herds along their customary routes
or watering them at their usual pools. The greatest possible freedom must be
granted to the nomads in this regard; all that matters is that they should not be
37MN, p. 62, para. 5.3.
371
MN, Anns., SeriesC, No.38 (Report No.416 from the Commander of Dori cercle on the difficulties
created by the delimitation estab lished in 1927 between the Colonies of Niger and Upper Volta Arrêté of
31August1927) regarding the boundaries between Dori cercle and Tillabéry cercle, 7July1930) ⎯ emphasis
added. Interestingly, this report i ndicates that even the work doAdministrators Delbos and Prudon only
gives a very imperfect reflection othe actual situation on the ground: “But during this field work, the two
Administrators had had to move very quickly and the heads of neighbouring cantons had not accompanied them
everywhere. Furthermore it seems that they had not been properly questioned about issues of dispute which might
[arise] following this delimitation.” - 68 -
allowed to evade their administrative obligations372 crossing a theoretical and
artificial frontier at an opportune moment.”
3.61. More broadly, the specific nature of the human geography in the disputed area
deprives the arguments based on possession of any probative value. It is indeed an
undisputed fact that the human geography of th e frontier area has always been characterized
by mobility on the part of the local people. This is an everyday occurrence and also follows a
more general pattern. Popul ation groups move according to weather conditions or the
economic situation. The consequence is the exis tence of “fossilized” or “ghost” villages, and
also a degree of vagueness with regard to the names of places in the frontier zone, to mention
just these two aspects. Besides, even the most sedentary groups may live in different villages
according to the season, and those villages may in some instances be on different sides of the 373
colonial frontier. As a result, “territorial boundaries [were] meaningless to the nomads” ;
“one does not need to be a very old colonial hand to know that the territories to which the
native groupements lay claim, in particular in semi-desert savannah areas, have traditional
boundaries which are somewhat imprecise. There are areas where they in terlock or overlap:
they are not drawn with the precision of urban concessions” 374. More radically, on 10 August
1954, the Head of Téra Subdivision admitted to,
“think[ing] like most of [his] predecessors that an exact delimitation of this
canton of Diagourou is absolutely impossible despite the never-ending claims
and disputes to which this situation giv es rise. Besides, these disputes have
never been very serious (except with regard to the canton of Téra, due to
relations between the current leaders); undertaking any sort of demarcation
work, even on a localized basis, would mean pointlessly stirring up old quarrels
which have more or less been forgotten.” 375
3.62. Niger does not call into question this particular characteris tic of the region’s
human geography. On the contrary, it fully endorses it, and this makes even more surprising
the value it believes it can give to territorial possession, which on its own admission was not
governed in any way by the position of the frontier, particularly at the time when the
1927 Arrêté was adopted:
“The problems of the frontier area are conditioned by various dominant
forms of production, namely: itinerant nomadism; seasonal trans-frontier
pastoral transhumance, conducted on a pendular basis; semi-nomadism;
sedentary field agriculture; itinerant agriculture; gold prospection and
extraction . . . The expansion and dispersal of villages makes it more difficult to
determine the course of the frontier. The exhaustion of the soil on the plateaux is
another movement factor. This frequently causes the inhabitants of a village to
transfer to a new site, situated a few kilometres from the previous one. It is not
37MN, Anns., Series C, No.42 (Letter No.2259 AGI from the Lieutenant-Governor of Niger to the
Lieutenant-Governor of Upper Volta, dated 27 Septembe r 1929) (emphasis added). Tax collection was still an
unresolved problem between the two Colonies the follow ing year (see, for exampl e, MN, Anns., Series C,
No. 37 ⎯ Letter No.362 from the Commander of Dori cercle to the Governor of UpperVolta dated
11 June 1930).
37MN, Anns., SeriesC, No.45 (Letter No.112 of 10April1932 and Tour Report from Civil Service
Deputy Roser, Acting Commander of Dori cercle, to the Governor of Upper Volta (Political Office), p. 3).
37MN, Anns., SeriesC, No.25 (Letter No.411 from the Commander of Dori cercle to the Governor of
Upper Volta dated 14 August 1929).
37MN, Anns., SeriesC, No.84 (Report from the Head of Téra Subdivision on the census of Diagourou
canton, dated 10 August 1954, pp. 5-6). - 69 -
unusual, in such cases, for the hamlets attached to the main village of origin to
376
have similar or identical names to the latter.”
And also according to Niger,
“The territorial partition did not create problems for the villages, which
were concentrated in a relatively confined space (a few hectares). However, for
peoples whose homelands were spread over more extended areas (covering
dozens, if not hundreds, of square k ilometres), their partition was socially
disruptive and provoked population movements motivated by the preservation of
communal or cultural identities, or th e safeguard of interests. Each cercle, now
wishing to know the precise number of its inhabitants, was impelled to carry out
censuses. The instability of the populati ons of areas close to the shared
boundaries or territories resulted in multiple registrations and the use of
contradictory criteria for defining admi nistrative links (place of temporary
settlement or village of origin).
Apart from traditional nomadic movements or the search for new land,
there were various factors impelling popula tions to change from one territory to
another: differences in régime as between colonies in the matter of compulsory
service or of human or livestock taxation, the existence of basic infrastructure in
the neighbouring territory (access to wate r, vaccination facilities for livestock,
schools, health centres, etc.), power relationships within tribes, etc. Thus, all
along the frontier, a game of cat-and-mouse developed between colonial
administrators and frontier populations.” 377
3.63. In such circumstances, the choice of an artificial boundary, despite its alleged
disadvantages, probably turned out to be the wisest one. As it was not possible to draw a
boundary following a practice alien to any concep t of territorial delimitation, the option of a
straight line linking identified points appeared to be the only way of meeting the objective
laid down by the author of the 1927 Arrêté: to delimit, for the purposes of administrative
management, the two districts of Niger and Upper Volta.
3.64. Moreover, Niger does not provide any factual evidence which would justify its
proposed course and stand in the way of the most obvious interpretation (adopted precisely
for that reason by the colonial authorities) of the amended Arrêté of 1927, according to which
a straight line formed the inter-colonial bounda ry from the Tao marker onwards. Firstly,
Niger does not take the trouble systematically to provide foundations for its assertions;
secondly, when it does do this, it is solely by referring to a practice subsequent to the adoption
of the 1927 Arrêté ⎯ supported by an extraordinarily small number of documents; it also
happens that Niger itself acknowledges that possession may have changed hands between
1927 and the subsequent period, which goes to show the complete irrelevance of practice
subsequent to the adoption of the Arrêté. Finally, these few documents are not in themselves
at all convincing, in any event.
3.65. Burkina will specifically demonstrate in the next part of this text how the
proposed indications of possession invoked by Niger in no way justify, in factual terms, the
course of the boundary claimed by the latter. It will do this by following, step by step, the
376
MN, pp. 81-82, para. 6.7.
37MN, pp. 80-81, para. 6.6. - 70 -
line described by Niger in paragraphs6.21 to 6.25 of its Memorial ⎯ whilst noting at the
outset the complexity of its course, which Niger’s Memorial hardly makes any easier to
understand 378. More precisely, it will show that Nige r’s line diverges considerably from that
defined in the 1927 Erratum as well as from the line featured on the 1960 map, which itself
fails to comply with the text of the Erratum, since it adds a considerable number of frontier
points which the latter does not indicate as su ch (see sketch-map No.4, Course of the
boundary after the Tao marker, reproduced again below).
37The recapitulation in paragraph 6.26 of Niger’s Memorial does not actually contain the same description
of the course as that given in the preceding paragraphs. - 72 -
A. “[from the Tao astronomic marker] the IGN line passes to the west of Petelkolé (the
village’s co-ordinates are 14°00'35.7"N, 00°24'52.6"E), which it leaves to
Niger . . . The frontier line follows the IGN line as far as the outskirts of Petelkolé. It
then deviates slightly to the west so as to meet the endpoint of the upgraded stretch of
the Téra-Dori road constructed by Nig er (co-ordinates: 140 °00'4.2"N,
00°24'16.3"E). It then rejoins the IGN line at the point having co-ordinates
13° 59' 39" N, 00° 25' 12" E.” (Memorial of Niger, p. 94.)
3.66. From this very first sector, Niger adopts a course which deviates, to its advantage,
from the straight-line course set out in the Erratu m, as well as from the line of the 1960 map.
Niger does not explain the reason for this devia tion. Nor does it justify the purpose of the
frontier crossing points it claims, for example, “the endpoint of the upgraded stretch of the
379
Téra-Dori road constructed by Niger”, which the 1927 Erratum does not mention .
3.67. This deviation is obviously to be expl ained by Niger’s wish to enclose within its
borders the village of Petelkolé, which however lies to the west, so on the Burkina side, of the
straight line laid down in the 1927 Erratum.
3.68. According to Niger, the fact that th e line it claims leaves this village to the east,
on Niger380side, is “in accordance with the ad ministrative information from the colonial
period” . However, this village is not listed in the Niger fascicle of the General List of
Localities of French West Africa at the time of the 1927 Arrêté. While it is not listed in the
Upper Volta fascicle either, it is indicated on381e 1954 sketch-map of Diagourou canton
(Niger) as a village “outside the canton” . Similarly, it is not mentioned in the Arrêté of
1 January 1956 establishing polling stations and di stricts for the elections to Niger’s National
382
Assembly . Furthermore, this village, which is featured twice on the 1960 map, is in both
cases located to the west of the line shown on it, hence on the Upper Volta side, so that Niger
is mistaken when it writes that “the IGN line passes to the west of Petelkolé...which it
383
leaves to Niger” .
384
3.69. Niger only puts forward two points to support the opposite view :
(i) first of all, it invokes “the Roser/Boy er agreement” of 1932, which, however, as is
385
known , intended to modify the boundary laid down by the 1927 Erratum and not
to interpret it; this document therefore does not indicate that the boundary passes to
the west of Petelkolé, as Niger claims, but that this is the “frontier in the area in
question [which] should then be defined as follows” in the “ further erratum”
37This new road was upgraded very recently. It does not correspond to the “motor road” mentioned in the
Erratum.
38MN, p. 94, para. 6.22.
381
MN, Anns., Series D, No. 21 (Diagourou canton: scale 1:250,000, 1954).
382
MN, Anns. SeriesB, No. 31 ( Arrêté No.2794/APA establishing polling stations and districts for the
elections to the National Assembly (Official Journal of Niger, No. 304, 1 January 1956).
38MN, pp. 93-94, para. 6.22.
38MN, p. 94, para. 6.22.
38See para. 3.33 above. - 73 -
386
proposed, which was never to be adopted ; incidentally, Administrator Delbos’s
sketch-map, for its part, placed Petelkolé to the west of the boundary proposed, thus
387
on the Upper Volta side ;
(ii) the 1953 Tour Report 388does not provide any further backing for Niger’s position.
Firstly, and once again, the points mentioned in this report were made on the basis of
the delimitation proposed by Administrator Delbos and not on that finally adopted in
the 1927 Erratum. Secondly, the information contained in this report is vague, since
at the time “the astronomic markers mentioned in the Arrêté could not be located”.
It should be recalled in this regard that the boundary referred to by what Niger calls
“the Roser/Boyer agreement” of 1932 does not arrive at the Tao astronomic marker,
as the Arrêté indicates, but at a “frontier marker situated 5.75km from the Tao
astronomic marker” 389. Finally, it is difficult to ascribe any meaning to the reference
made in the report to the boundary passi ng between “the permanent hamlets” of
Petelkarkalé and Petelkolé, since Petelk arkalé does not appear on the 1960 map.
Thus there is nothing to justify the enclaving of Petelkolé, which neither the Erratum
nor, for that matter, the line on the 1960 map allocates to Niger.
B. “The frontier then follows the IGN line, leaving Fetokarkale (Burkina Faso) to the
west. It then passes through a frontier point known as Baobab (13°58'38.9"N,
00° 26' 03.5" E), and through Tindiki (13° 57' 15.4.9" N, 00° 26' 23.6" E), as far as
the break in the line of crosses in the vicinity of Ihouchaltane (Oulsalta on the
1960 IGN map, Sebba sheet).” (Memorial of Niger, p. 94.)
3.70. Niger claims to go back to the line of the 1960 map from this point onwards. It
first of all expressly leaves the village of Fetokarkale to Bu rkina Faso (“[t]he
frontier . . . leav[es] Fetokarkale (Burkina Faso) to the west”), as does the straight line linking
the Tao marker to the River Sirba at Bossébangou. Niger also mentions two crossing points
(the “Baobab” and “Tindiki” 390 points) which, not being mentioned in the Erratum, cannot be
regarded as “frontier points”. Furthermore, these two names do not appear on the 1960 map.
3.71. At the level of the Oulsalta encamp ment, Niger itself states that the line on the
map breaks off, then resumes a little further al ong. If a straight line were to be drawn
between the two broken-off sections of the line on the map, following the methodology
391
adopted by Niger itself in other sectors , Oulsalta would very clearly be situated on the
Upper Volta side of the line. This is also where it is located if the straight-line course adopted
by the Erratum is followed.
386MN Anns., Series C, No. 45 (Letter No. 112 and Tour Report from Roser, Civil Service Deputy, Acting
Commander of Dori cercle, to the Governor of Upper Volta (Political Office) dated 10 April 1932, p. 6) (emphasis
added).
387MN, Anns., SeriesC, No.14 (sketc h-map prepared by Administrator Delbos of the route followed by
the Administrators of Dori and Tillabéry on a mission in June 1927 with a view to delimitation between Dori and
Tillabéry cercles).
388MN Anns., SeriesC, No.79 (Report of a tour conducted from 16 to 23 November 1953 by
Deputy-Administrator Lacroix (Tillabéry cercle), dated 24 December 1953).
389MN Anns., SeriesC, No.73 (Offici al telegram/letter No.70 from th e Head of Téra Subdivision to
Tillabéry cercle dated 11 July 1951).
390It should be noted that the co-ordinates of this latter point given by Niger situate it not on the frontier but
in Burkina, if the line on the 1960 map is followed as Niger suggests.
391
See MN, p. 100 in fine, para. 6.26: “From there . . . the IGN line, connecting the gaps between sections
with straight lines, as far as . . .” (emphasis added). - 74 -
C.“The frontier passes through a point situ ated on the river to the west of the
encampment, whose co-ordinates are 13°55' 36.4" N, 00° 27' 07.2" E . . . The
boundary passes through the point havi ng co-ordinates 13° 53' 12.8" N,
00°28'13.5"E located on the Kalsatouma-Sidibébé road. It then rejoins the IGN
line at the point having co-ordinates 13°53'24"N, 00°29'58"E.” (Memorial of
Niger, p. 95.)
3.72. Despite the above, Niger makes its line depart once again from the line on the
1960 map so as completely to enclave Oulsalta (and its two districts, Banguel Ndao and
Dongobe) to its advantage. Instead of using a straight-line section to bridge the gap between
the broken-off sections of the line on the map, Niger’s course veers off to the south-west, then
to the south, then to the south-east, and finally to the north-east, thus forming a curve which is
almost three times longer than the straight line which would have connected the sections
directly. Absolutely nothing justifies this course, except the wish to place the Oulsalta
encampment on Niger’s side of the boundary.
3.73. Niger does itself point out, however, that “its ownership has been disputed”; “[i]t
appears on the frontier according to the sketch-map prepared by Delbos in June 1927”, that is,
just before the Arrêté was adopted ⎯ in actual fact the encampment is drawn as an oval shape
392
straddling the proposed boundary . This can be explained by the presence of the Ossolo
Pool in the north-east of this sector. As th e people here are nomadic tribes, their location
depends less on their camping places than on the position of watering holes where they water
393
their herds . This may justify the fact that in this case the location of Oulsalta may have
overlapped the delimitation line. Similarly, it is difficult to draw the slightest conclusion in
terms of inter-colonial delimitation from the fact that tribes cultivated certain fields in the
frontier area ⎯ particularly when these fields had been raided, as was the case for the lougans
(fields)394of the Logomaten. It also appears that management of the Oudalan Touareg and
the Logomaten came under the authority of Uppe r Volta, before its dismemberment, and not 395
of Niger, as is shown by a letter from the Governor-General of FWA dated 3 January 1934 .
3.74. The fact that the encampment was placed east of the boundary proposed by “the
Roser/Boyer Agreement of April1932, cited by the authorities of Niger Colony on
396
24 May 1935 and 11 July 1951” is also without relevance, for the same reasons as those set
forth with regard to the village of Petelkol é. The situation of a place in terms of a
delimitation which has not been confirmed cannot be used to call into question the confirmed
delimitation.
39MN, pp.95-97, para.6.23 and p.95, fn.266 (empha sis added). See MN, Anns., SeriesC, No.14
(sketch-map prepared by Administrator Delbos of the rout e followed by the Administrators of Dori and Tillabéry
on a mission in June 1927 with a view to delimitation between Dori and Tillabéry cercles).
39The Commander of Dori cercle’s protests against the boundary adopted by the 1927 Arrêté focused in
particular on the fact that this left the watering places to Niger (see para .1.31 above and MN , Anns., SeriesC,
No. 21, p. 2). This applies in particular to the Ossolo Pool, which is explicitly mentioned in the amended Arrêté of
1927.
39Incidentally, the only document cited by Niger in this regard (MN, pp. 95-96, para. 6.23 and MN, Anns.,
Series C, No. 64) dates from 1941, a time when the Colony of Upper Volta had been dismembered. In this respect
also, this document has no relevance whatsoever to the present case.
39Annex CMBF No. 4 (Letter No.2 AP/2 from the Gover nor-General of FWA to the Governor of Niger,
dated 3 January 1934, on the attitude of the Dori and Tillabéry Touareg).
39MN, p. 95, para. 6.23. - 75 -
3.75. Finally, Niger asserts that this “locality is shown on the sketch-map of Diagourou
canton in 1954”. This is true, apart from one decisive nuance: the locality in question is not
underlined on this sketch-map, which according to the legend of the latter means that it is a
“village outside the canton” 397.
3.76. Niger does not submit any other points which might justify enclaving Oulsalta.
D. “From that point the boundary follows the 1960IGN line as far as the point having
co-ordinates 13° 52' 04" N, 00° 31' 00" E, where the area of Komanti encampments
(Kamanti or Comanti on certain documents) commences . . . The frontier marked on
the 1960IGN sheets is drawn with many gaps, to indicate that its course is
particularly problematic in this sector... From the point having co-ordinates
13°52'04"N, 0°31'00"E, where there is a break in the line of crosses on the
1960IGN map, the boundary passes thro ugh the point having co-ordinates
13°48'55"N, 0°30'23"E, then reaches the point with co-ordinates 13°46'31"N,
0°30'27"E. It then runs to the point with co-ordinates 13°46'18"N, 0°32'47"E
located to the north of Ouro Sabou on the tributary arm of the River Tyekol
Dyongoltol. The frontier then follows that tributary until its confluence with the
Tyekol Dyongoltol at the point with co-ordinates 13°46'51"N, 00°35'53"E; from
there, it follows the IGN line as far as th e point with co-ordinates 13°46'22.5"N,
0°37'25.9"E, located at the level of Bangaré on the River Folko...” (Memorial of
Niger, pp. 95-97.)
3.77. The same technique of enclaving territory is again used by Niger here, in an even
more blatant way. The course of its claimed boundary actually follows the line on the 1960
map only for a very short distance. Once again, instead of using straight lines to bridge the
gap in the line of crosses ⎯ or even, failing that, following the watercourses present in this
sector ⎯ Niger draws a line which makes off in the opposite direction: it heads in a
south-westerly direction, then turns towards the south, the east, and finally the north-east.
The triangle thus drawn forms a broad salient, about which the 1927 Erratum, once more,
does not say a word. And the line of this sa lient and the points through which it passes are
not given any justification or even explanation by Niger. Nor does Niger explain why it takes
no account of the crosses appearing on the 1960 map between the two sections of the line on
the map which Niger links via an enclave.
3.78. The only argument put forward by Niger to justify the claim that, as a result of
this salient, “the localities of Komanti, Kamanti, also called Ouro Toupé, Zongouweitan, also
called Kamanti Fété Tao, and Dingui-Dingui, also called Ouro Tanbella” belong to Niger is as
follows: these encampments have, it claims, been “administered by Niger since the colonial
period” 398. In other words, it is claiming that possession is equivalent to title ⎯ assuming
that such possession is proved, something which Niger entirely fails to do. There can be no
clearer indication of Niger’s attempt to disregar d the law applicable to the delimitation, as
expressly specified by the Parties in their 1987 Agreement.
397
MN, Anns., Series C, No. 21 (Diagourou canton: scale 1:250,000, 1954).
39MN, pp. 96-97, para. 6.23. - 76 -
3.79. Niger does add that “Zongowaetan is shown on the 1954sketch-map of
399
Diagourou canton” . Yet this encampment is shown, in 1954, to the north-east of Bangaré,
hence north-east of the line on the 1960 map. On no account is this encampment situated
within the broad salient which Niger allocates itself south-west of this line.
E. “At this point, the frontier line takes a clear south-west orientation. The co-ordinates
of the point where the frontier line changes direction are the following:
13° 46' 22.5" N, 00° 37' 25.9" E. To the south of Bangaré, the boundary returns to
the IGN line.” (Memorial of Niger, p. 98.)
3.80. On this short section Niger adopts the line on the 1960 map, which appears to
leave the village of Bangaré to Niger. Th e latter is described by Niger as a “large,
cosmopolitan village of over 1,000souls” 400. While Bangaré was known to the colonial
authorities in 1927, it is important to point out that at the time it was no more than a small
market, which only became a village in 1945. In his report date d 10August1954 on the
census of Diagourou canton, the Head of Téra Subdivision noted:
“ Recently formed villages
There are four of them . . . The fourth, Bangaré, was created by Mr. Garat
in1945. It is made up of several components; heads of families from the
Diagourou or Yagha tribes and heads of Mossi and Gourmantché families,
settled there around the little market of Bangaré, for the development of which
401
the village was created.”
3.81. According to Niger, the line on the 1960 map is justified, since “Bangaré has
always been located in the territory of Niger”. However, in this regard Niger is working on
the basis of post-1950 documents 402, as well as giving them implications which they do not
403
have . These documents are made even less co nvincing by the fact that they are
contradicted by documents which are contemporaneous with the amended Arrêté of 1927.
Niger acknowledges as much: “On the sketch-map draw404y Prudon in 1927, Bangaré was,
however, shown in Upper Volta territory[ ]. In the sketch-map s prepared by Delbos
in June[ 405] and August 1927[ 406], this name appears on the boundary.” 407
399MN, p. 96, para. 6.23 and MN Anns., Series D, No. 21 (Diagourou canton: scale 1:250,000, 1954).
400
MN, p. 98, para. 6. 24.
401
MN, Anns., SeriesC, No.84 (Report from the Head of Téra Subdivision on the census of Diagourou
canton, dated 10 August 1954), p. 9.
402
MN, pp. 97-98, para. 6.24.
403For example, the 1953 Tour Report does not, contrary to what Niger claims, state that “the frontier
passes through it [this village]” (MN, p.97, para.6.24).It indicates more vaguely:“There are no distinctive
geographical features in the area between Bangaré and Tao. The ‘ frontier area’ first passes through Bangaré, to
the north-west of Ousaltan, across a plain with extensive lateritic areas…” (MN, Anns., Series C, No. 79, Report of
a tour conducted from 16 to 23 November 1953 by Deputy-AdministratorLacroix (Tillabéry cercle), dated
24 December 1953, p. 2) (emphasis added).
404
MN, Anns., Series D, No. 3. (Tillabéry cercle 1:200,000 sketch-map prepared by Administrator Prudon,
June 1927).
405
MN, Anns., Series C, No. 3 (1:500,000 sketch-map representing Say cercle, no title, author or date).
406
MN, Anns., Series C, No. 16 (Letter from Delbos, Commander of Dori cercle, to the Governor of Upper
Volta, dated 27 August 1927). - 77 -
3.82. It may also be added that the “List of Niger cantons and villages forwarded to the
Minister for Overseas France in 1948” does not include Bangaré ⎯ any more than it includes
408
Oulsaltan or Petelkolé ⎯ among the villages in Diagourou c409on . The same applies to the
list of villages in Téra Subdivision dated 6 July 1933 .
410
3.83. In these circumstances , it is clear that on the date of adoption of the amended
Arrêté of 1927, Bangaré was regarded as having to fall within the Colony of Upper Volta.
This is the solution adopted by the 1927 Erratum, as it does not indicate any intermediate
point between the Tao marker and the River Si rba at Bossébangou. There is no reason to
replace it with another line, whether this is the line on the 1960 map, which diverges from it,
or the one claimed by Niger, which is not based on any relevant factors.
F.“Following the watercourses, where there are no crosses, it passes between
Kolangoldagabé, in Burkina Faso (co-ordinates 13° 43' 52.3" N, 00° 36' 14.5" E) and
Lolnan[g]o, in Niger (co-ordinates 13°43'50.3"N, 00°36'49.0"E).” (Memorial of
Niger, p. 98.)
3.84. The course defended by Niger in this sector follows the very intermittent line on
the 1960 map, filling in the gaps411t with strai ght lines but by “[f]ollowing the watercourses,
where there are no crosses” , which is especially curious because in this sector such
watercourses do not always exist between two broken-off sections of the line. Furthermore,
that does not in any case correspond to the c ourse claimed by Niger, which proceeds by
means of artificial lines. Here, Niger lays claim to the villages of Lolnando, Kolmangol Nore
Ole and Pate Bolga 412, once again without basing this claim on a single document.
G.“The frontier then passes through the locality of Sénobellabé (geographical
co-ordinates: 13° 36' 52.6" N, 00° 50' 00.8" E).” (Memorial of Niger, p. 98.)
3.85. The locality of Sénobellabé constitutes another difficulty for Niger. The line on
the 1960 map actually places this locality on Niger’s side, even though it was indisputably
acknowledged as being on the Upper Volta side of the inter-colonial boundary when the latter
was adopted in 1927. Niger expressly concedes as much: this “crop-growing area”
“was regarded as belonging to Upper Vo lta by the Roser/Boyer Agreement of
April 1932, in reliance on the Delbos line of 1927. This view was confirmed by
the tour report of the Head of Té ra subdivision dated 8November1933,
forwarded to the Governor of Niger by the Commander of Tillabéry cercle. The
same view was taken in the Record of Agreement between the Administrators of
Dori and Téra of 25 April 1935. The result was a transfer of population of Niger
40MN, p. 97, para. 6.24.
40MN, Anns., Series C, No. 71 (List of Niger cantons and villages forwarded to the Minister for Overseas
France (Diagourou, Tamou and Torodi cantons), undated, 1948).
409
MN, Anns., Series C, No. 50 (List of villages in Téra Subdivision, 6 July 1933).
410
See also para. 3.36 above.
41MN, p. 98, para. 6.24.
41MN, p. 98, para. 6.24 in fine. - 78 -
origin to the localities of Taka and Yolo, situated in Niger territory. The same
happened with the Record of Agreement of 8 December 1943.” 413
3.86. In such circumstances, it is hard to understand how Niger could possibly claim a
line passing to the west of this locality. However, this is what Niger actually does . . . simply
by moving the village, both in space and time! According to Niger,
“these are farming hamlets, which do not remain in the same place. The sites
change according to the seasons and reta in the same toponyms. Before 1960,
Sénobellabé was further north. Today, the former site has been abandoned and
the hamlets which continue to bear that name are to be found on the Burkina side
414
of the IGN line.”
3.87. This “reasoning” calls for three comments to be made:
(i)Niger is acknowledging here that the “method” which consists of trying to
rediscover “actual boundaries... in prac tice” is meaningless in a region where
localities “change according to the seasons a nd retain the same toponyms”; Burkina
takes note of this;
(ii) this “method” is applied by Niger in a most curious way: it relies on the site of the
locality today, not on its site in 1927;
(iii) whilst this locality was considered to belong to Upper Volta in 1927, this was not on
the basis of the site where it is today, but according to its 1927 site. As a result, the
line claimed by Niger which passes west of the location occupied by Sénobellabé in
1927 is unfounded, since it passes to the west of a locality which at the relevant
period was indisputably on the Upper Volta side and hence on the other side of the
inter-colonial boundary. The straight-line boundary coming from the Tao marker for
its part leaves the village of Sénobellabé to Burkina.
3.88. Niger relies on the fact that this locality is still listed, it claims, amongst the
villages of Diagourou canton (Niger) in 1933 and 1948 415. This locality does not however
416
appear on the 1954 sketch-map of Diagourou canton . Furthermore, there were disputes
between the two colonies as to which one it belonged to, up until at least 1932 417. And as
Niger recalls, in all cases, these disputes led to the reaffirmation that Sénobellabé was part of
Upper Volta territory.
3.89. Once again, examination of the factual elements produced by Niger thus leads
one to dismiss as entirely without foundation the course it claims, as well as the line on the
1960 map in this sector. On the other hand, these elements fully corroborate the
interpretation, endorsed by Burkina, which th e colonial authorities gave to the Erratum,
413
MN, p.98, para.6.25. See also the Record of Agreement of 25 April 1935, MN, Anns., SeriesC,
No. 57, which also allocates this locality to Dori cercle.
414
MN, p. 99, para. 6.25.
41MN, p. 100, para. 6.25.
41MN, Anns., Series D, No. 21 (Diagourou canton: scale 1:250,000, 1954).
41MN Anns., SeriesC, No.46 ( Bulletin de renseignements politiques of Tillabéry cercle dated
11 October 1932). - 79 -
according to which the latter adopted a course in two straight-line sections between the
418
Tong-Tong marker, the Tao marker and the River Sirba at Bossébangou .
H.“The IGN line meets the line which at the time constituted the boundary of Say
(tripoint for the cantons of Tillabéry, Dori and Say) at the point with co-ordinates
13° 29' 08" N, 01° 01' 00" E.” (Memorial of Niger, p. 99.)
3.90. This meeting “point” is purely arbitrary in three ways. As has been recalled, the
amended Arrêté of 1927 makes no reference to it. Fu rthermore, the 1960 map does not show
“the line which at the time constituted the bounda ry of Say”. Lastly, Niger does not explain
what method it selected in order to situate the point in question at the co-ordinates that it
gives. As will be established in the next chapter, the last section of the boundary in this sector
does not stop at the so-called “tripoint” mentioned by Niger, but at the point where the
boundary reaches the River Sirba at Bossébangou, according to the unequivocal wording of
the 1927 Erratum.
*
3.91. In conclusion, it is clear that the course of the boundary claimed by Niger departs
from the text of the 1927 Erratum even when the la tter is entirely sufficient. It also departs
from the line on the 1960 map ⎯ which does not comply in this sector with the Erratum. The
fact that Niger has decided to disregard both the course made up of straight lines in the Arrêté
and the line on the 1960 map (which it nonetheless claims to follow, contrary to all reason) is
a blatant infringement of the law applicable to the Parties. Such a strategy is perhaps aimed at
obtaining from the Court a boundary which, by way of compromise, would follow the line on
the map which runs between the line claimed by Niger and the straight line connecting the
Tao marker to the River Sirba at Bossébangou. Burkina is confident that the Court will
determine its course in accordan ce with the agreement of the Pa rties laid down in the 1987
Agreement and reaffirmed in the Special Agreement seising the Court: as a considerable
number of colonial administrators noted, some with regret, the inter-colonial boundary
decided upon in 1927 took the form of artific ial lines linking the frontier points expressly
designated in the Arrêté via straight-line sections. This constitutes the frontier between
Burkina and Niger in this sector, as Niger had furthermore agreed on various occasions before
its sudden change of mind in 1990.
3.92. As a consequence, on the basis of the above findings, Burkina concludes,
⎯ firstly, that the course of the frontier describ ed by Niger in the part of its Memorial
relating to the “Téra sector” is devoid of any legal or factual foundation;
418
See in particular para. 3.2 above. - 80 -
⎯ secondly, that the course of the frontier in this sector is that described on pages 158 and
160 of Burkina’s Memorial: from the Tong-Tong astronomic marker, the frontier
follows a straight line as far as the Tao astronomic marker (Lat.: 14º 03' 04.7" N; Long.:
0º 22' 51.8" E) 419 and then, from that point, a straight line up to the point
where the frontier reaches the River Sirba at Bossébangou (Lat.: 13°21'06.5"N;
Long. 1° 17' 11.0" E) 420.
419
The co-ordinates of this point were measured by GPS by Burkina. The co-ordinates of this marker on
the Clarke 1880 ellipsoid are: Lat. 14º 03' 13'' N; Long. 00 22' 53'' E.
420
The co-ordinates of this point are given on the Clarke 1880 ellipsoid. C HAPTER IV
THE COURSE OF THE FRONTIER IN THE “S AY SECTOR ”
4.1. In the sector presented by Niger as “the Say sector”, which corresponds to the
section of the frontier separating the two States at the level of the Yagha, Komandjari and
Tapoa départements on the Burkina side and the départements of Téra (wit421egard to the
section of the frontier reaching the River Sirba at Bossébangou) and Say on its own side,
Niger only appears to acknowledge the f act that the frontier is defined by Arrêté No. 2336 of
the Governor-General of FWA, dated 31 August1927, as amended by the Erratum of
5October1927, which states that the frontier “reach[es] the River Sirba at Bossebangou. It
almost immediately turns back up towards the nor th-west, leaving to Niger, on the left bank
of that river, a salient which includes the villages of Alfassi, Kouro, Tokalan, and Tankouro;
then, turning back to the south, it again cuts the Sirba at the level of the Say parallel”. Niger
does indeed affirm that the 1927 Arrêté “remained, at the time when the two States became 422
independent, the only reference text for th e determination of their common frontier” , and
that the use of this text for the determination of the frontier is “in accordance with the terms
423
of the 2009 Special Agreement and of the 1987 Agreement between the two States” .
However, Niger immediately contradicts this affirmation ⎯ with which Burkina is in full
agreement ⎯ by stating that whilst the text of the Erratum “will constitute the primary basis
for determination of the course of the frontier between the two States in this second
sector” 424, the existence of “parts of that text [which] are problematic” and the finding that the
1960 map “in part deviates markedly from that d escribed in the Erratum” gives it reasons for
“not following it in certain respects” 425.
4.2. In actual fact, the line claimed by Ni ger has nothing in common with that laid
down in the text which is nonetheless recognized as the “reference text”:
⎯ whereas the text indicates that after cro ssing the Téra-Dori motor road at the Tao
astronomic marker, the frontier “reach[es] the River Sirba at Bossebangou”, Niger claims
that it follows an entirely different course, stopping around 30kms north of the River
Sirba426;
⎯ whereas according to the text, the frontier, having reached the River Sirba at
Bossébangou, “almost immediately turns back up towards the north-west, leaving to
Niger, on the left bank of that river, a salient which includes the villages of Alfassi,
Kouro, Tokalan and Tankouro”, according to Niger, on the other hand, “the frontier
427
cannot create a salient in this area” and instead of turning back up towards the
north-west, “[i]t simply turns in a south-westerly direction” 428;
42See MBF, p. 13, sketch-map No. 4 ⎯ Administrative divisions adjoining the frontier.
42MN, p. 104, para. 7.12.
42Ibid.
424
MN, p. 105, para. 7.12.
425
Ibid.
42MN, p. 111, para. 7.23.
42MN, p. 112, para. 7.26.
428
MN, p.112, para. 7.26. - 82 -
⎯ whereas the Erratum states that the frontier, “t urning back to the south, … again cuts the
Sirba at the level of the Say parallel”, the text should, according to Niger, be rewritten in
such a way that the frontier moves in a south-westerly direction so as to arrive
429
“approximately [at] the Sirba at the level of the Say parallel” ⎯ moreover without
cutting it “again”, as it would not have already reached it at Bossébangou as required by
the Erratum;
⎯ whereas the text indicates that the frontier “following an east-south-east direction,
continues in a straight line up to a point located 1,200m to the west of the village of
Tchenguiliba”, Niger “maintains its claim here to a frontier in two straigh430ine sections,
as it appears on those maps and sketch-maps of the colonial period” .
4.3. The submissions thus presented by Ni ger are incompatible with the principle
accepted without reservation by the Parties and laid down in Article6 of the Special
Agreement seising the Court, according to which the frontier between the two States is
defined by international law, including “the principle of the intangibility of boundaries
inherited from colonization . . . and the Agreement of 28 March 1987”. This “applicable law”
actually leads one to disregard the documents on which Niger bases its arguments with a view
to establishing that the course of the frontier is not that described in the Erratum, for at least
four cumulative reasons.
4.4. Firstly, the application of the uti possidetis, in so far as the definition of the frontier
is concerned, requires reference to be made to the title constituted by the Arrêté of
31August1927 as amended by the Erratum of 5October1927 ⎯ recognized moreover by 431
Niger as “the only reference text” at the tim e when the two States gained independence .
Therefore it is entirely immaterial whether the c ourse of the frontier described by this text is
contradicted by that suggested by certain non -regulatory texts, or whether some maps and
sketch-maps show a different line. And the fact that the regulatory line may have thwarted
the expectations of certain officials is also of scant importance. At most, the documents and
facts thus illustrated could be categorized as effectivités ⎯ something that Niger prudently
refrains from doing ⎯ which are at odds with the title. However, it is well established that
when applying the uti possidetis principle, such effectivités have to be disregarded in favour
432
of title .
4.5. Secondly, Article 1 of the 1987 Agreement refers exclusively to the Arrêté and its
Erratum as far as the description of the frontier is concerned, and in so doing rules out any
possibility of disregarding it, particularly when it comes to the course followed in the Say
sector, on the basis of any other document. Sixty years on, the 1987 Agreement thus upholds
the 1927 title through an express agreement of the tw o Parties to this dispute, 27years after
they gained independence.
4.6. Thirdly, Article 2 of the 1987 Agre ement, concerning not the definition of the
frontier but rather its demarcation, indicates th at this should be done “following the course
described” by the Arrêté and its Erratum, and only mentions the 1960 map “and/or any other
relevant document accepted by joint agreement of the Parties” in order to mitigate any
42MN, p. 115, para. 7.32 (emphasis added).
430
MN, p.120, para. 7.40 (emphasis added).
431
See paras. 1.48-1.50 above.
43See paras. 1.51-1.64 above. - 83 -
instances where the description might “ not suffice”, not in order to contradict it. This means
that where the text “suffices” to provide a clear description of the course of the frontier, there
can be no question of referring to any other description arising from a map or contained in
433
another document with a view to amending that course .
4.7. Finally, the documents referred to by Niger must also be disregarded, as Article 2
of the 1987 Agreement stipulates that the only doc uments which may be used to establish the
course of the frontier are those which have been “accepted by joint agreement of the Parties”,
something which canno434e said for any of the documents used by Niger to support its
position in this area .
4.8. Niger is therefore misguided in believing that it can base its entire position on
documentary material which the Parties rejected as inapplicable and irrelevant when they set
out the applicable law. In law, it is indisputable that the line constituting the frontier between
the two States is that described by the Arrêté as amended by its Erratum, supplemented
alternatively, and only where it does not suffice, by the line shown on the 1960 map.
However, the line claimed by Niger is based ne ither on the description in the Erratum, which
is systematically disregarded, nor on the 1960 map. Niger’s position is therefore flawed from
the outset by an error in law which affects its entire line of argument 43.
4.9. These observations by themselves suffice to respond to Niger’s arguments and to
refute in its entirety the frontier line it claims in the Say sector. Burkina thus remains firmly
convinced that the course of the frontier betw een the Parties is that which results from the
amended Arrêté of 1927, as it has established in its Memorial 436.
4.10. Moreover, a detailed analysis of Niger’s reasoning reveals that, even were it to be
conceded that the course of the frontier coul d be determined by taking into account the
documentation produced by Niger ⎯ quod non, contrary to the latter’s assertions:
⎯ the frontier reaches the River Sirba at Bossébangou (Section 1);
⎯ the course claimed by Niger in the sector of the salient is entirely unjustified (Section 2);
⎯ the frontier between the intersection of th e River Sirba and the Say parallel and the
beginning of the Botou bend consists of a single straight line (Section 3).
S ECTION 1
T HE FRONTIER REACHES THE R IVER SIRBA AT BOSSÉBANGOU
4.11. In the Bossébangou area, Niger’s approach openly consists of giving a supposed
“traditional boundary” of cercles precedence over the inter-colonial boundary established by
the Erratum. As a result, the frontier would not be able to reach the River Sirba at
43Ibid.
434
Ibid.
43See Chap. I above.
43MBF, Submissions, pp. 160-162. - 84 -
Bossébangou, stopping instead around 30km north of this locality, the section of line
continuing to the River Sirba at Bossébangou being one of Niger’s internal boundaries,
separating the former cercles of Say and Tillabéry.
4.12. Four preliminary observations are called for.
4.13. First of all, it is important to point out that although this position was previously
defended by Niger at the extraordinary m437ing of the Joint Technical Commission on
Demarcation held on 14May 1990 , it was swiftly abandoned as untenable. At the end of
July1990, Niger accepted that “the line of the frontier reaches the River Sirba at
438
Bossébangou” .
4.14. Secondly, Niger’s position lacks consis tency and varies from page to page of the
Memorial. It is first stated, on page65, that “in the area of Bossébangou,... [the line
included] a part of the internal boundaries of Say cercle”; then, on page107, we read that
Bossébangou is apparently not in Say cercle , but on the boundary between the cercles of
Tillabéry (including the cantons of Dori cercle incorporated into Niger in 1926) and Say,
which would imply that the contested boundary does not constitute an incursion into Say
cercle, but separates Say and Tillabéry cercles.
4.15. This indecision on the part of Niger ma kes its argument rather difficult to grasp.
Only once does Niger suggest that, by desce nding as far as the River Sirba at Bossébangou,
the frontier resulting from the Erratum actually enters the cercles of Say or Tillabéry. In
contrast, it makes more consistent criticism of this frontier for descending too far to the south
and thus mistakenly following (Niger’s internal) boundary between Say and Tillabéry cercles
(thus without actually entering either of them) for some 30 km within Niger’s territory. It is
this second argument which should probably be taken as reflecting Niger’s position.
4.16. Thirdly, it is not the Erratum but Niger itself which confuses the inter-colonial
boundary with internal boundaries within Niger, when it completely ignores the fact that
Upper Volta had disappeared from 1932 to 1947. For example, to demonstrate its idea that
“the inter-col onial boundary in this sector was also very strongly maintained after 1927” and
that this boundary departs from the text of the Erratum 43, it invokes a Record of Agreement
of 8 December 1943 440, even though this document can have absolutely nothing to do with
the inter-colonial boundary between Niger a nd Upper Volta, since the latter had been
dissolved and dismembered in 1932 and remained so in 1943 441.
43Ann. MBF 85.
438
Ann. MBF87 (Report of the second ordinary session of the Joint Technical Commission on
Demarcation of the Frontier, 23-28 July 1990, 28 July 1990), p. 3 in fine. See also para. 2.17 above.
439
MN, p. 109, para. 7.20.
44MN, p. 110, para. 7.20.
44See paras. 3.57-3.58 above. - 85 -
4.17. Fourthly, Niger refers to maps a nd sketch-maps which are irrelevant for the
purpose of determining the colonial frontier. Three of the maps put forward by Niger to
support its argument date from th e period between 1932 and 1947 442, when Upper Volta no
longer existed. Furthermore, one of them, the title of which, according to Niger, is “road map
443
of the Colony of Upper Volta to a scale of 1:1,000,000”, dating from 1936 , does not exist,
since this is actually the road map of the Colony of Niger in its 1936 edition.
4.18. On the basis of these imprecise assertions in particular, and many others which
will be mentioned below, Niger’s contention that the Erratum contains a “mistake” hinges on
the following reasoning: despite the clear term s of the Erratum, from which it follows that,
coming from the north-west, the frontier “reaches the River Sirba at Bossébangou”, this line
is said to be entirely unjustified, as it calls into question the “traditional course of the
boundaries” of Say cercle 444. As far as Niger is concerned, the Arrêté of August 1927 was
prepared on the basis of the Lefilliatre-Choteau Record of Agreement on delimitation, which
sets out all the boundaries of Say cercle without restricting itself to those which separate this
cercle from Upper Volta 445; the Erratum only partially corrected this mistake, since it did not
take account of the fact that some of the Dori446 cantons which bordered Say cercle had also
been incorporated into Niger in 1926 . Consequently, by stipulating that the frontier
reaches the River Sirba at Bossébangou, it unwittingly sends it inside the territory of Niger 447.
However, according to Niger, the Erratum coul d not have defined the inter-colonial boundary
without following the pre-existing outlines of the cercles, since it is an implementing text of
448
the 1926 Decree , which by incorporating the Dori cercle cantons and Say cercle into Niger
automatically had the result that the new inte r-colonial frontier followed their “traditional”
boundaries. In Niger’s view, this mistake means that the 1927 Arrêté and its Erratum have no
effect in the Bossébangou region. Furthermore, Niger claims, documents dating from the
colonial period bear this out, in particular by not reproducing this “mistake”, and as far as the
1960 map is concerned, it should be disregarded, since it too is erroneous. As a result, Niger
directs its efforts towards rediscovering “traditional boundaries” and claims that the
inter-colonial frontier should be established in accordance with the lines they follow.
4.19. This argument collapses by itself, since the Erratum does not contain any
mistake (1) and, in addition, the documents on which Niger bases its argument do not support
the line that it proposes (2).
442
MN, p. 107, para.7.17; these are the 1:2,500,000 road maps of the Colony of Niger, 1934edition
(MN, Anns., Series D, No. 16) and 1936 edition (MN, Anns., Series D, No. 17) and the map Niger entitles “road
map of the Colony of Upper Volta to a scale of 1:1,000,000, 1936 edition, FWA Geographical Department, Dakar”
on p. 108, para. 7.17 of its Memorial (and refers to as MN, Anns. Series D, No. 17).
443MN, p. 108, para. 7.17.
444
MN, p. 110, para. 7.21.
445
MN, pp. 106-107, paras. 7.15-7.16.
446Ibid.
447MN, p. 107, para. 7.16.
448
MN, p. 111, para. 7.22. - 86 -
1. There is no mistake in the Erratum on the point where the frontier
reaches the River Sirba at Bossébangou
4.20. It is indeed the case th at, in its initial version, the Arrêté of August 1927
contained too many details regarding the boundaries of Say cercle, since even though it stated
its aim as determining the frontier between Nige r and Upper Volta, some parts of the text
established purely internal boundaries within Niger, in particular those of Say cercle in the
part which did not adjoin Upper Volta. One of the Erratum’s specific aims was to correct it
on that point (it also clarified certain aspects of the course of the frontier). Nonetheless,
according to Niger, despite that correction, the part of the frontier described by the Erratum as
reaching the River Sirba at Bossébangou remained a purely internal boundary within Niger.
A mistake therefore persisted, in Niger’s view, due to the fact that the Erratum had not
properly taken account of the fact that some cantons of Dori cercle had been transferred to
Niger at the same time as Say cercle 44.
4.21. The theory that a mistake wa s made is, however, untenable: the Arrêté and the
Erratum fully and explicitly took account of the transfer to Niger not only of Say cercle but
also of certain cantons of Dori cercle.
4.22. To try to convince the Court of the contrary, Niger firstly states that the
Lefilliatre-Choteau Record of Agreement is the source of the drafting error concealed within
450
the August 1927 Arrêté . This appears to be an established fact, since the terms of the
Record of Agreement of 10 February 1927 are identical to those of Article 1, paragraph 2, of
the Arrêté of August1927 (apart from the section on Botou canton) 451. It is therefore
probably because it reproduced the text signed by Lefilliatre and Choteau, which did not have
the exclusive aim of establishing the boundary between Say cercle and Upper Volta but
indicated all the boundaries of territories making up Say cercle, that the Arrêté went beyond
its own objective of determining the inter-colonial boundary between Niger and Upper Volta.
4.23. However, as Niger sees it, this mist ake was only partially corrected by the
Erratum, since the latter, in Niger’s view, was drafted without account being taken of the fact
that some Dori cantons had also been transferred to Niger, which would have meant that
452
Bossébangou would no longer be on the boundary with Upper Volta .
4.24. Yet this is to forget that the Arrêté of 31 August 1927 drew 453 only on the
Lefilliatre-Choteau Record of Agreement of 10 February 1927 , but also on the
Lefilliatre-Brévié Record of Agreement of 2 February 1927 454. The letter submitting the draft
Arrêté, sent in July 1927 to the Governor-General by the acting Director of Political and
Administrative Affairs of the Standing Committee of the Government Council, leaves no
room for doubt in this regard, as it states that the draft Arrêté “has been established in
44MN, p. 107, para. 7.16.
450
MN, p. 106, para. 7.15.
451
This is also confirmed by the letter of July 1927 submitting the draft Arrêté determining the boundaries
of the Colonies of Upper Volta and Niger, from the Director of Political and Administrative Affairs of the
Standing Committee of the Government C ouncil, which indicates that the draftArrêté was drawn up in
accordance, inter alia, with that Record of Agreement (CMBF Ann. 2).
452
MN, p. 107, para. 7.16.
45MN, Anns., Series C, No. 8.
45MN, Anns., Series C, No. 7. - 87 -
accordance with the following Records of Agreement: the first dated 2February1927,
determining the boundaries of the new Tillabéry cercle with Upper Volta; the second dated
10 February 1927, fixing the boundaries of Say cercle and Upper Volta; and the third dated
9May1927, indicating the boundaries of Botou canton with that same Colony” 455. The
Lefilliatre-Brévié Record of Agreement does expressly take account of the transfer of Dori
cantons to Niger, since its specific objective is to define the new boundaries between Niger
and Upper Volta which arise as a result. It establishes, having explicit regard to the Decree
dated 28 December 1926, that “[t]he cantons belonging to the former Tillabéry cercle on
22June[1910] shall be incorporated into the Colony of Niger. These cantons . . . are
bounded [to the west by the line which, coming from the north] reaches the River Sirba
(boundary of Say cercle) near to and to the south of Boulkalo.”
4.25. Moreover, it suffices to read Article1, paragraph1, of the Arrêté of
456
31 August 1927 to see that it exactly reproduces the terms of that Record of Agreement,
and therefore takes due account of the new boundaries between Tillabéry cercle and Upper
Volta resulting from the transfer to Niger of certain Dori cantons, specifying that:
“this boundary... descend[s] in a north-south direction, cutting the Téra-Dori
motor road to the west of the Ossolo Pool, until it reaches the River Sirba
(boundary of Say cercle), near to and to the south of Boulkalo”.
4.26. Consequently, when drafting the Arrêté of 31August1927, the colonial
authorities could in no way have made the mistake of forgetting that, outside Say cercle, some
cantons from Dori cercle had been incorporated into Niger by the Decree of
28 December 1926, since this Arrêté defines the new boundary between Upper Volta and
Tillabéry cercle, which henceforth contains them 457. The same applies, a fortiori, to the
Erratum.
4.27. Furthermore, all the Erratum did was to rationalize the course of the inter-colonial
boundary, by explicitly making the point located on the River Sirba near to and to the south of
Boulkalo coincide with Bossébangou, a point thus referred to as a “tripoint” in the
terminology used by Niger. It thereby clarified the boundaries of Tillabéry and Say cercles at
the points where they merge with the inter-c olonial boundary with Upper Volta, describing
the form taken by the salient from Bossébangou onwards ⎯ something the Arrêté did not do
in its initial version.
4.28. In so doing, it is also in order to show unequivocally that the boundary coming
from the Tao marker is inter-colonial when it r eaches the River Sirba at Bossébangou that the
Erratum states that, having reached that point, the frontier does not promptly turn back up to
form the salient, but “almost immediately” turns back up. In this way, having descended
from the Tao marker to the River Sirba at Bossébangou, the frontier briefly runs in a westerly
direction before turning back up towards the north-west, which as clearly as possible prevents
the line that “ascends” towards the north-west to form the salient from overlapping with the
455
CMBF Ann.2. This letter of su bmission is moreover probably the s ource of the mistake made by the
Arrêté, as it implies that the Record of Agreement of 10 February 1927 fixes the boundary between Say cercle and
Upper Volta, whereas it actually fixes all the boundaries of Say cercle.
456
MBF, Ann. 34.
45The Dori cantons which were transferred to Niger by th e Decree of 28 December 1926 are those which
belonged to Niger before 22 June 1910, the date of the Arrêté allocating them to Upper Volta. These cantons were
incorporated into Tillabéry cercle by an Arrêté of 22 January 1927. - 88 -
line “descending” from the north-west from the Tao marker. This ensures that by arriving
from the north to reach the River Sirba at Bossébangou, the inter-colonial frontier separates
Upper Volta (to the west) from the territory of Niger corresponding to Tillabéry cercle (to the
east), while by almost immediately turning back up to the north-west to form the salient, the
frontier separates Upper Volta (to the north) from the territory of Niger comprising Say cercle
(to the south). The next map illustrates the cour se of the frontier in this area (see sketch-map
No. 6 below ⎯ Course of the boundary in the area of the salient).
4.29. There is therefore no mistake in the Erratum; rather, it was drafted in full
knowledge of the fact that the boundary whic h it establishes separates Upper Volta from
Niger, including both Say cercle and Tillabéry cercle. Moreover, the “traditional course”
claimed by Niger is entirely without justification.
2. The documents invoked by Niger to contradict the terms of the Arrêté of
31 August 1927 as amended by its Erratum do not support
the course it claims
458
4.30. As Burkina has already pointed out , Niger, relying on various documents,
contends that the Erratum is erroneous in that it does not comply with “the traditional course
459
of the boundaries of Say cercle” . In order to establish the said “traditional course”, Niger
calls on a variety of documents. As Burkina has also pointed out already, Niger cannot, in
law, call upon these documents in order to contradict the title 460. But even assuming that it
were able to do so, these documents ⎯ which Burkina in no way recognizes as having been
“accepted by joint agreement of the Parties” under the terms of the 1987 Agreement and
which it will only mention for the purposes of the discussion ⎯ would not in any way support
the course claimed by Niger, whether they date from before or after the Erratum.
458
See para. 4.18 above.
45MN, p. 110, para. 7.21.
46See paras. 1.51-1.64 and 4.3-4.18 above. - 90 -
A. Documents prior to the Erratum
4.31. Niger firstly seeks to demonstrate that all the documents prio r to the adoption of
the Erratum take the same approach and already confirm a “view of the course of the
461
inter-colonial boundary” which would be very strongly maintained after 1927 . But in this
respect it refers only to the work done by Delbos 462 and Prudon 463, conveniently ignoring
other texts from the same period which show that, contrary to its own assertions,
Bossébangou was considered to be a locality on the boundary of Say and Dori cercles. This
is especially the case with the Choteau -Lefilliatre Record of Agreement of
464
10 February 1927 , which establishes the boundaries of Say cercle and attests that
Bossébangou was considered to be on the boundary of Say and Dori cercles before the
adoption of the Erratum.
4.32. Niger also relies on four sketch-maps and maps and suggests that these show that
before 1927 the boundary between Upper Volta and Niger, or, to be more precise, between
Say cercle and Dori cercle , never descended as far as Bossébangou 465.
4.33. Nevertheless, precisely the opposite is shown by three of the four documents
466
mentioned ⎯ Commander Truchard’s 1915 sketch-map , map No.60 of the 1926 Atlas of
Cercles 467and the 1926 Blondel-La Rougery map 468. Apart from the fact that none of these
sketch-maps and maps shows the boundary between Upper Volta and Niger on the right-hand
side of the River Niger, since neither in 1915 nor in 1926 did Niger possess territory on the
right bank of the river, it can be seen from these documents that the boundary between Dori
and Say cercles does indeed “descend” as far as Bossébangou. Moreover, they could not
make a “tripoint” appear between Dori, Say and Tillabéry cercles, since the cantons of
Tillabéry cercle situated on the right bank of the rive r were included within the territory of
Dori cercle at the time.
469
4.34. As for Captain Boutiq’s 1909 sketch-map , this could not be seen as an
illustration of the inter-colonial frontier either, since the three cercles (Say, Dori and
Tillabéry) belonged at that time to the single colony of Upper Senegal and Niger. It is also
the only one out of all the documents submitted by Niger to feature a “tripoint” between the
cercles of Dori, Tillabéry and Say to the north-west of Bossébangou. It is all the more
unreliable because:
⎯ the cercle boundaries illustrated on this sketch-map, which accompanies a “report of
CaptainBoutiq, Commander of Djerma cercle, on the possible transition of the military
régime to a civil one for the right bank of the Niger” and illustrates “the course of the
461
MN, p. 109, para. 7.20.
46MN, p. 108, paras. 7.19-7.20.
46MN, p. 109, para. 7.19.
46MN, Anns., Series C, No. 8.
465
MN, p. 110, para. 7.21.
466
MN, Anns., Series D, No. 4.
46MN, Anns., Series D, No. 6.
46MN, Anns., Series D, No. 9.
469
MN, Anns., Series D, No. 1. - 91 -
Niger through Djerma cercle” hardly amount to anything more than the boundaries
proposed by the Commander of Djerma cercle and only represent his view of the area in
1909; besides,
⎯ other proposals regarding the borders of cercles in the region were to be made before the
470
end of 1927, as is shown for example by Administrator Delbos’s sketch-map of 1927 .
On this sketch-map, it can be clearly seen that the borders of Say are presented differently
according to whether they represent the view of the Colony of Niger (in red) or that of the
Colony of Upper Volta (in black); moreover, the second sketch-map produced by
Captain Delbos and appended to his repor t of 17 December 1927 clearly shows the
boundary descending south of Nabambori and as far as the River Sirba 471; finally,
⎯ despite claiming that Captain Boutiq’s map illustrates the “traditional boundary” of Say
cercle, Niger disregards it totally as far as the southern boundary of this cercle is
concerned: whereas the 1909 map shows it as a broken line with an angle pointing
towards the north, Niger claims the exact oppos ite, that is, a broken line with an angle
pointing towards the south 47.
4.35. The theory of a “traditional boundary”, “strongly established” before the adoption
of the Erratum, therefore does not stand up. Subsequent documents do not provide any
support for this theory either.
B. Documents subsequent to the Erratum
4.36. Niger believes it possible to assert that the “view of the course of the
inter-colonial boundary” in the Say sector as it appeared, according to Niger, before the
adoption of the Arrêté and its Erratum, was “strongly maintained” after 1927 473. Besides the
fact that, as Burkina has just shown, ther e was certainly no “firmly established” boundary
before 1927, the documents Niger which relies on for the subsequent period cannot overturn
that conclusion.
4.37. Niger refers first of all to a report from the Acting Commander of Dori cercle
474 475
drawn up on 10 April 1932 after a tour, and quotes three lines from a nine-page report .
Yet a more attentive perusal of this report is all that is required to see that:
⎯ firstly, it confirms, rather than invalidati ng, as Niger suggests, the delimitation made by
the Erratum, since the signatory proposes a fu rther erratum to align the boundary more
closely with his own views; this means that he had no doubt at all about the fact that the
course established in 1927 constituted the law and that any change to it would require a
new text to be adopted ⎯ something that was never done;
⎯ secondly, this text proposes a revision of the legal boundary on the basis of an erroneous
premise. Its author indeed states that:
470MN, Anns., series D, No. 2.
471MN, Anns., Series C, No. 20.
472
See Section 3 below.
473
MN, p. 109, para. 7.20.
474Ibid.
475MN, Anns., Series C, No. 45. - 92 -
“[T]here are two possibilities: either the Arrêté général of
31August1927 and the Erratum that followed were intended to endorse the
work of Administrators Delbos and Prud’hon and officially establish the
boundary they proposed after their inspec tion tour, or those texts refer to other
documents.
The second option would476em to be h azardous and implausible, while the
first is the only logical one.”
In actual fact, it is the second option which should have been accepted, since the Arrêté and
its Erratum were drawn up on the basis of the Brévié-Lefilliatre and Choteau-Lefilliatre
Records of Agreement;
⎯ lastly, this redrawn boundary is entirely ne w and has nothing to do with the “traditional
boundaries” claimed by Niger with reference to the 1909 sketch-map, since Niger
proposes a line between Tillabéry and Dori cercles which would have its starting point
not at Nababori, but at Alfassi.
4.38. The Record of Agreement of 8 December1943 recounting the delimitation
exercises between Dori and Tillabéry carried out by Administrators Delmond, Texier and
Garat provides no further support for Niger’s position 47. Not only does it suggest that the
point of contact between Dori, Tillabéry and Say cercles is situated at another entirely new
location, which does not correspond to Nababori or Alfassi but would be at a distance of
some 6.5km from Nababori, but also it was signed more than ten years after the
disappearance of Upper Volta; that disqualifies this entirely new “traditional boundary”,
which can have no relevance, since Upper Volta was reconstituted in 1947 within its
478
1932 boundaries .
4.39. Niger’s arguments relating to the Bossébangou area therefore appear to be
unfounded, as is the case for its arguments regarding the area of the salient.
SECTION 2
T HE LINE CLAIMED BY N IGER IN THE “SALIENT ” SECTOR
IS NOT JUSTIFIED
4.40. Even though Niger makes as if seeking “to determine the precise course of the
‘salient’ as defined by the Erratum of October 1927” 479, its position is clearly that there is no
salient in this area. Referring to the line proposed in its earlier argument, Niger maintains that
“[g]iven that this boundary came not from Bos sébangou but ran directly from the Tao marker
to the ‘tripoint’ between the cercles of Dori, Tillabéry and Say, as identified above, the
480
frontier cannot create a salient in this area” . However, precisely the opposite conclusion
must be drawn: since there is a salient in this area, as is expressly indicated by the legislative
text of 1927, the boundary coming down from the Tao marker cannot stop at the “tripoint”
invented by Niger, but must of necessity reach the River Sirba at Bossébangou , which it does
in any event, as previous explanations have shown.
47Page 5 of the document.
47MN, p. 110, para. 7.20; MN Anns., Series C, No. 69.
478
See paras. 3.57-3.58 above.
47MN, p. 114, para. 7.30.
48MN, p. 112, para. 7.26. - 93 -
4.41. Still, one more invention means nothing to Niger: whereas the title constituted by
the Erratum states that from the River Sirba at Bossébangou, the frontier “turns back up”
towards the north-west, Niger’s Memorial maintains that there is no turn481 back up, and that
the frontier instead “simply turns in a south-westerly direction” from a point situated to the
north of Bossébangou. Finally, having reached the point of the salient, the frontier does not
482
turn back southwards but runs in a “roughly NNE/SSW” direction. Niger’s approach here
is once again quite simply to disregard the cour se of the frontier as defined by the text of the
Erratum, and to give precedence to a cour se which results from a few non-legislative
documents.
4.42. In this respect, Niger endow s the 1915 and 1927 maps with particular
significance. It is on these, according to Niger, that “reliance should be placed in order to
483
determine the precise course of the ‘salient’” . For Niger, they are more relevant than the
1960 map, which shou ld be disregarded since it “makes the frontier in this area run
significantly further to the east than that shown on the previous maps” 484. In other words, it
should be disregarded because it does not suit Niger. It will be agreed that this cannot be a
sufficient reason.
4.43. This argument is once again in total contradiction with the terms of the 1987
Agreement. Neither the 1915 map nor the 1 927 map are documents accepted by joint
agreement of the Parties.
4.44. Furthermore, in order to justify the line it claims, Niger purports to have
identified the exact location of the villages mentioned in the Arrêté as amended by the
Erratum 485. In actual fact, the location of two of the four villages, Alfassi and Kouro, is not a
486
matter of debate, as Niger’s siting of them places them incontrovertibly on the eastern side
of the lines claimed by each of the Parties. Ho wever, the same does not apply to the villages
of Tokalan and Tankouro.
4.45. As far as the village of Tokalan is concerned, Burkina considers that it is
completely impossible to locate, whereas Niger claims the opposite, but is unable to do so
convincingly. Thus it stated in 1990, with great authority, that this village “corresponds to the
487
exact location of the curren488illage of Takatami” , but then claims in its Memorial that it is
very close to Tangangari , albeit without explaining its change of opinion.
4.46. It is no longer possible to identify the village of Tankouro as it existed at the time
when the Arrêté and its Erratum were adopted, and it is not identified by Niger either.
According to Niger, the sketch-map of Say cercle drawn up by Administrator Truchard in
481
Ibid.
482MN, sketch-map on p. 114, and p. 114, para. 7.31.
483MN, p. 114, para. 7.30.
484
Ibid.
485
MN, p. 112, para. 7.27.
486MN, map reproduced facing p. 114.
487Ann. MBF 87.
488MN, p. 115, para. 7.31. - 94 -
489
1915 does feature it , but it appears likely that, in the same way as Tokalan, it “simply
disappeared during the period contemporary with the adoption of the 1927 Erratum, doubtless
490
as a result of the very unfavourable health c onditions prevailing at the time in this sector” .
This is probably what did happen, but Niger fails to see the consequence, which is that these
villages, which no longer existed in 1927, cannot provide the slightest piece of useful
information in determining the course described by the Erratum.
491
4.47. Niger also relies, in order to determ ine “the depth of the ‘salient’ at its point”
on a telegram/letter sent in 1935 by Say subdivision to Dori cercle 49. According to this text,
“[a]fter Bosseibangou, Say Subdivision encroaches on the left bank [of the River Sirba] to a
depth of some 15 km ⎯ the village of Alfassi (Torodi canton) is the only Say village located
on that bank”. However, this document is in capable of providing the slightest information
regarding the inter-colonial boundary, since it is subsequent to 1932 and prior to 1947.
4.48. Niger then claims that it is appropriate to disregard the description in the Erratum,
according to which, “turning back to the south [ on leaving the salient], [the line] again cuts
the Sirba at the level of the Say parallel”, on the twofold grounds that this description does
not allow the inclusion of the four villages and that “the Record of Agreement of
10February1927, which served as a preparatory document for the Arrêté général of
August 1927 and for the Erratum which corrected the latter, was evidently less precise on the
matter” 493. Yet not only are two of the four villages impossible to locate ⎯ and they could
certainly not have been located in 1927 since, as Niger itself concedes, they no longer existed
at that date 494⎯ but furthermore it would be utterly absurd to give precedence, for the
purpose of determining the frontier, to a previous non-legislative text over a subsequent
legislative text, especially as the former is in fact less precise than the latter.
4.49. Niger also asserts that the line it claims is confirmed by a sketch-map drawn by
Sergeant Labitte in 1930 495. However, not only is this sketch-map not accepted “by joint
agreement of the Parties” under the terms of the 1987 Agreement, it is also erroneous. It
actually includes within the salient the village of Boborgou Saba, even though this village is
mentioned neither in the Arrêté nor in the Erratum as having to be located within the salient.
Consequently, the form given to the salient by the drafter of the sketch-map is of no
relevance.
4.50. Finally, Niger asserts that “reference has to be made to the representations of the
Say parallel as it was shown on the maps of the period ⎯ and not on modern maps ⎯ in
order to determine the point where the frontier changes direction in this sector”, referring here
to the 1926 Blondel-La Rougerie map 496. Quite apart from the fact that Niger offers no
justification to support it, this assertion is entirely wrong.
489MN, p. 113, para. 7.28.
490Ibid.
491MN, p. 114, para. 7.31.
492
MN, Anns., Series C, No. 61.
493
MN, p. 115, para. 7.32.
494See para. 4.46 above.
495MN, p. 115, para. 7.33, fn. 329.
496
MN, p. 115, para. 7.33; MN Anns., Series D, No. 9. - 95 -
4.51. First of all, the Agreement reached between the Parties in 1987 does not in any
way identify this map as relevant, nor is it “[an ]other relevant document accepted by joint
agreement of the Parties”.
4.52. Secondly, it is obvious that the map could not have inspired the drafting of either
the Arrêté of August 1927 or its Erratum, nor can it justify any interpretation running counter
to their clear terms, since:
⎯ this map only represents a small portion of the area delimited by the Erratum;
⎯ it shows the village of Kalba located on the le ft bank of the Sirba and at the same time
within the salient ⎯ yet Kalba is not mentioned as appearing in the salient in the Arrêté
or in its Erratum;
⎯ the village of Tankouro is not featured on it; and
⎯ the Say parallel does not meet the River Sirba, whose source is shown much further to the
north.
4.53. None of Niger’s arguments can provide any basis for its position that there is no
salient in the area, a position directly contradicted by the terms of the Erratum. Nor can they
be used to demonstrate that the straight line fo rmed by the frontier in the last section of the
Say area is in fact made up of two straight lines.
S ECTION 3
THE FRONTIER BETWEEN THE INTERSECTION OF THE R IVER S IRBA AND THE S AY
PARALLEL AND THE BEGINNING OF THE BOTOU BEND IS MADE
UP OF A SINGLE STRAIGHT LINE
4.54. The text of the Erratum is extrem ely clear about the course of the frontier
between the end of the salient and the Botou be nd. It states that from the point where it
intersects with the River Sirba at the level of the Say parallel “the frontier, following an
east-south-east direction, continues in a straight line up to a point located 1,200 m to the west
of the village of Tchenguiliba”. Niger acknow ledges moreover that this text is “of great
simplicity”497and defines “the boundary in this area as a single straight line” 49, statements
with which Burkina agrees wholeheartedly.
4.55. Nevertheless, the line described by the Erratum is probably too simple for Niger,
since it rejects it, arguing that it “appears to have no basis in th499ituation prior to the adoption
of the Erratum and was never confirmed in the subsequent practice” .
4.56. Once again, Niger is mistakenly seeking to give precedence over legal title to
documents, mainly maps, which have not been accepted “by joint agreement of the Parties”
under the terms of the 1987 Agreement. Niger’s arguments therefore h
ave absolutely no legal
foundation, and can only be rejected.
497
MN, p. 116, para. 7.35.
498
MN, p. 116, para. 7.36.
49MN, p. 116, para. 7.35. - 96 -
4.57. Above and beyond th e legal error made by Niger, none of the “evidence” it
submits to justify a divided line is convincing.
4.58. Niger refers firstly to a series of maps and sketch-maps which purportedly show
the line in this area not as one straight line, but as a line made up of two sections.
4.59. The very first thing which strikes one upon looking at the list on page117 of
Niger’s Memorial is that it does not include either Captain Boutiq’s 1915 sketch-map ⎯
500
which Niger nonetheless judges to be “of fundamental importance” ⎯ or the 1927map
entitled “New frontier between Upper Volta a nd Niger”, on which it relies for support both
501
for its theory that the fron tier does not reach Bossébangou and for its claim regarding the
course of the “salient” 502. This absence is easily explained, since these documents militate
strongly against its case: the first of them s hows a line changing direction to form an angle
which points north-east, thereby giving more territo ry to Burkina than th e straight line does,
while the second one shows a straight line. Ni ger’s embarrassed silence here confirms the
utter incoherence of its position.
4.60. Furthermore, Niger also omits from its list the following maps:
503
⎯ the 1927 road map of the Colony of Upper Volta and the gene504 political and
administrative map, to a scale of 1:2,500,000, Second Edition 1928 , which in no way
confirm a line in two straight sections but show a single straight line as far as the Botou
bend; and
505
⎯ the 1930 “French West Africa” map, which shows a perfectly straight line .
4.61. With regard to the maps to which Niger does refer:
⎯ the “Niamey sheet of the ‘Sketch-Maps of the Sahara and Neighbouring Regions on a
506
scale of 1:1,000,000’ (ND-31, Army Geographical Section, 1926-1927)” and the
Blondel-La Rougerie map 507are both prior to the adoption of the Arrêté and its Erratum;
⎯ the “Government-General of French West A frica, Niger Colony, road map to a scale of
1:2,500,000, 1934 Edition” 508, for its part, illustrates cercle boundaries which do not
correspond at all to the course of the inter-colonial boundary described by the Erratum;
moreover, as it dates from 1934, it could neither take precedence over the title nor have
the slightest effect on the determination of the frontier of Upper Volta, which was
reconstituted in 1947 within its 1932 boundaries;
50MN, p. 111, para. 7.22.
50MN, p. 107, para. 7.17.
50MN, p. 114, para. 7.30.
503
MN, Anns., Series D, No. 11.
504
MN, Anns., Series D, No. 14.
50MN, Anns., Series D, No. 15.
50MN, Anns., Series D, No. 10.
50MN, Anns., Series D, No. 9.
508
MN, Anns., Series D, No. 16. - 97 -
⎯ the “French West Africa: General Political and Administrative Map to a scale of
1:2,500,000, Fourth Edition 1939, FWA Geographical Department, Dakar” 509 calls for
the same comment; furthermore, it invalid ates Niger’s theory, since it shows a line
divided into not two but three sections.
4.62. As far as the sketch-maps are concerned, virtually all of them fail to indicate the
author or the date, which disqualifies them fr om the outset. It should nonetheless be pointed
out that:
⎯ the 1:500,000 sketch-map entitled “Say Cercle”, with no author or date 51, does not show
any boundaries; what Niger has apparently taken for a cercle boundary actually
represents tracks or paths linking villages;
511
⎯ the 1:400,000 sketch-map entitled “Tour of 17 to 27 May 1943” , with no author or
date, provides proof that cercle boundaries were not considered very stable during the
1932-1947period (when Upper Volta was no longer in existence), since it very clearly
shows a boundary with an entirely new shape to the north-east of Bossébangou.
4.63. Niger ends the part of its argument based on maps and sketch-maps by
contending that:
“Moreover the point where the frontier changes direction, which appears,
inter alia, on the 1960 IGN map, is an undisputed frontier point between the two
States . . . That point is, moreover, very clearly identified on the completion
surveys carried out by the IGN during its 1958-1959 season. The survey entitled
‘Diapaga Information’ corr esponding to this sector of the frontier does in fact
include the indication ‘frontier marker’ at the precise place where the line
changes direction before subsequently connecting with the start of the Botou
Loop.” 512
513
4.64. However, a glance at the document in question reveals that the orange line
which is supposed to represent the inter-col onial boundary and which passes through the
point described as “frontier marker” is labe lled “Uncertain boundary to be maintained ⎯
territorial boundary according to the Protocol of Agreement (not surveyed on the ground)”.
This wording clearly shows that the orange lin e is far from being an inter-colonial boundary
firmly established along a divided line, since it is both “uncertain” and based on a “Protocol”
of which nothing is known, apart from the fact that it was not “surveyed on the ground”.
4.65. To conclude, Niger adduces a series of alleged colonial and post-colonial
effectivités, purportedly to demonstrate that the demarcation line in practice on the ground has
always been the line divided into two sections that it claims, rather than the single line set out
50MN, Anns., Series D, No. 18.
510
MN, Anns., Series C, No. 1.
511
MN, Anns., Series C, No. 68.
51MN, p. 118, para. 7.38.
51MN, Anns., Series D, No. 30. - 98 -
very clearly by the Erratum. The villages in the triangular area between the straight line and
the divided line have, according to Niger, always been regarded as part of its territory. These
villages are Dissi, Fombon, Latti, Tabaré and Tiaboungou 51. However, an analysis of the
documents shows that Niger is drawing erroneous conclusions from them.
⎯ The village of Dissi
4.66. The first of the references put forward by Niger to provide justification for Dissi
being under the administration of Nige515s entitled “Directory of the villages of Say
subdivision, Tamou canton, 1941” . The additional information provided by Niger about
this document is merely that it is reproduced in Annex Series C, No. 63, which is presented as
being an “(extract), undated, 1941”. This document, whose origins are unknown and which is
hence unusable, shows, according to Niger, that the village of Dissi was part of Tamou canton
in 1941. However, the village mentioned in th is document is actually Dissirire and not Dissi,
and it is highly unlikely that Dissi could be the same village as Dissiriré, in Tamou canton:
not only is Tamou very much to the east of Dissi, but there are also a considerable number of
villages much closer to Tamou than Dissi wh ich do not appear on the list of villages in
Tamou canton, such as, for example, Ouro Bambalé, Kankani, or indeed Latti or Tabaré.
4.67. Niger also produces the “Census tour of Say Subdivision, Tamou canton, dated
516
23 March 1947” , in an attempt, once again, to prove that the village of Dissiriré
corresponds to Dissi and has been administered by Niger for many years. However, the
document confirms on the contrary that Dissiriré could not possibly be Dissi. The report
states that the tour carried out by the admi nistrator led him to survey the villages of
OuroHesso, Sadima and Dissiriré in a singl e day, 13February19 47, having ended his
previous day’s tour in Kotaki. This helps to locate the sector where Dissiriré would logically
be found (not far from Kotaki), and indeed, a close examination of the map of the region does
show that the village of Dissiriré is located to the east of Kotaki and to the north of the Botou
bend. Sketch-map No. 7 in the present Counter-Memorial, shows the areas through which the
administrator travelled on 10, 11, 12 and 13Fe bruary 1947, and demonstrates conclusively
that while the village of Dissiriré (or Dissiri dé) is indeed in Niger’s territory, in Tamou
canton, it is definitely not, on the other hand, situated within the triangle between the straight
line laid down by the Erratum and the divided lin e claimed by Niger. This village is not
Dissi.
514
MN, pp. 119-120, para. 7.39.
515
MN, p. 119, fn. 343.
51MN, Anns., Series C, No. 70. - 100 -
4.68. The three other documents which Niger be lieves it can put forward to prove that Dissi
has always been administered by Niger, i.e., the “Census tour of Tamou canton by the Head of Say
517
Subdivision, 25 Mar5181954” , the “Record of tax receipts, Say District, Tamou canton, da519 3
September 1971” and the lists of “Localities in Tamou canton, 1987, 1991 and 2001” are not
convincing evidence either, as they too refer to the location of Dissiriré which, contrary to Niger’s
assertions, cannot be mistaken for Dissi.
⎯ The villages of Fombon, Latti and Tabaré
4.69. The first of the documents which Niger refers to as proof that Fombon (which cannot
be located on the maps available to Burkina), Latti and Tabaré are under the administration of
Niger is prior to the title, since it dates from 1921. This is the “List of cercle villages by canton,
prepared on 1 October 1921 ⎯ Torodi canton” 520. Not only is this of no use in interpreting the
Erratum, due to its date, but Niger also reads thi ngs into it which the document does not state: in
the version it produces, this document does ind eed mention Fonbougou (which cannot be found on
the maps at Burkina’s disposal), but not Fombon. Nor, for that matter, does it mention Latti or
Tabaré. A place called “Taboura” is mentioned, bu t nothing allows one to assume that this is
Tabaré. It will also be noted that neither Di ssi nor Tiaboungou are mentioned, even though these
are also claimed by Niger.
4.70. The second document is the “List of localities of T521di canton, extract from the
General List of Localities of French West Africa , 1927” . This document does not, however,
mention any of the villages that Niger claims it mentions. It features “Fombougou”, which cannot
be found on the maps available to Burkina, “L ati”, which is not “Latti” and cannot be found on
those maps either, and “Taboura”, which is not “T abaré” and also cannot be found on those maps.
Moreover, it appears that the largest villages in the “triangle”, that is, Golongana, Dissi, Deguema,
Faltyangou or Dyaya, and Kankani, are not mentioned in the list, even though they should be, if
Niger is to be believed, since they are situated in the immediate vicinity of the villages of Latti and
Tabaré. It is therefore highly unlikely that the villages of “Lati” and “Taboura” in Torodi canton
correspond to Latti and Tabaré.
522
4.71. Niger then relies on a “Tour Report, Say Subdivision, 13-27 September 1933” . With
regard to the name “Fombonou” (which Burkina has not found on a map), this appears only
fleetingly and inconsistently in the report. The document in fact begins with a list of the villages in
the various cantons, with population figures for each of them. Fombonou is not on that list. Next,
the livestock numbers for each village are given. Here, Fombonou appears in the list of villages in
Tamou canton (whereas, if the information in the prev ious document is to be accepted, it should
instead be located in Torodi canton), but the livestock numbers are as follows:
⎯ sheep: none
⎯ goats: none
51MN, Anns., Series C, No. 81.
51MN, Anns., Series C, No. 101.
51MN, Anns., Series C, Nos. 104, 107, and 108.
520
MN, Anns., Series C, No. 4, referred to in MN, p. 119, fn. 344.
52MN, Anns., Series C, No. 6.
52MN, Anns., Series C, No. 51. - 101 -
⎯ cattle: none
⎯ donkeys: none
⎯ horses: none.
Finally, the document provides a table of crops for each village. Fombonou does not appear in that
table. In conclusion, this document shows only that the village of Fombonou, which has neither
population, nor livestock, nor crops, did not exist at the time when the survey in question was
made.
4.72. It is probably on the basis of this list mentioning the ghost village of Fombonou that the
523
“List of Niger cantons and villages forwarded in 1948 to the Minister for Overseas France” and
the “Alphabetical list of villages by canton, Torodi canton, updated to 1January1954” 524 were
drawn up. They therefore prove nothing. Furthermore, for the same reasons as those set out in
paragraph4.70 above, the villages of “Lati” a nd “Taboura”, mentioned as appearing in Torodi
canton, cannot be assimilated with Latti and Taba ré. Incidentally, the second of these two
documents no longer spells the village’s name as “Lati” but as “Pati”.
4.73. Niger also refers to “ Arrêté No.2794/APA est525ishing polling stations and districts
for the elections to the National Assembly, 1955” . However, this document is of particular
interest in that it definitively proves that “Taboura” cannot be considered to be the same place as
“Tabaré”, since it states that for electoral purposes, Taboura is part of Bolsi, the electoral district
covering the villages of Alfassi, Bolsi and Bosseybangou. In fact, the village of Bolsi is at such a
distance from the village of Tabaré that it could not possibly provide its polling station, which
proves that Taboura and Tabaré are definitely two different villages.
526
4.74. Finally, Niger refers to a “Record of tax receipts, Torodi canton, 1971” , a “List of
villages in Torodi canton, 19 August 1973” 527and a “List of polling stations in Say District,
528
1 November 1989” . However, as they were drawn up after the Colonies gained independence,
these documents cannot be considered relevant. The same applies to the documents put forward by
Niger to justify effectivités in the village of Tiabogou 529.
4.75. In any case, the argument based on effectivités collapses automatically once Niger
claims that they take precedence over the Erratum, which was never called into question during the
colonial period and whose relevance for the descrip tion of the course of the frontier was officially
reaffirmed by the 1987 Agreement between the Par ties. And the text of the Erratum is perfectly
clear: the line “continues in a straight line”.
52MN Anns., Series C, No. 71; “Fombounou” is mentioned here as located in Tamou canton.
52MN Anns., Series C, No. 80; “Fombongou” is mentioned here as located in Torodi canton.
525
MN, Anns., Series B, No. 31.
526
MN, Anns., Series C, No. 102.
52MN, Anns., Series C, No. 103.
52MN, Anns., Series C, No. 106.
529
Lists of localities in Torodi canton, Say district, Tillabéry département , 1991 and 2001 (MN. Anns., Series C,
Nos. 107 and 108). - 102 -
4.76. For all these reasons, Burkina concludes, with regard to the “Say sector” that
⎯ the course of the frontier described by Niger is entirely unjustified;
⎯ the course of the frontier in the Say sector is that described on page 158 of Burkina’s Memorial
and follows:
⎯ a straight line from530e Tao astronomic marker (Lat.: 14° 030' 4.7N; Long.:
0° 22' 51.8" E) as far as the point where the frontie r reaches the River Sirba at Bossébangou
(Lat.: 13° 21' 06.5" N; Long.: 1° 17' 11.0" E) 531;
⎯ from that point, the frontier follows the River Si rba from east to west as far as the point on its
right bank with co-ordinates: Lat.: 13° 19' 53.5" N; Long.: 1° 07' 20.4" E;
⎯ from that point, the frontier follows the course shown on the 1:200,000-scale map of the
Institut géographique national de France , 1960 edition, as far as the point with co-ordinates:
Lat.: 13° 22' 30.0" N; Long.: 0° 59' 40.0" E;
⎯ from that point, the frontier runs in a straight line southwards, ending at the intersection
of the right bank of the River Sirba with the Say parallel (Lat.: 13°06'10.7"N; Long.:
0° 59' 40.0" E);
⎯ from that point, the frontier runs in a strai ght line as far as the start of the Botou bend
532
(Tyenkilibi) (Lat.: 12° 36' 19.2" N; Long.: 1° 52' 06.9" E) .
530The co-ordinates of this point have been measured by GPS by Burkina. The co-ordinates of this marker on the
Clarke 1880 ellipsoid are: Lat.: 14° 03' 13'' N; Long.: 00° 22' 53'' E.
531The co-ordinates of this point, and of those which follow, are given on the Clarke 1880 ellipsoid.
532The co-ordinates of this point are those used in the record of the work of the Joint Survey Mission of
3 July 2009 (Ann. MBF 101), as measured by GPS (WGS 84 ellipsoid). - 103 -
SUBMISSIONS
5.1. In view of all the considerations contained in its Memorial and in the present
Counter-Memorial, Burkina Faso stands by the subm issions set forth in paragraphs5.1 and 5.2 of
its Memorial in their entirety and requests the Court to find in its favour and to reject any contrary
submissions from the Republic of Niger.
20 January 2012
(Signed) Jérôme T RAORÉ ,
Minister of Justice and Promotion of Human Rights,
Keeper of the Seals, Agent of Burkina Faso.
(Signed) Jérôme B OUGOUMA , (Signed) Kadré Désiré O UEDRAGO ,
Minister for Territorial Administration, Ambassador of Burkina Faso to the Kingdom
Decentralization and Security, Co-Agent. of the Netherlands, Co-Agent.
___________ - 104 -
SUMMARY OF SKETCH -MAPS
Sketch-map No. 1 Positions of the Parties
Sketch-map No. 2 Comparison of the sketch-maps proposed by Delbos and Prudon with
the lines defended by the two Parties
Sketch-map No. 3 Course of the boundary between the Tong-Tong and Tao markers
Sketch-map No. 4 Course of the boundary after the Tao marker
Sketch-map No. 5 Course of the boundary in the area of the salient
Sketch-map No. 6 Course of the boundary in the area of the salient
Sketch-map No. 7 Census tour of Say Subdivision, Tamou canton, 10 to 13 February 1947 - 105 -
LIST OF ANNEXES
Annex CMBF 1. Letter from the Governor-General of FWA to the Governor of Niger,
2 April 1927
Annex CMBF 2. Letter from the acting Director of Political and Administrative Affairs to the
Governor-General of FWA, July 1927
Annex CMBF 3. Annual political report of Tillabéry cercle, 1931
AnnexCMBF4. Letter No.2 AP/2 from the Governor-General of FWA to the Governor of
Niger, 3 January 1934
AnnexCMBF5. Letter No.418 AP/2 from the Director of Political and Administrative
Affairs of the Government-General to the Military Chef du Cabinet ,
7 June 1938
Annex CMBF 6. Note No. 521 CM2 from the FW A Geographical Department to the Director
of Political and Administrative Affairs of the Government-General,
25 June 1938
___________
Counter-Memorial of Burkina Faso