Delimiting

delimitation:
Lessons learned from the ITLOS Chamber judgment on the
Dispute concerning delimitation of the maritime boundary between Ghana
and Côte d’Ivoire in the Atlantic Ocean

Maria Gavouneli*

On 23 September 2017 an ITLOS Special Chamber issued its decision on the
delimitation of the maritime boundary between Ghana and Côte d’Ivoire in the
Atlantic Ocean. At stake were millions of barrels of oil in the Tweneboa Enyenra
Ntomme (TEN) oilfields, an area discovered in 2009 and brought to production
in 2016 – at a time the case was already brought before ITLOS with a request for
provisional measures, which put a temporary halt to the full development of the
field.
This is the second delimitation case brought before ITLOS and the first before a
Special Chamber of the Tribunal, constituted pursuant to article 15(2) of the
ITLOS Statute upon an agreement of the Parties. In reaching its final decision, the
tribunal fully reviewed both the delimitation provisions of the Law of the Sea
Convention (LOSC) and the extensive delimitation jurisprudence produced by
other international courts and tribunals. To a certain extent, the judgment reads
like a consolidation of the law of maritime delimitation, solidly erected on
established principles and practices over the determination of a single maritime
boundary and the three-stage process, involving a provisional equidistance line
adjusted on the basis of relevant circumstances and controlled through a
disproportionality test to ensure an equitable result. The interesting elements to
be found in this judgment are not new in content but rather in the emphasis
accorded to their normative scope: the primacy of State consent remains the
foundation myth of international law – and it is reaffirmed in this case. And
arising therefrom, as a rational extension of the principle but certainly in a novel
formulation, the acts and omissions of each State in pursuit of its rights may
trigger the responsibility of the State concerned. I will deal with each one of
these questions in turn.
A tacit agreement – or not
The present case started its judicial life in 2014 as a request instituting arbitral
proceedings under Annex VII of the Law of the Sea Convention and was
eventually brought before a Special Chamber of the International Tribunal on the
Law of the Sea on the basis of an agreement between the Parties. They confirmed
–and the tribunal concurred [para. 99]– that the applicable law comprised the
delimitation provisions of the Convention to be found in articles 15 (territorial
sea), 74 (EEZ), 76 (outer continental shelf) and 83 (continental shelf) as well as
‘other rules of international law not incompatible with the Convention’ under
article 293 LOSC. The starting point remains in all cases a possible agreement

* Associate Professor of International Law, Faculty of Law, National & Kapodistrian University of

Athens; Associate Research Fellow, Institute of Advanced Legal Studies, University of London.

Electronic copy available at: https://ssrn.com/abstract=3054575

between the Parties; and only lacking such agreement is the corresponding
procedure triggered.
Ghana claimed that a ‘customary equidistance boundary’ already existed
between the Parties in their Atlantic coast [para. 102]. It argued that the limits of
the Parties’ oil concessions over the years constitute but a reflection of a
mutually agreed and recognized pre-existing maritime boundary [para. 111],
such oil practice a tacit assertion of a valid boundary, albeit in the absence of a
formal boundary treaty [para. 110]. In contrast, Côte d’Ivoire suggested that oil
concessions, requests and authorizations for seismic surveys, intermittent or no
protest to drilling activities and the consistent use of oil concession maps
reflected “caution in a context of uncertainty relating to an undelimited area
rather than to a formal request or authorization to cross a delimited boundary”
[para. 128] and not a common understanding of the Parties to that effect.
The Special Chamber concurred. It found that the oil practice of the Parties as
well as their subsequent bilateral exchanges do not constitute ‘compelling
evidence’ that a tacit legal agreement was indeed in place [para. 212]. Citing the
ICJ judgment in the Territorial and Maritime Dispute between Nicaragua and
Honduras in the Caribbean Sea (Nicaragua v. Honduras, ICJ Reports 2007), it
affirmed that “[t]he establishment of a permanent maritime boundary is a matter
of grave importance and agreement is not to be easily presumed” (ibid, para.
253). Furthermore, since such actions do not amount to compelling evidence of
the presumed consent for the Parties, it follows that they do not constitute
“conduct [that] has created the appearance of particular situation and another
State, relying on such conduct in good faith, has acted or abstained from a action
to its detriment” [para. 242, citing Delimitation of the maritime boundary in the
Bay of Bengal (Bangladesh/Myanmar, Judgment, 2012, para. 124)] either; and
consequently Côte d’Ivoire is not estopped from requesting the delimitation of
the heretofore non-existent maritime boundary [para. 245].
In essence, the Special Chamber shows itself both traditional and pragmatic in
outlook. Traditional because it appears to rely extensively and perhaps
excessively on the expressed will of the State, setting aside other indications of
presumed consent. In doing so, it remains wilfully blind to the unwillingness of
States to enunciate and fully qualify actions and omissions, which are
nevertheless significant of their sovereign choices. Nevertheless, it is also
pragmatic, as it reinforces legal security and allows the States sufficient leeway
to pursue their lawful interests without the choking effect of evidentiary value. It
is worth remembering in this respect that in indicating provisional measures
(Order of 25 April 2015, paras 94-95), the same Chamber had maintained a
delicate balance between disallowing new drilling in the contested area and
maintaining operations in the existing installations, thus safeguarding both their
continued commercial value and effective protection of the marine environment
without putting ‘undue burden’ on Ghana (ibid, paras 99-100). Taken in
conjunction with the freezing effect of the obligation under articles 74(3) and
83(3) LOSC not to jeopardize or hamper the reaching of a final agreement while

Electronic copy available at: https://ssrn.com/abstract=3054575

making every effort to enter into provisional arrangements of a practical nature,
this exacting standard allows States operating in undelimited areas to engage in
exploration and exploitation activities related to their natural resources,
provided that they operate in good faith in strict observance of all other
applicable rules, such as the obligation to conduct an Environmental Impact
Assessment in advance and during such operations – see also the BIICL Report on
the Obligations of States under Articles 74(3) and 83(3) of UNCLOS in respect of
undelimited maritime areas, 2016, pp. 19-34. Otherwise, any claim over a
maritime area or even a mala fide attempt at negotiation would effectively
prohibit the affected State from engaging in any such lawful activities – one could
hardly consider a more drastic disincentive to the peaceful settlement of
disputes…
Responsibility for acts undertaken in undelimited areas
The full repercussions of this approach are further explored in the final part of
the judgment, where the Special Chamber is called upon to decide whether the
international responsibility of Ghana has been engaged by conducting or
licensing hydrocarbon activities in an area over which Côte d’Ivoire claims to
have sovereign rights [para. 589]. The tribunal “emphasizes that in a case of
overlap both States concerned have an entitlement to the relevant continental
shelf on the basis of their respective coasts. Only a decision on delimitation
establishes which part of the continental shelf under dispute appertains to which
of the claiming States” [para. 591], clearly indicating that only such a judicial
pronouncement has a constitutive nature, which grants one State entitlement
priority over the other. Although the Special Chamber does not refer to it, the
same conclusion was also reached in the Matter of the South China Sea
Arbitration between the Republic of Philippines and the People’s Republic of China
(Award, 12 July 2016, paras 950-993): the arbitral tribunal found China in
breach of article 300 LOSC, not for asserting its claims on disputed maritime
areas but for its flagrant breach of other obligations under international law,
including those on the protection of endangered species.
Côte d’Ivoire also argued that the international responsibility of Ghana was
triggered by its alleged obligation to negotiate in good faith with a view to
achieving an equitable solution to the dispute as per article 83(1) LOSC; and,
while doing so, not to jeopardize or hamper the conclusion of an agreement as
per article 83(3) LOSC. In the first case, the Special Chamber reiterated “that the
obligation to negotiate in good faith is an obligation of conduct and not one of
result” [para. 604] and consequently a State cannot be held accountable for
failing to reach an agreement. In the second instance, the tribunal identified two
interlinked obligations, ‘to make every effort to enter into practical
arrangements of a practical nature’ and ‘not to jeopardize or hamper the
reaching of the final agreement’ during the transitional period [para. 626], both
obligations of conduct [paras 627 and 629] and subject to a duty to act in good
faith regardless of the final outcome. It concluded that activities undertaken in
the area claimed by Ghana could not be considered in breach of the obligation to
jeopardize or hamper [para. 633], thus effectively reading together the two
connected obligations. Judge Paik in his Separate Opinion did not appear that
convinced: he qualified the obligation as ‘a result-oriented notion’ [para. 6] and
concentrated on the need to reach such decisions on a case-by-case basis, going
beyond the ‘permanent physical change’ standard suggested in the Award in the
arbitration regarding the delimitation of the maritime boundary between Guyana
and Suriname (award, 17 September 2007, XXX RIIA, para. 467); or the
‘significant and permanent modification of the physical character of the area in
dispute’ employed by the Special Chamber in the provisional measures phase
[para. 8, citing Order, 25 April 2015, para. 89].
All these actions were allegedly carried out once the tribunal was instituted and
the dispute brought before it: The Special Chamber held that issues of
international responsibility arising from the ‘dispute concerning the delimitation
of [their] maritime boundary’ would stretch the meaning of the words [para.
548]; it accepted, however, that Ghana had not objected to the relevant claims
brought by Côte d’Ivoire, affirmed jurisdiction on the basis of forum prorogatum
[para. 552] and proceeded to apply the ILC Articles on State Responsibility on
the basis of both article 304 LOSC [para. 556] and customary international law
[para. 558, citing the ITLOS Seabed Disputes Chamber advisory opinion on the
Responsibilities and obligations of States with respect to activities in the Area,
Advisory Opinion, 1 February 2011, para. 169].
In contrast, the Special Chamber had no difficulty to affirm that jurisdiction to
adjudicate over the alleged violation of the provisional measures prescribed in
2015 falls within the inherent competence of the tribunal [para. 546]; upon
detailed consideration, it again absolved Ghana of any such breach [para. 658].
A single maritime boundary
The Tribunal was asked by the Parties to delimit a single maritime boundary
with respect to the territorial sea, the exclusive economic zone and the
continental shelf [para. 74], a fairly typical request although not always followed
in recent practice: in the Delimitation of the maritime boundary in the Bay of
Bengal (Bangladesh/Myanmar, Judgment, 2012) the Tribunal traced three
delimitation lines, each with its own methodology, for the territorial sea, the
continental shelf and EEZ and the outer continental shelf respectively, and ended
up creating a grey zone where the respective continental shelves and EEZs did
not coincide. The LOSC does distinguish between the rules applicable to the
delimitation of the territorial sea (article 15 LOSC) and those applicable to the
delimitation of the exclusive economic zone and the continental shelf (articles 74
and 83 LOSC). The Tribunal took note of the distinction [para. 260] but it
concluded that the Parties had “implicitly agreed” [para. 262] to follow the same
methodology for the whole delimitation process and that was the methodology
developed for the latter set of maritime zones [para. 259]. It did, however, note
that the rights of the coastal State on the territorial sea entail sovereignty over
the seabed, the superjacent waters and the air column above and are not
functional; and that no sovereignty-raised questions were raised by the Parties.
Is there an implication here that, had sovereignty issues been raised, the
outcome would have been different in terms of applicable law? One could safely
assume so.
Of particular interest is the Tribunal’s treatment of the request for delimitation
of the continental shelf beyond 200 nm. Ghana has already addressed itself to the
Commission on the Limits of the Continental Shelf (CLCS) and the relevant
procedure has been completed. Côte d’Ivoire has also made its submission but it
has not received as yet any recommendations. Both Parties agreed that the
Tribunal was competent to delimit the boundary between those portions of the
outer continental shelf – and the Tribunal concurred, having examined its
jurisdiction proprio motu. In essence, the Special Chamber treated the
continental shelf as unitary, correctly emphasizing that: “there is in law only a
single continental shelf rather than an inner continental shelf ad a separate
extended or outer continental shelf [para. 490, citing Arbitration between
Barbados and the Republic of Trinidad and Tobago relating to the delimitation of
the exclusive economic zone and the continental shelf between them, Decision of
11 April 2006, XXVII RIAA, para. 213]. While respectful of any CLCS
recommendations, issued under article 76(8) LOSC, it concluded that the
delimitation line extends beyond the 200nm limit because
“the functions of the CLCS and those of the Special Chamber differ. Whereas
the former differ with the delineation of the continental shelf beyond
200nm, the latter decides on delimitation with a neighbouring State, that is
to say, on the course of the lateral limits” [para 517].
One would be justified to seek the reasons for the different approach taken by
ITLOS in the Delimitation of the maritime boundary in the Bay of Bengal
(Bangladesh/Myanmar, Judgment of 14 March 2012, paras 390-392) lie in the
common position of the Parties that the Special Chamber had jurisdiction to
decide on the delimitation of the continental shelf beyond 200nm [para. 489].
A common methodology
In any case, the Special Chamber proceeded to decide on the methodology to be
used for such unitary delimitation. Rejecting the arguments of Côte d’Ivoire for
the application of the angle bisector methodology, it affirmed the choice made by
ITLOS in the Delimitation of the maritime boundary in the Bay of Bengal
(Bangladesh/Myanmar) case, where it was clearly stated that the
equidistance/relevant circumstances method was the favoured by international
courts and tribunals (Judgment, 2012, para. 238). Indeed, the tribunal treated
the few cases where the equidistance line was not followed as due to the
particular circumstances of those cases, which the adjudicating court or tribunal
had “not only the possibility but also the obligation to take into account” [para.
323]: the configuration of the land, the unstable character of the mouth of the
river Coco and the competing claims over title to several small islands in the area
in the case of the Territorial and Maritime Dispute between Nicaragua and
Honduras in the Caribbean Sea (Nicaragua v. Honduras, Judgment, 2007, para.
275]; the half effect given to the Kerkennah islands in the Continental Shelf case
(Tunisia/Libyan Arab Jamahiriya, Judgment, 1982, para. 129); the specific impact
of the demarcation line in the Delimitation of the Maritime Boundary in the Gulf of
Maine Area (Canada/USA, Judgment, 1984); or even the geographical factors in
the Case concerning Delimitation of the Maritime Boundary between Guinea and
Guinea-Bissau (Award, 1985, 77 ILR 684). Neither the scarcity of base points and
their location nor the supposed instability of the coastline was considered
adequate.
Instead, the Special Chamber endorsed the by now traditional three-stage
approach to delimitation [para. 360] and proceeded to apply it in the present
case. First, It drew a provisional equidistance line in what it considered a single
continental shelf [para. 373], deciding on the extent and location of the relevant
coasts and area and determining base points: boldly stating that it was under no
obligation to accept base points suggested by the Parties [para. 393], it
proceeded to select them on the basis of the geographic peculiarities of the
relevant coast, which it had already defined as those seaward projections of the
coast which overlap with those of the other party [para. 361, citing the Maritime
Delimitation in the Black Sea case (Romania v. Ukraine, Judgment, 2009, paras
96-97, 99); para. 372, citing the Delimitation of the maritime boundary in the Bay
of Bengal case (Bangladesh/Myanmar, Judgment, 2012, para. 198)].
Moving into the second stage of determining relevant circumstances, on the basis
of which the provisional line is to be adjusted, the tribunal clarified its view of
delimitation: it “must not completely refashion geography or compensate for the
inequalities of nature” [para. 409]. As a result, it found that the concavity of the
Ivorian coast and the convexity of the Ghanaian coast [para. 426] or the
geography of the Jomoro peninsula [para. 436] did not constitute such relevant
circumstances. Of particular interest is the clear decision of the Special Chamber
not to be swayed by the location and distribution of hydrocarbon resources:
“Maritime delimitation is not a means for distributing justice” [para. 452]. The
rule is that
“… in principle, … maritime delimitation … is based on geographical
considerations. Only in extreme situations – in the words of the Chamber of
the ICJ in the Gulf of Maine case – if the envisaged delimitation was “likely
to entail catastrophic repercussions for the livelihood and economic well-
being of the population of the countries concerned” …, may considerations
other than geographical ones become relevant…” [para.453].
Albeit most forceful in its delivery, the tribunal is not innovative in this approach.
Ever since the Continental Shelf case (Tunisia/Libyan Arab Jamahiriya, ICJ
Reports 1982), the presence of energy installations is considered “an element to
be taken into account in the process of weighing all relevant factors to achieve an
equitable result” (para. 107) but it is clear that
“these economic considerations cannot be taken into account for the
delimitation of the continental shelf areas appertaining to each Party. They
 are virtually extraneous factors since they are variables which
unpredictable fortune or calamity, as the case may be, might at any time
cause to tilt he scale one way or another. A country may be poor today and
become rich tomorrow as a result of an event such as the discovery of a
valuable economic resource” (ibid).
The emphasis in the present case is put on access to resources rather than
existing exploitation efforts but the spirit remains the same. The Special
Chamber concentrates on the scope of application of possible exceptions, going
back to the arbitral award in the Grisbadarna case (Norway/Sweden, decision of
23 October 1909, XI RIAA 147), in order to fully justify its rejection of any such
relevant circumstances. The practice of the Parties in according oil concessions
in the same Area is also treated in the same manner [para. 478].
In the final stage of the delimitation process, the Special Chamber found that the
relevant area it had calculated on the basis of the length of the relevant coasts it
had indicated is in fact not disproportionate to the equidistance line it traced and
has now become permanent since there were no relevant circumstances
advocating its adjustment. In fact, the final result is slightly in favour of Cote
d’Ivoire, which had the longer coastline (352 km as compared to the 139km of
the Ghanaian coast), but the ratio of the length of the relevant coasts at
approximately 1:2,53 is almost similar to the ratio of the allocated areas at
approximately 1:2,02 [para. 537].
It is interesting to note in that respect that the tribunal accorded particularly
emphasis on transparency and predictability, which it also considered objectives
of the delimitation process [para. 281, citing the Bay of Bengal Maritime
Boundary Arbitration between Bangladesh and India, Award of 7 July 2014, para.
339]. The impact of these objectives seems to permeate the system, covering not
only the three-phase process but also applying when States are pursuing an
equitable solution as per articles 74(1) and 83(1) LOSC. Indeed, the Special
Chamber seems to consider adherence to the established delimitation
jurisprudence as an inherent element of the transparency and predictability
requirements. Truth be told, there would be a lot of relieved States and experts if
that were the case, as occasionally court watching and the related ability to
foresee the slings and arrows of outrageous (judicatory) fortune amounts to an
obscure art.
All in all, the casual reader of this judgment would be treated to a comprehensive
treatise of the state of play in delimitation, as it stands today. Not a bad result for
a day in court…