DAYS after the Russian invasion began, Ukraine instituted proceedings before the International Court of Justice (ICJ) under the 1948 Genocide Convention (GC), arguing that Russia cannot lawfully rely on false claims of genocide to justify military action in Ukrainian territory. Russia did not appear at the oral proceedings but sent a document asserting the ICJ's lack of jurisdiction. On 16 March 2022, the ICJ delivered its Order on Provisional Measures, akin to an injunction in national law. A 13:2 majority concluded that the court had prima facie jurisdiction to rule on the merits of the dispute, and ordered Russia to suspend its military operations, and to ensure that persons directed, supported, and/or controlled by Russia cease military operations (Judge Xue and Vice-President Gevorgian dissenting in part). By a unanimous vote, the court also required both states to refrain from actions that might aggravate, extend or otherwise make the dispute difficult to resolve. The court's application of the law on provisional measures is relatively straightforward. What is remarkable is that, so far at least, Ukraine has been successful in converting false claims that it has breached international law into a basis for the court's jurisdiction – a feat which raises important and difficult questions about the potential broader applicability of such jurisdictional manoeuvres, and the standards by which the court should assess whether states are acting in good faith in cases that are less clear cut than the present one.
The court may indicate provisional measures only if the provisions relied upon by the applicant afford a prima facie basis on which jurisdiction to decide the dispute could be founded but it “need not satisfy itself in a definitive manner” (at [24]), leaving open the possibility that it may reach a different conclusion in the merits phase. The court accepted Ukraine's characterisation of the disagreement as being whether Ukraine is committing genocide and whether Russia's unilateral assertion that it is provides a lawful basis for military action to prevent and punish that genocide pursuant to Article I GC. This was therefore a dispute “relating to the interpretation, application or fulfilment” of the GC, over which the court can exercise jurisdiction by the terms of the compromissory clause in Article IX. In its document to the court, Russia argued that it was not relying on the GC to justify its military operation but was instead acting in self-defence, and references to genocide in President Putin's speech were “not equal to the invocation of the [GC] or the existence of a dispute under it”. However, the court appeared unconvinced, noting the circumstances in which Russia had claimed that Ukraine was committing genocide contrary to international law and had stated that the purpose of its military action was to prevent and punish genocide (at [37]–[41]). It also confirmed that states need only refer to the “subject matter of the treaty with sufficient clarity” to enable the other party to determine that there is a dispute about that treaty, and that in the present case there is prima facie sufficient evidence to satisfy this condition (at [44]–[47]). Given that genocide as a legal concept is bounded by the definition in the GC, which is one of the most important instruments of the international legal order (and one to which both Ukraine and Russia are parties) and there are statements by Russian officials that appear to refer to aspects of the Convention, it is hard to see how Russia could not have had the Convention in mind when it accused Ukraine of genocide.
After finding that it had prima facie jurisdiction to hear the case on the merits, the court turned to the other conditions for granting provisional measures: that the rights claimed are at least plausible; that there is a link between them and the requested measures; and that, without provisional measures, there is a real and imminent risk that prejudice would be caused. The majority agreed that Article I GC must be implemented in “good faith, taking into account other parts of the Convention” (including the preamble, Art. VIII on seeking assistance from UN organs and Art. IX on dispute resolution) and that, in discharging the duty to prevent and punish genocide, “State[s] may only act within the limits permitted by international law … [and] in conformity with the spirit and aims of the UN” (at [56]–[58]). This good faith condition is crucial for establishing jurisdiction because it forecloses the possibility of arguing (in a style reminiscent of the Kosovo Advisory Opinion, ICJ Reports 2010, p. 403) that Russia's military operation is at least in accordance with the GC insofar as it does not prohibit force, even if it also does not confer a right to use force. The court made it clear that the GC is not an island entire of itself; rather, it must be interpreted in harmony with the UN Charter and customary international law, both of which regulate the use of force. In a political win for Ukraine, the court explained that it is “not in possession of evidence substantiating the allegation of the Russian Federation that genocide has been committed on Ukrainian territory” and “it is doubtful that the Convention, in light of its object and purpose, authorises a Contacting Party's unilateral use of force in the territory of another state for the purpose of preventing or punishing an alleged genocide” (at [59], emphasis added). The upshot is that Ukraine has at least a plausible right not to be subject to military operations for the purpose of preventing and punishing an alleged genocide in Ukraine (at [60]). To an extent, the court's reasoning helps clarify its earlier decision in Bosnia and Herzegovina v Serbia and Montenegro, where it held that the positive obligation under the GC to prevent genocide is one of due diligence, which is determined in part by the state's capacity to influence effectively the actions of those committing genocide. Crucially, a state “may only act [to prevent genocide] within the limits permitted by international law” (ICJ Reports 2007, p. 43, at [430]). Given this, it is notable that the court in Ukraine says only that it is “doubtful” that the GC authorises unilateral force in the territory of another state to prevent or punish an alleged genocide (perhaps it did not want to appear to have predetermined the issue at the provisional measures stage).
The majority were satisfied that there was a link between the rights claimed and the provisional measures requested – that is, for Russia, and those it directs, controls, influences and/or supports, to cease military operations for the purpose of preventing or punishing genocide (at [64]). If such measures were not granted, there was a real and imminent risk that prejudice would be caused before the court is able to reach a decision on the merits (at [66]–[77]). The full court ordered both states to refrain from action that might aggravate or extend the dispute. This somewhat striking concession was the subject of strong criticism by Judges Daudet and Robinson, with the former noting that the UN General Assembly had designated Russia the aggressor and expressing concern that Ukrainians fighting for survival might find this measure insulting (at [6]). It is difficult to justify the court's position here, but perhaps this measure was designed to ensure that at least part of the Order received a unanimous vote. In this respect, Judge Xue, reserving her position, “wonder[ed]” how provisional measures “can be meaningfully and effectively implemented by only one Party to the conflict” (at [6]). The court also decided not to require Russia to provide a report on implementing the provisional measures (at [83]), likely because it was concerned about creating a potentially unhelpful precedent and because it knew that Russia would not comply anyway.
At first sight, Ukraine is similar to so-called “round-peg-square-hole” cases in which the applicant “creatively” describes facts to fit a claim within a compromissory clause. In Georgia v Russia, ICJ Reports 2011, p. 70, Georgia argued that in August 2008 Russia practised, sponsored and supported racial discrimination through its use of force in the South Ossetia and Abkhazia regions. A divided court accepted that it had prima facie jurisdiction under Article 22 of the Convention for the Elimination of All Forms of Racial Discrimination (CERD) at the provisional measures stage but later concluded that the preconditions for establishing jurisdiction under that provision – namely, negotiation with respect to the dispute over the interpretation or application of CERD and/or use of the dispute procedures in the treaty – had not been satisfied at the relevant time, thereby foreclosing jurisdiction. The court thus applied the textual conditions in the compromissory clause strictly, rather than concluding that the dispute was really one that fell outside the scope of CERD. For some, requiring negotiations between warring parties to deal specifically with CERD was somewhat myopic and disappointingly formalistic. For others, it was a signal that the court does not look kindly on parties who seek to squeeze apparently unrelated disputes into jurisdictional clauses in order to attract international attention. This anxiety is reflected in the opinions of some of the judges in Ukraine: Judge Bennouna warned that “artificially linking a dispute concerning the unlawful use of force to the Genocide Convention does nothing to strengthen that instrument” and Judge Xue considered that the relevant acts do “not appear to the capable of falling within the scope of the [GC]”. Vice-President Gevorgian considered that Ukraine was attempting to circumvent this difficulty altogether by claiming that the Convention “embodies a right ‘not to be subjected to another States’ military operations on its territory based on a brazen abuse of Article I [GC]’”. For him, this manoeuvre undermines state consent and raises the possibility that “any purportedly illegal act…could be shoehorned into a random treaty as long the as the subject-matter regulated by the treaty had some role in the political considerations preceding the respective act” (at [7], emphasis in original). Arguably, he misses the point: Russia justified its military operation by reference to false claims of genocide and thus the GC is not a “random” treaty into which Ukraine's claim is being “shoehorned”. However, he is right that Ukraine is not really a case in which the facts are being re-described to fit a compromissory clause – rather, the case is about whether the obligation in Article I GC is bounded by the international law on the use of force, so that Russia is precluded from relying on the GC to justify the lawfulness of its military action. Hence it will be crucial at the merits phrase to substantiate the argument that Article I GC must be carried out in good faith and in accordance with general international law (of course this is not the first time that the court has interpreted a treaty provision in this way: Oil Platforms (Islamic Republic of Iran v United States of America ICJ Reports 2003, p. 161)).
Some judges relied on Legality of the Use of Force cases (in particular, Order of 2 June 1999, ICJ Reports 1999, p. 124), where the court found it had no jurisdiction to consider whether states had committed genocide through their use of force, to suggest an apparent contradiction between that case and the present one. However, as Judge Nolte argued in his Declaration, Legality centred on an allegation that states engaging in the use of force were committing genocide through their use of force. In contrast, the claim in Ukraine is whether actions taken purportedly to prevent and punish genocide were lawful under the GC. Robinson explicitly linked this to the question of false claims of international law, pointing out that Ukraine has not asked the court a general question about Russia's force; rather, it is arguing that Russia cannot lawfully rely on a false claim to act to prevent genocide under the Convention.
The court's Order is a clear win for Ukraine, even though Russia is highly unlikely to comply. The way in which Ukraine relies on the idea of falsity to characterise its rights – the right “not to be subject to a false claim of genocide” (at [52]) and for military action not to be “launched on a pretext of genocide” (at [68]) – is perhaps the most remarkable aspect of the case. Although the court stops short of directly embracing the idea of a right not to be subject to false claims of international law, it does conclude that Ukraine has “a plausible right not to be subject to military operations by the Russian Federation for the purpose of preventing and punishing an alleged genocide in the territory of Ukraine” (at [60]). While jurisdiction is clearly based on the existence of a dispute as to whether genocide is taking place and what actions Russia may lawfully take in response, the court nevertheless signalled that Article I GC must be carried out in good faith and in a way that is compatible with general international law. At the merits stage, it may have to grapple with what this means in practice, including what it means to make “bad faith” and pre-textual claims about law and fact as opposed to making claims that are plausible but wrong. Drawing these distinctions and characterising the impact that deception has on the validity of international legal arguments is a perennial challenge for international law, as too many past examples of unilateral force have shown.
