I. Maritime Delimitation in the Indian Ocean (Somalia v. Kenya)

Somalia asked the Court to determine the complete course of the single maritime boundary between itself and Kenya, including the outer continental shelf beyond 200 nautical miles (nm).Footnote ² Kenya challenged the Court's jurisdiction and also contested the admissibility of Somalia's Application. This was the first time that either state had appeared before the ICJ. The case raised important questions about the role of the Court in maritime boundary cases, and in particular whether it should yield priority to the 1982 UN Convention on the Law of the Sea (UNCLOS)Footnote ³ dispute settlement bodies. The Court adopted a flexible approach to treaty interpretation to avoid such a result.

Both states had made declarations under the Optional Clause accepting the Court's jurisdiction and the case turned mainly on the interpretation of one of Kenya's reservations to its declaration. The reservation excluded from the Court's jurisdiction “[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.”Footnote ⁴ Kenya argued that the parties had in fact agreed to have recourse to some other method of settlement in a 2009 Memorandum of Understanding (MOU) and also in UNCLOS Part XV, and that therefore the Court's jurisdiction was excluded by its reservation. This particular type of reservation is very common,Footnote ⁵ but the Court had never before found that it lacked jurisdiction on the basis of a reservation of this kind, and it did not do so in this case.Footnote ⁶

First, Kenya's argument with regard to the MOU. The general legal framework for the Court's judgment was provided by UNCLOS, to which both Somalia and Kenya are parties and which therefore governs the delimitation of their single maritime boundary. UNCLOS provides for states to claim an outer continental shelf beyond 200 nm. Article 76 assigns a special role to the Commission on the Limits of the Continental Shelf (CLCS)Footnote ⁷ with regard to the “delineation” of the outer limits of the shelf. A state party intending to establish the outer limits of its continental shelf shall submit information on such limits, including scientific and technical data, to the CLCS which is then to make recommendations to the coastal state; “[t]he limits of the shelf established by [the] coastal state on the basis of these recommendations shall be final and binding.”Footnote ⁸ Article 76(10) provides that: “The provisions of this article are without prejudice to the question of delimitation of the continental shelf between States with opposite of adjacent coasts.”Footnote ⁹

The requirement to provide scientific and technical data to the CLCS proved a problem for several developing states. They risked losing their claims to a continental shelf beyond 200 nm if they were not able to submit the necessary information to the CLCS within the deadline set under the Law of the Sea Convention.Footnote ¹⁰ Somalia faced particular difficulties because it was only beginning to emerge from the violence and instability prevalent since 1991, it was still in the process of establishing an effective government, and it lacked the resources and necessary expertise. Norway accordingly provided assistance to certain African states, including Somalia, in the preparation of their submissions to the CLCS.Footnote ¹¹

Under its Rules of Procedure, the CLCS is not to delineate the outer limit of a state's continental shelf if there is a dispute, unless there is “prior consent given by all States that are parties to such a dispute.”Footnote ¹² Somalia argued that there was such a dispute: the parties had radically different positions on the method for the determination of the boundary, with Kenya asserting a line on the (unusual) basis of a parallel of latitude, whereas Somalia's own claim was based on equidistance, to be determined by the three-stage process generally followed by the Court in recent maritime boundary cases.

In order to allow the delineation of their outer continental shelves to go ahead, Somalia and Kenya therefore concluded an MOU on April 7, 2009. This agreement was drawn up with the assistance of NorwayFootnote ¹³ and it was registered and published in accordance with Article 102 of the UN Charter. The MOU is set out in full in the Court's Judgment because its interpretation was crucial to the case.Footnote ¹⁴ The question before the Court was whether this MOU was “an agreement to have recourse to some other method or methods of settlement” that excluded the Court's jurisdiction under Kenya's reservation.

Soon after its signature, the MOU gave rise to controversy in Somalia and it was rejected by the Transitional Federal Parliament on August 1, 2009. The prime minister of Somalia requested that the UN secretary-general treat the MOU “as non-actionable.”Footnote ¹⁵ Somalia then objected to the CLCS's consideration of Kenya's submission and initiated the current proceedings before the ICJ.Footnote ¹⁶ In explaining why it had acted in what might be seen as a precipitate manner, Somalia said that there was a danger Kenya would make a new reservation altering its Optional Clause acceptance in order to exclude the Court's jurisdiction over maritime boundary disputes (in the same way that the UK and Japan had taken action to exclude the Court's jurisdiction after cases were brought against them).Footnote ¹⁷ Somalia's suspicions were partly borne out by later developments. On January 24, 2017 Kenya lodged a new reservation under UNCLOS Article 298(1)(a)(i) opting out of Part XV compulsory dispute settlement procedures with respect to maritime boundary delimitations.Footnote ¹⁸

Somalia did not actually rely on the alleged invalidity of the MOU as a reason for rejecting Kenya's preliminary objections.Footnote ¹⁹ Instead, it argued that the MOU did not constitute an agreement on a method for settling maritime boundary disputes under Kenya's reservation. Nevertheless, Kenya devoted much of its written pleadings to an account of the MOU's background and status,Footnote ²⁰ and the Court said that it was appropriate to address “the issue [of] whether the MOU constitute[d] a treaty in force between the Parties.”Footnote ²¹ Obviously, if it was not in force, Kenya would not be able to argue that it established a method of settlement and thus deprived the Court of jurisdiction.

Somalia had put forward a series of unconvincing arguments on the invalidity of the MOU. These shifted over time and included claims that the MOU required ratification by its parliament and that its minister who had signed the MOU was not duly authorized to do so.Footnote ²² The Court applied customary international law to these arguments, because neither Somalia nor Kenya was a party to the Vienna Convention on the Law of Treaties (VCLT).Footnote ²³ It briefly and firmly dismissed Somalia's assertions and held that the minister who had signed the MOU had been given full powers to do so. There was no need for ratification: the MOU itself in its final paragraph had expressly provided that signature was enough. Nor was Somalia allowed to rely on any provisions of its own law requiring ratification; the Court reaffirmed the strict approach it had taken in Cameroon v. Nigeria on the rule codified in VCLT Article 46. It also relied on the rule set out in VCLT Article 45 that a state may not invoke a ground for invalidating a treaty if it must, by reason of its own conduct, be considered as having acquiesced in the validity of that treaty.

The Court then went on to the crucial question of the interpretation of the MOU, applying VCLT Articles 31 and 32 as “reflective of customary international law.”Footnote ²⁴ Did the MOU amount to an agreement to settle the dispute by other means so that Kenya's reservation deprived the ICJ of jurisdiction? By 13 votes to 3 the Court held that it did not. Its judgment reveals a fundamental difference of approach between the states parties on the appropriate method of treaty interpretation. For Kenya, the ordinary meaning of the text of the sixth paragraph of the MOU was decisive. In its written pleadings Kenya was surprisingly brief on this question,Footnote ²⁵ but it went into more detail in its oral pleadings.Footnote ²⁶ Paragraph six provides that

Kenya argued that this provided for a two-step procedure: first, the CLCS was to delineate the outer limits of the continental shelf; then the parties were to agree on delimitation of the whole of their maritime boundary. The parties had thus agreed to settle their dispute by negotiated agreement after receipt of the CLCS's recommendations. According to Kenya, the object and purpose of the MOU was to “organize the procedures for both delineation and delimitation.”Footnote ²⁷

In contrast, Somalia argued that paragraph six did not establish a method for the settlement of the boundary dispute. The object and purpose of the MOU was demonstrated in its long title, “to grant each other No-Objection in respect of submissions on the Outer Limits of the Continental Shelf Beyond 200 Nautical Miles to the Commission on the Limits of the Continental Shelf.” Moreover, the MOU only concerned the delineation of the outer continental shelf, and not any other zones.Footnote ²⁸

The Court adopted Somalia's approach. It held that the elements of VCLT Article 31 on treaty interpretation—ordinary meaning, context, and object and purpose—were to be considered as a whole. Paragraph six was “difficult to understand … without a prior analysis of the text of the MOU as a whole, which provide[d] the context … and [gave] insight into the object and purpose of the [treaty.]”Footnote ²⁹ The purpose of the MOU was to allow Somalia and Kenya each to make a submission on the outer limits of the continental shelf to the CLCS without objection from the other. None of the references to delimitation in the first five paragraphs supported Kenya's contention that the MOU provided a method for settling the dispute on the delimitation of the whole maritime boundary.Footnote ³⁰ Rather, they defined the delimitation dispute between the parties and made it clear that CLCS delineation was without prejudice to future delimitation of the maritime boundaries. As for paragraph six, it did not have the significance attributed to it by Kenya. The Court did not give decisive weight to the word “after.”Footnote ³¹ The Court's crucial finding was that although paragraph six referred to delimitation after the CLCS had made its recommendations, and this language suggested that the parties contemplated that delimitation would occur after the delineation of the outer limits of their continental shelves, “this [did] not necessarily mean that they intended to bind themselves to proceed to delimitation only in that way.”Footnote ³²

The Court also turned to UNCLOS Article 83(1), in accordance with the rule of treaty interpretation set out in VCLT Article 31(3)(c), which allows account to be taken of “any relevant rules of international law applicable in the relations between the parties.”Footnote ³³ Both Article 83 and paragraph six of the MOU provided for delimitation by agreement, and so the Court could read the latter in light of the former.Footnote ³⁴ It concluded that neither provision prescribed a method of dispute settlement. As Somalia pointed out, Kenya's arguments would have the dramatic effect of depriving the ICJ of a role in maritime delimitation between parties to UNCLOS.Footnote ³⁵

The Court said that Kenya itself had admitted that paragraph six did not preclude it from engaging in negotiations with Somalia on their maritime boundary before the CLCS had made its recommendations. Kenya had in fact engaged in negotiations concerning all maritime zones, including the continental shelf in 2014 before Somalia filed its Application. This engagement “confirme[d] that Kenya did not consider itself bound to wait for the [CLCS's] recommendations before [entering] into negotiations.”Footnote ³⁶ A lack of certainty about the outer limits of the shelf did not prevent the parties or the Court from undertaking the delimitation of the boundary before the CLCS had made its recommendations. “The parties could have reached an agreement on their maritime boundary at any time by mutual consent.”Footnote ³⁷

Kenya also invoked UNCLOS Part XV to make a second argument that its reservation excluded the Court's jurisdiction. It made only a very brief passing reference to this argument in its written pleadingsFootnote ³⁸ but developed it further in its oral pleadings.Footnote ³⁹ Kenya's argument was that UNCLOS Part XV provided an “agreement on a method of settlement” under Kenya's reservation. The Part XV scheme is a complex one and the Court went painstakingly through its provisions.Footnote ⁴⁰ Article 282 allows states to agree “through a general, regional or bilateral agreement or otherwise” to submit a dispute to a procedure entailing a binding result. All agreed that even though there was no express mention of the ICJ in this article, the term “or otherwise” allowed for agreement to the jurisdiction of the Court on the basis of Optional Clause declarations.Footnote ⁴¹ But Kenya argued that it had not in fact chosen the Court under Article 282 because the Optional Clause declarations of the parties did not coincide—owing to the Kenyan reservation. It claimed that both parties had therefore accepted arbitration under Article 287 as a default method of settlement. It is not surprising that Kenya's argument did not appeal to the Court because it would have the result that any case between UNCLOS parties in which there was a Kenyan-type reservation could not go to the ICJ.Footnote ⁴² The Court rejected Kenya's argument by 15 votes to 1, with only Judge Robinson (Jamaica) dissenting.

Finally, the Court considered Kenya's two arguments on admissibility and rejected these by a vote of 15–1. The first argument was quickly dismissed as it was essentially a reiteration of Kenya's claim about the interpretation of paragraph six of the MOU.Footnote ⁴³ The second was an argument that Somalia did not come to the Court with “clean hands” because it had violated the MOU when it withdrew its consent to CLCS consideration of Kenya's submission.Footnote ⁴⁴ The Court held that “the fact that an applicant may have breached a treaty at issue in a case does not per se affect the admissibility of its application.” Moreover, Somalia had not actually invoked the MOU to give jurisdiction to Court or as a source of substantive law. There was no need for the Court to go into the general doctrine of clean hands, an argument never accepted by the Court as undermining the admissibility of a case. Somalia's position that it is never bad faith to go to the Court proved more appealing.Footnote ⁴⁵ The fact that Kenya resisted the jurisdiction of the Court raised the question whether it was confident in the legality of its own maritime boundary claims and of the exploration licenses it had granted to oil and gas companies in the disputed area.

II. Alleged Violations of Maritime Rights and Sovereign Spaces in the Caribbean Sea (Nicaragua v. Colombia)

The Court issued an Order on the admissibility of counterclaims in Nicaragua v. Colombia.Footnote ⁴⁶ This was the latest stage in a series of interconnected cases between the two states. The complex judicial saga began in November 2012 when the ICJ issued a judgment determining the delimitation of the maritime boundaries between the two states.Footnote ⁴⁷ That judgment left unresolved the delimitation of the outer continental shelf. Nicaragua brought that additional issue to the ICJ in 2013 and on March 17, 2016 the Court found that it had jurisdiction to delimit the outer shelf.Footnote ⁴⁸ In 2013, Nicaragua also brought a new case, accusing Colombia of violating its rights in the maritime zones that the Court had declared to belong to Nicaragua in its November 2012 judgment. In 2016, at the Preliminary Objections stage, the Court held that it had jurisdiction under the Pact of Bogotá to decide on Nicaragua's application.Footnote ⁴⁹ In 2017 Colombia then brought four counterclaims. The Court held that two of these were admissible and two were inadmissible. Its reasoning—and that of the individual opinions—shows the uncertainty remaining about the rules governing counterclaims and the wide discretion the Court can exercise under the brief provisions of Article 80 of the Rules of Court on counterclaims. Article 80 provides that two requirements must be met for the Court to entertain a counterclaim: that it “comes within the jurisdiction of the Court” and that it is “directly connected with the subject-matter of the claim of the other party.” The meaning of this provision has to some extent been clarified in the Court's jurisprudence: it decided two applications for counterclaims in the 1950s, then there was a gap before it decided on several further applications since the late 1990s.Footnote ⁵⁰

In its Order on Colombia's counterclaims the Court addressed the question of direct connection first.Footnote ⁵¹ In its earlier decisions on counterclaims, it had identified “a range of factors that could establish a direct connection in fact and in law between a counter-claim and [the initial application].”Footnote ⁵² With respect to the connection in fact, the Court had in the past considered

In the current case, Colombia's first and second counterclaims were that Nicaragua had violated its duty of due diligence to protect and preserve the marine environment in the Southwestern Caribbean Sea, and to protect the rights of the inhabitants of the San Andrés Archipelago to benefit from a healthy, sound, and sustainable environment. The Court accepted that a majority of the incidents allegedly occurred in Nicaragua's exclusive economic zone (EEZ), and thus Colombia's first and second counterclaims related to the same geographical area as Nicaragua's principal claims.Footnote ⁵⁵ However, because the counterclaims were based on the actions of private Nicaraguan vessels engaged in predatory fishing and destroying the marine environment, whereas Nicaragua's claims were based upon actions by Colombia's navy allegedly interfering with Nicaragua's exclusive sovereign rights and jurisdiction in its EEZ, the Court took a strict view and held that the facts did not relate to the same factual complex.Footnote ⁵⁶ Furthermore, it said that the legal principles relied upon by the parties were different: Colombia invoked rules relating to the preservation and protection of the environment; Nicaragua referred to rules relating to sovereign rights, jurisdiction, and duties of a coastal state within its maritime area.Footnote ⁵⁷ Nor were the parties pursuing the same legal aim by their respective claims. “Colombia [sought] to establish that Nicaragua [had] failed to comply with its obligations to protect and preserve the maritime environment,” whereas Nicaragua argued that Colombia had violated its sovereign rights and jurisdiction within its maritime areas.Footnote ⁵⁸ Therefore, Colombia had not shown that its first two counterclaims met the condition of a direct connection, and the Court declared them inadmissible. Only Judge ad hoc Caron (Colombia) dissented on this point.

Colombia's third counterclaim was that Nicaragua's intimidation and harassment had infringed the customary fishing rights of the inhabitants of the San Andrés Archipelago to access and exploit their traditional fishing grounds.Footnote ⁵⁹ The Court observed that the facts relied on by the parties related to the same time period (following delivery of the 2012 Judgment) and the same geographical area (Nicaragua's EEZ). The claims of the parties were of the same nature insofar as they alleged similar types of conduct of the naval forces of one party against nationals of the other party. The respective claims of the parties concerned the scope of the rights and obligations of a coastal state in its EEZ; both parties were pursuing the same legal aim. Accordingly, there was “a direct legal connection between Colombia's third counterclaim and Nicaragua's principal claims.”Footnote ⁶⁰

More radically, Colombia's fourth counterclaim asked the Court to declare that Nicaragua's 2013 decree establishing straight baselines had the effect of extending its internal waters and maritime zones beyond what was permitted by international law and therefore that it violated Colombia's sovereign rights and jurisdiction. The Court held that there was a direct connection because both parties were complaining about provisions of domestic law with regard to the delineation of their maritime spaces in the same geographical area and in the same time period.Footnote ⁶¹ Both parties were alleging violations of sovereign rights under customary international law rules on the EEZ and contiguous zone, and both parties were pursuing the same legal aim since each was seeking a declaration that the other's decree was in violation of international law. Judge Greenwood dissented on this point and gave a compelling argument against the Court's position.Footnote ⁶² He agreed that there was a dispute between the parties regarding Nicaragua's decree establishing a system of straight baseline, but that dispute was entirely “separate and distinct” from the dispute that had given rise to the principal claim.Footnote ⁶³ The status of the area in which the incidents at the heart of Nicaragua's claim and Colombia's third counterclaim were said to have taken place would not be affected by any decision about Nicaragua's baselines. But it seems that the majority were keen to dispose of all possible remaining maritime boundary issues between the two parties to limit the prospect of yet more litigation between them.

In the second part of its Order, the Court considered whether Colombia's third and fourth counterclaims came within the jurisdiction of the Court.Footnote ⁶⁴ The parties disagreed as to the critical date for the determination of jurisdiction: was it enough that the Court had already asserted its jurisdiction under the Pact of Bogotá at the time Nicaragua's Application was made in 2013 (as Colombia claimedFootnote ⁶⁵), or was it necessary for the Court now to consider the question of its jurisdiction again, at the date the counterclaims were submitted (as Nicaragua arguedFootnote ⁶⁶)? The issue was critical, and the Court was divided. The question arose because Colombia had denounced the Pact of Bogotá, and it had ceased to be in force between the parties three years before Colombia brought its counterclaims. The Court invoked the reasoning in the Nottebohm case where it had famously held that the lapse of a party's Optional Clause declaration once proceedings had begun did not mean that the Court lost jurisdiction over the claim “with all its aspects.”Footnote ⁶⁷ The Court now used this in support of its determination that once it had established jurisdiction to entertain a case, it had jurisdiction over all its phases.Footnote ⁶⁸ Although counterclaims were autonomous legal acts, the object of which was to submit new claims to the Court, they were linked to the principal claims. The termination of the Pact of Bogotá as between the parties did not deprive the Court of its jurisdiction to entertain the counterclaims.Footnote ⁶⁹

Nevertheless, the Court accepted Nicaragua's argument that it still had to satisfy itself that the counterclaims met the conditions in the Pact of Bogotá. First, was there a dispute with regard to the subject matter of the third and fourth counterclaims?Footnote ⁷⁰ With regard to the third counterclaim, the Court held that a dispute had existed since November 2013 because Nicaragua was aware that its views were positively opposed by Colombia, thus reasserting without comment the controversial “awareness” test used in the Marshall Islands cases.Footnote ⁷¹ With regard to the fourth counterclaim, the Court considered that diplomatic correspondence showed that “the Parties held opposing views on the … delineation of their respective maritime spaces … following the Court's 2012 Judgment,” and that there was therefore a dispute.Footnote ⁷²

Second, was the matter one that could not in the opinion of the parties be settled by negotiations?Footnote ⁷³ The Court dealt with this in a summary fashion, as it had in its 2016 judgment establishing jurisdiction in the current case. It followed the test adopted there that “neither of the Parties could plausibly maintain that the dispute between them could be settled by direct negotiations through the usual diplomatic channels.”Footnote ⁷⁴ With regard to the third counterclaim, the Court simply asserted that the parties had never initiated direct negotiations following the 2012 Judgment. This alone indicated that they did not consider there was a possibility of finding a resolution of their dispute about traditional fishing rights through the usual diplomatic channels. By 11 votes to 5, it held the third counterclaim admissible. With regard to the fourth counterclaim, Nicaragua's adoption of its 2013 decree and Colombia's rejection of it by a diplomatic note of protest showed that it was no longer useful for the parties to engage in direct negotiations. The Court's reasoning here is very sketchy and not entirely convincing. By 9 votes to 7, the Court held the fourth counterclaim admissible. The fundamental division may be depicted as one between those judges keen to resolve all the issues between the parties and those taking a stricter line on the need for consent.

III. Application of the International Convention for the Suppression of the Financing of Terrorism and of the International Convention on the Elimination of All Forms of Racial Discrimination (Ukraine v. Russian Federation)

The Court decided on two applications for provisional measures. First, Ukraine v. Russia.Footnote ⁷⁵ This case raised many difficult issues, but the Court avoided them at the provisional measures stage or dealt with them only briefly. Ukraine instituted proceedings alleging violations of the 1999 International Convention for the Suppression of the Financing of Terrorism (ICSFT)Footnote ⁷⁶ in eastern Ukraine and of the 1965 International Convention on the Elimination of All Forms of Racial Discrimination (CERD) in Crimea.Footnote ⁷⁷ Russia argued convincingly that these were two separate claims, and that Ukraine's main focus was on the ongoing conflict between pro-government and pro-Russian groups in eastern Ukraine and the annexation of Crimea.Footnote ⁷⁸ The Court declared that it was well aware of the context of extensive fighting in eastern Ukraine and the extent of the human tragedy. At the end of its Order, it reminded the parties that the Security Council had endorsed the 2015 Minsk Agreements on a peace plan for eastern Ukraine and said that it expected them to work for the full implementation of the agreements in order to achieve a peaceful settlement of the conflict.Footnote ⁷⁹

On the same day that it filed its Application with the Court, Ukraine requested provisional measures to safeguard some of the rights it claimed under the two conventions. First, it asked the Court to order Russia not to aggravate or extend the dispute. Under the ICSFT, it also requested the Court to order Russia to: exercise appropriate control over its border to prevent further acts of financing of terrorism; halt and prevent all transfers of money, weapons, vehicles, equipment, training, or personnel from its territory; and take all measures at its disposal to ensure that any groups operating in Ukraine that have received transfers of money, weapons, etc. shall refrain from carrying out acts of terrorism against civilians in Ukraine. Under CERD, Ukraine requested the Court to order Russia to: refrain from any acts of racial discrimination; cease acts of political and cultural suppression against the Crimean Tatars, including suspending the ban on the Meijlis of the Crimean Tatar people;Footnote ⁸⁰ end the disappearances of Crimean Tatars; and cease political and cultural suppression against the ethnic Ukrainians in Crimea, including suspending restrictions on Ukrainian-language education.

The Ukraine v. Russia case was strongly reminiscent of the earlier Georgia v. Russia case in that both arose out of an armed conflict involving the parties, and in both the applicants tried to bring their claims within the scope of the specialized treaties under which the Court might have jurisdiction. In Georgia v. Russia, the Court had held, with only a brief explanation, that it had prima facie jurisdiction at the provisional measures stage.Footnote ⁸¹ But at the Preliminary Objections stage, a deeply divided Court held by 8 votes to 7 that it had no jurisdiction to decide on Georgia's claim of violations of CERD by Russia through its support for separatists carrying out ethnic cleansing of Georgians in South Ossetia and Abkhazia. The Court upheld Russia's objections to jurisdiction on the basis that the CERD requirements had not been met: there was no dispute under CERD until August 2008 because Georgia had not made specific claims under that Convention,Footnote ⁸² and so only those claims relating to actions after that date could be admissible, and moreover the preconditions stipulated by Article 22 CERD were not met.Footnote ⁸³

Ukraine's Application and oral pleadings in the current case showed that it had learned the lessons of Georgia v. Russia, to which it made frequent references.Footnote ⁸⁴ It took care to submit evidence proving that there was a dispute going back to spring 2014 and that the necessary procedural preconditions under the ICSFT and CERD had been met. Indeed Russia alleged that Ukraine had not engaged in consultations on the ICSFTin good faith, “but solely for the purpose of claiming to have exhausted the requirements of the dispute resolution procedure provided for by the Convention.”Footnote ⁸⁵ Ukraine's case under the ICSFT was that Russia had failed to take appropriate measures to prevent the provision of funds for the financing of terrorism in Ukraine, and that it had itself supplied weapons and other assistance to groups such as the Donetsk People's Republic that had carried out terrorist acts.Footnote ⁸⁶ Ukraine listed terrorist attacks such as the shelling of civilians and the shooting down of Malaysian Airlines Flight MH17. It took a wide view of the obligations on states parties under the ICSFT. First, it argued that “the provision of funds” under the ICSFT was an extremely broad term and that it incorporated the supply of weapons, equipment and personnel as well as the provision of money. Second, Ukraine interpreted the ICSFT as not only requiring states to regulate the private financing of terrorism but also prohibiting states themselves from financing terrorism. It relied on the Bosnia Genocide case to support its argument that if a treaty requires a state to prevent an act, this necessarily brings a duty on the state itself not to perform that act.Footnote ⁸⁷

Russia denied that there was any dispute between the parties as to the interpretation and application of the ICSFT. Ukraine had not shown that the acts of indiscriminate shelling and other violations of international humanitarian law fell within the definition of acts of terrorism under the Convention. Russia also challenged Ukraine's interpretation of the ICSFT. It said that the ICSFT obliges states to cooperate in the prevention and punishment of the financing by private actors of terrorist activities but that it does not cover matters of state responsibility for the financing of such activities by the state itself. In contrast to the Genocide Convention, the text of the ICSFT, its drafting history, and subsequent practice confirmed that it only addressed state obligations with respect to private actors rather than state responsibility for its own acts.Footnote ⁸⁸

The Court avoided these fundamental questions about the ICSFT, saying that it did not need to decide these problematic issues at this stage.Footnote ⁸⁹ It simply held that it was clear from the proceedings that the parties differed on the question whether events in eastern Ukraine since the spring of 2014 gave rise to issues relating to their rights and obligations under the ICSFT. It asserted that at least some of the allegations made by Ukraine appeared to be capable of falling within the provisions of the Convention, and the record was sufficient to establish prima facie the existence of a dispute concerning the interpretation and application of the ICSFT.

The Court then went on to examine whether the procedural preconditions in ICSFT Article 24 were satisfied: a dispute that “cannot be settled through negotiation within a reasonable time” shall be submitted to arbitration at the request of one of the parties and may be referred to the Court only if the parties are unable to agree on the organization of the arbitration within six months.Footnote ⁹⁰ Ukraine contended that it had made efforts to negotiate a resolution to the dispute, sending more than forty diplomatic notes and engaging in four rounds of bilateral negotiations and that it had suggested arbitration. More than six months had passed since then. Russia replied that Ukraine had not negotiated in good faith and had not made a good faith effort to set up an arbitral tribunal. Moreover, Ukraine had put forward allegations that went beyond the scope of the Convention and its notes included accusations regarding the prohibition of the use of force. These arguments were similar to those made by Russia in Georgia v. Russia, that the real dispute between the parties was about the use of force. Russia also denied that Ukraine had submitted concrete proposals for an arbitration agreement, saying that it had instead proposed resort to an ad hoc Chamber of the ICJ, and that this did not qualify as arbitration under Article 24.Footnote ⁹¹ Again, the Court did not go into this question in any detail. It found that issues relating to the ICSFThad been raised in contacts between the parties and this demonstrated that the parties had engaged in negotiations. It appeared from the facts on the record that these issues could not be resolved by negotiations, and that the parties were unable to agree on the organization of arbitration. Prima facie the procedural preconditions had been met.Footnote ⁹²

However, Ukraine failed at the next stage with regard to its claims under the ICSFT: it did not show that the rights it claimed were “plausible.”Footnote ⁹³ This requirement for the award of provisional measures was expressly introduced by the Court (without discussion) in Belgium v. Senegal Footnote ⁹⁴ and is now applied as a matter of course. But it remains obscure.Footnote ⁹⁵ This was the first time that the Court refused provisional measures on the ground that the rights in question were not plausible. Judges Owada (Japan) and Bhandari (India) and Judge ad hoc Pocar (Ukraine) rejected the Court's view that Ukraine's rights were not plausible. Judge Owada's understanding of this requirement was that it set a fairly low standard.Footnote ⁹⁶ For him the “‘so-called’ test of plausibility” was a shorthand term to refer to the condition that “a link must exist between the right whose protection is sought and the measures requested.” He argued that the term plausible was deliberately chosen to distinguish the test from the prima facie test for jurisdiction, and that it was a lower threshold, in order to avoid prejudging the merits. Judge ad hoc Pocar took a similar line. He warned that the current case might reinforce the risk that parties would over-address the merits of the case at the provisional measures stage in order to meet the plausibility requirement.Footnote ⁹⁷ Judge Bhandari's position was less clear.Footnote ⁹⁸ He set out his own view of plausibility on the basis of Judge Abraham's seminal Separate Opinion in Pulp Mills.Footnote ⁹⁹ Finally, Judge Cançado Trindade argued at some length that the proper test should be “human vulnerability” rather than plausibility of rights. He was not concerned with the danger of prejudgment.Footnote ¹⁰⁰

In its request for provisional measures, Ukraine said that it was seeking to protect its rights under Article 18: its right to Russia's cooperation in preventing the financing of terrorism, that is, “the provision or collection of funds with the intention that they should be used, or in the knowledge that they will be used, in order to carry out acts of terrorism, defined in Article 2.”Footnote ¹⁰¹ The parties disagreed on the classification of the incidents listed by Ukraine as acts of terrorism, and as to whether the applicable law was the ICSFTor international humanitarian law. But the Court did not have to decide these difficult questions. Russia's decisive argument was that there was no plausible allegation that it had financed terrorism because Article 2 was concerned solely with funds supplied with the knowledge or intention that they were to be used for acts of terrorism.Footnote ¹⁰²

The Court held that a state party may rely on Article 18 to require another party to cooperate with it in the prevention of certain types of acts only if it is plausible that such acts constitute offenses under Article 2. Ukraine had not provided evidence which provided “a sufficient basis to find it plausible” that the elements of intention or knowledge were present.Footnote ¹⁰³ This part of the Court's reasoning resembles that in the Legality of Use of Force cases.Footnote ¹⁰⁴ There Yugoslavia sued ten NATO states for the 1999 military campaign over Kosovo. It invoked the Genocide Convention as the basis for the Court's jurisdiction but the Court held that there was no prima facie jurisdiction as regards Yugoslavia's claim that the NATO airstrikes on its territory constituted genocide in violation of Article II of the Genocide Convention. Yugoslavia had not shown that the NATO states had the requisite intent to destroy all or part of an ethnic, racial, or religious group, and the Court was not prepared to infer this from their use of force.Footnote ¹⁰⁵ Similarly, in the current case it was not enough for Ukraine simply to assert that Russia had the necessary knowledge or intent under ICSFT Article 2. It must bring evidence.Footnote ¹⁰⁶ Ukraine had not addressed this issue in any depth in its pleadings.Footnote ¹⁰⁷ Accordingly, the Court did not indicate provisional measures under the ICSFT.Footnote ¹⁰⁸ As in the NATO cases, the applicant state's failure to bring evidence on questions of intent or knowledge allowed the Court to avoid a decision on a very sensitive issue.

In contrast, the Court did order provisional measures under CERD. Ukraine claimed that Russia had discriminated against Crimean Tatars and ethnic Ukrainians in Crimea in pursuit of the “cultural erasure” of non-Russian communities. First, in its consideration of prima facie jurisdiction, the Court held that there was a dispute.Footnote ¹⁰⁹ The parties differed on the question of whether the events in Crimea had given rise to issues relating to their rights and obligations under Articles 2 and 5 CERD.Footnote ¹¹⁰ The Court simply asserted without explanation that some of the acts referred to by Ukraine, in particular the banning of the Meijlis and alleged restrictions on the educational rights of ethnic Ukrainians, were capable of falling within the scope of CERD.

As for the procedural preconditions set out in CERD Article 22 that the dispute must be one “not settled by negotiation or by the procedures expressly provided for in this Convention,” Ukraine said it had made extensive efforts to negotiate a resolution through diplomatic notes and negotiations, and that it had become apparent that further negotiations would be futile.Footnote ¹¹¹ The Court referred to Georgia v. Russia for the proposition that the terms of Article 22 established preconditions to be fulfilled.Footnote ¹¹² Here the record showed the issues had not been resolved by negotiations.Footnote ¹¹³ The more difficult question was whether Ukraine was also obliged by Article 22 to attempt proceedings before the CERD Committee as a “procedure expressly provided for in this Convention.” Ukraine acknowledged that it had not brought the issues to the attention of the CERD Committee but argued that this was only an alternative precondition.Footnote ¹¹⁴ Russia argued that the two preconditions in Article 22 were cumulative, and therefore the ICJ had no jurisdiction. The Court held that “it need not make a pronouncement on the issue at this stage of the proceedings.”Footnote ¹¹⁵ This seems a little strange. It is true that the Court had earlier avoided this question in Georgia v. Russia but there it did not make any difference to the conclusion whether the two conditions were cumulative or alternative,Footnote ¹¹⁶ whereas here it did. If the preconditions were cumulative and Ukraine had not satisfied both of them because it had not gone to the Committee, then the Court would not have jurisdiction.

Next, the Court considered whether the rights for which Ukraine was seeking protection were plausible.Footnote ¹¹⁷ Was it plausible that the acts complained of constituted acts of racial discrimination? Russia argued that even if human rights violations had occurred, Ukraine had not proved that these actions were discriminatory: it had not proved a differentiation of treatment between those communities and other residents in Crimea. Nevertheless, the Court held that some of the acts fulfilled this condition of plausibility and that there was a link between the measures requested and the rights that were the subject of the main action.Footnote ¹¹⁸

Finally, with regard to the requirement that the applicant show that provisional measures are urgently required to prevent a danger of irreparable prejudice to the rights in dispute, Russia argued that Ukraine had not made any reference to this throughout the two years of “consultations.”Footnote ¹¹⁹ Nor had the CERD Committee triggered its urgent action procedure. Nevertheless, the Court held that certain rights—political, civil, economic, social, and cultural rights—were of such a nature that prejudice to them was capable of causing irreparable harm.Footnote ¹²⁰ The Crimean Tatars and ethnic Ukrainians in Crimea appeared to be vulnerable. The Court took account of a UN General Assembly resolution expressing serious concern over the banning of the Meijlis, and also of reports from the Office of the UN High Commissioner for Human Rights and the Organization for Security and Co-operation in Europe, which were critical of Russian behavior in Crimea. These reports showed the limits on the ability of Tatars to choose representative institutions, and restrictions on the availability of Ukrainian-language education in Crimean schools. Thus, there was an imminent risk of irreparable harm. The Court ordered provisional measures under CERD by a vote of 13–3. Judges Tomka (Slovakia), Xue (China), and Skotnikov (Russia) dissented; these judges had also dissented in Georgia v. Russia.Footnote ¹²¹

In the current case, the Court used the power under Article 75 of its Rules to indicate measures “in whole or in part other than those requested,” as it had done in Georgia v. Russia where it ordered provisional measures to be taken by both parties, not just by Russia, and not in the terms Georgia had requested.Footnote ¹²² The Court now ordered Russia to suspend the ban on the Meijlis and to ensure the availability of education in the Ukrainian language. It also indicated to both parties—not just to Russia—that “they shall refrain from any action which might aggravate or extend the dispute before the Court or make it more difficult to resolve.” This last type of measure is common in the practice of the Court, but has been the cause of some controversy in the past.Footnote ¹²³

IV. Jadhav Case (India v. Pakistan)

The other provisional measures case was more straightforward and followed the same approach as three earlier death penalty cases against the United States. India filed an Application on May 8, 2017 alleging that Pakistan had violated Article 36 of the 1963 Vienna Convention on Consular Relations (VCCR) during its detention and trial of an Indian national, Kulbhushan Jadhav, who was sentenced to death in Pakistan. On the same day India also requested the Court to order provisional measures that Pakistan should: first, take all measures necessary to ensure that Jadhav was not executed; second, report its action to the Court; and third, ensure that no action was taken that might prejudice the rights of India or Jadhav with respect to any decision on the merits of the case.Footnote ¹²⁴

Jadhav had been detained by the Pakistani authorities on March 3, 2016 on suspicion of espionage and terrorism. India was informed of his arrest on March 25 and repeatedly requested consular access to no avail. Pakistan formally requested India's assistance in the investigation process and informed India that consular access would be considered in light of India's response to this request.Footnote ¹²⁵ Jadhav was sentenced to death for espionage, sabotage and terrorism on April 14.Footnote ¹²⁶ The Court first considered whether it had prima facie jurisdiction. India and Pakistan were both parties to the VCCR and to its Optional Protocol, which provided for the compulsory jurisdiction of the ICJ over disputes arising out of the interpretation or application of the Convention. However, Pakistan challenged the Court's jurisdiction.Footnote ¹²⁷ Its arguments were radical and disturbing.Footnote ¹²⁸ It claimed that there was no dispute about the interpretation or application of the VCCR because it did not apply to those suspected of espionage or terrorism. It also argued that a 2008 bilateral Agreement on Consular Access between India and Pakistan limited the protection provided by the VCCR. Pakistan relied on VCCR Article 73(2), which provides that “nothing in the present Convention shall preclude states from concluding international agreements confirming or supplementing or extending or amplifying” its provisions. It interpreted this as allowing subsequent agreements that limited the protections of the VCCR. The 2008 Agreement provided that “in case of arrest, detention or sentence made on political or security grounds, each side may examine the case on its merits.” Pakistan claimed that this provision applied to Jadhav and meant that the Court lacked jurisdiction under the Optional Protocol.

The Court did not enter into these controversial questions. It simply noted that the parties appeared to have differed on the question of India's consular assistance to Jadhav.Footnote ¹²⁹ India had maintained that Jadhav should have been afforded consular assistance under the VCCR; Pakistan had stated that such assistance would be considered in the light of India's response to its request for assistance in its investigation. Those elements were sufficient to establish that a dispute existed at the date of the Application with regard to the arrest, detention, trial, and sentencing of Jadhav. Moreover, the dispute was one over which the Court might have subject matter jurisdiction under the Optional Protocol because Pakistan's alleged failures were capable of falling within the scope of the VCCR.Footnote ¹³⁰ As for Pakistan's controversial arguments, the Court noted that the VCCR “does not contain express provisions excluding from its protection persons suspected of espionage or terrorism,” so at this stage it could not be concluded that the VCCR did not apply in the case of Jadhav.Footnote ¹³¹ It did not need to decide whether Article 73 allowed a bilateral agreement to limit the rights under the VCCR. It was sufficient to note that the 2008 Agreement did not expressly impose such a limitation, and therefore there was no basis to conclude that the 2008 Agreement prevented it from exercising jurisdiction under the Optional Protocol over disputes about Article 36 VCCR.Footnote ¹³²

The Court then considered whether the rights asserted by India were plausible and held that they were.Footnote ¹³³ Pakistan's arguments that Article 36 did not apply to persons suspected of espionage or terrorism and that the 2008 Agreement governed the situation did not mean that the rights claimed by India were not plausible. No legal analysis had been advanced on these points by the parties. The Court also found without difficulty that there was a link between the rights pursuant to which protection was sought under the VCCR and the provisional measures requested to prevent Jadhav from being executed before the Court's final decision.Footnote ¹³⁴

Provisional measures were urgently required because there was a real and imminent risk of irreparable prejudice to the rights in dispute. Pakistan claimed that there was no urgency because Jadhav had up to 150 days to apply for clemency. There would be no execution before the end of August 2017. It said that it was not in its interests to “stop the canary from singing.”Footnote ¹³⁵ But the Court held that the mere fact that there was a death sentence, and that Jadhav might be executed, was sufficient to demonstrate the existence of such a risk. There was considerable uncertainty as to when an appeal could be heard. Pakistan had given no assurance that Jadhav would not be executed before the ICJ's final decision. The facts that Jadhav could petition for clemency and that the date of his execution had not been fixed did not preclude the Court from indicating provisional measures.Footnote ¹³⁶ At first sight, this seemed to diverge from the ICJ's decision in the Avena case.Footnote ¹³⁷ There the Court had indicated provisional measures only in respect of those individuals whose executions had actually been scheduled. But, as Judge Bhandari pointed out, the facts and circumstances of this case were different: in the United States, execution dates are communicated to the public, generally with several weeks of notice. In Pakistan, it is unclear whether this would be done.Footnote ¹³⁸ The Court ordered unanimously that Pakistan “shall take all measures at its disposal to ensure that Jadhav is not executed pending the final decision in these proceedings.” It reaffirmed that its orders were binding, following its decision in LaGrand (2001). Thus, without discussion, the Court indicated only the first of the three measures requested by India. Although it did not comment on Pakistan's controversial arguments limiting the consular protection of foreign nationals suspected of espionage or terrorism, the signs are that they will not appeal to the Court if and when it comes to decide on the merits of the case.