I read Daniel Halberstam’s eloquent and erudite defence of Opinion 2/13 with great interest and I agree that (some of) the Court’s arguments can be rationally explained. What struck me about his piece, however, is that while it is centred on the concept of autonomy, he doesn’t seem to regard it necessary to provide us with a definition of it. In order to mount an effective defence of the Court’s position, it would have surely been a good starting point to defend the Court’s conception of autonomy as expressed in the Opinion.
I think that this conception is at the core of the concerns that many of the Court’s critics have voiced. In other words, the question is what autonomy we are talking about. It seems to me as though with Opinion 2/13 the Court has tightened the screws on the autonomy concept quite a bit. In order to demonstrate this, it is perhaps worth recalling that in Opinion 1/00 the Court defined the external aspect of autonomy thus:
Preservation of the autonomy of the Community legal order requires therefore, first, that the essential character of the powers of the Community and its institutions as conceived in the Treaty remain unaltered … Second, it requires that the procedures for ensuring uniform interpretation of the rules of the […] Agreement and for resolving disputes will not have the effect of binding the Community and its institutions, in the exercise of their internal powers, to a particular interpretation of the rules of Community law referred to in that agreement. (at paras 12 and 13)
According to this, autonomy means that an international court must not be given jurisdiction to render internally binding interpretations of EU law. It does not say here that it must not consider EU law at all. Rather it is about the power to interpret European Union law in an authoritative manner.
Opinion 2/13 seems to extend this notion quite considerably. Take for example the ‘plausibility’ review by the ECtHR in the context of the co-respondent mechanism under Article 3 (5) of the draft accession agreement. What if the ECtHR decided that a request by the European Union to be joined as co-respondent were implausible? Would this lead to an internally binding interpretation of the distribution of competences between the EU and its Member States? Or would the consequences of such a finding be confined to the dispute before the ECtHR with the result that in the event of a judgment against the Member State concerned that Member State would bear the sole responsibility under the ECHR to remove the violation and/or pay just satisfaction? It seems to me that the latter would be the case. Moreover, it seems clear that the plausibility test was included in the accession agreement precisely in order to avoid a violation of the autonomy of EU law. I do not wish to deny here that the plausibility test could easily be done without. After all, how many instances of misuse of the co-respondent procedure can we conceive of? But I find it difficult to agree that its inclusion is contrary to the autonomy of the EU legal order.
Another example for a tightening of the autonomy concept is the Court’s insistence that the accession agreement contain an express clause safeguarding its own exclusive jurisdiction, which Daniel Halberstam defends as plausible. Again, I do not consider the solution found in the accession agreement to be the most elegant one. But in my eyes it does the trick: it ‘switches off’ the competing exclusivity clause of Article 55 ECHR. Admittedly, it leaves intact the ECtHR’s jurisdiction over inter-party cases, including cases between two Member States or even between the Union and a Member State. Yet Article 344 TFEU makes it clear for that it would be in violation of EU law if such disputes were brought before the ECtHR if they dealt with ‘the ECHR as an integral part of EU law’. Having ‘switched off’ Article 55 ECHR results in the parties concerned not operating under a conflict of jurisdiction between two potentially competent courts. Hence from their point of view the situation is clear: they must bring their case to the Court of Justice. If they do not, they are in breach of their obligations under EU law and the European Commission can bring infringement proceedings under Article 258 TFEU. Let us recall that this is exactly what happened in the Mox Plant case. Apart from this arguable tightening of the autonomy concept, it is also remarkable that the Court does not seem to trust its own legal system any longer to provide effective remedies in such cases.
It would thus have been welcome if Daniel Halberstam had provided us with a stronger normative case for this stricter conception of autonomy.