18 Dezember 2014

Oops! We did it again – the CJEU’s Opinion on EU Accession to the ECHR

Today the CJEU answered the European Commission’s question “Is the Draft Agreement on the Accession of the European Union to the Convention for the Protection of Human Rights and Fundamental Freedoms compatible with the Treaties?” with a resounding “No”. This response probably comes as a surprise to many, not least the drafters of the Draft Accession Agreement (DAA), whose ambition to produce an agreement coupling the constitutional requirements of EU law with the Convention system proved unsuccessful. Having declared a previous attempt incompatible with the Treaties in Opinion 2/94 the Court did it again: it has thus reaffirmed its reluctance to subject the EU legal order (and most importantly its own judgments) to an external scrutiny by the ECtHR. The Court found fault with almost every aspect of the DAA, including its core features, the co-respondent and prior involvement mechanisms.

The Court’s Opinion revolves around two main themes: the autonomy of EU law and the exclusive jurisdiction of the CJEU. The relevance of the autonomy of EU law for this Opinion can be boiled down to this: an international court must not be given jurisdiction to interpret EU law in a binding fashion, in particular not on the division of powers between the EU and the Member States (for details see here). Moreover, under Article 344 TFEU the CJEU has exclusive jurisdiction to interpret EU law. Given that EU accession to the ECHR would give the ECtHR jurisdiction to review the compatibility of EU actions and omissions with the ECHR, it is not hard to see why both of these limits guided the accession negotiations and were of central relevance in Opinion 2/13.

This short blog entry cannot do justice to the enormous complexity of the Opinion (let alone the Advocate General’s view). I will therefore confine myself to outlining the key issues identified by the Court and briefly analyzing them (though not in the same order as the Court) before making some more general remarks.

The first three points discussed here concern the key features of the DAA.

First, the CJEU considered the co-respondent mechanism to fall foul of the autonomy of EU law. The mechanism allows for the EU to be involved as a co-respondent in cases where a Member State is held responsible before the ECtHR if the compatibility with the ECHR of a provision of EU law is at issue (Article 3 (3) DAA) in order to allow the EU to fully partake in proceedings concerning its laws. The Court criticized the specificities for involving the EU in that if the Member State requested the Union to become a co-respondent the ECtHR would have to carry out an assessment whether it is ‘plausible’ that the conditions for the involvement were met. In the eyes of the CJEU this would involve an assessment of the rules governing the division of powers between the EU and the Member States and was therefore not possible without a violation of the autonomy of EU law. It is submitted that this constitutes an overly strict approach. A plausibility review is superficial at best and does not lead to a binding assessment of the rules on the division of powers.

In addition, and perhaps more convincingly, the Court criticized the rules on the responsibility of the Union and the Member States if a violation is found. Generally, the DAA foresees a joint responsibility, but there is a way out if the EU and the Member State concerned give reasons and the ECtHR decides otherwise on the basis of these reasons. This can more plausibly be regarded as being in violation of the autonomy of EU law.

Moreover, the Court considered that the DAA would allow for a situation to occur in which a Member State that has declared a reservation to a part of the Convention could be held responsible under the co-respondent mechanism regardless. This it considered to be contrary to Protocol No. 8 to the Lisbon Treaty, which requires that the DAA must not affect the situation of the Member States in relation to the ECHR. This seems to concern an unlikely situation given that the EU can (logically) only be a co-respondent where there is a Member State as respondent, which presupposes that the Member State is under an obligation under the Convention, i.e. it has not made a reservation. So it seems that this problem would be pre-empted on the basis of the ECHR anyway.

Second, the Court took issue with the mechanism for its own prior involvement in co-respondent cases. Considering that most cases reach the ECtHR after the conclusion of proceedings in the (respondent) Member State, there is no guarantee that the CJEU has had the opportunity to consider the case because the system for preliminary references is not watertight. In order to account for the subsidiarity of the ECHR system, the DAA foresees an involvement of the CJEU in cases in which it has not yet assessed the compatibility with the ECHR of the provision of EU law concerned (Article 3 (6) DAA). Under the DAA the ECtHR decides whether these conditions are met. The CJEU did not accept this as it would be ‘tantamount to conferring on [the ECtHR] jurisdiction to interpret the case-law of the Court of Justice’. This would be incompatible with the Treaties. Moreover, the CJEU also criticised that its prior involvement was limited to cases concerning the validity of EU norms and requested that a prior involvement must also be possible in cases where the interpretation of EU law is necessary (Daniel Thym has thus been proven right). Otherwise the Court’s exclusive jurisdiction would be violated.

Third, the preservation of its exclusive jurisdiction was also at the heart of the Court’s scrutiny of Article 5 DAA, which has as its aim the exclusion of inter-party disputes between the EU Member States (and between the EU and a Member State). Because Article 5 does not explicitly exclude the ECtHR’s jurisdiction, it may not be too surprising that the CJEU found a direct infringement of Article 344 TFEU even though the explanations to the DAA make it clear that the operation of Article 344 TFEU would not be affected.

The remaining four failures identified by the CJEU are not directly concerned with any of the provisions of the DAA and some can be considered a little more esoteric. They are, however, testament to the very robust way in which the Court has rejected the agreement before it.

First, the CJEU considered that Protocol No 16, which has not yet entered into force (!), would be problematic (on this see Thomas Streinz). The Protocol will allow the highest courts of the parties signed up to it to request an advisory opinion from the ECtHR. If a Member State court were to make use of this option, the CJEU fears that the preliminary reference procedure could be circumvented given that the advisory opinion could trigger a prior involvement instead. It is not clear at all, how this should be possible given that the prior involvement mechanism is restricted to a co-respondent situation. Moreover, the CJEU did not take into account that an opinion provided under Protocol 16 would not be binding, nor would it relieve the referring highest court from its obligation under Article 267 (3) TFEU (see Marten Breuer). This passage of the Opinion is thus hard to comprehend.

Second, the CJEU considered that the DAA would have to make amends for the specificities of the Charter of Fundamental Rights (CFR), in particular Article 53 CFR. The background is that Article 53 ECHR allows the parties to the Convention to provide for more extensive human rights protection. The CJEU saw a danger that this would enable Member States to circumvent the limits set by Article 53 CFR. To an innocent reader, this does not make a lot of sense given that Article 53 CFR appears to say more or less the same. Yet one must not ignore the Court’s Melloni decision, on which it explicitly relies: only if the primacy, unity and effectiveness of EU law are not affected, may a Member State invoke its own (higher) constitutional standards. Hence the Court requires a coordination (whatever this would mean in practice) between the two Articles 53 so that the level of protection guaranteed by the Charter would not be compromised. This argument is somewhat bizarrely constructed and worded in a slightly misleading way considering that in effect the Court is trying to avoid that a higher standard of human rights protection than that ensured by the Charter should prevail.

Third, the Court was adamant that the principle of mutual trust underlying in particular the Area of Freedom, Security and Justice (AFSJ) be maintained at all costs. This principle rules out, for instance, that a Member State court refuses to execute a European Arrest Warrant on fundamental rights grounds. This is because the Member State is under an obligation to trust the requesting Member State’s legal order and the protection of fundamental rights provided by it. The Court held this principle to be so fundamental that the DAA would have to ensure that after accession a Member State court (and after the exhaustion of domestic remedies the ECtHR) would not be in a position to check whether the requesting Member State has complied with fundamental rights as this would upset the balance of the EU and thus undermine the autonomy of EU law. Given that AFSJ measures can be considered to be one of the Union’s weak spots in terms of fundamental rights protection, this can be read as an attempt by the CJEU to remove one of the stickiest issues in practice from the jurisdiction of the ECtHR.

Fourth, the CJEU required that the ECtHR not be given jurisdiction over CFSP measures, which the CJEU is currently not competent to hear (see Articles 40 TEU and 275 TFEU; an example are EU military operations). As far as such measures are concerned, the Court held that entrusting the ECtHR with the jurisdiction to hear them would effectively give sole jurisdiction over these issues to a non-EU body. It considered this an impossibility. This is somewhat baffling given that a lot of Member States in principle accept the ECtHR’s jurisdiction over foreign policy matters even though their domestic courts may not be competent to hear such cases. Moreover, one cannot help but have the sneaking suspicion that the Court has thereby extended its exclusive jurisdiction to cases over which it does not have jurisdiction (which is somewhat paradoxical).

Having outlined the main issues with the DAA as identified by the CJEU, some more general (and very brief) remarks are called for. Firstly, it seems as though the Court was determined to put a spanner in the works of the accession agreement. Its criticism is not punctual, but goes to the very heart of the DAA and may make accession a practical impossibility. There is no reason why, for instance, the non-EU parties to the ECHR should politically agree to the wide-ranging exclusions (of many aspects of the AFSJ, the CFSP, Article 53 ECHR) called for by the CJEU. The Court’s non-accommodating stance is secondly confirmed in its multiple requests to be presented with a watertight accession agreement. Arguably, many of the issues identified by the Court could be resolved EU-internally either by concluding internally binding rules or by applying the duty of loyal cooperation. This would certainly be true of the Protocol No 16 issue, but also of the defects in Article 5 DAA and in the co-respondent mechanism. This externalization of internally resolvable issues is new and worrying because it makes the EU a difficult partner to deal with. Thirdly, the Opinion is underpinned by a certain degree of irony: at the very start of the substantial part of the Court’s reasoning, it pointed out that the EU is not a state under international law, which would suggest that it should perhaps be more open to international integration. However, the CJEU’s unaccommodating stance and insistence on the preservation of its autonomous legal order, suggest the opposite.

So, what next? It is clear that the drafters of the DAA will have to return to the negotiating table. While some of the problems can be resolved by some minor redrafting, others may prove more difficult deal with (the Opinion requires more analysis in this respect) and, more importantly, to push through politically. One cannot but help the conclusion that the Court’s agenda was driven by a desire to make accession if not impossible, but very difficult indeed. This is somewhat surprising given that the Court had been very much involved in the accession negotiations by producing a discussion document (here) and a joint communication of the presidents of the two European courts (here), and that it seemed to have got what it asked for.

SUGGESTED CITATION  Lock, Tobias: Oops! We did it again – the CJEU’s Opinion on EU Accession to the ECHR, VerfBlog, 2014/12/18, https://verfassungsblog.de/oops-das-gutachten-des-eugh-zum-emrk-beitritt-der-eu-2/, DOI: 10.17176/20170602-153047.


  1. […] Thomas Streinz). Die ausgetauschten Argumente bedürfen an dieser Stelle keiner Wiederholung, zumal Tobias Lock die betreffenden Fragen mit Blick auf das Gutachten aufgegriffen hat und ich seiner Analyse in ganz […]

  2. […] what is next? Tobias Lock in his very fast and intelligent comment answered that question by stating that ‘[i]t is clear […]

  3. […] law as it currently stands. This ruling has been critically received, including on this blog (see here (Lock), here (Besselink), here (Michl), here (Douglas-Scott), here (Peers) and here (O’Neill)). […]

  4. Kurt Roth Fr 29 Jan 2016 at 16:10 - Reply

    Habe einen Brief vom Herrn Müller-Elschner erhalten. Ich habe als Mittelloser Bürger nach Artikel 35 Absatz 3b geklagt. Meine Unterlagen wurden vernichtet, lt. Hern Müller-Elschner, da ich mich nicht an den Artikel 47 , den ich nicht kannte, gehalten habe!! Ich besitze kein Kleingeld, wurde aber von der DRV betrogen und bestohlen.Ich habe nur verlangt, sich an die EU-Konvention zu halten (Artl.35 Abs.3b).

Leave A Comment


1. We welcome your comments but you do so as our guest. Please note that we will exercise our property rights to make sure that Verfassungsblog remains a safe and attractive place for everyone. Your comment will not appear immediately but will be moderated by us. Just as with posts, we make a choice. That means not all submitted comments will be published.

2. We expect comments to be matter-of-fact, on-topic and free of sarcasm, innuendo and ad personam arguments.

3. Racist, sexist and otherwise discriminatory comments will not be published.

4. Comments under pseudonym are allowed but a valid email address is obligatory. The use of more than one pseudonym is not allowed.

Other posts about this region:

Other posts about this region:
4 comments Join the discussion