Separate opinion of Judge Yusuf

SEPARATE OPINION OF JUDGE YUSUF
Agreement with rejection of acquiescence  Also, with dismissal of application of parallel of latitude  Disagreement with certain aspects of implementation of methodology  Selection of base points for median line in territorial sea deviates from applicable law and jurisprudence of Court  Erroneous adjustment of provisional equidistance line by reference to “broader geographical configuration”  Taking account of extraneous relevant circumstances  Refashioning of geography in search of concavity and elusive cut-off  Incorrect use of the concept of “cut-off”  Flawed reasoning for delimitation within 200 nautical miles extended into delimitation of continental shelf beyond 200 nautical miles  Possible “grey area” may create new problems between Parties.
I. Introduction
1. I am in agreement with the decision of the Court to reject Kenya’s claim that Somalia had acquiesced to a maritime boundary that follows the parallel of latitude described in paragraph 35 of the Judgment.
2. I also concur in the decision of the Court to deny Kenya’s request to adjudge and declare that the maritime boundary between Somalia and Kenya in the Indian Ocean shall follow the parallel of latitude at 1° 39' 43.2" S. As noted in paragraph 130 of the Judgment, “the Court does not consider that the use of the parallel of latitude is the appropriate methodology to achieve an equitable solution”.
3. Consequently, the Court has decided to apply a median line in the territorial sea as prescribed by Article 15 of the United Nations Convention on the Law of the Sea (hereinafter “UNCLOS”) and to use its usual three-stage methodology for the establishment of the maritime boundary in the exclusive economic zone (hereinafter “EEZ”) and continental shelf. However, the way in which the base points have been selected in the construction of the median line for the territorial sea departs from the provisions of UNCLOS and from the jurisprudence of the Court. I will therefore address this matter in the present opinion.
4. As indicated in my vote against subparagraphs (4) and (5) of the dispositif, my main disagreement relates to the manner in which the three-stage methodology has been implemented, particularly with regard to the adjustment of the equidistance line. I could not agree with the unprecedented search for a concavity in a so-called “broader geographical configuration” (cf. paragraphs 164-168 of the Judgment), which has nothing to do with the geography and coastlines of Somalia and Kenya, but can only be understood as an attempt to justify a “judicial refashioning of geography”.
5. This is further compounded by a substantial adjustment of the provisional equidistance line constructed for the delimitation of the EEZ and continental shelf, without any reasons given except that it has been done on the basis of an allegedly “serious cut-off” of the coastal projections of Kenya (cf. paragraphs 168 and 171 of the Judgment). However, no such “serious cut-off” can be visualized within 200 nautical miles, even on sketch-map 10 of the Judgment entitled “Geographical configuration and its effect on equidistance lines”. This is a very regrettable and unprecedented use of the words “serious cut-off” for something different from what they actually mean.
- 2 -
6. The use of a geodetic line based on the incorrectly adjusted equidistance line brings into the delimitation of the area beyond 200 nautical miles the same flawed reasoning used for the area within the 200-nautical-mile zone. This reasoning does not take into account the fact that any “cut-off” effect of Kenya’s coastal projections in the outer continental shelf could solely be due to its agreement with Tanzania, which should have no legal effect on the delimitation between Somalia and Kenya. Moreover, the incorrect adjustment of the equidistance line gives rise to what the Judgment refers to as a “possible grey area” on the edge of the 200-nautical-mile delimitation. This “possible grey area”, which is depicted in sketch-map No. 12, may also lead in the future to a “Court-created” new problem between the Parties.
7. The reasons for my above reservations and disagreements are further elaborated below.
II. The construction of a median line in the territorial sea
8. The approach taken in the Judgment in the selection of base points for the construction of a median line is questionable for a number of reasons. First, according to Article 15 of UNCLOS, the median line in the territorial sea shall be constructed by reference to “the nearest points on the baselines from which the breadth of the territorial seas of each of the two States is measured”1. In this regard, Article 5 of UNCLOS states that, in principle, “the normal baseline for measuring the breadth of the territorial sea is the low-water line along the coast” (emphasis added). In Qatar v. Bahrain, the Court referred to Article 5 of UNCLOS and stated that “under the applicable rules of international law the normal baseline for measuring this breadth is the low-water line along the coast”2.
9. Similarly, in Eritrea/Yemen, the arbitral tribunal rejected Yemen’s argument that it should establish the median line boundary from base points on the high-water line, noting that “the use of the low-water line is laid down by a general international rule in the Convention’s article 5” and this “accords with long practice and with the well-established customary rule of the law of the sea”3. In Bangladesh v. India, the UNCLOS Annex VII arbitral tribunal referred to Eritrea/Yemen and reaffirmed that it would “determine the appropriate base points by reference to the physical geography at the time of the delimitation and to the low-water line of the relevant coasts4”.
1 Cf. Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain), Merits, Judgment, I.C.J. Reports 2001, p. 94. para. 177; see also Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial Guinea intervening), Judgment, I.C.J. Reports 2002, p. 442, para. 290.
2 Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain), Merits, Judgment, I.C.J. Reports 2001, p. 97, para. 184.
3 Second stage of the proceedings between Eritrea and Yemen (Maritime Delimitation), Award 17 December 1999, United Nations, Reports of International Arbitral Awards (RIAA), Vol. XXII, p. 338, para. 14 and p. 366, paras. 133-135.
4 Bay of Bengal Maritime Boundary Arbitration (Bangladesh v. India), Award of 7 July 2014, RIAA, Vol. XXXII, p. 75, paras. 221-223.
- 3 -
10. Thus, both this Court and other international courts and tribunals have plotted a provisional equidistance or median line in the territorial sea by reference to such base points on the low-water line in accordance with Article 5 of UNCLOS and general international law5. It follows that, in accordance with the provisions of UNCLOS, its own jurisprudence and the practice of other courts and tribunals, the Court should have constructed a provisional median line by reference to such base points on the low-water line from which the breadth of the territorial sea is measured. However, the Judgment deviates from this practice without providing adequate reasons for its seemingly random selection of base points.
11. Secondly, it is well established that, when the parties agree on a particular point, such as the placement of base points on the coast for the purposes of maritime delimitation, the respective court or tribunal will respect that agreement6, unless particular reasons warrant a different conclusion. In the present case, both Parties have proposed base points for the construction of the provisional median line, which reflect the geographical reality in the immediate vicinity of the land boundary terminus (hereinafter “LBT”). In its Memorial, Somalia identified three base points on the Somali side, two of which were located on the Diua Damasciaca Islands (S1 and S2), while the third (S3) was located on a low-tide elevation near the southernmost tip of Ras Chiamboni7. On the Kenyan side, Somalia identified two base points (K1 and K2) “on the most seaward points on the charted low-tide coast” of Kenya’s mainland8.
12. While Kenya did not originally identify any base points for the construction of the provisional median line, it provided such co-ordinates in the additional document it submitted as Appendix 2, Volume 1 (hereinafter “KAD”)9. First, Kenya proposed two base points on the mainland in the immediate vicinity of the LBT (K1 and S1), both of which are less than 1 km away from the LBT. On the Somali side, Kenya proposed a base point (S2) on the Diua Damasciaca Islands, and another base point (S3) on the low-water line of Ras Chiamboni. On the Kenyan side, Kenya has —
5 Delimitation of the maritime boundary in the Bay of Bengal (Bangladesh/Myanmar), Judgment, ITLOS Reports 2012, pp. 47-48, paras. 155-156 (“the Tribunal . . . will draw an equidistance line from the low-water line indicated on the Admiralty Chart 817 used by the Parties”); Arbitration regarding the Delimitation of the Maritime Boundary between Guyana and Suriname, Award of 17 September 2007, RIAA, Vol. XXX, p. 109, para. 393 (“The Tribunal accepts the basepoints for the low-water lines of Suriname and Guyana provided by the Parties that are relevant to the drawing of the equidistance line”); Arbitration between Barbados and the Republic of Trinidad and Tobago, Decision of 11 April 2006, RIAA, Vol. XXVII, p. 245, para. 381 (“The line of delimitation then proceeds generally south-easterly as a series of geodetic line segments, each turning point being equidistant from the low water line of Barbados and from the nearest turning point or points of the archipelagic baselines of Trinidad and Tobago”); see also ibid., p. 248, Appendix, Technical Report of the Tribunal’s Hydrographer, para. 1; Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial Guinea intervening), Judgment, I.C.J. Reports 2002, p. 443, para. 292; Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain), Merits, Judgment, I.C.J. Reports 2001, p. 97, para. 184, pp. 100-101, paras. 201-202 and p. 104, paras. 216 and 219. (All emphases added.)
6 Cf. Maritime Delimitation in the Caribbean Sea and the Pacific Ocean (Costa Rica v. Nicaragua) and Land Boundary in the Northern Part of Isla Portillos (Costa Rica v. Nicaragua), Judgment, I.C.J. Reports 2018 (I), p. 191, paras. 139-140 and pp. 206-207, para. 173; Delimitation of the maritime boundary in the Bay of Bengal (Bangladesh/Myanmar), Judgment, ITLOS Reports 2012, pp. 47-48, paras. 155-156. See also Arbitration regarding the Delimitation of the Maritime Boundary between Guyana and Suriname, Award of 17 September 2007, RIAA, Vol. XXX, p. 109, para. 393.
7 MS, paras. 5.19-5.20. On the Somali side, Somalia’s base points S1 and S2 are located at 1° 39' 43.30" S -41° 34' 35.40" E and 1° 39' 35.90" S - 41° 34' 45.29" E respectively. Base point S3 is located at 1° 39' 14.99" S - 41° 35' 15.68" E.
8 MS, paras. 5.19-5.20. On the Kenyan side, Somalia’s base point K1 is located at 1° 42' 00.06" S - 41° 32' 47.38" E. Somalia’s base point K2 is located at 1° 43' 04.77" S - 41° 32' 37.18" E.
9 KAD, pp. 188-189. On the Somali side, Kenya’s base point S1 is located at 1° 39' 36.3" S - 41° 33' 40.4" E. Kenya’s base point S2 is located at 1° 39' 40.9" S - 41° 34' 35.4" E. Kenya’s base point S3 is located at 1° 38' 57.0" S - 41° 35' 21.9" E.
- 4 -
just like Somalia — proposed base points on the low-water line of the mainland’s coastline (K3, K4 and K5)
10.
13. The base points proposed by the Parties for the construction of the provisional median line are largely concordant. Both Somalia and Kenya have placed base points on the Diua Damasciaca Islands and the low-water line of the mainland. Indeed, Kenya’s S2 is some 74 metres away from Somalia’s S1, and about 342 metres away from Somalia’s S2. The distance between Kenya’s S3 and Somalia’s S3 is approximately 587 metres. Somalia’s K2 is just about 775 metres away from Kenya’s K3 and 230 metres away from Kenya’s K411. As a result, the provisional median lines constructed by both Parties are very similar, as the Parties themselves have acknowledged12.
14. Notwithstanding the Parties’ general agreement on this point, the Judgment disregards the base points proposed by the Parties both on the mainland low-water line and the southernmost tip of Ras Chiamboni, as well as the Diua Damasciaca Islands, and departs both from the provisions of UNCLOS and from the jurisprudence of the Court regarding base points. Instead, a median line is constructed using base points which are located exclusively on the Parties’ terra firma (S1 to S4 and K1 to K4) and spread across an artificially straight line on the coast (paragraphs 115-116 of the Judgment). This is justified in the Judgment with the following statement: “Although in the identification of base points the Court will have regard to the proposals of the parties, it need not select a particular base point, even if the parties are in agreement thereon, if it does not consider that base point to be appropriate” (paragraph 111).
15. It may be true that in the Black Sea case, the Court stated that it “should not base itself solely on the choice of base points made by one of those Parties”13. The Court, however, did not suggest that it enjoys an unlimited discretion in selecting whichever base points it likes, nor did it elevate the criterion of “appropriateness” or what the Court “consider[s] . . . to be appropriate” (see paragraphs 111-112) to an all-encompassing standard on the basis of which the identification of base points should be made. On the contrary, the Court stated that it “must . . . select base points by reference to the physical geography of the relevant coasts14”. In the same vein, the Court stressed that
“the geometrical nature of the first stage of the delimitation exercise leads it to use as base points those which the geography of the coast identifies as a physical reality at the time of the delimitation. That geographical reality covers not only the physical elements produced by geodynamics and the movements of the sea, but also any other material factors that are present.15”
10 KAD, pp. 187-189 and fig. 11. On the Kenyan side, Kenya’s base point K1 is located at 1° 39' 51.6" S - 41° 33' 28.4" E. Kenya’s base point K2 is located at 1° 40' 39.6" S - 41° 32' 55.3" E. Kenya’s base point K3 is located at 1° 42' 40.1" S - 41° 32' 41.8" E. Kenya’s base point K4 is located at 1° 43' 12.2" S - 41° 32' 38.5" E.
11 The distance between the different co-ordinates in this paragraph were calculated using the following software: United States, Federal Communications Commission, ‘Distance and Azimuths Between Two Sets of Coordinates’, https://www.fcc.gov/media/radio/distance-and-azimuths.
12 Cf. CR 2021/3, p. 12, para. 13 (Reichler); KAD, pp. 187-189, para. 369 and figs. 11 and 12.
13 Maritime Delimitation in the Black Sea (Romania v. Ukraine), Judgment, I.C.J. Reports 2009, p. 108, para 137.
14 Ibid.
15 Ibid., p. 106, para. 131.
- 5 -
16. Thirdly, it is difficult to understand the decision to ignore the base points proposed by the Parties on the southernmost tip of Ras Chiamboni, a protuberance on the Somali mainland near the LBT. By ignoring Ras Chiamboni, the Judgment has disregarded a material feature in Somalia’s coastline which marks a “significant shift” in the direction of its coast. Even more confusingly, while paragraph 114 of the Judgment discounts Ras Chiamboni as a “minor protuberance” for the purposes of a median line, paragraph 146 of the Judgment places base point S6 on a much smaller protuberance in Somalia’s coast (opposite the Umfaali islets) for the purposes of constructing a provisional equidistance line in the EEZ and continental shelf. No explanation is given in the Judgment for such an inconsistent selection of base points.
17. Fourthly, this inconsistent approach is further repeated with regard to the base points proposed by the Parties on the Diua Damasciaca Islands, which are equally set aside. According to the Judgment, these islands are “tiny [and] arid” and “would have a disproportionate impact on the course of the median line” (paragraph 114). Curiously, however, paragraph 146 of the Judgment has placed base points K5 and K6 on Shakani Island off Kenya’s main coast, without giving reasons for this manifest inconsistency.
18. The fact that the Diua Damasciaca Islands are “tiny [and] arid” does not, ipso facto, preclude the Parties or the Court from selecting appropriate base points thereon as reflected in the past practice of the Court. It should indeed be recalled that the Court has considered appropriate to place base points on small insular features that were located in the immediate vicinity of the coast. This was the case, for example, in the Black Sea case, where the Court considered appropriate to use the south-eastern tip of Tsyganka Island as a base point, “because in this area of adjacency it [was] the most prominent point on the Ukrainian coast”16. In Nicaragua v. Colombia, the Court also considered that the “islands fringing the Nicaraguan coast” formed part of the “relevant coast” of Nicaragua, and consequently placed the base points on the Edinburgh Reef, Muerto Cay, Miskitos Cays, Ned Thomas Cay, Roca Tyra, Little Corn Island and Great Corn Island17. Also, while the Court refrained from placing base points on sandy features that are relatively unstable, it observed in Costa Rica v. Nicaragua that it would “construct the provisional median line for delimiting the territorial sea only on the basis of points situated on the natural coast, which may include points placed on islands or rocks”18.
19. The Judgment’s approach in the selection of base points has resulted in a contrived median line, the construction of which appears to have been aimed at producing a line which comes as close as possible to a bisector line, although there is nothing that justifies the use of a bisector for the delimitation of the territorial sea between Somalia and Kenya. Paragraph 118 in the Judgment reinforces this impression. Indeed, this paragraph suggests that the approach adopted by the Court for the construction of the median line may have been dictated by the search for a median line that “corresponds closely to the course of a line ‘at right angles to the general trend of the coastline’, assuming that the 1927/1933 treaty arrangement, in using this phrase, had as an objective to draw a line that continues into the territorial sea, a question that the Court need not decide”.
16 Maritime Delimitation in the Black Sea (Romania v. Ukraine), Judgment, I.C.J. Reports 2009, p. 109, para. 143 and p. 115, sketch-map No. 7.
17 Territorial and Maritime Dispute (Nicaragua v. Colombia), Judgment, I.C.J. Reports 2012 (II), p. 678, para. 145, pp. 698-699, para. 201 and p. 701, sketch-map No. 8.
18 Maritime Delimitation in the Caribbean Sea and the Pacific Ocean (Costa Rica v. Nicaragua) and Land Boundary in the Northern Part of Isla Portillos (Costa Rica v. Nicaragua), Judgment, I.C.J. Reports 2018 (I), p. 177, para. 100.
- 6 -
20. The 1927/1933 land boundary demarcation arrangements concluded between the former colonial Powers (Great Britain and Italy) have no relevance whatsoever to the dispute between Somalia and Kenya or to the delimitation of their maritime boundaries, because no maritime boundary between them was ever established by such arrangements, nor by the land boundary agreement concluded between the two colonial Powers in 1924, on which the 1927/1933 arrangements are based. As stated by Somalia in its reply to the question posed by a Member of the Court, “[n]either [it] nor Kenya, since their independence and at all times thereafter, has ever claimed that the maritime boundary in the territorial sea follows a line perpendicular to the coast at Dar es Salam, for any distance”.
21. The reference to such arrangements in paragraph 118, and the manner in which it is phrased, can only create misunderstandings. That is particularly the case because the Judgment itself, several paragraphs earlier, discounts the relevance of such colonial land demarcation agreements to the maritime boundary, inter alia, on the basis of the positions taken by the two neighbouring States both in their national legislation and in their negotiations and statements (see paragraphs 106 to 109). It is indeed concluded at the end of these paragraphs that “the Court therefore considers it unnecessary to decide whether the 1927/1933 treaty arrangement had as an objective the delimitation of the boundary in the territorial sea” (paragraph 109). Such an objective could not manifestly exist in a land demarcation agreement. What purpose is then served by invoking the same arrangement again in paragraph 118 in connection with the course of the median line as constructed by the Court? None whatsoever, in my view, if not to cast unwarranted doubt on the validity of the positions taken, and so clearly expressed, by two independent African States in their national legislation following their independence and the consequent exercise of their right to self-determination 60 years ago.
III. Delimitation of the EEZ and continental shelf within 200 nautical miles
22. As pointed out above, I disagree with the flawed reasoning used in the Judgment to justify the adjustment of the provisional equidistance line. It is wrong both as a matter of fact and of law. As a matter of fact, it entails an arbitrary refashioning of the geography by engaging in a search for a purported “concavity” in a so-called “broader geographical configuration” beyond the area of delimitation, which appears to be aimed at achieving preconceived results. As a matter of law, the reasoning deviates not only from the Court’s long-standing jurisprudence on the delimitation of the EEZ and continental shelf, but also from that of other international tribunals, without offering any rationalization for doing so. I will address each of these points in turn.
A. Refashioning of geography in search of a concavity and an elusive cut-off
23. The Court had hitherto applied its dictum that there should be no question of “a judicial refashioning of geography, which neither the law nor the practice of maritime delimitation authorizes”19. It is unfortunate that the Judgment breaks with that tradition. It does so by engaging in a search for a concavity beyond the coasts of the Parties and an elusive cut-off effect that could justify the adjustment of the equidistance line. Thus, it is stated in paragraph 164 that “[i]f the examination of the coastline is limited only to the coasts of Kenya and Somalia, any concavity is not conspicuous”. This is quite correct, and the story should have ended there because as was stressed by the Court in Cameroon v. Nigeria, “the concavity of the coastline may be a circumstance relevant to delimitation”, but “this can only be the case when such concavity lies within the area to be
19 Territorial and Maritime Dispute (Nicaragua v. Colombia), Judgment, I.C.J. Reports 2012 (II), p. 699, para. 202; see also Maritime Delimitation in the Black Sea (Romania v. Ukraine), Judgment, I.C.J. Reports 2009, p. 110, para. 149.
- 7 -
delimited”
20. However, the Judgment then goes on to say: “examining only the coastlines of the two States concerned to assess the extent of any cut-off effect resulting from the geographical configuration of the coastline may be an overly narrow approach”. It is not clear why the analysis is suddenly shifted to an assessment of a cut-off effect, or what is exactly meant by “geographical configuration of the coastline” in this context. In any case, “examining only the coastlines of the two States concerned” is not a narrow approach, but is in conformity with the scope of the jurisdiction of the Court which cannot be extended to the coastlines of third States. Somalia and Kenya requested the Court to delimit their maritime boundary, not that of third States. It is also legally erroneous to look for a concavity outside of the area to be delimited or to try to import it into a geographical area where it does not exist in order to achieve preconceived results.
24. The relevant circumstances that may justify the adjustment of a provisional equidistance line are essentially of a geographical nature. Indeed, the construction or adjustment of an equidistance line is dictated by the particular geography of the area to be delimited. It must faithfully reflect that geography, and that geography only. For such circumstances to be taken into account in order to achieve an equitable solution, they must also arise within the area to be delimited. It is the geographical situation of that area, its coastal configuration, the length of the coast and the presence of any special or unusual maritime features therein that may give rise to relevant circumstances to be taken into account in the construction or adjustment of an equidistance line. The importation of extraneous geographical factors lying beyond the Parties’ relevant coasts and the relevant area plainly contradicts the cardinal principle that “the land dominates the sea”.
25. The reliance of the Judgment on a so-called “broader geographical configuration”, which is not defined anywhere and the scope of which is not clearly indicated, effectively disconnects its analysis from the reality of the geographical circumstances prevailing in the relevant coasts and the relevant area of the maritime dispute between Somalia and Kenya. Moreover, by expanding the scope of enquiry into the coastline of a third State, the Judgment has reduced into irrelevance the role and function of the central concepts of “relevant coasts” and “relevant area” in the three-step methodology developed over the years by the Court for maritime delimitation, while paying lip service to their use in the present case.
26. Both the meaning of the word “concave” and the concept of “concavity” in maritime delimitation are also misused in the Judgment. First, in order to be described as “concave”, a coastline must look indented, hollowed or recessed in the middle, and curve inward like the inside of a bowl. According to the Oxford English Dictionary, concave means “having an outline or surface that curves inward like the interior of a circle or sphere”. Is there any coastal area which curves inward or looks like the inside of a bowl or the interior of a circle in the coastline of Somalia or Kenya? The answer is negative. The Judgment itself recognizes as much in paragraph 164. However, in an attempt at judicial refashioning of geography, it continues in its relentless, yet unjustified, search of such “concavity” in what it refers to as a “broader geographical configuration”.
27. Secondly, a concavity is a geographical given. It either exists or not in the area to be delimited. For it to be acknowledged or taken into account in the context of a maritime delimitation, it must belong to the geographical reality of such an area. It cannot be grafted onto the area by importing it from a “broader geographical configuration”, whatever such an expression may mean. The only coastline on which one can find a slight concavity in East Africa is that of Tanzania; but
20 Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial Guinea intervening), Judgment, I.C.J. Reports 2002, p. 445, para. 297.
- 8 -
this country is not a party to the dispute before the Court. The coastline of Tanzania has nothing to do with a maritime delimitation between Somalia and Kenya.
28. Nevertheless, it becomes eventually clear in paragraph 168 of the Judgment that it is indeed the Tanzanian coastline that is taken into account in order to justify the existence of a concavity in this part of the East African coast which the three States share. Thus, it is stated in paragraph 168: “The potential cut-off of Kenya’s maritime entitlements cannot be properly observed by examining the coasts of Kenya and Somalia in isolation. When the mainland coasts of Somalia, Kenya and Tanzania are observed together, as a whole, the coastline is undoubtedly concave”. From this observation, which practically includes the Tanzanian coast into the area to be delimited, contrary to the long-standing practice of the Court, the conclusion is drawn that “Kenya faces a cut-off of its maritime entitlements as the middle State located between Somalia and Tanzania”.
29. According to this reasoning, a strict equidistance line would be suitable for the delimitation of the coasts of Kenya and Somalia alone since they show no observable concavity when taken by themselves; but when the coast of Tanzania is taken into account, such an equidistance line would create a disadvantage for Kenya. This means that the Court has to take the coastline of a third State — not party to the dispute or to this case — into account in order to justify this artificial disadvantage which Kenya would suffer if an unadjusted equidistance line were used. However, what is overlooked by this erroneous analysis is that, for a concavity and its potential cut-off effect to be taken into account as a relevant circumstance in the delimitation of maritime areas, it must be rooted in the coastline of one of the Parties. The involvement in the delimitation process of coasts other than those of the Parties will have the effect of extending the area to be delimited to a coastline which has in fact nothing to do with it.
30. As was observed by Judge Koretsky in the North Sea Continental Shelf cases, macrogeographical considerations are “entirely irrelevant” in maritime delimitation, “except in the improbable framework of a desire to redraw the political map of one or more regions of the world”21. The arbitrary refashioning of geography to achieve preconceived results does not only distort the concept of relevant circumstances in the usual methodology of the Court for maritime delimitation in the EEZ and continental shelf, but it clearly contradicts its jurisprudence. As was correctly emphasized by the Court in Cameroon v. Nigeria: “The geographical configuration of the maritime areas that the Court is called upon to delimit is a given. It is not an element open to modification by the Court but a fact on the basis of which the Court must effect the delimitation.”22
B. The departure from the Court’s settled jurisprudence
31. I am equally in disagreement with the Judgment with regard to the adjustment of the equidistance line on the basis of the above-described considerations since, in doing so, it departs from the settled jurisprudence of the Court and of other international tribunals. In the Continental Shelf (Tunisia/Libyan Arab Jamahiriya) case, the Court observed that
21 North Sea Continental Shelf (Federal Republic of Germany/Denmark; Federal Republic of Germany/Netherlands), Judgment, I.C.J. Reports 1969, dissenting opinion of Vice-President Koretsky, p. 162.
22 Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial Guinea intervening), Judgment, I.C.J. Reports 2002, pp. 443-445, para. 295.
- 9 -
“[t]he only areas which can be relevant for the determination of the claims of Libya and Tunisia to the continental shelf in front of their respective coasts are those which can be considered as lying either off the Tunisian or off the Libyan coast. These areas form together the area which is relevant to the decision of the dispute.”23
Similarly, in Cameroon v. Nigeria, the Court noted that “the maritime boundary between Cameroon and Nigeria can only be determined by reference to points on the coastlines of these two States and not of third States”24. In Costa Rica v. Nicaragua, Costa Rica argued that it found itself in the situation of a “three-State concavity” where the “coastal concavity and the cut-off created by that concavity in conjunction with a notional delimitation with a third State” (Panama) would lead to an inequitable result25. The Court rejected this argument observing that
“[t]he overall concavity of Costa Rica’s coast and its relations with Panama cannot justify an adjustment of the equidistance line in its relations with Nicaragua. When constructing the maritime boundary between the Parties, the relevant issue is whether the seaward projections from Nicaragua’s coast create a cut-off for the projections from Costa Rica’s coast as a result of the concavity of that coast.26” (Emphases added.)
32. However, in an attempt to justify the departure from the Court’s practice, it is stated in paragraph 167 of the Judgment that “[i]n the present case, the potential cut-off of Kenya’s maritime entitlements should be assessed in a broader geographical configuration. This was also the approach adopted by the Arbitral Tribunal in the Guinea/Guinea-Bissau case.” It is rather strange that the Court should rely as judicial precedent on an award which considered the equidistance methodology, generally used by the Court, inapplicable to the delimitation of the coasts of Guinea and Guinea-Bissau because of their concavity27. More specifically, the award did not treat the concavity of the coastline of a third State in the region — since, in any event, the concavity was located within the relevant coasts of the parties28 — as a relevant circumstance for the adjustment of the provisional equidistance line. Rather, the tribunal adopted a methodology “looking at the whole of West Africa” as “a long coastline”, leading “towards a delimitation which [wa]s integrated into the present or future delimitations of the region as a whole”29.
33. Moreover, as was pointed out by the ITLOS Special Chamber in Ghana/Côte d’Ivoire, “the approach taken by that Award was not followed by subsequent international jurisprudence”30. Indeed, the Chamber was “not convinced that Côte d’Ivoire c[ould] rely on the jurisprudence of this Arbitral Award [in Guinea/Guinea-Bissau] to sustain its reasoning”, especially since “the maritime area off the coasts of Guinea and Guinea-Bissau is geographically complex, whereas the coasts of
23 Continental Shelf (Tunisia/Libyan Arab Jamahiriya), Judgment, I.C.J. Reports 1982, p. 61, para. 74.
24 Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v. Nigeria: Equatorial Guinea intervening), Judgment, I.C.J. Reports 2002, p. 442, para. 291.
25 Maritime Delimitation in the Caribbean Sea and the Pacific Ocean (Costa Rica v. Nicaragua) and Land Boundary in the Northern Part of Isla Portillos (Costa Rica v. Nicaragua), Judgment, I.C.J. Reports 2018 (I), p. 195, para. 150.
26 Ibid., p. 196, para. 156.
27 Delimitation of the Maritime Boundary between Guinea and Guinea-Bissau, Award of 14 February 1985, International Law Reports (ILR), Vol. 77, pp. 683-684, para. 108 (noting that its preferred approach “condemns the equidistance method as seen by Guinea-Bissau”).
28 Ibid., pp. 681-683, paras. 103-104 and 108 (“If the coasts of each country are examined separately, it can be seen that the Guinea-Bissau coastline is convex, when the Bijagos are taken into account, and that that of Guinea is concave”).
29 Ibid., pp. 683-684, para. 108; emphasis in the original.
30 Delimitation of the Maritime Boundary in the Atlantic Ocean (Ghana/Côte d’Ivoire), Judgment, ITLOS Reports 2017, p. 89, para. 287.
- 10 -
Ghana and Côte d’Ivoire are straight rather than indented”
31, which is the case also of the coasts of Somalia and Kenya. It is therefore difficult to understand why the International Court of Justice would have recourse to such an award, which flies in the face of its own jurisprudence in the delimitation of maritime boundaries through the use of the equidistance line in a three-stage methodology.
34. The other judgments and awards relied upon to justify the adjustment of the equidistance line are similarly either inapposite or have nothing to do with the circumstances of the present case, and do not provide authority for taking into account as a relevant circumstance, the coastline of a third State which is not party to these proceedings and which is situated well beyond the relevant coasts and area. First, reference is made to the North Sea Continental Shelf cases (paragraph 165). It may be true that in the North Sea Continental Shelf cases, the Court recognized that the marked concavity or convexity of the coastline may amount to an equitable consideration for the adjustment of the equidistance line32. But, in that case, the marked concavity and convexity of the coastline existed in the relevant coasts of all three States that were parties to the proceedings; it did not involve the coastlines of a third State far from the relevant area, such as the United Kingdom, Norway or Belgium.
35. Secondly, paragraph 166 refers to the cases of Bangladesh/Myanmar and Bangladesh v. India. Leaving aside the fact that the Bay of Bengal, with its marked concavities and sinuosities, bears no resemblance to the — almost linear — coastlines of Somalia and Kenya, in those cases the respective tribunals limited their enquiry specifically to the coasts of the parties to these proceedings. They did not consider the potential effect of the concavity of the Bay of Bengal vis-à-vis the coasts of third States. As stated by ITLOS in Bangladesh/Myanmar, “concavity per se is not necessarily a relevant circumstance”33. The Tribunal stressed that an adjustment may be necessary “when an equidistance line drawn between two States produces a cut-off effect on the maritime entitlement of one of those States, as a result of the concavity of the coast”34; the “coast in question”, however, was understood as “the coast of Bangladesh”, a party to these proceedings, not the coastline of India, which was not mentioned in the relevant analysis35.
36. In Bangladesh v. India, the arbitral tribunal also referred to Cameroon v. Nigeria and Bangladesh/Myanmar, “not[ing] the common view in international jurisprudence that concavity as such does not necessarily constitute a relevant circumstance requiring the adjustment of a provisional equidistance line”36. The tribunal stressed that “the existence of a cut-off effect should be established on an objective basis and in a transparent manner”, whereas “a decision as to the existence of a cut-off effect must take into account the whole area in which competing claims have been made”37.
31 Delimitation of the Maritime Boundary in the Atlantic Ocean (Ghana/Côte d’Ivoire), Judgment, ITLOS Reports 2017, p. 89, para. 287.
32 North Sea Continental Shelf (Federal Republic of Germany/Denmark; Federal Republic of Germany/Netherlands), Judgment, I.C.J. Reports 1969, p. 49, para. 89 (a).
33 Delimitation of the maritime boundary in the Bay of Bengal (Bangladesh/Myanmar), Judgment, ITLOS Reports 2012, p. 81, para. 292.
34 Ibid.
35 Ibid., pp. 81-82, paras. 293 and 297; see also ibid., p. 87, paras. 323 and 325 (“the coast of Bangladesh . . . is decidedly concave. This concavity causes the provisional equidistance line to cut across Bangladesh’s coastal front” and “The Tribunal . . . takes the position that . . . an adjustment must be made to its provisional equidistance line to abate the cut-off effect of the line on Bangladesh’s concave coast . . . in light of the coastal geography of the Parties”) and p. 89, para. 333, referring to the “coasts of the Parties”. (All emphases added.)
36 Bay of Bengal Maritime Boundary Arbitration (Bangladesh v. India), Award of 7 July 2014, RIAA, Vol. XXXII, p. 120, para. 402.
37 Ibid., p. 121, para. 404.
- 11 -
In assessing the concavity as a relevant circumstance, the arbitral tribunal examined the projections of the “coast of Bangladesh”, which was “manifestly concave”, and the projections of the “south-east-facing coasts of India”
38. It did not take into account the coastlines of Myanmar, which was not mentioned in the relevant analysis.
37. Thirdly, even if it were to be assumed, arguendo, that a concavity exists in the area to be delimited, which of course is not the case here, such a concavity would have to be, in the first instance, a marked one; and secondly it would have to produce a severe or serious cut-off effect to be considered as a relevant circumstance. Neither a marked concavity in East Africa, including the Tanzanian coast, nor a serious cut-off or shut-off of the seaward projection of Kenya’s maritime boundary can be identified on a map in the instant case within the 200-nautical-mile area. A strict equidistance line between Somalia and Kenya allows the seaward projection of their coasts to proceed in the same general direction, and does not stop or cut off Kenya’s potential entitlements (see sketch-map No. 10 in the Judgment). Words must have a meaning, and a slight narrowing of the coastal projections of a country cannot be characterized as a “serious cut-off”. It is not fitting for a court to claim, as Humpty-Dumpty did in Alice in Wonderland, that “when I use a word, it means just what I choose it to mean  neither more nor less”.
38. Fourthly, as was observed by the arbitral tribunal in Bangladesh v. India, two criteria must be met for a cut-off produced by a provisional equidistance line to warrant adjustment of the provisional equidistance line:
“First, the line must prevent a coastal State from extending its maritime boundary as far seaward as international law permits. Second, the line must be such that — if not adjusted — it would fail to achieve the equitable solution required by articles 74 and 83 of the Convention. This requires an assessment of where the disadvantage of the cut-off materializes and of its seriousness.” 39
Neither of these criteria is met in the present case. A cut-off must be capable of causing something to end or to be stopped. However, no such effect is produced by an unadjusted equidistance line between Somalia and Kenya within the 200-nautical-mile zone, whether the Kenya-Tanzania parallel of latitude or a strict equidistance line is used. At 200 nautical miles, the distance between the Kenya-Tanzania parallel of latitude and the unadjusted equidistance line with Somalia would still be, according to Kenya, about 180 km wide40. Thus, an unadjusted equidistance line would not prevent Kenya from extending its maritime boundary “as far seaward as international law permits”.
39. Indeed, paragraph 171 recognizes that “the cut-off effect in the present case is less pronounced than in some other cases” but goes on to say that “it is nonetheless still serious enough to warrant some adjustment to address the substantial narrowing of Kenya’s potential entitlement”. It is not clear what is meant by a “serious enough” cut-off in this context, nor is this notion elaborated in the Judgment. However, its use is not, in any case, consistent either with the ordinary meaning of the word “cut-off” in English nor with international jurisprudence. According to the Oxford Advanced Learner’s Dictionary, to “cut off” means to “remove something from something larger by cutting”, to “block or get in the way of something”. The central idea in a “cut-off” in maritime
38 Bay of Bengal Maritime Boundary Arbitration (Bangladesh v. India), Award of 7 July 2014, RIAA, Vol. XXXII, pp. 121-122, paras. 406-407.
39 Ibid., p. 124, para. 417; see also Delimitation of the Maritime Boundary in the Atlantic Ocean (Ghana/Côte d’Ivoire), Judgment, ITLOS Reports 2017, p. 120, para. 422.
40 Cf. CMK, para. 343 and fig. 3-1; CR 2021/3, p. 19, para. 31 (Reichler).
- 12 -
delimitation is to preclude the coastline of a State from projecting seaward as far as international law permits, such that a mere narrowing of a seaward projection would not qualify as a “cut-off”.
40. The jurisprudence of the Court and of other international tribunals is quite clear on the meaning and implications of a cut-off in maritime delimitation. The first reference to a “cut-off” in the jurisprudence of the Court was in the North Sea Continental Shelf cases, where the Court stated that “in the case of a concave or recessing coast . . . the effect of the use of the equidistance method is to pull the line of the boundary inwards, in the direction of the concavity”, causing the area enclosed by the equidistance lines “to take the form approximately of a triangle with its apex to seaward and, as it was put on behalf of the Federal Republic, ‘cutting off ’ the coastal State from the further areas of the continental shelf outside of and beyond this triangle”41.
41. In Bangladesh v. India, Bangladesh found itself in a situation similar to the one described in the North Sea Continental Shelf cases, consisting of a triangle with its apex to seaward, as a result of a strict application of the provisional equidistance line. The tribunal noted that
“in the present case, the seaward projections of the west-facing coast of Bangladesh on the north-eastern margins of the Bay of Bengal . . . are affected by the provisional equidistance line. The effect is even more pronounced in respect of the southward projection of the south-facing coast of Bangladesh . . . as far as the area beyond 200 [nautical miles] is concerned. The cut-off effect is evidently more pronounced from point Prov-3 southwards, where the provisional equidistance line bends eastwards to the detriment of Bangladesh, influenced by base point I-2 on the Indian coast and the receding coast of Bangladesh in the inner part of the Bay . . . On the basis of the foregoing, the Tribunal concludes that, as a result of the concavity of the coast, the provisional equidistance line it constructed in fact produces a cut-off effect on the seaward projections of the coast of Bangladesh.”42
42. Likewise, in the case concerning the Delimitation of the Maritime Boundary between Guinea and Guinea-Bissau, the arbitral tribunal stated that
“[w]hen in fact . . . there are three adjacent States along a concave coastline, the equidistance method has the other drawback of resulting in the middle country being enclaved by the other two and thus prevented from extending its maritime territory as far seaward as international law permits”43.
43. The non-existence of a “cut-off” in the present case — much less a serious one — is further demonstrated by the use of the concept in Article 7, paragraph 6 (on straight baselines), and Article 47, paragraph 5 (on archipelagic baselines), of UNCLOS, which read as follows:
41 North Sea Continental Shelf (Federal Republic of Germany/Denmark; Federal Republic of Germany/Netherlands), Judgment, I.C.J. Reports 1969, p. 17, para. 8.
42 Bay of Bengal Maritime Boundary Arbitration (Bangladesh v. India), Award of 7 July 2014, RIAA, Vol. XXXII, pp. 122-123, paras. 407-408. See also ibid., p. 122, map 6 (Projections from coasts).
43 Delimitation of the Maritime Boundary between Guinea and Guinea-Bissau, Award of 14 February 1985, ILR, Vol. 77, p. 682, para. 104.
- 13 -
“The system of straight baselines may not be applied by a State in such a manner as to cut off the territorial sea of another State from the high seas or an exclusive economic zone.”
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
“The system of such baselines shall not be applied by an archipelagic State in such a manner as to cut off from the high seas or the exclusive economic zone the territorial sea of another State.” (Emphases added.)
44. These provisions reproduce almost verbatim the text of Article 4 (5) of the 1958 Convention on the Territorial Sea and the Contiguous Zone44. Originally, the idea of a “cut-off” effect was not envisaged by the International Law Commission’s 1956 “Articles concerning the Law of the Sea”45. The idea of a “cut-off” in the Convention originates from a Portuguese proposal at the 1958 United Nations Conference on the Law of the Sea, with very similar wording46. In explaining its proposal, the delegate of Portugal stated that “it would be absurd if one coastal State were able to deny another coastal State access to the high seas”47. These provisions were taken as the basis for UNCLOS III, without much debate as to their substance or content.
45. According to the Virginia Commentary on Article 7 of UNCLOS:
“Paragraph 6 [of Article 7 of UNCLOS] is based on article 4, paragraph 5, of the 1958 Convention, with the addition of the reference to the exclusive economic zone. Its purpose is to protect the access of a coastal State to any open sea area where it enjoys the freedom of navigation. The additional reference to the exclusive economic zone in paragraph 6 is justified since the freedom of navigation is exercised also in that zone under article 58, paragraph 1.”48 (Emphasis added.)
46. The question therefore arises whether there is any area in the coastal projections of Somalia and Kenya within 200 nautical miles or beyond it which, because of the use of the equidistance line, takes the form “approximately of a triangle with its apex to seaward”, thus cutting off Kenya from further areas of the EEZ or continental shelf beyond this triangle, or which results in Kenya being enclaved. The answer is manifestly negative. Neither a serious cut-off nor an enclavement can be visualized even on sketch-map No. 10 of the Judgment, which only shows a slight narrowing of the coastal projections of Kenya that cannot realistically be claimed to cut it off from, or block its access to, any maritime zone within or beyond 200 nautical miles.
44 United Nations, Treaty Series (UNTS), Vol. 516, Convention on the Territorial Sea and the Contiguous Zone, 1958. Article 4, paragraph 5, reads as follows: “The system of straight baselines may not be applied by a State in such a manner as to cut off from the high seas the territorial sea of another State.”
45 Yearbook of the International Law Commission, 1956, Vol. II, p. 265 et seq.
46 Official Records of the United Nations Conference on the Law of the Sea, (Volume III, First Committee (Territorial Sea and Contiguous Zone)), Summary records of meetings and Annexes, UN doc. A/CONF.13/39, p. 240, doc. A/CONF.13/C.1/L.101, Portugal: proposal (Article 5), second point: “Insert a new paragraph as follows: ‘4. The system of straight baselines may never be drawn by a State in such a manner as to cut off from the high seas the territorial sea of another State.’”
47 Ibid., p. 148, para. 27.
48 Nandan, Satya N., S. Rosenne and N. R. Grandy (volume eds), United Nations Convention on the Law of the Sea, 1982: A Commentary, Vol. II, 1993, Center of Oceans Law and Policy, University of Virginia, Martinus Nijhoff, Dordrecht, p. 103, para. 7.9 (h).
- 14 -
47. To conclude, it should be recalled that the Court has repeatedly emphasized in the past the need to “be faithful to the actual geographical situation”49 in defining the relevant coast and relevant area and to avoid “completely refashioning nature”50. The present Judgment engages in such refashioning by importing into the area to be delimited between Somalia and Kenya, the characteristics of the coastline of a third State, namely the existence of a slight concavity off the coast of Tanzania. The law and methodology hitherto developed by the Court for the purposes of delimitation between adjacent or opposite coasts have given rise to a high degree of predictability and a normative coherence in the interpretation and application of the international law of the sea.
48. This long-standing predictability and coherence risk to be shattered by the incorrect and unprecedented approach used in the adjustment of the equidistance line in the present Judgment by disregarding a cardinal principle of maritime delimitation, that “the land dominates the sea”. By introducing into the analysis of the overlapping claims of Somalia and Kenya extraneous coastal configurations and geographical circumstances well beyond the relevant coasts of the Parties, and beyond the relevant area, the Judgment has introduced into the law and process of maritime delimitation considerations which are “strange to its nature”51 and undermine the reliable methodology developed by the Court.
IV. Delimitation of the continental shelf beyond 200 nautical miles
49. I agree that the Court should proceed to a delimitation of the continental shelf beyond 200 nautical miles as requested by both Parties. I disagree, however, with the manner in which the delimitation has been implemented for the following reasons.
50. First, for the same reasons as described above, I disagree with the extension of the same geodetic line that was unjustifiably adjusted within the 200 nautical miles. There was no valid reason to do so. The Court cannot simply assert that a delimitation line should take a certain course without justifying it or giving convincing reasons for it. The narrowing of the coastal projections of Kenya is in fact more pronounced after the 200 nautical miles due to Kenya’s maritime delimitation agreement in 2009 with Tanzania. However, this is not specifically mentioned in the Judgment.
51. It should be recalled, in this connection, that in that agreement Kenya deliberately chose the parallel of latitude delimitation instead of an equidistance line in order to gain about 10,000 sq km within 200 nautical miles, which, however, made it lose more than 25,000 sq km of maritime space beyond 200 nautical miles. Thus, if there is a cut-off effect in the area beyond 200 nautical miles, it is purely and simply due to Kenya’s choice in 2009. Moreover, the agreement between Kenya and Tanzania cannot have any legal effect for Somalia in accordance with the principle pacta tertiis nec nocent nec prosunt. For this reason, Somalia cannot be required to compensate Kenya for the maritime area it surrendered on the basis of its agreement with Tanzania by shifting the equidistance line northwards in its favour as has been done in the Judgment.
49 Continental Shelf (Libyan Arab Jamahiriya/Malta), Judgment, I.C.J. Reports 1985, p. 45, para. 57.
50 Ibid. See also North Sea Continental Shelf (Federal Republic of Germany/Denmark; Federal Republic of Germany/Netherlands), Judgment, I.C.J. Reports 1969, p. 49, para. 91.
51 Continental Shelf (Libyan Arab Jamahiriya/Malta), Judgment, I.C.J. Reports 1985, p. 40, para. 48; Maritime Delimitation in the Area between Greenland and Jan Mayen (Denmark v. Norway), Judgment, I.C.J. Reports 1993, p. 63, para. 57.
- 15 -
52. Secondly, the extension of the adjusted equidistance line beyond 200 nautical miles along the above-mentioned geodetic line also creates a new problem with regard to what the Judgment refers to as the “grey area”. It is the erroneous manner in which the adjustment of the equidistance line is made in the present case that produces this “grey area” as depicted in sketch-map No. 12. Although it is stated in the Judgment that such a “grey area” is only a possibility, and therefore the Court “does not consider it necessary . . . to pronounce itself on the legal régime that would be applicable in that area” (paragraph 197), the mere reference to it and its representation in a sketch-map which is an integral part of the Judgment may create a new and unnecessary controversy between these two neighbouring States in the future.
(Signed) Abdulqawi A. YUSUF.
___________