The Nagorno-Karabakh Cases

Last week I wrote about one particular aspect of the recent Grand Chamber judgments of the European Court of Human Rights in two cases dealing with the aftermath of the Nagorno-Karabakh conflict: Chiragov and Others v. Armenia, no. 13216/05 and Sargsyan v. Azerbaijan, no. 40167/06, namely the Court’s conclusion that belligerent occupation necessarily requires troops on the ground. I also promised a more comprehensive look at the two (very important) judgments, and here it is. The two cases concerned the aftermath in the conflict, in the sense that they dealt with the right of persons displaced by the conflict to access their property (under Article 1 of Protocol 1 to the Convention), rather than with the conflict itself, which was outside the Court’s temporal jurisdiction. That said, there are numerous noteworthy aspects of these two judgments.

First, there is the cases’ basic structure. Both cases were brought by individuals, but there are more than a thousand other applications pending before the Court with essentially the same issues. While these are formally not pilot judgments in the sense the Court uses the term, they are in fact test cases on the basis of which the Court is set to resolve all of the other pending cases, unless the parties choose to settle them first. And while the cases were brought by individuals, they have a strong interstate dimension, not only because of their politically controversial subject-matter, but because Armenia and Azerbaijan both intervened as third parties in the case in which the other state was the respondent (i.e. Armenia intervened in Sargasyan and Azerbaijan in Chiragov). These were, if you will, interstate cases by proxy.

Second, the Court took a very relaxed approach to the question of non-exhaustion of effective domestic remedies for admissibility purposes, finding rather summarily that no such remedies existed. Chiragov, para. 119 in particular is worth quoting in full:

It should also be noted that the Republic of Armenia has denied that their armed forces or other authorities have been involved in the events giving rise to the complaints in the present case or that Armenia exercises – or have at any point in time exercised – jurisdiction over Nagorno-Karabakh and the surrounding territories. Given such a denial of involvement or jurisdiction, it would not be reasonable to expect the applicants to bring claims for restitution or compensation before the Armenian courts and authorities. Regard must further be had to the political and general context. As a consequence of the war, virtually all Azerbaijanis have left the disputed territories. No political solution of the conflict has been reached. Rather, the hostile rhetoric between the leaders of Armenia and Azerbaijan appear to have intensified, ceasefire breaches are recurrent and the military build-up in the region has escalated in recent years. In these circumstances, it is not realistic that any possible remedy in the unrecognised “NKR” entity in practice could afford displaced Azerbaijanis effective redress.

Emphasis mine; similarly, see Sargsyan, paras. 117-119, where the Court finds the Azerbaijan did not discharge its burden of showing that effective remedies specific to the applicants’ situation existed domestically. The flexible attitude that the Court has shown here, particularly in the two italicized sentences, has clear implications for other similar (and also politically very controversial) cases currently pending before it, including interstate cases between Georgia and Russia and Ukraine and Russia, and many individual applications brought against these states with regard to the relevant conflicts. It now seems significantly less likely that the Court will use non-exhaustion as a vehicle for avoiding to deal with such cases. Accordingly, many, if not most of them will proceed to the merits.

Third, there is the Court’s approach to the Convention’s extraterritorial application in Chiragov. Recall the two basic conceptions of the notion of state jurisdiction in Article 1 of the Convention, as either effective control of an area (spatial model of jurisdiction) or as authority and control over an individual by a state’s agents (personal model of jurisdiction). In Chiragov the Court (quite rightly) decided to look at the question of jurisdiction from a spatial perspective; the fundamental issue in dispute was that while the applicants claimed that Armenia exercised effective overall control, and thus jurisdiction, over Nagorno-Karabakh, Armenia claimed that the ‘Nagorno-Karabakh Republic’ was an independent entity which it did not control, even though it did provide it with some support, but without the stationing of any Armenian troops in NK. The Court found against Armenia (Chiragov, paras. 168 ff), finding that it did have control over the area of NK.

Fourth, it’s very interesting to observe just how the Court evaluated the evidence that proved Armenia’s control over NK. Here, as with the issue of non-exhaustion of domestic remedies, the Court takes a very flexible approach:

• With regard to Armenia’s military involvement, the Court noted that it could not conclusively establish the composition of the armed forces that occupied NK during the conflict (Chiragov, para. 173), but the involvement of Armenian forces was nonetheless confirmed by a Human Rights Watch report and the statement of an Armenian official. Moreover, (para. 174): ‘it is hardly conceivable that Nagorno-Karabakh – an entity with a population of less than 150,000 ethnic Armenians – was able, without the substantial military support of Armenia, to set up a defence force in early 1992 that, against the country of Azerbaijan with approximately seven million people, not only established control of the former NKAO but also, before the end of 1993, conquered the whole or major parts of seven surrounding Azerbaijani districts.’
• A military agreement was signed between Armenia and NK, which provided that conscripts from one entity can serve in the other (para. 176).
• Many declarations and decisions of relevant international bodies confirmed Armenian presence in NK (para. 177). Notably, in one breath the Court cites a report of the International Crisis Group together with decisions of e.g. the Parliamentary Assembly of the Council of Europe, apparently seeing them of equivalent probative value.
• A statement by the Armenian President referring to the victory of ‘our Army’ in the Nagorno-Karabakh war (para. 178 citing para. 72). The Court thus finds, without ruling on just how many Armenian troops are deployed in NK, that (para. 180): ‘The Court need not solve this issue as, based on the numerous reports and statements presented above, it finds it established that the Republic of Armenia, through its military presence and the provision of military equipment and expertise, has been significantly involved in the Nagorno-Karabakh conflict from an early date. This military support has been – and continues to be – decisive for the conquest of and continued control over the territories in issue, and the evidence, not the least the 1994 military co-operation agreement, convincingly shows that the armed forces of Armenia and the “NKR” are highly integrated.’
• General political support by Armenia to the NKR (para. 181).
• Interchange of prominent politicians; granting of Armenian passports to NK residents; operation of Armenian law enforcement agents in NK (para. 182).
• Financial support which represents a substantial (roughly half) portion of the NKR budget (para. 183). Notably, the one source that the Court cites for this finding is a report by the International Crisis Group. The Court also mentions a fund established by the Armenian president (para. 184), and concludes that (para. 185): ‘It is true that substantial financial assistance to “NKR” is also coming from other sources, including the US government and direct contributions from the Armenian diaspora. Nevertheless, the figures mentioned above show that the “NKR” would not be able to subsist economically without the substantial support stemming from Armenia’

The Court’s ultimate conclusion was thus that (para. 186):

All of the above reveals that the Republic of Armenia, from the early days of the Nagorno-Karabakh conflict, has had a significant and decisive influence over the “NKR”, that the two entities are highly integrated in virtually all important matters and that this situation persists to this day. In other words, the “NKR” and its administration survives by virtue of the military, political, financial and other support given to it by Armenia which, consequently, exercises effective control over Nagorno‑Karabakh and the surrounding territories, including the district of Lachin. The matters complained of therefore come within the jurisdiction of Armenia for the purposes of Article 1 of the Convention.

The ease with which the Court establishes Armenia’s control over NK is, I think, remarkable. So is the Court’s reliance on reports by NGOs such as HRW and ICG, without requiring much (if anything) by way of corroboration. This has clear implications for future cases raising similar issues (e.g. Georgia/Russia – does Russia control Abkhazia or South Ossetia; Ukraine/Russia – does Russia control the Donbas region, etc – particularly in light of the Court’s reasoning on whether it’s ‘hardly conceivable’ that a smaller/less populous entity could on its own militarily defeat a much larger one without outside support). This is especially the case because the Court did not consider it necessary to conduct a fact-finding mission or to hear any witnesses.

The Court’s rather loose approach to evidence provoked a very strong dissent from Judge Pinto de Albuquerque (also dissenting in Sargsyan), who developed something of an allergic reaction to the two cases as a whole – but is I think right in noting that the Court’s approach to issues of proof is far from rigorous. Similarly, Judge Ziemele notes in her opinion (para. 5) that ‘Unlike the particularly scrupulous establishment of the facts normally carried out by the International Court of Justice (“the ICJ”) in cases concerning disputes over territories, jurisdiction and attribution of responsibility, the Court appears to be watering down certain evidentiary standards in highly controversial situations.’

Fifth, the Court’s bottom line in establishing a violation of Article 1 of Protocol 1 in Chiragov was that (para. 199): ‘pending a comprehensive peace agreement, it would appear particularly important to establish a property claims mechanism, which should be easily accessible and provide procedures operating with flexible evidentiary standards, allowing the applicants and others in their situation to have their property rights restored and to obtain compensation for the loss of their enjoyment.’ The question of reparations was reserved for further decision.

While there is close parallelism of the analysis in the two judgments, the end result in Sargsyan is somewhat different, but the Court still clearly tries to strike a politically palatable balance. First, the Court ‘accepts that refusing civilians, including the applicant, access to Gulistan is justified by safety considerations, in particular restricting access to a mined area and protecting civilians against the dangers existing in such an area’ (para. 233), but then notes that the state has a duty to take alternative measures in order to secure property rights of displaced persons even if it is not itself responsible for the displacement (para. 234), and then concludes that Azerbaijan failed to take such measures with respect to displaced ethnic Armenians (para. 241).

Sixth, the overall picture coming from Chiragov is that it confirms the post Al-Skeini trend that the Court is now likely to find Article 1 jurisdiction and is being increasingly generous on threshold questions of the Convention’s extraterritorial application. This means, as I have argued in respect of Jaloud, that the focus of such cases must shift to the merits. Litigants that devote too much attention to the threshold question of jurisdiction, and accordingly less to the merits, are likely to find that this is a losing strategy. (Note that in Chiragov 14 of the 17 judges agreed that Armenia had jurisdiction). In Sargsyan, on the other hand, the Court required a persuasive showing to rebut the presumption that a state remains in control of its own territory, a showing that was not made on the facts. (See also my earlier post on the implications of the Court’s rather traditional approach to the definition of belligerent occupation).

Seventh, the unfortunate confusion between jurisdiction and responsibility that has long plagued the Court’s case law persists in the two judgments, with this whole set of issues provoking a number of separate opinions. For example, in Chiragov, para. 201, ‘the Court finds that there has been and continues to be a breach of the applicants’ rights under Article 1 of Protocol No. 1 for which the Republic of Armenia is responsible.’ What is not clear from the judgment is whether the Court believes that the conduct of NKR separatists, which prevents the applicants from accessing their property, is attributable to Armenia, or rather whether Armenia is found responsible for failing to fulfil its positive obligations and protect the applicants from the conduct of third parties in areas under its jurisdiction.

In other words, the Court seems to think that its jurisdiction analysis – which relates to Armenia’s control over the territory of Nagorno-Karabakh – suffices for establishing Armenia’s responsibility for human rights violations committed by an entity that claims to be independent from Armenia, without examining Armenia’s control over that actor under the standards established by the International Law Commission and the International Court of Justice, standard which are both stricter and qualitatively different from those applied by the European Court to the jurisdiction issue. This confusion has pervaded the Court’s jurisprudence from earlier cases such as Loizidou, to Ilascu, Catan, and most recently Jaloud. I’ve said what I think on this in my post on the Jaloud case, so I won’t repeat that here.

A number of judges grappled with this relationship between jurisdiction and responsibility. For example, in her separate opinion, Judge Motoc does examine the different standard employed by the ICJ in Nicaragua and Bosnian Genocide, but considers that it was unnecessary to apply them on the facts of this case since jurisdiction was clearly established – while I agree that it may have been unnecessary to apply these standards, some conduct still had to be attributed to Armenia and the Court does not make it clear which conduct this is. In this regard I am in agreement with the separate opinion of Judge Ziemele, who considers that the requirements of the ICJ/ILC tests of attribution do not seem to be met on the facts, and that Armenia should be held responsible for failing to comply with positive obligations, viz. to exert the influence it has over the NKR in order to secure the applicants’ right of access to their property. As an aside, this question of the relationship between jurisdiction and responsibility provoked much discussion at a one-day conference recently organized jointly by the European Court and the European Society of International Law, with speakers including judges of the European Court and the ICJ, as well as academics, including myself. The videos of the conference can be found here.

To sum up, the Nagorno-Karabakh cases are an important addition to the Court’s jurisprudence. Although they are not as foundational as was, say, Al-Skeini, I am sure that they their impact will be felt for many years to come.