On Friday the European Court of Human Rights delivered its Grand Chamber judgment in Catan and Others v. Moldova and Russia, nos. 43370/04, 8252/05 and 18454/06, yet another case on the ECHR’s extraterritorial application, dealing in particular with the Convention’s application to the separatist republic of Trandniestria in Moldova (link to judgment). The case is in effect a sequel to the Court’s earlier judgments on Transdniestria in Ilascu and Ivantoc, this time dealing however with a significantly different factual pattern.
The applicants were Moldovans who lived in Transdniestria and who were at the time of lodging the application pupils at three Moldovan-language schools and their parents. They complained under Article 2 of Protocol No. 1 to the Convention and Article 8 of the Convention, taken alone and in conjunction with Article 14 about the closure of their schools and their harassment by the separatist Transdniestrian authorities. The reason for this harassment was basically a policy of Russification by the Transdniestrian authorities whereby schools in the region could only operate in and teach the Moldovan (i.e. Romanian) language as written in the Cyrillic alphabet, rather than the much more commonly used Latin one. In short, the applicants’ education became embroilled in language politics, very similar for instance to those in the Balkans.
What makes this case particularly interesting is the relationship between Article 1 ECHR notion of state jurisdiction, as the threshold for the existence of (all or some) state obligations under the Convention, and the attribution of conduct under the secondary rules of the law of state responsibility. In Ilascu, paras 392-3, the Court held that
[T]he “MRT” [Transdniestria], set up in 1991-92 with the support of the Russian Federation, vested with organs of power and its own administration, remains under the effective authority, or at the very least under the decisive influence, of the Russian Federation, and in any event that it survives by virtue of the military, economic, financial and political support given to it by the Russian Federation. … [T]here is a continuous and uninterrupted link of responsibility on the part of the Russian Federation for the applicants’ fate, as the Russian Federation’s policy of support for the regime and collaboration with it continued beyond 5 May 1998, and after that date the Russian Federation made no attempt to put an end to the applicants’ situation brought about by its agents, and did not act to prevent the violations allegedly committed after 5 May 1998.
Ilascu was notable for several reasons. First, it apparently applied the spatial model of Article 1 jurisdiction as control of an area while lowering the threshold of the needed control (the ‘decisive influence’ bit). Secondly, it completely confused jurisdiction with responsibility; it was utterly unclear from the case whether the Court considered all acts of the MRT to be attributable to Russia, apparently on the basis of a sui generis rule on attribution of conduct that hardly seemed compliant with the ILC’s work on state responsibility or the jurisprudence of the ICJ, or rather whether Russia was held responsible for failing to comply with a positive obligation to prevent human rights violations by non-state actors (the MRT) operating in an area under its jurisdiction. Third, the Court also found that Moldova had positive obligations in the MRT despite having lost control of the territory, a (human rights-friendly) ruling that in my view compromised the purely factual nature of the Art 1 jurisdictional tests for the sake of a rather vague positive obligation which did not amount to much in practice anyway.
Here comes Catan, which provided the Court with the opportunity to revisit some of these points. What distinguishes Catan and Ilascu is primarily the lapse in time with regard to the facts of the two cases, during which Russia’s control over Transdniestria arguably decreased. Moreover, unlike in Ilascu Russian authorities had no involvement in the harassment of the applicants and the interference with their right to education. The Court thus had to build upon Ilascu, and that it did, producing a rather mixed (if again human rights-friendly) outcome. In brief, it found that both Moldova and Russia retained jurisdiction over Transdniestria; that Moldova this time did comply with its positive obligations; but that Russia was to be held reponsible for a violation of Art 2 of Protocol 1, and was as a consequence liable for significant damages.
The Court first proceeds to recapitulate its case law on extraterritorial application (para. 103 ff). Somewhat amusingly, it states that it ‘has established a number of clear [!] principles in its case-law under Article 1.’ Paraphrasing Al-Skeini, it considers that ‘“Jurisdiction” under Article 1 is a threshold criterion. The exercise of jurisdiction is a necessary condition for a Contracting State to be able to be held responsible for acts or omissions imputable to it which give rise to an allegation of the infringement of rights and freedoms set forth in the Convention.’ Note how this clearly distinguishes between jurisdiction and attribution, and so far so good.
It then affirms the Ilascu principle that title over territory alone is grounds for a limited type of Art. 1 jurisdiction (para. 110): ‘Although Moldova has no effective control over the acts of the “MRT” in Transdniestria, the fact that the region is recognised under public international law as part of Moldova’s territory gives rise to an obligation, under Article 1 of the Convention, to use all legal and diplomatic means available to it to continue to guarantee the enjoyment of the rights and freedoms defined in the Convention to those living there.’
As for Russia, the Court considers it crucial that the events at issue in Catan took place in 2002-4, a period included, if somewhat peripherally, in its ruling in Ilascu (para. 111). Therefore, ‘where the Court has already concluded that the Russian Federation had jurisdiction over certain events in Transdniestria during the relevant period, it considers that the burden now lies on the Russian Government to establish that Russia did not exercise jurisdiction in relation to the events complained of by the present applicants’ (para. 112). This pretty deft reversal of the burden of proof allowed the Court to simplify the evolution of Russia’s involvement in the MRT. The Court similarly makes it clear that it is applying a spatial, rather than personal conception of Art. 1 jurisdiction (para. 114), and that unlike in Al-Skeini there is no evidence of a direct involvement of Russian agents in the acts against the applicants’ schools.
The Court then had to respond to a rather powerful argument made by Russia, namely that its approach in Ilascu was at odds with the work of the ILC and the ICJ’s Nicaragua and Bosnian Genocide rulings, in which it found that the acts of a prima facie non-state actor could become attributable to a state if it (1) either exercises complete control over the non-state actor, which is totally dependent on it, so that all of its acts are those of the state, of which it is essentially a de facto organ; or (2) effective control over the specific conduct in question. (For more discussion on all this see pp. 41-53 of my book on extraterritorial application; see also here). To this the Court responds that (para. 115):
The Court recalls that in the judgment relied upon by the Government of the Russian Federation, the International Court of Justice was concerned with determining when the conduct of a person or group of persons could be attributed to a State, so that the State could be held responsible under international law in respect of that conduct. In the instant case, however, the Court is concerned with a different question, namely whether facts complained of by an applicant fell within the jurisdiction of a respondent State within the meaning of Article 1 of the Convention. As the summary of the Court’s case-law set out above demonstrates, the test for establishing the existence of “jurisdiction” under Article 1 of the Convention has never been equated with the test for establishing a State’s responsibility for an internationally wrongful act under international law.
Aside from the rather tiresome refusal of the Court to acknowledge any inconsistency in its prior case law (viz. jurisdiction ‘has never been equated’ with attribution, when nobody who ever read Ilascu was sure what exactly the Court was doing, and I won’t even mention other such cases), this is a welcome development. As in Al-Skeini, the Court distinguishes between jurisdiction and attribution, and this is a good thing. It then proceeds to find that on the facts of the case Russia maintained jurisdiction over Transdniestria (paras. 120-121):
As mentioned above, the Court in the Ilaşcu judgment also found that the “MRT” only survived during the period in question by virtue of Russia’s economic support, inter alia (see paragraph 111 above). The Court does not consider that the Russian Government have discharged the burden of proof upon them and established that this finding was incorrect. In particular, it is not denied by the Russian Government that the Russian public corporation Gazprom supplied gas to the region and that the “MRT” paid for only a tiny fraction of the gas consumed, both by individual households and by the large industrial complexes established in Transdniestria, many of them found by the Court to be Russian-owned (see paragraphs 39-40 above). The Russian Government accepts that it spends USD millions every year in the form of humanitarian aid to the population of Transdniestria, including the payment of old age pensions and financial assistance to schools, hospitals and prisons. In the light of the statistic, supplied by the Moldovan Government and undisputed by the Russian Government, that only approximately 20% of the population of the “MRT” are economically active, the importance for the local economy of Russian pensions and other aid can be better appreciated. Finally, the Court notes that the Russian Government do not take issue with the applicants’ statistics regarding nationality, according to which almost one fifth of those living in the region controlled by the “MRT” have been granted Russian nationality (see paragraphs 41-42 above).
In summary, therefore, the Russian Government have not persuaded the Court that the conclusions it reached in 2004 in the Ilaşcu judgment (cited above) were inaccurate. The “MRT” was established as a result of Russian military assistance. The continued Russian military and armaments presence in the region sent a strong signal, to the “MRT” leaders, the Moldovan Government and international observers, of Russia’s continued military support for the separatists. In addition, the population were dependent on free or highly subsidised gas supplies, pensions and other financial aid from Russia.
The Court then went on to examine the merits of the applicants’ complaint under Article 2 of Protocol No. 1, and I will deal with this only very briefly, even though its analysis of the right to education merits a more detailed examinatiom. The Court found that the sole purpose of the interference with the applicants education by the MRT authorities was ‘entrenching the separatist ideology’ (para. 144), and accordingly found a violation on the facts. It however saw no need to venture further into discussing the case under Articles 8 and 14 of the Convention, with a number of judges jointly dissenting on this pont.
What I find most intriguing here is the Court’s approach to state responsibility. While it found that Moldova had discharged its (paltry) positive obligation ‘having made considerable effort to support the applicants’ (para. 147), with regard to Russia it ruled as follows (paras. 149-150):
The Court notes that there is no evidence of any direct participation by Russian agents in the measures taken against the applicants. Nor is there any evidence of Russian involvement in or approbation for the “MRT”‘s language policy in general. Indeed, it was through efforts made by Russian mediators, acting together with mediators from Ukraine and the OSCE, that the “MRT” authorities permitted the schools to reopen as “foreign institutions of private education” (see paragraphs 49, 56 and 66 above).
Nonetheless, the Court has established that Russia exercised effective control over the “MRT” during the period in question. In the light of this conclusion, and in accordance with the Court’s case-law, it is not necessary to determine whether or not Russia exercised detailed control over the policies and actions of the subordinate local administration (see paragraph 106 above). By virtue of its continued military, economic and political support for the “MRT”, which could not otherwise survive, Russia incurs responsibility under the Convention for the violation of the applicants’ rights to education. In conclusion, the Court holds that there has been a violation of Article 2 of Protocol No. 1 to the Convention in respect of the Russian Federation.
Note how this brings us back to attribution. Is the Court here saying that Russia was responsible for everything that the MRT did, i.e. that all of its acts were attributable to Russia, by virtue of some ECHR-specific rule of attribution? Not only would this go against what the ILC and the ICJ had to say on the matter, but this would also contradict the earlier passages in Catan in which the Court draws the distinction between jurisdiction and responsibility. Or is the Court here saying (as I would argue is the much better view) that Russia is responsible for failing to prevent the acts of the MRT that violated the applicants’ rights? But if that is so, why does the Court not even mention positive obligations with regard to Russia, nor judge it according to a due diligence standard, through which the efforts made by Russian mediators could certainly be taken into account? In effect, the Court would appear to have treated this case in exactly the same way as if Russian authorities were directly involved in the closing of the schools, and that just does not seem right to me.
To conclude, while the Court’s efforts to clarify its Article 1 case law continue, and while parts of Catan are on the right track (with a 16 to 1 majority in the Grand Chamber on top), ultimately we again have a judgment that is conceptually open to various interpretations. And that is a real problem. It is one thing to agree or disagree with the Court on its assessments of the facts, and specifically as how the reality on the ground changed (or not) between Ilascu and Catan. But it quite another not to know (yet again!) what the Court is exactly doing and how its methods figure into the broader framework of international law, besides reaching a result that it thought appropriate under the circumstances.