On 2 February 2017, the International Court of Justice handed down its Judgment on preliminary objections in the case concerning Maritime Delimitation in the Indian Ocean (Somalia v. Kenya). Somalia had brought the case to request that the Court determine its single maritime boundary with neighbouring Kenya. The ICJ held that it may proceed to the merits phase, thereby rejecting the respondent’s submissions. Among other arguments, Kenya raised an objection rooted in Part XV (“Settlement of disputes”) of the 1982 United Nations Convention on the Law of the Sea (LOSC). It contended that the Convention’s dispute settlement system is an agreement on the method of settlement for its maritime boundary dispute with Somalia and therefore falls within the scope of Kenya’s reservation to its optional clause declaration made pursuant to Art. 36(2) of the ICJ Statute, which excludes “[d]isputes in regard to which the parties to the dispute have agreed or shall agree to have recourse to some other method or methods of settlement”.
The fact that Kenya relied on this argument is noteworthy in and of itself, as it was the first time that the Court faced a LOSC-based jurisdictional challenge. Moreover, we believe that the way in which the Court disposed of this argument has far-reaching implications since it casts a long shadow over dispute resolution in the law of the sea. But before delving into the ICJ’s reasoning and its ramifications, we will highlight some essentials of the LOSC dispute settlement system.
Part XV of the LOSC
Part XV is famous for establishing a range of compulsory dispute settlement procedures entailing binding decisions. States parties may choose from among several such procedures: the International Tribunal for the Law of the Sea (ITLOS), the ICJ, arbitration and special arbitration. It should be stressed that arbitration, regulated in Annex VII of the LOSC, is the “default option”. This means that (a) if a State party has not voiced a preference it is considered to have selected arbitration and (b) if the parties to the dispute have not accepted the same procedure, the legal dispute may be submitted only to an arbitral tribunal.
This obligatory scheme of adjudication is not all-encompassing. While certain categories of disputes are excluded from the compulsory system, States parties may opt out of others, for instance maritime delimitation. Art. 282 of Part XV offers parties to a dispute yet another way out:
If the States Parties which are parties to a dispute concerning the interpretation or application of this Convention have agreed, through a general, regional or bilateral agreement or otherwise, that such dispute shall, at the request of any party to the dispute, be submitted to a procedure that entails a binding decision, that procedure shall apply in lieu of the procedures provided for in this Part, unless the parties to the dispute otherwise agree. (emphasis added)
It is widely understood that the wording “or otherwise” covers the situation of optional clause declarations made by both parties to a dispute under Art. 36(2) of the ICJ Statute (see e.g. M.H. Nordquist, S. Rosenne & L.B. Sohn, United Nations Convention on the Law of the Sea, 1982: A Commentary (Vol. V, Martinus Nijhoff, 1989, pp. 26-27)).
The ICJ’s Approach to Part XV of the LOSC
The Court had to tackle the following conundrum: do the parties’ optional clause declarations form an agreement to appear before the ICJ and consequently exclude the LOSC dispute settlement system (Art. 282 of the LOSC) despite the Kenyan reservation? The core of the Court’s reasoning derives from the travaux préparatoires of the Convention, i.e. the Third United Nations Conference on the Law of the Sea (UNCLOS) (1973-1982). More than half of the optional clause declarations that existed at the time of the UNCLOS had a Kenyan-type reservation. Notwithstanding their prevalence, the Court found that the travaux préparatoires reveal no intention on the part of the UNCLOS participants to exclude most of the existing optional clause declarations in that period, namely those including Kenyan-type reservations. Thus, the Court held, such reservations cannot bar the application of Art. 282.
Several difficulties arise from this aspect of the ICJ’s reasoning. Generally, it can be questioned whether travaux préparatoires provide a strong footing on which to base the jurisdiction of the Court, which consistently applies a stringent test of “preponderance”. Turning to the specifics, it stands to reason that the silence of the Conference on this issue could have been construed differently. After all, silence can ‘say’ many things. The absence of any debate on Kenyan-type reservations during the UNCLOS might simply stem from a broadly shared belief that such reservations rendered Art. 282 inapplicable and that this needed no further comment. Judge Robinson, the sole dissenter on the matter of Part XV in Somalia v. Kenya and a former Ambassador to the UNCLOS, drives similar points home in convincing detail. It is also worth observing that the Judgment does not contain references to the case law of the ITLOS and Annex VII Arbitral Tribunals, which throughout the past two decades have done much to shed light on the interpretation of Part XV.
There was another legal basis for accepting jurisdiction, which was mentioned by the Court itself but only in subsidiary order. The ICJ took its cue from its predecessor, the Permanent Court of International Justice, which affirmed the Court’s ability to entertain a case should the alternative be a denial of justice resulting from a negative conflict of jurisdiction (Factory at Chorzów (Jurisdiction), Judgment, p. 30). This approach proved prescient in light of a last minute declaration under the LOSC filed by Kenya that excludes maritime delimitation from the Convention’s compulsory system. It was deposited nine days before the ICJ delivered its Judgment and is not mentioned in the ruling. The denial of justice line might serve as an antidote to strategically timed tinkering with declarations under the LOSC and/or the ICJ Statute (cf. e.g. the recent reservation added to the UK’s optional clause declaration in response to proceedings instituted by the Marshall Islands: “any dispute which is substantially the same as a dispute previously submitted to the Court by the same or another Party.”).
The Impact on LOSC Dispute Settlement
There is the real prospect of Somalia v. Kenya making waves in LOSC dispute settlement where both parties have entered optional clause declarations under Art. 36(2) of the ICJ Statute and a Kenyan-type reservation applies between them. This class of reservation is not only widespread but also good form as its inclusion in authoritative model clauses would suggest (see Handbook on accepting the jurisdiction of the International Court of Justice, para. 29 and a Council of Europe Recommendation).
First, the Court’s ratio does not seem to consider States’ choice of forum under the LOSC. What does this mean for a country like Canada, for instance? Canada has inserted a Kenyan-style reservation into its optional clause declaration. It has also selected the ITLOS and arbitration as its preferred fora under the LOSC to the exclusion of the ICJ. Could Canada see maritime disputes being brought against it before the ICJ despite its express intention not to have the World Court settle such disputes? To avoid such an outcome, States are well-advised to list the various categories of sea disputes they wish to exclude in their optional clause declarations (see para. 128 of the 2017 Judgment implying as much).
Second, reluctant respondents appearing before the ITLOS or an Annex VII Arbitral Tribunal now have a powerful card up their sleeve. They could plausibly argue that the dispute should have been submitted to the ICJ. Given ITLOS Judges’ and Annex VII Arbitrators’ steady reliance on the Court’s jurisprudence, it is quite possible they will decline jurisdiction lest they open the door to a procedural fragmentation of sorts.
Beyond the practical ramifications of Somalia v. Kenya, systemic shifts are afoot. In 1945, the founders of the United Nations sought to give pride of place to the World Court for the settlement of legal disputes (see Art. 36(3) of the UN Charter). During the UNCLOS the delegations could not reach the same consensus in favour of the ICJ, nor the ITLOS for that matter. After painstaking and protracted discussions, they arrived at the following compromise in 1982: there would come a compulsory system, but priority would be given to arbitration (unless parties to a dispute agree otherwise). This part of the package deal has shown great vitality, suffice it to mention the instrumental role of Annex VII Arbitral Tribunals in “heating up” the once dormant Permanent Court of Arbitration. Fast-forward to the present day, the ICJ’s Somalia v. Kenya Judgment has in one fell swoop salvaged a case that otherwise might not have been adjudicated while strengthening the Court’s position in the pantheon of international courts and tribunals. But at what cost?
The authors would like to thank Tamar Meshel and Matina Papadaki (both of MPI Luxembourg) for their insightful comments.