Comment on Kokott/Sobotta “The Kadi Case – Constitutional Core Values and International Law – Finding the Balance?”

Nele Yang is a PhD candidate and research fellow at the Max Planck Institute for Comparative and International Law in Heidelberg.

The title of Kokott and Sobotta’s article (available free here) seems to suggest that this is another analysis of whether the 2008 Kadi decision succeeds in balancing core EU constitutional values and the effective implementation of UN anti-terrorist sanctions. Innovatively, however, the authors choose not to venture down this well-trodden path. Instead they ask what can be drawn from the Kadi judgment and the developments it has triggered in order to reduce conflict between the EU legal order and the UN Al-Qaida sanctions regime in general.

Solange – the Importance of Conditions

Kokott and Sobotta assert that the ECJ’s 2008 Kadi decision contains an approach similar to the German Federal Constitutional Court’s famous Solange. The question whether the 2008 Kadi judgment can be read as spelling out a Solange approach or at least leaving this approach open as one alternative for future action is less controversial than the article makes it out to be. Considering how much of the Kadi debate has been conducted around this issue, it is to be regretted that Kokott and Sobotta do not clarify that there is no “either/or” relationship between an approach à la Solange and a dualist approach. Moreover, Solange does not necessarily imply an attenuated form of dualism. To the contrary, depending on the conditions contained in the relative clause starting with “as long as”, the stance that it is this legal order which determines under which circumstances the other legal order will have any bearing on it might even be intensified. This is why we distinguish between Solange II and Solange I. Thus, “solange” in itself only expresses that interaction is made conditional. It does not say anything about the conditions, but the conditions are the crucial part. So the question is not so much whether a Solange approach can be drawn from the 2008 judgment but rather under which conditions the Court might in future cases practice deference.

At least implicitly, this question seems to have been on the authors’ minds, too. They interpret the 2008 Kadi judgment as implying that the CJEU might take a different position if the UN delisting procedure guaranteed judicial protection. But they do not ponder what the standard of this protection might be, and consequently, what could serve as a yardstick to assess the current state of affairs. The Kadi decision has certainly had an observable impact on the UN delisting procedure which since 2008 has evolved from a procedure only accessible through States to a mechanism involving the listed individuals themselves and a so-called Ombudsperson. The authors’ only guideline in analyzing this development seems to be that “[t]o envisage a Solange II relationship between the EU and the SC, the judicial protection at the UN level has to improve substantially in comparison to the situation examined by the Court in Kadi”. The analysis remains open-ended. Kokott and Sobotta conclude that “[the Ombudsperson’s] role has not attained the quality of a court of law” and that the standards on which the Ombudsperson bases her recommendations for delisting remain intransparent. It remains unclear, however, whether the authors consider these aspects as deficiencies which would have to be remedied before the CJEU could play the Solange II card. The authors conclude that “it is unclear whether the CJEU would consider this review sufficient to refrain from controlling implementing acts of the EU”. In other words, it is doubtful whether the procedure has fully risen to the challenge yet because we do not know what exactly the challenge is.

Ombudsperson Integration as a Separate Issue?

But Kokott and Sobotta submit that the Ombudsperson mechanism could be used to reduce conflict between the UNSC sanctions regime and CJEU judicial review “[n]evertheless”. This would be realized, the authors suggest, by incorporating this mechanism as an admissibility criterion for an action for annulment under art. 263 TFEU. Not only since Solange have conjunctions played an important role in expressing relationships. The implication of “nevertheless” here is that the question whether the current state of affairs at UN level would require the CJEU’s deference can be separated entirely from the question whether and how the Ombudsperson could be used to decrease the potential for conflict. I submit that such a detachment is not possible. Quite to the contrary: the justification for integrating the Ombudsperson procedure as an admissibility criterion is tied to the question of its effectiveness as a remedy. I will show this by examining the authors’ factual and legal arguments.

Rationality – an Elusive Common Ground

To illustrate the reasons why their proposal will reduce conflict, Kokott and Sobotta build on three notions: rationality of decision, that the Ombudsperson is “closer to the case” than the CJEU, and that the UN regime should be given a chance to activate its self-healing forces.

For the authors, the reason why all of this would serve to reduce conflict is that “[i]f the decision to maintain the listing at UN level is taken rationally, it should be possible to confirm it most of the time in the EU Courts”. If the CJEU decides based on the information on which the Sanctions Committee has also based its rational decision to maintain an individual’s listing, the possibility of conflicting outcomes reduces, the argument goes. Rationality, however, is an elusive concept. It remains unclear which meaning the authors ascribe to it. Perhaps it is this lack of clarity which produces the questionable premise that decisions taken rationally based on the same information will most of the time yield the same outcome. This premise leads to the conclusion that a decision which is taken rationally at UN level will in most cases also be in conformity with EU law. However, this really depends. If rationality means taking into account all of the available information when making a decision, the outcomes can still vary widely depending on which aspects the decision-makers assign more importance to. The outcome of the rational decision of a Sanctions Committee primarily concerned with international security can differ substantially from the outcome of the rational decision of a CJEU that considers individual rights protection a priority, even if using the same information. Therefore, one cannot stipulate the reduction of “the risk of divergent decisions”  here, neither can one equate it with the reduction of “the risk of conflict between UN sanctions and EU judicial protection”.

Plus, the decisions would not even be based on the same information. Kokott and Sobotta maintain that the Ombudsperson mechanism could help the CJEU overcome the communication deficit in its relationship with the Sanctions Committee. The Ombudsperson is indeed closer to the case in the sense that it communicates directly with the Sanctions Committee. The CJEU, on the other hand, has to communicate through the EU Commission – which is reluctant on its own to initiative requests for more information from the Sanctions Committee, an attitude clearly shown in the public hearing of the Kadi II case before the ECJ’s Grand Chamber. However, the CJEU still would not be deciding based on the same information as the Sanctions Committee, or even the Ombudsperson, because of confidentiality issues. The authors concede that “[p]roblems are likely to arise if the listing cannot be justified exclusively on the basis of non-classified information but also depends on confidential information”. They do not elaborate on the important role confidential information plays in the listing of individuals. The Ombudsperson has concluded agreements and arrangements with a small number of states to allow her access to confidential information, but the US as a major actor in listing initiatives is only willing to share confidential information on an ad hoc basis, and institutions external to the process – like the CJEU – neither see this information nor the recommendation resulting from it. If one cannot know what all of the available information is, one cannot assess whether all of the available information has duly been taken into account. Thus, for the CJEU the assessment of the Sanctions Committee’s decision is still a matter of educated guessing.

The Conditionalities of Loyal Cooperation

The authors admit that their proposal is about the exhaustion of international remedies. This is a fundamental difference. With the exhaustion of local remedies, the remedies which first have to be exhausted are those of the legal order in which the impugned act originates. An EU Regulation implementing UN sanctions clearly is the act of an EU institution. Actions for annulment can only be brought against EU acts. That a remedy of a different legal order should be exhausted before this is possible would have to be justified on grounds other than the origin of the act. If one wants to draw inspiration from the exhaustion of local remedies, one will have to ask what the rationale of this mechanism is and whether this rationale can be applied here to require the exhaustion of international remedies as envisaged by the authors. As one possible source of analogy, Kokott and Sobotta cite GC caselaw which obligates the Council to give precedence to Member States’ remedies (T-348/07 Stichting Al-Aqsa, para. 163). The constellation there already eludes the classic local/non-local distinction because the impugned act is an EU Regulation, not a Member State act. What, then, is the rationale of the Council’s obligation? Tracing back the line of caselaw leads to OMPI (T-228/02 Organisation des Modjahedines du peuple d’Iran, para. 124), which “by analogy” refers to the Postbank decision (T-353/94). Here, the then CFI asserts that a prohibition to produce, in national court proceedings, confidential documents from prior administrative procedure involving the Commission would run counter to the principle of sincere cooperation. The CFI argues that the fact that Member States are bound by the standard of the EU rule of law allows the assumption that they will also apply this standard effectively (at para. 69). This assumption justifies EU authorities’ obligation to cooperate in good faith. It is this rationale which OMPI takes up by analogy. – The principle of sincere cooperation is not a one-way street. This must be kept in mind when applying it to structure the relationship between the CJEU and the UN Sanctions Committee and using it to justify the mechanism suggested by the authors. For an obligation of the CJEU to cooperate loyally to arise, a certain standard of rights protection would have to be met by the UN delisting procedure. Clemens A. Feinäugle has examined how the notion of “UN loyalty” or “justice loyalty” might be employed to structure relations between the CJEU and the UN Sanctions Committee – tellingly, all of this is based on the condition that an independent and impartial control panel be established at UN level in accordance with certain rule of law standards in the first place (Clemens A. Feinäugle, Hoheitsgewalt im Völkerrecht, 2011).

As further source of inspiration, Kokott and Sobotta point to primary law provisions and paras 292 ff. of the 2008 Kadi decision. Here, they see traces of a principle of loyal cooperation in EU-UN relations which could require the exhaustion of international remedies. However, as shown above, such cooperation cannot be unconditional. This is reflected by the wording of the Treaty provisions and the ECJ decision (“respect”, “take due account of”, “appropriate forms of cooperation”). Consequently, it is difficult to see how these could be used as the basis for an obligation to exhaust UN remedies if it remains unclear whether these remedies are effective in terms of rights protection. All the more so since their effectiveness would play a role in determining their conformity with art. 47 of the EU Charter of Fundamental Rights (see Commentary by the EU Network of Independent Experts on Fundamental Rights, p. 360, available through the EU Fundamental Rights Agency).

Effectiveness as a Condition for the Local Remedies Rule in International Law

The local remedies rule in international law which Kokott and Sobotta draw on also points to the question of the effectiveness of the remedy. One of the arguments justifying this rule, apart from considerations of fairness to the other legal order, is “that in legal systems subject to the rule of law, local remedies will normally be more effective than international ones” (Crawford and Grant, ‘Local Remedies, Exhaustion of’, in: Max Planck Encyclopaedia of Public International Law, para. 7). This is comparable to the rationale extracted from the GC’s caselaw above: that a remedy in one order operates according to the rule of law is essential for the assumption that this remedy will be more effective, which in turn justifies its precedence.

Back to the Beginning

The article does much to identify the delisting procedure as a major parameter the CJEU should consider when reviewing implementation of Al-Qaida sanctions. However, it does not investigate the standards that the Court should apply to assess the compatibility of the delisting procedure with the rule of law. But this is a necessary exercise in order to decide whether that procedure should be elevated to an admissibility criterion for CJEU scrutiny. So the mechanism proposed by the authors leads back to the question of the paper’s first part: Has the delisting procedure evolved enough for the CJEU to practice deference? One might spy a contradiction here. If the UN remedy met the CJEU’s Solange conditions, would the CJEU not have to cede to it? And would this not render obsolete the question of the exhaustion of the UN remedy as an admissibility criterion? This is not the case. Deference does not have to mean that the Court would have to fully abstain from reviewing EU acts that implement UN Al-Qaida listings – it could also just mean modifying the intensity of review. Thus, there would still be room for the Ombudsperson mechanism as an admissibility requirement.

The essential question remains what the ECJ requires when it laments the absence of “guarantees of judicial protection” in the UN delisting procedure (C-402/05 P Kadi, para. 322). The ECJ’s Kadi II judgment forthcoming in 2013 might shed more light on this.