Sharing Responsibility for UN Targeted Sanctions

 Posted by: Tzanakopoulos Antonios

Cross-posted on EJIL:Talk!

UN targeted sanctions, especially those related to terrorism, have had their fair share of the limelight lately, particularly in view of important decisions by the ECJ, the ECtHR, the UK Supreme Court and others in cases such as Kadi, Nada, and Ahmed. Here, I try to look at this jurisprudence through the lens of the project on shared responsibility (SHARES). After introducing the relevant sanctions regime, I argue that the complex conduct of the UN and its member-states in designing, imposing, and implementing the sanctions leads to them sharing international responsibility for the resulting breach of aspects of the internationally protected right to a fair trial. This is so because states are ‘held responsible’ in their own domestic courts or in regional international courts, which then forces them to turn to the UN and seek to implement the organisation’s international responsibility. In this manner, the international responsibility for what is in effect ‘shared’ conduct is itself shared, in practice.

The UN Security Council has imposed obligations on member-states to implement asset freezes, travel bans, and arms embargoes on ‘listed’ individuals and legal entities ‘associated’ with the Taliban, Al-Qaida, and Osama bin Laden, by virtue of Resolution 1267 (1999). This sanctions regime, conveniently abbreviated until recently as the ‘1267 sanctions regime’, continues in force, even though it has now been split in two by Resolutions 1988 and 1989 (2011). The former resolution (and associated regime) refers to the Taliban and the latter refers to Al-Qaida. The differences between the two regimes focus primarily on the delisting procedure, which is significantly more advanced in the 1989 regime regarding Al-Qaida. In the remainder of this short contribution I will focus on this ‘1267/1989 sanctions regime’.

The targeted sanctions imposed by the 1267/1989 sanctions regime, both as they operated in the past and as they operate today, may be seen as being in violation of the right of access to a court and the right to an effective remedy of the targeted individuals and legal entities (hereinafter, when I refer to individuals this should be understood as encompassing legal entities as well) as this exists in customary international law (apart from its entrenchment in various instruments for the protection of human rights, notably the ICCPR, the ECHR, and the ACHR). Τhe sanctions regime has targeted individuals for a significant period of time (for some of them over a decade) and imposes significant restrictions on them, which may even be characterized as criminal in nature, in particular in view of their duration. Targeted individuals had no real recourse against these restrictions until Resolution 1904 (2009) introduced an ‘Office of the Ombudsperson’. This independent and impartial organ has been established to receive delisting requests from individuals, look into them, and make ‘recommendations’ to the competent Sanctions Committee of the Security Council regarding delisting. However, despite further enhancement of the delisting process in Resolution 1989 for the Al-Qaida regime, recourse remains limited, since in the final analysis delisting remains within the discretion of the Security Council.

We are dealing here with complex conduct, which involves both Security Council and member-state conduct. For its part, the Council sets up a sanctions regime and imposes an obligation on member-states to implement it. For their part, member-states comply with their obligations by implementing the regime against individuals found in any way within the reach of their jurisdiction. The outcome of this complex conduct is the denial of any effective avenue for challenge of restrictive measures on the part of targeted entities, and thus the violation of certain aspects of the right to a fair trial, a right under customary international law. This obligation to respect the right to a fair trial, which includes the right of access to a court and the right to an effective remedy, is incumbent on both the Security Council and the member-states, independently of any relevant (and overlapping) treaty obligations (of member-states only, in practice). This obligation is breached through conduct taken, prima facie, both by the Council and the member-states. How is responsibility for this internationally wrongful act to be shared between the United Nations and its member-states?

In order to answer this question we must untangle the complex conduct, and seek to attribute it to the various implicated entities. The conduct of the UN (through the Council) in setting up the regime and imposing obligations on member-states to implement it is normative in nature. Such normative conduct may itself constitute a breach of the obligation to respect the right to a fair trial: the keeping in force of normative acts in breach of international obligations may constitute an internationally wrongful act even if the normative act has not (yet) been implemented. But if not implemented, there is little chance of any real impact of the act in the ‘real’ world, and thus responsibility for it, while engaged, may remain theoretical. The normative act in our case, however, is implemented by the member-states through their organs, in compliance with their international obligations under Article 25 of the UN Charter. They take both normative and physical conduct in implementing the regime designed by the Council. Already at this point we can detect some ‘sharing’: the UN and the member-states must co-operate, act together, in order to create effects in the real world, that is in order to actually freeze accounts and ban travels, leaving those targeted with no real remedy.

The conduct of the Security Council in setting up the binding regime is attributable to the UN, because the Security Council is a UN organ. The conduct of member-states’ organs in implementing the regime is attributable in the first instance to their states, even though this conduct is strictly conditioned by the obligation imposed upon the states under Article 25 of the UN Charter to comply with binding Security Council decisions.

If we can establish that the normative conduct of the Council is enough to violate the UN’s obligation under customary law to respect the right to a fair trial, then there is little problem: the UN becomes responsible for that violation, while the member-states become responsible for the same violation by virtue of their own conduct in implementing. However, it may be argued that the normative conduct of the Council is not wrongful in and of itself. It is rather its implementation by the member-states that first breaches the international obligation to respect the right to a fair trial. But of course this member-state conduct is imposed by the Security Council. The question is then, how may the UN become responsible for conduct which is, in the first instance, attributable to the member-states (the implementation of the regime)?

There is no reason why the implementing conduct of member-states should not be concurrently attributable to the United Nations. This conduct is taken by member-states’ organs, but it is effectively controlled by means of a binding decision of the UN which imposes on these organs a strict obligation to achieve a given result (the freezing of the accounts of a named individual, among others). The member-state then is acting as an ‘agent’ of the UN (‘a person through whom the organisation acts’). However, the ILC Draft Articles on the Responsibility of International Organisations block this simple concurrent attribution (even though they do in principle admit the potential for concurrent attribution) by requiring, in cases where the alleged agent is an organ of a state, that the international organisation exercise effective factual control over that organ (ie not normative, but actual control on the ground along the lines of Nicaragua and Article 8 of the ILC Articles on the Responsibility of States). In any event, this still allows the UN to be held responsible for the wrongful act of the member-states through the provision on ‘circumvention’ (Article 17 DARIO). This is a problematic provision that does not reflect customary law. It also confuses primary with secondary norms, since it purports to allocate responsibility for the acts of member-states to the international organisation, when in reality what it does is to prohibit ‘circumvention’ of obligations by the international organisation (an act that remains wrongful even when the member-state acting in implementation of the binding decision of the organisation does not perpetrate an internationally wrongful act itself). But the bottom line is that both entities, international organisation and implementing member-states, will engage their international responsibility in the end.

Whether through the parallel perpetration of an internationally wrongful act by independent conduct, then, or through concurrent attribution of conduct, or even through derivative responsibility for the act of another, both UN and member-states are responsible for violating the right of access to a court and the right to an effective remedy. How can this responsibility be implemented in practice?

This is the point where we can see shared responsibility in action, and it’s all on account of significant practice on the part of domestic and regional international courts. Taking domestic courts first, in which I include the EU courts, these may not have jurisdiction over the UN, but they do have jurisdiction over domestic implementing measures of the member-states of the UN. Individuals targeted by the 1267/1989 regime have brought cases before these courts seeking to have the domestic implementing measures struck out or set aside. Here is how domestic and regional international courts have dealt with such claims:

I. If courts recognize that state conduct is conditioned by the UN-imposed obligation, they have difficulty holding the state responsible for the violation of the right to a fair trial, even if they do attribute the implementing act to the state. In doing this, they confirm concurrent, or ‘shared’, attribution of the conduct complained of, but then they either (i) accept that by reviewing the implementing act they must also review the UN act, and thus lapse into a very lax standard of review (see eg Kadi in the CFI, Nada in the Swiss Federal Tribunal); or (ii) they accept that they must incidentally review the act of the international organisation (as above) but then defer to allegedly equivalent protection at the level of the international organisation (Bosphorus in the ECtHR).

II. Neither of the two options above allows courts to properly review either the conduct of the UN, or the implementing conduct of the member-state. So eventually courts abandoned the approach under I. and started pretending that the state implementing conduct was not conditioned by the UN act. This radical disengagement of the two courses of conduct is of course wrong—to claim, as the ECJ did in Kadi, or the UK Supreme Court in Ahmed, or the ECtHR in Nada, that the 1267/1989 regime allows any margin of discretion in its implementation cannot withstand scrutiny. When the Council demands that a particular person must have her account frozen, the obligation is one of result. If the account is not frozen, then the obligation is violated, and so no margin of appreciation or discretion is allowed to the implementing state. And yet, domestic courts have quashed the implementing acts for violation of the right to a fair trial (as guaranteed under domestic law, which in substance coincides with the guarantee in international law). In effect, they can be seen as offering ‘juridical restitution’ (withdrawal of the offending normative act), which is but a method of implementing the international responsibility of the state for violation of the right to a fair trial. Even if it can be argued that such implementation is ‘internal’ to the state (through the action of its domestic courts), the decision of the ECtHR in Nada is a clear example where a state is held internationally responsible for not offering an effective remedy, even though it was internationally bound to act in the (wrongful) way that it did.

In this second set of cases, ‘sharing’ is not limited to attribution (as in the first set, where avoidance mechanisms kick in), but it is taken to its logical conclusion: a type of ‘shared’ responsibility. Here is how: the domestic court implements the responsibility of the state for the violation of the right to a fair trial by effecting ‘juridical restitution’ in striking down the domestic implementing measure. (Even better, the regional international court holds the state internationally responsible and requires that it effect juridical restitution and make compensation in favour of the beneficiary of the rule.) This is the state responsibility ‘part’. The state, however, is forced in this way to breach its obligation to comply with the Security Council decision. Aside from being able to justify this disobedience as a countermeasure, the state is in any event forced in practice to put pressure on the UN to reform the sanctions regime to have it comply with the right to a fair trial. This leads to the implementation of UN responsibility for the violation of the right, and constitutes the international organisation responsibility ‘part’ of the shared responsibility regime. Indeed, the model is not imaginary, but reflects what has happened in practice over the last years, even with the admission of the Security Council itself (which in Resolution 1904 recognises that the reform of the delisting regime is in response to continued legal challenges in member state courts). Responsibility then for the violation of aspects of the right to a fair trial (access to a court and right to an effective remedy) is shared in practice between the member-state and the UN. The member-state is forced by courts implementing its responsibility to implement itself the responsibility of the UN.

The ECtHR Finds Macedonia Responsible in Connection with Torture by the CIA, but on What Basis?The ECtHR Finds Macedonia Responsible in Connection with Torture by the CIA, but on What Basis?

 Posted by: Nollkaemper André

Cross-posted from EJIL: Talk!

On 13 December 2012, the European Court of Human Rights (‘the Court’) found that the Former Yugoslav Republic of Macedonia (‘Macedonia’) was responsible in connection with the ill-treatment and torture of Khaled El-Masri. The judgment adds a further chapter to the Court’s rich case law on situations where a state party is held responsible in connection with the (wrongful) acts of another state.

El-Masri, a Lebanese-born German national, alleged that in the period from 31 December 2003 to 29 May 2004 he had been subjected to a secret rendition operation, in which agents of Macedonia had arrested him, held him incommunicado, questioned and ill-treated him, and handed him over at Skopje Airport to CIA agents who then transferred him to Afghanistan, where he had been detained and ill-treated for over four months.

No one who reads the facts of the case will argue with the Court’s conclusion that Macedonia had to bear international responsibility. The question is on what grounds one can base this conclusion.

The approach chosen by the Court may surprise many international lawyers. Influenced by decades of work of the International Law Commission (‘ILC’), their approach would be a combination of attribution of conduct on the one hand and the breach of an international obligation, on the other: Macedonia then would be responsible for handing over El-Masri to the CIA, in the face of risk (if not certainty) that he would be ill-treated and tortured. They would not normally say that the act of ill-treatment at the hands of the CIA itself is attributed to Macedonia, but limit Macedonia’s responsibility to its own wrongful conduct. This distinction may seem a legal nicety, but it may have practical relevance (for questions of evidence and reparation) and also reflects that what is essentially a sovereignty-based consideration: it should not easily be presumed that a state is responsible for acts committed by another subject of international law.

The Court takes a somewhat different approach. But it is quite difficult to figure out exactly what this approach is. While the fact that the Court does not feel compelled to follow the ILC’s conceptual straightjacket is in many respects refreshing, its own line is at times somewhat inconsistent and confusing. For one thing, it is difficult to see why the Court uses interchangeably the terms ‘attribution’ and ‘imputation’ – one may guess that the Court uses the latter when it seeks to leave aside the ILC’s approach, but it would be nice if the Court would not invite us to speculate.

A more substantive point is that, without clear rationale and justification, the Court intertwines the concepts of attribution of conduct and attribution of responsibility. For instance, the Court found that the responsibility of Macedonia was engaged with regard to the applicant’s transfer into the custody of the United States authorities and eventually his transfer to Afghanistan, despite the existence of a real risk that he would be subjected to further treatment contrary to Article 3 of the European Convention on Human Rights (‘ECHR’) (para. 223). This follows the Soering case law, and fits in traditional doctrine: Macedonia would only be  responsible under Article 3 for its own conduct and not for the torture in Afghanistan itself. But it is then hard to understand why the Court speaks in this context of attribution of responsibility (para. 215), rather than attribution of conduct.

Nonetheless, amidst all this confusion, an interesting thought emerges. In regard to two allegations, the Court concluded that Macedonia was not (only) responsible for the act of handing over El-Masri, but that it was responsible for conduct that clearly was not its own. Macedonia had handed over El-Masri to a CIA rendition team at Skopje Airport, where he was ill-treated before being sent off to Afghanistan. The Court found Macedonia ‘responsible for the ill-treatment to which the applicant was subjected at Skopje Airport by CIA agents’. And in respect to the detention in Kabul, once the CIA had flown EL-Masri there, it found Macedonia to be responsible for a violation of Article 5 during the entire period of his captivity in Kabul (para. 240), and imputed the detention (an act committed by the CIA) as such to Macedonia (see also para. 235).

This approach of making a state responsible for acts that are not its own may be explained by the scope and contents of the particular obligations of states under the ECHR. That is particularly clear for the events at Skopje airport. The Court could have approached this as a Soering type situation, and could have held Macedonia responsible for handing over a person to a foreign state in the face of a risk of torture. However, the difference was that the CIA mistreated El-Masri at the airport itself, and this was therefore not just a question of handing someone over in the face that he would be mistreated in another state. The Court thus had to take a different approach and stated in para. 206 that:

the acts complained of were carried out in the presence of officials of the respondent State and within its jurisdiction. Consequently, the respondent State must be regarded as responsible under the Convention for acts performed by foreign officials on its territory with the acquiescence or connivance of its authorities.

The Court’s finding that Macedonia was responsible under the Convention for the acts performed by foreign officials on its territory with the acquiescence, or connivance of its authorities, reminds us that there is a world outside the conceptual framework established by the ILC. The Court did not attempt to explain its approach in ‘ILC terms’ – none of the four Articles of the Articles on State Responsibility (‘ASR’) that the Court referred to (Arts. 7, 14, 15 or 16) is relevant in this context. Driven by the general obligation of states to protect rights of persons under their jurisdiction, the Court’s approach seems to be Macedonia is responsible for the wrongful act of handing over El-Masri and that because of its acquiescence or connivance, this responsibility extends to El-Masri’s ill-treatment by the CIA.

This approach follows the Human Rights Committee, which in a 2006 case against Sweden determined that ‘a State party is responsible for acts of foreign officials exercising acts of sovereign authority on its territory, if such acts are performed with the consent or acquiescence of the State party’ (Mohammed Alzery v. Sweden, para. 11.6).

A striking aspect of the Court’s reasoning is that it equates the responsibility of a state vis-à-vis the conduct of another state (here the US), with the responsibility of a state vis-à-vis the acts of private persons. It refers, for instance to Ilascu v. Moldova and Russia, in which the Court said that ‘the acquiescence or connivance of the authorities of a Contracting State in the acts of private individuals which violate the Convention rights of other individuals within its jurisdiction may engage the State’s responsibility under the Convention’ (para. 318).

The Court’s equation of state responsibility vis-à-vis conduct of private actors on the one hand, and vis-à-vis foreign states on its territory, is logical from the perspective of the ECHR. Once the scope of states’ obligations to protect rights of persons within their jurisdiction was extended to obligations to secure, that third parties within their jurisdiction did not interfere with rights and freedoms of individuals, there is no compelling basis for distinguishing between private parties and states that acted within the territory of the state party.

The justification of the construction then lies in the combination of the (positive) obligations of states party under the Convention, and the fact that the conduct in question took place on its territory with its acquiescence or connivance, which in turn was incompatible with the positive obligations. The primary rules thus in a way incorporate questions that in the ILC texts are considered as freestanding secondary rules, just as the Court considers the question of facilitation of ill-treatment of El-Masri as a breach of positive obligations (e.g. para. 211), not in terms of a responsibility under Article 16 of the ILC Articles (even though the Court did list this as a relevant norm of international law).

Although all of this thus can be explained with the framework of the Convention, a reading of the judgment raises many questions concerning this approach. The statement that Macedonia was responsible under the Convention for acts performed by foreign officials on its territory is somewhat ambiguous. Since the Court did not go as far as attributing CIA conduct to Macedonia, this wording may be taken to suggest that Macedonia would be responsible without itself having committed a wrongful act. That of course is not what the Court intended to say, but a more solid explanation would help.

Somewhat mysterious is also that whereas Macedonia’s responsibility for handing over El-Masri to the CIA was based on the traditional Soering case law (Macedonia was responsible for the act of transferring El-Masri into the hands of the CIA in the face of risk of torture (para. 223), not for the torture in Kabul, just as in a Soering case, the extraditing state is not responsible for the death row or the death penalty that later occurs), it took a different line in regard to the responsibility of Macedonia for the illegal detention by the CIA of El-Masri in Kabul. Here the Court found Macedonia to be responsible for a violation of Article 5 during the entire period of his captivity (para. 240) and suggested that it imputed the detention (in the hands of the CIA) as such to Macedonia (see also para. 235).

This could be explained as extension of responsibility based on criteria of foreseeability and causation, but then it is not obvious why these factors were not applied in regard of the removal of the applicant to Kabul, where the Court only held Macedonia responsible for the removal, not for the torture in Kabul itself. All of this seems a connection of unfinished and not systemically developed thoughts.

Yet, in this midst of all this there is a thought that matters. For all its incoherence and lack of clarity, the Court’s language has a hint of normative power that the general law of responsibility lacks. The general law of responsibility by its conception of responsibility-based-on-wrongfulness, prefers determinations that one is responsible for the handing over of a person or for its inaction, not for the resulting torture itself.  In contrast, the Court’s approach may allow us to say that if a state hands over a person to another state in the knowledge that the person is tortured, and stands by when that torture happens, it bears responsibility for the torture itself.

The hint of a fresh approach to the law of responsibility deserves to be taken forward. There is no other international Court that has a similarly rich case-law on responsibility of states in connection to conduct of other states. It is about time that the Court starts clarifying the grounds of responsibility of states under the Convention, and in particular the grounds for extending such responsibility to acts that were committed by other states.