Memorial of Burkina Faso

Document Number
17108
Document Type
Date of the Document
Document File
Document

12377

INTERNATIONAL COURT OF JUSTICE

CASE CONCERNING THE FRONTIER DISPUTE
(BURKINA FASO/NIGER)

MEMORIAL OF BURKINA FASO

VOLUME I

20 APRIL 2011

[Translation by the Registry] T ABLE OF C ONTENTS

INTRODUCTION ................................................................................................................................... 1

Section 1 Conclusion of the Special Agreement........................................................................... 2

1. Border incidents and maintaining the status quo ................................................................. 2

2. Preparation and adoption of the Special Agreement............................................................ 3

Section 2 The subject of the dispute ............................................................................................. 3

Section 3 Geography of the Niger-Burkina frontier...................................................................... 7

Section 4 Structure of the Memorial........................................................................................... 13

C HAPTER I HISTORICAL BACKGROUND ........................................................................................... 14

Section 1 Fixing the colonial boundaries between Upper Volta and Niger................................ 14

1. The French colonial conquest............................................................................................ 15

2. The incarnations of Upper Volta and the colonial subdivisions ........................................ 18

3. The colonial boundaries of Upper Volta and Niger........................................................... 22

Section 2 Discussions on fixing the frontier............................................................................... 27

1. The Protocol of Agreement of 23 June 1964..................................................................... 28

2. The Agreement and Protocol of Agreement of 28 March 1987......................................... 34

3. The decision of 15 May 1991 and its denunciation by Niger............................................ 40

4. The conclusions of the fourth ordinary session of the Joint Technical Commission
on Demarcation (2001)..................................................................................................... 41

CHAPTER II A PPLICABLE LAW ......................................................................................................... 45

Section 1 The principle of respect for colonial boundaries......................................................... 46

1. The “methodology” of uti possidetis ................................................................................. 46

2. Determination of the uti possidetis line (the relationship between title and
effectivités)........................................................................................................................ 48

Section 2 Sources of applicable law............................................................................................ 50

1. The agreement of the Parties on relevant sources.............................................................. 50

2. The relevant documents..................................................................................................... 53

A. The frontier title: the Arrêté of 1927 and its Erratum................................................. 54
B. The course of the boundary should the Arrêté and its Erratum not suffice.................. 57

C HAPTER III THE TASK OF THE COURT IN RESPECT OF THE DEMARCATED PORTION OF THE
FRONTIER ..................................................................................................................................... 59

Section 1 The Parties’ agreement................................................................................................59

1. The methodology behind the Parties’ agreement (reminder)............................................. 60 - ii -

2. From the heights of N’Gouma to the Tong-Tong astronomic marker............................... 65

3. From the Botou bend to the River Mekrou........................................................................ 68

Section 2 The task of the Court Confirmation of the Parties’ agreement ................................... 72

1. The Court’s participation in the comprehensive settlement of the frontier dispute

between the Parties ........................................................................................................... 72
2. Determination of the endpoints of the frontier................................................................... 74

C HAPTER IV T HE PART OF THE FRONTIER WHOSE COURSE IS TO BE DETERMINED BY THE

C OURT ......................................................................................................................................... 80

Section 1 The course of the frontier from the Tong-Tong astronomic marker to the point
where it reaches the River Sirba at Bossébangou .................................................................. 83

Section 1 The course of the frontier from the Tong-Tong astronomic marker to the point
where it reaches the River Sirba at Bossébangou .................................................................. 83

1. The location of the frontier points designated in the 1927 Arrêté ..................................... 84

A. The Tong-Tong astronomic marker............................................................................. 84

B. The Tao astronomic marker......................................................................................... 84

C. The point where the boundary reaches the River Sirba at Bossébangou ..................... 85
2. The line connecting the three points designated in the Arrêté........................................... 90

A. The 1927 Arrêté adopts an artificial boundary in this sector....................................... 90

B. The course defined in the Arrêté consists of two straight lines.................................... 92

C. The interpretation whereby the course of the frontier consists of two straight
lines was officially endorsed by the two Parties......................................................... 96

D. The course of the frontier as two straight lines corresponds to the practice
followed in the field of boundary delimitation......................................................... 100

Section 2 The course of the frontier from the point where it reaches the River Sirba at
Bossébangou as far as the intersection of the Sirba with the Say parallel........................... 108

1. The description of the frontier ......................................................................................... 110

A. “almost immediately turns back up towards the north-west” .................................... 110

(a) The course of the frontier between point P and the point where it “turns
back up” towards the north-west ....................................................................... 110

(b) Point P1 from which the frontier begins to turn back up towards the
north-west........................................................................................................... 114

(c) “leaving to Niger, on the left bank of that river, a salient which includes the
villages of Alfassi, Kouro, Tokalan, and Tankouro”.......................................... 115

(d) “then, turning back to the south, it again cuts the Sirba at the level of the
Say parallel” ...................................................................................................... 115

2. The course of the frontier................................................................................................. 118

3. The line described in the Erratum and the consensual line of 1988 are identical............ 128

Section 3 From the intersection of the Sirba with the Say parallel to the west of the village
of Tchenguiliba.................................................................................................................... 129 - iii -

SUBMISSIONS .................................................................................................................................. 133

LIST OF ANNEXES ........................................................................................................................... 136

LIST OF CARTOGRAPHIC A NNEXES ................................................................................................. 144 INTRODUCTION

0.1. By a joint letter dated 12 May 2010, filed in the Registry on 20 July 2010, Burkina Faso

(hereinafter Burkina) and the Republic of Niger (hereinafter Niger) notified to the Registry of the
Court a Special Agreement, which was signed in Niamey on 24February2009 and entered into
force on 20 November 2009, with a view to the full and final settlement of the frontier dispute

between them.

0.2. The Parties have thereby once again shown their confidence in the distinguished Court,
to which both have previously had occasion to refer other frontier disputes with other neighbours:

 by an Order of 22December1986, a Chamber of the Court settled the Frontier Dispute
1
between Burkina and Mali ; and

 the Order of 12 July 2005 issued by another Chamber of the Court brought an end to the case
2
concerning the Frontier Dispute between Niger and Benin .

Not only were these decisions welcomed by the parties, they also set precedents which are
particularly relevant for the purposes of resolving this case, both in respect of establishing the
endpoints of the frontier between Niger and Burkina, and in respect of the methodology adopted.

0.3. However, that methodology cannot be fully transposed to the present dispute. In the

cases dealt with by the Judgments cited above, the Parties were disputing the existence and scope
of the legal titles establishing their frontier. They were disputes relating to boundary delimitation

in the strict sense of the term. The present case is different in that both Niger and Burkina accept
that the Arrêté of 31August1927 3 of the Governor-General of French West Africa (FWA), as
4
clarified by the Erratum of 5 October 1927 , describes the course of their comm5n frontier, which 6
only remains to be demarcated, as the Parties state in the Agreement and Protocol of Agreement
signed on 28 March 1987 and referred to explicitly in the Special Agreement seising the Court. As

the delimitation has been established, it merely remains for the distinguished Court to clarify the
interpretation of those instruments with a view to the demarcation of the frontier between the two
7
countries .

8
0.4. As the Parties agreed under Article 3 of the Special Agreement , the Court, by Order of
14September2010, fixed 20 April 2011 as the time-limit for each Party to file a Memorial. The

present Memorial is submitted in accordance with that Order.

1
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 554.
2Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 90.

3Ann. MBF 34.

4Ann. MBF 35.
5
Ann. MBF 72.
6
Ann. MBF 73.
7
Which is, moreover, the title of the 1987 Agreement and Protocol “on the demarcation of the frontier
between the two countries”. For further details, see below, paras. 0.17 to 0.19.
8
“1. Without prejudice to any question as to the burden of proof, the Parties request the Court to authorize
the following procedure for the written pleadings: (a) a Memorial filed by each Party not later than
nine (9) months after the seising of the Court.” - 2 -

SECTION 1
C ONCLUSION OF THE S PECIAL A GREEMENT

1. Border incidents and maintaining the status quo

0.5. The Special Agreement whereby the two Parties seised the Court of the present dispute

has its historical origins in the fact that the boundary is not demarcated. This has led to territorial
encroachments, of which both States have rapidly sought to prevent any recurrence and limit the

effects. Just four years after gaining independence, Burkina and Niger expressly included a clause
to that effect in the Protocol of Agreement of 23June1964 concerning the demarcation of their
common frontier, whereby “[t]he security forces of each of the Contracting Parties shall make no

incursion i9to the neighbouring territory without prior express authorization from the responsible
authority” . It was further agreed in the same Protocol that “[t]he status quo shall be maintained in
respect of land utilisation rights until the frontier has been demarcated, and those rights shall
10
terminate forthwith on signature of the demarcation report” .

0.6. Notwithstanding these undertakings, there have been constant problems on the frontier
since 1964. Nevertheless, the two Parties have always agreed that these problems should not affect
the process of demarcating the frontier, especially as the incursions were “unintentional” 11and due

solely to the absence of demarcation.

0.7. When the two Parties concluded the Special Agreement seising the Court
inFebruary2009, they considered it appropriate, first, to reiterate the undertaking of 1964 12and,

secondly, to reassert the need f13 increased cross-border co-operation— which they have always
sought — in the disputed area .

0.8. Pursuant to Article 10 of the Special Agreement:

“Pending the Judgment of the Court, the Parties undertake to maintain peace,
security and tranquillity among the populations of the two States in the frontier region,
refraining from any act of incursion into the disputed areas and organizing regular

meetings of administrative officials and the security services.

With regard to the creation of socio-economic infrastructure, the Parties

undertake to hold preliminary consultations prior to implementation”.

Ann. MBF 45, point 4, second paragraph, of the Protocol.

1Ibid., point 3 of the Protocol.

1Communication from the Minister for the Interior and Decentralization of the Republic of Niger during work on
the framework for consultation on cross-border transhumance between the Republic of Niger and Burkina Faso,
25 January 2003, p. 3, Ann. MBF 95.

1The Protocol of 13June1964 is specifically mentioned in the first paragraph of the Special Agreement of
24 February 2009.
13
See, for example, the report of the meeting between the Minister for the Interior and Planning (Republic of
Niger) and the Minister for Territorial Administration and Security (Burkina Faso) held in Tenkodogo (Burkina Faso)
from 24 to 26 May 2000, Ann. MBF 94. - 3 -

0.9 This formal undertaking merely sets out for the Parties

“the principle universally accepted by international tribunals and likewise laid down in
many conventions... to the effect that the parties to a case must... not allow any
step of any kind to be taken which might aggravate or extend the dispute” . 14

2. Preparation and adoption of the Special Agreement

0.10. The negotiations aimed at settling the frontier dispute between the Parties and their
partial deadlock are described in detail in Chapter I, section two, of this Memorial. In order to
15
break the deadlock , in 2006 the two States began to consider submitting the dispute to the
International Court of Justice. The final negotiations resulted in the Foreign Ministers of the two
countries signing the Special Agreement in Niamey on 24 February 2009 16.

0.11. Nevertheless, this did not mark the end of the discussions on seising the Court. Thus in

Article 2, paragraph 2, of the Special Agreement, reference is made t17the Parties’ “agreement” on
the demarcation of the two outermost sectors of the boundary ; that agreement, however, was
never formalized in a written document, pending overall demarcation of the frontier. Only after an

excha18e of letters between the Ministers concerned did the Parties reach a final agreement in
2009 .

S ECTION 2
T HE SUBJECT OF THE DISPUTE

0.12. Article 2 of the Special Agreement describes the subject of the dispute as follows:

“The Court is requested to:

1. determine the course of the boundary between the two countries in the sector from
the astronomic marker of Tong-Tong (latitude 14° 25' 04" N;

longitud147') to the beginning of the Botou bend
(latitude 12° 36' 18" N; longitude 01° 52' 07" E);

2. place on record the Parties’ agreement on the results of the work of the Joint
Technical Commission on Demarcation of the Burkina Faso-Niger boundary with
regard to the following sectors:

1Electricity Company of Sofia and Bulgaria, Order of 5December1939, P.C.I.J., SeriesA/B, No., p. 199;

LaGrand (Germany v. United States of America), Judgment, I.C.J. Reports 2001, p. 503, para. 103.
1The last meeting of the Joint Technical Commission on Demarcation established by the Agreement of
28 March 1987 took place in Niamey from 28 to 30 September 2004 (see the report of the meeting of the Joint Technical

Commission on Demarcation of the Niger-Burkina Frontier, Niamey, 28-30 September 2004, Ann. MBF 97).
1See also the Joint Communiqué of the meeting of the Foreign Ministers for negotiation and signature of the
Special Agreement seising the In ternational Court of Justice of the frontier dispute between Niger and Burkina Faso,
Niamey, 22-24 February 2009, Ann. MBF 98.

1See below, paras. 0.12 to 0.14.

1See the record of the work of the Joint Survey Mission of 3July2009, Ann.MBF101, and the record of the
meeting to ascertain the co-ordinates of the unmarked points in sectorB, 15October2009, Ann.MBF105. Details on
how this “agreement” came about are given in Chapter III of this Memorial (paras. 3.12-3.13). - 4 -

(a)the sector from the heights of N’Gouma to the astronomic marker of
Tong-Tong;

(b) the sector from the beginning of the Botou bend to the River Mekrou”.

0.13. The Special Agreement thus divides the boundary into three sectors. From north-west

to south-east (see sketch-map No.1 on page7 of the French text— General view of the frontier
between Burkina and Niger, also reproduced in Cartographic Annex 18), they consist of:

 the sector from the heights of Mount N’Gouma to the Tong-Tong astronomic marker;

 the sector from the Tong-Tong marker to the beginning of the Botou bend; and

 the sector from the beginning of the Botou bend to the River Mekrou.

0.14. Whilst Burkina would have preferred a simpler formulation, whereby the Parties would

have requested the Court to determine the entire course of their common frontier, it agreed to the
one proposed by Niger. However, Burkina wishes to emphasize that the subject of the dispute
remains unchanged: the purpose of the judgment that the Court is requested to render is to put an
end, with the authority of res judicata, to the entire frontier dispute between the Parties.

0.15. Moreover, the Special Agreement has two further distinctive features worthy of note:

 it pays particular attention to the applicable law and emphasizes the importance of the

Agreement of 28 March 1987; and

 it envisages ex ante the arrangements for implementing the Judgment.

0.16. Paragraphs 2 to 4 of Article 7 of the Special Agreement, describing those
arrangements, read thus:

“2. From the day on which the Judgment is rendered, the Parties shall have
eighteen(18)months in which to commence the work of demarcating the
boundary.

3. In case of difficulty in the implementation of the Judgment, either Party may seise
the Court pursuant to Article 60 of its Statute.

4. The Parties request the Court to nominate, in its Judgment, three (3)experts to

assist them as necessary in the demarcation.” - 5 -

Sketch-map No. 1 - General view of the frontier between Burkina and Niger

GENERAL VIEW OF THE FRONTIER BETWEEN THE TWO STATES
  
   

 
 



$

Scale : 1/2,000,000

 !""
% 


   

N'Gouma Heights
    #

  



Republic of Niger

  

 

Point where the frontier reaches the
  
  


 "#    sébangou

Inter he

  " $"
Say parallel with the Sirba


Start of the

   
Botou bend

D

 

Section to be determined by the Court
     


Marked sections

 
 
Intersection with the

 
 
River Mekrou

 
   
Frontiers with third countries

n

 
o


ses only

7 - 6 -

Paragraphs 2 and 3 reflect the wording of paragraphs 2 a19 3 of Article 7 of the Special
Agreement of 15June2001 in the Benin/Niger case , while paragraph 4, whose application
ensured that the Judgment of 22 December 1986 was enforced in the best possible conditions 20, is
21
taken directly from the Special Agreement of 16 September 1983 in the Burkina/Mali case .

0.17. Furthermore, while it is not unusual for the Parties to devote a provision of the Special

Agreement seising the Court to applicable law, it is noteworthy that, in this instance, they focus on
a specific, earlier treaty, in which they had given a precise and exhaustive list of the instruments to
be taken into consideration for the demarcation of the frontier: the Agreement of 28 March 1987.

0.18. That treaty is mentioned no fewer than four times in the Special Agreement:

 in Article 6 on Applicable law, it is singled out:

“The rules and principles of international law applicable to the dispute are those
referred to in Article 38, paragraph 1, of the Statute of the International Court of
Justice, including: the principle of the intangibility of boundaries inherited from

colonization; and the Agreement of 28 March 1987;”

but it is also mentioned three times in the preamble:

 in the very first paragraph, the Parties recall that:

“by agreements signed at Niamey on 23June1964 and at Ouagadougou on
28 March 1987, the two Governments agreed to mark out their common boundary and
to that end created a Joint Technical Commission on Demarcation”;

 in the next paragraph, they cite in full the first two articles of that Agreement:

“Whereas Articles 1 and 2 of the Agreement of 28March1987 provide as
follows:

‘Article 1

The frontier between the two States shall run from the heights of N’Gouma,

situated to the north of the Kabia ford, to the intersection of the former boundary of
the cercles of Fada and Say with the course of the Mekrou, as described in the Arrêté
of 31 August 1927, as clarified by the Erratum of 5 October 1927.

Article 2

The frontier shall be demarcated by boundary markers following the course
described by Arrêté 2336 of 31/08/1927, as clarified by Erratum 2602/APA of
5/10/1927. Should the Arrêté and Erratum not suffice, the course shall be that shown

19Article 7, paragraphs 2 and 3, available on the Court’s website at
http://www.icj-cij.org/docket/files/125/7068.pdfe also paragraph 2 of Article IV of the Special Agreement of

16 September 1983 in the Burkina/Mali case, also available on the Court’s website at
http://www.icj-cij.org/docket/files/69/10665.pdf
20See Affef Ben Mansour, La mise en Œuvre des arrêts et des sentences des juridictions internationales , thesis,
Paris West University, Nanterre La Défense, 2009, 664 p., esp. pp. 487-497 and 498; see also the working documents of

the demarcation commission, pp. 589-606 (forthcoming from Larcier, Brussels, in 2011).
21Article IV, paragraph 3, ibid. - 7 -

on the 1:200,000-scale map of the Institut Géographique National de France, 1960
22
edition , and/or any other relevant document accepted by joint agreement of the
Parties’;”

 the following paragraph states that “thanks to the work of the Joint Technical Commission on
Demarcation established pursuant to these provisions, the Parties have been able to reach
agreement” in respect of those sectors of the boundary that have been marked out (which are
23
referred to once again in Article 2, paragraph 2, of the Special Agreement );

 finally, in the last paragraph of the preamble, the Parties state that in concluding the Special
Agreement, they have thus applied “Article8 of the Agreement of 28March1987 referred to

above”.

0.19. This shows the extreme importance of the Agreement of 28March1987 and of the
instruments to which it refers for the purposes of settling the dispute submitted by the Parties to the
Court— in particular with regard to the actual definition of its subject-matter. Thus it follows
from Articles 1 and 2 of the Agreement— cited in full in the second paragraph of the Special

Agreement  that the Parties consider that the frontier between them has already been “described
in the Arrêté of 31August1927, as clarified by the Erratum of 5October1927”. They are
therefore not in disagreement as to the delimitation of their common frontier, but in respect of

certain points of detail requiring clarification with a view to the demarcation of that frontier.

0.20. The remaining procedural provisions of the Special Agreement of 24February2009
are fairly standard. Burkina would nevertheless like to point out that it regrets Niger’s rejection of
its proposal to file written pleadings not simultaneously, but consecutively, pursuant to Practice

Direction I. Burkina, for its part, had indicated that it was equally willing to produce either the
Memorial and Reply or the Counter-Memorial and Rejoinder, as required under the consecutive
procedure, without prejudice, of course, to “any issue in the case, including the issue of burden of
proof”.

S ECTION 3

G EOGRAPHY OF THE NIGER -BURKINA FRONTIER

0.21. For the Court to have a clear picture of the background to the frontier dispute between
Burkina and Niger, a brief description of the geography of the Niger-Burkina frontier is

appropriate.

0.22. The frontier, whose course the Court is requested to establish in the Special Agreement

in this case, lies between two West African States, former French colonies which gained full
sovereignty in 1960.

0.23. Burkina, formerly Upper Volta (until 1984), has been an independent State since
5 August 1960. It shares its frontiers, which stretch over more than 3,500 km, with six States: to
the south, Benin, Togo, Ghana and Côte d’Ivoire; to the west and north, Mali; to the east, Niger

(see sketch-map No.2 on the following page— General situation of the territories of the Parties;
also reproduced in Cartographic Annex 19).

22
This refers to all the 1:200,000-scale maps produced by IGN France in 1960; these are reproduced in
Cartographic Annexes 7 to 12.
23See above, para. 0.12. - 8 -

Sketch-map No. 2 - Situation of the parties - Territorial overview



  
Chad

          

CeR nte rapluA blfircican

 


Libya

C      

     

   

 




Map prepared for illustrative purposes only

  

!    
l      

A

  



  

     
1/20,000,000



 


          

        



    

     
Guinea

      
Mauritania

Morocco  
     

    
Senegal

    
    gtreiers

SITUATION OF THE PARTIES - TERRITORIAL OVERVIEW








 
         " !  # # 
Gambia

11 - 9 -

0.24. Present-day Burkina, which is home to a population of some fifteen million and has its
capital at Ouagadougou, is divided into 13 administrative regions, which are further subdivided
into 45 provinces (see sketch-map No. 3 on page 12 of the French text— Administrative Map of
Burkina Faso; also reproduced in Cartographic Annex20). Five provinces border Niger. From

north to south, these include the provinces of Oudalan (administrative centre: Gorom-Gorom),
Seno (administrative centre: Dori), Yagha (administrative centre: Sebba), Komondjari
(administrative centre: Gayeri) and Tapoa (administrative centre: Diapaga). The first three of

these provinces are in the Sahel region (administrative centre: Dori) and the last two in the Eastern
region (administrative centre: Fada N’Gourma) (see sketch-map No.4 on page13 of the French
text — Administrative Divisions on the Frontier; also reproduced in Cartographic Annex 21).

0.25. Most of Burkina is usually described as forming a peneplain. It consists of a plateau at
an average of 300m above sea level. There are few hills. The frontier region with Niger lies
within the western basin of the River Niger. A number of the Niger’s tributaries that are relevant to

this case — none of which are navigable — such as the Sirba, the Tapoa and the Dyamongou, have
their source in Burkina .4

0.26. The Republic of Niger, for its part, has been independent since 3 August 1960. It also
shares its frontiers with six States: it is bounded to the south by Nigeria, to the east by Chad, to the
north by Libya and Algeria, to the north-west by Mali, and finally to the south-east by Burkina and

Benin (see sketch-map No. 2 on page 11 of the French text — General Situation of the Territories
of the Parties; also reproduced in Cartographic Annex 19).

0.27. The territory of Niger, which has a population of some 11 million people, is currently
divided into three regions, which are subdivided into 36 départements (see sketch-map No. 5 on the
following page — Administrative Map of the Republic of Niger; also reproduced in Cartographic

Annex 22). The region bordering Burkina is Tillabéry (which encloses the region that has Niamey
as its administrative centre, on the east bank of the River Niger). Within that region, the two
départements of Téra (in the north) and Say (in the south) border Burkina (see sketch-map No. 4 on
page13 of the French text— Administrative Divisions on the Frontier; also reproduced in

Cartographic Annex 21).

0.28. The majority of Niger’s territory lies within the Sahara or the Sahel. It has only one

river running through it, the River Niger, in the south-west of the country.

0.29. The frontier between Burkina and Niger runs, broadly speaking, in a north-westerly
south-easterly direction and is some 590 km in length. It can be described very approximately as
following the same direction as the right bank of the River Niger, at a distance varying in different
sectors of twenty to a hundred kilometres from the river bank.

2The Sirba, the Tapoa and the Dyamongou join the River Niger respectively at Sorbon Goungou, Wékouré and

north-west of Kirtachi Seybou. - 10 -

Sketch-map No. 3 - Administrative map of Burkina





(

    

Benin





  

    
    

 
 

&&

Map prepared for illustrative purposes only

 
  





     
    


  

  

 
    


          


     

 

  



  


  



        

  
   ’
   

 
  

  




   
  
    

 
 





  

  


 
 



 

        

 

 
 

     


 

   
 








  




  


     

1
 ! #$%&

0/



 
dministrative centres of   

 
 

  Rea gd iom A nia ndépartementsa utn iv de arcie es nta re ns d 

International frontier
National capital

  

1/4,000,000

     

/
  



   Scale*+ , - . / / / . / / /
    

DMINISTRATIVE MAP OF BURKINA FASO
0 /
 
  


  
 


A

12 - 11 -

Sketch-map No. 4 - Administrative divisions adjoining the frontier

ADMINISTRATIVE DIVISIONS ADJOINING THE FRONTIER


 

 
   

     





   



 



        








  



 









    















  

 
   

 

   



 





 



 







Map prepared for illustrative purposes only
  

 
 
  




 



National capital     

 

  

        !  " #




administrative centres


Frontier between the two

$  
 

 " 

States

 



Boundaries and administrative centres of

 

  

   "    
 #
Frontiers with third


$   "  
 "
 

regions and départements

countries

13 - 12 -

Sketch-map No. 5 - Administrative map of the Republic of Niger

 
  
ADMINISTRATIVE MAP OF THE REPUBLIC OF NIGER

Libya




     ! " " " ! " " "

 
    

        
National capital and urban community

              



International frontiers  

Algeria



Regional boundaries and
    

             

administrative centres

Administrative centres of

            

départements

  


  

  
    
 




  

  
          

   

        


Chad

 



    

    


 






        



   

    

 
    

    
   


 

    
    




    

        


        

 

  
 

   


   

 



      
  



     
 

Nigeria

       
  

  



    

Map prepared for illustrative purposes only

            

    
  
Benin

  
      



15 - 13 -

0.30. The frontier between Burkina and Niger lies between two other frontiers whose course
the International Court of Justice has previously defined . It begins in the north at the intersection

with the eastern end of the Burkina-Mali frontier, a tripoint whose location was reserved by the
Court in its Judgment of 22 December 1986 in the case concerning the Frontier Dispute (Burkina
Faso/Republic of Mali ) , and ends in the south at the River Mekrou, which forms the western

frontier between Benin and Niger in this area, as noted by the Court in its Judgment of
12July 2005 in the case concerning the Frontier Dispute (Benin/Niger). The Court also reserved
the precise location of the tripoint between the three States concerned in that case .

S ECTION 4
STRUCTURE OF THE M EMORIAL

0.31. The present Memorial comprises four chapters:

 Chapter I sets out the historical background to the case: both from a long-term perspective in
relating the circumstances in which the colonial boundaries between Upper Volta and Niger
were determined by the Arrêté of 31 August 1927 and its Erratum of 5 October 1927, and how

those boundaries survived Upper Volta’s turbulent history; and from a more recent perspective
in describing the discussions on the frontier held by the Parties after independence and prior to
seising the Court;

 in Chapter II, Burkina reviews the applicable law, focusing in particular on the principle of the
intangibility of frontiers inherited from colonization and the Agreement of 28March1987,

both expressly mentioned in Article 6 of the Special Agreement;

 Chapter III describes the task of the Court in respect of the demarcated portion of the

boundary; thus, it recounts the circumstances in which the Parties reached the “agreement”
placed on record in Article 2, paragraph 2, of the Special Agreement and describes the precise
object of that agreement, before going on to explain in greater detail what is expected of the

distinguished Court in that regard;

 finally, in Chapter IV, Burkina describes the response that it believes the Court should give to

the matter which lies at the heart of the present dispute and which is set out in the first
paragraph of the same Article 2; it describes the course of the frontier from the Tong-Tong
astronomic marker to the beginning of the Botou bend.

0.32. These four chapters are followed by Burkina Faso’s Submissions.

25
See above, para. 0.2.
2I.C.J. Reports 1986, pp.576-580, paras.44-50. In contrast to the 2005 Judgment (see following note), the
Chamber of the Court nevertheless determined the precise location of the endpoint of the frontier between the parties in

this case.
2I.C.J. Reports 2005, p. 151, para. 146, point 4. C HAPTER I

H ISTORICAL BACKGROUND

1.1. The present chapter aims to place thedispute submitted to the Court in its historical
context, both in the more distant past by recounting the circumstances in which the boundaries
between the French colonies of Upper Volta and Niger were determined, and more recently, by

describing the discussions on fixing the frontier held between the two countries from the time they
gained independence to the conclusion of the Special Agreement of 24February2009. It
establishes that the negotiations between Burkina and Niger on delimiting the frontier formalized

the delimitation resulting from the Arrêté of the Governor-General of FWA of 31 August 1927, as
modified by its Erratum of 5 October 1927 (section 2). That delimitation has survived the turbulent
history of Upper Volta and the cercles of the two French colonies of Upper Volta and Niger prior

to independence (section 1).

SECTION 1

FIXING THE COLONIAL BOUNDARIES BETWEEN
U PPER V OLTA AND N IGER

1.2. A Chamber of the Court has already had occasion to consider the delimitation of

Burkina’s frontier with another of its neighbours, the Republic of Mali. In its Judgment of
22December1986, the Chamber noted that the characteristic feature of the legal context of the
frontier determination it was to undertake was that:

“both States involved derive their existence from the process of decolonization which
has been unfolding in Africa during the past 30 years. Their territories, and that of

Niger, were formerly part of the French colonies which were grouped together under
the name of French West Africa (AOF). Considering only the situation which
prevailed immediately before the accession to independence of the two States, and
disregarding previous administrative changes, it can be said that Burkina Faso

corresponds to the colony of Upper Volta, and the Republic of Mali to the colony of
Sudan (formerly French Sudan).” 28

Another Chamber of the Court, for its part ruling on the frontier dispute between Niger and Benin,
stated that it was:

“set within a historical context marked by the accession to independence of the
territories that were formerly part of French West Africa (“Afrique occidentale
française”, hereinafter “AOF”). Benin, which has been independent since
1August1960, corresponds to the former colony of Dahomey, and Niger, which has

been independent since 3August1960, corresponds to a territory which underwent
various administrative transformations during the colonial period.”

1.3. Burkina and Niger, having both acceded to independence in 1960 (on 5 and
3Augustrespectively), emerged from territories colonized by France (1). Their frontier results
from the delimitations between the administrative divisions of the two colonies undertaken by the

former colonial power (3), in an often arbitrary manner, as the many “incarnations” of Upper Volta
almost comically demonstrate (2).

28
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 564, para. 19.
2Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 107, para. 20. - 15 -

1. The French colonial conquest

1.4. Together with Chad 30, Upper Volta and Niger are among the last territories to have been

colonized by France. The Judgments of the Chambers of the Court cited above provide a brief
summary of their respective colonial histories.

1.5. The 1986 Judgment describes the French colonial conquest and colonization of Upper
Volta in the following terms:

“For the purpose of determining in broad terms what for each of the two Parties
[Burkina and Mali] was the colonial heritage to which the uti possidetis was to apply,
the origins of the French colonies concerned will be briefly retraced. For this purpose,

however, it is unnecessary to go further back in the history of the colonies of French
West Africa than 1919. At that time, the present territories of Mali and Burkina Faso
both formed part of the colony of Upper Senegal and Niger. By virtue of a decree of

the President of the French Republic dated 1March1919, the cercles of Gaoua,
Bobo-Dioulasso, Dédougou, Ouagadougou, Dori and Fada N’Gourma, which had
until then been part of Upper Senegal and Niger, were established as a separate colony
31
with the name of Upper Volta[ ]. By a decree of 4December1920, the remaining
territories, comprising what was left of Upper Senegal and Niger, were given the name
of French Sudan, and by a decree of 13 October 1922 the Civil Territory of the Niger

became an independent colony. (…) [T]he decree of 1 March 1919 which had created
Upper Volta was rescinded by a decree of 5September1932, and the cercles which
had comprised Upper Volta were incorporated, in whole or in part, into Niger and into

French Sudan or the Ivory Coast. The Chamber refers to paragraph 73 below and to
sketch-map No.2[ 32], which shows the distribution of the cercles in the disputed
frontier region. Upper Volta was reconstituted in 1947 by the law 47-1707 of

4September 1947, which rescinded outright the decree of 5September1932 that had
abolished the colony of Upper Volta, and stated that the boundaries of ‘the
re-established territory of Upper Volta’ were to be ‘those of the former colony of

Upper Volta on 5September1932’. It was this reconstituted Upper Volta which
subsequently obtained independence on 5 August 1960, and took the name of Burkina
Faso in 1984.” 33

1.6. For its part, the 2005 Judgment of the Chamber of the Court described the evolution of

the legal status of the territories of Dahomey and Niger during the colonial period in the following
terms:

“33. In the second half of the nineteenth century, France initially established
settlements along the coast of Dahomey, at Cotonou and Porto Novo. Following an
armed conflict with the local chieftain in the 1880s and 1890s, it consolidated its

presence in the region first by placing Dahomey under protectorate (1892), and then
by creating the ‘colony of Dahomey and dependencies’ (decree of 22June1894).
France subsequently launched expeditions northwards from its possessions in

Dahomey, as well as southwards and eastwards from Sudan, which enabled it, in the

30
For a brief description of the colonization of Chad by France, see Territorial Dispute (Libyan Arab
Jamahiriya/Chad), Judgment, I.C.J. Reports 1994, pp. 17-18, paras. 24-32.
31
It is noteworthy that the Court, in omitting to mention the Say cercle, gives an incomplete list from the decree
of 1 March 1919.
32
See I.C.J. Reports 1986, p. 593.
3Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 569, para. 32. - 16 -

autumn of 1897, to occupy the valley of the River Niger (in particular the sector
between Say and Boussa).

The French occupation was expressly formalized, as regards the region of
north-western Dahomey, by a convention concluded with Germany on 23July1897,

and as regards north-eastern Dahomey, by a convention concluded with Great Britain
on 14 June 1898. By means of a convention of 8 April 1904, certain adjustments were
made to the line established in 1898 in order to separate the French and British areas
of influence. The parties to that convention fixed the boundaries of their respective

possessions in accordance with those adjustments by means of a convention of
29 May 1906 in respect of the region to the east of the River Niger, and by means of
an agreement of 19October1906 in respect of the territories between the Gulf of
Guinea and that river; demarcation operations, documented in an official record dated

19February1910, were subsequently carried out by the Anglo-French Commission
for the delimitation of the territories situated between the Niger and Lake Chad.

34. At the end of the nineteenth century, when the colony of Dahomey was

incorporated into the AOF by decree of 17October1899, it encompassed, in the
region concerned by the present dispute, territories situated on both banks of the River
Niger. By virtue of the same decree, which had provided for the dissolution of French

Sudan and the apportionment of the territories it had comprised among different
colonies and two specially created military territories, the territory of Say was also
attributed to Dahomey. This territorial incorporation was put into effect by an arrêté
of the Governor of Dahomey dated 20 March 1901.

By arrêté of 23July1900, the Governor-General of the AOF decided to
establish a third military territory encompassing the regions on the left bank of the
River Niger from Say to Lake Chad. That 1900 arrêté was followed by a decree of

the President of the French Republic dated 20December1900 with the same object.
The boundary between the Third Military Territory and the First Military Territory
created in 1899 was subsequently determined by an arrêté of the Governor-General of

the AOF, dated 20 March 1902.

By a decree of 18October1904 on the reorganization of the General
Government of the AOF, the President of the French Republic inter alia established

the colony of Haut-Sénégal et Niger comprising ‘the former territories of
Haut-Sénégal and Moyen-Niger and those which form[ed] the Third Military
Territory’. The newly created colony was composed of ‘ cercles under civil
administration’ as well as the ‘Military Territory of Niger’, constituted by the former

First and Third Military Territories.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The colony of Haute-Volta was abolished by decree of 5 September 1932, then
reconstituted with the same territorial basis by Law No. 47-1707 of 4 September 1947;
in the intervening period, the cercles of Fada and Dori (excluding the canton of
34
Aribinda) were incorporated into Niger.”

34
Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 112-113, paras. 33-35. - 17 -

1.7. Notwithstanding a number of omissions 35, these two Judgments provide some useful
historical insights and describe how France, starting with the trading posts it possessed in Senegal,

Côte d’Ivoire and Gabon, proceeded to occupy central Africa from the 1880s onwards. After the
Berlin Conference (15 November 1884 — 26 February 1885), France, Britain and Germany shared

out western Africa. The Franco-German Convention of 23 July1987 drew the boundary between
the French and German spheres of influence, whilst the Franco-British Convention of
14 June 1898, modified in 1904 and 1905, assigned the inland portion of the River Niger to France

and the delta to Great Britain. More specifically, the French and British zones of influence were
separated by a line running from Say (Niger) to Baroua (on the north-east shore of Lake Chad): the
territory to the north of that line belonged to France, and the territory to the south to Great Britain.

1.8. The Mossi (or Moose), the Gourmantché (Gulmance) and the Peuls (Fulbe) are the main

ethnic groups living in the Niger Bend. According to the work of Michel Izard, the present-day
Mossi have been established in the area since the 15th century, when Ouedraogo, the grandson of a
Dagomba King (an ethnic group from the north of present-day Ghana), came to Tenkodogo and
36
established the founding kingdom of the Mossi people . His descendants founded the kingdoms
of Ouagadougou (in the centre of the region) and Yatenga (in the north) between the 15th and

16thcenturies. Later on, in the 18th century, a fourth entity emerged alongside these three
kingdoms and, taking advantage of their decline, became established to the east of Ouagadougou:
the Mossi kingdom of Boussouma. According to MauriceDelafosse, the Mossi kingdoms were

characterized by their strength, their ethnic and linguistic homogeneity, and their great stability: 37
their independence only came to an end with French occupation at the close of the 19th century .
The royal dynasties — which had common roots — held power for several centuries, during which

time their political and social institutions demonstrated great continuity thanks to the cement of
religion (ancestor worship).

1.9. The birth of the Gourmantché kingdoms (Gourma being the right bank of the river, as
opposed to the “Haoussa” bank), to the east of present-day Burkina, also dates back to the

15th century and its founding ancestor, Diaba Lompo. On the eve of the European conquests, the
most influential of these kingdoms were Nungu in the centre (capital Fada N’Gourma), Bilanga
38
(bordering eastern Mossi country), Pama, Matiacoali and Koala (in the north) . The latter
kingdom in particular exerted a wide-ranging influence on the Peul lineages that settled in the Dori
region towards the beginning of the 18th century. However, once Usman Dan Fodio had founded

the Sokoto Caliphate in 1810, the Peul peoples sought his support against the Gourmantché. The
ensuing Jihad (holy war) led to the foundation of the Peul Emirates of Liptako (capital Dori) and
Yagha (capital Sebba), and finally put an end to the rule of the Gourmantché kingdom of Koala 39.
40
Nevertheless, internecine struggles among the Mossi principalities , fratricidal warfare among the

35It should be noted in particular that the 1986 Judgment makes no mention of the important decree of
28 December 1926, which is referred to, however, in the 2005 Judgment — see below, para. 1.16.

36See M. Izard, Introduction à l’histoire des royaumes mossi , Paris/CNRS-Ouagadougou, CVRS, 1970, Vol.1,
pp. 1-212.

37M. Delafosse, Haut-Sénégal-Niger (Soudan français), Première série. Tome II: l’histoire , Paris, Emile Larose,
1912, pp.122-124. See also the map of these empires, p.151, reproduced in Cartographic AnnexMBF1 and available
on the website Gallica.bnf.fr.

38See Y. G. Madiéga, Contribution à l’histoire précoloniale du Gulmu (Haute-Volta) , Wiesbaden, Frantz Steiner
Verlag, 1982, 260 pages.

39H. Diallo, Les Fulbe de la Haute-Volta de la fin du 18è siècle à la fin du 19è siècle , doctoral thesis, Université
Paris I, 1979, pp. 78-94.
40
V. K. Arhin and J. Ki-Zerbo, “ Etats et peoples de la boucle du Niger et de la Volta ”, in J. F. Ade Ajawi (dir.),
Histoire générale de l’Afrique, VI: l’Afrique au XIXè siècle jusque vers les années 1880 , UNESCO, Paris, 1996,
esp. pp. 119-723. - 18 -

Gourmantché kingdoms and conflicts of succession within the Peul Emirates enabled France to
take control of Voltan territory between 1888 (when contact was first made by Captain Binger,
known for his famous book Du Niger au golfe de Guinée par le pays de Kong et le Mossi 41) and

1897. France gradually established its dominance over the region, and the conclusion of
agreements with Germany and Great Britain formalized its conquest and occupation of these
territories42.

1.10. Immediately after its conquest of Upper Volta, France launched military expeditions

from Dahomey and French Sudan, and established its presence in Niger, thus gaining control of
both banks of the Niger in the zone of influence accorded to it under the agreements concluded
with Germany and Great Britain. Next, deploying to the east along the northern boundary of the

Sokoto Caliphate, which had been reserved for the British to conquer, the expeditionary force from
French Sudan conquered Songhai land — also known as “zaberma territories” (djerma) — and then
the Haoussa States of Gobir, Maradi and Damagaram, whose capital Zinder was captured in 1899.

The Joalland-Meynier column continued its push eastwards and entered Bornou where, on the
western shore of Lake Chad, it joined the Foureau-Lamy and Gentil columns, which had come

from North Africa and Gabon respectively. By the end of 1900, following victories against
Rabih— the warlord leading the resistance in Borno— and against the main Tuareg rebel
strongholds, the conquest of Niger was complete and the Military Territory of Niger established . 43

2. The incarnations of Upper Volta and the colonial subdivisions

1.11. It was only relatively late on that Upper Volta and Niger were established as “fully
fledged” colonies.

1.12. Once France had occupied both banks of the Niger Bend in the latter half of the 1890s,
three “military territories” (as opposed to civil cercles, because they were not completely pacified)

were created in the region:

 under Article 1 of the Arrêté of the Governor-General of FWA of 23December1899 44, the

First Military Territory comprised the cercles of Timbuktu, Sumpi, Bamba, Gao and Sinder
(which correspond to the southern part of present-day Mali) and the résidences of Dori, Macina

and Yatenga (which today cover the northern half of Burkina);

 the Second Military Territory, for its part, was established by the same Arrêté of

23 December 1899 and consisted of the “ cercles of Koutiala, Sikasso, Bobo-Dioulasso, Koury,
Diébougou . . . and the résidences of Mossi and Gourounsi” (Article 2); it corresponds roughly
to the south of Burkina;

41
L.-G. Binger, Du Niger au golfe de Guinée par le pays de Kong et le Mossi : 1887-1889, Hachette, Paris, 1892.
42See above, para. 1.7, and M. Delafosse, op. cit. note 36, pp. 420-423; see also P.-C. Hien, Le jeu des frontiers
en Afrique occidentale: Cent ans de situations conflictuelles au Burkina Faso actuel (1886-1986), doctoral thesis,

Université Paris I, 1996, . 80-207, and the map of French protectorate treaties in Upper Volta, in
J.-M. Kambou-Ferrand, Peuples voltaïques et conquête coloniale 1885-1914, L’Harmattan, Paris, 1993, p. 260.
43By way of example, see the map “ La conquête de l’Afrique de l’Ouest(1880-1900)”, in B. Lugan, Atlas
historique de l’Afrique des origins à nos jours, Ed. Du Rocher, Monaco, 2001, p. 156 (Cartographic Annex MBF 2); and

the map “ States and peoples of West Africa at the outset of European partiti” in J. D. Fage, An Atlas of African
History, London, Arnould, 2nd edition, 1978 (Cartographic Annex MBF 3).
44Ann. MBF 2. - 19 -

 the Third Military Territory was created at a later date by an Arrêté général of 23 July 1900 45,
46
the terms of which were confirmed by the Decree of 20December1900 : it encompassed
“the areas on the left bank of the Niger, from Say to Lake Chad, which [had been] placed under
47
the French sphere of influence by the Convention of 14June1898” and consisted of the
territories newly conquered by France and which had yet to be pacified; its administrative

centre was Zinder and it included most of the territories that form the present-day Republic of
Niger.

1.13. A decree of 18 October 1904 of the President of the French Republic, providing for the
48
reorganization of the General Government of FWA , established the Colony of Haut-Sénégal et
Niger, comprising “the former territories of Haut-Sénégal and Moyen-Niger, as well as those
forming the Third Military Territory” 49. This new colony was composed of “ cercles under civil

administration”— with no further details— and the “Military Territory of Niger”, consisting of
the former First and Third Military Territories.

1.14. As the Chamber of the Court explained in the 2005 Judgment in the Frontier Dispute
between Benin and Niger:

50
“By decree of 2March1907[ ], the cercles of Fada-N’Gourma and Say were
detached from Dahomey and incorporated into the colony of Haut-Sénégal et Niger.

The intercolonial boundary fixed by that decree was revised on two occasions in its
western part, first by a decree of 12 August 1909[ 51], and subsequently by a decree of
52
23 April 1913[ ].

35. On 7September1911, a further decree[ 53] separated the Military Territory

of Niger from the colony of Haut-Sénégal et Niger (the cercle of Say remaining as a
district of that colony), in order to make it an administrative subdivision under the

direct control of the office of the Government-General of the AOF. On 1 January
1921[ ], that military territory became the Civil Territory of Niger, and was then
55
made an autonomous colony by decree of 13 October 1922[ ].

In the meantime, the decree of 1March1919[ 56] had provided for the

establishment of the colony of Haute-Volta, to which were attributed, inter alia, the

45
Ann. MBF 3.
46
On this point, see ICJ Judgment of 12July2005, Frontier Dispute (Benin/Niger), I.C.J.Reports2005, p. 121,
para. 50.
47
Article 1 of the Arrêté general of 23 July 1900 (Ann. MBF 3).
48
French West Africa was created by decree of 16June1895 and initially comprised Senegal, French Sudan
(present-day Mali), Guinea and Côte d’Ivoire.
49
Ann. MBF 4.
50Ann. MBF 8.

51Ann. MBF 11.

52Ann. MBF 15.

53Ann. MBF 13.
54
Ann. MBF 18.
55
Ann. MBF 19.
56
Ann. MBF 16. - 20 -

cercles of Say and Fada-N’Gourma, which had hitherto formed part of Haut-Sénégal
57
et Niger.”

1.15. The invaluable Atlas of the cercles of FWA, “compiled and drawn at the Geographical
Service of FWA in Dakar under the direction of Commander Edouard de MARTONNE”, contains

representations of all the cercles of FWA at the time the Atlas was produced in 1926. The frontier
between Upper Volta and Niger was then formed by the River Niger, thus all the cercles concerned

in the present dispute, namely Dori (in the north-east), Fada— or Fada N’Gourma— (in the
south-west), and Say (in the south-east), were in what was then Upper Volta. The sketch-maps of
58
these cercles are reproduced in the annexes to the present Memorial .

59
1.16. “By decree of 28December1926[ ], certain cantons in the cercle of Dori and the
cercle of Say (with the exception of the canton of Gourmanché-de-Botou) were detached from
60
Haute-Volta and incorporated into Niger” . That decree is of particular importance for the
purposes of this case: up until that point, Upper Volta had extended to the River Niger. It was
established within its present-day boundaries (which were to be defined by the 1927 Arrêté and its
61 62
Erratum ) as from 1 January 1927 , since the Law of 4September1947 reconstituting Upper
Volta returned it to within its 1932 boundaries 63(which had remained unchanged since 1927) . 64

1.17. Regarding Niger, the Chamber of the Court states in its 2005 Judgment:

“The internal reorganization of Niger was the subject of successive arrêtés of

the Governor-General of the AOF, dated 26December1904, 31December1907,
14 December 1908, 22 June 1910, 23 November 1912 and 22 January 1927 65. On the

eve of independence, as a result of an arrêté général of 30 March 1956 adding seven
new cercles to the colony, Niger comprised 16 cercles.” 66

1.18. In 1932, Upper Volta was abolished altogether by the decree of 5Septemberand its
67
constituent territories were divided among Niger, French Sudan and Côte d’Ivoire . In the words
of the Chamber which ruled on the Frontier Dispute between Burkina and Mali, one of the effects

of the decree of 5 September 1932:

57Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 113, paras. 34-35.

58See Cartographic Annex MBF 4.

59Ann. MBF 26.
60
Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 113, para. 35.
61
See below, paras. 1.27 and 1.30.
62
Under Article 2 of the Decree of 28 December 1926: “The following territories, which are currently part of the
Colony of Upper Volta, are incorporated into the Colony of Niger, beginning from 1January1927, namely...”
(Followed by a list of the territories concerned.)
63
See below, para. 1.19.
64
Today, the border provinces are, from north to south: in Burkina Faso: Oudalan, Seno, Yagha, Komondjari and
Tapoa; and in Niger: the region of Tillabéri — or Tillabéry — (comprising the départements of Téra in the north and Say
in the south), whose western boundary forms the Burkina-Niger frontier.
65
Anns. 5, 9, 10, 12, 14, 29.
66
Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, pp. 113-114, para. 36.
67
Ann. MBF 36. - 21 -

“was the outright abrogation of the decree of 1March1919 which had created the

colony of Upper Volta, and hence the abolition of that colony. The new decree, which
came into force on 1 January 1933, also provided as follows :

‘Thrte. 2 cercles of Fada and Dori (except the canton of
Aribinda) are annexed to the colony of Niger.

The cercle of Ouahigouya, the canton of Aribinda within the cercle
of Dori and that part of the cercle of Dedougou located on the left bank of
the Black Volta are annexed to the colony of French Sudan . . .’

By an Order of the Governor-General of French West Africa dated
17November 1932, the territories of the colony of Upper Volta which had been
annexed to French Sudan by the above-mentioned decree were reorganized as follows :

T‘he. cercle of Ouahigouya, at present forming part of Upper
Volta, and the canton of Aribinda, detached from the cercle of Dori, are

to form a single unit under the na68 of cercle of Ouahigouya, with its
chief town at Ouahigouya . . .’.”

1.19. Upper Volta was reconstituted within its 1932 boundaries by Law47-1707 of
4 September 1947 , which abrogated the decree of 5 September 1932. Under Article 2:

“The re-established territory of Upper Volta shall have administrative and
financial autonomy under the same conditions as the other territories of the French
West African group.

Its administrative centre shall be at Ouagadougou and its boundaries shall be
those of the former colony of Upper Volta on 5 September 1932.”

As the Court noted, “[i]t was this reconstituted Upper Volta which subsequently obtained
independence on 5 August 1960 and took the name of Burkina Faso in 1984” . 70

1.20. As the Chamber of the Court underscored in its Judgment of 22 December 1986:

“The law of 4September1947 ‘for the re-establishment of the territory of
Upper Volta’ made no reference to any map; al1 it contained was a reference in
general terms to the boundaries ‘of the former colony . . .on 5 September 1932’.

Neither of the two Parties has been able to identify the map, if there was one, which
was used by the French lawmakers in 1947 in order to obtain a clearer picture of those
boundaries.” 71

Burkina has been unable to identify such a map for the purposes of these proceedings, just as
it was unable to do so at the time of the case which gave rise to that decision.

6Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports , p.592, para.73. See in
particular the sketch-map on page 593 of the Judgment illustrating the changes made to Dori cercle.

6Ann. MBF 39.
70
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 569, para. 32.
7Ibid., p. 583, para. 57. - 22 -

3. The colonial boundaries of Upper Volta and Niger

1.21. Since the boundaries between the two colonies remained unchanged between 1947 and

1960 when they gained independence, it must be ascertained what the boundaries between Upper
Volta and Niger were at the time the former colony was temporarily abolished on
5 September 1932.

1.22. A document found in the archives of IGN France in Paris provides a concise and
accurate summary of the situation. It offers an analysis of the Niger-Upper Volta frontier, dated

3November1972 and undertaken by Mr.BernardMillot, a cartographic artist who had been
appointed to the IGN’s map library at the beginning of the 1970s to collate documentation on the
frontiers between the States of the former “French Community”:

“Falling under the same administrative regime, these two territories were
governed by the system of arrêtés and decrees. Boundary problems were thus the

subject of numerous arrêtés, the first of which appears to date back to
18October1904. This was followed by others, such as the decrees of
13October1922 and 28December1926, which incorporated some of the cercles of

Upper Volta into Niger.

Arrêté No.184 of 22 January 1927[ 72] concerned the cercles and cantons of

Dori, Tillabéry, Say, Gourmantché Botou and Fada, and thus affected the boundary
between the two territories. The text of the Arrêté was published in the Journal
officiel of FWA of 12February1927 on page 124. Following the intervention of the

cercle commanders, Arrêté N7341 of 5March1927 was issued regarding the canton
of Gourmantché Botou[ ] and its attachment to the cercles of Fada and Say.
However, it was the Arrêté of 31August1927 which provided the first definition of

the frontier and more specifically that of the frontier cercles and cantons74It was
published in the Journal officiel of FWA No. 1201 of 24 September 1927[ ]. After it
was contested, an erratum to the Arrêté of 31August1927 was published and

appeared i75the76 Journal officiel of FWA No.1205 of 15October1927 on
page 718[ ].”

1.23. In fact, the 1927 Arrêté and its Erratum give both the first and last definition of the
frontier and it has been neither modified nor clarified by any document since that time.

1.24. Before then, the colonies and their constituent cercles or résidences were defined
simply by listing their respective subdivisions. As the Chamber of the Court noted in 2005:

“In reality, the 1919 decree defines the territory of Haute-Volta solely by
reference to the cercles which compose it, and it is thus also by this means that it

indirectly defines the boundaries between Haute-Volta and the neighbouring colonies,
and in particular Dahomey. It is by the precise delimitation of the cercles mentioned

7Ann. MBF 29.

7The wording is misleading and confuses ethnic group with administrative division. What should have been
written was “the Gourmantché canton of Botou”.
74
Ann. MBF 34.
75
Ann. MBF 35.
7Ann. MBF 58. - 23 -

in Article 1 of the decree of 1 March 1919 ņ a delimitation not effected by the decree
77
itself ņ that, from this date, the intercolonial boundary could be defined.”

1.25. Similarly, for Niger the Arrêté of 23 November 1912 reorganized the territory into

seven cercles but did not draw their boundaries, any more than did the Decree of
28 December 1926 incorporating Say and certain cantons of Dori cercle into Niger.

1.26. As from the Arrêté of 1927, however, it was no longer a matter of administrative
divisions but of the frontier. Once the Erratum was adopted, that frontier was established ne
varietur and those texts constitute the frontier title which has never been modified since.

1.27. According to the Arrêté of 31August1927 of the Governor-General ad interim of
78
FWA :

“Article 1— The boundaries of the Colonies of Niger and Upper Volta shall
henceforth be determined as follows:

1. Boundaries between the Tillabéry cercle and Upper Volta:

This boundary is determined to the north by the current boundary with Sudan (Gao
cercle) as far as the heights of N’Gourma, and to the west by a line passing
through the Kabia ford, Mount Darouskoy and Mount Balébanguia, west of the
ruins of the village of Tokébangou, and Mount Doumafondé, which then turns

towards the south-east, leaving the ruins of Tong-Tong to the east and descending
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.

2. Boundaries between the Say cercle and Upper Volta:

The villages of Botou canton are excluded from this boundary.

To the north and to the east, by the current boundary with Niger (Niamey cercle),

from Sorbohaoussa to the mouth of the River Mekrou;

To the north-west, by the River Sirba from its mouth as far as the village of
Bossébangou. From this point a salient, including on the left bank of the Sirba the

villages of Afassi, Kouro, Takalan and Tankouro;

To the south-west, a line starting approximately from the Sirba at the level of the

Say parallel and running as far as the Mekrou;

To the south-east, by the Mekrou from that point as far as its confluence with the
Niger.

77
Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 146, para. 134.
7Ann. MBF 34. - 24 -

3. Boundaries of Botou canton:

To the west: the furthest point is marked by the intersection of the Fada-Say road
with the former boundary of the two cercles and the Tiéguelofonou marigot. That
point is located 1,200 m west of the village of Tchenguiliba.

From that point, the boundary turns back up towards the north, running in a
straight line in a marked SSW-NNE direction.

It passes approximately 2 km west of the village of Berni-Oueli and terminates in

the north approximately 2km south of the village of Vendou Mama at the top of
the northernmost spur of the Héni-Djoari (Gourma) massif or Jackal Mountain.

To the north: the boundary runs in a marked west-east direction. It passes 1 km
south of Mount Tambado Djoaga, follows the course of the Dantiabonga marigot,

passes south of Dantiandou, follows the line of the Yoga Djoaga hills as far as the
confluence of the Dantiabouga and Diamoungou marigots, and continues along the
latter up to the confluence of the Diamoungou and Boulelfonou marigots

approximately 5 km north of the latter village;

To the north-east: the boundary follows the crests of the Djoapienga hills up to the
source of the Boulelfonou marigot, runs up the north slope of the Tounga Djoaga

massif and terminates at the point known as Niobo-Farou (Caiman Pool), a sort of
broad basin, which is traversed during the dry season by the track from Botou to
Fombonou;

To the east: the boundary follows the eastern crests of the Tounga Djoaga massif
and runs towards the River Tapoa in a precise north-south direction. It passes
approximately 5km east of the village of Royori (a relatively dispersed farming
village) and reaches the Tapoa at a point which it is not possible to define

precisely;

To the south-east and to the south: the boundary follows the course of the Tapoa
upstream until it meets the former boundary of the Fada and Say cercles.

This end-point cannot be defined, as the southern region of Botou is completely
empty, and virtually unexplored . . .”

1.28. This text, which draws the boundaries between Upper Volta (as defined by the Decree
of 28 December 1926) and Niger, has its distant origins in a circular letter dated 28 July 1920 from
the Lieutenant-Governor of Upper Volta, Hesling, to the Commanders of the Colony’s cercles. In
that letter he states that:

“The Territory of Upper Volta was constituted by the grouping together of a
number of cercles of Haut-Sénégal-Niger, without any changes being made to the
administrative structure of those divisions. As a result there are some overlaps among

the ethnic groups living along the boundaries of the cercles; this did not raise any
problems when the area of settlement to either side of those boundaries fell under the
responsibility of one and the same Governor, but creates more serious problems in the
case of the boundary between the two Colonies.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - 25 -

It is thus clearly worthwhile to make the necessary adjustments to the boundary
to resolve ethnic issues, whenever circumstances permit and particularly when such
partial adjustments would have no significant effect on the general administrative

structure of the colony. Adjustments should be made in all cases where a very small
or subsidiary group has been separated from the main groupement for no particular

reason: such as the Tuareg and the Oudala [sic] Bella in Dori cercle, the Bariba and
the Atakora Takomba in Fada cercle, the Bouna Lobi in Côte d’Ivoire, and the Peuhls
of the left bank of the Niger in the Military Territory.” 79

1Th9e. Arrêté of 31 August 1927 was adopted following various exchanges of letters and
80 81
telegrams and after the Lieutenant-Governors of Niger and Upper Volta had given their
agreement in principle. The latter gave his agreement in principle in a letter dated 1 February 1926
to the Governor-General of FWA 82, though he expressed some misgivings in the face of concerns

voiced by the local populations and laid down the condition that Botou canton should remain in the
territory of Upper Volta; his agreement was confirmed at a meeting of the representatives of the
Lieutenant-Governors of Upper Volta and Niger on 10 February 1927 . 83

1.30. The Erratum of 5 October 1927 84replaced the previous text of 31 August, cited above,

with the following:

“Article 1 of the Arrêté of 31 August 1927 fixing the boundaries of the Colonies

of Niger and Upper Volta, published in the Official Journal of French West Africa
No. 1201, of 24 September 1927, page 638, should read as follows:

Article 1— The boundaries of the Colonies of Niger and Upper Volta are
determined as follows:

A line starting from the heights of N’Gouma, passing through the Kabia ford
(astronomic point), Mount Arounskoye and Mount Balébanguia, to the west of the

ruins of the village of Tokebangou, Mount Doumafende and 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. It 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; then, turning back to

the south, it again cuts the Sirba at the level of the Say parallel.

79Ann. MBF 17.
80
See the letter dated 22March1923 from the Lieutenant-Governor of Upper Volta to the Commander of Dori
cercle (which suggested transferring to Niger the part of Dori situated on the right bank of the River Niger),
Ann. MBF 20; the reply dated 7 April 1923 from the Commander of Dori cercle (stating his hostility to the suggestion as
he did not want to lose Gothèye for financial reasons), Ann.MBF21; the letter dated 7June1923 (from the
Lieutenant-Governor of Upper Volta to the Governor-General of FWA), Ann.MBF22; the telegram of 13June1923
(from the Lieutenant-Governor of Upper Volta to the Commander of Dori cercle (there were no plans to transfer Dori),

Ann.MBF23; the letter dated 26January1926 from the Lieutenant-Governor of Niger to the Governor-General of
FWA, Ann. MBF 24.
81See espt.he draft Arrêté submitted on 27June1927 to the Governor-General of FWA by the

Lieutenant-Governor of Niger (Ann. MBF 33).
82Ann. MBF 25.

83Ann.MBF31. On that occasion they expressed their agreement on a line that prefigured that ofArrêté for
the northern part of the boundary — they were attempting to execute the Decree of 28 December 1926 — and decided to
leave it to the administrators of Say and Fada to fix the boundaries of Botou canton.

84Ann. MBF 35. - 26 -

From that point 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.

From that point it turns back up in a straight line that runs in a marked
SSW-NNE direction; it passes approximately 2km west of the village of Birniouoli
and, approximately 2km to the south of the south of the village of Vendou Mama,

reaches the top of the northernmost spur of the Heni-Djouri (Gourma) massif or Jackal
Mountain.

Running then in a west-east direction, it passes 1 km south of Mount Tambado

Djoaga, follows the course of the Dantiabonga marigot, passes south of Dantiandou,
follows the line of the Yoga Djoaga hills as far as the confluence of the Dantiabonga
and Diamongou marigots, and runs along the latter as far as the confluence of the
Dialongou and Boulelfonou marigots approximately 5 km north of the latter village.

From that point, the boundary follows the crests of the Djoapionga hills as far as
the source of the Boulolfonou marigot, runs up the northern slope of the Tounga and
Djoaga massif and terminates at the point known as Niobo-Farou (Caiman Pool), a

sort of broad basin, which is traversed during the dry season by the track from Botou
to Fombonou.

It is then determined by the eastern crests of the Tounga Djoaga massif, before

running towards the River Tapoa in a precise north-south direction. It passes
approximately 5 km east of the village of Kogori and reaches the Tapoa approximately
4 km south of the aforementioned village.

It then follows the course of the Tapoa upstream until it meets the 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.”

1.31. The differences between the two texts are as follows:

 whereas, in the north, the Arrêté refers to “the current boundary with Sudan (Gao cercle) as far

as the heights of N’Gourma [sic]”, the Erratum specifies “[a] line starting from the heights of
N’Gouma”; while in the first draft the tripoint with French Sudan is imprecise, in the second it
is clearly located on Mount N’Gouma;

 still in the northern sector, some points of reference are more precise: the frontier passes
through the Tong-Tong and Tao astronomic markers and reaches the River Sirba directly at
Bossébangou — this last clarification makes the phrase “near to and to the south of Boulkalo”
which appeared in the Arrêté redundant;

 the Arrêté was more verbose in respect of the boundaries of Say cercle: it did not merely
describe the parts of the boundary that form the frontier with Upper Volta, but described in
detail the entire boundary of the cercle, notwithstanding the fact that this was not the purpose

of the Arrêté as stated in its title, which was to “[fix] the boundaries of the Colonies of Upper
Volta and Niger”; however, the description of the course of the boundary between the two
Colonies is even more precise when it comes to the salient comprising the four villages;

 at the level of Botou canton, the frontier is defined in the same manner, with the exception of
the location where it first reaches the Tapoa— ra ther than speak of a “point which it is not
possible to define precisely” (Arrêté), according to the Erratum it is a point which is - 27 -

approximately four kilometres south of the village of Kogori (also cited in the Arrêté85);

finally, the Arrêté left the endpoint on the former Fada-Say boundary and gave no further
details (as it was deemed impossible to define, given that the region was empty and
unexplored), whilst the Erratum states that the line follows the former Fada-Say boundary up to

the point where it intersects with the Mekrou, and thus determines the tripoint with Dahomey.

1.32. As the Chamber of the Court noted in its Judgment of 12 July 2005:

“the erratum would seem in effect to have been motivated not by the fact that
the Governor-General did not mean to fix the south-eastern boundary of the cercle of

Say along the Mekrou, but rather by a wish not to define the boundary between
Dahomey and Niger in an arrêté whose purpose, as was clear from its title, was to fix
the boundary between Niger and Haute-Volta... [T]he erratum thus [did] not

contradict the fact that the boundaries of the cercle of Say were as indicated the
previous August . . .”86

1.33. The delimitation settled by the Erratum of 5 October 1927 has not been modified since.

SECTION 2
DISCUSSIONS ON FIXING THE FRONTIER

1.34. Upper Volta and Niger rapidly made contact after independence in order to establish
their common frontier in a clear and definitive manner. It immediately became apparent that the
matter they had to deal with together was basically a technical one, as the two States agreed from

the outset to consider that the “primary text” was the Arrêté of 31 August 1927 and its Erratum of
5Octoberof the same year. Just a few years after independence, the Protocol of Agreement of
23 June 1964 placed on record their shared vision and laid the foundations of a process intended to

lead to the mutually agreed demarcation of the frontier (1.).

1.35. Largely for financial reasons the process was not as rapid as the two States had initially

planned, while at the same time there were recurrent problems in the border zone due to the fact
that the course of the frontier had not been marked ou t on the ground. The Parties re-opened
discussions on demarcation when they adopted the Agreement and Protocol of Agreement of

28 March 1987 (2.).

1.36. This was not without effect on the demarcation process, since the campaign to mark
out the frontier, which was implemented by the Joint Technical Commission on Demarcation
established by the Protocol of Agreement of 28March1987, resulted in the installation of
87
23 markers out of an initially planned total of 45 . However, placing the remaining markers was
to prove far more problematic. Niger developed misgivings about implementation of the frontier as
described by the Arrêté and its Erratum. Thus in 1990 the Niger experts began putting forward
new claims, disputing some of what had hitherto been consensual and ascribing new meaning to the

words of the Erratum. This led the political authorities of the two countries to intervene once
again. Anxious to end the debate once and for all, the Ministers in charge decided to adopt an
amicable solution and to depart partially from the text of the 1927 Arrêté and its Erratum in such a

85
Although the Arrêté uses the toponym “Royori”, the wording of the sentence that mentions that village leaves
no doubt that it is referring to the village of Kogori.
86
Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 147, para. 136.
8Regarding the figure of 45 markers, see below para. 1.71. - 28 -

way as to meet Niger’s expectations, though without incorporating its latest interpretation of that
text. That is the thrust of the decision duly recorded in the Joint Communiqué of 16May1991.
Nevertheless, regardless of what it stood to gain, Niger denounced that decision shortly after its

adoption (3.).

1.37. The demarcation process initiated by the Protocol of Agreement of 28 March 1987 was

thus brought to a standstill. The positions of the Parties in respect of the demarcation of the
frontier took their final form inJuly2001, at the fourth ordinary session of the Joint Technical
Commission on Demarcation. Its conclusions reflect the situation as it stood at the time the Court

was seised: the Parties’ final agreement on the two sectors of the frontier already fully marked out,
and their disagreement as to the remainder (4.). The Special Agreement of 24February2009
seising the International Court of Justice marked the end of discussions on demarcation of the

frontier and the Parties have since placed themselves in the hands of the Court for a definitive
settlement of their dispute regarding the implementation of the Agreement of 28 March 1987.

1. The Protocol of Agreement of 23 June 1964

1.38. It was problems with registration that led Niger to approach Upper Volta very soon
after independence with a view to marking out the frontier.

1.39. Thus, on 7 June 1963, the President of the Republic of Niger sent an urgent letter to the
President of the Republic of Upper Volta, drawing his attention to the fact that a number of Voltan

families of the Gourmantché ethnic group living in Niger (Tamou canton) refused to be registered
there. In the interests of maintaining good neighbourly relations, Niger said it wished to avoid
creating tensions with its neighbour by expelling those families, but nevertheless insisted that they

should be registered, at least on a provisional basis pending a political solution to the problem. The
issue was purely fiscal, as, according to the Head of State of Niger, registration would be “without
electoral or military implications, [and] without prejudice to nationality” 88. Niger’s sole aim was

thus to subject the families in question to the country’s various taxes89nd charges, regardless of
their protests that they had already paid them in Upper Volta . He further suggested that in the
event the said families were effectively subject to double taxation, they should be removed from

the registration books in their canton of origin in Upper Volta.

1.40. In a “note on the frontier problems between the Republics of Niger and Upper Volta”

annexed to the President of Niger’s letter, a further issue was raised, namely the fact that people
from Téra, in Niger, tended to register in Dori, where the taxes were less onerous. In respect of the
frontier, the note also pointed out that it was not visible on the ground since the Tong-Tong and
90
Tao astronomic markers had disappeared , and recommended that a delimitation commission
should establish the frontier, “which is not to be modified but to be marked out” 91.

88Letter No.1052/M1/AI and Note from the President of Niger to the President of Upper Volta, 7June1963,
Ann. MBF 42.
89
Ibid.
90
Ibid.
91Ibid. - 29 -

1.41. On 12 September 1963, the Minister for the Interior of Upper Volta, wishing to be kept

informed of these matters, asked the cercle commanders to provide him with an objective account
of all the frontier problems that may have come to their attention . 92

1.42. On 17 June 1964, the Minister for Foreign Affairs of Niger took an initiative which this
time was to lead to the Protocol of Agreement of 23 June of the same year. In a Note Verbale to

the Minister for Foreign Affairs of Upper Volta, he stated that the matter of the frontier between the
two countries was simply a problem of demarcation, which could easily be settled by technical
experts:

“The frontier between Upper Volta and Niger is fixed by an Arrêté dated
31 August 1927 of the Governor-General and clarified by an erratum published in the
Journal Officiel of FWA No. 1201 of 24 September 1927, page 638.

Thus a fairly precise basic text exists, but the markers are lost, and it is currently
not possible for the representatives of the two Republics to locate the frontier on the

ground precisely.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

To settle all these arguments, a delegation from Niger led by the Head of Say
District and a Voltan delegation led by the Commander of Diapaga cercle met in the
village of Boulel (on the Niger-Volta frontier) on 10 April 1964.

While the questions of registration and those concerning the transhumance of
the Peulh, etc., were resolved, the problem of the frontier remains unchanged. In

accordance with the spirit of the Addis Ababa conference, the two delegations wished
to retain the boundary drawn by the colonial power, but requested that the two
Governments should appoint technical experts so that the frontier could be marked out

definitively.

Accordingly, the question is whether Upper Volta agrees that the frontier

established by colonization should be respected. If so, Niger suggests to Upper Volta
that the aim of the conference be confined to settling the questions of registration, tax
collection and transhumance between the cercles of Téra and Dori (Upper Volta), and
93
to appointing surveyors from both States to mark out the frontier.”

1.43. Upper Volta did not intend to call the 1927 Arrêté and its Erratum of 5October1927

into question; thus the Protocol of Agreement of 23June1964 was adopted very rapidly, and
signed on behalf of Upper Volta by the Minister for the Interior and Security, and on behalf of
Niger by the Minister for Saharan and Nomad Affairs and Post and Telecommunications . 94

1.44. This Protocol of Agreement — which the signatories wanted to apply immediately and
whose legal effect, in its own words, was to “[replace] all prior provisions which are inconsistent

therewith” (the last sentence of the Protocol of Agreem ent)— notes first of all that “there is no
indication on the ground of the theoretical boundaries between the two States”, and goes on to state

92
Circular letter No.34/PRES/IS/DI from the Minister for the Interior and Security of the Republic of Upper
Volta to the cercle commanders.
93
Note Verbale No.2216/MAE/DAPEC from the Minister for Foreign Affairs of Niger to the Minister for
Foreign Affairs of Upper Volta, 17 June 1964, Ann. MBF 44.
94Protocol of Agreement on the delimitation of the frontier, Niamey, 23 June 1964, Ann. MBF 45. - 30 -

that, in order to deal with the resulting situation, the two States had reached a number of

“decisions”. The first, relating to the delimitation of the frontier, states:

“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 Erratum 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
demarcation, starting in mid-November 1964 and beginning with the disputed sectors,

in particular the stretch of the frontier lying between Téra and Dori.”

1.45. In the eyes of the Parties, the reference texts for the purposes of determining the course

of the frontier were thus clearly the Arrêté of 31 August 1927 and the Erratum of 5 October of the
same year. The reference to the Paris IGN 1:200,000-scale map as a “basic document” can be
explained by the fact that it is the only map base precise enough to serve as a working document on

which the course described by the above-mentioned texts could be plotted.

1.46. The two Parties also undertook to allow their nationals to move freely from one side of

the frontier to the other, provided they were carrying identity papers, and to recognize the right of
all nationals of the other Party to enter their territory, travel within it and establish their residence
there, without being obliged to have a visa or permit of any kind, with the exception of

transhumants, who should have a transhumance certificate (point 2 of the Protocol of Agreement).
It was further d ecided to maintain the status quo in respect of land utilization rights until the
frontier had been demarcated (point 3 of the Protocol of Agreement). The Parties also agreed that

the security forces of both Parties should make no incursion into the neighbouring territory without
prior express authorization from the responsible authority, and that any border disputes would be
settled through meetings between delegates or by an exchange of Notes (point 4 of the Protocol of
Agreement). To give substance to the latter point, the heads of the border districts were authorized

to contact one another directly and meet on a regular basis in order to deal with issues of local
concern (point 5 of the Protocol of Agreement). Finally, the Protocol of Agreement includes
provisions relating to census-taking and the collection of taxes from sedentary nationals and

nomads (point 6 of the Protocol of Agreement).

1.47. The two Parties promptly sought to mark out the frontier in accordance with the
schedule announced in the Protocol of Agreement of 23 June 1964. The Topographic Service and
Land Registry of Niger contacted the IGN Annex in Dakar by letter dated 20 July 1964 in order to
obtain a 1:1,000,000 map showing the main astronomic points along the frontier 95, whilst his

counterpart in Upper Volta enquired of the same IGN Annex in Dakar whether it could mark out
the frontier by placing a marker approximately every ten kilometres, and how much such an
operation would cost 96. The demarcation process was thus set in motion. This was, moreover,

considered necessary, as, notwithstanding the Protocol of Agreement of 23June1964, there were
still major problems in respect of tax collection. From Niger’s point of view, many of its nationals

95
Letter No.1 10/STC from the Director of the Topographic Service and Land Registry of Niger to the Director
of the IGN office in Dakar, Upper Volta-Niger, 20 July 1964, Ann. MBF 46.
9Letter No.00517/CT from the Director of the Land Registry, Urban Planning and Housing of Upper Volta to
the Director of the IGN office in Dakar, 25 July 1964, Ann. MBF 47. - 31 -

continued to avoid paying the taxes due in their own country by opting to be registered in Upper
Volta where the taxes, particularly on animals, were more favourable . 97

1.48. The actual process of demarcation, however, was not put in place as rapidly as the

Parties had originally intended, given the considerable cost of such an undertaking over a frontier
approximately 590km in length 98. Thus, the problems on the ground persisted, in particular in

respect of tax collection, as the President of Niger pointed out in a letter dated 6March1967,
whereby he suggested to his counterpart in Upper Volta that efforts should be made to implement
the Protocol of Agreement 99. On 16March1967 the Minister for the Interior and Security of

Upper Volta responded by promptly sending a circular to the Commanders of the cercles bordering
Niger, asking them to send him all the relevant documents and information, with a view to a
100
meeting of the Joint Commission on Demarcation of the Frontier at the end of April .

1.49. Nevertheless, no headway was made in the demarcation process and border relations
remained difficult at times, invariably on account of problems of tax collection. The Niger
authorities were often under the impression that Niger nationals were trying to avoid their civic

duties, and sometimes pursued them into Upper Volta without taking the precaution of obtaining
the prior agreement of the neighbouring authorities, as nevertheless provided under point 4 of the

Protocol of Agreement of 23 June10164. During one such incident, a Niger officer lost his rifle in
the territory of Upper Volta .

1.50. It was no doubt to avoid the risk of an incident caused by one of its officers seriously
affecting relations between the two countries that Niger again took the initiative three years later,

this time through its Minister for Foreign Affairs. By Note Verbale dated 7November1967, the
Minister for Foreign Affairs of Niger proposed a meeting with his counterpart from Upper Volta,
102
with a view to clarifying some of the terms of the 1964 Protocol of Agreement . The Minister for
Foreign Affairs of Upper Volta promptly agreed to the proposed meeting, which he felt was
“highly desirable in order to put an end to the numerous incidents that plague the border zone year

in, year out”, and indicated that it could take place on 5 January 1968, leaving it to Niger to choose
the venue 103.

1.51. That meeting eventually took place on 9 and 10 January 1968 in Niamey. The Niger

delegation was led by the Minister for Saharan and Nomad Affairs and Post and
Telecommunications, and that of Upper Volta by the Minister for the Interior and Security. As
indicated in the joint communiqué issued at the end of the meeting, the Parties noted that the

97Letter No. 32/MAE/DAPEC from the Minister for Foreign Affairs of Niger to the Minister for Foreign Affairs
of Upper Volta, 6 May 1965, Ann. MBF 48.
98
See the analysis of the Niger-Upper Volta frontier undertaken by IGN France, 3 November 1972,
Ann. MBF 58.
99
Letter No.0288/PRES/CAB from the President of Niger to the President of Upper Volta, 6March1967,
Ann. MBF 49.
100
Circular No.18/IS/DI from the Minister for the Interior and Security of the Republic of Upper Volta to the
Commanders of the cercles adjoining Niger (Diapaga, Fada N’Gourma, Bogande, Dori and Oudalan), 16March1967,
Ann. MBF 50.
101
Letter No. 8/CONF-DR from the Commander of Dori cercle to the Minister for the Interior,
3 November 1967, Ann. MBF 52.
102
Note Verbale No.02697/MAE/DAPEC from the Minister for Foreign Affairs of Niger to the Minister for
Foreign Affairs of Upper Volta, 7 November 1967, Ann. MBF 51.
103
Letter No. 23421/IS/DI from the Minister for the Interior to the Minister for Foreign Affairs of the Republic of
Upper Volta, 23 November 1967, Ann. MBF 53. - 32 -

Protocol of Agreement had not been fully implemented “in particular at the level of the heads of
the border districts of both countries”. They therefore agreed to clarify certain points, particularly
in respect of population movements, taxes, and various other issues. Regarding the frontier, the

record states:

“Following an exchange of correspondence between the Government of the
Republic of Upper Volta and the Institut Géographique National de Paris, it has been
agreed to entrust the latter body with the task of demarcating the frontier. The cost

has been estimated at some 10 million CFA, to be borne equally by the Governments
of Niger and Upper Volta.

As the Niger Government has already paid its share of the budget for the current
year, the Head of the Delegation of Upper Volta has undertaken to refer this matter to
his Government at the earliest opportunity in order to obtain the necessary funds.” 104

1.52. However, for financial reasons, the dema rcation process was not restarted, and the

meeting did not put an end to the difficult border relations between the two Parties. Among other
examples of these problems, the Deputy Commander of Dori cercle sent an official telegram dated
1 March 1968 to the Minister for the Interior, informing him that: “Niger authorities, in breach of
105
protocols, infiltrate Yatacou sectors to collect taxes without knowledge of Dori authorities.”

1.53. This was not an isolated incident and many such others occurred, mostly in connection
with Niger taxing individuals considered by local administrators to be Niger nationals or liable to

pay various charges. Some incidents were more serious than others, however. Niger was fully
aware of some of these incidents, as, in a letter dated 13 March 1970, the Upper Volta Minister for
Foreign Affairs informed the Minister for the Interior and Security of his country that his Niger

counterpart had officially expressed regret for an inciden106aused by the intervention of Niger
customs officers in the territory of Upper Volta . Other such incidents led the local Niger
authorities to acknowledge their failure to comply with the undertakings given 107.

1.54. Although the question of demarcation was complicated at local level, no progress was

made until 1982, and incidents were usually resolved through series of meetings between the local
authorities.

1.55. The principle of establishing a joint commission to identify the documents relating to
the boundary and demarcate the frontier was reiterated on 16 September 1982 at a working session

of delegations led by the Minister for the Interior and Security of Upper Volta and the Secretary of
State for the Interior of Niger 10. Several months later, on 25 April 1983, a Treaty of Friendship
and Co-operation was signed between the Republic of Upper Volta and the Republic of Niger, as

104Report of the ministerial meeting between Niger and Upper Volta, Niamey, 9-10 January 1968, Ann. MBF 54.

105Official telegram No.17/DR from the Deputy Commander of Dori cercle to the Commander of Dori cercle,
1 March 1968, Ann. MBF 55.

106Letter No.00446/AE/SG/AAM from the Minister for Foreign Affairs of Upper Volta to the Minister for the
Interior and Security, 13 March 1970, Ann. MBF 56.

107Mission report No.86/DR-CONF from the Commander of Dori cercle (Upper Volta) to the Minister for the
Interior and Security of the Republic of Upper Volta, 30October1972, Ann.MBF57; Record of the meeting between
the sous-préfet of Say (Niger) and the sous-préfet of Sebba (Upper Volta), 12 November 1981, Ann. MBF 59.

108Report of the working meeting between the Minister for the Interior and Security of Upper Volta and the
Secretary of State for the Interior of Niger, Niamey, 16 September 1982, Ann. MBF 60. - 33 -

109
well as an Agreement establishing a Niger-Volta Joint Commission for Co-operation . These
documents were ratified in accordance with the law of Burkina by decree of 6 May 1983 110.

1.56. In February 1985, following a further ministerial meeting, the two Parties agreed that,
while most of the many incidents along the frontie r had hitherto been resolved by the local

authorities, it was still necessary to “revive the Joint Commission for Niger-Burkina Co-operation”,
and recommended that “the frontier between the two States be demarcated in accordance with the

provisions of the Protocol of Agreement of 23June1964”. In their recommendation, the Parties
called on their respective authorities to establish “a joint commission responsible for the
demarcation of the frontier between the two States” 111.

1.57. In the meantime, the conduct of the authorities on the ground remained confrontational.

Furthermore, on 6 May 1985, the Minister for Territorial Administration of Burkina was obliged to
send a letter of protest to his Niger counterpart pointing out to him:

“At the ministerial meeting held in Ouagadougou on 12, 13 and
14 February 1985, problems relating to security and peace in the gold-bearing regions
of Deba (Niger), Tangangari and Takatami (Burkina) were discussed. It was agreed in

particular that peace and security should be maintained in that area and more generally
that the Niger-Burkina frontier should be demarcated on the basis of the provisions of
the Protocol of Agreement of 23 June 1964.

I have the honour to inform you that on 18 February 1985, namely less than one

week after the said meeting, a number of new pickets were again erected in the
gold-bearing sector of Tangangari by a Niger technical service.

This operation, performed intentionally without prior notice to the Burkina
authorities, would appear to be provocative.

I would like to take this opportunity to ask you to intervene with the department
and border officials concerned in order to suspend the operation pending the creation
and establishment of the Joint Commission on Demarcation of the Frontier, pursuant

to the recommendations of the meeting held in Ouagadougou from 12 to
14 February 1985.” 112

In his reply of 14June1985, the Niger Minister denied the facts, stating that at no point
during the meeting of local officials held on 11 April 1985 had there been “any mention of pickets
113
being erected in Tangangari or Takatami” . The problem nevertheless persisted in Takatami, a
gold-bearing site coveted by Niger — which explains why Niger persists in considering it to be one
and the same village as Tokalan 114. A radio message of 8 February 1986 from the Commander of

109
Ann. MBF 61.
110Decree No.83-0207/CSP/PRES/AE.C ratifying the Treaty of Friendship and Co-operation between Upper

Volta and Niger and the Agreement Establishing a Niger-Volta Joint Commission on Co-operation, 6 May 1983,
Ann. MBF 62.
111Report of the meeting between the Minister Delegate for the Interior of Niger and the Minister for Territorial

Administration of Burkina Faso, Ouagadougou, 12-14 February 1985, Ann. MBF 63.
112Letter No.001198/MATS/CAB/SG/CF from the Minister for Territorial Administration and Security of
Burkina Faso to the Minister Delegate for the Interior of Niger, 6 May 1985, Ann. MBF 64.

113Letter No.518/MI/MDI/DAPA/CF from the Minister Delegate for the Interior of Niger to the Minister for
Territorial Administration and Security of Burkina Faso, 14 June 1985, Ann. MBF 65.

114See below para. 1.77 and paras. 4.115-4.119. - 34 -

the territorial brigade of the Sebba gendarmerie to his superiors stated: “honour to report stop
placing of wooden stakes on 6/2/86 Niger-Burkina border zone stop south-east side of Sebba
département stop by Niger geologists consisting of five (5) men one (1) of whom white stop
115
gold-bearing site Tangangary” .

1.58. A further meeting at ministerial level was held in Niamey in April 1986. It dealt in
particular with the arrangements for setting up the Joint Commission on Demarcation of the

Frontier, but the ministers went much further than they had at previous meetings where they had
merely recommended action, and drew up a draft treaty and protocol of agreement on the
demarcation of the frontier between the two countries. This draft was to form the basis of the 1987
116
Agreement and Protocol of Agreement .

1.59. The process of demarcating the frontier was once again underway. The experts met
from 21 to 23 May 1986 in order to estimate the cost of the work and, in the words of the report
they adopted, to proceed “to interpret the Arrêté of 31 August 1927, as clarified by its Erratum of
117
5 October 1927, in light of the 1:200,000 map” . However, strictly speaking, it was not a matter
of “interpreting” the Arrêté and its Erratum; in reality these texts did not raise any concerns
between the Parties, which, in keeping with their demarcation mission, sought only to undertake a

“map interpretation”, an operation which consisted in plotting on the map the description of the
frontier given by the text, and relying on the map itself if need be, in accordance with the Protocol

of Agreement of 23June1964. Thus, in respect of the locations of Mount N’Gouma, Mount
Arounskoye, Mount Balébanguia, the point to the w est of the ruins of Tokébangou and Mount
Doumefende, the list of defining points on the Niger-Burkina frontier drawn up by the experts and

annexed to the rep118 of the meeting carries the note: “co-ordinates to be measured on a
1:200,000 map” .

1.60. Shortly thereafter, the Burkina Minister for Territorial Administration and Security sent
a letter to his Niger counterpart, informing him that the President of Burkina had agreed to the

demarcation of the common frontier between the two countries, but added that it was nevertheless
preferable for the work to begin only once the judgment by the International Court of Justice in the
then ongoing case between Burkina and Mali had been rendered 119.

2. The Agreement and Protocol of Agreement of 28 March 1987

1.61. The Agreement and Protocol of Agreement of 28March1987 were signed by the
respective Ministers for Foreign Affairs 120. They definitively establish the delimitation of the

frontier. Article 21 of the Protocol of Agreement provides that it enters into force upon signature,

115Radio message No.11/4 from the territorial unit, Sebba gendarmerie, to Dori company, Seno High

Commissioner, 8 February 1986, Ann. MBF 66; see also Report No. 24/4 from the unit commander, Sebba gendarmerie,
to the company commander, Dori, 20 March 1986, Ann. MBF 67.
116Report of the meeting between the Minister for Territorial Administration and Security of Burkina Faso and the

Minister Delegate for the Interior of Niger, 9 April 1986, Ann. MBF 68; see below, paras. 1.61 et seq.
117Report of the meeting between technical experts of Niger and Burkina, Ouagadougou, 21-23 May 1986,
Ann. MBF 69.

118Ibid.

119Letter No. 0603/MATS/SG/DFR from the Minister for Territorial Administration and Security to the Minister
Delegate for the Interior of Niger, 3 October 1986, Ann. MBF 70.
120
Protocol of Agreement between Niger and Burkina Faso on the demarcation of the frontier, 28March1987,
Ann. MBF 73; Agreement of 28 March 1987 on the demarcation of the frontier between Burkina Faso and the Republic
of Niger, Ann. MBF 72. - 35 -

and Article 22 states that the Parties shall agree to settle by peaceful means any disputes which may

arise in its implementation. The Agreement includes similar provisions in its Articles 9 and 8.

1.62. The first two articles of these instruments are drafted in an almost identical manner 12:

Article 1 of both the Agreement and the Protocol of Agreement concern the delimitation of the
frontier; Article 2 sets out the principles applicable to its demarcation.

122
1.63. Under Article 1 of the Protocol of Agreement :

“The frontier between the two States shall run from the heights of N’Gouma,

situated to the north of the Kabia ford, to the intersection of the former boundary of
the cercles of Fada and Say with the course of the Mekrou, as described in the Arrêté
of 31 August 1927, as clarified by the Erratum of 5 October 1927.”

This text marks the agreement between the two States on the line of the frontier, and
describes the frontier in two ways. First, it notes the general course of the frontier (it “shall run
from the heights of N’Gouma . . . to the intersection of the former boundary of the cercles of Fada

and Say with the course of the Mekrou”). Then it states that the frontier is “as described in the
Arrêté of 31 August 1927, as clarified by the Erratum of 5 October 1927”.

1.64. Article 2 of the Protocol (identical to Article 2 of the Agreement) has a different
subject-matter. It establishes the method the Parties undertake to follow in marking out the

frontier. Thus, the frontier:

“shall be demarcated by boundary markers following the course described by
Arrêté 2336 of 31 August 1927, as clarified by Erratum 2602/APA of 5 October 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.”

1.65. The following articles of the Protocol of Agreement provide the framework for the
actual demarcation work:

“Article 3. There is hereby established a Joint Technical Commission on
Demarcation. It shall be composed of an equal number of members from each of the

two Parties. At its meetings, the Chair shall be taken by the host country and the
function of Rapporteur shall be filled by the visiting country.

Article 4. The Joint Technical Commission on Demarcation shall be

responsible for marking the frontier, in accordance with the provisions of Articles1
and 2 above, following the entry into force of this Protocol. It shall present a report
every six months to the Governments of the two States on the progress of works on the

ground.

Article 5. The Joint Technical Commission on Demarcation shall enjoy full
autonomy in executing the works of demarcation. However, any difficulties liable to

121
Except for the reference to the Erratum, which appears in Article 1 of the Protocol and not in Article 1 of the
Agreement. This difference is probably due to a drafting error. On this point, see below, paras.2.26 and 2.27 of the
present Memorial.
12Article 1 of the Agreement as cited in the Special Agreement seising the Court is identical. - 36 -

hinder the conduct of the works, as well as important decisions with potential financial
consequences or involving additional costs, must be submitted to the two
123
Governments.”

1.66. On that basis, a first meeting between technical experts from Niger and Burkina took
place in Ouagadougou from 14 to 17 May 1987, and the financial, organizational and
administrative aspects of the Joint Commission on Demarcation’s working procedures were
124 125
broached . On 19June1987, the President of Burkina adopted a kiti designating the
individuals who were to sit on the Joint Technical Commission on Demarcation of the Frontier 126.

The Commission’s first meeting, held in Niamey from 25 to 27June19127 dealt with matters
similar to those dealt with at the meeting of May 1987 . The next meeting, which had been
scheduled for October 1987, could not take place owing to a number of events in both Burkina and
128
Niger, notably the death of President Kountché .

1.67. Finally, on 9 March 1988 a delegation of 42 experts from both countries began visiting
various locations along the frontier to conduct topographical work 129. It was not possible to resolve
all the problems. At their meeting of 12 to 15 May 1988, the members of the Joint Commission

noted that the technical sub-committee had encountered a problem in determining the course of the
frontier:

“Two points were discussed at length, based on additional information
provided. These concerned, first, the geographic location of the village of

Takabangou and, second, the location of the frontier line from the Say parallel to
another village named Tyenkilibi or Tyenkiliba.

The Commission noted that the line along this part of the frontier was based on
that of the French National Geographic Institute (IGN France) 1:200,000 map, not on

Arrêté No.2326 of 31August1927, as clarified by its Erratum of 5October1927,
both of which were designated in the agreement signed by the two Governments
in March 1987 in Ouagadougou.

The technical specialists explained that their position was based on the frontier
line as recorded in the report of the meeting between specialists from Niger and

Burkina held from 21 to 23 May 1986 in Ouagadougou.

It was apparent that this line was an interpretation of the above-mentioned

Arrêté and Erratum. The Commission considered that the technical staff were not
authorized to adopt a procedure that deviated from the decisions of the two

Governments. They were accordingly instructed to reconsider the 110km portion in

123
Ibid.
124Report and recommendations of the meeting between technical experts from Niger and Burkina Faso held in

Ouagadougou from 14 to 17 May 1987, Ann.MBF74; see also the Report of the mission conducted from 24 to
28 June 1987 in Niamey in connection with the demarcation of the frontier, July 1987, Ann. MBF 77.
125“Kiti” is the term used under the revolutionary regime of Burkina Faso for “decree”.

126Kiti of 19June1987 of the President of Burkina Faso appointing the members of the Joint Commission on
Demarcation, Ann. MBF 75.

127Report of the first meeting of the Joint Technical Commission on Demarcation held in Niamey,
25-27 June 1987, Ann. MBF 76.

128Note No.ANV/36/103/MAT/SG/DGAT/DAF from the Director of Border Affairs to the Minister for
Territorial Administration of Burkina Faso, 19 January 1988, Ann. MBF 78.

129Radio message from the Director of Border Affairs on mission in Téra, March 1988, Ann. MBF 79. - 37 -

question within eight (8) days, complying with the texts designated in the Agreement
130
and Protocol of Agreement signed by the two Governments.”

1.68. This criticism of the method used by the experts in 1988, who had followed in the

footsteps of those of the 1986 experts, was due to the fact that the latter had relied on the
cartographic line established by the Paris IGN— mainly in order to establish the co-ordinates of
pertinent geographic points — not because the Arrêté and its Erratum did “not suffice”, but because

the 1964Protocol of Agreement they had to implement gave them a relatively free hand and did
not subject use of the map to inadequacies in the texts, and also because it was more convenient: it
was thus easier to rely on the map than to go to the locations concerned in order to determine the

co-ordinates of the defining points, especially in light of the budgetary situation. The 1988 experts,
on the other hand, had no choice but to comply with the letter of the 1987 Protocol of Agreement,
which they failed to do by merely repeating the by then outdated findings of 1986, notwithstanding

the fact that they could have visited the sites in question.

1.69. The Joint Commission met again in Niamey on 26, 27 and 28September1988, this

time in order to plot on a map, in accordance with the 1987 Protocol of Agreement: “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”. This is what they
131
were to do by formally adopting what they agreed to call the “consensual line” . The report of
their meeting states that:

“1. According to the basic texts ( Arrêté No. 2336 of 31 August 1927 and its

Erratum No 2602/APA of 5October1927) the boundary line passes three (3)
categories of points:

 astronomic markers;

 natural features;

 localities.

The experts are thus 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 28 March 1987.

Only the Tokébangou point, which could not be identified despite numerous
investigations, was subject to interpretation by the Technical Sub-Committee. Thus,
the Joint Technical Commission proposes the following line:

‘The frontier point described as passing to the west of the ruins of
Tokébangou shall be the midpoint in the straight line connecting the site
indicated by the villagers of Dolbel and Kossa and the iron marker post

placed at the foot of Mount Kombara, a hill located to the west of the
presumed site of the former village of Tokébangou as indicated by the
inhabitants of said village.’

13Report of the meeting of the Joint Technical Commission on Demarcation, Diapaga, 12-15 May 1988,

Ann. MBF 80.
13See Cartographic AnnexMBF15: this map was produced in two copies signed by the Chairmen of the
Commission; during the meeting, it was decided that each party should submit the course of the line to its Government
for approval (see Report No. 42/FP/MAT/SG/DCAF from the Minister for Territorial Administration to the Head of State
of Burkina Faso, 5 March 1991, Ann. MBF 88). - 38 -

2. Comparison of the line on the IGNF map and the line resulting from the field
survey, in accordance with the basic texts.

It should be noted that at several locations the frontier line resulting from the
basic texts does not coincide with the line on the IGNF 1:200,000 map or with certain
132
administrative realities on the ground.”

1.70. The Joint Commission had accomplished the work required by the 1987 Protocol of

Agreement, in accordance with its terms, and demarcation work therefore commenced on that
basis, and indeed with what were initially satisfactory results. At its meeting of 18 May 1989, the

Joint Technical Commission on Demarcation was able to draw up a list133 the co-ordinates of
thirty-two defining points which was to enable demarcation .

1.71. The original intention was to place a marker every five kilometres, and even every
kilometre in densely populated areas. However, that ambition was scaled down in the light of

budgetary constraints and eventually it was agreed to adopt the principle of placing 45markers.
According to the report of the fifth meeting of the Joint Technical Commission on Demarcation,
held in Niamey on 13 and 14 November 1989, it was thus possible to place a marker at every

change of direction (of which there were thirty-three) in the frontier line, and to add a further
twelve markers — “consolidation points” — in densely populated areas 134.

1.72. The procedure thus adopted was efficient. Moreover, in November 1989, the Joint
Technical Commission on Demarcation meeting in Niamey noted with satisfaction that:

“1. — The field survey of the entire frontier (approximately 600 km) took place
in 1988. A marker post was placed every 20 km.

2. — Field observations were carried out by the Joint Technical Sub-Committee

in 1989. A total of 8 reference markers were recorded and their co-ordinates
calculated by IGN France.

In the W Park, it was difficult to conduct observations because of the dense
vegetation, but it was easier in the north.

It 135uld be noted that the first marker was erected between Botou and
Tamou.”

1.73. The record of the second ordinary meeting of the Joint Technical Commission on
Demarcation of the Frontier held in Ouagadougou from 23 to 28 July 1990 was also able to report

good res136s, with twenty-three markers out of a planned total of forty-five having effectively been
erected .

132
Report of the fourth meeting of the Joint Technical Commission on Demarcation of the Niger-Burkina frontier,
held in Niamey on 26, 27 and 28 September 1988, Ann. MBF 81.
133
Report of the meeting of the Joint Technical Commission on Demarcation of 18 May 1989, Ann. MBF 83.
134Report of the fifth meeting of the Joint Technical Commission on Demarcation of 13 and 14 November 1989,
Ann. MBF 84.

135Ibid.

136Report of the Joint Technical Commission on Demarcation of the Niger-Burkina Frontier, Ouagadougou,
23-28 July 1990, Ann. MBF 87. - 39 -

1.74. However, in certain sectors the process came to a halt. The report from the Burkina
Minister for Territorial Administration to the Head of State marks the point at which Burkina

became fully aware of Niger’s misgivings:

“As ofFebruary1990, when the Niger Chairman of the Commission was in

France for training and had been temporarily replaced by Niger’s Topographical
Director— the counterpart of the Director-General of the Burkina Geographic
Institute — certain tensions began to arise, in particular among the technical experts in
137
the field.”

In April 1990, the Niger Chairman ad interim of the Commission voiced his reservations
about the course of the boundary, as noted by the extraordinary session of the Commission
convened in Niamey in May 138.

1.75. Thus, it was during the extraordinary meeting of the Joint Technical Commission on

Demarcation of 14 May 1990 that Niger made clear its intention to call into question the
consensual line of 1988 by all possible means. At the time it relied on an argument that it was
never to repeat thereafter. Its then short-lived belief was that:

“the course of the boundary[ 139] was no longer consistent with the above-mentioned
140
texts[ ] from the Tao marker to Bossébangou. He thus pointed out that an error in
interpretation of the Erratum brought the frontier to the River Sirba at Bossébangou.

However, the correct interpretation of the Erratum produced a salient, leaving to
Niger the villages of Alfassi, Kouro, Tokalan, and Tankouro. In support of its
position, Niger presented a map on a scale of 1:1,000,000 entitled ‘French West

Africa, new frontier between Upper Volta and Niger based on the Erratum of
5 October 1927 to the Arrêté of 31 August 1927.” 141

1.76. The consensual line of 1988 was thus rejected by Niger as being inaccurate, because it

ended the frontier at the River Sirba at Bossébangou, whereas in fact it should never reach that
point, since— according to Niger— in order to form the salient, the frontier had to deviate
towards the west well to the north of Bossébangou. That position was supported by a sketch-map
142
purportedly representing the frontier line resulting from the Erratum of 5October1927 , but
which is obviously inconsistent with the Erratum, since, for example, the line it proposes does not
reach the River Sirba at Bossébangou as is very clearly indicated in the Erratum. Furthermore, this

obvious inconsistency led Niger to argue that the sketch-map should be considered as being
officially annexed and “linked” to the Erratum and that, in its view, it was preferable to rely on the

sketch-map rather than on the letter of the Erratum which, according to Niger, “includes

137
Report No.42/FP/MAT/SG/DCAF from the Minister for Territorial Administration to the Head of State of
Burkina Faso, 5 March 1991, Ann. MBF 88.
138
Ann. MBF 87.
139
Namely the consensual line of 1988.
140Namely the Arrêté and the Erratum.

141Report of the extraordinary meeting of the Joint Technical Commission on Demarcation, 15 May 1990,
Ann. MBF 85.

142The sketch-map is reproduced in Cartographic Annex MBF 6. - 40 -

descriptions of143e internal boundaries of the Tillabéry and Say cercles” at the level of
Bossébangou .

1.77. That thesis was never put forward again thereafter, but the Niger members of the
Commission on Demarcation nevertheless remained hostile to the consensual line, with the result

that at the end of July 1990 the Joint Technical Commission on Demarcation of the Frontier could
do no more than take note of the two Parties’ different positions. On that occasion, Niger put
forward two completely new interpretations of the 1927 Arrêté and its Erratum: (i) it claimed that

the word “turns” [ s’infléchit in French] as used in the Erratum implied that the section of the
boundary between the Tong-Tong astronomic marker and the River Sirba at Bossébangou
described a curve; and (ii) Takalan and Takatami were allegedly one and the same village, and

therefore the expression “at the level of the Say parallel” as used in the Erratum should be
considered as being “merely indicative” 144. Burkina, for its part, maintained a position that was not
only consistent with what it had always held, but also with what had prevailed throughout the
145
discussions up until 1989, and which had moreover resulted in the consensual line of 1988 .

1.78. A further meeting was held inFebruary1991, but as the participants were unable to
bridge such clearly irreconcilable positions, all they could do was agree to refer the matter for
ministerial arbitration 146. In an attempt to end the deadlock, the Minister for Territorial

Administration of Burkina had already taken the initiative to submit the matter to the Minister for
the Interior of Niger, expressing his concern that the discussions were becoming confrontational.
Indeed, as the Burkina Minister stated in his report to the Head of State, at the time his Niger

counterpart seemed: “not . . . to understand the position of his compatriot, the Chairman ad interim
of the Commission” 147.

3. The decision of 15 May 1991 and its denunciation by Niger

1.79. At the ministerial consultative and working meeting of 14 and 15 May 1991, the
Minister for the Interior of Niger and the Minister for Territorial Administration of Burkina decided
on a political solution to break the deadlock in the work of the Joint Commission on Demarcation.

According to that decision:

“1. From the Tong-Tong astronomic marker to the River Sirba at Bossebangou,

passing through the Tao astronomic marker, the frontier shall consist of a series of
straight lines.

2. From the River Sirba at Bossebangou to the River Mékrou, the course of the
frontier adopted shall be that shown on the map to a scale of 1:200,000 of
IGN/France, 1960 edition.

143
Report of the extraordinary meeting of the Joint Technical Commission on Demarcation, 15 May 1990,
Ann. MBF 85.
144
Report of the Joint Technical Commission on Demarcation of the Niger-Burkina Frontier, Ouagadougou,
23-28July1990, Ann.MBF87; see also the account of Niger’s position on the different sections of the frontier,
28 July 1990, Ann. MBF 87.
145Ibid.

146That meeting is referred to in the Joint Communiqué of 16 May 1991 adopted following the Ministerial
consultative and working meeting of 14 and 15 May 1991, Ann. MBF 89.

147Report No.42/FP/MAT/SG/DCAF from the Minister for Territorial Administration to the Head of State of
Burkina Faso, 5 March 1991, Ann. MBF 88. - 41 -

To that end, two extracts from the 1:200,000 map of IGN/France, 1960 edition, on
which that course is shown have been initialled by both Ministers.

The Joint Technical Commission on Demarcation has accordingly been authorized
to continue its work forthwith on that basis.” 148

1.80. This solution is clearly not a straightforward application of the 1987 Protocol of
Agreement, which would have resulted in the consensual line of 1988 criticized by Niger. It was a
political solution aimed at giving partial satisfaction to Niger in order to restart the demarcation

process as quickly as possible: (i) the consensual line was reconfirmed in the first section of the
frontier, which consisted of straight lines; but (ii) from the River Sirba at Bossébangou, it was

decided, by way of compromise, to have recourse to the line shown on the 1960 map, even though
this would give rise to certain inconsistencies with the text of the Arrêté of 31 August 1927 and its
Erratum of 5 October of the same year.

1.81. On that basis, the Technical Commission held an extraordinary meeting, at which it
149
was planned to complete the demarcation work by the end of 1992 . However, work on the
ground was interrupted, as was noted by the Joint Technical Commission on Demarcation of the
150
Frontier meeting in Niamey on 2 No151ber 1994 , and acknowledged by the Ministers at a
meeting on 6 December 1997 . Once again the hold-up was due to the Niger experts, who, at a
meeting of the Joint Technical Commission on Demarcation from 2 to 4 November 1994 152, called

into question the solution arising from the decision of 16 May 1991 on the grounds, or rather on the
pretext, that it “was not entirely consistent with the terms of Articles 1 and 2 of the 1987 Protocol
153
of Agreement” — which was only to be expected given that it was a political compromise.

4. The conclusions of the fourth ordinary session of the Joint Technical
Commission on Demarcation (2001)

1.82. The ministerial recommendations made in 1997 to restart discussions were without
effect. It had indeed become a somewhat problematic undertaking after Niger had in turn
renounced the consensual line unanimously adopted by the experts in 1988 and the political

compromise of 1991. The Ministers nevertheless met again in May 2000 in an attempt to give
fresh impetus to the process by “strongly recommending a meeting of the Joint Technical

Commission on Demarcation of the Frontier in September 2000”. Furthermore, in their view it was
all the more pressing as the situation was once again giving rise to “administrative problems on the
frontier, particularly at locations between Téra and Falagountou, Botou and Say” 154.

148Ministerial consultative and working meeting between Niger and Burkina, Joint Communiqué of 16 May 1991,

Ann. MBF 89.
149Ministerial Council Report on the results of the meeting of ministers held in Niger on 2July1991,
Ann. MBF 90.

150Report of the Joint Technical Commission on Demarcation, 2 November 1994, Ann. MBF 91.

151Report of the meeting between the Minister for the Interior of Niger and the Minister for Territorial
Administration of Burkina Faso, 5-6 December 1997, Ann. MBF 92.

152Report of the Joint Technical Commission on Demarcation of the Niger-Burkina Frontier, Niamey,
2-4 November 1994, Ann. MBF 91.

153Report of the Joint Technical Commission on Demarcation, 2 November 1994, Ann. MBF 91.
154
Report of the meeting between the Minister for the Interior of Niger and the Minister for Territorial
Administration of Burkina, Tenkodogo, 24-26 May 2000, Ann. MBF 93. - 42 -

1.83. The fourth ordinary session of the Joint Technical Commission on Demarcation of the

Burkina-Niger Frontier was held in July 2001. It established a joint commission of 12 members to
review the theoretical line of the frontier. This review was conducted:

“in light of the basic documents, specifically:

 the Agreement and Protocol of Agreement of 28 March 1987,

 the Arrêté général of 31 Augus1 t 927, as clarified by its Erratum of
5 October 1927”.

1.84. The Commission’s conclusions on the frontier were as follows:

“1. The frontier was clearly defined from the heights of [Mount] N’Gouma to the
astronomic marker of Tong-Tong, with the exception of the ruins of Tokébangou,

which the frontier passes to the west. These ruins were not identified in the course
of the survey of the frontier line.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2. The frontier was clearly defined from Tchenguiliba to the River Mékrou, subject
to the survey team’s verification of the position of the village of Kogori.

3. From the Tong-Tong astronomic marker to the River Sirba at Bossébangou, the
phrase ‘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’ has resulted in two interpretations:

(a) the frontier is composed of two (2) straight lines:

 from the Tong-Tong astronomic marker to the Tao astronomic marker;

 from the Tao astronomic marker to the River Sirba at Bossébangou.

(b) the frontier consists of a curved line, starting from the Tong-Tong astronomic
marker, passing through the Tao marker and terminating at the River Sirba at

Bossébangou.

The Commission decided to continue the documentary research in order to clarify
the course of the frontier in this area.

4. From Bossébangou to Tchenguiliba, the Commission noted problems of
interpretation associated with the failure to identify the villages referred to in the
Erratum and with the identification of the point at which the frontier again cuts the

River Sirba at the level of the Say parallel. The technical survey team will also
visit the area in order to identify these villages or their 1927 sites. The villages
concerned are Alfassi, Kouro, Tokalan and Tankouro.” 155

15Report of the Joint Technical Commission on Demarcation of the Burkina-Niger Frontier, Ouagadougou,

18-20 July 2001, Ann. MBF 94. - 43 -

1.85. It is clear that the experts were of exactly the same opinion in respect of the first two
sectors mentioned in this report, and which were in fact no longer really at issue, since they had

actually been marked out: from the heights of N’Gouma to the Tong-Tong astronomic marker the
frontier was “clearly” defined. The same was tr ue of the portion from Tchenguiliba to the River
Mekrou in the south, which was also “clearly” defined. That consensus was not called into

question thereafter. The Parties formalized this when they signed the Special Agreement seising
the International Court of Justice, whereby:

“thanks to the work of the Joint Technical Commission on Demarcation..., the

Parties have been able to reach agreement in respect of the following sectors of the
frontier:

(a) from the heights of N’Gouma to the astronomic marker of Tong-Tong;

(b) from the beginning of the Botou bend to the River Mekrou;

Whereas the two Parties accept the re sults of the work carried out in those
sectors as definitive.” 156

1.86. However, no progress was made with work on the other sectors, neither from the

Tong-Tong astronomic marker to the River Sirba at Bossébangou, nor from Bossébangou to
Tchenguiliba.

1.87. The respective Ministers nevertheless called for work to recommence. The final
communiqué of 4 April 2003 concluding their consultative meeting in Téra suggests:

“performing the tasks recommended at the technical meeting in Ouagadougou
in 2001, namely:

 a joint survey of the frontier line;

 continuing documentary research;

 holding the next technical session in Niger;

157
 replenishing the joint fund” .

Further, at the fifth ordinary session of the Joint Technical Commission on Demarcation
158
of September 2004 , it was decided to appoint a field survey team to locate in particular the ruins
of the village of Tokébangou and the villages of Kouro, Alfassi, Tokalan, Tankouro and Kogori.
That decision was not implemented, however.

156
Ann. MBF 99.
15Final communiqué of 4 April 2003, Ann. MBF 96.

15Report of the Joint Technical Commission on Demarcation of the Niger-Burkina frontier, Niamey,
28-30 September 2004, Ann. MBF 97. - 44 -

1.88. The two Parties decided to turn to the International Court of Justice so that it might
both “place on record [their] agreement on the results of the work of the Joint Technical
Commission on Demarcation of the . . . boundary” 159, and settle matters relating to the frontier line

in the section from the Tong-Tong astronomic marker to the beginning of the Botou bend. The
Court is seised on the basis of Article 8 in particular of the Agreement of 28 March 1987, which is
expressly referred to in the preamble to the Special Agreement. This article states that:

“The Parties agree to submit all disputes or differences arising out of the
implementation or interpretation of this Agreement to modes of peaceful settlement, in

particular those provided for in the Charters of the Organization of African Unity and
of the United Nations.”

The dispute relating to the determination of the frontier line in the disputed sector can thus

be considered as a dispute as to the application of the Agreement of 28 March 1987.

15Special Agreement seising the International Court of Justice, Article 2.2, Ann. MBF 99. CHAPTER II

A PPLICABLE LAW

2.1. Under Article 6 on “Applicable law” of the Special Agreement of 24 February 2009:

“The rules and principles of international law applicable to the dispute are those
referred to in Article 38, paragraph 1, of the Statute of the International Court of
Justice, including: the principle of the intangibility of boundaries inherited from
colonization; and the Agreement of 28 March 1987.”

2.2. The reference to sources listed in Article 38, paragraph 1, of the Statute of the

International Court of Justice requires no particular comment, except to point out, as did the
Chamber of the Court in its Judgment of 22December1986, that while the determination of the
course of the boundary between two sovereign States is clearly a matter for international law, in

cases involving countries which emerged from the same French colonial empire, the question must
also be appraised

“in the light of French colonial law, ‘droit d'outre-mer’. Since the territories of the two
States had been part of French West Africa, the former boundary between them
became an international frontier only at the moment when they became independent.
The line which the Chamber is required to determine as being that which existed in

1959-1960, was at that time merely the administrative boundary dividing two former
French colonies, called territoires d'outre-mer from 1946; as such it had to be defined
not according to international law, but according to the French legislation which was
160
applicable to such territoires.”

Nevertheless, in accordance with the warning conveyed in the same Judgment, there should

be no mistake as to the scope of French colonial law:

“The principle of uti possidetis freezes the territorial title; it stops the clock, but

does not put back the hands. Hence international law does not effect any renvoi to the
law established by the colonizing State, nor indeed to any legal rule unilaterally
established by any State whatever; French law— especially legislation enacted by

France for its colonies and territoires d'outre-mer — may play a role not in itself (as if
there were a sort of continuum juris, a legal relay between such law and international
law), but only as one factual element among others, or as evidence indicative of what
has been called the ‘colonial heritage’, i.e., the ‘photograph of the territory’ at the
161
critical date.”

2.3. In the light of this observation, Burkina Faso will briefly recall in section 1 of this
chapter the scope of the principle of the intangibility of boundaries inherited from colonization. In
section 2, it will present the other sources of applicable law, with particular emphasis on the 1987

Agreement and how it relates to other relevant instr uments for the purposes of settling the present
dispute.

160
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 568, para. 29.
16Ibid., p.568, para.30; see also Frontier Dispute (Benin/Niger), Judgment, I.C.J.Reports2005p. 110,
para. 28, and p. 120, para. 46. - 46 -

SECTION 1
T HE PRINCIPLE OF RESPECT FOR COLONIAL BOUNDARIES

2.4. It is usual to refer to the principle of the intangibility of colonial boundaries in cases of
this kind. To mention only those cases submitted to this Court by African States: in the preamble

to the Special Agreement of 16September1983, the Republics of Upper Volta and Mali stated
their desire “to achieve as rapidly as possible a settle ment of the frontier dispute between them,
based in particular on respect for the principle of the intangibility of frontiers inherited from
162
colonization” ; and Article 6 on applicable law of the Special Agreement of 15June between
Benin and Niger, which is written in terms very similar to those of the Special Agreement whereby
Burkina and Niger seised the Court in the present dispute, stated:

“The rules and principles of international law applicable to the dispute are those
set out in Article 38, paragraph 1, of the Statute of the International Court of Justice,

including the principle of State succession to the boundaries inherited from
colonization, that is to say, the intangibility of those boundaries.” 163

2.5. The scope of this principle, often referred to by the Latin phrase uti possidetis juris 16,
can therefore be ascertained fairly precisely in the light of the Court’s jurisprudence which has

shaped it, particularly in respect of, firstly (1), the methods for implementing the uti possidetis juris
principle— what could be called the “methodology”—, and, secondly (2), the relationship
165
between title (here consisting of the 1927 Arrêté as amended by its Erratum ) and — colonial and
post-colonial — effectivités.

1. The “methodology” of uti possidetis

166
2.6. According to the well-known dictum of the 1986 Judgment , for the purposes of
applying the principle of the intangibility of boundaries inherited from colonization, it falls to the
Court to identify “the ‘photograph of the territory’ at the critical date”.

2.7. In these proceedings, there is no particular difficulty in determining the critical date — it

may be more precise to speak of the “first critical date”: Upper Volta (Burkina Faso as of 1984)
and Niger gained independence two days apart, on 5 and 3August1960 respectively. As no
significant event took place between those two dates, they therefore constitute together “the”
167
critical date at which the “colonial heritage” can be established . There should be no
disagreement between the Parties as to this point.

2.8. However, as no text was adopted establishing the frontier between them at the time of
their accession to independence, it is first necessary to step back in time to ascertain when (and if)

the boundary was established, and, secondly, to ensure that “downstream”, since that date, it has
not been modified. These two tasks are relatively simple in the present case, in contrast to those
cases involving disputes between States born out of decolonization where “ uti possidetis speaks for

16Ibid., p. 557.

16Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 96.
164
See the previous Judgment (note 160) of 1986, Reports 1986, pp. 565-566, paras. 20-23.
165
See below, paras. 2.16 and 2.37.
16Prev. para. 2.2.

16See the previous Judgment (note 160) of 1986, Reports 1986, p. 570, para. 33. - 47 -

once with an uncertain voice” 168: in this case, the boundary between the Parties was fully defined
169
in Arrêté général 170 2336 of 31 August 1927 which was superseded by the Erratum of
5 October 1927 , and it has never been modified since.

2.9. It is true that the principle of the “intangibility” of colonial boundaries, as embodied in
171
the OAU’s Cairo Resolution of 1964 and in Article 4 (b) of the Constitutive Act of the African
Union of 2000, is not absolute in so far as it is “obviously open to those States to vary the
boundaries between them by agreement” 172. But in this case, the Parties have always considered

their common frontier to be that which existed at the time of their accession to independence and
that the frontier line in question could be determined by reference to the line described in the
173
1927 Arrêté and its Erratum . It is significant in this regard that the Agreement of
28 March 1987 174— referred to expressly in the Special Agreement and cited as a source of law
175
that can be applied by the Court — stipulates that “[t]he frontier between the two States” is “as
described in the Arrêté of 31August1927, as clarified by the Erratum of 5October1927”. Far
from reviewing the boundary established by the Arrêté, the aim of the Agreement is, on the

contrary, to mark it out.

2.10. In contrast to other cases where the application of the principle of the intangibility of
colonial frontiers was called into question, here, the uti possidetis principle can speak with a certain

voice: a document issued by the Governor General of FWA describes in full the course of the
boundary between the colonies of Upper Volta and Niger. It has neither been modified nor called

into question since it was adopted. Furthermore, in the Agreement of 28 March 1987 cited in the
Special Agreement, both Parties officially considered that the present frontier, in its entirety, was

described in the 1927 Arrêté as clarified by its Erratum.

176
2.11. Of course, Upper Volta was dissolved by the decree of 5September1932 and its
territory divided among the neighbouring French colonies 177. But it was re-established by Law
178
No. 47-1707 of 4 September 1947 which abrogated the 1932 decree and expressly stated that the
boundaries of the “re-established territory of Upper Volta” would be “those of the former colony of
Upper Volta on 5 September 1932” 179.

2.12. As of 1960, the following questions should therefore be asked:

168Land, Island and Maritime Frontier Dispute (ElSalvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992, p. 386, para. 41.

169Ann. MBF 34.
170
Ann. MBF 35.
171
Resolution AHG/Res. 16 (I) adopted at the first session of the Conference of Heads of State and Government
of the OAU, meeting in Cairo from 17 to 21 July 1964.
172
Land, Island and Maritime Frontier Dispute (ElSalvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992, p. 408, para. 80; see also, p. 401, para. 67.
173
See below, para. 2.20 et seq.
174
Ann. MBF 72.
175See above, para. 2.1.

176Ann. MBF 36.

177See above, paras. 1.5 and 1.18.
178
Ann. MBF 39.
179
See above, paras. 1.5-1.6, 1.16 and 1.19. - 48 -

 what was the course of the frontier on that date? It was that of the reconstituted Upper Volta of
1947;

 what were the boundaries in 1947? They were “those of the former colony of Upper Volta on
5 September 1932”; and

 what were those boundaries in 1932? They had been described in the 1927 Arrêté and its
Erratum.

Those are the boundaries that constitute the frontier between Burkina and Niger today and

those two texts— to which the 1987 Agreement and Protocol accordingly refer— constitute the
legal title which has force of law between them 18.

2. Determination of the uti possidetis line (the relationship
between title and effectivités)

2.13. Since a clear and uncontested title exists— in this case, the 1927 Arrêté as clarified
(and partly modified) by its Erratum— the question of the relationship between title and
“effectivités”— a question which has played a significant role in a number of cases heard

previously by the Court — is of minor importance.

2.14. The jurisprudence on the question was established by the dictum of the Chamber of the
Court in the case concerning the Frontier Dispute (Burkina/Mali) , which has repeatedly been
affirmed since. Stating “in general terms, what legal relationship exists between such acts and the

titles on which the implementation of the principle of uti possidetis is grounded”, the Chamber
distinguishes four eventualities:

“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 other 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 effectivité can then play an essential role in showing how the title is
interpreted in practice.” 181

2.15. It emerges from this very logical and clear classification, which has frequently been
confirmed since 182, that, where a frontier title exists, effectivités can play a confirmatory role only,

and that, if the effective administration does not correspond to the title, “preference should be given
to the holder of the title”. As the Chamber of the Court recalled in the clearest possible terms in the
Benin/Niger case: “To determine the course of the intercolonial boundary at the critical date it is

180
On the relationship between the Arrêté and the Erratum, see above, paras.1.30 to 1.32; below, paras.2.41
et seq.; on the relationship between the Agreement and the Protocol, see below, paras. 2.25 to 2.28.
18Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, pp. 586-87, para. 63.

18Land, Island and Maritime Frontier Dispute (ElSalvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992 , p.398, para.61; see also, p.401, para; Territorial Dispute (Libyan Arab Jamahiriya/Chad),
Judgment, I.C.J. Reports 1994 , p. 38, para. 75-76Land and Maritime Boundary between Cameroon and Nigeria
(Cameroon v. Nigeria: Equatorial Guinea intervening), Judgment, I.C.J. Reports 2002 , p.353, para.68Sovereignty
over Pulau Ligitan and Pulau Sipadan (Indonesia/Malaysia), Judgment, I.C.J. Reports 2002, p. 678, para. 126. - 49 -

necessary to examine first the legal titles relied on by the Parties, with any effectivités being
considered only on a confirmatory or subsidiary basis.” 183

2.16. In the present case, there is no doubt that such a title exists: it is the Arrêté of 1927 and
184
its Erratum . It therefore prevails over any effectivités to the contrary, which could only create a
fait accompli in breach of the title on which the other Party may rely in a given area. This applies

to coloni185effectivités (“i.e., the conduct of the administrative authorities during the colonial 186
period” ), as much as to post-colonial ones — those that occurred after independence .

2.17. This position applies all the more forcefully in this case, given that, by the Protocol of
Agreement of 23June1964 relating to the demarcation of their common frontier 187, the Parties,

having “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 Paris188stitut Géographique National ”, had already decided to freeze
the situation on the ground . Further, in Article 10 of the Special Agreement they recalled their
“Special Undertaking”

“[p]ending the Judgment of the Court,... [to] undertake to maintain peace, security

and tranquillity among the populations of the two States in the frontier region,
refraining from any act of incursion into the disputed areas and organizing regular
meetings of administrative officials and the security services”.

2.18. The primacy of title was vividly confirmed in Cameroon v. Nigeria. In that case,

Nigeria relied on certain activities, both in the north in the Lake Chad area and in the south in the
Bakassi peninsula, which

“could normally be considered to be acts à titre de souverain. The Court notes,
however, that, as there was a pre-existing title held by Cameroon in this area of the

lake, the pertinent legal test is whether there was thus evidenced acquiescence by
Cameroon in the passing of title from itself to Nigeria” 189.

190
And, noting that Cameroon held the legal title to the territory , the Court concluded that
Cameroon had sovereignty, as “the situation was essentially one where the effectivités adduced by

Nigeria did not correspond to the law, and... accordingly ‘preference should be given to the

183
Frontier Dispute (Benin/Niger), Order of 16February2005, I.C.J.Reports2005, p.143, para.128; see also
p. 149, para. 141.
184
See below, paras. 2.36 et seq.
185
Land, Island and Maritime Frontier Dispute (ElSalvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992, p. 398, para. 610.
186
See esp. ibid, p. 399, para. 62.
187Ann. MBF 45.

188See above, Introduction, para. 0.5 and paras. 1.38 to 1.46.

189Land, Island and Maritime Frontier Dispute (ElSalvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992, p. 353, para. 67; see also p. 415, para. 223.
190
See ibid., p. 353, para. 68 ; see also pp. 415-416, para. 223. - 50 -

holder of the title’ (Fron191r Dispute (Burkina Faso/Republic of Mali), Judgment, I. C. J. Reports
1986, p. 557, para. 63)” .

2.19. Such titles thus take precedence over any claim to the contrary, no matter what kind of
effectivités they may be based on, including insidious possession of a territory, be it long-term or
not; titulus est justa causa possidendi id quod nostrum est 192.

SECTION 2

S OURCES OF APPLICABLE LAW

2.20. Not only did the Parties to the present dispute not derogate from these well-established

rules, they even specified that they should be applied in this case by agreeing to consider that the
1927 Arrêté and its Erratum describe their common frontier in its entirety. In effect, in the
Agreement of 28March1987 on the demarcation of the frontier, they expressly enumerated the

sources of law applicable for those purposes (1). Thereby, they established the pre-eminence of the
frontier title constituted by the Arrêté of 31 August 1927, as clarified by its Erratum, over any other
evidence of the frontier line between the two countries (2).

1. The agreement of the Parties on relevant sources

2.21. In addition to the principle of the intangibility of frontiers inherited from colonization
(which is a norm rather than a source), the Parties expressly invoked in the Special Agreement of
24February2009 a specific source of law that can be applied by the Court: “the Agreement of

28 March 1987”.

2.22. Th193ircumstances under which this Agreement was adopted are set out in Chapter I of
this Memorial .

2.23. The first two articles of this binding instrument make express reference to the Arrêté of
31 August 1927 of the Governor General of FWA. The provisions read thus:

“Article 1

The frontier between the two States shall run from the heights of N’Gouma,

situated to the north of the Kabia ford, to the intersection of the former boundary of
the cercles of Fada and Say with the course of the Mekrou, as described in the Arrêté
of 31 August 1927.

Article 2

The frontier shall be demarcated by boundary markers following the course
described by Arrêté2336 of 31August1927, as clarified by Erratum2602/APA of
5 October 1927. Should the Arrêté and Erratum not suffice, the course shall be that

shown on the 1:200,000-scalemap of the Institut Géographique National de France ,

191
Ibid., pp.354-355, para.70 and p.416, paras.223-224; see also Territorial Dispute (Libyan Arab
Jamahiriya/Chad), Judgment, I.C.J. Reports 1994, p. 38, para. 76.
192
V.R. Jennings, The Acquisition of Territory in International Law, Manchester U.P., 1963, p. 5.
19See paras. 1.58 to 1.60. - 51 -

1960edition, and/or any other relevant document accepted by joint agreement of the

Parties”.

2.24. Besides the final provisions, the other clauses of the Agreement concern the creation of
the Joint Technical Commission on Demarcation, which is tasked with demarcating the frontier
(Art. 3), and a fund for the demarcation of the frontier (Art. 4).

2.25. The Agreement stricto sensu is supplemented by a Protocol of Agreement signed on
the same day, which sets out the arrangements for the demarcation work. That instrument—

which is inseparable from the Agreement— also refers back to the 1927 Arrêté and its Erratum,
which are cited in the first paragraph of the preamble:

“The Government of the Republic of Niger and the Revolutionary Government
of BurkinaFaso, having signed a Protocol of Agreement on 24June1964 in Niamey
covering in particular the delimitation of their joint frontier, on the basis of Arrêté

général No.2336 of 31August1927, as clarified by Erratum No.2602/APA of
5 October 1927 and the 1:200,000-scale map of the Institut Géographique National de
France; . . .”

2.26. Articles 1 and 2 of the Protocol are written in very similar terms to the corresponding
provisions in the Agreement. However, the fullest expression of the Parties’ agreement on the

delimitation of their frontier is in the first two articles of the Protocol, which are a more rigorous
version of the first two articles of the Agreement. While Article 2 of the Protocol of Agreement is
identical to Article 2 of the Agreement, Article 1 of the Agreement is less precise than Article 1 of
194
the Protocol , since it merely states that the frontier is “as described in the Arrêté of
31August1927” and fails to include the words “as clarified by the Erratum of 5October1927”.
That is clearly a drafting error and of no particular significance.

2.27. If the Parties had meant to rely only on the Arrêté and not on the Erratum to establish
their frontier, they would obviously not have mentioned the Erratum in Article 2 of the Agreement,

nor in Articles1 and 2 of the Protocol of Agreement adopted on the same day. Moreover, a
reference to the Erratum was also included in Article 1 of the “draft treaty on demarcation of the
frontier between the two countries”, annexed to the record of the meeting of 9 April 1986 between

the Minister for Territorial Administration and Security of Burkina and the Minister Delegate for
the Interior of Niger and a key part of the preparatory work on the Agreement of 28 March 1987 195;
and there is nothing to indicate that the Parties had intended to repudiate the Erratum between

April1986 andMarch1987. Furthermore, the Parties clearly corrected their error when drafting
the Special Agreement seising the Court, as, when citing Article 1 of the Agreement, they include
the reference to the Erratum that was left out of the text of the Agreement signed on

23March1987. In any event, according to the principle of lex posterior priori derogat ¸ the
Erratum should be preferred over the original version of the Arrêté 196.

194
Furthermore, Article 1 of the Protocol of Agreement refers to the frontier between the “two countries”, whilst
Article 1 of the Agreement refers to the frontier between the “two States”.
195
Report of the meeting between the Minister for Territorial Administration and Security of Burkina and the
Minister Delegate for the Interior of Niger, 9 April 1986, Ann. MBF 68; see above, para. 1.58.
19The original Arrêté can however be used in the event of problems of interpretation of the Erratum, as is clear
from the Court’s 1986 Judgment in the case concerning the Frontier Dispute — see below, para. 2.42. - 52 -

2.28. For the rest, the Protocol sets out more general rules relating to the functioning of the
Joint Technical Commission on Demarcation (Arts. 3 to 5) and the fund for the demarcation of the

frontier (Arts. 6 to 10), and contains various “technical and transitional provisions” (Arts. 11 to 20).
The only provision among the latter that could have an indirect effect on the delimitation of the
frontier is the second paragraph of Article 14, whereby: “ The régime governing frontier

watercourses shall remain that applicable under the relevant international law.”

2.29. It is important to note that the 1987 Agreement, as its very title indicates, concerns “the
demarcation of the frontier between the two countries”. It is thus a demarcation agreement [accord
197
de démarcation] “whose purpose is to fix, in concreto, the frontier on the ground” 198 , or, if you
will, an agreement on marking out the frontier [accord d’abornement] , which, in both cases,
“presupposes the prior delimitation — in other words definition — of the frontier” 199. In this case,

that definition exists in the form of the Arrêté of 31August1927, as clarified by its Erratum of
5 October 1927, as the Parties formally recognized in the 1987 Agreement cited in Article 6 of the
Special Agreement.

2.30. Further, Articles 1 and 2 of the Agreement concern different subjects. Article1

expressly confirms that, by mutual agreement, the description of the frontier is that given in the
Arrêté, and only the Arrêté, as clarified or modified by its Errat um. It is (or they are) the only
200
relevant instrument(s) for the purposes of the— established— delimitation of the frontier .
Article 2, for its part, defines the method agreed upon by both Parties for marking out the frontier,
to be followed by the Joint Technical Commission on Demarcation; for the purposes of this

operation— which might require additional information to that given in the delimitation
instrument, it is provided that “[s]hould 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 , 1960 edition,
and/or any other relevant document accepted by joint agreement of the Parties”.

2.31. Moreover, it was because the technical experts did not follow that method and deviated
from the description in the Arrêté that, at its meeting in May 1988, the Joint Technical Commission

on Demarcation instructed them “to reconsider the 110km portion in question within eight
(8) days, complying with the texts designated in the Agreement and Protocol of Agreement signed
by the two Governments” 201— which they did 202. Thus, the “consensual line” adopted at the

fourth meeting of the Joint Technical Commission on Demarcation is the result of the unanimous

197
Daniel Bardonnet, “De la ‘densification’ des frontiers terrestres en Amérique latine”, Droits et libertés à la fin
du XXème siècle— Influence des données économiques et technologiques; Etudes offertes à Claude-Albert Colliard ,
Pedone, Paris, 1984, p.4 (also cited in Jean Salmon dir., Dictionnaire de droit international public , Bruylant/AUF,
Brussels, 2001, p.317); see also Patrick Daillier, Mathias Forteau and Alain PelletDroit international public, LGDJ,
Paris, 8ème édition, 2009, p.517; Sir Robert Jennings and Sir Arthur Watts, Oppenheim’s International Law ,
9th edition, Longman, Harlow, 1992, p. 662 or Charles de Visscher, Problèmes de confins en droit international public ,

Pedone, Paris, 1969, p. 12.
198According to the Dictionnaire de droit international public , prev. Note 197, marking out [ abornement] is a
“technical operation consisting in laying out on the ground, by means of posts and markers . . . the frontier between two

States. . . . [it] implements a prior delimitation; it is a method of demarcation” (p.2).
199Territorial Dispute (Libyan Arab Jamahiriya/Chad), Judgment, I.C.J. Reports 1994, p. 28, para. 56.

200As is explained above (paras.2.26 and 2.27), there is no reason to focus on the fact that the Erratum is not
expressly mentioned in Article 1 of the Agreement.

201Report of the meeting of the Joint Technical Commission on Demarcation, Diapaga, 12-15 May 1988, see
above, para. 1.67.

202See above, paras. 1.68-1.69. - 53 -

agreement of the technical experts, based on “[t]he basic 203ts ( Arrêté No. 2336 of 31 August 1927
and its Erratum No. 2602/APA of 5 October 1927)” .

2.32. Of course, that consensus was subsequently called into question. But the differences
did not touch on the principle that the frontier was defined by the Arrêté and its Erratum. The
consensual line was called into question as a result of Niger changing its interpretation of those
204
texts . And while the line adopted by the decision of 15 May 1991 following the ministerial
consultative and working meeting deviated from that described in the Arrêté and its Erratum 205
over one stretch of the frontier, that compromise text was again called into question by Niger 206.

2.33. After some hesitation, inJuly2001 the Joint Technical Commission on Demarcation

decided

“to create a joint committee composed of twelve (12) members to review the

theoretical line of the frontier in light of the basic documents, specifically:

 the Agreement and Protocol of Agreement of 28 March 1987,

 the Arrêté Général of 31 Augus1 t 927, as clarified by its Erratum of
5 October 1927”.

It was both “going back to basics”, at the same time as, in reality, repeating the same thing
twice, since the 1987 Agreement itself refers to the Arrêté and its Erratum; but this twofold

reference again confirms the particular importance the Parties attach to the Arrêté and its Erratum.

2.34. As they were unable to agree on the interpretation of the Arrêté, the Parties decided to

seise this Court by a Special Agreement, which nevertheless confirms the fact that they do agree to
consider that the Court should settle their dispute on the basis of the Arrêté and its Erratum.

2. The relevant documents

2.35. The Agreement of 28March1987, which describes in turn the law applicable to the
delimitation and to the demarcation, does not place on the same footing the 1927 Arrêté and its
Erratum, on the one hand, and the 1960 IGN France map and any other document accepted by joint

agreement of the Parties, on the other. This differentiation is justified: the Arrêté in its version
resulting from the Erratum constitutes the frontier title on which each Party is entitled to rely (A);
the other documents— including the 1:200,000-scale map— can only be considered on a

subsidiary basis (B).

203
Report of the fourth meeting of the Joint Technical Commission on Demarcation of the Niger-Burkina frontier,
Niamey, 26, 27 and 28 September 1988, Ann. MBF 81.
204
See above, paras. 1.75-1.78.
20See Ministerial consultative and working meeting between Niger and Burkina on the frontier line,
16 May 1991 (plus sketch-maps), Ann. MBF 89; see also the Ministerial Council Report on the results of the ministerial
meeting between Niger and Burkina on the course of the common frontier held in Ouagadougou, 2July1991,

Ann. MBF 90. See above, paras. 1.79-1.80.
20See above, para. 1.81. - 54 -

A. The frontier title: the Arrêté of 1927 and its Erratum

2.36. According to the definition given by the Chamber of the Court in the Burkina/Mali
207
case, and confirmed several times since , the concept of title can denote “documentary evidence”;
however,

“[i]t is hardly necessary to recall that this is not the only accepted meaning of the word

‘title’ . . . In fact, the concept of title may also, and more generally, comprehend both
any evidence which may establish the existence of a right, and the actual source of that
right.”208

2T.he. Arrêté of 31August1927 of the Governor General of FWA (as clarified by the

Erratum of 5October1927)— whose pro209ive value was accepted by the Chamber of the Court
in that same Judgment of 1986 , as well as by the Chamber which ruled in 2005 in the Frontier
Dispute between Benin and Niger 210 — should be considered as a title in the following two senses:

 it describes the boundary between the two French colonies of Upper Volta and Niger and
thereby constitutes the “documentary evidence” of the frontier line inherited by Burkina and

Niger; but,

 it also establishes the actual existence of a frontier — which was not so in the two cases cited

above— whose course has never been modified since; it not only describes the frontier, it
establishes it and constitutes its “source” and “basis”.

2.38. In many respects, the frontier defined by the amended Arrêté is artificial in nature. The
colonial authorities, wanting to establish a complete and precise boundary, were aware of the

implications of choosing such a boundary, and that choice was made by the Governor General of
FWA in full knowledge of the facts and in accordance with what was standard practice at the time.
In 1947, the date the colony of Upper Volta was reconstituted within its former boundaries of 1932,

it was once again recalled that “the boundaries that separated Upper Volta from its neighbours to
the north and the east, Sudan and Niger [in 1932], were purely theoretical and did not correspond to
any geographical reality” 211. In retaining this artificial delimitation, the 1927 Arrêté merely

followed what was standard practice for the colonizers in Africa. Indeed, it is important to

“bear in mind that most African frontiers are purely artificial... In most cases they

are boundaries put in place by the colonial powers as a result of agreements between
them or with indigenous peoples or through conquest or occupation. I hasten to add
that boundaries the world over are, most of them, artificial. But in Africa they are

patently even more artificial than elsewhere, since most of them are merely straight
lines traced on the drawing board with little relevance to the physical circumstances
on the ground. As far back as 1890, Lord Salisbury said: ‘We have been engaged . . .

in drawing lines upon maps where no white man’s feet have ever trod; we have been
giving away mountains and rivers and lakes to each other, but we have only been
hindered by the small impediment that we never knew exactly where those mountains

20Land, Island and Maritime Frontier Dispute (ElSalvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992, pp. 388-3898, para. 45.

20Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 564, para. 18.
209
Ibid., p. 581, para. 52 ; pp. 589-592, paras. 68-72, or pp. 642-643, paras. 167-168.
210
Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, pp. 146-147, para. 135.
21See the Mission report of 30 May 1947 (Ann. MBF 38), p. 11, and pp. 34-36. - 55 -

and rivers and lakes were.’ (Memorial of Libya, Vol.1, p. 25, para.3.01: quoted
from The Times, 7 August 1890.)” 212

2.39213he 1927 Arrêté is no exception to the rule and establishes an arbitrary and artificial
boundary . Nevertheless, as modified by its Erratum, it constitutes a title covering the whole of
the frontier — which it describes with a precision that stands in stark contrast to the ditherings of

many internal delimitations in former FWA and French Equatorial Africa (FEA). This
distinguishes the present case from numerous others where the uti possidetis spoke “with an
214
uncertain voice” . Thus:

 in the Burkina/Mali case, the Chamber stated that “neither the legislative and regulative texts,

nor the relevant administrative documents, contain any complete description of the course of
the boundary between French Sudan and Upper Volta during the two periods when these
215
colonies co-existed, i.e., between 1919 and 1932, and between 1947 and 1960” ;

 for its part, the Chamber called upon to rule on the Frontier Dispute between El Salvador and
Honduras noted that neither of the Parties had “produced any legislative or similar material
indicating specifically, with the authority of the Spanish Crown, the extent of the territories and
216
the location of the boundaries of the relevant provinces in each area of the land boundary” .

 in the case concerning Sovereignty over Pulau Ligitan and Pulau Sipadan, Indonesia invoked a
convention from 1891, which, according to the Court, “cannot be interpreted as establishing an
allocation line determining sovereignty over the islands out to sea, to the east of Sebatik” 217;

and

 in the Frontier Dispute between Benin and Niger, “the Chamber concluded that”, in respect of
the River Niger sector, “neither of the Parties ha[d] succeeded in providing evidence of title on
the basis of regulative or administrative acts during the colonial period” 218.

2.40. In contrast, in the Aouzou Strip case, having noted “that Article 3 of the 1955 Treaty
[between France and Libya] was aimed at settling all the frontier questions, and not just some of
them” 219, the Court relied entirely on that provision and the instruments it cites for the purposes of
220
defining the frontier thus delimited and refused to give consideration to anything else . It should
be noted, however, that the Franco-Libyan treaty, which was the applicable frontier title, did not

describe the disputed frontier, whereas in this case, firstly , the Parties have determined by mutual

21Territorial Dispute (Libyan Arab Jamahiriya/Chad), Judgment, I.C.J. Reports 1994 , individual opinion of
Judge Ajibola p. 53, para. 9.

21Burkina wishes to reiterate in this connection that it fully accepts that colonial heritage, no matter how arbitrary
the boundary delimitations. See Burkina’s Memorial in the case concerning the Frontier Dispute (Burkina/Mali ), in

I.C.J. Pleadings, Oral Arguments and Documents , Vol. II, p.63, paras.1-9. These arguments are equally valid in the
present case.
214
See above, para. 2.8.
21Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 580, para. 51.

21Land, Island and Maritime Frontier Dispute (ElSalvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992, p. 388, para. 44.

21Sovereignty over Pulau Ligitan and Pulau Sipadan (Indonesia/Malaysia), Judgment, I.C.J. Reports 2002 ,
pp. 652-653, para. 52; see also p. 668, para. 92.

21Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 127, para. 75.

21Territorial Dispute (Libyan Arab Jamahiriya/Chad), Judgment, I.C.J. Reports 1994, p. 24, para. 48.
220
See ibid. and pp. 38-39, paras. 75-76. - 56 -

agreement that the 1927 Arrêté is the only title they can rely on and, secondly, that same instrument
(as clarified by its Erratum) describes the entire common frontier.

2.41. It is true that the description of the line in the Arrêté and that in its Erratum differ in
parts 221and that, consequently, preference should be given to the text of 5 October 1927, which is
222 223
both subsequent to and more precise than that of 31 August .

2.42. Nevertheless, if the need arises, the original text may clarify the meaning to be ascribed
to the Erratum. Moreover, the Chamber of the Court dealt with that question in its 1986 Judgment

in the Burkina/Mali case. That decision is enlightening in two respects:

 firstly, it is recalled that:

“if the Chamber’s task were to interpret and apply the Order as amended on
5 October 1927 as a regulative text, for the purpose of establishing the boundaries of

Upper Volta in 1932, it would have to examine its scope and appraise the relevance of
the initial text of 31 August 1927, and of any travaux préparatoires, in the light of the
particular rules of the legal system from which the Order derives its force as a

regulation, i.e., French colonial law. But the Chamber recalls that the 1927 Order
does not directly concern the boundary between Sudan and Upper Volta, but only the
boundary between Upper Volta and Niger, and that for the purposes of this case, the

Chamber is 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 and Upper Volta. In addition, from a more general perspective, 224
Chamber has already had occasion to emphasize (paragraph 30[ ]) that if colonial
law has any role to play in this case it does so not in its own right, by way of a renvoi

from international law to colonial law, but solely as evidence of the situation which
existed at the time when the two Stat es Parties achieved independence. The Chamber
is therefore free to examine in this light the two successive versions of the 1927 Order,

while nonetheless attributing greater weight to the text as modified by the erratum as a
reflection of the definitive intentions of the colonial authorities, and to take the
225
travaux préparatoires into consideration if this proves to be necessary.”

The same applies in the present case: the Court may feel the need to resort to the original

text of 31August1927; but solely for the purpose of interpreting the text of the Erratum of
5 October;

 moreover, and secondly, the 1986 Judgment quite rightly refers to “the Order as amended” by
the Erratum of 5 October 1927 226; it is that text which constitutes the frontier title in this case.

221See above, paras. 1.27 and 1.30-1.31.
222
See above, para. 2.27.
223
For an explanation of the reasons for the adoption of the Erratum, see above, para. 1.32.
224See above note 161.

225Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986 , p. 590, para. 69 — italics
added.

226See the passage in italics in the previous citation; see also, for example: “It is clear from the actual wording of
the text of the amended Order that the starting-point of the boundary . . .” (Ibid, para. 70 — italics added.) - 57 -

2.43. The difference between the two texts is first of all one of style. Unlike the Erratum, the
text of 31Augustis written from the subjective perspective of an observer located in Niger
describing the boundaries of the colony of Niger rather than the boundary between the two

territories. Thus, for example, it refers to the boundary “to the north-west”, which means to the
north-west of the authors, who must be located to the south-east of said boundary, namely in Niger
in this instance. In contrast, the Erratum undertakes to describe the frontier not from a subjective

viewpoint but from an objective one, whereby the author is situated neither on one side or the other
of the line described. The author of the Erratum is on the boundary itself and describes its course
from north to south as if travelling along it from one end to the other.

2.44. Further, as explained in the previous chapter 227, the Erratum of 5October1927
clarifies a certain number points in the 31Augusttext, which is thereby modified in certain

respects. As the Chamber of the Court stated in 1986, this is the “amended Order”.

B. The course of the boundary should the Arrêté and its Erratum not suffice

2.45. While Article 1 of the Agreement of 28 March 1987 — which delimits the frontier —

refers only to the boundary described in the Arrêté of 31 August 1927, as modified by its Erratum
of 5 October 1927, Article 2 — which indicates the method to be used to mark out the boundary —
provides that “[s]hould the Arrêté and Erratum not suffice , the course shall be that shown on the

1:200,000-scalemap of the Institut Géographique National de France , 1960edi228n, and/or any
other relevant document accepted by joint agreement of the Parties” .

2.46. As the Parties have not agreed on any document other than the Arrêté of 1927 and its
Erratum 229 since the adoption of the 1987 Agreement, the latter provision is not applicable.

2.47. Thus, only the 1960 IGN France map can be used to clarify the course of the frontier
between the Parties. Nevertheless, it follows from the very text of the 1987 Agreement that it may

only be used on a subsidiary basis. There are two reasons for this:

 firstly, the Agreement only refers to the map in question when explaining the methodology for

demarcating the frontier; and demarcation “presupposes the prior delimitation... of the
frontier”230, a delimitation by which it is governed and which the Parties agree to consider, in
Article 1, as having been effected by the 1927 Arrêté; and

 secondly, even for the purposes of marking out the frontier, the 1:200,000-scale map can only
be used “[s]hould the Arrêté and Erratum not suffice”.

2.48. Such cautious use is all the more appropriate since, in its ruling on the frontier dispute

between Burkina and Mali in 1986, the Chamber of the Court noted that the probative value of the
map in question was relatively limited. It held that: “the map compiled in 1958-1960 by the

227
Paras. 1.30-1.31.
22Italics added.

22With the sole exception of the formalization, in 2009, of their “agreement” within the meaning of Article 2 of
the Special Agreement — see below, paras. 3.12-3.13. The “consensual line” of 1988 could also be included as such an
agreement (see the record of the fourth meeting of the Joint Technical Commission for the Demarcation of the
Niger-Burkina frontier, Niamey, 26, 27 and 28 September 1988, Ann. MBF 81), but it was called into question by Niger

(see above, paras. 1.74 to 1.78).
23See above, para. 2.29. - 58 -

IGN— a body neutral towards the Parties to the present dispute—, although it does not possess
the status of a legal title, is a visual portrayal both of the available texts and of information obtained
on the ground”.

However, it immediately adds that:

“This in itself is not sufficient to permit the Chamber to infer that the frontier

line depicted in the form of small crosses, whether in a continuous or a broken series,
in the successive editions of the IGN map, corresponds entirely with the boundary
inherited from the colonial administration. It has to consider how far the evidence

offered by this or any map corroborates the other evidence produced. The Chamber
cannot uphold the information given by the map where it is contradicted by other
trustworthy information concerning the intentions of the colonial power. However,

having regard to the date on which the surveys were made and the neutrality of the
source, the Chamber considers that where all other evidence is lacking, or is not
sufficient to show an exact line, the probative value of the IGN map becomes
231
decisive.”

This was stating what the 1987 Agreement was to lay down a few months later, in a slightly

different form.

2.49. Thus, even though the 1987 Agreement did not state it expressly, the map could in any
event only be used, should the Arrêté as modified by the Erratum contain lacunae or not suffice (it
being recalled that an equivalent title was absent in the Burkina/Mali case 23). Further on in the

same decision, moreover, the Chamber questions “the validity of the indications233ven by the
1960 IGN map” for the purposes of construing a text dating from 1927 .

2.50. The case brought before the Court by the Special Agreement of 24 February 2009 has
some quite unique features compared to the majority of other frontier delimitation cases that have

been brought before it. Indeed, such disputes generally arise for want of a title establishing the
frontier, or because the evidence is insufficient or imprecise, or because one of the parties disputes
the existence or validity of such a title. This was true in the Aouzou Strip case, where Libya

considered that the 1955 Treaty of Friendship234d Good Neighbourliness between France and
Libya did not constitute a frontier title . This is not at all the case in the present proceedings,
where Burkina and Niger explicitly recognize in a binding treaty, to which the Special Agreement

explicitly refers, that a frontier title exists not only as “documentary evidence” (as it describes the
course of their common frontier) but also as the source of their respective territorial rights: the
1927 Arrêté in the version resulting from the Erratum.

23Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 586, para. 62.
232
See above, para. 2.39.
23Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 647, para. 173.

23Territorial Dispute (Libyan Arab Jamahiriya/Chad), Judgment, I.C.J. Reports 1994 , esp. p. 20, para. 36, and
pp. 22-23, paras. 43-46. CHAPTER III

T HE TASK OF THE COURT IN RESPECT OF THE DEMARCATED

PORTION OF THE FRONTIER

3.1. Under Article 2 of the Special Agreement whereby the Parties seised the Court, which
defines the “Subject of the dispute”:

“The Court is requested to:

1. determine the course of the boundary between the two countries in the sector from

the astronomic marker of Tong-Tong (latitude 14° 25' 04" N;
longitd147) to the beginning of the Botou bend
(latitude 12° 36' 18" N; longitude 01° 52' 07" E);

2. place on record the Parties’ agreement on the results of the work of the Joint
Technical Commission on Demarcation of the Burkina Faso-Niger boundary with

regard to the following sectors:

(a)the sector from the heights of N’Gouma to the astronomic marker of
Tong-Tong;

(b) the sector from the beginning of the Botou bend to the River Mekrou”.

3.2. The following chapter gives a precise description of the course of the frontier, in
Burkina’s view, in the middle sector (between the Tong-Tong astronomic marker and the beginning
of the Botou bend), on which the Parties were unable to reach agreement 23. The aim of this

chapter is to recall briefly the substance of the agreement reached on the two extremities of the
frontier (from the tripoint with the Republic of Mali to the Tong-Tong marker and from the
beginning of the Botou bend to the Mekrou) (Sec. 1), and to comment on the role assigned to the

Court by the Special Agreement in respect of these two sectors (Sec. 2).

SECTION 1

THE P ARTIES ’AGREEMENT

3.3. Under Article 2, paragraph 2, of the Special Agreement, the Parties’ agreement “on the

results of the work of the Joint Technical Commission on Demarcation of the Burkina Faso-Niger
boundary” concerns two sectors, one of 63 and one of 162km in length, situated at either end of
their common frontier (some 590 km overall236):

 the sector from the heights of N’Gouma to the Tong-Tong astronomic marker (2); and

 the sector from the beginning of the Botou bend to the River Mekrou (3).

In both cases, the Parties relied on the same instruments and adopted the same method (1).

235
With the exception of the agreement reached in 1988, called into question by Niger from 1990 onwards.
23Report of the second ordinary session of the Joint Technical Commission on Demarcation of the Niger-Burkina
Frontier, Ouagadougou, 23-28 July 1990, Ann. MBF 87. - 60 -

1. The methodology behind the Parties’ agreement (reminder)

3.4. Without it being necessary here to describe the details of the negotiations that resulted in
the Parties’ agreement on demarcating and marking out the two sectors of the boundary to the north
of the Tong-Tong astronomic marker and to the south of the Botou bend 237, it is sufficient to recall

that the agreement was achieved on the basis of the methodology established by Article 2 of the
Agreement of 28 March 1987, to which the Special Agreement refers 238:

 the course of the frontier is that described in the Arrêté of 31August1927, as clarified by its
Erratum of 5 October 1927;

 the frontier should therefore be marked out in conformity with that course;

 it is only in the event that the description in the 1927 Arrêté and its Erratum should not suffice
that recourse should be had to the 1960 IGN France 1:200,000-scale map “and/or any other

relevant document accepted by joint agreement of the Parties”;

 it being noted that no such document has been accepted by the Parties 239.

3.5. It should be pointed out that throughout the process which led to their agreement on the

northern and southern portions of the frontier the Parties alw ays based themselves systematically
on the line described in the 1927 Erratum. Notwithstanding the progress and setbacks between the
Parties in reaching agreement (some points which appeared settled were called into question by

Niger, which led to this Court being seised), the method used has been consistently the same:

 identificati240of “defining points” on the frontier based on the Arrêté as amended by the
Erratum ;

 plotting on the 1:200,000 map defining points with known co-ordinates (for example
astronomic markers) and the resulting line (which in fact entails not taking account of the

boundary line shown on the map, which the Parties use simply as a base map); in other words,
the course plotted on the map represents the technical experts’ cartographic interpretation of
the Arrêté as clarified and amended by the Erratum 241;

 field survey and identification of other defining points (for example the intersection of
242 243
watercourses) and measurement of their precise geodetic co-ordinates ;

237
See Section 2 of Chapter 1 above.
238
See above, paras. 2.21 et seq., in particular 2.29-2.30. The undertaking to rely on these documents alone was
reiterated at the fourth ordinary session of the Joint Technical Commission in July 2001, Ann. MBF 94.
23With the exception, of course, of those indicating their agreement on the two sectors of the frontier which were

marked out and the subject of the 1988 agreement disputed by Niger.
24See esp.the report of the meeting between technical experts of Niger and Burkina, Ouagadougou,

21-23May1986, Ann.MBF(List of defining points on the Niger-Burkina frontier ( Arrêté of 31 August 1927 and its
Erratum of 5 October 1927)), Ann. MBF 69.
24See esp.the report of the meeting between Niger and Burkina technical experts, Ouagadougou,

21-23May1986, previous note, and above, para.1.59. On the only occasion that the technical experts wanted, for the
sake of convenience, to give precedence to the map over the text of the Arrêté (see the report annexed to the report of the
meeting of Joint Technical Commission on Demarcation, Diapaga, 12-15 May 1988, Ann.MBF80), the Joint
Commission set things to rights and instructed them “to reconsider the 110 km portion in question within eight (8) days,
complying with the texts designated in the Agreement and Protocol of Agreement signed by the two Governments”
(ibid.); on this episode, see above, paras. 1.67 and 1.68.

24See esp. the report of the previously cited meeting, 12-15 May 1988, and above, para. 1.67. - 61 -

244
 plotting, also on the 1960 map, the points thus determined and the resulting course ;

245
 installation of markers in the field .

3.6. Thus, the Joint Technical Commission could see the frontier line as described in the
Erratum on the 1960 map and compare it to the line printed on the map.

3.7. It should be noted that two types of defining frontier points were taken into

consideration:

 firstly, those corresponding to reference points installed for the purposes of mapping the
territory (astronomic points or markers); the co-ordinates of these points are already known
(geographical co-ordinates (latitude and longitude) and/or rectangular co-ordinates in the UTM

projection system (X, Y and Z)) and represented on the map by a specific symbol
(conventional sign); they had to be located on the ground and it had to be ensured that they

were actually situated at the original sites as defined by the co-ordinates noted in the records;
if the markers had disappeared, the technical experts had to install new ones based on the
known co-ordinates; and,

 secondly, the defining points described by the Arrêté and/or its Erratum (intersection points,

ruins, natural features such as peaks, ridges, watercourses, etc.).

In these instances, the technical experts from the two countries searched together for the

details on the ground through visual observation and, if necessary, by cross-checking information
gathered from local inhabitants. Once the experts had reached a consensus on the location of a
physical detail, the defining point was indicated by a temporary marker, its co-ordinates measured

and calculated by both parties, then plotted on the 1:200,000 map to establish the course of the
frontier in accordance with the Erratum.

3.8. This method, used by the Joint Technical Commission on Demarcation and its
246
sub-committee, led to the unanimous agreement on the “consensual line” , adopted by the
Commission at its meeting in Niamey from 26 to 28September1988 247 and confirmed at two
further meetings held in Téra and Niamey respectively from 16 to 18 May 1989 248 and on 13 and
249
14 November 1989 .

243
See the reports of the fourth meeting of the Joint Technical Commission on Demarcation of the Niger-Burkina
Frontier, Niamey, 26, 27 and 28September1988, Ann.MBF81, and of the meeting of the Joint Technical
Sub-Committee on Demarcation of 18 May 1989, Ann.MBF83. See also the record of the work of the Joint Survey
Mission of 3 July 2009, Ann. MBF 101.

244See the report on completion of the survey of the proposed demarcation of the Niger-Burkina frontier,
28September1988, Ann.MBF82, and its Cartographic AnnexMBF15; see also report No.42/FP/SG/DCAF from the
Minister for Territorial Administration to the Head of State of Burkina Faso, 5 March 1991, Ann. MBF 88.

245See the report of the fourth meeting of the Joint Technical Commission on Demarcation of the Niger-Burkina
Frontier, Niamey, 26, 27 and 28 September 1988, Ann. MBF 81; and the record of the 2 ndsession of the Joint Technical
Commission on Demarcation, 23-28 July 1990, Ann. MBF 87.

246For further details on the events leading up to this agreement, see above, paras. 1.69-1.70.

247Report of the fourth meeting of the Joint Technical Commission on Demarcation of the Niger-Burkina
Frontier, Niamey, 26, 27 and 28 September 1988, Ann. MBF 81.

248Report of the meeting of the Joint Technical Sub-Committee on Demarcation, 18 May 1989, Ann. MBF 83.
249
Report of the fifth meeting of the Joint Technical Commission on Demarcation, Niamey, 13 and
14 November 1989, Ann. MBF 84. - 62 -

250
3.9. That overall agreement was called into question by Niger , however, and the Joint
Technical Commission on Demarcation, at its fourth session from 18 to 20 July 2001, was obliged
to note that there were differences in interpretation in respect of the sectors of the frontier from the

Tong-Tong astronomic marker to the River Sirba at Bossébangou and from Bossébangou to
Tchenguiliba. However:

“1. The frontier was clearly defined from the heights of [Mount] N’Gouma to the
astronomic marker of Tong-Tong, with the exception of the ruins of Tokébangou,
which the frontier passes to the west. These ruins were not identified in the course

of the survey of the frontier line. The Commission decided to send a survey team
into the field to identify the ruins of the village of Tokébangou.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2. The frontier was clearly defined from Tchenguiliba to the River Mékrou, subject
251
to the survey team’s verification of the position of the village of Kogori.”

3.10. Thus, for those two sectors of the frontier, the Joint Technical Commission on
Demarcation once again endorsed the 1988 agreement, which was already reflected in the map
annexed to the report of the fourth meeting ofSeptember1988. That agreement would be

confirmed in 2009, both in the Special Agreement of 24February2009 and in the exchange of
letters between the Ministers for Foreign Affairs of both countries of 29Octoberand
2 November 2009, embodying the agreement of 3 July 2009 between the technical experts.

3.11. Although Burkina had indicated its preference to submit the entire frontier to the Court,

it agreed to Niger’s proposal not to take the risk of calling into question the lines over which there
was no disagreement between the Parties 252. The precise object of their “agreement”, which at the
time the Special Agreement was signed had never been formalized, had yet to be ascertained.

3.12. The circumstances in which the “agreement” mentioned in Article 2, paragraph 2, of

the Special Agreement was reached are as follows:

 At the meeting held in Niamey from 22 to 24February2009 for the purposes of agreeing on

the contents of the Special Agreement seising the Court, and approving and signing the text of
that Agreement, the Ministers for Foreign Affairs of the two States “also adopted the experts’
[present in each delegation] proposal for a joint field mission to agree on the co-ordinates of
253
the boundary markers in Sectors A and B referred to in the Special Agreement” .

 In their report on that ministerial meeting sent to the national commission monitoring the
frontier dispute, the Burkina experts noted that there were “errors in respect of the co-ordinates
of markers placed in the field, as transmitted by Niger. It was decided to send a joint team into

250
For details, see above, paras. 1.74 to 1.81.
25Report of the fourth ordinary session of the Joint Technical Commission on Demarcation of the Niger-Burkina

Frontier, Ouagadougou, 18-20 July 2001, Ann. MBF 94.
25See above, para. 0.14 of the introduction to this Memorial.

25Joint Communiqué of the meeting of Foreign Ministers for negotiation and signature of the Special Agreement
seising the International Court of Justice, 24 February 2009, Ann. MBF 98, p. 3. - 63 -

the field to determine the position of these markers by GPS and obtain mutually agreed
254
co-ordinates” .

 This was done, and after working together on excellent terms, the technical directors of the

National Geographic Institutes of Burkina and Niger signed a joint communiqué on
3July2009 indicating the GPS co-ordinates of the markers constructed on the frontier and
describing the methodology used 255.

 Upon receiving this document, the Burkina Foreign Minister, by letter dated 17 July 2009, took
the initiative to submit to his Niger counterpart a “draft exchange of Notes embodying the

agreement between the Parties on the delimited sectors”:

“We agreed at our meeting at Niamey from 22 to 24February2009 that we

would endorse the experts’ proposal to carry out a joint mission on the ground with a
view to reaching agreement on the co-ordinates of the boundary markers in sectors A
and B referred to in the Special Agreement.

The experts performed their assignment by carrying out the joint mission of
determining the co-ordinates of the markers erected along the Burkina Faso-Niger
frontier and they laid down their agreement in a record (procès-verbal) signed at

Diapaga (Burkina Faso) on 3 July 2009, a copy of which is attached hereto.

In consequence of the foregoing, I propose that we consider this record of the

experts’ work to represent our agreement within the meaning of the above-quoted
provision [namely Article 2, paragraph 2] of the Special Agreement and I would be
grateful to you for confirming to me your concurrence on this point.” 256

 By a letter dated 8 September 2009, the Niger Minister gave his agreement in principle to these
proposals; he suggested a partial rewording and added some details in respect of the two

endpoints of the southern sector of the frontier line:

“while expressing my agreement with this proposal, I would suggest the text be

shortened as follows:

Pursuant to Article 2, paragraph 2, of the Special Agreement seising the
International Court of Justice of the Frontier Dispute between Burkina Faso and the

Republic of Niger, signed at Niamey (Niger) on 24February2009, the Parties place
on record their agreement on the results of the work of the Joint Technical
Commission on Demarcation as reflected in the record of the Joint Mission of Experts,

signed at Diapaga (Burkina Faso) on 3 July 2009 and which describes the location of
the markers in sectors A and B referred to in the Special Agreement: [table of GPS
co-ordinates]

Furthermore, as the two endpoints of sector B are not marked out, the two
Parties agree to consider their co-ordinates to be those derived from the IGN France

1:200,000-scale map, 1960 edition, Kirtachi sheet.

254
Report of meeting No.11 of the Committee of Exper ts of the National Commission monitoring the frontier
dispute between Burkina Faso and the Republic of Niger, 15 May 2009, Ann. MBF 100.
25Record of the work of the Joint Survey Mission to ascertain the co-ordinates of the markers constructed on the

frontier between Burkina Faso and the Republic of Niger, 3 July 2009, Ann. MBF 101.
25Letter from the Minister for Foreign Affairs of Burkina to the Minister for Foreign Affairs of Niger regarding
the draft exchange of Notes embodying the agreement between th e Parties on the delimited sectors of the frontier, dated
17 July 2009, Ann. MBF 102. - 64 -

Threes:e

 the point where the Tapoa intersects with the former boundary of the Fada and Say
cercles;

 the point where the former boundary of the Fada and Say cercles intersects with
the River Mekrou.

If this new wording meets with your approval, please consider my present reply
and your letter cited above as constituting an agreement between the Republic of
257
Niger and Burkina Faso on the matter in question.”

 The Burkina Minister’s reply, dated 8October2009, proposes endorsing the agreement

between the Parties on that basis:

“While expressing my agreement with the new wording of my proposal, I

would suggest that the document containing the co-ordinates of the markers already
erected should be the one initialled by the technical experts of the two Parties on
258
3 July 2009[ ].

As regards sector B, Burkina Faso has no objection to considering the line to be

that derived from the IGN France 1:200,000-scale map, 1960 edition, Kirtachi sheet.

However, as this sector has not been marked out, I propose that the heads of the

two Institutes responsible for mapping should meet . . . with a view to ascertaining the
co-ordinates of those frontier points in a concerted and consensual manner.

The record of that work and the record of 3 July 2009 will then be considered as
representing our understanding under the Special Agreement signed at Niamey on
259
24 February 2009.”

3.13. As the Minister of Burki260did not receive a reply, he recalled his proposa261o his Niger
counterpart on 29October2009 , who replied favourably on 2 November . The exchange of
letters of 29Octoberand 2 November 2009 thus formally embody the “Parties’ agreement”

mentioned in Article 2, paragraph 2, of the Special Agreement, and accordingly that “agreement” is
constituted by the documents referred to in those letters:

257
Letter from the Minister for Foreign Affairs of Niger to the Minister for Foreign Affairs of Burkina regarding
the draft exchange of Notes embodying the agreement between th e Parties on the delimited sectors of the frontier, dated
8 September 2009, Ann. MBF 103.
258
The reason for this clarification is that the table of co-ordinates attached by the Niger Minister to his letter of
8 September 2009 was not a copy of the table initialled by the directors of the National Geographic Institutes of the two
countries on 3 July 2009.
259
Letter from the Minister for Foreign Affairs of Burkina to the Minister for Foreign Affairs of Niger regarding
the draft exchange of Notes embodying the agreement between the Parties on the delimited sectors of the frontier,
8 October 2009, Ann. MBF 104. The record of the work to ascertain the co-ordinates of the Tapoa point and the Mekrou
point is reproduced in the record of the meeting to ascertain the co-ordinates of the unmarked points in Sector B,
15October2009, Ann.MBF105. The co-ordinates are derived from the IGN France 1:200,000-scale map, 1955 Kandi
sheet and 1960 Kirtachi sheet.

260Ann. MBF 106.

261Ann. MBF 107. - 65 -

 the table of co-ordinates appearing in the record of the work of the Joint Survey Mission to
ascertain the co-ordinates of the markers constructed on the Burkina-Niger frontier of
3 July 2009 262; and

 in respect of the southern sector (“from the beginning of the Botou bend to the River Mekrou”),

the co-ordinates derived from the IGN 1:200,000-scale map as shown in the record of the
meeting to ascertain the co-ordinates of the unmarked points in SectoB r of
15 October 2009 26.

3.14. The full list of co-ordinates is annexed to this chapter.

2. From the heights of N’Gouma to the Tong-Tong
astronomic marker

3.15. The northern sector of the frontier, on whose course the Parties agree, separates
Burkina’s present-day Oudalan province, which corresponds to the northern part of the former Dori

cercle, from the Tillabéri region of Niger (which corresponds to the former cercle of the same
name).

3.16. It is described as follows in the first sentence of the first paragraph of the Erratum of
5 October 1927:

“The boundaries of the Colonies of Niger and Upper Volta are determined as
follows:

A line starting from the heights of N’Gouma, passing through the Kabia ford
(astronomic point), Mount Arounskoye and Mount Balébanguia, to the west of the

ruins of the village of Tokebangou, Mount Doumafende and the Tong-Tong
astronomic marker.”

3.17. The different “defining points” of this sector were identified at the meeting on
demarcation work between the technical experts on 21 May 1986 264. These points, listed from

262
Ann.MBF101. It should be noted that the Parties evidently consider the 23 points shown in the record to be
connected by straight lines (except when, pursuant to the 19Arrêté, the boundary follows a watercourse, i.e., “from
Nonbokoli to the confluence of the Rivers Dyamongou and Dantyabongou; from there to the confluence of the Rivers
Dyamongou and Boulel Fouanou” and “from the intersection of the Botou bend with the River Tapoa as far as the former
boundary of the Fada and Say cercles”).

263The co-ordinates of a point vary according to the ellipsoid used to establish them. There are two geodetic
systems (among others, but these are the ones that are relevant to this case): one, older system, known as the “Clarke
1880 ellipsoid”, which was used to establish the 1960 map, and which is used to extract points from a map (for example
the last two points on the boundary, 15 October 2009); the other, modern system, known as “WGS 84 ellipsoid” (WGS
stands for World Geodetic System), associated with the GPS concept.

Even though the figures may differ, it would not be correct to say that the co-ordinates are, strictly speaking,
different, as the point’s location does not change; it is simply that its co-ordinates depend on the method of measurement.
Thus the same point will have different co-ordinates depending on the geodetic system used, namely in 1986 (and 1988):
Clarke 1880; in 2009: WGS 84 (the co-ordinates were measured by GPS).

It should be noted that (i) the margin of error in the former system is 40 metres on the ground, while in the second
it is approximately one metre and (ii) it is possible to calculate the transformation parameters in order to switch from one
system to the other.

264See above, note 240. - 66 -

north to south, correspond exactly to each locality, natural feature and marker cited in that
instrument:

“1. Mount N’Gouma (co-ordinates to be measured on a 1:200,000 map)

2. Astronomic marker at the Kabia Ford (ij = 14º 53' 09"N; Ȝ = 0º 13' 04")

(X = 200,623; Y = 1,647,421)

3. Mount Arounskoye or Arwaskoy

(co-ordinates to be measured on a 1:200,000 map)

4. Mount Balebanguia (" " " )

5. Point to the west of the Tokébangou or Takabangou ruins

(" " )

6. Mount Doumafendé (" " )

7. Tong-Tongastronomicmarker ( ij = 14º 25' 04"; Ȝ = 0º 12' 47")

265
(X = 199,477; Y = 1,595,611)” .

3.18. The only co-ordinates known with certainty at that time were those of the Kabia ford
and Tong-Tong astronomic markers, thanks to the “data sheets, sheet at 1:200000, ND-31-XIII,
266
Observer Nevière, 1927” . Nevertheless, in September 1988, the Joint Technical Commission on
Demarcation placed on record the findings of the te chnical experts in respect of six out of these
seven points and noted the difficulties of locating the Tokébangou ruins 267. As the ruins were no

longer there, it was neither possible to apply the texts mechanically nor to rely on the 1960 IGN
map, which contains no mention of them; thus a location had to be selected on the basis of the field
survey which would satisfy. both parties 26.

3.19. That sector of the frontier has never been disputed since 269, as the Commission

confirmed at the end of its fourth ordinary session in July 2001: reviewing the “theoretical line” of
the frontier, it states that “[t]he frontier was clearly defined from the heights of N’Gouma to the
astronomic marker of Tong-Tong, with the exception of the ruins of Tokébangou, which the

265
Report of the meeting between technical experts of Niger and Burkina, Ouagadougou, 21-23 May 1986,
Ann.MBF(List of defining points on the Niger-Burkina frontier ( Arrêté of 31August1927 and its Erratum of
5 October 1927)), Ann. MBF 69.
266
Ann. MBF 41.
267See the two composites of IGNF 1:200,000 boundary maps, 1960 edition, annexed to the report on completion
of the survey of the proposed demarcation of the Niger-Burkina frontier signed by the two Chairmen of the Joint
Technical Commission on Demarcation, 28 September 1988, Cartographic Annex MBF 15 and above, para. 3.8.

268See ibid.

269The northern sector is thus the only portion of the frontier not to have been subject to the comprehensive
review undertaken by the Joint Technical Commission at its second ordinary session from 23-28 July 1990 in the wake of
criticism from Niger. - 67 -

frontier passes to the west. These ruins were not identified in the course of the survey of the
frontier line.”270

3.20. That uncertainty nevertheless did not prevent the Parties from establishing the
271
co-ordinates of si272f the seven defining points in the sector (measured first of all on th273
1:200,000 map) and proceeding to mark them out between December 1989 and May 1990 . A
joint mission undertaken in June-July 2009 measured the GPS co-ordinates of those markers as

identified in the report of 28 July 1990. Measured on the WGS 84 ellipsoid, they are as follows:

 Mount N’Gouma (0° 14' 36.4'' east; 14° 54' 46.0'' north)

 Kabia Ford (0° 13' 06.3'' east; 14° 53' 09.8'' north)

 Mount Arwaskoye (0° 10' 35.8'' east; 14° 50' 44.7'' north)

 Mount Bellé Banguia (0° 14' 09.6'' east; 14° 45' 05.2'' north)

 Takabougou (0° 10' 16.1'' east; 14° 37' 54.5'' north)

 Mount Douma Fendé (0° 09' 42.1'' east; 14° 32' 00.6'' north)

 Tong Tong (0° 12' 51.7'' east; 14° 24' 53.2'' north).

3.21. In respect of Mount N’Gouma, the Parties were also able to rely on the conclusions of

the Chamber of the Court in the case of the Frontier Dispute between Burkina and Mali, whereby:

“the Governor-General, in the 1927 Order, as modified by the erratum, and hence in

letter 191 CM2 of 1935, described an existing boundary which passed through
elevations situated north of the Kabia ford, and that the administrators, rightly or
wrongly, considered that these elevations were called by the local people the ‘heights

of N'Gouma’. The Chamber has simply to ascertain, therefore, the point where the
boundary defined by the texts in question terminates within the above-described

ensemble of elevations environing the ford. After minutely examining the topography
shown on the IGN map, the Chamber finds that this point should be fixed 3 kilometres
north of the ford, at the spot defined by the co-ordinates 0"143 ' 9"E and
274
14" 54' 48" N.”

270
Ann. MBF 94.
271With the exception of Mount N’Gouma.

272See the report of the meeting of the Joint Technical Sub-Committee on Demarcation, signed at Téra,
18 May 1989, Ann. MBF 83. The technical experts had “adopted the average of the various co-ordinates obtained by the

two teams within tolerable limits” (p.1) for each of the 27 defining points (out of the 32 planned) which were actually
measured. The list of defining points is annexed to the report.
273See Ann.MBFII of the report of the 2nd session of the Joint Technical Commission on Demarcation,

23-28 July 1990, Ann. MBF 87, which designates the boundary markers that were effectively installed on the ground. It
should be noted that there is no marker at Mount N’Gouma— nor at the southern tripoint —, as “[t]he markers at the
two endpoints are scheduled to be installed after tripartite agreements have been reached”. See also below, para.3.28;
on the endpoints, see below, paras. 3.41-3.42.
274
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986 , p. 647, para. 173. See also
point M on the map mentioned in para.175 of the Judgment and annexed thereto, which is reproduced in Cartographic
Annex MBF 14. - 68 -

275
3.22. There is no need to focus on the apparent (minimal ) differences between the
co-ordinates of the heights of N’Gouma adopted by the Chamber of the Court in 1986 and those
276
established by the ex277ts in 2009 : these can be explained by the different methods of
measurement used . The same explanation accounts for the (minute) differences between the
co-ordinates agreed on in 1988 278 and those eventually adopted in 2009 279. In any event, it is the

Parties’ agreement of 2009 that the Court is requested to place on record.

3. From the Botou bend to the River Mekrou

3.23. The southern sector of the frontier between the two Parties, whose course they also

agree on, corresponds to Burkina’s present-day Tapoa province (which includes, from north to
south, the frontier départements of Kantchari, Botou and Diapaga) (and to the former Fada cercle)
and to the southern part of the Tillabéri region of Niger (which includes the départements of Téra

in the north and Say in the south). It was no mo re difficult for the Parties’ to reach agreement on
this line than it was for the northern sector.

3.24. However, contrary to the northern sector, reading the basic texts was not sufficient to
identify the defining points. Notwithstanding the clarifications of the Erratum of 5 October1927,

the Arrêté of 31 August 1927 gives very few salient topographical details, considering the length of
the line and the changes in direction. It is described in the Erratum of 5 October 1927 thus:

“[from a point located 1,200m to the west of the village of Tchenguiliba, the
frontier turns back up] in a straight line that runs in a marked SSW-NNE direction; it
passes approximately 2 km west of the village of Birniouoli and, approximately 2 km

south of the south of the village of Vendou Mama, reaches the top of the northernmost
spur of the Heni-Djouri (Gourma) massif or Jackal Mountain.

Running then in a west-east direction, it passes 1 km south of Mount Tambado
Djoaga, follows the course of the Dantiabonga marigot, passes south of Dantiandou,
follows the line of the Yoga Djoaga hills as far as the confluence of the Dantiabonga

and Diamongou marigots, and runs along the latter as far as the confluence of the
Dialongou and Boulelfonou marigots approximately 5 km north of the latter village.

From that point, the boundary follows the crests of the Djoapionga hills as far as
the source of the Boulolfonou marigot, runs up the northern slope of the Tounga and
Djoaga massif and terminates at the point known as Niobo-Farou (Caiman Pool), a

sort of broad basin, which is traversed during the dry season by the track from Botou
to Fombonou.

It is then determined by the eastern crests of the Tounga Djoaga massif, before
running towards the River Tapoa in a precise north-south direction. It passes

275
The co-ordinates of Mount N’Gouma are, by the Court’s calculation: 14° 54' 48'' N, 0° 14' 39'' E, whilst the
joint mission of June-July 2009 gives: 14° 54' 46,0'' N, 0° 14' 36,4'' E.
276
See above, para. 3.12 and the table annexed to this chapter.
27A single location may have different co-ordinates depending on the method of measurement used (Clarke 1880

ellipsoid in 1960, 1986 and 1988, and WGS 84 in 2009). See above, note 263.
27See above, para.3.8, and the report on completion of the survey of the proposed demarcation of the
Niger-Burkina frontier, 28 September 1988, Ann. MBF 82 and its Cartographic Annex MBF 14.

27See above, para. 3.12 and the table provided at the end of this chapter. - 69 -

approximately 5km east of the village of Kogori [sic] and reaches the Tapoa
approximately 4 km south of the aforementioned village.

It then follows the course of the Tapoa upstream until it meets the 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.”

3.25. Given the relative complexity of this sector, the technical experts of Burkina and Niger

meeting on 21 May 1986 refrained from drawing up a list of the defining points in this area, and,
after enumerating the first nine points on the frontier, merely referred to the “other defining
280
points” . It was not until the fourth meeting of the Joint Technical Commission on 26, 27 and
28September1988 that this portion of the boundary was actually identified with the help of the
IGN map: “the section situated to the west of Tc henguiliba as far as the frontier with Benin [is]
281
identical to the line on the IGNF 1:200,000 map, 1960 edition” .

3.26. The village of Tchenguiliba — which is mentioned under the toponym of Tyenkilibi or
Tyenkiliba in some documents and on some maps, and which, to the west, marks the beginning of
the Botou bend — had been identified by the Joint Technical Commission at a meeting in Diapaga
282
from 12-15 May 1988 . The list of specific points marked out between December 1989 and May
1990 appears in the report of the meeting of the Joint Technical Sub-Committee on Demarcation
held at Téra on 18 May 1989 283. It is reproduced in Annex II to the report of the 2nd session of the
284
Joint Technical Commission on Demarcation from 23-28 July 1990 :

“16. Tyenkilibi

17. JackalMountain

18. Loguil[Laguil]

19. Nonbokoli

20. Confluence of the Dyamongou and Dantyabongou Rivers

21. Confluence of the Dyamongou and Boubel Fouanou Rivers

22. Boubel

23. BoubelEast

24. Teylinga

25. DyapiongaNorth

26. DyapiongaSouth

280
See above, para. 3.17.
281
Report on completion of the survey of the proposed demarcation of the Niger-Burkina frontier,
28 September 1988, Ann. MBF 82, p. 3, para. 3.
28Report of the meeting of the Joint Technical Commission on Demarcation, Diapaga, 12-15 May 1988,

Ann. MBF 80.
28Ann. MBF 83.

28Ann. MBF 84. - 70 -

27. Kanleyenou

28. Nyobo Farou (Caiman Pool)

29. Eastern crests of Mount Tambouadyoaga

30. Banindyididouana

31. Confluence of the Banindyidi Fouanou and Tapoa Rivers”.

3.27. The majority of the markers were placed at the defining points mentioned in the

1927texts; however, others were added to avoid any uncertainty as to the course of the frontier,
namely:

 at the villages of Laguil and Nonbokoli, which show where the frontier passes “1 km south of
Mount Tambado Djoaga” and follows the course of the Dantiabonga marigot; and

 at the villages of Teylinga and Kanleyenou.

285
As was the case with the village of Tokébangou mentioned above , the doubts286rrounding
the precise location of the village of Kogori— which still persisted inJuly2001 — did not
prevent the Parties from reaching an agreement on this portion of the frontier.

3.28. The southern extremity of the frontier on the River Mekrou— i.e., the tripoint with

Benin — was defined in the following terms in the Judgment of the International Court of Justice
of 12 July 2005, which is fairly brief on the matter:

“[T]he boundary between the Republic of Benin and the Republic of Niger in
the River Mekrou sector follows the median line of that river, from the intersection of
the said line with the line of deepest soundings of the main navigable channel of the
287
River Niger as far as the boundary of the Parties with Burkina Faso.”

3.29. As for the northern sector, the points marked out in 1990 were measured by GPS in
2009, and a number of points added on either side of the confluence of the Rivers Dyamongou and

Boulel Fouanou, both on the Niger side and the Burkina side, with marker 21 being “triplicated”; 288
the same procedure was used to mark the “Boulel” point which is also located in a river :

“16. Tyenkilibi (1° 52' 06.9'' east; 12° 36' 19.2'' north)

17. Jackal Mountain (1° 55' 43.9'' east; 12° 41' 33.1'' north)

18. Loguil (1° 57' 01.3'' east; 12° 41' 31.9'' north)

28See paras. 3.18-3.19.
286
See the report of the fourth ordinary session of the Joint Technical Commission of 20 July 2001, Ann. MBF 94,
p.4: the only question mark at that time hung over Kogori, and it was planned to resolve the matter by carrying out a
survey of the course of the southern sector of the frontier.
287
Frontier Dispute (Benin/Niger), Judgment, I.C.J. Reports 2005, p. 151, para. 146. On p. 144 there is a — very
simplified — sketch-map.
288
Strictly speaking there is only one frontier point at each of these two places; the “RN” and “BF” markers were
installed on either side of the frontier point to ascertain the co-ordinates (by calculating the average of the co-ordinates of
the two markers) of that point, where it was impossible to place a marker given that it was in the middle of a river. See
below, note 308. - 71 -

19. Nonbokoli (1° 58' 47.0'' east; 12° 44' 12.9'' north)

20. Confluence of the Dyamongou and Dantyabongou Rivers (2° 05' 14.9'' east;
12° 43' 15.1'' north)

21. BF- Confluence of the Dyamongou and Boubel Fouanou Rivers — Burkina side
(2° 06' 23.4'' east; 12° 43' 43.1'' north)

21. Confluence of the Dyamongou and Boubel Fouanou Rivers (2° 06' 23.9'' east;
12° 43' 44.0'' north)

21. RN- Confluence of the Dyamongou and Boubel Fouanou Rivers— Niger side
(2° 06' 24.4'' east; 12° 43' 44.8'' north)

22. BF- Boulel — Burkina side (2° 06' 52.8'' east; 12° 42' 14.6'' north)

22. Boulel (2° 06' 53.3'' east; 12° 42' 15.1'' north)

22. RN- Boulel — Niger side (2° 06' 53.7'' east; 12° 42' 15.5'' north)

23. Boulel East (= Teylinga) (2° 09' 43.2'' east; 12° 41' 09.5'' north)

24. Dyapionga North (2° 09' 37.3'' east; 12° 39' 42.3'' north)

25. Dyapionga South (2° 09' 08.1'' east; 12° 38' 55.4'' north)

26. Kanleyenou (2° 11' 57.1'' east; 12° 37' 21.7'' north)

27. Nyobo Farou (Caiman Pool) (2° 13' 23.9'' east; 12° 35' 19.6'' north)

28. Eatern crests of Mount Tambouadyoaga (2° 13' 48.0'' east; 12° 31' 19.7'' north)

30. Banindyididouana (2° 16' 27.2'' east; 12° 27' 52.7'' north)

31. Confluence of the Banindyidi Fouanou and Tapoa Rivers (2° 16' 40.6'' east;
289
12° 25' 30.5'' north)” .

290
3.30. As explained above , the consensual definition of the southern sector was completed
on 15 October 2009 with the determination of the last two points, not by the WGS 84 method, but
by extracting their co-ordinates from the IGN France 1:200,000-scale map, 1955 Kandi sheet and

1960 Kirtachi sheet. The co-ordinates of those points are:

 31- Intersection of the former boundary of the Say and Fada cercles with the course of the

River Tapoa: 2° 04' 12.77'' east; 12° 21' 04.88'' north

289The above list was drawn up on the basis of the record of 3July2009, taking account of the discrepancy—
noted in the record by the technical experts —, in respect of points 23 and 24, between the report of 28 July 1990 and the
reality on the ground: “where the point designated as ‘Boulel East’ corresponds to the ‘Teylinga’ point. This produces a
shift upward in all the designations from that point onward. As a sult, No.31 disappears from the list” (emphasis

added). The technical experts propose “that the designations of the points that appear in the Report of 28July 1990 be
corrected to conform to the reality in the field”. See th e record of the work of the Joint Survey Mission to ascertain the
co-ordinates of the markers constructed on the Burkina-Niger frontier, 3 July 2009, p. 7, Ann. MBF 101.
290Para. 3.12, note [259]. - 72 -

 32- Intersection of the former boundary of the Say and Fada cercles with the course of the
Mekrou: 2° 24' 15.25'' east; 11° 54' 07.83'' north.

SECTION 2
T HE TASK OF THE C OURT
C ONFIRMATION OF THE P ARTIES ’AGREEMENT

3.31. The task entrusted to the Court by Article 2, paragraph 2, of the Special Agreement is
somewhat unusual, but fully within the functions of the Court as the principal judicial organ of the

United Nations (1). The same is true of the — always delicate — task of determining the endpoints
of the frontier, which are the subject of the “Part ies’ agreement” within the meaning of that
provision (2).

1. The Court’s participation in the comprehensive settlement of the
frontier dispute between the Parties

3.32. As the PCIJ emphasised, “[i]t is... natural that any article designed to fix a frontier
should, if possible, be so interpreted that the result of the application of its provisions in their
291
entirety should be the establishment of a precise, complete and definitive frontier” . This is no
less true when two States turn to the Court with a view to the delimitation of their common frontier,
as

“[a] judicial decision, which ‘is simply an alternative to the direct and friendly
settlement’ of the dispute between the Parties ( P.C.I.J., Series A, No.22 , p.13),
merely substitutes for the solution stemming directly from their shared intention, the

solution arrived at by a court under the mandate which they have given it. In both
instances, the solution only has legal and binding effect as between the States which
have accepted it, either directly or as a consequence of having accepted the court’s
292
jurisdiction to decide the case.”

3.33. In the present proceedings, the two Parties have a general frontier dispute which they
have always intended to settle in its entirety. That is precisely the reason why they have not
entered into a formal agreement establishing their consensus on the two northern and southern

sectors of their common frontier: they only agreed in this regard to define their “agreement”,
which they have requested the Court to place on record. The definition of their agreement was
undertaken after the adoption of the Special Agreement, and was formally embodied in the
exchange of letters of 29Octoberand 2 November 2009 which took place between the Foreign

Ministers of the two countries and was sealed by the signature of the record of the meeting to
ascertain the co-ordinates of the unmarked points of Sector B by the directors of the national
geographic institutes of Burkina and Niger on 15 October 2009 293.

3.34. It should also be pointed out that a decision by the Court on the entire frontier would

seem all the more logical and necessary, given that two points come under both “categories” of
sections of the frontier distinguished by Article 2 of the Special Agreement:

291
Interpretation of Article 3, Paragraph 2, of the Treaty of Lausanne, Advisory Opinion, 1925, P.C.I.J., Series B,
No. 12, p. 20.
292
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 572, para. 46.
29See above, paras. 3.12-3.13. - 73 -

 the Tong-Tong astronomic marker is both the last point of the frontier’s northern sector, on
which the Parties reached agreement, and the first point of the disputed sector; and

 the beginning of the Botou bend is the last point of the portion whose course is in dispute and
the first point of the second demarcated sector on which the Parties have agreed.

3.35. As the oft-cited dictum of the Court in the case concerning the Temple of Preah Vihear

states, “[i]n general, when two countries establish a frontier between them, one of the primary
objects is to achieve stability and finality” 294. In this case, only the Court’s intervention is liable to

achieve the end sought by the Parties to these proceedings.

3.36. Fur295r, “[t]hat the Court may, in an appropriate case, make a declaratory judgment is
indisputable” , and it has frequently fallen to the Court to record the agreement of the parties’ on
some aspects of a dispute that has been submitted to it 29. Thus, in the case of the Société

commerciale de Belgique , the PCIJ declared that the two Governments were agreed “in
contemplating the possibility of negotiations with a view to a friendly settlement, in which regard
297
would be had, amongst other things, to Greece’s capacity to pay” , and, in the case of the Free
Zones it quite clearly considered that “there seems nothing to prevent the Court from embodying in
its judgment an agreement previously concluded between the Parties” and that “a ‘judgment by

consent’, though not expressly provided for by the Statute, is in accordance with the spirit of that
instrument . . .” 298. Further still, in the case of the Free Zones , the Permanent Court did not
hesitate, in the actual dispositif of its Judgment, to decide that: “the declaration made in regard to

this question by the Agent of the Swiss Government before the Court at the hearing on
April 22nd, 1932, shall be placed on record” 299.

That was a unilateral declaration made on behalf of one of the parties but which formed a
part of the agreement recorded by the Court; the same should apply, a fortiori, for an “agreement”

which the Parties would like to acquire the force of res judicata.

3.37. Moreover, while it would be going too far to claim that the frontier dispute before the
Court has ever really posed a threat to peace and security in the region, it has been the source of —
sometimes serious— tension between the two countries and there have been a considerable
300
number of border incidents . By removing such threats once and for all, the Court will be doing
no more than contributing to the objectives of the United Nations, of which it is the principal

judicial organ. By resolving finally and completely the dispute the Parties have brought before it, it
will contribute to the consolidation of peace between the two countries, and to the development of
friendly and cooperative relations between them; these are precisely the objectives of the United

Nations, as set out in paragraphs 1 to 3 of the Charter.

294
Temple of Preah Vihear (Cambodia v. Thailand), Merits, Judgment, I.C.J. Reports 1962, p. 34.
295
Northern Cameroons (Cameroon v. United Kingdom), Preliminary Objections, Judgment, I.C.J. Reports 1963,
p.37. See also Interpretation of Judgments Nos.7 and8 (Factory at Chorzów), Judgment No.11, 1927, P.C.I.J.,
Series A, No. 13, p. 20; or Nuclear Tests (Australia v. France), Judgment, I.C.J. Reports 1974, p. 163, para. 30.
296
The very principle of forum prorogatum is associated with this idea.
297
Société Commerciale de Belgique, Judgment, 1939, P.C.I.J., Series A/B, No. 78, p. 178.
298Free Zones of Upper Savoy and the Distr ict of Gex, Order of 6December1930, P.C.I.J., SeriesA, No.24 ,
p. 14.

299Free Zones of Upper Savoy and the District of Gex, Judgment, 1932, P.C.I.J., Series A/B, No. 46, p. 172.

300See above, Introduction, paras. 0.5-0.6. - 74 -

3.38. In its recent Advisory Opinion on Kosovo, the Court recalled that: “its answer to a

request for an advisory opinion ‘represents its participation in the activities of the Organization,
and, in principle, should not be refused” (Interpretation of Peace Treaties with Bulgaria, Hungary
and Romania, First Phase, Advisory Opinion, I.C.J. Reports 1950 , p.71; Difference Relating to

Immunity from Legal Process of a Special Rapporteur of the Commission on Human Rights,
Advisory Opinion, I.C.J. Reports 1999 (I) , pp. 78-79, para. 29; Legal Consequences of the
Construction of a Wall in the Occupied Palestinian Territory, Advisory Opinion, I.C.J. Reports

2004 (I), p. 156, para. 44). Accordingly, the consistent jurisprudence of the Court has determined
that only ‘compelling reasons’ should lead the Court to refuse its opinion in response to a request
falling within its jurisdiction (Judgments of the Administrative Tribunal of the ILO upon complaints

made against the Unesco, I.C.J. Reports 1956 ,p. 86; Legal Consequences of the Construction of a
Wall in the Occupied Palestinian Territory, Advisory Opinion, I.C.J. Reports 2004 (I) , p.156,
para. 44)” 301.

3.39. Both Parties are convinced that the Court, being seised of a contentious case (in

relation to which it does not have the power that it has under Article 65, paragraph 1, of its Statute
to consider the propriety of giving an advisory opinion), will not decline to fulfil the task entrusted
to it and thereby completely and finally resolve the frontier dispute between them.

2. Determination of the endpoints of the frontier

3.40. The Parties’ agreement within the meaning of Article 2, paragraph 2, of the Special
Agreement covers the two extremities of their common frontier.

3.41. In the north, where it joins the frontiers between them and Mali, the co-ordinates of
Mount N’Gouma, the first point mentioned in the Arrêté of 1927, were fixed in the record of the
work of the Joint Survey Mission to ascertain the co-ordinates of the markers constructed on the

Burkina-Niger frontier of 3July2009, confirmed in the exchange of letters of 29October and
2November 2009 between the Foreign Ministers of the two countries, at 14°54' 46.0'' latitude
north and 0° 14' 36.4'' longitude east 302. These co-ordinates correspond to those of the marker

constructed by Mali and Burkina, pursuant to the 1986 Judgment303t Point M, which represents the
endpoint of their common frontier under that Judgment . At the meeting of technical experts in
conjunction with the final meeting to negotiate the Special Agreement inFebruary2009, Niger

agreed that this marker also marked the starting point of the Burkina-Niger frontier.

3.42. In accordance with the Erratum to the 1927 Arrêté, the southern extremity of the

frontier is situated at the intersection of the former boundary of the cercles of Fada and Say with
the River Mekrou. The Parties’ agreement on th ese precise co-ordinates was only finally reached
at the meeting of the heads of the two cartographic institutes of Burkina and Niger held at
304
Kantchari on 15October2009 , following the exchange of letters between the Parties’ Foreign
Ministers of 8 September and 8 October of the same year. Accordingly, that point is situated at the
intersection of the meridian 2° 24' 15.25'' longitude east and the parallel 11° 54' 07.83'' latitude

north.

30Accordance with International Law of the Unilateral Declaration of Independence in respect of Kosovo,
Advisory Opinion, I.C.J. Report 2010, para. 30.
302
As noted above (para. 3.22), these co-ordinates are not identical to those retained by the Chamber of the Court
in paragraph 173 of the 1986 Judgment (14° 54' 48'' N, 0° 14' 39'' EI.C.J. Reports 1986 , p.647). These minimal
variations are due to the difference in methodology used: see above, note 263.
303
I.C.J. Reports 1986, p. 650, para. 179.A.8 (operative text).
30See above, paras. 3.12 and 3.30. - 75 -

3.43. Burkina is aware of the fact that in placing on record the Parties’ agreement on the
endpoints of their common frontier, the Court will not fix the tripoints with Mali and Benin, as
neither of those two countries are party to the proceedings. However, as the Chamber of the Court

noted in the Burkina-Mali case, it is required “not to fix a tripoint, which would necessitate the
consent of all the States concerned, but to ascertain, in the light of the evidence which the Parties
have made available to it, how far the frontier which they inherited from the colonial power
305
extends” . Furthermore, the rights of neighbouring States “are in any event safeguarded by the
operation of Article 59 of the Statute of the Court” 306, and upholding the points adopted by the
Parties as marking the endpoints of their common frontier will present no problem for the Court in

the present proceedings, given that both of those points correspond to points fixed by previous
decisions of the Court in cases involving respectively one of the Parties to these proceedings and
307
one of the third States concerned .

3.44. The course of the frontier on which the Parties agree and which the Court has been

requested to place on record consists of watercourses or straight lines connecting the points
mentioned in the Erratum to the 1927 Arrêté and those added by mutual agreement of the Parties.

The relevant co-ordinates are reproduced in the table on the following page of this Memorial. The
resulting illustrative line is shown on two sketch-maps (sketch-map No.6 showing the northern
sector, called “Sector A”, and sketch-map No.7 showing the southern sector, called “Sector B”;

also reproduced in Cartographic Annexes 23 and 24) reproduced on pages 96 and 97 of the French
text, and in a sketch-map showing the entire course of the boundary reproduced in Cartographic
Annex 25.

305Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 579, para. 49.

306Ibid., p. 577, para. 46.
307
The 1986 Judgment fixes the co-ordinates of point M very precisely (see above, note 274). For its part, the
Judgment of 12July 2005 simply states “that the boundary between the Republic of Benin and the Republic of Niger in
the River Mekrou sector follows the median line of that river, from the intersection of the said line with the line of
deepest soundings of the main navigable channel of the River Niger as far as the boundary of the Parties with Burkina
Faso” (I.C.J. Reports 2005, p. 151, para. 146.4) (operative text). - 76 -

CO -ORDINATES OF THE FRONTIER POINTS CONSTITUTING THE P ARTIES ’AGREEMENT W308IN
THE MEANING OF ARTICLE 2, PARAGRAPH 2,OF THE SPECIAL A GREEMENT

309
Number Name on the report of 28 July 1990 Latitude (DMS) Longitude (DMS)

1 Mount N’Gouma 14º 54' 46.0" N 0º 14' 36.4" E

2 Kabia ford (astronomic marker) 14º 53' 09.8" N 0º 13' 06.3" E

3 Mount Arwaskoye 14º 50' 44.7" N 0º 10' 35.8" E

4 Bellé Banguïa 14º 45' 05.2" N 0º 14' 09.6" E

5 Takabougou 14º 37' 54.5" N 0º 10' 16.1" E

6 Mount Douma Fendé 14º 32' 00.6" N 0º 09' 42.1" E

7 Tong Tong (astronomic marker) 14º 24' 53.2" N 0º 12' 51.7" E

16 Tyenkilibi 12º 36' 19.2" N 1º 52' 06.9" E

17 Jackal Mountain 12º 41' 33.1" N 1º 55' 43.9" E
18 Laguil 12º 41' 31.9" N 1º 57' 01.3" E

19 Nonboloki 12º 44' 12.9" N 1º 58' 47.0" E

20 Confluence of Dyamongou and 12º 43' 15.1" N 2º 05' 14.9" E

Dantiabongou Rivers

21 BF ConfDyam/Boulel — Burkina side 12º 43' 43.1" N 2º 06' 23.4" E

21 Confluence of Dyamongou and Boulel 12º 43' 44.0" N 2º 06' 23.9" E
Fouanou Rivers

21 RN ConfDyam/Boulel — Niger side 12º 43' 44.8" N 2º 06' 24.4" E

22 BF Boulel — Burkina side 12º 42' 14.6" N 2º 06' 52.8" E

22 Boulel 12º 42' 15.1" N 2º 06' 53.3" E

22 RN Boulel — Niger side 12º 42' 15.5" N 2º 06' 53.7" E

23 Boulel East (=Teylinga) 12º 41' 09.5" N 2º 09' 43.2" E

24 Dyapionga North 12º 39' 42.3" N 2º 09' 37.3" E

25 Dyapionga South 12º 38' 55.4" N 2º 09' 08.1" E

30The co-ordinates of these points, which appear in the Annexto the record of the Joint Survey Mission to
ascertain the co-ordinates of the markers constructed on the Burkina Faso-Republic of Niger frontier, 3July 2009, were
established by GPS, except for:

 points 21 and 22, whose co-ordinates were calbased on the average of the co-ordinates of
21BF and 21 RN, and 22 BF and 22 RN respectively. This procedure was used because of the
difficulty of constructing a marker at points 21 and 22 which are located in a river (point 21 is at the
confluence of the Dyamongou and the Boulel Fouanou, while point 22 is in the Boulel Fouanou
which runs through the village of Boulel). Consequently, the sole purpose of markers 21 BF and
21RN and 22, and markers 22 BF and 22 RN is to identify the respective positions of frontier
markers 21 and 22.

 points 31 and 32, whose co-ordinates were derived from a composite of IGN 1:200,000 maps.

The list of defining points reproduced in the above table was drawn up on the basis of the third observation of the
technical experts in the record of 3 July 2009, p. 7; see above, note 288.
30Apart from points 31 and 32, which were not in the record and were derived from the IGN 1:200,000 map on

15 October 2009 and determined on the Clarke 1880 ellipsoid. - 77 -

309
Number Name on the report of 28 July 1990 Latitude (DMS) Longitude (DMS)

26 Kanleyenou 12º 37' 21.7" N 2º 11' 57.1" E

27 Niobo Farou 12º 35' 19.6" N 2º 13' 23.9" E
28 Eastern crests of Mount Tambouadyoaga 12º 31' 19.7" N 2º 13' 48.0" E

29 Banindyididouana 12º 27' 52.7" N 2º 16' 27.2" E

30 Confluence of Banindyidi Fouanou and 12º 25' 30.5" N 2º 16' 40.6" E

Tapoa Rivers
31 Intersection of the Tapoa with the former 12º 21' 04.88" N 2º 04' 12.77" E

boundary of Fada and Say cercles

32 Intersection of the former boundary of 11º 54' 07.83" N 2º 24' 15.25" E
Fada and Say cercles with the Mekrou - 78 -

Sketch-map No. 6 - View of the northern sector - Sector A - of the part of the

frontier agreed by the Parties



 View Sector A




Scale: 1:200,000 














 

  



96 - 79 -

Sketch-map No. 7 - View of the southern sector - Sector B - of the part of the

frontier agreed by the Parties


 


View Sector B
 

   
    





Scale: 1:250,000



 

 

 



 


































 


 

97 C HAPTER IV

T HE PART OF THE FRONTIER WHOSE COURSE IS TO BE
DETERMINED BY THE C OURT

4.1. The part of the frontier of which the Court is requested, under the Special Agreement, to
“determine the course” lies between the two sectors discussed in the previous chapter 310. It
consists of the intermediate section which starts at the Tong-Tong astronomic marker and

terminates at the beginning of the Botou bend.

4.2. Under Article 2, paragraph 1, of the Special Agreement, the Court is thus requested to
“determine the course of the boundary between the two countries in the sector from the astronomic
marker of Tong-Tong (latitude14°25'04"N; longitude00°12'47"E) to the beginning of the
Botou bend (latitude 12° 36' 18" N; longitude 01° 52' 07" E)”.

4.3. This sector of the frontier is delimited as follows in the 192Arrêté (according to the
Erratum):

“The boundaries of the Colonies of Niger and Upper Volta are determined as
follows:

[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. It 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; then, turning back to the south, it again cuts the Sirba at the level of the

Say parallel.

From that point 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 . . .”

4.4. In the light of this description of the course of the boundary, the portion referred to in

Article 2, paragraph 1, of the Special Agreement can be divided into three sections (see page 100 of
the French text, sketch-map No.8— Disputed sector of the frontier; also reproduced in
Cartographic Annex 26):

 the first section goes from the Tong-Tong astronomic marker to the point where the frontier
reaches the River Sirba at Bossébangou (Sec. 1);

 the second starts at the latter point and ends at the intersection of the River Sirba with the Say
parallel (Sec. 2);

 the third and final section runs from the intersection of the River Sirba with the Say parallel to
the beginning of the Botou bend (Sec. 3).

31See above, sketch-map No. 1 showing an overview of the frontier between the two States, p. 7, reproduced in

Cartographic Annex MBF 18. - 81 -

4.5. It should be pointed out that in the Special Agreement seising the Court, the co-ordinates

of some of the points through which the boundary passes are given on the Clarke 1880 ellipsoid.
These co-ordinates were derived from the 1960 IGN 1:200,000 map. They are different from the
GPS co-ordinates, which are based on the WGS 84 ellipsoid. Since the Special Agreement was

signed, the Parties have in particular been able to measure the GPS co-ordinates of the Tong-Tong
astronomic marker and of the point corresponding to the beginning of the Botou bend 31. Burkina
has also obtained the GPS reading of the position of the Tao marker. Burkina will use these

co-ordinates for these three points in the rest of this chapter. As the position of the other points can
only be derived at this stage from the 1960 IGN 1:200,000 map, their co-ordinates will be given
systematically on the Clarke 1880 ellipsoid.

31In this connection, see the record of the work of the Joint Survey Mission of 3 July 2009, Ann. MBF 101. - 82 -

Sketch-map No. 8 - Disputed sector of the frontier

DISPUTED SECTOR OF THE FRONTIER


 

Scale : 1/1,000,000
 




  



Section 1
 



  

Re public of Niger

Burkina Faso
 

Point where the boundary reaches the

   

River Sirba at Bossébangou


  



Section 2



 

Intersection of the Say Parallel 



 
with the Sirba


Section 3

Beginni

e


Botou bend

▲ Markers constructed by the Joint Technical Commission 

 


● Po  
  




Map prepared for illustrative purposes only

100 - 83 -

S ECTION 1
T HE COURSE OF THE FRONTIER FROM THE T ONG -TONG ASTRONOMIC MARKER TO
THE POINT WHERE IT REACHES THE R IVER SIRBA AT B OSSÉBANGOU

4.6. In its original version, the first article of Arrêté of 31August1927 delimited this
section as follows:

“The boundaries of the Colonies of Niger and Upper Volta shall henceforth be
determined as follows:

1. Boundaries between the Tillabéry cercle and Upper Volta:

This boundary is determined to the north by the current boundary with Sudan (Gao

cercle) as far as the heights of N’Gourma, and to the west by a line . . . which then
turns towards the south-east, leaving the ruins of Tong-Tong to the east and
descending 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.

2. Boundaries between the Say cercle and Upper Volta:

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

To the north-west, by the River Sirba from its mouth as far as the village of
Bossébangou. From this point a salient . . .”2

4.7. The amended Arrêté follows the same course, which it describes in almost identical
terms, apart from a few nuances and clarifications:

“The boundaries of the Colonies of Niger and Upper Volta are determined as
follows:

A line starting from the heights of N’Gouma, passing through... the
Tong-Tong astronomic marker; this line then turns towards the south-east, cutting the

Téra-Dori motor road at the Tao astronomic marker locat313to the west of the Ossolo
Pool, and reaching the River Sirba at Bossebangou . . .”

4.8. As is clear from the text, the 1927 Arrêté provides a precise delimitation of the boundary
between the two Parties in this sector. Moreover, the Parties have not disputed it. Neither of them
has suggested that the Arrêté would not suffice to identify the course of the boundary in this sector

and, in particular, neither of them has deemed it necessary to have recourse to the line shown on the
1960 map — which, moreover, in this location does not coincide with the text of the Arrêté14.

312
Ann. MBF 34 (italics added).
31Ann. MBF 35.

31In this sector, the line on the map in fact passes through a range of points which are not mentioned at all in the
Arrêté; it also goes through a number of changes of direction which are not described in the text either. - 84 -

4.9. Thus, in this sector, the task of the Court is:

 first to determine the location of the three points described in the Arrêté through which the
boundary passes successively, namely the Tong-Tong astronomic marker, the Tao astronomic
315
marker and the point where the boundary reaches the River Sirba at Bossébangou (1);

 and then to ascertain how these three points are connected pursuant to the 1927 Arrêté. For

reasons which Burkina will develop hereafter, there is no doubt that these three points are
connected by two successive straight lines (2.).

1. The location of the frontier points designated in the 1927 Arrêté

4.10. The location of the Tong-Tong (A) and Tao (B) astronomic markers and the point

where the frontier reaches the River Sirba at Bossébangou (C) will be identified in turn, with a
view to ascertaining precisely where the boundary line delimited by the 1927 Arrêté passes.

A. The Tong-Tong astronomic marker

4.11. The location of the starting point of the boundary in this sector is defined in Article 2,
paragraph 1, of the Special Agreement seising the Court. Under the terms of that article, the
starting point consists of the “astronomic marker of Tong-Tong”.

4.12. This corresponds exactly to the provision of the 1927 Erratum. In its original version,
the text of the 1927 Arrêté simply stated that the inter-colonial boundary “le[ft] the ruins of

Tong-Tong to the east” without identifying the exact point through which the frontier passes. The
Erratum made it clear that the boundary passes through “the Tong-Tong astronomic marker”.

4.13. The co-ordinates of that marker are given in the Special Agreement:
“latitude 14° 25' 04" N; longitude 00° 12' 47" E” 316. According to the GPS reading taken during
the joint mission of June-July 2009 317, the co-ordinates are: latitude 14°24' 3.2N;

longitude 0° 12' 51.7" E.

B. The Tao astronomic marker

4.14. The 1927 Arrêté (according to the Erratum) describes the second point through which

the boundary passes in this sector in the following manner: the boundary “cut[s] the Téra-Dori
motor road at the Tao astronomic marker located to the west of the Ossolo Pool”. With this
description, the Erratum provides a further useful clarification to the original version of the Arrêté
in stating that it is at the Tao astronomic marker that the boundary “cut[s] the the Téra-Dori motor

road to the west of the Ossolo Pool” (according to the less precise version of the August text).

315
On the fact that it is these three successive points through which the frontier should pass, see in particular the
report of the second ordinary session of the Joint Technical Commission on Demarcation of the Niger-Burkina Frontier,
held in Ouagadougou from 23 to 28 July 1990, Ann. MBF 87, in particular p. 3 and p.5 and Ann. MBF II, p. 1.
316
These are the co-ordinates based on the Clarke 1880 ellipsoid.
31See above, Chapter III, esp. paras. 3.12, 3.20 and 3.29. - 85 -

4.15. In retaining this marker as the point through which the inter-colonia318oundary
passes — as it did when it chose to refer directly to the Tong-Tong marker — the Erratum makes
two clarifications:

 it indicates exactly at which location on the motor road between Téra and Dori and at which
distance “to the west of the Ossolo Pool” the boundary should pass;

 it provides a point which is very easy to locate geographically, since an astronomic marker,
better than a road or a pool, is by its very essence a geographical reference point defined with

precision for the purposes of mapping the territory.

4.16. The location of the Tao ma319r, which was marked out in May 1990 when the two
Parties installed a frontier marker , can be ascertained by referring to the data sheet on the
astronomic markers 320. The Tao marker data sheet, drawn up by the Nevière mission in 1927,

reads as follows:

“Name of point: Tao marker

Lat: 14º 0313" N Long: 0º2253" E

217.186 1555.3 087Zone:

Tamarind tree with indicator plate to the north of the Dori-Téra road. New
Upper Volta-Niger frontier.” 321

The co-ordinates of this marker, measured by GPS by Burkina, are: Lat.: 14º 03' 04.7" N;
Long: 0º 22' 51.8" E.

C. The point where the boundary reaches the River Sirba at Bossébangou

4.17. The original version of the 1927 Arrêté defined the endpoint of the boundary in this
sector using two additional periphrases:

 “This boundary... then turns towards the south-east, leaving the ruins of
Tong-Tong to the east and descending 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”.

 “Boundaries between the Say cercle and Upper Volta: . . . To the north-west, by

the River Sirba from its mouth as far as the village of Bossébangou. From this
point a salient . . .”

4.18. Three pieces of information were provided to determine the location of the point
concerned. The line must: reach the River Sirba; near to and to the south of Boulkalo; at the

village of Bossébangou.

318
See above, para. 4.12.
319
See below, para. 4.51.
32When astronomic markers are installed, a survey is always carried out and a data sheet drawn up showing the
co-ordinates of the markers.

32Ann.MBF41. The data sheet also gives the co-ordinates of the Tong-Tong astronomic marker. The
co-ordinates correspond to those adopted in the Special Agreement (see above, para. 4.13). - 86 -

4.19. The Erratum simplifies and clarifies this description, without modifying it, by stating

that after the Tao astronomic marker, the boundary “reach[es] the River Sirba at Bossebangou”.

4.20. Given that the location of the point at which the boundary arrives in this sector is

identified by two pieces of information in conjunction— it “reach[es] the River Sirba” and
“reach[es it] . . . at Bossébangou” — both elements should be taken into account when defining its
location.

4.21. As the village of Bossébangou is situated on the right bank of the River Sirba, the only
possible way of making sense of the reference to a line “reaching the River Sirba” “at

Bossebangou” is to consider that the frontier, coming from the Tao astronomic marker, crosses the
River Sirba and stops on the right bank of that river at the village of Bossébangou:

 the boundary cannot actually go to the centre of the village of Bossébangou, as it would then
extend beyond the river that it is supposed to reach (and also cut the village in two, for which
there is no provision in the Arrêté which leaves the village to the Colony of Niger 322);

 nor can it stop on the left bank or even in the middle of the River Sirba, as it would then not be
considered to end “at Bossébangou”.

4.22. This interpretation is corroborated by two facts in particular:

 in the first place, it is the solution adopted by the 1960 map, which ends the boundary line on
the right bank of the river, which the line then follows up to the point where the salient begins
(see below, Section 2);

 it is the only way to understand the part of the amended Arrêté which states that after the
salient of four villages, the boundary “again cuts the Sirba at the level of the Say parallel”. For

it to do this, the boundary must necessarily have cut the river at least once already. There is
thus only one possible hypothesis: the boundary must join the right bank of the river when it
reaches Bossébangou. Only on this condition can the boundary then leave, at the level of the
salient, certain villages to Niger “on the left ba nk” of the River Sirba and then cross it “again”

at the level of the Say parallel to reach the right bank from where it then “continues in a
straight line up to a point located 1,200m to the west of the village of Tchenguiliba”, as
provided in the Arrêté.

4.23. As regards ascertaining precisely where this point is on the right bank of the river, its
location depends on the course of the line connecting it to the Tao astronomic marker. Given that,

for the reasons set out hereafter (2 below), the 1927 Arrêté defines a course which follows a
straight line between the Tao astronomic marker and the point where the frontier “reach[es] the

322
The General List of Localities of French West Africa, which was based on the delimitation carried out in 1927,
registers the village of “Bossébangou” in Fascicle VI on the colony of Niger (Ann. MBF 28, p. 8). This village does not
appear in Fascicle IV on the colony of Upper Volta (Ann.MBF27, p.9). The village of Bossébangou was already
known to the French colonizers at the end of the 19th century: see Captain Y. Urvoy, Histoire des populations du
Soudan central (Colonie du Niger), Librarie Larose, Paris, 1936, p. 120: “The first mission was that of Captain Monteil,
sent to reconnoitre the Say-Barroua line on the ground. He left Ségou in December 1890, went to Dori, where on 22 May
1891 he signed a protectorate treaty for Liptako; on 24 July, he crossed the River Sirba at Kakou, near Bossébangou,
reached Say on 19Augustand continued to Argougou, Kano, Chad and Tripoli” (reproduced in Ann.MBFE.14 of
Niger’s Counter-Memorial of 28 May 2004, submitted in the case of the Frontier Dispute (Benin/Niger) . See
Ann. MBF 37. It is therefore no surprise that this village served as a suitable point for the inter-colonial boundary line to
pass through. - 87 -

River Sirba at Bossébangou”, that point is necessarily defined in the following manner (see the
explanatory diagrams on the following pages, also reproduced in Cartographic Annexes 27 and 28):

 first a straight line should be drawn between the Tao astronomic marker and the centre of the
village of Bossébangou;

 the endpoint of the frontier in this sector is then constituted by the point where this straight line
intersects with the right bank of the River Sirba. - 88 -

Diagram No. 1 - View of the line between Tao and the River Sirba

Tao astronomic marker

View of the line between Tao and the River Sirba
Scale 1:1,000,000

Diagrams for illustrative purposes only “Point where the frontier reaches the River
Sirba at Bossébangou”

107 - 89 -

Diagram No. 2 - Illustration of the method used to determine the point where the
frontier reaches the River Sirba at Bossébangou

Illustration of the method used to determine the point where the
frontier reaches the River Sirba at Bossébangou

Scale 1:20,000

Left bank
of the Sirba

To the TAO
astronomic
marke r

Right bank
of the Sirba

“Point where the frontier reaches
the River Sirba at Bossébangou”

Centre of the village of
Bossébangou

Diagrams for illustrative purposes only

108 - 90 -

4.24. In short, the frontier in this sector, pursuant to the 1927 Arrêté, thus passes through the

following three points in succession:

 the Tong-Tong astronomic marker (Lat.: 142 °45'3.2" ; Long0.°25'1.7")
(WGS 84 ellipsoid) 323;

 the Tao astronomic marker (Lat.: 14 03' .N"; Lon0g2.' E")
(WGS 84 ellipsoid) 324;

 and the point where the right bank of the River Sirba intersects with the straight line
connecting the Tao astronomic marker to the village of Bossébangou (Lat.: 13°21'06.5"N;

Long. 1° 17' 11.0" E) (Clarke 1880 ellipsoid).

4.25. These three points having been located, it now remains to define the line connecting

them.

2. The line connecting the three points designated in the Arrêté

4.26. The brevity with which the French colonizer defined the course of the inter-colonial
boundary in this sector in 1927 (“[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 not the least
doubt that the boundary adopted was of an artificial nature (A). It is equally clear that this artificial
boundary consists of two straight lines connecting the three points designated in the Arrêté (B).

This most natural interpretation of the text of the 1927 Arrêté was, furthermore, officially endorsed
by the Parties (C) and corresponds in every respect to the practice followed in the field of boundary
delimitation (D).

A. The 1927 Arrêté adopts an artificial boundary in this sector

4.27. It is clear simply from reading the 1927 Arrêté that, in the sector concerned, it is an

artificial and not a natural boundary that has been adopted. Indeed no natural feature is referred to,
contrary to other sectors of the boundary, where the Arrêté deliberately describes its course as
following a natural feature, such as peaks (for example, the “the crests of the Djoapionga hills” in

the fifth paragraph) or watercourses (for example, the “course of the Tapoa” in the seventh
paragraph). In contrast, there is no such reference in the part of the Arrêté relating to this sector of
the frontier.

4.28. The amended Arrêté thus states that “[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”. If
we extract from this passage the references that concern only the course of the boundary strictly
speaking and exclude any description of the points through which it passes, it is obvious that the
Arrêté delimits the boundary in the form of an artificial line alone: in accordance with the Arrêté,

“this line then turns towards the south-east, cutting . . . and reaching . . .”.

4.29. Two facts confirm this very obvious interpretation of the Arrêté.

323
See above, para. 4.13.
324
See above, para. 4.16. - 91 -

4.30. Firstly, the only detail given in the Arrêté concerning the course of the boundary is that

it goes in a “south-east” direction from the Tong-Tong astronomic marker. This shows that the
course was simplified, defined solely— yet adequately— by ascertaining the points through
which the boundary, whose general direction was also described, passes.

4.31. Further, in 1923, the Commander of Dori cercle had argued in favour of the natural

boundary which was then formed by the River325ger — and which separated the Colonies of Niger
and Upper Volta untilDecember1926 — over the arbitrary and artificial boundary that was
eventually adopted in 1926 and 1927:

“Politically, the measure envisaged [which was implemented in
December 1926] would, no doubt, be to the advantage of the neighbouring Colony of
Niger by giving it control over both banks of the river; but it would, on the other

hand, be to the clear disadvantage of the Colony of Upper Volta by removing from
Dargol and Kokoro cantons a portion of territory which has always been part thereof
and which the Niger delimits geographically and naturally. Here there is in fact a

natural boundary and not one of those arbitrary and somewhat “fictitious” boundaries,
which take account of neither watercourses nor mountains, and to which, most often,
we pay little heed. In this instance, the Niger clearly marks the boundary between the

Sonhraïs cantons of Dori and the left bank of the river, commonly known as the
‘Haoussa bank ’, as opposed to the right bank or Gourma bank . This is a specific fact
which should, it seems, be taken into account.” 326

4.32. In his letter of 1February1926 to the Governor General of FWA, the Governor of

Upper Volta repeated these reservations in respect of the incorporation of the part of Dori cercle on
the left bank of the River Niger into the colony of Niger. However, he could only bow to the
decision of the highest authority of FWA who put forward economic arguments to justify the
incorporation of part of Dori cercle into the colony of Niger and the abandonment of the natural

frontier formed by the River Niger:

“In respect of the part of Dori Cercle on the bank of the Niger . . . I have always

been opposed to its incorporation into the neighbouring colony... However, you
informed me verbally that, after your recent visit to Niamey and considering the
situation in that part of the Niger basin, in light of the possibilities for economic

development, your decision was, in a manner of speaking, alr327y taken, in principle.
It would of course be unseemly for me to dispute it . . .”

4.33. The artificial nature of that delimitation did not seem to pose any great difficulties to
the colonial administration. Any territorial problems they might have encountered in the sector in
question involved the distribution of land ownership among the inhabitants of the two colonies

rather than being boundary problems strictly speaking— which, if they did arise, concerned the
demarcation, and not the actual delimitation, of the 1 927 boundary line, which was then only an
internal boundary of FWA. That much is clear from the telegram/letter of 23 January 1951 from

the Head of Téra subdivision to the Commander of Tillabéry cercle:

325
See above, para. 1.16.
32See the letter from the Commander of Dori cercle to the Governor of Upper Volta of 7 April 1923,

Ann.MBF21, p.1. These remarks were repeated by the Governor of Upper Volta in his letter of 7June1923 to the
Governor General of FWA (Ann. MBF 22, p. 4).
32See the letter dated 1February1926 to the Governor General of FWA from the Governor of Upper Volta,
Ann. MBF 25. - 92 -

“From the exchange of views which then took place, it emerged that the
Commander of Dori cercle agreed not to make the settlement of these disputes [these
were disputes as to the ownership of fields in the Oussaltan region] a question of

boundaries, while at the same time he reserved the right to instruct a liaison officer to
reconnoitre the boundary and mark out its course with the help of concrete markers.

Without contesting the appropriateness of such operations, they could be
considered to be part of the overall works which should probably be undertaken by
qualified staff once the heads of the respective Territories have reached agreement.

The theoretical boundaries they would mark out would not really help resolve
the land disputes which have arisen between the inhabitants of the two districts, whose
rights must be ascertained according to the rules of custom rather than by seeking to

determine whether the disputed territory is located in their territory of origin or
not.” 328

4.34. In adopting this artificial delimitation, the 1927 Arrêté merely reflected a practice
frequently followed by the colonizer in Africa, which consisted of adopting as boundaries “[mere]

straight lines traced on the drawing board with little relevance to the physical circumstances on the
ground” 329.

B. The course defined in the Arrêté consists of two straight lines

4.35. There are several facts which help establish that the artificial boundary defined in the
1927 Arrêté consists of two straight lines.

4.36. Firstly, the fact that the Arrêté does not specify the form to be ascribed to the artificial
line connecting the points through which the boundary passes — it merely states that it “then turns
towards the south-east, cutting... and reaching...”— can only be interpreted as referring to

straight lines:

 since two points are sufficient to define a straight line, the fact that the 1927 Arrêté only

designates the points through which the boundary passes, without providing any other details,
necessarily implies that those points are connected by a series of straight lines;

 a straight line is also the most obvious and the simplest solution to implement and visualize.
The fact that it is the shortest path between two points explains why it is the solution most
readily adopted. In the absence of any other indication in the Arrêté, there is no reason not to
330
presume that this most simple and obvious solution was adopted ;

328Telegram/letter from the Head of the Téra subdivision to the Tillabéry cercle, dated 23January1951,
Ann. MBF 40, p. 1.

329Territorial Dispute (Libyan Arab Jamahiriya/Chad), Judgment, I.C.J. Reports 1994 , Separate Opinion of
Judge Ajibola, p. 53, para. 9, and above, para. 2.38.
330
See, in the same connection, mutatis mutandis , P.C.A., The Government of Sudan/the Sudan People’s
Liberation Movement/Army (Abyei Arbitration ), Final Award of 22July2009, www.pca-cpa.org, para.746: “ By
delimiting the eastern and western boundaries of the Abyei Area in the foregoing manner, the Tribunal adopts the ABC
Experts’ use of lines of longitude in its delimitation of tribal boundaries, as the Tribunal finds that it was reasonable to
do so for both logical and practical reasons . . .”. - 93 -

 it is, more331r, the solution retained by the Parties for the sectors of their frontier not in
dispute : both in the sector running from the frontier with Mali to the Tong-Tong astronomic
marker and in the sector beginning at the Botou bend and ending at the River Mekrou, Burkina

and Niger agreed that the boundary delimited by the 1927 Arrêté consisted of a series of
straight lines connecting the frontier points identified, with the sole exception of those stretches
where the boundary follows a natural feature in accordance with the Arrêté. This justifies the

same (common-sense) interpretation being adopted for the sector between the Tong-Tong
astronomic marker and the point where the frontier reaches the River Sirba at Bossébangou. In
fact, in describing the course of the boundary, the Arrêté does not differentiate between the

different sectors: both in respect of the former (“non-disputed” sector332and the sector which
lies between them, it merely states that the line (in the singular ) passes successively through
a certain number of points. This was interpreted by the Parties as meaning that the line consists

of a series of straight lines connecting those points. What applied to the demarcated sectors
should thus also apply to the intermediate sector, in the absence of any indication to the
contrary in the Arrêté.

4.37. Secondly, the representation of the boundary as two separate straight lines, with one

changing direction in relation to the other at the turning point formed by the Tao astronomic
marker, is the one that Niger used in the atlas of maps annexed to its Memorial in the case of the
Frontier Dispute (Benin/Niger), in order to illustrate on a contemporary map the 1927 delimitation
333
between the Colony of Upper Volta and the Colony of Niger . It is also the form given by Niger,
in the same atlas, to the boundaries of Upper Volta as it was reconstituted in 1947. Given the scale
of the 1932 map used by Niger for illustrative purposes in its atlas, the course of the 1947 (and thus

the 1932) boundary can only be approximate. But what is highlighted by Niger is quite clear:
starting from the Tong-Tong astronomic marker, there is one straight line to the Tao astronomic
marker, then, after another turning point, there is a second straight line which ends at the River
334
Sirba .

4.38. Thirdly, the fact that the Arrêté does not specify the form the boundary should take

necessarily implies that the straight line option excludes all others. If any other artificial line had
been chosen (from a wide variety of possible types), it would necessarily have been accompanied,
firstly, by clarifications making it possible to determine the form of the artificial line, and,

secondly, by the information required to establish its course. The Niger delegation, a contrario,
acknowledged at the second ordinary session of the Joint Technical Commission in 1990 that such

clarifications would have been essential if anything other than the straight-line solution had been
retained in the Arrêté. It in fact claimed that the line, in its view a curve, adopted by the Arrêté of
1927 “is the arc of a circle, with a well-defined centre and radius” 335. Yet nowhere does the

1927 Arrêté actually state that the line is a curve, nor does it say what its centre and radius are.

4.39. The few cases in which an artificial boundary other than a straight line was adopted in

colonial practice attest to the fact that such a boundary cannot exist without being defined
exactly— if only because it is necessary to define what type of non-straight line (and there is a

331See above, Chapter III, para. 3.13.
332
“The boundaries of the Colonies of Niger and Upper Volta are determined as follows: A line starting from the
heights of . . .; this line then turns towards the south-east, crossing . . . and reaching . . . It [still the same line] almost
immediately turns back up [etc.]” (italics added).
333
See p. 105 of the atlas, reproduced in Cartographic Annex MBF 16.
334Ibid., p. 129, reproduced in Cartographic Annex MBF 17.

335Report of the second ordinary session of the Joint Technical Commission on Demarcation of the Niger-Burkina
Frontier, held in Ouagadougou from 23 to 28 July 1990, Ann. MBF 87, p. 3. - 94 -

wide variety of possible forms) had been chosen. One example is the Franco-British Convention of
14 June 1898 on the Niger, Article 4 of which described the boundary between the two States to the
east of the River Niger thus:

“Starting from the point on the left bank of the Niger indicated in the previous
Article, viz., the median line [of the Dallul Mauri, the frontier shall follow this median

line] until it meets the circumference of a circle drawn from the centre of the town of
Sokoto with a radius of 100 miles (160.932 metres). From this point it shall follow the
northern arc of this circle as far as its second intersection with the 14 parallel of north
336
latitude.”

There is nothing similar or comparable in the 1927 Arrêté. Moreover, it is instructive to note

that the boundary adopted in 1898 was abandoned in 1906 in favour of a simpler delimitation
consisting of straight lines connecting different points 337.

4.40. In the same connection, the Agreement of 19October1906 relative to the Frontier
between the British and French Possessions from the Gulf of Guinea to the Niger could be cited to

show that when an artificial boundary other than a straight line is selected as a frontier, it always
comes with corresponding clarifications:

“ArIticle

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

26. Thence it follows the circumference of a circle of 8½ kilom. radius, having
for centre the centre of the town of Okuta (Okouta), to a point situated 1 kilom. from

the direct road from Okuta (Okouta) to Boria, measured on the west of and at right
angles to this road;

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

33. Thence it follows on the west side of Guri (Goré) the circumference of a

circle of 4 kilom. radius, described round the centre of that village as centre, to a point
situated 1 kilom. from the road from Guri (Goré) to Yashikira (Yasssikéré), measured
at right angles to, and on the west of, this road

338
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .”

Once again, there is no such information— which is impossible to assume— in the
1927 Arrêté.

336
This Convention is reproduced in Ann.B.8 of Niger’s Memorial ofAugust2003 in the case of the Frontier
Dispute (Benin/Niger). See Ann. MBF 1.
33See the first article of the Convention of 29 May 1906 between Great Britain and France, respecting the

Delimitation of the Frontier between the British and French Possessions to the East of the Niger, signed at London,
9 April 1906, reproduced in Ann. B.21 of Niger’s Memorial of August 2003 submitted in the case of the Frontier Dispute
(Benin/Niger). See Ann. MBF 6.
338
The Agreement adopts another delimitation in the form of a circle (or “tangent to a circle”), the centre and
radius of which it defines in paragraphs 35, 36, 42, 44, 47, 49, 53, 55 and 57 to 60. This Agreement was reproduced in
Ann.B.22 of Niger’s Memorial ofAugust2003, submitted in the case of the Frontier Dispute (Benin/Niger) . See
Ann. MBF 7. See also, among other examples, Article 1 of the Agreement in respect of the delimitation of the frontiers
between Cyrenaica and Egypt, signed at Cairo on 6 December 1925: the frontier “shall follow the direction of the arc of
a circle drawn with Beacon Point (Ezl et El Gattara) as its centre and with a radius of 10 (ten) kilometres from the
abovementioned point to . . .” (cited in I. Brownlie, African Boundaries, 1979, pp. 104-105.) - 95 -

4.41. Fourthly, the drafts written preparatory to the Arrêté and prior to the Erratum confirm
that the boundary is formed by two straight lines which undergo two successive changes in
direction, one at the Tong-Tong astronomic marker and the other at the Tao astronomic marker.

4.42. The original version of the Arrêté of 31 August 339stated the following:

“This boundary is determined... by a line passing through... Mount
Doumafondé, which then turns towards the south-east, leaving the ruins of Tong-Tong
to the east and descending in a north-south direction, cutting the Téra-Dori motor road
340
to the west of the Ossolo Pool, until it reaches the River Sirba . . .”

4.43. That line had already been approved by the governing authorities of the Colonies of
Niger and Upper Volta on 2 February 1927 in a report which reads thus:

“Record of agreement

In the year one thousand nine hundred and twenty-seven, on the second of
February,

Between the Governor of the Colony of Niger and the Inspector of
Administrative Affairs, Mr.L EFILLIATRE , representative of the Governor of Upper

Volta, at Téra.

Having regard to the Decree dated the twenty-eighth of December, one

thousand nine hundred and twenty-six.

The following was agreed:

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

[The cantons transferred to Niger] are bounded to the north by the current
boundary with Sudan (Gao cercle) as far as the heights of N’Gourma, and to the west
by a line passing through the Kabia ford . . ., which then turns towards the south-east,

leaving the ruins of Tong-Tong to the east and descending 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.

No opposition on the part of the local inhabitants having been encountered, this
record was closed and signed by the Parties.” 341

4.44. These two documents invite the following comments:

 in the first place, it should be noted that they use the same frontier points as those described,
with greater precision, in the Erratum;

339
In its Judgment in the case of thFrontier Dispute (Burkina/Mali) , the Chamber of the Court accepted the
subsidiary and probative character of the initial version of the 1927 Arrêté (see the Judgment of 22 December 1986, I.C.J.
Reports 1986, p. 590, para. 69).
340
Italics added.
34Ann. MBF 30. - 96 -

 secondly, it is clear from the way they are written that the boundary line consists of two
sections with different orientations: the first descends “in a north-south direction”, the second

continues “until it reaches” the River Sirba in a direction that can only be “south-east”, since
that is the general direction of the frontier from the Tong-Tong astronomic marker (“which
then turns towards the south-east”, as both texts specify);

 it necessarily follows that the statement whereby the line turns towards the south-east—
repeated in the Erratum — was in no way intended to describe the form of the line connecting
the three frontier points designated in the Arrêté. Its sole purpose is to indicate the existence of

a turning point in the boundary, between the section that arrives at the Tong-Tong astronomic
marker and the section that runs from there and ends at the River Sirba at Bossébangou. If the
reference to turning was meant to describe the form of the line (for example, to switch from a
straight line to a curved line), it would then be inconsistent with the clarification coming

immediately afterwards, whereby the first stretch of this new section of the boundary follows a
“north-south” direction — one line cannot curve in two different directions at once. However,
there would be no inconsistency in stating, firstly, that from the point where the line turns at the
Tong-Tong astronomic marker this section of the boundary undergoes an overall change in

direction to the south-east in relation to the previous section, and, secondly, that this overall
stretch consists of two sections with different orientations, one “north-south”, the other
south-east;

 in this regard the instruments preparatory to or preceding the Erratum confirm that the two
sections described are indeed straight lines: on the one hand, it is stated that the first section
follows a “north-south” direction (“and descending in a north-south direction”); and on the
other, that the second section runs “until it reaches” the River Sirba at Bossébangou. It would

be difficult to interpret these phrases as meaning anything other than straight lines.

4.45. There is no reason to believe that a different solution prevailed in the corrected version

of the Arrêté. The corrected version merely gives a more precise description of the points through
which the line passes (from then on the Tong-Tong and the Tao astronomic markers are referred to
specifically) and it was thus no longer necessary to describe the orientation of each section. The
precise identification of the points through which the boundary passes was henceforth sufficient to

define its course precisely. In the absence of any other clarification, the way the text of the Arrêté
is structured (“this line then turns towards the south-east, cutting . . ., and reaching . . .”) leaves no
doubt that the boundary consists of two straight lines connecting the three points.

C. The interpretation whereby the course of the frontier consists of two straight lines was
officially endorsed by the two Parties

4.46. In light of the preceding evidence, it is not surprising that Burkina and Niger agreed to
consider that the 1927 Arrêté delimited their common frontier with two straight lines in this sector.

4.47. That was the interpretation adopted when the boundary line was first considered by the
Joint Technical Commission on Demarcation established in 1987. Further, inSeptember1988, it
was unanimously decided to begin demarcation of the course in the form of two straight lines
derived by the two Parties from their interpretation of the 1927 Arrêté, by plotting on the 1960 map

a course consisting of two successive straight lines connecting the Tong-Tong and Tao astronomic
markers, and the Tao marker to the point where the frontier reaches the Sirba at Bossébangou.

4.48. According to the record signed by both delegations, the very purpose of the Joint

Commission’s fourth meeting of 26, 27 and 28September1988 was in fact “to plot, on the map, - 97 -

the line resulting from the field survey conducted by the Technical Sub-Committee and to submit342
the results of said work to both Governments for final selection of the frontier line” . More
specifically,

“The agenda included one item:

 to plot on the map the boundary line, as surveyed by the Technical
Sub-Committee, pursuant to the additional work requested of it at the meeting of
12-15 May 1988 at Diapaga.

The Technical Sub-Committee then reported on the work of the survey mission
and presented on the map the boundary line it had surveyed.

From the attached Report on completion of the field survey work, the following
key points emerged:

1. According to the basic texts ( Arrêté No. 2336 of 31 August 1927 and its
Erratum No 2602/APA of 5October1927) the boundary line passes three (3)
categories of points:

astronommaikers;

nafeartlres;

localities.

The experts are thus 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 28 March 1987.” 343

4.49. Moreover, the report stated that it should “be noted that at several locations the frontier

line resulting from the basic texts does not coincide with 344 line on the IGNF 1:200,000 map or
with certain administrative realities on the ground” , as in some “specific cases”, “certain
administrative entities are clearly outside the two boundary lines” 345. However, given that, where

the two lines did not coincide, the line defined in the Arrêté — as interpreted consensually by the
Parties — prevailed over the line on the map, it was appropriate to adopt the former line. That is
precisely what the authors of the report did by plotting the line defined in the Arrêté (straight lines

connecting the points cited in the Arrêté) on the 1960 map.

4.50. As was explained in the report in question, in paragraph II.5:

“Two composites of IGNF 1:200,000 boundary maps, 1960 edition, a map
346
overlay and transparency are attached to this report[ ]. They show:

 the boundary line as it exists on the IGNF map, 1960 edition

34Ann. MBF 81, p. 1.

34Ibid, p. 2.
344
Ann. MBF 81, p. 3.
345
Ann. MBF 82.
34See Cartographic Annex MBF 14. - 98 -

 the boundary line resulting from the interpretation of the basic texts and from the
field survey mission . . .” 347

It is clear from these composites and overlays that the consensual line “resulting from the
interpretation of the basic texts and from the field survey mission” connected the three points cited

in the Arrêté with two straight lines.

4.51. In November 1989, the Commission arranged for this line to be established officially
on the ground and planned for the installation of just 45 markers at first (because of budgetary

constraints), 33 of which were to be placed at each “turning” p348t in the line, with the remaining
12 supplementing those in densely populated areas . It was as part of this process that,
betweenDecember1989 and May 1990, “23 markers” were effectively installed “where the

frontier changed direction”, including in the “northern section”:

“* No. 007 Tong-Tong (astronomic marker)

* No. 008 Tao (astronomic marker)” 349.

The choice of these two points as “turning points” was thus to confirm the 1988 line and the
obvious interpretation of the 1927 Arrêté by making the line undergo two successive changes of

direction at these two points.

4.52. The consensual line and interpretation adopted in 1988 were nevertheless called into
question by the Niger delegation in July 1990, on the grounds of an alleged misinterpretation of the
1927 Arrêté 350. This about-turn calls for two comments:

 it was very sudden, since, less than three months earlier, the Niger delegation had reasserted
the validity of the 1988 consensual line, voicing its disagreement with only one point (despite

having adopted it in 1988): according to the Niger delegation, “a misinterpretation of the
Erratum makes the boundary terminate at the River Sirba at Bossébangou” 351. This

criticism— the only one made in May 1990 by the Niger delegation in respect of the 1988
consensual line in this sector— was very surprising, given that the line thereby did exactly
what the text of the Erratum provides for (“and reaching the River Sirba at Bossebangou” 352).

That is most probably why this isolated criticism was not subsequently repeated by the Niger
delegation, which, on this point, thereafter relied on the Arrêté’s clear text;

 the Niger delegation’s new position, adopted less than three months later, inJuly1990, was
promptly submitted to the relevant ministers of the two Parties which, in May 1991, simply

rejected it and once again endorsed the interpretation made official in 1988.

347
Ann. MBF 42 (italics added).
348See the Report of the fifth meeting of the Joint Technical Commission on Demarcation held on 13 and

14 November 1989, Ann. MBF 84; see also above, para. 1.71.
349See the report of the second ordinary meeting of the Joint Technical Commission on Demarcation of the

Niger-Burkina Frontier, held at Ouagadougou from 23 to 28 July 1990, Ann. MBF II; Ann. MBF 87.
350See above, paras. 1.75 to 1.78, and para. 2.32.

351See the record of the extraordinary meeting of the Joint Technical Commission on Demarcation, 15 May 1990,
Ann. MBF 85, p. 1.

352See above, section 1, C. - 99 -

4.53. In May 1991, as Burkina recalled earlier 353, the ministers agreed a compromise which

consisted of two distinct parts:

 the first part, relating to the present sector, retained the boundary as described in the Arrêté,

and recalled the consensual interpretation given in 1988;

 the second part departed from the line agreed in 1988 in accordance with the Arrêté in the

sector between the salient of the four villages and the River Mekrou, and replaced it with the
line on the 1960 map, contrary to the provisions of the Arrêté which, unlike the line on the

map, notably adopts a “straight line” between the intersection of the River Sirba and the Say354
parallel and the point situated at 1,200 metres west of the village of Tchenguiliba .

4.54. The decision of May 1991 represented a compromise in so far as it retained the line of
the 1927 Arrêté in its entirety in the first sector and partly abandoned it in the second 355. Thus, in

respect of the first sector (the one at issue here) it was drafted in a declaratory manner (“the frontier
consists of a series of straight lines”), whilst for the second it was drafted in a constitutive mode

(“From the River Sirba at Bossébangou to the River Mekrou, it was agreed to adopt the frontier
line as shown on the IGN/France map, 1960 edition). This difference is justified: in the first
instance, it was necessary only to recall the interp retation of the line which the two Parties had

considered in 1987 as binding between them (and which has never ceased to be so); in the second
instance, this line was being set aside in favour of another. This second decision does not affect the
interpretation of the 1927 Arrêté, as it in fact sought not to apply part of it 35. The first decision, on

the other hand, is fully binding on the Parties to the present proceedings, since it adopts a
consensual interpretation of the Arrêté which constitutes the frontier title 357.

4.55. The interpretation of the amended Arrêté adopted by mutual agreement by the

competent ministers of the two Parties in this sector of the frontier confirmed the consensual
interpretation that had been adopted in 1988: Burkina and Niger once again recognized that the
1927 Arrêté delimited their frontier with two straight lines: “1. From the Tong-Tong astronomic

marker to the River Sirba at Bossebangou, passing through the Tao astronomic marker, the frontier
shall consist of a series of straight lines.” 358

4.56. As the Minister for Territorial Administration of Burkina wrote in his ministerial
359
council report of 2 July 1991 , he and his counterpart thus decided to confirm the consensual line
of 1988:

“1. Part of the frontier from the Tong-Tong astronomic marker to the River
Sirba at Bossébangou

353
See above, Chapter I, Section 2, para. 1.79.
354
See below, Sec. 3, paras. 4.146-4.151.
355See above, Chap. I, paras. 1.79-1.80.

356Moreover, the political solution envisaged in May 1991 to restart demarcation work was later rejected by Niger
(see above, Chap. II, Sec. 2, para. 2.32).

357See above, Chap. II, paras. 2.36 et seq.

358Joint Communiqué on the Ministerial consultative and working meeting between Niger and Burkina, held on
14 and 15 May 1991 in Ouagadougou, Ann. MBF 89.
359
Ministerial Council Report dated 2July1991, by the Minister for Territorial Administration of Burkina,
Ann. MBF 90. - 100 -

This part of the frontier consists of two straight lines from the Tong-Tong

astronomic marker to the Tao astronomic mar360 and from the latter to the River Sirba
at Bossébangou. (As initially agreed).” ,

that is to say, as it had been agreed as part of the consensual line of 1988.

4.57. Since then, the Niger delegation has once again called into question this (doubly)
361
consensual interpretation , which explains why the ensuing dispute has been submitted to the
Court. However, the Niger delegation did not have the power to challenge an interpretation that is
fully binding on the State of Niger. That interpretation thus continues to be binding between the

Parties.

D. The course of the frontier as two straight lines corresponds to the practice followed in the

field of boundary delimitation

4.58. The interpretation resulting from the terms of the text of the 1927 Arrêté, and which
was officially endorsed by the two Parties in 1988 and again in 1991, is corroborated in every

respect by examining boundary delimitation practice. This shows that when a text delimiting a
boundary designates points without specifying that they are connected by a natural feature or by an
artificial line of a specific kind, that text is interpreted as connecting the points in question by a

series of straight lines. This same boundary practice also confirms that reference to a “turn” in a
delimitation text does not designate an artificial line that is not straight; it refers on the contrary to
the existence of a turning point between two artificial lines which, in the absence of any indication

to the contrary, are straight lines. This twofold conclusion can be deduced from jurisprudence
relating to both land and maritime delimitation.

4.59. In the case of the Frontier Dispute (Burkina/Mali ), for example, the Chamber of the
Court had to interpret and apply colonial texts which, in several places, used the verb “turn”
[s’infléchir in French]. At no point in its Judgment of 22December1986 did the Chamber posit

that the use of this verb referred to any course other than one consisting of straight lines. On the
contrary, consultation of the line adopted by the Court confirms that, when the verb “turn” is used,
the frontier simply changes direction. The Court at times even equates “turn” and “straight line” in

the operative part of its Judgment:

“179. For these reasons, The Chamber

Unanimously,

Decides

A. That the frontier line between Burkina Faso and the Republic of Mali in the
disputed area, as defined in the Special Agreement concluded on 16September1983
between those two States, is as follows:

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

360
Ibid. (italics added).
36Report of the 3rd ordinary session of the Joint Technical Commission on Demarcation of the Frontier held in
Niamey from 2-4 November 1994, Ann. MBF 91, p. 3; Report 18 to 21 July 2001 of the Joint Technical Commission on
Demarcation, p. 5, Ann. MBF 94; Report of the 5th ordinary session of the Joint Technical Commission on Demarcation
from 2-4 November 1994, Ann. MBF 97. - 101 -

(2) At point B, the line turns eastwards and intersects the track connecting

Dionouga and Diguel at approximately 7.5 kilometres from Dionouga at a point with
the geographical co-ordinates 1° 54' 24" W and 14° 29' 20" N (point C).

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

(6) From point H, the line follows the IGN line as far as the point with the
geographical co-ordinates 0° 26' 35" W and 15° 05' 00" N (point 1); from there it

turns towards the south-east and continues straight as far as point J defined below.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

(8) At point K the line turns towards the north-east and continues straight as far
as the point with the geographical co-ordinates 0° 14' 44" W and 15° 04' 42" N (point
L), and, from that point, continues straight to a point with the geographical

co-ordinates 0° 14' 39" E and 14° 54' 48" N (point M362situated approximately
3 kilometres to the north of the Kabia ford . . .”

4.60. In the reasoning of its Judgment, the Court also expressed the idea that a turn
[infléchissement] is not only not inconsistent with the choice of straight lines, but, on the contrary,
in principle is indicative of such a choice, since it should be assumed that the colonial authorities

adopted the least complex solution, which, according to the Chamber of the Court, is confirmed by
colonial practice contemporary to the 1927 Arrêté (in particular the 1926 Atlas des cercles):

“Provided it observes the administrative appurtenance of the villages, a
boundary may also follow a straight line or consist of a series of straight lines all
running in the same general direction, with some minor deviations. The colonial maps
of the period, for example, the 1926 Atlas des cercles, show clearly that the latter was

the form most frequently taken by the cercle boundaries. It is also of relevance that
the description given by the administrator of the cercle of Mopti of the subdivision
boundary corresponding to the boundary contemplated in Order 2728 AP refers to a

single line starting from the village of Yoro and subsequently ‘heading northeastward
as far as the pool of Toussougou’. Th e Chamber concludes that in adding the detail
that the line was to ‘leave’ to the cercle of Mopti the village of Yoro and the ‘four

villages’, the Geographical Service of Fren363West Africa did not intend the line to
take a more complex form as a result.”

4.61. The Court was again faced with the verb “turn” [s’infléchir] in the case of
Cameroon v. Nigeria. Certain paragraphs of the Thomson-Marchand Declaration of 1929-1930
(from exactly the same era as that of the 1927 Arrêté) read thus:

“21. Thence [the frontier line] in a south-south-westerly direction [ Ensuite elle
[la ligne frontalière s’infléchit au sud-sud-ouest], leaving the village of Dile on the
British side, the village of Libam on the French side to the hill of Matakam.

22. Thence [still the same line, as in the 1927 Arrêté] running due west to a
point to the south of the village of Wisik where it turns to the south on a line running

along the watershed and passing by Mabas on the French side, after which it leaves

362
Previously cited Judgment, pp. 649-650, para. 179 (italics added). In respect of the final stretch of the frontier,
the Court found in favour of Burkina which had considered that “the IGN line veers slightly northward” and concluded
that “this straight line must constitute the final segment of the line which it is required to draw” (ibid., p. 648, para. 174).
36Previously cited Judgment, p. 624, para. 130 (italics added). - 102 -

Wula on the English side runn364 south and bounded by cultivated land to the east of
the line of the watershed.”

4.62. It is noteworthy here that the Thomson-Marchand Declaration describes a line which
turns in a single direction, in exactly the same way as the 1927 Arrêté does, and does not deem the
wording inconsistent with the fact that this line passes through several points and runs in several

directions in turn (all towards the south), including along the line of a watershed.

4.63. In paragraph 179 of the same Judgment, the Court also chose to use the verb “turn” for
the sole purpose of indicating the change in direction of the frontier line which, previously running
in a west-north-west direction, bifurcates to the north and follows the course of the same river.

According to the Court,

“From that point, the boundary follows the course of the River Mburi. It...

takes a . . . course . . . west-north-west along a stretch where the river is also called the
Maven or the Ntem. Then, some 2km further on, it turns to run due north where the
River Mburi is also called the Manton or Ntem (see below, p. 396, sketch-map
365
No. 10).”

As in the 1927 Arrêté, the reference to turning has no other purpose here than to describe the

change of direction undergone by one section of the frontier in relation to the next, and it is not
intended to delimit the frontier as such. In this instance, the frontier follows the course of the River
Mburi, whose orientation the Court describes by specifying that it turns “due north” at the end of

this sector of the frontier.

4.64. Similarly, in paragraph 147 of its 2002 Judgment, the Court again cites the

Thomson-Marchand Declaration, which reads as follows:

“38. Thence from this point along the line of the watershed between the River

Baleo and the River Noumbero along the crest of the Tschapeu Range, to a point
2kilometres to the north of Namberu, turning by this village, which is in Nigeria,
going up a valley north-east and then south-east, which crosses the Banglang range
366
about a kilometre to the south of the source of the Kordo River.”

4.65. The Court considered that “the final part of paragraph 38 of the Thomson-Marchand
Declaration pose[d] problems of interpretation”, not because the term “turning” was used, but on
the grounds that it contained “fundamental errors of a material nature” regarding the fact that there
367
was no valley “in the area running ‘north-east, then south-east’” . The method used by the Court
to establish the boundary up to the final point in this sector is very interesting to note. Faced with
an uncertainty, the Court determined the arrival point of the boundary and then connected it, with a

straight line, to the terminal point which it had succeeded in locating:

“The Court observes that, while the text of the Thomson-Marchand Declaration

contains scant information enabling it to determine the precise course of the boundary
in this sector, the description of it in the Logan-Le Brun procès-verbal is, however, far

36Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial Guinea
intervening), Judgment, I.C.J. Reports 2002, p. 369, para. 103 — italics added.
365
Ibid., p. 395, para. 179 — italics added.
366
Ibid., p. 384, para. 147 — italics added.
36Ibid., p. 384, para. 150. - 103 -

more detailed and enables such a determination to be made. The Court... notes in

particular that the Logan-Le Brun procès-verbal and paragraph 38 of the
Thomson-Marchand Declaration appear to make the boundary in this sector terminate
at the same point. Thus the Logan-Le Brun procès-verbal provides that the boundary
runs to Mount Tapere, situated ‘about a mile to the south of the source of the Mayo

Kordo’, whilst the English text of the Thomson-Marchand Declaration provides for
the boundary to pass through a point ‘about a kilometre to the south of the source of
the Kordo River’. The French text of paragraph 38 omits the phrase ‘to the south of’.

The Court is bound moreover to note in this regard that the part of the boundary
situated to the north of the source of the Noumberou, on which the Parties are in
agreement, follows the boundary established by the Logan-Le Brun procès-verbal.
The Court considers that it is the boundary line proposed by Nigeria which is to be

preferred. That is the line which runs most directly to Hosere Tapere . . . indicated by
the Logan-Le Brun procès-verbal as the terminal for this section of the boundary...
The Court accordingly concludes that the final part of paragraph 38 of the

Thomson-Marchand Declaration must be interpreted as providing for the boundary to
follow the course of the River Noumberou as far as its source, and then from that point
to run in a straight line as far as Hosere Tapere as identified by the Court (see below,
368
p. 386, sketch-map No. 9).”

4.66. The Judgment handed down on 11 September 1992 by the Chamber of the Court in the

case of the Land, Island and Maritime Frontier Dispute between El Salvador and Honduras leads to
the same conclusions. In respect of the “Third Sector— Eastern Sub-Sector”, the Chamber
describes the “Frontier delimited in the Judgment” as follows:

“The boundary thus runs from the confluence of that quebrada with the river
Sazalapa (point A) to the hill shown on the maps with a spot height of 875 metres
(point B), and then turns southwards to the landmark described as the ‘crest of a hill

that has an opening [portezuelo] through which passes the road that leads to the town
of Gracias a Dios’. Of the several hills in the region, the most likely is, in the
Chamber’s view, the one which El Salvador’s interpretation identifies with the
Guampa hill (marked as point C on sketch-map No.C-4) .... From there the line,

inclining still more to the south , runs through the triangulation point marked as La
Canada (point D on sketch-map No.C-4), to the ridge joining... (point E on
sketch-map No. C-4).” 369

4.67. The situation is very similar here to that described in the 1927 Arrêté: the line, while
turning towards the south, must pass through one point before reaching another point at which the

line undergoes a more marked change in direction (“inclining still more to the south”). The Court
connects these different points by straight lines, as can be seen on sketch-map No. C-4 on page 457
of the Court’s Reports.

4.68. In the operative part of the Judgment, points A, B, C and D of sketch-map No.C-4
become points H, I, J and K of the frontier. The Court defines the boundary line in this sector thus:

“and thence upstream along the middle of the river Sazalapa to the confluence of the
quebrada Llano Negro with that river (point H on Map No.III annexed . . .); from
there south-eastwards to the top of the hill (point 1 on Map No. III annexed . . .), and

thence south-eastwards to the crest of the hill marked on the map as a spot height of

368
Ibid., p. 385, paras. 151-152 — italics added.
369
I.C.J. Reports 1992, pp. 456-457, para. 156 — italics added. - 104 -

1,017 metres (point J on Map No. III annexed . . .); from there the boundary, inclining

still more to the south, runs through the triangulation point known as La Canada
(point K on Map No. III annexed; co-ordinates . . .)” 370.

4.69. Here the Court replaces “turns southwards” with “south-eastwards” and retains exactly
the same line (the lines connecting points I to K on sketch-map No. III annexed to the Judgment are

the same straight lines in sketch-map No.C-4). It follows that these two phrases are
interchangeable and that in using them, the Chamber of the Court merely intends to show that the
frontier, formed by a series of straight lines connecting defined points, changes direction.

4.70. This conclusion is fully borne out by the jurisprudence on maritime delimitation.
International judges, and the principal judicial organ of the United Nations in particular, in fact

commonly use the verb “turn” to describe a change in direction of a boundary formed of several
straight lines. When, on the other hand, they describe a boundary that has a different form, they
take care to provide a precise description so that it may be identified. This consistent practice

confirms that, when used without any other description or clarification, the verb “turn” designates a
turning point in a series of straight lines.

4.71. Again in the case of Cameroon v. Nigeria, the Court thus used the verb “turn” to define
the course of the maritime frontier:

“from point G the delimitation line should directly join the equidistance line at a point
with co-ordinates 8° 2l' 0" longitude east and 4° 17' 00" latitude north, which will be
called X. The boundary between the respective maritime areas of Cameroon and

Nigeria will therefore continue beyond point G in a westward direction until it reaches
point X at the above-mentioned co-ordinates. The boundary will turn at point X and
continue southwards along the equidistance line.” 371

4.72. As evidenced by illustrative sketch-map No.12 annexed to the Judgment 37, the line
connecting points G and X, and the line running southwards from point X, the turning point, are

straight lines.

4.73. The Court had already done the same the previous year in its Judgment in the
Qatar/Bahrain case. In paragraph 222 of its Judgment of 16March2001, the Court decided as
follows:

“Taking account of al1 of the foregoing, the Court decides that, from the point
of intersection of the respective maritime limits of Saudi Arabia on the one hand and
of Bahrain and Qatar on the other, which cannot be fixed, the boundary will follow a

north-easterly direction, then immediately turn in an easterly direction, after which it
will pass between Jazirat Hawar and Janan; it will subsequently turn to the north and
pass between the Hawar Islands and the Qatar peninsula and continue in a northerly

direction, leaving the low-tide elevation of Fasht Bu Thur, and Fasht al Azm, on the
Bahraini side, and the low-tide elevations of Qita’a el Erge and Qit’at ash Shajarah on

37Ibid., p. 612, para. 427.
371
Previously cited Judgment of 10 October 2002, p. 448, para. 307 — italics added.
37Ibid., p. 449. - 105 -

the Qatari side; finally it will pass between Qit’at Jaradah and Fasht ad Dibal, leaving
373
Qit’at Jaradah on the Bahraini side and Fasht ad Dibal on the Qatari side.”

This description is represented, on the sketch-map annexed to the Court’s Judgment for the

purpose of illustr374ng the delimitation line adopted, by a series of straight lines connecting
specified points . In particular, the provision whereby the boundary will “immediately turn in an
easterly direction” is illustrated by a turning point between two straight lines (the section

connecting points 2 and 3 undergoes a change of direction in relation to the section connecting
points 1 and 2). Similarly, the provision whereby the boundary “will subsequently turn to the
north” is represented in the form of a turning point between the straight line connecting

points 10 and 11 and the straight line connecting 9 and 10.

4.74. In the case concerning the Territorial and Maritime Dispute between Nicaragua and

Honduras in the Caribbean Sea , the Court once again used the term “ s’infléchir” to designate a
change in direction of a section of the boundary in relation to the preceding one. On this occasion,
one of the sections (the one starting from the turning point) did not take the form of a straight line,

which is precisely the reason why the Court on this occasion defined the precise form of this
section:

“La ligne de délimitation doit commencer au point de départ fixé sur la
bissectrice à 3 milles marins au large (…). A partir de ce point, elle suit la bissectrice
jusqu’à ce qu’elle rejoigne la limite extérieure de la mer territoriale de 12 milles

marins de Bobel Cay. Elle s’infléchit alors vers le sud pour suivre le pourtour de cette
mer territoriale jusqu’à ce qu’elle rencontre (…)”

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

“The line of delimitation is to begin at the starting-point 3 nautical miles
offshore on the bisector (…). From there it continues along the bisector until it

reaches the outer limit of the 12-nautical-mile territorial sea of Bobel Cay. It then
traces this territorial sea round to the south until it reaches (…)” 375

4.75. The Court again used the verb “ s’infléchir” to designate a change of direction (here of
two straight lines, in the absence of any other clarification) in the case concerning the Maritime

Delimitation in the Black Sea between Romania and Ukraine:

“Le segment initial de la ligne d’équidistance provisoire entre les côtes
adjacentes de la Roumanie et de l’Ukraine est construit à partir des points de base

constitués, pour la côte roumaine, par la base de la digue de Sulina et, pour la côte
ukrainienne, par la pointe sud-est de l’île de Tsyganka. A partir d’un point situé à
mi-distance de ces deux points de base, il se dirige vers le sud-est jusqu’au point A

(…), où son tracé s’infléchit sous l’effet d’un point de base situé sur la péninsule de
Sacaline, sur la côte roumaine. Au point A, la ligne d’équidistance change légèrement
de direction pour se poursuivre jusqu’au point B (…), où son tracé s’infléchit sous

l’effet du point de base situé sur le cap Tarkhankut, sur la côte opposée de l’Ukraine.

In its initial segment the provisional equidistance line between the Romanian

and Ukrainian adjacent coasts is controlled by base points located on the landward end
of the Sulina dyke on the Romanian coast and south-eastern tip of Tsyganka Island on

37I.C.J. Reports 2001, p. 109, para. 222 — italics added.
374
Ibid., p. 116, para. 251.
37Judgment of 8 October 2007, I.C.J. Reports 2007, p. 759, para. 320 — italics added. - 106 -

the Ukrainian coast. It runs in a south-easterly direction, from a point lying midway
between these two base points, until Point A (…) where it becomes affected by a base
point located on the Sacalin Peninsula on the Romanian coast. At Point A the

equidistance line slightly changes direction and continues to Point B (…) where it
becomes affected by the base point located on Cape Tarkhankut on Ukraine’s opposite
376
coasts.”

Further on, the Court summarizes this line by stating that it “ suit cette ligne jusqu’à ce que

son tracé s’infléchisse sous l’effet de points de base situés sur les côtes de la Roumanie et de
l’Ukraine qui se font face ” “follows that line [the equidistance line] until it becomes affected by
base points [in the plural, since here the Court is describing the A-B and B-C segments at the same
377
time] on the opposite coasts of Romania and Ukraine” . And as is clearly apparent on
sketch-maps No.6 and 7 annexed to the Court’s Judgment 378, the segments connecting points A
and B and B and C are, once again, straight lines 37.

4.76. In view of this unequivocal practice, the 1927 Arrêté can only be interpreted as

defining a course consisting of two successive straight lines connecting the three designated points
through which it passes.

4.77. Moreover, that is precisely the interpretation of the 1927 Arrêté reached by the Court
itself in its 1986 Judgment in the case between Burkina and Mali.

4.78. In that case, Mali claimed that Mount N’Gouma was to the east of the Kabia ford, and

not to the north as claimed by Burkina. The Court ruled out that interpretation of the 1927 Arrêté,
and, while not binding as such on Niger under the principle of res judicata, it nevertheless confirms

the interpretation of the Arrêté to be adopted in this sector.

4.79. To dispose of Mali’s argument, the Court held that the notion of turning used in the

1927 Arrêté did not describe the form of the inter-colonial boundary, but merely a change of
direction between two straight lines at a precise point.

4.80. It is indeed on the basis of this interpretation

 that the Court was able to compare the angle of the turn at the Tong-Tong astronomic marker,
which it deemed to be 155° and which it represented on an illustrative sketch-map showing the
380
inter-colonial boundary following two straight lines , with the angle (considered to be 90
degrees) of the change of direction the frontier would have undergone between the two sections

376Judgment of 3 February 2009, I.C.J. Reports 2009, pp. 111-112, para. 154.

377Ibid., p. 128, para. 206.
378
Ibid., pp. 114-115.
379
See also ibid., pp. 130-131, para. 218 and sketch-maps No. 8 and 9, pp. 132-133.
380Sketch-map No.6 on page 645 of the Court’s Reports only gives a very diagrammatic representation of the

boundary line between Burkina and Niger. The Court in fact draws a straight line between the Kabia ford and the
Tong-Tong marker where the boundary passes through other intermediate points and terminates at the Tong-Tong marker
by a straight line running first in a north-west south-east direction, and not north-south. Nevertheless, if a fictitious line is
drawn from the tripoint with Mali to the Tong-Tong marker, then a straight line from there to the Tao marker, the degree
of the general turn of the boundary from Tong-Tong roughly corresponds to that described by the Court in its
1986 Judgment. - 107 -

connecting Mount N’Gouma to the Kabia ford, and Kabia ford to Mount Darouskoy (if, as
Mali claimed, Mount N’Gouma was to the east of the Kabia ford);

 and that it consequently rejected Mali’s interpretation on the following grounds:

“if, as Mali suggests, one starts from the hypothesis that mount N’Gourna was to the
east of the Kabia ford, any boundary which started from mount N’Gouma, passed
through the ford, and then ran in the direction of mount Darouskoy (Arounskaye)
381 382
would turn sharply — through something like 90 degrees[ ] — at the ford[ ], since
mount Darouskoy lies south of the ford. The text of the Order of 31August1927
states that the boundary ‘then tums to the south-east’ in the neighbourhood of
383
Tong-Tong, a turn which is much less sharp (approximately 155 degrees)[ ] than the
supposed turn at the Kabia ford (see sketch-map No. 6 below). It is therefore difficult
to see how the draftsman of the Order could have failed to mention that the Kabia ford
384 385
was the position of such a marked turn[ ], if that had really been the case.”

4.81. At no point in this passage of its 1986 Judgment does the Court hypothesize that the
turn mentioned in the 1927 Arrêté at the Tong-Tong astronomic marker refers to anything other

than a turning point between one straight line and another. On the contrary, the sketch-map drawn
to illustrate its reasoning, and which is annexed at page 645 of its Judgment, clearly establishes
that, in the mind of the Court, the turn indicated in the 1927 Arrêté constitutes a change of direction

between one straight line and another. And rightly so, as this is the clearest and most obvious
meaning of the 1927 Arrêté.

4.82. For all the above reasons, between the Tong-Tong marker and the village of
Bossébangou, the frontier takes the following course:

 a straight line connecting the Tong-Tong astronomic marker (Lat.: 14° 24' 53.2" N; Long.:
0° 12' 51.7" E) to the Tao astronomic marker (Lat.: 14° 03' 04.7" N; Long.: 0° 22' 51.8" E)

(WGS 84 ellipsoid);

 a straight line connecting the latter point with the right bank of the River Sirba at the point
where it intersects with the straight line connecting the Tao astronomic marker and the village
of Bossébangou (Lat.: 13° 21' 06.5" N; Long.: 1° 17' 11.0" E) (Clarke 1880 ellipsoid).

381This clarification confirms that the turn designates a change of direction (here of 90 degrees) between two

straight lines at a precise point.
382The turn thus occurs between two lines at a precise location (it marks a point between two straight lines
according to the Court); it does not designate the general form of a line.

383Again, the turn designates here a change of direction at a precise point. That is why the Court can describe the
turn as “sharp” — which would not make sense if the turn referred to the form of the boundary line between two points
(if the turn describes only the form of the line, it is meaningless to state that it goes through a “sharp” turn; if the turn in a

line is sharp, it is necessarily in relation to another line).
384Here the Court again equates turning with turning point.

385I.C.J. Reports 1986, p. 644, para. 170. - 108 -

SECTION 2

T HE COURSE OF THE FRONTIER FROM THE POINT WHERE IT REACHES THE R IVER S IRBA
AT B OSSÉBANGOU AS FAR AS THE INTERSECTION OF THE SIRBA

WITH THE S AY PARALLEL

Sketch-map No. 9 - Location of the point on the River Sirba at Bossébangou

where the section of the frontier which starts at the Tao astronomic marke r

terminates, called point “P”

To Tao marker

 



 

133 - 109 -

4.83. From the end-point of the segment of the frontier which descends from the north-west

after passing through the Tao astronomic marker and “reach[es386he River Sirba at Bossébangou”, a
point whose location is described in the previous section and which for convenience sake will be
called “P” in the rest of this section (see sketch-map No.9— reproduced above and in
Cartographic Annex29), the frontier described by the Erratum of 5October1927 takes the

following course:

“It 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; then, turning back to the south, it again cuts the Sirba at the
level of the Say parallel.”

4.84. This description differs from that given in the original text of the Arrêté of
32 August 1927, which stated:

“To the north-west, by the River Sirba from its mouth as far as the village of
Bossébangou. From this point a salient, including on the left bank of the Sirba the
villages of Afassi, Kouro, Takalan and Tankouro;

To the south-west, a line starting approximately from the Sirba at the level of
the Say parallel and running as far as the Mekrou.”

4.85. It is clear from a comparison of the two texts that the Erratum is more precise than the
text it replaces. Thus, the Erratum:

 specifies that the salient of villages does not begin “from” Bossébangou, but that the course of

the boundary creates this salient “almost immediately” after it reaches the River Sirba at
Bossébangou; it also:

 defines the direction taken by this salient, and states that in order to form it the frontier runs
“towards the north-west”; moreover, the Erratum:

 corrects the names of two villages in the salient: it is not Afassi but Alfassi, and not Takalan

but Tokalan; it then

 specifies that after forming a salient leaving to Niger the four villages mentioned, the frontier

turns back to the south; finally, it

 states that after turning back to the south, the frontier again cuts the Sirba at the level of the Say
parallel, whereas the Arrêté of August 1927 mentioned an approximate point here.

4.86. The Erratum’s description of the frontier is thus far more precise than that of the
original text of the Arrêté of August 1927. It is in the light of this fact that consideration should be

given to the meaning to be ascribed to the terms of the Erratum (1), before describing the resulting
line (2), and then noting that this line corresponds in every respect to the consensual line of
1988 (3).

386
See above, paras. 4.17 to 4.24. - 110 -

1. The description of the frontier

4.87. According to the description of the frontier in the Erratum, after reaching the River
Sirba at Bossébangou (point P), the frontier:

 “almost immediately turns back up towards the north-west” (A);

 “leaving to Niger, on the left bank of that river, a salient which includes the

villages of Alfassi, Kouro, Tokalan, and Tankouro” (B);

 “then, turning back to the south, it again cuts the Sirba at the level of the Say

parallel” (C).

The meaning of each of these provisions requires clarification.

A. “almost immediately turns back up towards the north-west”

4.88. This description comprises three closely linked elements: from point P, the frontier
“turns back up”; but it does not turn back immediately: it does so “almost immediately”; and, it
turns back in an approximate direction, which is “towards the north-west”.

4.89. The first question to be determined in order to fully understand the meaning of this
description concerns the course of the frontier between point P and the point from which the

frontier “turns back up” (a). The second question to ask is: where is the point from which the
frontier “turns back up” after following that course in order to form the beginning of the
salient (b)? The third question concerns the meaning of the words “towards the north-west” (c).

(a) The course of the frontier between point P and the point where it “turns back up” towards
the north-west

4.90. It is perfectly clear that the frontier includes a section between point P and the point
from which it begins to “turn back up”. If that were not the case, i.e., if it were to begin turning
back up immediately after it reached point P, in the terms of the Erratum the frontier should then

“turn back on itself”: thus, given that it runs in a south-easterly direction before reaching this
point, if it were to turn back up towards the north-west at this point, it would almost inevitably have
to retrace its steps and in fact cancel out the last part of its course and never reach the River Sirba at

Bossébangou.

4.91. This error was reproduced in the 1:1,000,000 sketch-map of the “New frontier between

Upper Volta and Niger (according to the Erratum of 5October1927 to the Arrêté of
31 August 1927)” 387, and the 1927 edition of the 1:1,000,000 road map of the colony of Upper
Volta 388. On this sketch-map and map, the first of which clearly states its intention to reproduce

the course of the boundary as it results from the Arrêté of 31August1927 and its Erratum, the
frontier is clearly seen to descend from the Tao astronomic marker in a straight line toward
Bossébangou, but never reaches the River Sirba and instead turns south-south-west some fifteen

kilometres to the north-west of the village of Bossébangou, and forms a bulge which starts to the
north-east of Alfassi.

387
Cartographic Annex MBF 6.
38Cartographic Annex MBF 5. - 111 -

4.92. These representations of the boundary line, allegedly based on the description in the
Erratum, are completely erroneous, as they deviate repeatedly from the description of the frontier in
this sector given in the Erratum. Thus, according to the lines seen on this sketch-map and map:

 the frontier never reaches the River Sirba at Bossébangou, contrary to what is stated in the text
of the Erratum;

 at no point does it “turn back up” in a north-westerly direction, whilst the text clearly states that
it does so;

 it only cuts the River Sirba once, at the level of the Say parallel, whilst the text states that at
this location the frontier “again” cuts the River Sirba, which necessarily means that it has

already cut it once before;

 it does not turn back to the south to cut the River Sirba at the level of the Say parallel, contrary

to what is stated in the text, but reaches this point after running in a south-westerly direction.

4.93. Notwithstanding the gross misinterpretations of the text of the Arrêté and its Erratum
contained in the 1:1,000,000 sketch-map of the “New frontier between Upper Volta and Niger
(according to the Erratum of 5October1927 to the Arrêté of 31August1927)” 389, the Niger

experts in the Joint Technical Commission on Demarcation attempted to use it in support of one of
their arguments to reject the consensual line of 1988. At the extraordinary meeting of the
Commission on 14 May 1990 390, they thus maintained that this sketch-map should be considered to

be more reliable than the Erratum, because, in their view:

“notwithstanding the precision of the Erratum in some places, at the level of

Bossébangou it contains descriptions of the internal boundaries of Tillabéry and Say
cercles. This situation is reflected on the map in question which differentiates
between the frontiers between the two colonies and the boundaries of the cercles.” 391

4.94. This argument held, on the one hand, that the Erratum was quite simply wrong in that it
confused the boundaries of the cercles with the boundaries of the colonies, and on the other hand,

that the 1927 sketch-map restored the political reality by distinguishing between the boundaries of
the cercles (in particular at the level of Bossébangou) and the inter-colonial boundary (at the level
of the salient), and was therefore more reliable than the Erratum. In the view of the Niger experts

at the time, the 1927 sketch-map was thus not to be considered as plotting the course described by
the Erratum (since it is too blatantly inconsistent with it), but rather as correcting the alleged errors.

392
4.95. Such a position was of course untenable and, moreover, Niger soon abandoned it :

 firstly, while the Arrêté of 31 August 1927 did effectively confuse the boundaries of the cercles
with those of the colonies, that was precisely not the case with the Erratum, which was adopted
with a view to describing only the inter-colonial boundary— without any reference to the

cercle boundaries— and further clarifying its course; moreover, the text of the Erratum is
unambiguous in this regard, since it begins with the following statement: “The boundaries of

38Cartographic Annex MBF 6.
390
See above, para. 1.75.
39Report of the extraordinary meeting of the Joint Technical Commission on Demarcation, 15 May 1990,

Ann. MBF 85.
39See above, para. 1.77. - 112 -

the Colonies of Niger and Upper Volta are determined as follows...”; it is thus not the
boundaries of the cercles that it describes;

 secondly, if this argument were to be accepted, the 1927 sketch-map would be deemed to
replace the 1927 Arrêté and its Erratum; the sketch-map would then be recognized as a

territorial “title”, namely “a document endowed by inter393ional law with intrinsic legal force
for the purpose of establishing territorial rights” ; yet legally this is impossible, since it is
established that maps “cannot constitute” such a title, unless they “fall into the category of
394
physical expressions of the will of the State or States concerned” ; this is not the case with
the 1927 sketch-map which, moreover, “offers no information as to which official body
compiled it or which administrative authority approved the line shown on it” 395;

 finally, the Protocol of Agreement of 28 March 1987, which binds the Parties, establishes that

the frontier is “as396scribed in the Arrêté of 31August1927, as clarified by the Erratum of
5 October 1927” , and not as described by the 1927 sketch-map.

4.96. Thus the frontier that should be retained is that described in the 1927 Arrêté and its
Erratum, pursuant to Article 1 of the Protocol of Agreement of 28 March 1987. In accordance with

that description, after reaching point P, the frontier does not retrace its steps, but “almost
immediately” turns back up towards the north-west. That means — and it is key to understanding
the course of the frontier — that it does not turn back up “immediately”. Thus there is necessarily

a section of the frontier which does not “turn back up”, but runs between point P and the point
where it “turns back up”, which for the sake of convenience shall be called point P1.

4.97. This statement of the obvious highlights another error in another argument put forward
by the Niger experts during their discussions on the frontier, this time in July 1990. In the report of

a meeting of the Joint Technical Commission on Demarcation held from 23 to 28July1990,
Niger’s position, which was different to the one it held in May of the same year 397, was recorded as

follows:

“Although the Erratum specifies that the frontier line reaches the River Sirba at

Bossébangou following a south-easterly direction, and then turns back in the opposite
direction (north-west), which would eliminate the frontier line over that short stretch,
Niger accepts that the line of the frontier reaches the River Sirba at Bossébangou and

then turns back on itself at an angle of some kind, but only on the condition that it
does not cut the loop of the River Sirb a. The line must return to a point at which,
turning south to cut the River Sirba again at the level of the Say parallel, it creates a
398
salient . . .”

4.98. This time, and contrary to the position it held in May 1990, Niger accepted that the
frontier had to reach the River Sirba at Bossébangou. However, it also maintained that it should

393Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986, p. 582, para. 54.
394
Ibid.
395
Frontier Dispute (Burkina Faso/Republic of Mali), Judgment, I.C.J. Reports 1986 , p. 583, para. 57; the Court
refers to this document as a “map” and not a sketch-map.
396Ann. MBF 73; on the Protocol of Agreement of 1987, see above, paras. 1.63 to 1.65.

397See above, para. 1.75.

398Report of the meeting of the Joint Technical Commission on Demarcation of the Frontier, held from 23 to
28 July 1990, 28 July 1990, Ann. MBF 87. - 113 -

turn back immediately at this point, not retracing its steps but at an “angle of some kind”, on

condition that it did not cut across the loops of the River Sirba.

4.99. This position was as untenable as the previous one, based as it was on a series of
equally blatant errors. First, as Burkina has already pointed out, the frontier does not turn back
“immediately” as Niger maintained, but “almost immediately”, which necessarily means that from

point P, over a certain distance, it follows a course that does not “turn back up”; secondly, when
the frontier does turn back up, it is not at an angle “of some kind”, but “towards the north-west”;
lastly, the condition whereby the frontier should not cut across the loops of the River Sirba is based

on speculation for which there is no basis in the text of the 1927 Arrêté or its Erratum.

4.100. While it is clear that the frontier described by the text does not turn back up

“immediately”, the description in the 1927 Arrêté and its Erratum does not explicitly state which
course should be followed between point P and point P1 where it begins to “turn back up” towards
the north-west. However, the frontier cannot follow any course here other than that followed by

the right bank of the River Sirba, from Bossébangou to its source.

4.101. Thus, while the text states that the frontier, coming from the north-west, “reaches the
River Sirba” at Bossébangou, there is not the least indication that it “leaves” the River Sirba before
it starts to “turn back up”. This must mean that the frontier follows the course of the river until it
starts to “turn back up” 399. The frontier thus also logically follows the right bank of the River
400
Sirba. In this connection, Burkina has already pointed out above that the point where the frontier
reaches the River Sirba at Bossébangou, point P, is situated on the right bank of the River Sirba.
Consequently, as there is no indication that the frontier turns back on itself to reach the median line

of the river, it must be understood that it necessarily stays on and follows the right bank up to the
point where it again crosses the river to start “turning back up” towards the north-west.

4.102. This interpretation is confirmed twice in the text.

4.103. First, it is explicitly stated therein that the frontier, after reaching the point from which
it begins to “turn back up”, produces a salient that includes part of the left bank of the River Sirba
in Niger. This information suggests a contrario that before the point from which it “turns back

up”, the frontier does not leave the left bank of the Sirba to Niger.

4.104. Secondly, since the frontier is described as “turning back up” “towards the
north-west” at the end of the section between point P and point P1 (the point at which it begins to
turn back up), it necessarily follows that this section, by contrast, runs generally in a “downward”

399
Texts which describe frontiers frequently use the verb “reach” to mean that the frontier joins a pre-existing line
that it is then to follow. Thus, among other examples, in the Treaty of Vienna of 9June1915 Article 66 on the
boundaries of the Kingdom of the Netherlands states that the boundary “reaches the limits between the old departments of
the Ourthe and the Roer; [and] then goes along these limits . . .” The arrangement of 12 July 1893 between Great Britain
and France fixing the boundary between British and French possessions on the Gold Coast established the demarcation
line between British and French possessions on the Gold Coast by stating, in Article 4, that the line “reaches the Volta at
the spot where that river is intersected by the road frBandagadi to Kirhindi. Thence it follows the thalweg of the
Volta to its intersection by the 9th degree of north latitude” (British and Foreign State Papers, compiled by The Librarian
and Keeper of the Papers, Foreign Office, vol. 85, London, 1899, H.M. Stationery Office, p. 31). Article 9 of the London

Agreement of 11March1913 between Great Britain and Germany respecting the Settlement of the Frontier between
Nigeria and the Cameroons, from Yola to the Sea, provides that the frontier “cross[es] the watershed to the source of the
River Maquari (Mekwer), [and] follows the thalweg of that river”.
400See above, para. 4.21. - 114 -

direction (towards the south) or at least in a “neutral” direction (to the west), since, if it did not, the

frontier could not “turn back up” (to the north) when it reaches the end of this section, since it
would already be “turning back up”. This downward or horizontal direction corresponds exactly to
the general direction of the River Sirba as it was known to the cartographers in 1927, at least over
401
the portion which follows its course upstream from Bossébangou .

4.105. It should finally be noted that this was also the logica402nterpretation formalized by the
Parties when they adopted the “consensual line” of 1988 . The line on the 1960 IGN
1:200,000map, to be consulted should the description given by the Arrêté and its Erratum not
403
suffice , also shows a line that is consistent with this interpretation.

(b) Point P1 from which the frontier begins to turn back up towards the north-west

4.106. The second question concerns the location of point P1, from where the frontier, after
following the course of the River Sirba upstream along its right bank, “turns back up” and begins to

form the salient.

4.107. As it results from the Erratum, this point can be determined by a combination of the

following three elements:

 the result of the frontier “turning back up” is to leave four villages to Niger: Alfassi, Kouro,

Tokalan, Tankouro;

 this section of the frontier starting from point P1 on the River Sirba should run “towards the

north-west”; and

 the frontier line turns back up “almost immediately” after it reaches the “Sirba at

Bossébangou”, namely at point P.

4.108. It is immediately apparent that these three indications are consistent and
complementary. The first indicates the reason why the frontier must “turn back up”: so that the
four villages are in Niger; the second indicates how the line can comply with the result sought (to

leave the four villages to Niger). Thus, for it to say that the frontier line runs “towards the
north-west” from the point where it changes direction on the River Sirba, means that it must be
sufficiently directed towards the north-west to leave the villages to Niger, given the location of the
point on the right back of the River Sirba from which the frontier turns back. It is clear that the

frontier will have to turn back even further to the north as it approaches Alfassi. The third
indication also adopts this manner of approximate description, when it uses the words “almost
immediately”: this is not a sign of chronic indecision, but, on the contrary, ensures that the frontier

turns back up soon enough, but not too soon, to enable the villages to be included in the territory of
Niger, by directing the beginning of the salient towards the north-west.

401
Map No. 60, Say cercle, from the Atlas des cercles of FWA, fascicle IV, Upper Volta, represents the course of
the River Sirba from Bossébangou as running in a south-westerly and thus “downward” direction; Cartographic
Annex MBF 4.
402
The map representing the consensual line is reproduced in Cartographic Annex MBF 15.
40The 1960 IGN 1:200,000 map which adopts this line is reproduced in Cartographic Annex MBF 11. - 115 -

(c) “leaving to Niger, on the left bank of that river, a salient which includes the villages of

Alfassi, Kouro, Tokalan, and Tankouro”

4.109. The text of the Arrêté of 31 August 1927 stated that the salient includes four villages,

but did not give any indication as to the shape the sa lient should take. It was thus very vague: by
simply stating the result expected of the frontier line without giving any information about the said
line, it left many options open.

4.110. The text of the Erratum takes a very different approach, and indicates the direction of
the line, not only in respect of the beginning of the salient (here the frontier runs “towards the

north-west”), but also in respect of the section which forms the west side of the salient, as, after
running towards the north-west, the frontier line turns back to the south, following a course from
north to south and passing through a precise point, which is the intersection of the River Sirba with
the Say parallel.

4.111. Accordingly, the Niger salient is clearly bounded, to the west, by the line running

north from the intersection of the River Sirba with the Say parallel and, to the north-east, by the
line turning back up from the River Sirba to the north-west.

(d) “then, turning back to the south, it again cuts the Sirba at the level of the Say parallel”

4.112. While the text of the Arrêté of 31 August 1927 gave no indication as to the course of

the frontier between what constitutes the salient and the starting point of the following section,
which was described as “starting approximately from the Sirba at the level of the Say parallel”, the
Erratum provides two key clarifications:

 after turning “towards the north-west” from point P1 on the River Sirba, leaving the four
villages on the left bank of the Sirba on the Niger side, the line “turns back” to the south;

 turning back to the south, it again cuts the River Sirba at the level of the Say parallel.

4.113. These clarifications are invaluable, as they give two defining geographical

descriptions.

4.114. Firstly, the point of arrival of the section of the frontier described here by the Erratum,
which is fixed at the intersection of the Say parallel with the River Sirba, cannot be considered as
being fixed “approximately”, as indicated in the 31 August 1927 version of the Arrêté: in contrast
to the latter, the Erratum on the contrary describes that point as being precisely where the line

arriving from the north “cuts the Sirba at the level of the Say parallel”; it is certainly possible to
locate this geographical point, since the location of Say is known 404.

4.115. Niger of course claimed the opposite inJuly1990 and maintained that “the
expression, ‘at the level of the Say parallel’ was merely indicative” 405. Their argument was that

according to the indications given for the salient, the frontier had to deviate to the west to include

404
The co-ordinates of Say are: Lat. 13° 06' 11'' N; Long. 2° 21' 07'' E (1880 Clarke). The Say parallel thus runs
along latitude 13° 06' 11'' N.
40Report of the meeting of the Joint Technical Commission on Demarcation of the Frontier, held from 23 to
28 July 1990, 28 July 1990, Ann. MBF 87. - 116 -

the village of Takatami, which Niger considered to be Takalan, so that it was then impossible for

the frontier, “turning back to the south”, to cross the River Sirba at the level of the Say parallel.
This southward-running line, descending from a point too far to the west, would in fact never cross
the River Sirba. More specifically, according to Niger:

“The field survey mission has already identified the sites of Alfassi, Kouro [,
Takalan] and Tankouro (see the Report and theJune1990 Mission Report). In view
of a comparison between the old and new cartographic documents, and on the basis of

all the physical details noted, the Takalan site corresponds to the exact location of the
current village of Takatami. The frontier line, in creating the salient, cannot cut the
Sirba at its exact intersection with the Say parallel and at the same time encompass the

four villages. This s406s that the expression, ‘at the level of the Say parallel’ was
merely indicative.”

4.116. This claim is erroneous on at least three counts.

4.117. First of all, it is clearly untrue that the reference in the Erratum to the geographical
point corresponding to the intersection of the River Sirba with the Say parallel was intended to be
“indicative”; if that had been the author’s intention, he would have used the far less precise phrase
in theAugust Arrêté, which does indeed mention a point “approximately” where the Sirba meets

the Say parallel. That was precisely not the case: the phrase used in the Erratum is anything but
vague, and on the contrary is intended to be indisputable.

4.118. Moreover, Niger’s argument is not in the least convincing when it further maintains,
on the basis of speculation about the village of Tokalan, of which there is no trace today, that the

Erratum contains an error which should be corrected by completely ignoring one of the clearest
reference points that it gives, namely the point corresponding to the intersection of the Sirba River
with the Say parallel. In interpreting the text of the Erratum, there is no room for speculation about
a clearly identifiable and undisputed toponym: in the event of doubt as to the meaning of one

toponym amongst others, it should be presumed that if the other— clear and undisputed—
indications are followed and that those indications enable the frontier to be drawn effectively, any
indications that cannot be verified should also be deemed correct. This is, moreover, the approach

taken by the Chamber of the Court in the case of the Land, Island and Maritime Frontier Dispute
between El Salvador and Honduras:

“The Chamber considers that it is impossible to reconcile all the landmarks,
distances and directions given in the various 18th century surveys in this region: the
most that can be achieved is a line which harmonizes with such features as are
identifiable with a high degree of probability, corresponds more or less to the recorded
407
distances, and does not leave any major discrepancy unexplained.”

4.119. Finally, the claim that the site of the lost village of Tokalan corresponds to the
present-day village of Takatami is completely unfounded. The topographic sub-committee’s report
on the mission from 5 to 12June1990 to identify the sites of Alfassi, Kouri, Takalan (Tokalan),
Tankouro and Kogori, on which Niger’s argument was based, states only that “[t]he individuals

questioned did not know of a village by that name, but were familiar with a village called Takatami

406
Report of the second ordinary session Joint Technical Commission on Demarcation, 28July1990,
Ann. MBF 87.
40Judgment of 11 September 1992, I.C. J. Reports 1992, p. 456, para. 155. - 117 -

beside the River Faga” 408. In no way does this demonstrate — rather it disproves — that Tokalan

and Takatami are one and the same village. Moreover, eleven years later, the report of the fourth
ordinary session of the Joint Technical Commission on Demarcation, meeting in Ouagadougou
from 18 to 20 July 2001, noted “the failure to identify the villages referred to in the Erratum” and
409
recommended conducting a further field survey mission . Above all, it is simply not possible to
equate the village of Takatami with the village of Tokalan, as, for one thing, Takatami already
existed in 1927 (it could therefore not be identified as being the village of Tokalan at the same

time), and, secondly, this village was obviously considered to be part of the colony of Upper Volta,
since it appears on the list of villages belonging to that colony, reproduced in the General List of
Localities of French West Africa 41. Thus Takatami has clearly never been considered to be in

Niger.

4.120. The second provision, whereby the frontier “turns back to the south” and crosses the
River Sirba at the level of the Say parallel, describes with certainty the course of the frontier from
the end of the section which “turns back up” towards the north-west after following the right bank
of the River Sirba. From this endpoint, which is logically to the north-west of Alfassi, the frontier

follows the line that runs from north to south and passes through the point of intersection of the
River Sirba and the Say parallel.

4.121. It is important to note that the provision whereby the frontier “turns back to the south”
is very precise in two respects.

4.122. Firstly, it means that the direction taken by the frontier at this spot is defined
precisely. Otherwise, the text would have stated that the frontier turns back “towards” the south

and not “to” the south. This is clearly the alternative used elsewhere in the text, for example where
it states that after the Tong-Tong astronomic marker the line turns “towards the south-east”, or
again where it states that after reaching the River Sirba at Bossébangou the frontier turns back up

“towards the north-west”. That is not the language used here, where the fact that the word
“towards” is not used means, in contrast, that the text intends to indicate a precisely defined
southerly direction, and not an approximate one.

4.123. It also means that over this stretch of the frontier the line runs due south, and not in

any other direction between south-west and south-east, such as south-south-east or
south-south-west. It is due south and no other direction. Otherwise, if the direction were not
exactly to the south, the text of the Erratum would have made that clear, as it does for the following
portion of the frontier, for example, when it states that it follows “an east-south-east direction”.

4.124. This dual precision is very useful, in that it enables a shape to be given to the salient,

so that its western part can be marked out, and a line to be established which is perfectly consistent
with the description given by the Erratum: while the four villages could not all be located because
some had disappeared, it is certain that they must all, by definition, lie to the east of the north-south

line which passes through the intersection of the Say parallel and the River Sirba.

408
Report of the mission of the Topographical Sub-Commission from 5 to 12 June 1990, Ann. MBF 86.
409
Report of the fourth ordinary session of the Joint Technical Commission on Demarcation, 18 to 20 July 2001,
Ann. MBF 94.
41General List of Localities of French West Africa, Upper Volta, fascicle IV, 1927, Ann. MBF 27, p. 54. - 118 -

2. The course of the frontier

4.125. The course of the frontier between point P and the intersection of the Say parallel with
the River Sirba can be clarified in the light of the description of the frontier just presented. It is

apparent that this line is subdivided into three sections:

 the first section follows the course of the River Sirba upstream along its right bank from point

P, which corresponds to the River Sirba at Bossébangou, as far as the point from which the
second section begins, point “P1”;

 the second section departs from point P1 on the right bank of the River Sirba, and turns back up
towards the north-west, in such a way as to leave the four villages on the Niger side, as far as a
point which will be called “P2”;

 a third section departs from point P2 and descends in a straight line to the south until it meets
the intersection of the River Sirba with the Say parallel, a point which will be called “P3” 411.

4.126. The first section is extremely simple to draw, as it follows the course of the right bank
of the River Sirba, from point P (Lat.: 13° 21' 06.5" N; Long. 1° 17' 11.0" E, Clarke 1880), which

corresponds to the “Sirba at Bossébangou”, as far as point P1, whose co-ordinates will be justified
below. Sketch-map No.10, reproduced below and at Cartographic Annex30, shows this first
section, though for the time being it does not show point P1.

411In the remaining description of the line, it will thus be considered that:

 the point where the River Sirba reaches Bossébangou is point P;

 the point from which the frontier “turns back up” towards the north-west is point P1;

 the endpoint of the section of the frontier which “turns back up” towards the north-west is point P2;
and
 the point of intersection between the River Sirba and the Say parallel is point P3. - 119 -

Sketch-map No. 10 - Section of the frontier following the course of the Rive r

Sirba upstream along its right bank from point P as far as the point from which
the frontier turns back up towards the north-west, leaving the four villages on
the Niger side, called point P1

To Tao marker

148 - 120 -

4.127. The second section, from point P1 to point P2, can only be clarified by referring to the
1960 IGN map, as it is not possible to establish one specific point P1 based on the description of
the frontier in the Arrêté and its Erratum. It is in fact perfectly clear that there are a number of

potential point P1s on the right bank of the River Sirba which fit the description of the frontier in
the Erratum. Point P1 needs only to be placed in such way that the section that turns back up
towards the north-west from that point passes to the north of Alfassi, leaving that village to Niger.
In this respect, it is important to note that the frontier is not, in any event, meant to go any further

north than necessary to leave Alfassi to the south, as the three villages cited after Alfassi in the
Erratum are without question to the south of Alfassi. This is particularly apparent on a map from
1926 which shows (very approximately) three of the four villages 41, and there is not a shred of

evidence in the records that might lead to believe otherwise. In any event, none of the proposed
frontier lines in this area suggests that there is a village to the north of Alfassi in the salient, neither
the — erroneous — line shown on the 1:1,000,000-scale sketch-map of the “New frontier between

Upper Volta and Nige413according to the Erratum of 5October1927 to the Arrêté of
31 August 1927)” , nor that on 1960 IGN map; and no different solution can be found in the
consensual line of 1988 414.

4.128. Not only can several points be retained for point P1, but the section which starts from
that point could take several directions and several forms, given that, in order for this section to be

consistent with the description in the Erratum, it suffices that it both passes to the north of Alfassi
and runs generally “towards the north-west”.

4.129. The variety of options offered here by the description in the Erratum does not
preclude the possibility of determining with certainty both point P1 and the course of the section
between points P1 and P2. Indeed, it was with a view to resolving such situations that the Parties

undertook to apply the method agreed in Article 2 of the 1987 Protocol of Agreement, which
provides:

“Shothueld 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.”

4.130. Since it is clear that the Arrêté and its Erratum do not suffice here to precisely

determine the course of the boundary for the purposes of marking it out, the line shown on the 1960
IGN map should be referred to.

4.131. According to that map, point P1 on the right bank of the River Sirba, as shown on
sketch-map No.11 below and reproduced in Cartographic Annex31, has the following
co-ordinates:

Lat.: 13° 19' 53.5" N; Long. 1° 07' 20.4" E (Clarke 1880 ellipsoid).

41On map No. 60, Say cercle, from the Atlas des cercles of FWA, fascicule IV, Upper Volta, Kouro and Takalan
are located well to the south of Alfassi; see Cartographic Annex MBF 4.
413
Cartographic Annex MBF 6.
41See Cartographic Annexes MBF 15 and 36. - 122 -

4.132. From this point P1, the frontier follows the course shown on the IGN France

1:200,000-scale map, 1960 edition; it heads in a north-north-west direction, then runs towards the
west over a short distance, before turning back up again to the north-west over a short distance;
then, returning to the west, it meets the north-south line that passes through the intersection of the
River Sirba with the Say parallel. Sketch-map No.12 reproduced below and in Cartographic

Annex 32, illustrates this line. - 123 -

Sketch-map No. 12 - Stretch of the frontier showing the section turning back up towards
the north-west from point P1, leaving the four villages on the Niger side, as far as the

point where it descends in a straight line to the south

To Tao marker

151 - 124 -

4.133. The third section consists of a north-south line connecting the end of the previous

section and point P3, which corresponds to the intersection of the River Sirba with the Say parallel.

4.134. The position of this point P3 will obviously depend on whether it is fixed at the
intersection of the Say parallel with the River Sirba’s right bank, its left bank, or its median line.

The text of the Erratum nevertheless provides a crucial piece of information which fixes this point
at the intersection of the Say parallel with the right bank of the Sirba. Thus the Erratum states that,
turning back to the south, the frontier “again cuts the Sirba at the level of the Say parallel”. The

frontier “cuts” the River Sirba, in other words it crosses it, going from the left bank— given its
general west-east direction, and ending on the right bank, at the spot where it meets the Say
parallel.

4.135. Point P3 thus defined has the following co-ordinates:

Lat.: 13° 06' 10.75" N; Long. 0° 59' 40.0" E (Clarke 1880 ellipsoid).

4.136. Point P2 can then easily be determined since it is situated at the intersection of the
section that departs from point P1 and turns back up towards the north-west, following the course
plotted on the 1960 IGN France 1:200,000-scale map, with the line that departs from point P3 and

turns back due north, as illustrated in sketch-map No.13 below, also reproduced in Cartographic
Annex 33. - 125 -

Sketch-map No. 13 - Line from the intersection of the River Sirba with the Say

parallel, called point P3, turning back due north

To Tao marker

152 - 126 -

4.137. The co-ordinates of point P2 are:

Lat.: 13° 22' 30.05" N; Long. 0° 59' 40.0" E (Clarke 1880 ellipsoid).

The course of the frontier described in the Arrêté and its Erratum between points P and P3 is
thus as shown on sketch-map No.14 below (also reproduced in Cartographic Annex34). Here,

this line is exactly the same as the consensual line of 1988. - 127 -

Sketch-map No. 14 - Course of the frontier from the point where it reaches the Rive r
Sirba at Bossébangou as far as the intersection of the River Sirba with the Say parallel

To Tao marker

Line according to Burkina Faso

Consensual line of 1988

153 - 128 -

3. The line described in the Erratum and the consensual
line of 1988 are identical

4.138. It should be recalled that the consen sual line was mutually agreed in 1988 by the
technical sub-committee of the Joint Technical Commission on Demarcation of the Frontier, which

consisted of highly qualified field experts led by Mr.Waziri, national Director of Topography of
Niger, and Mr. Lohouara, Director of Territorial Cartography of Burkina, who was acting as a
delegate of the Director General of the Geographical Institute of Burkina. This team conducted a

survey of the Niger-Burkina frontier in 1988, which it then recorded in a report on completion of
the survey of the proposed demarcation of the Niger-Burkina frontier, dated 28 September 1988 41.

4.139. This work was considered to be highly conclusive:

“Apart from that single point (Tokébangou), the technical experts are
unanimous in respect of the map interpretation and the field survey of the frontier line
as defined in the basic documents cited in the Agreement signed at Ouagadougou on
416
29 March 1987.”

4.140. The experts were thus “unanimous” in considering that they had accurately located
the various geographical reference points which were mentioned in the Arrêté of 31 August1927,
as clarified by its Erratum, and were needed to determine the course of the frontier, apart from one:

the Tokébangou point. The fact that they were in complete agreement was confirmed by the Joint
Technical Commission on Demarcation of the Frontier, whose fourth meeting noted 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 28 March 1987” 417.

4.141. On the basis of this agreement, the experts were able to plot on a base map the line
described by the texts and surveyed on the ground. They also indicated the line of the 1960

1:200,000 map on the same base map, so that they could compare “the line on the IGNF map and
the line resulting from the field survey in accordance with the basic texts”, and concluded that:

“The section from Bossébangou to the frontier point on the River Sirba at the
level of the Say parallel and the section situated to the west of Tchenguiliba as far as
the frontier with Benin are identical to the line on the IGNF 1:200,000 map, 1960
418
edition.”

44192. The experts in fact considered— quite rightly and as does Burkina as explained
above — that, given the uncertainty over the precise direction taken by the beginning of the
salient at this particular location (between points P1 and P2), the Arrêté as clarified by its Erratum

did not “suffice” to determine the course of the frontier. They thus adopted the solution plotted on

41Ann. MBF 82.
416
Ibid.
41Report of the fourth meeting of the Joint Technical Commission on Demarcation of the Niger-Burkina

Frontier, 26-28 September 1988, Ann. MBF 81.
41Report on completion of the survey of the proposed demarcation of the Niger-Burkina frontier,
28 September 1988, Ann. MBF 82.

41Paras. 4.127-4.128. - 129 -

the 1960 IGN Paris map, in accordance with the method they were bound to apply under Article 2

of the1987 Protocol of Agreement. The Burkina Minister for Territorial Administration further
states in his report to the Head of State of 5 March 1991: “The Burkina delegation considers that
the Commission, faced with a lack of clarity which renders the text insufficient, was right to adopt
420
the line on the map in accordance with the Protocol of Agreement” .

4.143. Burkina remains convinced that the 1988 line, which is identical in this location to the
line described by Burkina in the present Memorial, is the line which the Parties are bound to adopt
under the 1987 Agreement and Protocol of Agreement.

4.144. Thus in this sector, the frontier line that Burkina requests the Court to determine is as
follows:

 from point P (Lat.: 13° 21' 06.5" N; Long. 1° 17' 11.0" E; Clarke 1880), the frontier follows
the right bank of the River Sirba, from east to west, as far as point P1 (Lat.13°19'53.5'' N;

Long. 1° 07' 20.4'' E; Clarke 1880), located on its right bank;

 from that point, the frontier follows the line on the 1:200,000-scale map of the Institut

géographique national de France , 1960 edition, as far as point P2, whose co-ordinates are:
(Lat. 13° 22' 30.0'' N; Long. 0° 59' 40.0'' E; Clarke 1880);

 from that point, the frontier runs in a straight line south, ending at point P3 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; Clarke 1880).

S ECTION 3
F ROM THE INTERSECTION OF THE S IRBA WITH THE SAY PARALLEL TO

THE WEST OF THE VILLAGE OF T CHENGUILIBA

4.145. The starting point of the last stretch of the frontier, which has not been marked out

and which the Court has been requested to determine, is the endpoint of the previous segment,
point P3. There is little doubt as to its location, as demonstrated in the preceding section: it is at
the intersection of the right bank of the River Sirba with the Say parallel, which is located at a point

with the co-ordinates: Lat. 13° 06' 10.7'' N; Long. 0° 59' 40.0'' E (Clarke 1880).

4.146. From that point, the Erratum describes the frontier thus: “From that point 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.”

The starting point is known, and the arrival point as well. The Parties in fact agree that the
point located 1,200 m to the west of the village of Tchenguiliba, which marks the beginning of the
“Botou bend”, has the following co-ordinates: Lat. 12° 36' 18'' N; Long. 1° 52' 07'' E 421, on the
422
Clarke 1880 ellipsoid; and Lat. 12° 36' 19.2'' N; Long. 1° 52' 06.9'' E , on the WGS 84 ellipsoid.

420
Ann. MBF 88.
421
These are the co-ordinates given by the Parties in the Special Agreement.
42These are the GPS co-ordinates noted in the record of the work of the Joint Survey Mission of 3July2009,
Ann. MBF 101; these co-ordinates will be used in the rest of this chapter. - 130 -

4.147. According to the Erratum, from the intersection of the River Sirba with the Say
parallel (point P3), the frontier follows “an east-south-east direction, [and] continues in a straight
line” up to the point whose geographical co-ordinates are given in the preceding paragraph.

4.148. There are two further pieces of information to be gained.

4.149. First of all, the line “follow[s] an east-south-east direction”. It might seem

superfluous to specify this, given that the starting point and arrival point are provided and that the
frontier runs in a straight line. However, State practice shows that it is extremely common to
423 424
describe frontiers in such a manner . The Court also has recourse to the same technique .

4.150. Secondly, it is stated that the line “continues in a straight line”; this is crystal clear

and does not require any particular comment.

4.151. Furthermore, this reading of the text was not a subject of debate among the experts of
the Technical Sub-Committee inSeptember1988 425. Plotting this line on a map— as does the

“consensual line” of 1988— or on the ground is straightfoward. The frontier follows a perfectly
straight line between the intersection of the right bank of the River Sirba with the Say parallel,
namely point P3, and the point corresponding to the beginning of the Botou bend.

4.152. In this sector, the frontier line which Burkina requests the Court to determine is thus

as follows:

 from 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; Clarke 1880), the frontier runs in a straight line as

423
For example, Article 2 of the Agreement between the Kingdom of Saudi Arabia and the United Arab Emirates
on the delimitation of boundaries, 21 August 1974, provides: “The land boundary between the Kingdom of Saudi Arabia
and the United Arab Emirates starts from point (a) on the coast of the Arabian Gulf at the approximate geographical
location of: Parallel 24 14' 58" north; Meridian 51 35' 26"east.

It extends from this point in a straight line proceeding in a southerly direction to point (b) at the geographical

location of: Parallel 24 07' 24" north; Meridian 51 35' 26" east.

It extends from this point in a straight line proceeding in a south-easterly direction to point (c) at the geographical
location of: Parallel 26 56' 09" north; Meridian 52 34' 52" east.

It extends from this point in a straight line proceeding east by south to point (d) at the geographical location of:
Parallel 22 37' 41" north; Meridian 55 08' 14" east” (U.N.T.S., 1993, p. 37). The same procedure is used in Article 2 of
the Agreement on the delimitation of the offshore and land boundaries between the Kingdom of Saudi Arabia and Qatar
(U.N.T.S., 1993, pp. 21-22).

424
Land, Island and Maritime Frontier Dispute (ElSalvador/Honduras: Nicaragua intervening), Judgment,
I.C.J. Reports 1992, p. 611, para. 425: “from there in a straight line in a south-easterly direction to the boundary marker
of the Cerro Piedra Menuda...”, p.611, para.426: “From the Pefia de Cayaguanca (point A on Map No.II annexed;
co-ordinates: 14" 2 l' 54" N, 89" 10' l 1 " W), the boundary runs in a straight line somewhat south of east to the Loma de
Los Encinos (point B on Map No.II annexed; co-ordinates: 14" 21'08" N, 89"08'54" W)”; or p.614, para.429: “and
thence in a straight line somewhat north of east to a hill some 1,100 metres high (point D on Map No.V annexed;
co-ordinates: 13" 55' 03" N, 87" 49' 50" W)”.

42Report of the fourth meeting of the Joint Technical Commission on Demarcation of the Niger-Burkina
Frontier, 26-28 September 1988, Ann. MBF 81. - 131 -

far as the start of the Botou bend (Tyenkilibi) (Lat. 12° 36' 19.2'' N; Long.1°52'06.9''E;
WGS 84 ellipsoid).

4.153. For all the reasons set out in this chapter, from the Tong-Tong astronomic marker the
frontier line follows, in succession:

 a straight line as far as the Tao astronomic marker (Lat. 14° 03' 04.7'' N;
Long. 0° 22' 51.8'' E) 426;

 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) 427;

 from that point, the frontier follows the right bank of the River Sirba, from east to west, up to
the point on the right bank with the co-ordinates: Lat. 13° 19' 53.5'' N; Long. 1° 07' 20.4'' E;

 from that point, the frontier follows the line shown on the 1:200,000-scale map of the Institut

géographique national de France , 1960 edition, as far as the point with the co-ordinates:
Lat. 13° 22' 30.0'' N; Long. 0° 59' 40.0'' E;

 from that point, the frontier runs in a straight line south, 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 straight line up to the beginning of the Botou bend
(Tyenkilibi) (Lat. 12° 36' 19.2'' N; Long. 1° 52' 06.9'' E) 428.

Sketch-map No.15 on the following page (also reproduced in Cartographic Annex35)
illustrates this line, in the disputed portion of the frontier in particular. Sketch-map No. 16, inserted
after the submissions and also reproduced in Cartographic Annex 36, shows the entire course of the

frontier.

42The co-ordinates of this point were 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.
42The co-ordinates of this point, and the following ones, are given on the Clarke 1880 ellipsoid.

42The co-ordinates of this point are thoseadopted 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). - 132 -

Sketch-map No. 15 - General view of the frontier

General view of the frontier

 
 



Scale: 1:1,000,000

Scale: 1/1,000,000
   














 









































 


 

 


 




 
 





  














 

 


      

159 - 133 -

S UBMISSIONS

5.1. In view of all the above considerations, Burkina Faso requests the Court to adjudge and
declare that the frontier between Burkina Faso an d the Republic of Niger follows the course
described hereafter:

1. from the heights of N’Gouma to the Tong-Tong astronomic marker, the frontier takes the
following course: a series of straight lines connecting the following points in turn 429: Mount

N’Gouma (Lat. 14°54' 46.0" N; Long. 0°14' 36.4" E), Kabia Ford (Lat. 14°53' 09.8" N;
Long. 0° 13' 06.3" E), Mount Arwaskoye (Lat. 14° 50' 44.7" N; Long. 0° 10' 35.8" E), Mount
Bellé Banguia (Lat. 14° 45' 05.2" N; Long. 0° 14' 09.6" E), Takabougou (Lat. 14° 37' 54.5" N;

Long. 0°10' 16.1" E), Mount Douma Fendé (Lat. 14° 32' 00.6" N; Long. 0° 09' 42.1" E) and
the Tong-Tong astronomic marker (Lat. 14° 24' 53.2" N; Long. 0° 12' 51.7" E);

2. from the Tong-Tong astronomic marker to the beginning of the Botou bend, the frontier takes
the following course:

 a straight line as far as the Tao astronomic marker (Lat. 14° 03' 04.7'' N;
Long. 0° 22' 51.8'' E) 430;

 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 431;

 from that point, the frontier follows the right bank of the River Sirba, from east to west, up to
the point on the right bank with the co-ordinates: Lat. 13° 19' 53.5'' N; Long. 1° 07' 20.4'' E;

 from that point, the frontier follows the line on the 1:200,000-scale map of the Institut
géographique national de France , 1960 edition, as far as the point with the co-ordinates:

Lat. 13° 22' 30.0'' N; Long. 0° 59' 40.0'' E;

 from that point, the frontier runs south in a straight line, 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'');

 from that point, the frontier runs in a straight line up to the beginning of the Botou bend
432
(Tyenkilibi) (Lat. 12° 36' 19.2'' N; Long. 1° 52' 06.9'' E) ;

3. from the beginning of the Botou bend as far as the River Mekrou, the frontier takes the

following course:

 a series of straight lines connecting the following points in turn: Jackal Mountain
(Lat. 12° 41' 33.1" N; Long. 1° 55' 43.9" E), Laguil (Lat. 12° 41' 31.9" N; Long. 1° 57' 1.3" E)
and Nonbokoli (Lat. 12° 44' 12.9" N; Long. 1° 58' 47.0" E);

 from the latter point, the frontier follows the median line of the Dantiabonga marigot, passes to
the south of Dantiandou and then follows the line of the Yoga Djoaga hills as far as the

429
The co-ordinates which follow are those adopted in the record of the work of the Joint Survey Mission of the
erected markers, 3 July 2009, Ann. MBF 101. The co-ordinates are measured by GPS.
43The co-ordinates of this point were 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.

43The co-ordinates of this point, and the following ones, are given on the Clarke 1880 ellipsoid.

43The co-ordinates of this point, and the following ones, are those adopted in the record of the work of Joint
Survey Mission of the markers erected, 3July2009, Ann.MBF101. The co-ordinates were measured by GPS
(WGS84 ellipsoid). - 134 -

confluence of the Dyamongou and Dantiabonga rivers (Lat. 124 °3' 15.1" N;

Long. 2° 05' 14.9" E);

 from that point, the frontier follows the median line of the River Dyamongou as far as the
confluence of the Dyamongou marigot and the Boulel Fouanou (Lat. 12°43' 44.0" N;

Long. 2° 06' 23.9" E;

 from that point, the frontier runs in a series of straight lines connecting the following points in

turn: Boulel (Lat. 12°42' 15.1" N; Long. 2° 06' 53.3'' E), Boulel East (Teylinga)
(Lat.12°41'09.5" N; Long. 2°09' 43.2" E), Dyapionga North (Lat. 12°39' 42.3" N;
Long.2°09' 37.3" E), Dyapionga South (Lat. 12°38' 55.4" N; Long. 2°09' 08.1" E),

Kanleyenou (Lat. 12°37' 21.7" N; Long. 2°11' 57.1" E), Niobo Farou (Caiman Pool)
(Lat.12°35'19.6" N; Long. 2°13' 23.9" E), the eastern crests of Mount Tambouadyoaga
(Lat.12°31'19.7" N; Long. 2°13' 48.0" E), Banindyididouana (Lat. 12°27' 52.7" N;
Long.2°16' 27.2" E) and the confluence of the Banindyidi Fouanou and Tapoa rivers

(Lat. 12° 25' 30.5" N; Long. 2° 16' 40.6" E);

 from the latter of those points, the frontier follows the median line of the River Tapoa as far as
433
the point where it intersects with the former boundary of the Fada and Say cercles
(Lat. 12° 21' 04.88" N; Long. 2° 04' 12.77" E);

 from the latter point, the frontier runs in a straight line, corresponding to the former boundary

of the Fada and Say cercles, up to the point where it intersects with the River Mekrou
(Lat. 11° 54' 07.83" N; Long. 2° 24' 15.25" E).

5.2. Pursuant to Article 7, paragraph 4, of the Special Agreement, Burkina Faso further
requests the Court, in its Judgment, to nominate three experts to assist the Parties as necessary for
the purposes of demarcation.

20 April 2011

Signed [Signature illegible] Stamp of Minister of Justice

Jérôme T RAORÉ ,
Minister of Justice, Keeper of the Seals,

Agent of Burkina Faso.

Stamp and signature [illegible] of Minister for Stamp and signature [illegible] of Ambassador
Territorial Administration of Burkina Faso in Brussels
and Decentralization

Jérôme B OUGOUMA , Kadré Désiré O UEDRAGO ,
Minister for Territorial Administration and Ambassador of Burkina Faso to the Kingdom
Decentralization, Co-Agent. of the Netherlands, Co-Agent.

___________

43The co-ordinates of the following points are those adopted in the record of the meeting to ascertain the
co-ordinates of the unmarked points in SectorB, 15October2009, Ann.MBF105. They were derived from the IGN

France 1:200,000-scale map (Clarke 1880).  
  


General view of the frontier

    !
Scale: 1:1,350,000





 

 












 













 



 













 

 





 















 











 







&  ’  ( ) 

Line according to Burkina Faso


"     #   $ %
" 

    

Map prepared for illustrative purposes only

Consensual line of 1988
     
  
  
         

   "     - 136 -

L IST OF ANNEXES

Annex MBF 1. Franco-British Convention for the Delimitation of their Possessions west and
east of the Niger, 14 June 1898, 186 Consolidated Treaty Series, 313.

Annex MBF 2. Arrêté of the Governor-General of French West Africa of

23 December 1899, providing for the organization of two Military Territories
in Haut-Sénégal and Moyen-Niger ( Official Journal of French West Africa ,
year 1899 (date uncertain), p. 65).

Annex MBF 3. Arrêté of the Governor-General of French West Africa creating a Third
Military Territory, with its administrative centre at Zinder, 23July1900
(Official Journal of French West Africa, year 1900 (date uncertain), p. 313).

AnnexMBF4. Decree of 18October1904 of the President of the French Republic,
reorganizing the Government-General of French West Africa,
18 October 1904 (Official Journal of French West Africa , No.287 of
21 October 1904, pp. 6250-6252).

Annex MBF 5. Arrêté of the Governor-General of French West Africa organizing the
Military Territory of Niger, 26December1904 (O fficial Journal of Senegal
and Dependencies, pp. 718-719).

AnnexMBF6. Convention of 29May1906 between France and Great Britain for the
Delimitation of the Frontier between French and British Possessions to the
east of the Niger, signed at London on 9April1906 ( Consolidated Treaty

Series, Vol. 202, pp. 7-13).

Annex MBF 7. Agreement of 19 October 1906 between France and Great Britain relative to
the Frontier between French and British Possessions from the Gulf of Guinea

to the Niger (Southern Nigeria and Dahomey), signed at Paris ( Consolidated
Treaty Series, Vol. 203, pp. 51-63).

Annex MBF 8. Decree of the President of the French Republic incorporating into the Colony

of Haut-Sénégal et Niger the cercles of Fada N’Gourma and Say,
2 March 1907 (Official Journal of French West Africa , No. 24 of
30 March 1907, p. 135).

Annex MBF 9. Arrêté No. 1277 of the Governor-General of French West Africa defining the
various administrative divisions of the Military Territory of Niger,
31 December 1907 (Official Journal of French West Africa , No.158 of
11 January 1908, pp. 12-13).

Annex MBF 10. Arrêté No.1241 bis of the Governor-General of French West Africa
reorganizing the administrative divisions of the Military Territory of Niger,
14 December 1908 (Official Journal of French West Africa , No.209 of

2 January 1909, pp. 1-3).

AnnexMBF11. Decree of the President of the French Republic fixing the boundaries
between Dahomey and Haut-Sénégal et Niger, 12August1909 ( Official

Journal of French West Africa, No. 245 of 11 September 1909, pp. 419-420). - 137 -

Annex MBF 12. Arrêté No. 672 of the Governor-General of French West Africa incorporating
the region of Timbuktu into the Civil Territory of Haut-Sénégal et Niger,

22 June 1910 (Official Journal of French West Africa , No2. 93 of
30 July 1910, p. 475).

AnnexMBF13. Decree of the President of the French Republic incorporating the Military

Territory of Niger into the Government-General of French West Africa with
effect from 1 January 1912, 7 September 1911 (Official Journal of the
French Republic, No. 247 of 12 September 1911, p. 7437).

Annex MBF 14. Arrêté No1.728 of the Governor-General of French West Africa
reorganizing the internal administration of the Military Territory of Niger,
23 November 1912 (Official Journal of French West Africa , No.422 of
11 January 1913, pp. 68-69).

AnnexMBF15. Decree of the President of the French Republic modifying the boundary
between Dahomey and Haut-Sénégal et Niger, 23April1913 ( Official
Journal of French West Africa, date uncertain, p. 572).

AnnexMBF16. Decree of the President of the French Republic dividing the Colony of
Haut-Sénégal et Niger and creating the Colony of Upper Volta,
1 March 1919 (Official Journal of French West Africa , No7. 68 of

23 August 1919, pp. 549-551).

AnnexMBF17. Circular letter No.713 A.G. of the Lieutenant-Governor of Upper Volta to
the Administrators of the cercles of the Colony, 28 July 1920.

AnnexMBF18. Decree of the President of the French Republic reorganizing the Military
Territory of Niger and converting it into a Colony of the Civil Territory of
Mauritania with effect from 1January1921, 4December1920 ( Official

Journal of the French Republic, No. 335 of 9 December 1920, p. 20244).

Annex MBF 19. Decree of the President of the French Republic converting the Civil Territory
of Niger into an autonomous Colony, 13October1922 ( Official Journal of

French West Africa, No. 955 of 20 January 1923, p. 58).

AnnexMBF20. Letter No.682 A.G. from the Lieutenant-Governor of Upper Volta to the
Commander of Dori cercle, 22 March 1923.

AnneMx B2F1.Letter No1.58 from the Commander of Dori cercle to the
Lieutenant-Governor of Upper Volta, 7 April 1923.

AnnexMBF22. Letter No.1270 A.G. from the Lieutenant-Governor of Upper Volta to the

Governor-General of French West Africa, 7 June 1923.

Annex MBF 23. Telegram-letter No. 1305 A.G. from the Lieutenant-Governor of Upper Volta
to the Commander of Dori cercle, 13 June 1923.

Annex MBF 24. Letter No. A.G. from the Lieutenant-Governor of Niger to the
Governor-General of French West Africa, 26 January 1926.

AnnexMBF25. Letter No.307 A.G. from the Lieutenant-Governor of Upper Volta to the
Governor-General of French West Africa, 1 February 1926. - 138 -

Annex MBF 26. Decree of the President of the French Republic establishing the
administrative centre of Niger at Niamey and incorporating certain

Territories of Upper Volta into the Colony of Niger, 28December1926
(Official Journal of the French Republic , No.2, 50th year, of 2, 3 and
4 January 27, pp. 197-198) and Arrêté of the Governor-General of French
West Africa promulgating in French West Africa the Decree of

28December1926 transferring the administrative centre of the Colony of
Niger and providing for territorial changes in French West Africa,
21 January 1927 (Official Journal of French West Africa , No1. 167,
year 1927, p. 92).

Annex MBF 27. General List of Localities of French West Africa , fascicleIV, Colony of
Upper Volta, 1927.

Annex MBF 28. General List of Localities of French West Africa , fascicleVI, Colony of
Niger, 1927.

Annex MBF 29. Arrêté No. 184 of the Governor-General of French West Africa providing for

territorial changes to the Colonies of Upper Volta and Niger,
22 January 1927 (Official Journal of French West Africa , No.1169 of
12 February 1927, p. 124).

Annex MBF 30. Record of Agreement 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, 2 February 1927.

AnneMx B3F1.Record of the meeting between the representatives of the
Lieutenant-Governors of Upper Volta and Niger, 10 February 1927.

Annex MBF 32. Arrêté No4. 1 of the Governor-General of French West Africa of

5 March 1927 (Official Journal of Upper Volta , No.178 of 15March1927,
p. 110).

AnnexMBF33. Letter No.1040 A.G./I from the Lieutenant-Governor of Niger to the

Governor-General of French West Africa, 27 June 1927.

Annex MBF 34. Arrêté of the Governor-General of French West Africa of 31August1927
fixing the boundaries of the Colonies of Upper Volta and Niger ( Official

Journal of French West Africa , No.201 of 24Septembe1r927,
pp. 658-659).

Annex MBF 35. Erratum to the Arrêté of 31August1927 fixing the boundaries of the

Colonies of Niger and Upper Volta, 5October1927 ( Official Journal of
French West Africa, No. 1205 of 15 October 1927, p. 718).

AnnexMBF36. Decree of the President of the French Republic of 5September1932
dissolving the Colony of Upper Volta and distributing its territory among the

Colonies of Niger, French Sudan and Côte d’Ivoire ( Official Journal of
French West Africa, No. 1471 of 15 October 1932, p. 902).

AnnexMBF37. Extract from the book by Captain Y. Urvoy, Histoire des populations du

Soudan central (Colonie du Niger), Librairie Larose, Paris, 1936, p. 120.

AnnexMBF38. Mission report No.103 by Inspector of ColoniesBargues on the possible
reconstitution of Upper Volta, 30 May 1947. - 139 -

Annex MBF 39. Law No. 47-1707 of 4 September 1947 reconstituting Upper Volta within its
boundaries of 5September1932 ( Official Journal of French West Africa ,

No. 2311 of 27 September 1947, p. 948).

AnnexMBF40. Telegram-letter No.7/C from the Head of Téra Subdivision to Tillabéry
cercle, 23 January 1951.

AnnexMBF41. Data sheets for astronomic markers, sheet at 1:200,000, ND-31-XIII,
Observer Nevière, 1927, IGN 1955.

Annex MBF 42. Letter No. 1052/M1/AI and Note from the President of the Republic of Niger
to the President of the Republic of Upper Volta, 7 June 1963.

AnnexMBF43. Circular letter No.34/PRES/IS/DI from the Minister for the Interior and
Security of the Republic of Upper Volta to cercle Commanders,

12 September 1963.

AnnexMBF44. Note Verbale No.2216/MAE/DAPEC from the Ministry of Foreign Affairs
of the Republic of Niger to the Ministry of Foreign Affairs of the Republic of

Upper Volta, 17 June 1964.

AnnexMBF45. Protocol of Agreement on the demarcation of the frontier between the
Republic of Upper Volta and the Republic of Niger, Niamey, 23 June 1964.

Annex MBF 46. Letter No. 1 10/STC from the Director of the Topographic Service and Land
Registry of Niger to the Director of the IGN office in Dakar, 20 July 1964.

AnnexMBF47. Letter No.00517/CT from the Director for the Land Registry, Urban
Planning and Housing of Upper Volta to the Director of the IGN office in
Dakar, 25 July 1964.

AnnexMBF48. Letter No.32/MAE/DAPEC from the Ministry of Foreign Affairs of the
Republic of Niger to the Ministry of Foreign Affairs of the Republic of
Upper Volta, 6 May 1965.

AnnexMBF49. Letter No. 0288/PRES/CAB from the President of Niger to the President of
Upper Volta, 6 March 1967.

AnnexMBF50. Circular No.18/IS/D1 from the Minister for the Interior and Security of the

Republic of Upper Volta to the Commanders of the cercles adjoining Niger
(Diapaga, Fada N’Gourma, Bogande, Dori and Oudalan), 16 March 1967.

Annex MBF 51. Note Verbale No. 02697/MAE/DAPEC from the Ministry of Foreign Affairs

of the Republic of Niger to the Minister for Foreign Affairs of the Republic
of Upper Volta, 7 November 1967.

Annex MBF 52. Letter No. 88/CONF-DR from the Commander of Dori cercle to the Minister

for the Interior of the Republic of Upper Volta, 8 November 1967.

Annex MBF 53. Letter No. 23421/IS/DI from the Minister for the Interior to the Minister for
Foreign Affairs of the Republic of Upper Volta, 23 November 1967.

AnnexMBF54. Report of the ministerial meeting between Niger and Upper Volta, Niamey,
9-10 January 1968. - 140 -

Annex MBF 55. Official telegram No. 17/DR from the Deputy-Commander of Dori cercle to
the Commander of Dori cercle, 1 March 1968.

AnnexMBF56. Letter No.00446/AE/SG/AAM from the Minister for Foreign Affairs of the
Republic of Upper Volta to the Minister for the Interior and Security of the
Republic of Upper Volta, 13 March 1970.

Annex MBF 57. Mission report No. 86/DR-CONF from the Commander of Dori cercle to the
Minister for the Interior and Security of the Republic of Upper Volta,
30 October 1972.

AnnexMBF58. Letter from BernardMillot, cartographic artist at the IGN archive in Paris,
3 November 1972.

AnnexMBF59. Record of the meeting between the sous-préfet of Say (Republic of Niger)

and the sous-préfet of Sebba (Republic of Upper Volta), 12 November 1981.

AnnexMBF60. Report of the working meeting between the Minister for the Interior and
Security of Upper Volta and the Secretary of State for the Interior of the

Republic of Niger, Niamey, 16 September 1982.

Annex MBF 61. Treaty of Friendship and Co-operation between the Republic of Upper Volta
and the Republic of Niger, and the Agreement Establishing a Niger-Volta

Joint Commission on Co-operation, 25 April 1983.

Annex MBF 62. Decree No. 83-0207/CSP/PRES/AE.C ratifying the Treaty of Friendship and
Co-operation between the Republic of Upper Volta and the Republic of

Niger, and the Agreement Establishing a Niger-Volta Joint Commission on
Co-operation, 6 May 1983.

AnnexMBF63. Report of the meeting between the Minister Delegate for the Interior of the

Republic of Niger and the Minister for Territorial Administration of Burkina
Faso, Ouagadougou, 12-14 February 1985.

AnnexMBF64. Letter No.001198/MATS/CAB/SG/CF from the Minister for Territorial

Administration and Security of Burkina Faso to the Minister Delegate for the
Interior of the Republic of Niger, 6 May 1985.

AnnexMBF65. Letter No.518/MI/MDI/DAPA/CF from the Minister Delegate for the

Interior of the Republic of Niger to the Minister for Territorial
Administration and Security of Burkina Faso, 14 June 1985.

AnnexMBF66. Radio message No.11/4 from the Territorial Unit, Sebba gendarmerie, to

Dori company, Seno High Commissioner, 8 February 1986.

Annex MBF 67. Report No. 24/4 from the Territorial Unit Commander, Sebba gendarmerie,
to the Company Commander, Dori, 20 March 1986.

AnnexMBF68. Report of the meeting between the Minister for Territorial Administration
and Security of Burkina Faso and the Minister Delegate for the Interior of the
Republic of Niger, 9 April 1986. - 141 -

AnnexMBF69. Report of the meeting between technical experts of the Republics of Niger
and Burkina Faso, Ouagadougou, 21-23May1986, and annex thereto (List

of defining points on the Niger-Burkina frontier, ( Arrêté of 31August1927
and its Erratum of 5 October 1927)), 23 May 1986.

AnnexMBF 70. Letter No0.603/MATS/SG/DFR from the Minister for Territorial

Administration and Security of Burkina Faso to the Minister Delegate for the
Interior of the Republic of Niger, 3 October 1986.

AnnexMBF 71. Preliminary drafts of the Treaty and Protocol of Agreement of

28March1987 between the Government of the Republic of Niger and the
Government of Burkina Faso on the demarcation of the State frontier
between the two countries, 1986 (precise date uncertain).

Annex MBF 72. Agreement of 28 March 1987 on the demarcation of the frontier between
Burkina Faso and the Republic of Niger.

AnnexMBF73. Protocol of Agreement of 28March1987 on the demarcation of the frontier

between Burkina Faso and the Republic of Niger.

Annex MBF 74. Report and recommendations of the meeting between technical experts from
Niger and Burkina Faso held in Ouagadougou from 14 to 17May1987,

17 May 1987.

Annex MBF 75. Kiti of the President of Burkina Faso appointing the members of the Joint
Commission on Demarcation of the Frontier between Niger and Burkina

Faso, 19 June 1987.

AnnexMBF76. Report of the first meeting of the Joint Technical Commission on
Demarcation held in Niamey, 25-27 June 1987, 27 June 1987.

AnnexMBF77. Report of the mission conducted from 24 to 28June1987 in Niamey in
connection with the demarcation of the frontier, July 1987.

AnnexMBF78. Note No.ANV/36/103/MAT/SG/DGAT/DAF from the Director of Border

Affairs to the Minister for Territorial Administration of Burkina Faso,
19 January 1988.

AnnexMBF79. Radio message from the Director of Border Affairs on mission in Téra

(Niger) to the Director-General for Territorial Administration in
Ouagadougou, March 1988 (date uncertain).

Annex MBF 80. Report of the meeting of the Joint Technical Commission on Demarcation of

the Burkina-Niger Frontier, Diapaga, 12-15 May 1988, 15 May 1988.

AnnexMBF81. Report of the fourth meeting of the Joint Technical Commission on
Demarcation of the Niger-Burkina Frontier, Niamey, 26-28 September 1988,

28 September 1988.

AnnexMBF82. Report on completion of the survey of the proposed demarcation of the
Niger-Burkina frontier, 28 September 1988.

Annex MBF 83. Report of the meeting of the Joint Technical Sub-Committee on Demarcation
of the Burkina-Niger Frontier, Téra, 16-18 May 1989, 18 May 1989. - 142 -

AnnexMBF84. Report of the fifth meeting of the Joint Technical Commission on
Demarcation of the Niger-Burkina Frontier, Niamey, 13-14 November 1989,

14 November 1989.

AnnexMBF85. Report of the extraordinary meeting of the Joint Technical Commission on
Demarcation of the Niger-Burkina Frontier, Niamey, 1May1990,

15 May 1990.

AnnexMBF86. Mission report of the Topographical Sub-Committee of the Joint Technical
Commission on Demarcation of the Niger-Burkina Frontier, 5-12 June 1990,

12 June 1990.

Annex MBF 87. Report of the second ordinary session of the Joint Technical Commission on
Demarcation of the Niger-Burkina Frontier, Ouagadougou, 23-28 July 1990,

28 July 1990.

AnnexMBF 88. Report No.2/FP/MAT/SG/DCAF from the Minister for Territorial
Administration to the Head of State of Burkina Faso, 5 March 1991.

AnnexMBF89. Joint Communiqué of the Ministerial consultative and working meeting
between Niger and Burkina Faso held in Ouagadougou from 14 to
15 May 1991, 2 July 1991.

AnnexMBF90. Report No.61/MAG/SG/DCAF by the Minister-in-Council on the results of
the Ministerial meeting between Niger and Burkina, held at Ouagadougou
from 14 to 15 May 1991, 2 July 1991.

AnnexMBF91. Report of the third ordinary session of the Joint Technical Commission on
Demarcation of the Niger-Burkina Frontier, held in Niamey from 2 to
4 November 1994, 4 November 1994.

AnnexMBF92. Report of the meeting between the Ministers for Territorial Administration
and Security of the Republics of Niger and Burkina Faso held in Kompienga,
Kompienga Province, from 5 to 6 December 1997, 6 December 1997.

Annex MBF 93. Report of the meeting between the Minister for the Interior of Niger and the
Minister for Territorial Administration of Burkina Faso, Tenkodogo,
24-26 May 2000.

AnnexMBF94. Report of the fourth ordinary session of the Joint Technical Commission on
Demarcation of the Niger-Burkina Frontier, held in Ouagadougou from 18 to
21 July 2001, 20 July 2001.

Annex MBF 95. Communication from the Minister for the Interior and Decentralisation of the
Republic of Niger on the work of the Consultation Framework for cross-
frontier transhumance between the Republic of Niger and Burkina Faso,
25 January 2003

Annex MBF 96. Final Communiqué of the meeting between the Minister for the Interior and
Decentralization of the Republic of Niger and the Minister for Territorial
Administration and Decentralization of Burkina Faso, 4 April 2003.

AnnexMBF97. Report of the fifth ordinary session of the Joint Technical Commission on
Demarcation of the Niger-Burkina Frontier, Niamey, 28-30 September 2004,
30 September 2004. - 143 -

AnnexMBF98. Joint Communiqué of the meeting of Foreign Ministers for negotiation and
signature of the Special Agreement seising the International Court of Justice

of the Frontier Dispute between the Republic of Niger and Burkina Faso,
Niamey, 22-24 February 2009, 24 February 2009.

AnnexMBF99. Special Agreement seising the International Court of Justice of the frontier

dispute between Burkina Faso and the Republic of Niger of
24 February 2009.

AnnexMBF100. Extracts from the report of meeting No.11 of the Committee of Experts of

the National Commission monitoring the frontier dispute between Burkina
Faso and the Republic of Niger, 15 May 2009.

Annex MBF 101. Record of the work of the Joint Survey Mission to ascertain the co-ordinates

of the markers constructed on the frontier between Burkina Faso and the
Republic of Niger, 23 June to 3 July 2009, 3 July 2009.

Annex MBF 102. Letter from the Minister for Foreign Affairs of Burkina Faso to the Minister

for Foreign Affairs of Niger regarding the draft exchange of Notes
embodying the agreement between the Parties on the delimited sectors of the
frontier, 17 July 2009.

AnnexMBF103. Letter No.00618/MAE/C/DAJC/DIR from the Minister for Foreign Affairs
of Niger to the Minister for Foreign Affairs of Burkina Faso,
8 September 2009.

Annex MBF 104. Letter No. 2009-004607/MAECR/SG/DGAJC from the Minister for Foreign
Affairs of Burkina Faso to the Minister for Foreign Affairs of the Republic of
Niger, 8 October 2009.

Annex MBF 105. Record of the meeting to ascertain the co-ordinates of the unmarked points in

Sector B, 15 October 2009.

Annex MBF 106. Letter No. 2009-004874/MAECR/SG/DGAJC from the Minister for Foreign
Affairs of Burkina Faso to the Minister for Foreign Affairs of the Republic of

Niger, 29 October 2009.

Annex MBF 107. Letter No. 007505/MAE/C/DAJC/DIR from the Minister for Foreign Affairs
of the Republic of Niger to the Minister for Foreign Affairs of Burkina Faso,

2 November 2009.

___________ - 144 -

LIST OF CARTOGRAPHIC A NNEXES

Cartographic Annex MBF 1. “The Mossi and Gourmantché Empires”, map in M. Delafosse,
Haut-Sénégal-Niger (French Sudan), First Series, Vol.II:
History, Paris, Emile Larose, 1912, p. 151.

Cartographic Annex MBF 2. “The Conquest of West Africa (1880-1900)”, map in B. Lugan,
Atlas historique de l’Afrique des origines à nos jours , Ed.
Du Rocher, Monaco, 2001, p. 156.

CartographicAnnexMBF3. “The States and peoples of West Africa on the eve of the
European partition”, map in J.D.Fage, An Atlas of African
History, London, Arnold, 2nd ed., 1978, reproduced in M’Baye
Gueye and Albert Adu Boahen, “Initiatives et résistances

africaines en Afrique occidentale de 1880 à 1914”, in General
History of Africa, Vol. VII, UNESCO, 1987, p. 139.

Cartographic Annex MBF 4. Map of the cercles of Dori, Fada, Say, in L’Atlas des cercles de

L’AOF, 1926, fascicle IV, Upper Volta.

CartographicAnnexMBF5. Road map at 1:1,000,000 of the Colony of Upper Volta,
published by the Geographical Department of French West

Africa, Dakar, 1927.

CartographicAnnexMBF6. Sketch-map at 1:1,000,000, new frontier between Upper Volta
and Niger (according to the Erratum of 8October1927 to the

Arrêté of 32 August 1927).

CartographicAnnexMBF7. Map of Kandi at 1:200,000, Republics of Niger, Dahomey and
Burkina Faso, sheet NC-31-XXI, published by IGN Paris, 1st

ed. 1955 (reprinted December 1985).

Cartographic Annex MBF 8. Map of Kirtachi at 1:200,000, Republic of Niger, Burkina Faso,
Republic of Dahomey, sheet ND-31-III, published by IGN Paris,

November 1960 (reprinted December 1985).

CartographicAnnexMBF9. Map of Diapaga at 1:200,000, Republics of Niger and Upper
Volta, sheetND-31-II, published by the FWA Geographical

Department, Dakar, 1960 (reprinted July 1970).

CartographicAnnexMBF10. Map of Sebba at 1:200,000, Republics of Niger and Upper
Volta, sheet ND-31VII, published by the FWA Geographical

Department, Dakar, 1960.

CartographicAnnexMBF11. Map of Gothèye at 1:200,000, Republics of Niger and Upper
Volta, sheet ND-31-VIII, published IGN Paris, 2nd quarter 1960

(reprinted November 1969 after partial update).

Cartographic Annex MBF 12. Map of Téra at 1:200,000, Republics of Mali, Niger and Burkina
Faso, sheet ND-31-XIII, published by IGN Paris, 1st ed.

July 1960 (reprinted in September 1969 after partial update). - 145 -

CartographicAnnexMBF13. “The suppression of Upper Volta (1932)”, sketch-map No.2 in
ICJ Judgment, 22December1986, Frontier Dispute (Burkina

Faso/Republic of Mali), I.C.J. Reports 1986, p. 593.

CartographiAnnex MBF 14. Map mentioned at paragraph175 of the Judgment of
22 December 1986 rendered by the Chamber of the International

Court of Justice formed to deal with the case concerning the
Frontier Dispute (Burkina Faso/Republic of Mali) , I.C.J.
Reports 1986, Annex.

Cartographic Annex MBF 15. Photographs of the montage of the 1:200,000-scale maps of IGN
France from 1955 and 1960, showing the consensual line of the
frontier adopted following the fourth meeting of the Joint
Technical Commission on the Demarcation of the

Niger-Burkina Frontier held in Niamey on 26, 27 and
28September1988, signed by the Niger and Burkina Chairs of
the Commission.

CartographicAnnexMBF16. Map with overlay illustrating the situation of the Colony of
Niger following the Decree of 28December1926, extracted
from the map of French West Africa on a scale of 1:250,000,
2nd ed. 1928, as produced by the Republic of Niger in the atlas

appended to its Memorial in the case of the Frontier Dispute
(Benin/Niger), p. 105.

Cartographic Annex MBF 17. Map with overlay illustrating the “Administrative boundaries of

Côte d’Ivoire, French Sudan and Niger”, from the Bulletin du
comité de l’Afrique française , December1932, as produced by
the Republic of Niger in the atlas appended to its Memorial in
the case concerning the Frontier Dispute (Benin/Niger), p. 129.

Cartographic Annex MBF 18. Sketch-map No.1  General view of the frontier between
Burkina Faso and the Republic of Niger.

Cartographic Annex MBF 19. Sketch-map No.2  Situation of the Parties  Territorial
overview.

Cartographic Annex MBF 20.
Sketch-map No. 3  Administrative map of Burkina Faso.

Cartographic Annex MBF 21. Sketch-map No.4  Administrative divisions adjoining the
frontier.

Cartographic Annex MBF 22. Sketch-map No.5  Administrative map of the Republic of
Niger.

Cartographic Annex MBF 23. Sketch-map No6 .  View of the northern Sector 
Sector A  of the part of the frontier agreed by the Parties.

Cartographic Annex MBF 24. Sketch-map No7 .  View of the southern Sector 
Sector B  of the part of the frontier agreed by the Parties.

Cartographic Annex MBF 25. Sketch-map showing the entire course of the marked section.

Cartographic Annex MBF 26. Sketch-map No. 8  Disputed Sector of the frontier. - 146 -

Cartographic Annex MBF 27. Diagram No.1  View of the line between Tao and the River
Sirba.

Cartographic Annex MBF 28. Diagram No.2  Illustration of the method used to determine
the point where the frontier reaches the River Sirba at
Bossébangou.

Cartographic Annex MBF 29. Sketch-map No. 9  Location of the point on the River Sirba at
Bossébangou where the section of the frontier which starts at the

Tao astronomic marker terminates, called point “P”.

Cartographic Annex MBF 30. Sketch-map No.10  Section of the frontier following the
course of the River Sirba upstream along its right bank from

point P as far as the point from which the frontier turns back up
towards the northwest, leaving the four villages on the Niger
side, called point “P 1”.

Cartographic Annex MBF 31. Sketch-map No. 11  Location of point P 1.

Cartographic Annex MBF 32. Sketch-map No.12  Stretch of the frontier showing the
section turning back up towards the northwest from pointP1,

leaving the four villages on the Niger side, as far as the point
where it descends in a straight line to the south.

Cartographic Annex MBF 33.
Sketch-map No.13  Line from the intersection of the River
Sirba with the Say parallel, known as pointP3, turning back
due north.

Cartographic Annex MBF 34. Sketch-map No.14  Course of the frontier from the point
where it reaches the River Sirba at Bossébangou as far as the
intersection of the River Sirba with the Say parallel.

Cartographic Annex MBF 35. Sketch-map No. 15  General view of the frontier.

Cartographic Annex MBF 36. Sketch-map No.16  General view of the frontier and

consensual line of the frontier adopted following the fourth
meeting of the Joint Technical Commission on Demarcation of
the Niger-Burkina Frontier held in Niamey on 26, 27 and
28 September 1988, plotted on a map at 1:350,000.

___________

Document file FR
Document
Document Long Title

Memorial of Burkina Faso

Links