Jurisdiction, Attribution and Responsibility in Jaloud

In my last post on the Jaloud v. Netherlands case, I looked primarily at the bottom line of the case and what it will mean for the future. In this post, however, I would like to try to clarify the conceptual framework of jurisdiction, attribution and responsibility (which Aurel also looked at in his post) that the Court used (or should have used) in the case. The key parts of the judgment in that regard are paras. 140-155, which I will not reproduce in full here, plus the concurring opinion of Judge Spielmann, joined by Judge Raimondi.

It is clear from even a cursory read of the Spielmann opinion (as well as the concurring opinion of Judge Motoc, who writes in opposition to the two other Judges), that the judges of the Grand Chamber found the question of the relationship between the Article 1 ECHR concept of state jurisdiction, and general international law concepts such as attribution of conduct and responsibility for wrongful acts, to be particularly vexing. Judging by the language used, there must have been quite the internal debate. Judges Spielmann and Raimondi found the Court’s use of the attribution concept and its references to the case law of the ICJ (para. 95-97) and the ILC’s Articles on State Responsibility (para. 98, quoting articles 2, 6 and 8) to have been exceptionally objectionable, indeed ‘ambiguous, subsidiary and incomprehensible.‘ For the two Judges, attribution was a ‘non-issue’ in the case, which the Court should have avoided:

There was therefore no need to examine the non-issue of “attribution”, which is completely separate from the question of “jurisdiction”. More fundamentally, the Court should in any event be careful not to conflate the notions of jurisdiction under Article 1 with the concept of State responsibility under general international law. Efforts to seek to elucidate the former by reference to the latter are conceptually unsound and likely to cause further confusion in an already difficult area of law.

Contrary to the two Judges, I will try to show that attribution was, in fact, a central issue in the case, and that the Court’s approach, including references to the ILC’s work on state responsibility, was generally sound. However, I will also show that the Court could have been clearer in explaining what it was actually doing, which would have had the salutary effect of avoiding potentially confusing points for future cases. In fact, at least to an external observer, the divide between the majority and the two Judges is not as great as it might first seem, and the important conceptual points that they raise in the separate opinion can and should be adequately addressed.

What, then, is the proper relationship between the Article 1 ECHR concept of jurisdiction and the concept of attribution? (Health warning: this is a long post).

I. The two Judges are correct that the two concepts are separate, and that we should be careful not to conflate them. It is in fact the Court itself that has over the decades been the greatest culprit in conflating jurisdiction and state responsibility. As I explain in detail in my book, pp. 41-53, the Court has serially confused jurisdiction and responsibility in cases such as Loizidou and Ilascu, often by using hopelessly imprecise terminology (e.g. by talking about the responsibility of state parties under Article 1 – as if that article set standards of responsibility – or saying that a respondent state’s responsibility was engaged (whatever that may mean exactly)).

It has thus been exceedingly difficult for outside observers to understand what the Court was actually doing in such cases, e.g. whether the respondent states in cases such as Loizidou and Ilascu were found directly responsible for the conduct of non-state actors (TRNC or Transdniestria), or rather for their own failing to prevent such conduct by third parties. When the counsel for Russia in the Catan case (see more here) pointed out the ambiguities in the Court’s approach and its possible inconsistency with attribution tests devised by the ICJ and endorsed by the ILC (recall the whole Nicaragua/Tadic/Bosnian Genocide saga), the Court responded with a commendable, if disingenuous, paragraph which it now reaffirmed in Jaloud, para. 154: ‘The Court reiterates that 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 general international law (see Catan, cited above, § 115).‘ The two Judges also cite the same passage approvingly in para. 5 of their opinion. As I said, while the passage is commendable, since jurisdiction and attribution (and/or responsibility) are not the same concepts, the disingenuousness of the ‘has never been equated’ bit is manifest.

II. How exactly are these concepts different? Because they are intellectual operations which involve different questions:

Attribution of conduct is a question of whether the acts or omissions of a living, breathing human being (or group of human beings) should in law be regarded as the conduct of a state, a constructed, abstract entity, which (just like a corporation in domestic law) can in reality only act through living, breathing human beings.

State jurisdiction, whether under Article 1 ECHR or under the comparable clauses of other human rights treaties, is a concept which poses two different questions – does a state exercise effective control of an area or territory in which the victim of the alleged human rights violation is located (spatial model of jurisdiction), or does the state through its agents exercise authority and control over the individual victim of the alleged human rights violation (personal model of jurisdiction).

Attribution and jurisdiction may at times use similar terminology (e.g. effective control), but they actually pose different questions (control over what? – an actor, a territory, a victim). If we look at the matter through the ILC’s conceptual framework, attribution is an issue under Article 2(a) of the ILC Articles (is the conduct that of the relevant state?), while jurisdiction is an issue under Article 2(b) (did the conduct breach an obligation of the state, i.e. did that obligation even apply to that particular conduct?).

To give an example, when Osama bin Laden was killed by US Navy Seals in Pakistan, the attribution issue was perfectly clear: the conduct in question is attributable to the United States, because the deed was done by organs of the United States. But the jurisdiction issue is a separate one, and is not as clear – did, through killing him, US agents exercise authority/power/control (and thus jurisdiction) over OBL as an individual, so that OBL had rights under the ICCPR vis-a-vis the United States? Or, to give a different example, when the GCHQ intercepts my email while I am in Switzerland, the attribution issue is again clear – the conduct in question is undoubtedly that of the United Kingdom. But there’s a separate jurisdiction issue – whether by reading my email the UK exercised jurisdiction over me, so that I would have privacy rights under the ECHR vis-a-vis the UK.

III. So far so good. But the two Judges are not correct when they say that attribution was a ‘non-issue’ in Jaloud. In fact, attribution is an issue in every single case, Jaloud included. For a state to be responsible for any violation of the Convention, some conduct of a living, breathing human being has to be attributed to that state. Again, there is no way that a state can act except through individuals, or groups of individuals, and the conduct of these individuals must be assigned in law to the state (whether we call this process attribution, imputation, something else, or don’t give it a name at all). What is true is that in the vast majority of Strasbourg cases (say 99%) the attribution issue is not explicitly raised by the parties or by the Court, since on the facts of the case the answers to the attribution inquiry are obvious: the conduct which violates human rights is that of the state’s de jure organs (Art. 4, ILC ASR). In other words, when Italy is found responsible because Italian courts unduly delay a trial, or the Bulgarian police kill an unarmed fugitive, or the UK authorities try to deport someone to a place where they may be tortured, nobody really bothers with an attribution inquiry or invokes the ILC Articles. But the attribution inquiry is nonetheless still done sub silentio, since without it there can be no state responsibility for an internationally wrongful act (see also in that regard para. 135 of Al-Skeini).

IV. Attribution was in fact a very important issue in Jaloud. But to understand how this was so we must also understand which conduct exactly is being attributed, and what the relationship between jurisdiction and attribution can be (saying that they are separate, as the two Judges correctly do, does not mean that the are always entirely unrelated).

First, recall that we have defined jurisdiction as either effective control of an area by a state, or as authority, power or control over individuals by a state. Both the spatial and personal types of jurisdiction can only be exercised by a state through its own organs or agents, i.e. persons whose acts are attributable to it. For example, Turkey could not have exercised effective control over Northern Cyprus if there weren’t Turkish soldiers and other agents in that part of the island.

Second, depending on the circumstances, in a single case there may well be several different active or omissive instances of conduct which can be be attributed, expressly or implicitly. There is jurisdiction-establishing conduct, i.e. acts or omissions by which a state, through its organs or agents, establishes control over an area or control over an individual (in our Northern Cyprus example, the deployment of Turkish troops in that part of the island; or in Ilascu, the provision of vast amounts of military and economic aid by Russia to the Transdniestrian separatists). Then there is the violation-establishing conduct, i.e. the act or omission that can actually constitute a violation of the ECHR (e.g. failing to investigate in Al-Skeini or Jaloud; or the killing of the victims in Al-Skeini and Jaloud, had the violation of the negative aspect of Article 2 ECHR actually been alleged). In some cases, especially those that involve the personal conception of jurisdiction, the jurisdiction-establishing conduct and the violation-establishing conduct may actually be one and the same (e.g. the act of detaining the individual, as in Hassan, or of killing them).

Third, this means that an attribution inquiry actually logically precedes the jurisdiction inquiry when it comes to the conduct which is itself constitutive of jurisdiction. Had there been doubts, for example, that Turkey had deployed its troops in Northern Cyprus, the Court would first have had to establish whether the soldiers on the island were, in fact, acting on behalf of the Turkish state.

Fourth, this was exactly the situation in Jaloud, since the Netherlands was arguing that the conduct of its soldiers deployed in Iraq was not its own, because they were operating under the command of a UK officer. In other words, if the conduct of Dutch soldiers in Iraq was as a general matter no longer attributable to the state of the Netherlands (for whatever reason), the Netherlands could not have exercised jurisdiction under Article 1 either under the personal or under the spatial models (cf. Behrami).

Fifth, this is also exactly why the Court in Jaloud actually resolves an attribution issue in para. 151 before resolving the jurisdiction issue in para. 152 (even though the Court’s analysis in preceding paras. 140-150 is not entirely clear in how it is framed):

151.  That being so, the Court cannot find that the Netherlands troops were placed “at the disposal” of any foreign power, whether it be Iraq or the United Kingdom or any other power, or that they were “under the exclusive direction or control” of any other State (compare, mutatis mutandis, Article 6 of the International Law Commission’s Articles on State Responsibility, see paragraph 98 above; see also Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment, I.C.J. Reports 2007, p. 43, § 406, paragraph 97 above)).

152.  The Court now turns to the circumstances surrounding the death of Mr Azhar Sabah Jaloud. It notes that Mr Azhar Sabah Jaloud met his death when a vehicle in which he was a passenger was fired upon while passing through a checkpoint manned by personnel under the command and direct supervision of a Netherlands Royal Army officer. The checkpoint had been set up in the execution of SFIR’s mission, under United Nations Security Council Resolution 1483 (see paragraph 93 above), to restore conditions of stability and security conducive to the creation of an effective administration in the country. The Court is satisfied that the respondent Party exercised its “jurisdiction” within the limits of its SFIR mission and for the purpose of asserting authority and control over persons passing through the checkpoint. That being the case, the Court finds that the death of Mr Azhar Sabah Jaloud occurred within the “jurisdiction” of the Netherlands, as that expression is to be construed within the meaning of Article 1 of the Convention.

In other words, in para. 151 the Court establishes that the conduct of Dutch troops in Iraq is still that of the Netherlands; while it does not use the word ‘attribution’ as such, it clearly invokes two different rules of attribution in the ILC Articles (Arts. 6 and 8, the latter via the reference to the Bosnian Genocide case). The Court then proceeds to establish, in para. 152, that Dutch soldiers exercised authority and control, i.e. personal jurisdiction, over persons passing through the checkpoint that they set up (this is the jurisdiction-establishing conduct which is attributable to the Netherlands).

Finally, after having established jurisdiction, and under a new sub-heading called ‘attribution,’ the Court in para. 155 establishes the attribution to the Netherlands of the actual alleged violations: ‘The facts giving rise to the applicant’s complaints derive from alleged acts and omissions of Netherlands military personnel and investigative and judicial authorities. As such they are capable of giving rise to the responsibility of the Netherlands under the Convention.‘

V. In sum, Jaloud becomes more comprehensible when we see that the case raised not one, but two, attribution inquiries, one of which was logically prior to the establishment of jurisdiction. The Court was thus perfectly correct in doing what it did, although it could have expressed itself more clearly. In that regard, the two Judges are also correct in saying that the second attribution inquiry (paras. 154-155) does not properly fall within a jurisdiction analysis – structurally it shouldn’t have been placed as a sub-heading within the big ‘jurisdiction’ heading of the judgment. It is also correct that once the Court established that the conduct of Dutch troops in Iraq was still attributable to the Netherlands (para. 151), the attribution inquiry that followed was obvious and trivial (the alleged acts and omissions with regard to the investigation were still attributable to the Netherlands). But while its outcome may have been obvious (especially because it was done on the basis of organ status, and because the violation-constitutive conduct was mostly one of omission), this is not to say, as the two Judges do, that no such issue existed, or that it was improper for the Court to mention it.