EU accession to the ECHR: How to square the circle?

Brussels 9 March 2020
  • Diminuer la taille du texte
  • Augmenter la taille du texte
  • Imprimer la page
  • Imprimer en PDF
EU accession to the ECHR: How to square the circle?

Mr Polakiewicz made a presentation at the FREMP/COHOM meeting today in Brussels. Please find below the text of his speech and click here to view his presentation.




EU accession to the ECHR: How to square the circle?

Ladies & gentlemen,

It is a great pleasure to share with you some reflections on a topic that has been fascinating me for many years.

In my intervention I would like to focus on the way forward, how to square the circle and address the objections formulated by the Court of Justice (ECJ) in its famous opinion 2/13.

From the outset, I would like to emphasise that the only realistic way forward is to address all objections one by one. I have prepared a presentation which covers all objections,[1] but I can focus today on only a few of them.

But let me start with a general point which in my view is essential for the understanding of the opinion.

It is not the Council of Europe that has requested the accession of the European Union to the Convention. It is your constitutional instrument, the Treaty on European Union, that requires the Union to accede to the ECHR. Article 6 (2) of the TEU is crystal-clear: “The European Union shall accede to the European Convention for the Protection of Human Rights and Fundamental Freedoms.”

What is particularly surprising in opinion 2/13 is the absence of any argument referring to the constitutional significance of article 6 (2) TEU. Instead, the ECJ raised a series of objections which are not only of an exceptional magnitude, but also, at least partially, of questionable legal relevance.

To give just two examples:

According to the ECJ, the mechanism of advisory opinions established by Protocol 16 to the ECHR would affect the autonomy and effectiveness of the ECJ’s preliminary ruling procedure. This argument is first of all beyond the scope of the accession agreement which does not encompass accession to Protocol No. 16. There is moreover a clear distribution of roles between the two courts. While the courts in Luxembourg interpret EU law, including the fundamental rights of the EU Charter, the ECtHR interprets the ECHR.

When it comes to the interpretation of human rights standards, it is not decisive which court speaks first but what it says, i.e. whether it remains within the limits of its jurisdiction. However, assuming for arguments sake that the ECtHR were to pronounce itself on an issue of EU law, the non-binding nature of its advisory opinions would mean that, as a matter of law, findings by the ECtHR cannot prevail over binding EU law including the interpretations given by the ECJ in preliminary reference proceedings under article 267 TFEU.

It is significant that already eight EU member states ratified Protocol 16 without any reservation or declaration. In 2018, a report by the French Senate concluded: “La mise en œuvre du protocole n°16 n’affecte aucunement les obligations incombant aux juridictions nationales au titre de l’article 267 TFUE.”[2]

The ECJ’s objection regarding the EU member states’ reservations seems to confuse the general reservation regime under the Vienna Convention on the Law of Treaties with the specific one under the ECHR. Under the ECHR, High Contracting Parties cannot generally derogate from a treaty provision but may do so only in respect of a specific piece of national legislation in force when the reservation is being made (article 57 (1) ECHR).[3] All reservations made by individual EU member states refer therefore to a specific national law in force when that reservation was made, and never to a provision of (primary or secondary) EU law.

This brings me to a very important point. If you take all the ECJ’s objections at face value and try to overcome them one by one by formal amendments to the draft accession agreement, there is a real risk that, as a result, the ECtHR’s jurisdiction over EU legal acts will be more restricted than it is today.

Such a solution would not only undermine the whole purpose of accession but would also be unacceptable to non-EU member states (NEUMS). It must not be forgotten that whatever concrete proposals the European Commission will present, they will be subject of negotiations in Strasbourg. As a former director of your legal service, Prof Jacqué remarked so pertinently in an early comment to opinion 2/13, it “needs two to tango.”[4]

We feel very much reassured that the letter by former Commission President Jean-Claude Juncker and Vice-President Frans Timmermans requesting the resumption of negotiations emphasises that the European Union will limit its request for amendments to the draft accession agreement to what is strictly necessary to address the objections raised by the ECJ.[5]

At the same time, I can only encourage you to proceed with the drafting of internal rules. Those rules could contain answers to some of the ECJ’s objections. An international treaty to which NEUMS will be parties and that will ultimately be interpreted by the ECtHR is not the right place to codify provisions whose main purpose it is to ensure that EU member states comply with EU law. With hindsight, I would argue that the absence of such rules made it easier for the ECJ to declare the draft accession agreement incompatible with EU law.

Arguably the ECJ’s most serious objections concern mutual trust and the Common Foreign and Security Policy (CFSP).

Regarding the principle of mutual trust, the ECJ argues that accession could undermine the autonomy of EU law as regards cross-border cooperation between EU member states, in particular the transfer of persons and the recognition and enforcement of judicial decisions.[6]

I would like to emphasise that in its case law, both prior and subsequent to opinion 2/13, the ECJ has consistently upheld respect for fundamental rights. It did so in many cases concerning child abduction[7] or the recognition of civil judgments.[8] It is true that the ECJ had in the past been more cautious when it came to the execution of European arrest warrants (EAW) or the return of refugees under the Dublin regulations. However, its more recent case-law is converging with that of the ECtHR.

As the ECJ’s president himself acknowledged, “mutual trust is not blind trust.”[9] You must not execute an EAW whenever where there is a real risk of inhuman or degrading treatment, for example because of detention conditions in the issuing member state.

In Dorobantu,[10] the ECJ recently underlined again the absolute nature of the prohibition of inhuman or degrading treatment and the duty of executing judicial authorities to “determine, specifically and precisely, whether in the circumstances of a particular case, there is a real risk that that person will be subjected in the issuing member state to inhuman or degrading treatment.”[11] In that context, the ECJ has specifically acknowledged that the relevant risk analysis should be based on ECHR judgments, judgments of other international and national courts as well as decisions or reports of organs of the Council of Europe or the United Nations.[12]

ECHR case law pursues the same objectives.[13] How can it be otherwise? EU law and ECHR are based on the same principles and values. Respect for fundamental rights is a key component of the area of freedom, security and justice and EU member states are not immune from being occasionally found in violation of even the most serious human rights violations.

In 2019 alone, ECtHR found

- 29 violations of articles 2 and 3 of the ECHR (right to life and prohibition of torture, including for lack of effective investigation, conditional violations and inhuman or degrading treatment);

- 16 violations of article 5 ECHR (right of liberty and security);

- 23 violations of article 6 ECHR (right to a fair trial) of EU member states.[14]

That being said, it may be useful to have a provision in the accession agreement which reflects the fact that EU law may provide for a specific distribution of responsibilities between EU member states involved in cross-border judicial cooperation. In child abduction cases for example, the relevant provisions require that objections regarding the lawfulness of a return judgment must be raised before the courts of the member state of origin.[15]

Regarding CFSP, the ECJ found that giving the ECtHR jurisdiction in this policy field would be incompatible with its own monopoly of jurisdiction over EU acts. The drafters of the Lisbon Treaty had, however, intentionally created a situation where national courts take, as regards certain CFSP decisions, the place of the EU courts. National courts, which are also Union and Convention courts, are perfectly suited to ensure effective judicial protection.

Questions of jurisdiction, attribution and responsibility are particularly complex in the CFSP framework, a rapidly evolving policy field which is uniquely situated between intergovernmentalism and supranationalism. You know much better than me that the ECJ’s jurisdiction in this field is restricted (article 275 (1) TFEU).

When drafting the accession agreement, the negotiating parties included a rule of attribution. Article 1 (4) of the draft accession agreement provided that “an act, measure or omission of organs of a member state of the European Union or of persons acting on its behalf shall be attributed to that state, even if such act, measure or omission occurs when the state implements the law of the European Union.” It was however added that “[t]his shall not preclude the European Union from being responsible as a correspondent.”

This approach is perfectly in line with international law. In its ‘Draft articles on the responsibility of international organizations’ (2011), the International Law Commission acknowledges that there may be cases of dual responsibility of member states and an international organisation.[16] It is therefore possible to agree, in the draft accession agreement, on specific rules of attribution that would bind the ECtHR. It is primarily an internal matter of the EU and its member states to agree on such rules before negotiating in Strasbourg.

We must not forget that many acts, in particular legislative instruments, adopted in the CFSP framework will only exceptionally affect the rights of individuals personally, directly and immediately in such a way that they can claim to be ‘victims’ of a Convention violation within the meaning of article 34 of the ECHR.[17] Concrete measures that may potentially interfere with human rights will typically be taken by EU member states and are consequently already now subject to judicial review before national courts and ultimately the ECtHR.

I am convinced that, provided there is the necessary political will, practical solutions can be found for this and the other objections. Only excluding CFSP matters would cross a red line. Such an exception could hardly be reconciled with the idea of comprehensive and effective external supervision by the ECtHR and the principle that the EU should accede to the ECHR on an equal footing with the other High Contracting Parties.[18]

* * *

This brings me to my last point which corresponds to the first question in your paper: what are the benefits of the EU’s accession to the ECHR for individual member states and Europe as a whole?

I would like to approach this question the other way around. What are the risks if accession will not happen?

Your governments can of course let the courts in Luxembourg and Strasbourg continue to develop their case law. They must, however, be aware that such an approach entails real risks for legal certainty and the coherence of human rights protection on our continent.

While both courts do their utmost to maintain a consistent approach, there are already now certain areas where commentators noticed divergences in case-law. Two examples are data protection[19] and the balancing of rights and duties of employees of private establishments run by the catholic church (freedoms of conscience, religion and expression).[20] Governmental experts from all 47 Council of Europe member states, including all EU member states, concluded already in 2015 that “[i]n case of non-accession, there is a real risk of the two main European legal systems drifting apart.[21]

As Prof Eeckhout put it so pertinently, the concept of the autonomy of EU law as developed in opinion 2/13 “risks undermining the very authority of law in the European legal space. It is one thing to conceive of European legal orders or systems - national law, EU law, and Convention law - as having their own identity and autonomy. It is another to conceive of them as self-contained and unbridgeable. The territorial and personal space in which they operate is unitary.”[22]

These legal systems may all claim authority over a single case, be it an asylum seeker in a refugee camp in Greece, a criminal suspect to be surrendered to a country in which s/he may suffer from inhuman prison conditions or an ordinary citizen whose personal data are intercepted illegally. “If the answer to their claims depends on which set of norms is applied, and which court hears their case, the rule of law will become relative and contingent, and the very idea of inalienable human rights will suffer.”[23]

Already now, it is difficult for practitioners to keep abreast with the developing case-law in Luxembourg and Strasbourg. Acting often in urgency, judges will have to identify in each individual case the applicable standards and accommodate the various legal sources which will exist wholly independently from each other. To quote a practitioner, “it is a demanding methodological task for a Supreme Court Justice to untangle the intricate and dynamic web of legal material that continuously are being spun within European law, layer by layer.”[24] In the worst case, domestic authorities may be held responsible under the Convention for action that is imposed by EU law.

* * *

As they say in Croatian, “[k]ad se male ruke slože, sve se može” (“when small hands work together, you can achieve everything”).

Let us all work together, over and above dividing institutional lines, to square the circle and bring this “never-ending story[25] to a happy end!

Thank you for your attention.


[1] See also J. Polakiewicz ‘Accession to the European Convention on Human Rights – An Insider’s View Adressing One by One the CJEU’s Objections in Opinion 2/13’ 36 Human Rights Law Journal 10-22 (2016).

[3] See J. Polakiewicz ‘Collective Responsibility and Reservations in a Common European Human Rights Area’ in I. Ziemele (ed.) Reservations to Human Rights Treaties and the Vienna Convention Regime (Springer Dordrecht 2004), 95 et seq.

[6] Opinion A-2/13, para 194.

[7] CJEU Judgment in Rinau,C-195/08 PPU, paras 48, 52, 85, (11 July 2008); CJEU Judgment in Povse v. Alpago C-211/10, para 43, (1 July 2010),

[8] CJEU Judgment in Krombach, , C-7/98, para 44, (28 May 2000); CJEU Judgment in De Baecker v Bowman, C-49/84, para. 12 (11 June 1985).

[9] See K Lenaerts La Vie Après l’Avis : Exploring The Principle of Mutual (Yet Not Blind) Trust (2017) Common Market Law Review, 806.

[10] CJEU Judgment in Dorobantu, C-128/18 (15 October 2019).

[11] Ibid, para. 63.

[12]Ibid, para 52; CJEU Judgment in Aranyosi, C-404/15 and C-659/15, para. 89 (5 April 2016).

[13] ECtHR, Avotiņš v. Latvia (17502/07, Grand Chamber judgment of 23 May 2016, para. 114; Romeo Castaño v Belgium (8351/17), judgment of 9 July 2019.

[15] CJEU Judgment in Zarraga Aguire, C-491/10, para. 50 (22 December 2010), by reference to CJEU Judgment in Rinau, C-195/08, para. 85 (11 July 2008); articles 42 (1) and 43 (2) of Regulation No 2201/2003.

[16] See article 48 of the “Draft articles on the responsibility of international organisations”, available at

[17] Segi and Gestoras ProAmnistia and Others v. Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden and the United Kingdom (dec.), nos. 6422/02 and 9916/02, ECHR 2002V.

[18] These were only two of the principles formulated by all negotiating parties at the outset of negotiations back in 2010, others included the subsidiary nature of the ECHR protection mechanism and the principle that amendments to and adaptations of the ECHR should be limited to what is strictly necessary for the purpose of the accession of the EU as a nonstate entity, see See Report of the CDDH of its 70th meeting, 15-18 June 2010, CDDH(2010)010, para. 31.

[19] ECtHR Judgment in Big Brother Watch and Others v. the United Kingdom (applications nos. 58170/13, 62322/14 and 24960/15), Joint Partly Dissenting and Partly Concurring Opinion of Judges Pardalos and Eicke, para. 22, p. 200.

[20] CJEU Judgment in IR v JQ, C-68/17 (11 September 2018). Deciding the same case previously, the German Federal Constitutional Court (BVerfG, Order of 22 October 2014, 2 BvR 661/12, paras 127 et seq.) referred extensively to ECHR case-law as providing interpretive guidelines for balancing the rights of the individual with those of the Catholic employer (arts. 9, 11 ECHR).

[21] CDDH(2015)R84 of 11 December 2015, para. 179.

[22] P. Eeckhout ‘Opinion 2/13 on EU Accession to the ECHR and Judicial Dialogue: Autonomy or Autarky’ 38 Fordham Int’l L.J. 955 (2015), 991.

[23] Ibid., 991-992.

[24] Justice A Bårdsen (Norwegian Supreme Court) ‘Fundamental Rights in EEA Law – The Perspective of a National Supreme Court Justice’ (12 June 2015), para. 26.

[25] S. Leutheusser-Schnarrenberger ‘Der Beitritt der EU zur EMRK: Eine schier unendliche Geschichte’ in: C. Hohmann-Dennhardt/P. Masuch/M. Villiger Grundrechte und Solidarität. Festschrift für Renate Jaeger (Kehl, Engel 2010), 135-146.

Visual ID description

Directorate of
Legal Advice and
Public International Law