The 2023 Draft Agreement on the EU Accession to the ECHR: Some Comments from the Perspective of EU Law

Joseph Ktenidis[*]

1. Introductory Remarks

The accession of the European Union (EU) to the European Convention on Human Rights (ECHR, the Convention) has been in the EU political agenda since the late 1970s.[1] However, the EU has not completed the accession process for fifteen years after the ratification of the Lisbon Treaty, which added a second paragraph to Article 6(2) of the Treaty on European Union (TEU) stipulating that “[t]he Union shall accede to the European Convention for the Protection of Human Rights and Fundamental Freedoms”.[2] Because of the wording of this provision, several commentators have argued that EU’s accession to the ECHR is mandatory, that is, the EU is under a legal obligation to accede to the Convention and the accession should have been completed shortly after the ratification of the Lisbon Treaty, in the end of 2009.[3] The lack of accession has therefore been considered by critics of the EU commitment to the rule of law a major deficit of the EU legal order. One commentator went as far as to argue that “only two other legal systems in Europe think of themselves as naturally placed above ECHR guarantees. These are Belarus and the Russian Federation”.[4]

I do not agree with this criticism. The EU accession to the ECHR is a delicate legal and political exercise. The accession agreement must balance the constitutional significance (for the EU legal order) of the accession to the Convention with the primary law reservations of Article 6(2) TEU which, right after providing the legal basis for the accession, stipulates that “[s]uch accession shall not affect the Union’s competences as defined in the Treaties”. The accession must also comply with the requirements of Protocol (No 8)Relating to Article 6(2) TEU, which forms part of primary EU law and stipulates that the accession agreement “shall make provision for preserving the specific characteristics of the Union and Union law, in particular with regard to: (a) the specific arrangements for the Union’s possible participation in the control bodies of the European Convention; (b) the mechanisms necessary to ensure that proceedings by non-Member States and individual applications are correctly addressed to Member States and/or the Union as appropriate”.[5]

In light of this highly complicated legal environment, the 2023 Draft Accession Agreement (DAA)[6] – concluded almost ten years after the Court of Justice of the European Union (CJEU, the EU Court) issued Opinion 2/13[7] which pronounced the incompatibility of the previous draft accession agreement with primary EU law – aspires to accommodate the ambiguous mandates of EU constitutional law, in order to avoid further objections by the EU Court. This highly positive – for the EU legal order and for the protection of fundamental rights in Europe – perspective is conditioned by various reservations about the DAA and the complexities of its application. Some of these complexities have been highlighted in the workshop that led to this special issue. Within this framework, I am making here three short comments.

2. Procedural “Hurdles” Before the ECtHR: Scenarios for the Applicants

My first comment is related to the title of the workshop,[8] particularly to the idea that there will be procedural “hurdles”, obviously for the applicants, when they access the European Court of Human Rights (ECtHR, the Strasbourg Court) after the DAA has entered into force. For the identification and discussion of “hurdles”, the procedural pathways available to applicants under the DAA when addressing the ECtHR should be examined.

For applicants, the accession of the EU to the ECHR and the application of the DAA shall be particularly important if their rights are violated by conduct attributed solely to EU institutions, without the involvement of an EU member state. This situation may also occur if an act is not attributed to an EU member state according to Article 1(4) of the DAA.[9] After the accession, applicants will have the right to file an application before the ECtHR directly against the EU, while today these applications are inadmissible because the EU is not a Contracting Party to the ECHR. There is no doubt that the main contribution of the EU accession to the ECHR to human rights protection is the empowerment of individuals to hold the EU directly accountable before the ECtHR for human rights violations. From the EU perspective, this development shall close a very important gap in the EU system of legal protection, and it shall end the partial immunity of the EU with respect to human rights violations. There are numerous direct actions of individuals before the EU Courts against the EU institutions, in which they complain about breach of their fundamental rights by the Union, that are adjudicated only at the EU level, without external control by the Strasbourg Court. By addressing this very important issue, the entry into force of the DAA shall complete the EU human rights protection system. I will call this type of application as the “basic scenario” after the accession.

The basic scenario does not entail any procedural challenges for the applicants. Individuals who direct applications against the EU shall not encounter any particular hurdles because all the procedural rules which apply to the Contracting Parties shall also apply, in the same manner, to the EU. In this context, the only complication which might arise when an individual files an application against the EU is that the Union could argue, as a respondent, that the human rights violation for which the applicant complains is the result of the application of primary law by the EU institutions. Accordingly, the EU institutions cannot disregard obligations under primary law, which is drafted by the member states. Therefore, the EU would claim in essence that the violation ultimately owes to the member states, who are the “masters of the Treaties”. This scenario is covered by the co-respondent mechanism established by the DAA, which, in Article 3(3), provides that the EU member states can be co-respondents together with the EU.[10]

Such cases are expected to reach the ECtHR shortly after the accession. For instance, one of the oldest complaints by individuals is related to the strict standing requirements of Article 263 of the Treaty on the Functioning of the EU (TFEU), governing applications for annulment of acts of EU institutions filed by natural or legal persons.[11] It is regularly argued that this provision is incompatible with Article 6 ECHR because it limits access of individuals to the CJEU.[12] It is anticipated that this argument will be introduced before the ECtHR in the context of an application against the EU after the accession. I think that such a development will be welcomed, especially by the CJEU, in light of the criticism it faces regarding its case law on the interpretation of Article 263 TFEU.

If the EU argues – as a (co-)respondent before the ECtHR –that a human rights violation ultimately owes to the member states because it is the result of the application by the EU institutions of primary law for which the Union has no discretion, the co-respondent mechanism shall apply according to Article 3(3) of the DAA and the member states should become parties in the proceedings. It must be noted that the co-respondent mechanism is voluntary and, eventually, a member state might choose not to join the proceedings.[13] However, this shall constitute a breach of EU law by the member state, because the member state will not comply with the EU decision, issued under Article 3(5) of the DAA,[14] holding that the criteria for the application of the co-respondent mechanism are fulfilled. The breach of EU law by the member state in this particular circumstance shall be sanctioned in accordance with EU law.

In my opinion, the fact that one or more member states might violate EU law by deciding not to join the proceedings as co-respondents does not affect substantially the position of the applicant. If all member states participate as co-respondents, they will be held together with the EU jointly responsible for the violation of the ECHR by the Strasbourg Court according to Article 3(8) of the DAA.[15] Therefore, they will be forced to amend primary EU law to terminate the violation and comply with the ECtHR judgment. However, if the EU alone, or the EU and only some of its member states are found responsible for the violation (because some or all EU member states decide not to join the proceedings as co-respondents) the EU will still be bound to execute the judgment of the ECtHR, even without amending primary EU law. In this case, the execution of the ECtHR judgment could be realised through the interpretation of primary EU law by the CJEU in accordance with the ECtHR ruling. Accordingly, I think that, as far as the basic scenario is concerned, the DAA provides efficient judicial protection to the applicant. Additionally, I think that the Bosphorus[16] presumption of equivalent protection is of no relevance to this basic scenario as the Bosphorus doctrine applies when the conduct is attributable to the EU member state and not to the EU.

A second scenario arises when the applicant complains about an act that is attributed to a member state according to Article 1(4) of the DAA, but the member state has acted in compliance with its EU obligations. This scenario shall apply, in principle, only in cases in which the member state does not have discretion when it applies EU law.[17] My view is that this scenario is of secondary importance for applicants because such applications are admissible today in principle (although they must overcome the Bosphorus presumption of equivalent protection), without the EU being a Contracting Party to the Convention. It is the Bosphorus presumption, which is a test of the ECtHR and not of the EU legal order, which affects the position of the applicant in these cases by imposing a heavy burden of proof to the applicant. Since the presumption is a rule introduced by the Strasbourg Court, it is up to this Court to abandon the presumption and relief the applicant from this burden.[18] I would assume that as long as accession is not realised, the scope of the presumption shall be reduced, as it already happens to some extent. If accession is completed, it is anticipated that the presumption – which was developed by the ECtHR in order to accommodate the EU member states obligations under EU law with the absence of the Strasbourg Court’s jurisdiction over the EU – will be totally abandoned by the ECtHR.

In the second scenario, in which the application is directed against a member state which claims it acted in compliance with its EU obligations, the co-respondent mechanism should be initiated.[19] The applicant cannot trigger the co-respondent mechanism because, according to Article 3(5) of the DAA, the co-respondent mechanism is triggered by the EU itself, the EU member states or the ECtHR and the mechanism only applies if the EU considers that the conditions for its application are met.[20] The difference in the functioning of the co-respondent mechanism in this scenario in relation to the basic scenario is that the EU (in contrast to the member states in the basic scenario) must justify on legal grounds its decision not to join the proceedings. This EU decision should be subject to judicial review by the EU Court through the action for annulment of Article 263 TFEU.

According to my opinion, it is of minor significance for the applicant whether the EU joins the proceedings before the ECtHR as a co-respondent. The applicant will not be facing the higher standard of proof arising from the Bosphorus presumption – unless the Strasbourg Court maintains the presumption after accession, which I do not believe is likely to happen. The Bosphorus presumption is the only hurdle the applicant is facing today, i.e., prior to accession, in cases falling under the second scenario; and this hurdle is expected to be removed.

In the case the EU does not join the proceedings as a co-respondent, compliance with the Strasbourg Court’s judgment shall be an obligation solely of the member state which was found to violate the Convention. It is possible that the member state will be “forced” to breach EU law in order to comply with the ECtHR judgment. The EU can avoid this tension by deciding to participate as a co-respondent in the first place. For this reason, I believe that the EU will have an incentive to participate as co-respondent in such cases and will not act with the objective to avoid responsibility. If the EU joins the proceedings as a co-respondent, both the EU and the respondent member state will be held jointly responsible for the human rights violation found. Accordingly, they should act in a co-ordinated manner in order to comply with the ECtHR judgment.

Furthermore, when the EU participates as a co-respondent, the prior involvement mechanism provided in Article 3(7) of the DAA[21] shall be activated. This mechanism may, in cases where the CJEU has not been previously involved, allow the matter to be resolved internally within the EU, before the ECtHR’s involvement. Additionally, if the EU joins the proceedings, the friendly settlement mechanism[22] will provide a procedural option to close the case at the Strasbourg level with the participation of both the Union and the member state(s).

It is possible that the applicant has a particular incentive in having the EU participate as a co-respondent in the second scenario cases. This might occur if the applicant’s objective is to hold the EU, and not a particular member state, accountable for the violation. It appears that the applicant does not have any procedural tool in order to trigger the co-respondent mechanism. However, one option for the applicant could be to direct the application against both the EU and the member state to which the act is attributed according to article 1(4) of the DAA. This scenario is envisaged in the DAA, which stipulates in Article 3(4) that, where an application is directed against and notified to both the EU and a member state, the co-respondent mechanism may apply.[23] If the applicant sues both the EU and a member state, the Strasbourg Court might undertake a more decisive role, as it will have to decide whether it shall notify the application to the EU. By notifying the application to the EU, the Strasbourg Court will indicate that the co-respondent requirements are met (unless there is another reason which justifies the position of both parties as respondents). Otherwise, the Court would have declared the application inadmissible against the EU, and it would have notified the application only to the member state, while it would have just informed the EU about the application. The EU will be aware of the Strasbourg Court’s position when it will decide authoritatively on the fulfilment of the co-respondent requirements and will probably take into account the position of the ECtHR.

The fact that the EU joins a case as a co-respondent should not place any extra burden of proof on the applicant, because the applicant should be pursuing its claims on the basis of the application directed against the member state. It is true that, instead of pleading against one party (the member state), the applicant will have to plead against two (the member state and the EU), but this is the price for potentially being awarded a judgment against two parties instead of one, who will be held jointly responsible for the violation of the ECHR. In principle, if the applicant wins, it is for others (the Committee of Ministers and the parties who lost) to resolve the issues related to the execution of the judgment and the allocation of liability among the parties. Also, I do not think that these complications should primarily concern the Strasbourg Court, which should discharge its function by deciding whether the respondent and co-respondent are jointly responsible for the human rights violation.

3. EU Secondary Legislation for the Application of the Accession Agreement

The second set of comments I would like to make is related to the internal EU procedural rules that must be adopted, in order for the Accession Agreement to operate in the EU legal order. There are no public documents to my knowledge referring to this preparation. I assume that the delay in the accession progress for more than two years after March 2023, when the DAA was finalised, is related to this issue among other concerns.

There is considerable secondary legislation that will have to be adopted at the EU level for the Accession Agreement to operate. One important question is the potential legal basis of these acts of secondary law, in light of the absence of an explicit EU competence in the area of fundamental rights. The legal basis will also define the procedure for the adoption of this legislation and the balance of powers among the member states and the EU institutions, and among the EU institutions themselves in the legislative procedure. Internal EU litigation on these questions should not be ruled out.

Obviously, it must be decided which EU institution will be competent to decide on the fulfilment of the co-respondent criteria in a timely manner. If the institution is not the CJEU (which is unlikely to my opinion) its decisions will be subject to judicial review at the EU level, which will be a considerable hurdle for the progress of the procedure before the Strasbourg Court. Also, allocation of liability among the EU and the member states for the enforcement of Strasbourg judgments will probably require arrangements within the EU in addition to the remedies already provided in EU law. Furthermore, the functioning of the prior involvement mechanism will require amendments in the Statute of the CJEU. I anticipate these developments to introduce procedural questions such as which parties shall have standing to challenge before the EU Courts the decision that the co-respondent criteria are fulfilled or not.

4. The EU as a Contracting Party to the ECHR

A last comment is related to the specific status of the EU as a prospective Contracting Party to the ECHR. In the context of this workshop, the “special” or even “privileged” position granted to the EU – in relation to all the other Contracting Parties – by the provisions of the DAA has been criticized. Furthermore, it has been argued that the EU should accede to the ECHR in the same manner as the other Contracting Parties, without any specific arrangements.[24] Many points of criticism to the complex technical provisions of the DAA originate in this argument. With respect to this discussion, it should be reminded that the EU is not sovereign, in the way sovereignty is understood as an element of statehood. It is not the master of its own legal order, the masters of the EU legal system remain the member states. This is not clearly visible in the case law of the CJEU, because the EU Court is advocating for the autonomy of the EU legal order. However, a legal principle elaborated by the CJEU such as the autonomy of the EU and its law, should not divert the attention from the reality that the present and the future of the EU legal order remain primarily in the hands of its member states.

This contradiction between the autonomy of the EU legal order and its dependence on the member states, which is inherent in the EU system, is illustrated in the legal framework governing the EU accession to the ECHR. Protocol 8 [in conjunction with the ambiguous wording of Article 6(2) TEU], which is drafted by the member states, specifically imposes substantial obstacles to the accession. If the CJEU departs from the strict requirements of Protocol 8 or other provisions of EU primary law, the member states might react through the national judiciaries, as they often do, and the efficiency (let alone the autonomy) of the EU legal order could be jeopardised. This has been very obvious in the confrontation between the EU Court and the German Constitutional Court regarding the application the Economic and Monetary Union law.[25]

The continuous tension between the EU autonomy and its dependence on the member states is mitigated by the judicial dialogue between the EU Court and national courts, which serves the objectives of preserving the efficiency of the EU legal system and safeguarding the actual application of EU rules in the territories of the member states. The preservation of the efficiency of the EU legal system is more challenging in the current stage of the historical evolution of EU, because the Union’s competences are not limited to trade, tariffs and the common market, as was the case in the early days of the European Economic Community. EU law extends today to sensitive areas such as family and private life, immigration, personal data, economic policy choices and crime prevention, for which there are different views in each member state, affected by varying approaches to the content and the scope of human rights. For such matters, individual EU member states are often in confrontation with the ECtHR because of different national -often constitutional- perceptions regarding human rights protection. Within this context, the identification of the EU as a self-standing party which is bound by the Convention, as if it was a sovereign state, disregards the unique nature of the EU and oversimplifies the challenges of its accession to the ECHR.

[*] Professor of EU Law, Faculty of Law, Aristotle University of Thessaloniki.

[1] ‘Memorandum Concerning the Accession of the European Communities to the Convention for the Protection of Human Rights and Fundamental Freedoms’ (1979) COM/1979/0210 final.

[2] Consolidated Version of the Treaty on European Union [2012] OJ C326/13, Article 6(2).

[3] Inter alia, Paul Craig, “EU Accession to the ECHR: Competence, Procedure and Substance”, (2013) 36 Fordham International Law Journal 1114.

[4] Dimitry Kochenov, “Restoring Dialogical Rule of Law in the European Union: Janus in the Mirror”, (2024) Cambridge Yearbook of European Legal Studies 1, 11.

[5] Consolidated Version of the Treaty on the Functioning of the European Union [2012] OJ C326/47, Protocol (No 8) Relating to Article 6(2) of the Treaty on European Union on the Accession of the Union to the European Convention on Human Rights and Fundamental Freedoms.

[6] Council of Europe, ‘Final Consolidated Version of the Draft Accession Instruments’ (17 March 2023) 46+1(2023)36 [2023 DAA].

[7] Opinion 2/13 Accession of the European Union to the European Convention for the Protection of Human Rights and Fundamental Freedoms [2014] ECLI:EU:C:2014:2454

[8] EU Accession to the ECHR: Procedural Hurdles and Prospects before the ECtHR.

[9] Article 1(4) 2023 DAA, note 6.

[10] Ibid., Article 3(3).

[11] Article 263 TFEU, note 5.

[12] Inter alia, Giulia Gentile, “The Power of Procedure: Fundamental Rights in the Action for Annulment before EU Courts” in Melanie Fink (ed.), “Redressing Fundamental Rights Violations by the EU. The Promise of the ‘Complete System of Remedies’” (CUP, 2024) 36.

[13] 2023 DAA, note 6, Explanatory Report, para. 62. This issue is discussed by Demi-Lee Franklin and Vassilis P. Tzevelekos, “The 2023 Draft Agreement on the EU Accession to the ECHR: Possible “Gaps” and “Cracks” in the Co-respondent Mechanism and the Implications for the Bosphorus Doctrine”, (2024) 9 European Papers 745, 762-763.

[14] Article 3(5) 2023 DAA, note 6.

[15] Ibid., Article 3(8).

[16] ECtHR, Bosphorus Hava Yolları Turizm ve Ticaret Anonim Şirketi v Ireland [GC], application no. 45036/98, 30 June 2005, paras 152-158.

[17] Article 3(2) 2023 DAA, note 6, and Explanatory Report, para. 56.

[18] Vassilis P. Tzevelekos, “The EU’s Accession to the ECHR: The Future of the Revised Draft Accession Agreement and a Call to End the Bosphorus Doctrine” (2025) 6(1) European Convention on Human Rights Law Review 1.

[19] Article 3(2) 2023 DAA, note 6.

[20] Ibid., Article 3(5).

[21] Ibid., Article 3(7).

[22] 2023 DAA, note 6, Explanatory Report, para. 69.

[23] Article 3(4) 2023 DAA, note 6 and Explanatory Report, para. 51.

[24] The Explanatory Report accompanying the DAA notes: “The EU should, as a matter of principle, accede to the Convention on an equal footing with the other High Contracting Parties, that is, with the same rights and obligations. It was, however, acknowledged that, because the EU is not a State, some adaptations would be necessary.” 2023 DAA, note 6, Explanatory Report, para. 7.

[25] Judgment of the Court (Grand Chamber) of 16 June 2015, Peter Gauweiler and Others v. Deutscher Bundestag, Case C-62/14, ECLI:EU:C:2015:400, Judgment of the Court (Grand Chamber) of 11 December 2018, Heinrich Weiss and Others, Case C-493/17. ECLI:EU:C:2018:1000, BVerfG, Urteil des Zweiten Senats, 5 Μay 2020, 2BvR 859/15.