Recalibrating the ‘Strict Obligations’ Requirement of the Bosphorus Doctrine of Equivalent Protection: The Strasbourg Court vis-à-vis the EU Principle of Mutual Trust

Abstract

Bosphorus doctrine of equivalent protection activated when states implement strict international organisations’ obligations – EU principle of mutual trust requires executing member states to presume issuing member states’ human rights compliance – Presumption influences the Bosphorus doctrine’s application – The European Court of Human Rights drew four relevant scenarios (full discretion; strict obligations; qualified discretion; EU law breach) – Case law consistent with Bosphorus logic but unfair to applicants – Court’s subsequent recalibration of ‘strict obligations’ requirement renders it moot, empties Bosphorus doctrine of its substance and equally undermines mutual trust regimes – Absence of equivalence of mutual trust regimes or EU accession to the ECHR as sole remedies

Introduction

As the EU pursues projects of further integration, fundamental rights protection represents both a means for such integration and an impediment to it. The arduous negotiations for the EU accession to the European Convention of Human Rights (ECHR), as well as the subtle balancing between human rights and the principle of mutual trust in the Area of Freedom, Security and Justice testify to that effect. Against this backdrop, the European Court of Human Rights (the Court) has trodden carefully on the question of EU member states’ human rights obligations under the ECHR when implementing EU law.

The cornerstone of its approach thereto remains the famous Bosphorus doctrine of equivalent protection. ¹ Not much needs to be said about the doctrine’s logic. As the Court has determined, when a state does ‘no more than implement [strict] legal obligations flowing from its membership’ of an international organisation, the ambit of state responsibility for human rights violation should be restricted to manifest deficiencies only, provided the said organisation offers in general, substantively and procedurally, ² equivalent human rights protection. ³ Any other approach could impede inter-state institutionalised cooperation ⁴ and would have led to a full, indirect review of the institutional act. ⁵ Conversely, when a member state enjoys discretion in the application, execution or implementation of the acts of an international organisation, it can and should proceed in a way that will not violate the ECHR and if it does not, it bears full responsibility for its acts; otherwise it could shield itself from any responsibility by transferring powers to the organisation. ⁶

This condition is crucial in the case of the EU, a supranational organisation with extensive powers that regulates various aspects of the EU citizens’ lives and exercises a multi-faceted (normative) control over member states. ⁷ Moreover, member states’ authorities constitute the implementing and enforcing vehicles of EU law, a reality that translates into a sort of dédoublement fonctionnel, where it is not always clear whether state organs act in the exercise of their own sovereign powers with wide discretion or as the executing arm of the Union implementing strict obligations stemming from EU law. ⁸

Mutual trust regimes, one of the most distinctive emanations of EU’s ever-evolving integration, illustrate how member states’ margin of discretion when implementing and enforcing EU measures is virtually a ‘moving target’. Specifically, such regimes require member states to mutually recognise and enforce in their respective national legal orders each other’s judgments and decisions ‘automatically’, ⁹ because of the trust embedded among them by the fact that they share the same values, ¹⁰ as well as common standards and equivalent processes. ¹¹  As we will see, automatic recognition and enforcement means, in simple terms, no review regarding the human rights compliance of these judgments and decisions by the executing member states, which evidently poses extraordinary challenges for human rights protection. ¹²

As a result, the Court has been embroiled in a stained dialogue with the European Court of Justice over the impact of Union cooperation schemes on fundamental rights. Specifically, the Court had to achieve a fine balance between upending mutual trust regimes by insisting that EU member states always proceed to a review concerning the human rights compliance of judgments and decisions issued in another EU member state; or undermining human rights protection if it fully deferred to mutual trust. ¹³ Hence, depending on whether mutual trust is construed within a particular regime as blind, qualified or non-existent, ¹⁴ the Court had to activate or disregard the Bosphorus doctrine by reviewing the strict obligations condition in order to balance the aforementioned conflicting strategies.

This paper focuses precisely on this case law of the Court and illustrates how the strict obligations condition has acquired a variable geometry with a view to accommodating the particularities of mutual trust regimes. On that basis, I will first systematise the four ways mutual trust and the strict obligations condition interact, highlighting how the Court eventually got these interactions right and construed in a coherent manner its Bosphorus case law. I will equally showcase that, while coherent, the doctrine’s application became unfair since it built a double hurdle (a dual presumption of equivalence) that an applicant had to overcome when arguing before the Court human rights violations stemming from EU member states acts in the framework of mutual trust regimes. Then, I will highlight how the Court became aware of this injustice and suddenly started recalibrating the strict obligations condition and, hence, distorting the Bosphorus doctrine, from the Avotiņš case onwards. This section will follow a sectoral approach examining the effects of the tournaround in the Court’s case law in the Brussels I, Brussels II and the European Arrest Warrant regimes, in that particular order. It will be argued that this volte face has increased the doctrine’s fairness at the expense of its coherence. ¹⁵ Consequently, it will be established that this ‘new era’ of the Bosphorus doctrine is fraught with serious risks due to the doctrine’s misuse and the ensuing legal uncertainty.

Mutual trust and the Strasbourg Court: building a principled, yet untenable, jurisprudence

As already explained, the free circulation of judicial and administrative decisions between EU member states in the Area of Freedom, Security and Justice relies heavily on the principles of mutual recognition ¹⁶ and mutual trust. ¹⁷ While references to the latter are relegated to the recitals of the preambles of relevant secondary EU legislation, ¹⁸ its significance cannot be overstated. The European Court of Justice has declared that this principle is ‘the raison d’être of the European Union’, ¹⁹ ‘of fundamental importance in EU law, given that it allows an area without internal borders to be created and maintained’. ²⁰ This has led certain authors to even argue that it constitutes one of the constitutional principles of EU law. ²¹

More crucially, the consequences flowing from the principle’s application are of paramount importance. The principle of mutual trust imposes on the enforcing member state an obligation to presume that the issuing EU member state is in compliance with fundamental rights. ²² The presumption translates into an obligation of the enforcing state to ‘not check whether that other member state has actually, in a specific case, observed the fundamental rights guaranteed by the EU’ when issuing the judgment/decision under enforcement, save when a ground for refusal is provided for in the related EU instrument or exceptional circumstances warrant so. ²³  Any other exception, resulting in the systematic subjection of said decisions to review by the enforcing national courts, would undermine the enhanced cooperation and the automaticity achieved in the Area of Freedom Security and Justice.

This presumption of human rights observance and the ensuing obligation of non-review, means that deference to mutual trust regimes has to be tested against the imperative purpose of respect for human rights. ²⁴ In this context, the Court must first examine whether the executing states’ obligation to abstain from a review of the issued document during the process of recognition and enforcement is a strict one or, inversely, whether these states retain a margin of discretion to refuse recognition. The Court’s task is complicated by the fact that the grounds for refusal are not uniform among the various mutual trust mechanisms. As a result, its treatment of the strict obligations condition in the framework of the equivalent protection doctrine concerning mutual trust mechanisms has been fraught with inconsistencies that have exacerbated its convoluted nature. It is this case law that most pointedly corroborates how the Court approaches ‘strict obligations’ as a concept of variable geometry, constantly adjusted as the Court’s stance vis-à-vis mutual trust evolves.

Four scenarios emerge in the interaction between grounds for refusal and the strict obligations condition.

Scenario 1: full discretion of EU member states does not activate the Bosphorus doctrine

This scenario is exemplified by the intra-EU transfers of asylum seekers pursuant to the Dublin Regulations. ²⁵ These are founded on the idea that since all EU member states respect fundamental rights, and thus constitute safe countries for return, ²⁶ the state responsible for examining an asylum request is determined according to a series of allocation criteria regulating the ‘automatic’ return to that state of an asylum seeker found in another EU member state. ²⁷

Nevertheless, any such presumption of safety ²⁸ does not entail a strict obligation for the state, where the asylum seeker is found, to return him/her to the state responsible per the allocation criteria, ²⁹ that is, without reviewing the situation in that member state. ³⁰ Article 17(1) Dublin III Regulation (recasting Article 3(2) of the Dublin II Regulation) introduces a so-called ‘sovereignty clause’, allowing a state to examine an application for international protection, ‘even if such examination is not its responsibility under the [allocation] criteria laid down in this Regulation’ (emphasis added). Consequently, one cannot speak of a strict obligation to blindly follow the presumption of safe country and return the asylum seeker to the EU member state designated per the Dublin criteria. ³¹ Thus, the Bosphorus doctrine remains inapplicable and the Court must fully scrutinise the EU member state return act.

This was confirmed by the Court, not without some initial hesitation on whether the Dublin regime imposed a strict obligation of automatic return or allowed (or even forced) sending states to set aside the allocation criteria in case of a non-refoulement risk. ³² Specifically, in the 2011 M.S.S. case, the Court concluded that Belgium’s decision to return an asylum seeker to Greece – despite the fact that the latter no longer fulfilled its obligations under the ECHR regarding the reception of asylum seekers and the treatment of their asylum requests – ‘did not strictly fall within Belgium’s international legal obligations’ because of the existence of the ‘sovereignty clause’ and, thus, the presumption of equivalent protection was not applicable in the case at hand. ³³

Scenario 2: strict obligations stemming from EU law activate the Bosphorus doctrine

This scenario is exemplified by the rules on child abduction under the Brussels IIbis Regulation. Child abduction is, generally, regulated by the 1980 Hague Convention, ³⁴ which was complemented with regard to intra-EU child abductions by said Regulation (recently recast). The latter modified the Hague Convention regarding, among others, one crucial aspect, in that it granted the requesting court (court of origin, namely, of the member state of habitual residence before abduction) the right to overrule a judgment of non-return, that was issued for specific reasons reflecting the best interests of the child by the requested court (i.e. court of the member state where the child had been removed) on the basis of Article 13 of the Hague Convention. ³⁵ More particularly, a judgment of non-return could be overturned if the requesting judge subsequently required the return per Article 11(8) of the Regulation and issued a certificate of enforcement in accordance with procedural safeguards, such as allowing prior hearing of the abducting parent and the child and taking into consideration the reasons for the order of non-return. ³⁶ This certified return judgment enjoyed procedural autonomy and had to be automatically recognised and enforced, without a declaration of enforceability, by the requested judge (Article 42(1)). ³⁷ Hence, the latter could not review whether the certified judgments respected all requisite procedural guarantees and the relevant fundamental rights of the parties involved. ³⁸  The requested judge had to presume that the process before the requesting judge complied with the requisite procedural and human rights standards and therefore any challenges against the enforceable judgment as well as any rectifications on the certificate had to be submitted before this latter judge. ³⁹

The lack of review powers of the requested court raised serious questions over the mechanism’s compatibility with human rights. ⁴⁰ What if the courts of the state of habitual residence had been negligent concerning the rights of the child or procedural guarantees leading to a flagrant denial of justice? Should the courts where the child was present blindly presume that the above procedure was compliant?

In Šneersone and Kampanella, ⁴¹ regarding a child removed from Italy to Latvia, the Court evaded these questions despite the existence of an Italian certified return judgment under Articles 11(8), 42 and 47 and the European Commission’s declaration on the dispute that the Regulation gave Italy ‘“the final say” in ordering the return, even if his/her new country of residence had declined to order the return’. ⁴² Instead, the Court reminded that under the Hague Convention the return could not be automatically or mechanically ordered, ignoring the amendments brought about by the Regulation. ⁴³

The Court’s stance was clarified in Povse. ⁴⁴ In that case, the applicants complained that the Austrian courts violated their right to private life by automatically proceeding to the recognition and enforcement of the Italian courts’ return order certified as enforceable under Article 42. In appraising the acts of the Austrian courts (namely, the requested judge), the Court declared that Article 42 left no discretion to them. ⁴⁵ In doing so, the Court sharply distinguished the above case from the M.S.S. case, where it had determined that the ‘sovereignty clause’ of the Dublin Regulation allowed states to withhold the return of an asylum seeker. ⁴⁶

The absence of state discretion signified that the Bosphorus doctrine was applicable and a looser standard of review for human rights violations was therefore employed by the Court, meaning that Austria’s presumption of compliance with the ECHR could be rebutted only in case of manifest deficiency regarding the protection of human rights. This was the case, the applicants argued, since the Austrian courts – by refusing to exercise any power of review – had deprived them of any human rights protection. ⁴⁷ Nevertheless, the Court did not deem this situation overtly problematic because the Regulation ensured that the applicants could bring a human rights complaint before the courts of origin that could grant them a stay of enforcement. ⁴⁸ For the Court, this strict division of tasks between the courts of origin and the executing courts sufficiently protected human rights in general, and could have protected the applicants’ rights in the case at hand, had the latter not failed to appeal the Italian decision. ⁴⁹ Thus, the Court did not find the protection of the applicants’ human rights before the Austrian courts on the basis of the Regulation’s relevant stipulations manifestly deficient. ⁵⁰

The Court’s line of reasoning has been criticised for strongly deferring to the Regulation’s mutual trust logic and the European Court of Justice’s interpretation thereof. ⁵¹ Yet, its shift of emphasis with regard to the element of state discretion, from its earlier hesitation over the automatic and mechanical enforcement of a return order ⁵² to the current admission that the Austrian courts did not enjoy any margin of discretion, should be welcomed as it correctly clarified which mutual trust regimes give rise to the Bosphorus presumption and which do not. ⁵³ The distinction between Povse and M.S.S. illustrates the two opposite scenarios with regard to mutual trust regimes: those where the executing member state acts under strict EU obligations; and those where EU law grants member states the faculty (or the obligation) to set aside the premise upon which mutual trust hinges. ⁵⁴

Scenario 3: qualified discretion of EU member states does not activate the Bosphorus doctrine

In addition to the two aforementioned outcomes, there is a third category of mutual trust regimes where the answer on the strict obligations condition is not straightforward. This is the case of the European Arrest Warrant Framework Decision, for instance, which enacts a mutual trust regime, where the execution of a European Arrest Warrant issued in another EU member state is, in principle, automatic, unless one of the exhaustively enumerated grounds for refusal is at play. ⁵⁵ In this regime, the executing state is either obliged to refuse enforcement when one of the mandatory grounds is applicable (Article 3 Framework Decision) or free to do so by invoking one of the optional grounds for non-execution (Articles 4 and 4a). ⁵⁶ Whereas none of these grounds concerns non-compliance with fundamental rights, ⁵⁷ Article 1(3) stipulates that the Framework Decision ‘shall not have the effect of modifying the obligation to respect fundamental rights’, this opaque reference inviting states to preserve fundamental rights. ⁵⁸  Yet, for a long time, the European Court of Justice and other EU authorities considered that supplementing or interpretatively broadening the grounds for refusal was contrary to the spirit, if not the letter, of the Framework Decision, ⁵⁹ even arguing that the narrowing down of the optional grounds by domestic legislation reinforced cooperation to the advantage of a more effective Area of Freedom, Security and Justice. ⁶⁰ Moreover, the European Court of Justice has repeatedly maintained that the Framework Decision provides for procedures that are in compliance with fundamental rights, hence member states must be presumed to respect those standards regardless of the implementation measures taken. ⁶¹ Ultimately, any complaints regarding human rights were to be raised, per the European Court of Justice, in the courts of the issuing member state – the executing courts being obliged to simply enforce the European Arrest Warrant. ⁶²

The above raises the question whether the Framework Decision imposes strict obligations on member states to automatically execute a European Arrest Warrant, or whether the courts of the executing member states enjoy a margin of discretion thereon and hence, can take into account any deficiencies in fundamental rights standards and due process in the issuing court in order to deny execution of such a warrant. ⁶³

The Court was initially evasive on the Bosphorus doctrine’s applicability herein: it recognised the ‘practically automatic’ European Arrest Warrant execution process, which should have resulted in the doctrine’s applicability, while simultaneously accepting that domestic law could include further grounds for refusal, which pointed towards increased state discretion and the doctrine’s inapplicability. ⁶⁴ Specifically, the Court did not apply the Bosphorus doctrine in Stapleton, regarding the execution by the Irish authorities of a European Arrest Warrant issued in 2005 by the British authorities for charges of fraud allegedly committed three decades ago. There, the applicant claimed that the Irish authorities failed to consider whether the execution of the European Arrest Warrant could result in a British judicial process that violated his right to fair trial (within reasonable time), to which the Court replied that there was no risk of a ‘“flagrant denial” of his Article 6 rights in the United Kingdom’. ⁶⁵

The application of the narrower ‘flagrant denial’ standard of review instead of the wider ‘real risk of unfairness in criminal proceedings’ ⁶⁶ one, as well as the assertion that the UK was presumed to comply the ECHR, ⁶⁷ could be interpreted as implicit admissions that the Bosphorus doctrine was activated and Ireland could be held responsible only in case of manifest deficiency, namely of ‘flagrant denial of justice’. ⁶⁸ But the Court’s stance is not altogether clear, since it neither explicitly applied the Bosphorus doctrine nor explained whether Ireland was strictly obliged to execute the European Arrest Warrant or enjoyed a margin of discretion thereto. ⁶⁹

This confusion regarding mutual trust regimes that allowed executing courts to examine a list of exhaustively enumerated grounds for refusal not encompassing fundamental rights violations was cleared in Ignaoua. ⁷⁰ The case concerned a group of applicants arrested in the UK pursuant to a European Arrest Warrant issued by Italy. They subsequently challenged their extradition to Italy, arguing that if sent there they risked onward removal to Tunisia, in breach of non-refoulement, and requesting the UK authorities to withhold the warrant’s execution. After rejection of their request by the UK courts, they seized the Court for a violation of Article 3 of the ECHR.

The Court was confronted with a dilemma: Were the English courts obliged to execute the European Arrest Warrant because of the absence of a relevant ground for refusal or not? At first, the Court seemed to adhere to the logic that the UK was implementing strict obligations as a result of its membership of the EU in executing the warrant. Specifically, it stressed that UK courts could (or should, because of mutual trust) presume that the Italian authorities were human rights compliant, while it also argued that the risk of non-refoulement could be adequately countered before the Italian (issuing) courts. ⁷¹ Nevertheless, the Court added that the trust and the ensuing presumption underpinning the European Arrest Warrant regime had to be accorded only some weight and consequently, were not irrebuttable. ⁷² Hence, UK courts were not subjected to strict obligations of non-review, the Bosphorus doctrine and the manifest deficiency threshold were inapplicable, and the Court could fully review their acts.

In doing so, the Court, correctly in our opinion, rejected any allegation that a European Arrest Warrant imposed strict obligations on the executing state, since Article 1(3) of the Framework Decision constitutes sufficient legal basis for taking into account in the execution phase any fundamental rights concerns. ⁷³ By disassociating from the Bosphorus doctrine some mutual trust regimes, like the European Arrest Warrant, that (implicitly) allow human rights review in the execution phase because they provide grounds for refusal, the Court highlights the continuing responsibility of issuing and executing EU member states in ensuring that such regimes remain in conformity with human rights standards. ⁷⁴

Scenario 4: breach of EU law does not activate the Bosphorus doctrine

The final scenario is exemplified by the Romeo Castaño v Belgium decision. The case concerned the enforcement by the Belgian authorities of a European Arrest Warrant issued by the Spanish courts against an individual accused of participation in a terrorist organisation (ETA) and suspected of the assassination of the applicants’ father. In reviewing the applicants’ complaint that Belgium, by refusing to execute said warrant, violated the procedural limb of Article 2 of the ECHR, the Court examined the circumstances in which Belgium could lawfully rebut the presumption of human rights compliance stemming from mutual trust ⁷⁵ and, hence, override its obligation to execute the warrant. Specifically, the Court argued that Belgium should have: first, conducted a detailed inquiry to substantiate the existence either of an individualised risk for the sought person’s rights or of structural shortcomings in Spanish detention conditions; and, second, requested more information from the Spanish authorities before deciding to refuse the surrender of the person sought by the European Arrest Warrant. ⁷⁶

The case is noteworthy in that the Court correctly avoids any reference to the Bosphorus doctrine as the case does not fall within the doctrine’s ambit. Specifically, by defying its obligation to execute the European Arrest Warrant, Belgium acts at variance with EU law. In other words, it does not implement its EU obligations; it rather violates them. ⁷⁷ This explains why the Court employs its full review standard for Belgium’s acts vis-à-vis the applicants’ rights, and not the ‘manifest deficiency’ one of the Bosphorus doctrine. The above is confirmed by the reasoning in Spasov, concerning the erroneous application of the common fisheries policy by Romania, where the Court did not examine the application of the Bosphorus doctrine despite admitting that applicable EU law imposed strict obligations, since Romania was evidently violating said policy. ⁷⁸

Interim conclusion: a principled jurisprudence on mutual trust regimes…in search of logic and fairness

After some initial hesitation (in K.R.S., Šneersone and Kampanella, Stapleton), the Court understood the need for a more consistent stance on the relation between mutual trust and the strict obligations condition of the Bosphorus doctrine and quickly enunciated a more principled jurisprudence (in M.S.S., Povse, Ignaoua).

Nevertheless, in clarifying, to a certain extent, the legal position of the executing member state, the Court, perhaps inadvertently, highlighted a paradox of its review standards regarding mutual trust regimes in scenarios 1 to 3 above. It is here submitted that the interaction between lack of state discretion, as a trigger of the Bosphorus doctrine, and mutual trust results in a catch-22 situation for the alleged victims, where the effects of mutual trust producing a dual presumption of compliance ⁷⁹ can be doubly detrimental for human rights protection. More specifically, when mutual trust does not permit the rebuttal of presumption of human rights compliance, forcing executing member states to automatically recognise a foreign decision, the application of the Bosphorus doctrine and its looser standard of human rights protection becomes inevitable due to lack of state discretion. ⁸⁰ In other words, the more restricted the review powers of the executing courts are because of mutual trust, the more restricted the review powers of the Court will be by virtue of the Bosphorus doctrine. ⁸¹ Ultimately, the interaction between the Bosphorus doctrine and mutual trust creates a double hurdle for victims of human rights violations, since they are confronted not only with the presumption of compliance with fundamental rights emanating from the principle of mutual trust, but also with the presumption of equivalent protection imposed by the Bosphorus doctrine. ⁸² Such a state of affairs renders an applicant’s chances of human rights protection at any level extremely slim.

The Bosphorus doctrine, however, is not intended to work this way. Specifically, EU measures’ (quasi-)immunity from Court’s review ⁸³ is acceptable because it is compensated by the fact that the EU legal order is generally presumed to offer equivalent protection, in terms of substance (human rights standards) and procedure (judicial review mechanisms). Regarding the latter, the presumption is applicable ⁸⁴ only when ‘the full potential of the supervisory mechanism provided for by EU law’ is employed. ⁸⁵ Hence, it is widely admitted that the procedural prong of the equivalence presumption equally encompasses human rights review by national courts qua Union courts as well as the European Court of Justice. ⁸⁶

Yet, blind mutual trust deprives executing national courts of their role in the EU judicial architecture, a role that could have benefited the applicants in their quest for human rights protection. Consequently, the ‘full potential’ of the EU legal order should not be deemed attained and the presumption of equivalent protection should be inapplicable in the case at hand. This is not, however, how the European Court of Human Rights has proceeded: by enunciating that the EU enjoys a general abstract presumption of equivalent protection, it has refrained from declaring case-by-case whether the equivalence presumption is (in-)applicable in mutual trust regimes, ⁸⁷ instead trying to regulate the inadvertent effects of blind mutual trust via the other condition of the Bosphorus doctrine, that of strict obligations. Put differently, instead of rendering the doctrine inapplicable because part of the procedural equivalence is missing when national courts must blindly trust their counterparts under EU law, which would have been a radical move on its part, the Court instead focuses on the strict obligations condition to contain the impact of the mutual trust principle, with the paradoxical results that I have just highlighted.

Consequently, this stabilisation phase fleshes out with consistency the logic of mutual trust in the Court’s case law: when there are no grounds for refusal, there is no state discretion and the Bosphorus doctrine of equivalent protection is activated, whereas when there are such grounds (public policy or implicit/explicit human rights incompatibility ones), EU member states enjoy discretion and, thus, the Bosphorus doctrine is inapplicable. Yet, this consistent and conforming to mutual trust approach by the Court establishes a double presumption at the expense of the applicant, since the (quasi-)immunity before the Court is not compensated within the EU judicial architecture.

Mutual trust and the Strasbourg Court: the volte-face

The inadequate effects of the above correlation for human rights protection showcases the Court’s difficulty in accommodating mutual trust while countering its unintended results. It took, thus, a matter of time for the Court to recalibrate its jurisprudence on mutual trust by exploiting the many facets (or variable geometry) of state discretion within mutual trust regimes. The following parts do not follow the order of the previously expounded scenarios; instead, I adopt a sectoral approach (Brussels I, Brussels II and then European Arrest Warrant regimes), showcasing the logical inconsistencies caused in each regime by the shift in (and, at times, direct overruling of) the Court’s case law.

Avotinš v Latvia: the ‘strict obligations’ requirement and the Brussels Ibis Regulation

The Avotinš case constitutes the turning point for this recalibration. The application concerned a case of recognition and enforcement by the Latvian courts of a civil judgment issued in Cyprus. The applicant had asked the Latvian courts to reject enforcement, since he had allegedly not been summoned to appear before the Cypriot courts and thus, had not been able to exercise his defence rights. ⁸⁸ The Supreme Court of Latvia dismissed his claim without submitting a request for a preliminary ruling to the European Court of Justice, ⁸⁹ noting that he could have appealed the judgment in the Cypriot courts, and ordered the judgment’s recognition and enforcement, since the Latvian courts could not review it under the Brussels I Regulation. ⁹⁰ Mr Avotinš subsequently seized the Court, arguing that the Latvian courts had breached his right to a fair hearing (Article 6(1) ECHR).

The Court was called upon to adjudge the human rights compliance of the Latvian courts’ treatment of Mr Avotinš’ claims on the basis of the Regulation, whose underlying logic is that of mutual trust. ⁹¹ The said Regulation, in juxtaposition to Brussels IIbis Regulation, provided in Articles 34–35 a series of grounds for refusal to be considered when declaring enforceability. ⁹² Hence, the Court had to determine whether the Bosphorus doctrine was applicable or not – namely, whether the executing courts enjoyed a margin of discretion or not to assess if the fundamental rights of the concerned individual were respected in the issuing courts. ⁹³ The case resembled Ignaoua, where the Court had refrained from applying the Bosphorus doctrine. Nevertheless, in the present case, the Grand Chamber of the Court determined that Article 34(2) of the Regulation ‘did not confer any discretion on the court from which the declaration of enforceability was sought’ because it was a precise, limited ground for refusal and had been extensively analysed by the European Court of Justice. ⁹⁴ The Court’s conclusion could not be overridden, it argued, by the fact that Article 34(1) consecrated public policy as a further ground for refusal, because the applicant had not invoked that clause. ⁹⁵

There is no doubt that the Court’s analysis is permeated by the logic of mutual trust, which is mentioned when the Court highlights how the presumption of human rights observance resulting from mutual trust deprives the enforcing court of its discretion in the matter. Accordingly, the latter court is compelled to render the judgment enforceable in a virtually automatic way ‘after purely formal checks of the documents supplied, without there being any possibility for the court to raise of its own motion any of the grounds for non-enforcement provided for by this Regulation’. ⁹⁶ In the same vein, the 2014 judgment had observed that due to mutual trust, the Latvian Supreme Court was ‘under a duty to ensure the recognition and the rapid and effective enforcement of the Cypriot judgment in Latvia’. ⁹⁷

Nevertheless, mutual trust by itself cannot eradicate all margin of appreciation, as the Court readily admitted in Ignaoua. And the same applies to the case at hand, ⁹⁸ since the Regulation allows the requested state to refuse recognition or enforcement, in the mould of the Dublin Regulations, which allow states to depart from the responsibility allocation criteria. Yet, the Court curiously juxtaposes the two regimes. The Court argues that whereas the Dublin Regulations empower the member state to proceed to a full assessment of its own, that is not the case with this Regulation. Here, the Court asserts, the executing member state has no margin of discretion when the conditions for the application of a refusal ground are fulfilled; it can only abstain from enforcement. ⁹⁹ Conversely, if no ground for refusal is applicable, requested courts have no discretion to deny enforcement. In such a case, per the Court’s reasoning, the requested courts only enjoy a margin of discretion with regard to the factual assessment of the case – which is not the same as a margin of discretion to judicially review the about-to-be-enforced judgment. ¹⁰⁰ These restrictions are further compounded by the fact that, according to the Court, the European Court of Justice had already interpreted in detail the contours of the refusal grounds in Article 34. ¹⁰¹

With respect, this argument is flawed. ¹⁰² First, by fully assessing the factual background of the case, the executing judge must determine that the courts of origin properly served the defendant with the document so that he could prepare his defence or take steps to prevent a decision delivered in default of appearance. ¹⁰³ As the European Court of Justice has expounded, the executing court ‘must be satisfied that all investigations required by the principles of diligence and good faith have been undertaken to trace the defendant’, ¹⁰⁴ even if a certificate indicating the date of service has been issued by the courts of origin (Article 54 of the Regulation). ¹⁰⁵ Moreover, the executing judge must assess whether there were available, effective remedies in the member state of origin in order to challenge the defective judgment there; in case of absence, and if the default judgment is in manifest and disproportionate breach of the right to a fair trial under Article 6 of the ECHR and Article 47(2) of the Charter of Fundamental Rights, the executing judge must deny enforcement. ¹⁰⁶

Furthermore, the European Court of Justice’s interpretation of the relevant refusal grounds in the Regulation has not fully clarified their ambit. It is, for instance, unclear whether manifest violations of fundamental rights fall under Article 34(1) or 34(2) or whether they constitute a separate reason for refusing execution, ¹⁰⁷ or what the standard of proof for the applicant should be. ¹⁰⁸ More crucially, there is a continuing debate about the executing court’s power to take into account refusal grounds that were not raised by the concerned individual when challenging the foreign judgment’s enforcement, the answer to which impacts decisively on the state discretion requirement for the triggering of the Bosphorus doctrine. ¹⁰⁹ In Avotinš, for instance, whereas the Grand Chamber argued initially that since the applicant had not raised the public policy defence, the Latvian courts could not examine it ex officio, ¹¹⁰ it ultimately held that when there is a manifest deficiency in the protection of fundamental rights, the executing courts ‘cannot refrain from examining the complaint on the sole ground that they are applying EU law’. ¹¹¹ In between can be placed the view that the executing courts may take into account other grounds for refusal proprio motu. ¹¹² Thus, mutual trust might not be equated with blind trust, ¹¹³ even where the applicant has refrained from raising the most appropriate refusal ground.

In conclusion, the Court probably rejected correctly the applicant’s plea by considering that there was no case of manifest deficiency for the rebuttal of the Bosphorus doctrine because the applicant’s failure to pursue a remedy before the Cypriot courts showed lack of due diligence. ¹¹⁴ Yet, the doctrine’s application owing to the absence of a margin of discretion by the executing courts misrepresents the logic of the Brussels I Regulation. ¹¹⁵ More particularly, the Court should have juxtaposed Avotinš to Povse (and not to M.S.S.) for two reasons: first, the Latvian courts enjoyed a certain margin of discretion regarding the factual background; and second, they could have probably examined other grounds for refusing execution, in contrast to the Brussels IIbis Regulation, where the executing courts must automatically and mechanically recognise the foreign judgment. ¹¹⁶ In reality, the Latvian courts’ margin of discretion is surreptitiously recognised by the Court, when the latter criticises the former for not verifying whether the conditions for the disapplication of Article 34(2) were fulfilled. Thus, the Court invites the Latvian courts to sideline mutual trust for the sake of human rights protection, while simultaneously arguing that those courts are under a strict obligation to enforce the judgment, and hence the Bosphorus doctrine is applicable.

Royer v Hungary, Rinau v Lithuania et al.: child abduction cases and the Bosphorus doctrine

The Court’s change of course concerning the strict obligations condition of the Bosphorus doctrine was not made immediately apparent, because Avotinš was the first judgment on the Brussels I Regulation. Afterwards, when the Court revisited its case law on the Brussels IIbis Regulation and the European Arrest Warrant, the extent of the Court’s turnaround became evident.

First, the Court reversed its previous pronouncements on the Bosphorus doctrine regarding Article 11(8) Brussels IIbis Regulation. In Povse, it had applied said doctrine, correctly stating that the requested courts were implementing strict obligations under Article 11(8). ¹¹⁷ Consequently, mutual trust imposed return procedures, whose implementation could not be conditioned on fundamental rights considerations, and compelled those courts to execute return requests automatically. ¹¹⁸

This was called into question in Sévère, where the father of a child abducted by the mother from Italy to Austria complained that the Austrian courts had violated his right to family life, by refusing to enforce the child’s return on grounds that this could cause severe psychological harm to the child (Article 13(b) Hague Convention) despite the fact that Italian authorities had made arrangements for the child’s return per the request of the Austrian courts (Article 11(4) Brussels IIbis Regulation). Whereas the case did not concern the automatic enforcement procedure of Article 11(8), ¹¹⁹ the Court’s pronouncements departed from mutual trust’s logic as exemplified in Povse. Specifically, the Court observed that the immediate return of the child – envisaged by Article 11(4) – corresponded only to one specific conception of the best interests of the child, leaving the door open for the requested court to lawfully refuse return despite the provision’s language. ¹²⁰

Having made a first step away from the logic of mutual trust, the Court then extended the above reasoning to Article 11(8) in Royer. Mr Royer complained that the Hungarian courts, by refusing to execute a judgment ordering his son’s return (delivered by a French court and accompanied by a certificate of enforcement under Article 39 of the Regulation) on grounds that it was contrary to public policy (Article 23(a)), had violated his right to respect for family life (Article 8 ECHR). ¹²¹ The Court reiterated the Regulation’s reliance on mutual trust, but it then held in an obiter dictum that, even in the automatic procedure of Articles 11(8) and 42, the requested courts ‘must verify that the principle of mutual recognition is not applied automatically and mechanically’. ¹²² In other words, Povse’s reference to the strict obligation of the requested court in the framework of Article 11(8) to automatically enforce a judgment issued by the requesting court – which also meant the applicability of the Bosphorus doctrine ¹²³ – suddenly vanished into thin air. Instead, the Court returned to its pronouncements on child returns under the Hague Convention, ¹²⁴ which notably lacks the Article 11(8) procedure, and to its reasoning in Avotinš, which however concerned a Regulation that allowed for refusal grounds to be invoked and taken into account by the requested courts.

The departure from Povse in Royer is obvious, since the Court recognises the competence of the requested courts to review whether the application of the automatic recognition mechanism of Article 11(8) violates the fundamental rights of the child. ¹²⁵ Moreover, it avoids any reference whatsoever to the jargon and the conditions of the Bosphorus jurisprudence – save for the citation of the Avotinš case where Bosphorus was employed – giving, thus, the impression that the procedure under Article 11(8) does not involve the fulfilment of any strict obligations by the requested court. ¹²⁶

This latter point was revisited by the Court in two subsequent judgments. In the O.C.I. and Others v Romania case, which concerned a complaint that the execution of an Italian return judgment by the Romanian courts violated Article 8 ECHR as no proper consideration was given to the family’s situation and the risk of ill-treatment of the children by their Italian father, the Court retained its distance from the Bosphorus doctrine. It particularly stressed, when examining the impact of mutual trust on the Brussels IIbis Regulation, that nothing at all therein (i.e. not even the Articles 11(8) and 42 procedure) obliged the requested courts to enforce a return order that would expose the children to a grave risk of domestic violence. ¹²⁷ Hence, in O.C.I. the Court persisted upon the absence of any strict obligations of the executing courts to follow up on a return request and did not apply the Bosphorus doctrine. ¹²⁸

In contrast, in Rinau the Court implicitly adhered to the logic of the Bosphorus doctrine, since it held that ‘where the courts of a State … are called upon to apply a mutual recognition mechanism established by EU law, they must give full effect to that mechanism where the protection of Convention rights cannot be considered manifestly deficient’. Instead, when a complaint of manifest deficiency is raised before them, ‘they cannot refrain from examining that complaint on the sole ground that they are applying EU law’. ¹²⁹

If one accepts that the Court is confirming the Bosphorus doctrine’s applicability under Article 11(8) Brussels IIbis Regulation in Rinau, one is confronted with both a fundamental inconsistency on the Bosphorus conditions’ orthodoxy and an astonishing solution to the paradox of the double hurdle described above. In stressing that a mutual trust regime cannot translate into lack of review by the requested courts, the Court encourages those courts (and not only the requesting courts) to allow remedial action by the applicants, while still restricting under the Bosphorus doctrine the standard of review before it. Thus, a balance is achieved between a more or less generous review in domestic (requesting and/or requested) courts and a very narrow one before Strasbourg. This solution, however, is problematic in two respects. First, it undermines the mutual trust logic of those cooperation mechanisms, as it invites requested courts to review the issued judgment. Second, it erodes the essence of the Bosphorus doctrine, since it sidelines the condition of ‘implementing strict obligations’ as its cornerstone because requested states can/must proceed to a human rights review.

From Pirozzi v Belgium to Bivolaru and Moldovan v France: confirming the court’s new strategy when reviewing the execution of a European Arrest Warrant

The shift in the Court’s case law regarding mutual trust regimes can be further observed in cases concerning the execution of European Arrest Warrants. As has been shown (in Ignaoua), the Court abstained from applying the Bosphorus doctrine because it determined that the executing courts enjoyed a certain margin of discretion. It, then, proceeded to a full review, avoiding any reference to the standard of manifest deficiency. The Court revisited the question in Pirozzi, a case concerning the execution by the Belgian authorities of a European Arrest Warrant issued after the condemnation of Mr Pirozzi for drug trafficking in Italy in an in absentia trial. Mr Pirozzi argued that the execution of the warrant violated his fair trial rights as the in absentia Italian judgment could not be challenged, leading to a flagrant denial of justice. ¹³⁰

After having observed that the Framework Decision is premised on mutual trust, and having noted the latter’s importance for an integrated cooperation in the Area of Freedom, Security and Justice, the Court found the said mechanism, in principle, compliant with the ECHR. ¹³¹ It further declared that the Framework Decision imposed on the Belgian authorities an obligation to presume that their Italian counterparts respect human rights, thus allowing them no margin of discretion, a finding that placed their actions squarely within the ambit of the Bosphorus doctrine (but at odds with Ignaoua). ¹³² Having said this, the Court immediately tempered the aforementioned pronouncements. It actually stressed that when a serious claim of manifest deficiency was raised, the Belgian authorities could not use EU law as an excuse for ignoring the claim, but had to apply the Framework Decision in conformity with the ECHR. ¹³³ Specifically, they had to review the complaint and desist from implementing the warrant even if the manifest deficiency did not fall within the scope of one of the refusal grounds. ¹³⁴ In Pirozzi, the Court considered that the Belgian authorities had satisfied the aforementioned requirements and, hence, no violation was found since the applicant was represented by his lawyer in the Italian proceedings. ¹³⁵

The Court’s reasoning draws a fine line between the executing courts’ absence of a margin of discretion, since it is the issuing state that is presumed to have taken care of human rights considerations, ¹³⁶ and their opposite obligation to not validate a European Arrest Warrant tainted by a manifest deficiency concerning human rights guarantees. As explained above, it is not clear how these two aspects of the Court’s reasoning can be reconciled. An executing court would either enjoy no margin of discretion to review a European Arrest Warrant save for the refusal grounds mentioned in the Framework Decision (and hence, Bosphorus is applicable), or it would have the power to scrutinise allegations of manifest deficiencies beyond the refusal grounds (and hence, Bosphorus is non-applicable); ¹³⁷ it cannot do both as the two are mutually exclusive. ¹³⁸ In any case, the Court adheres here to the Bosphorus logic, implicitly accepting the absence of state discretion, since it employs the Bosphorus standard of ‘manifest deficiency’, as reflected in the flagrant denial of justice threshold. ¹³⁹

The ambiguity created by the Court’s oscillation between asserting the existence of a margin of discretion and, thus, rejecting the applicability of the Bosphorus doctrine, in Ignaoua, on the one hand, and fully endorsing the doctrine by virtue of the absence of a margin of discretion in Pirozzi, on the other, did not abate afterwards. In Romeo Castaño, the Court, while referring to the mutual trust logic underlying the European Arrest Warrant, ¹⁴⁰ pays lip service to it. Specifically, by finding Belgium in breach of the ECHR because its authorities did not proceed to a sufficiently thorough examination and did not request additional information before withholding the execution of the European Arrest Warrant, the Court evidently invites Belgium to exercise a wide margin of discretion on whether to execute said warrant or not. ¹⁴¹ Yet, the Court omits any explanation on how such a margin is compatible with mutual trust and the corollary presumption of human rights compliance that deprives executing courts of the power to review said warrants save for the strictly interpreted refusal grounds or under exceptional circumstances. Instead, it merely notes that the suggested course of action is afforded by Belgian law. ¹⁴²

The Bivolaru and Moldovan cases highlight the full revamping of the strict obligations condition. These concerned two European Arrest Warrants issued by Romanian authorities – and aiming at the return of the two applicants in order to serve their sentences – that the French authorities were called upon to execute; ¹⁴³ and that is virtually the only similarity between the two joined cases. In the Moldovan case, the applicant raised before the French judge a detailed defence against refoulement due to the systemic/generalised deficiencies concerning the incarceration conditions in Romania and the consequent individual risk of being subjected to inhumane and degrading treatment. Nevertheless, France executed the warrant after receiving further information and assurances by the Romanian authorities. ¹⁴⁴

The Court fleshed out its reasoning in two steps. First, it affirmed the application of the Bosphorus doctrine in relation to the execution of a European Arrest Warrant. In a déjà-vu from Avotinš, the Court adjudged that refusal grounds in the Framework Decision were restrictively defined and exhaustively interpreted by the European Court of Justice. ¹⁴⁵ Thus, while national authorities could refuse the execution of a European Arrest Warrant, their margin of discretion was radically circumscribed and related only to the factual appraisal of the case, leading the Court to conclude that the French authorities were under a strict obligation to give effect to the Romanian request. ¹⁴⁶ This conclusion equally foreshadowed the Court’s elaboration on the second prerequisite for the activation of the Bosphorus doctrine. Specifically, the Court considered that due to the repeated treatment of the relevant legal issues by the Luxembourg Court, there was no need for a preliminary reference, thus the full potential of EU law had been deployed. Consequently, both conditions of the Bosphorus doctrine were fulfilled. ¹⁴⁷ Despite the narrower review standard, the Court found that the presumption of conformity with the ECHR was rebutted and France had breached Article 3, because the procedure before the French authorities was tainted by manifest deficiencies regarding the applicant’s human rights. ¹⁴⁸

The Court’s treatment of the strict obligations requirement leaves again a lot to be desired. Specifically, its admission that the executing authorities enjoy a margin of discretion that is only factual, and thus not autonomous, because the European Court of Justice has methodically defined the contours of the refusal grounds, is problematic. ¹⁴⁹ On the one hand, the European Court of Human Rights itself concedes that this margin of factual appreciation is complemented by the power to determine the legal consequences thereof, which increases the discretionary power of the French authorities. ¹⁵⁰ This is all the more so if one takes into account that France was found in violation of the ECHR precisely because its authorities did not draw the correct consequences from the appraisal of the facts. In other words, while the Court invites the French judge to systematically review the situation in Romania and draw the right inferences therefrom, it still insists that this national judge is subject to a strict non-review obligation stemming from Union law concerning the execution of a European Arrest Warrant. ¹⁵¹ On the other hand, presenting the European Court of Justice’s case law on the fundamental rights exception to (blind) mutual trust as settled is misleading. For instance, there are still clear discrepancies between Strasbourg and Luxembourg on the necessary test and threshold for the rebuttal of mutual trust. ¹⁵² Moreover, the numerous preliminary references addressed to the European Court of Justice, before and after Bivolaru and Moldovan, concerning the fundamental rights exception to the execution of a European Arrest Warrant, highlight the gray zones remaining in the European Court of Justice’s case law post-Aranyosi and Căldăraru. ¹⁵³

Furthermore, it should be clarified that even if one concedes that the Aranyosi and Căldăraru case and its progenies have imposed on executing member states a strict obligation to review the European Arrest Warrant’s compliance with human rights in exceptional circumstances (and have not simply acknowledged their faculty to do so), ¹⁵⁴ such strict obligation does not activate the Bosphorus doctrine. The latter was devised as a leeway towards member states forced to implement a Union measure that runs counter to their human rights obligations; it was not, however, meant to apply when the EU enacted a strict member state obligation to respect human rights that these states overrode, thus violating human rights. Put it differently, by breaching the human rights safeguards enunciated in the relevant EU act and case law, member states already defied their strict EU obligations and, thus, they were enjoying a margin of discretion obtained by their unlawful conduct. This too confirms that the Court erred in not exercising its full review powers.

The Court’s denial of reality over the ‘strict obligations’ condition of the Bosphorus doctrine in the case of a European Arrest Warrant execution becomes even more glaring when examining other facets of the judgment. Specifically, in the Bivolaru case, the applicant had fled Romania and was granted refugee status by Sweden because of a risk of persecution for his religious and other convictions if returned there. When the French authorities were requested to implement the warrant, they had to take into account the above elements, as well as the detention conditions in Romania. ¹⁵⁵ They ultimately proceeded to the execution of the warrant while rejecting the applicant’s request for a preliminary reference to the European Court of Justice, though there were no clear instructions by the latter regarding the impact of the refugee status on the execution of a European Arrest Warrant. ¹⁵⁶ For that reason, the Court found that the full potential of the Union review mechanism was not deployed and, hence, the Bosphorus doctrine was inapplicable for this part of Mr Bivolaru’s claims due to the absence of equivalence. ¹⁵⁷ Nevertheless, France was not found in violation of Article 3 of the ECHR because the applicant had failed to fully substantiate his claims.

Thus, the outcome of the Bivolaru and Moldovan cases is that France was condemned when the Bosphorus doctrine was applicable and exculpated when it was not. As has been pertinently observed, this is due to the fact that, curiously, the Court’s review is more intensive when it applies the Bosphorus presumption than when it does not. ¹⁵⁸ Such an inconsistent attitude can be easily explained. On the one side, the Court still argues that member states are subject to strict obligations when executing a European Arrest Warrant, thus activating the Bosphorus doctrine. On the other, it simultaneously exercises pressure on the respondent states to undertake an increasingly expansive review of the warrants’ compliance with human rights and to set them aside if implementation reveals relevant deficiencies. Such an attitude, while showing at first deference to the principle of mutual trust, ends up seriously undermining it, as it requires a more intensive review by the executing courts. Ultimately, the Bosphorus doctrine is applied together with the recognition that states enjoy a considerable margin of manœuvre and, hence, can compensate through their own review powers for the Court’s narrower review due to the presumption of equivalent protection. Or, conversely, the Bosphorus doctrine is only nominally applied, since the Court exercises an intensive review when national courts do not employ their margin of manœuvre to review any warrants, a power that should have disqualified the application of the doctrine in the first place.

Concluding observations

The above considerations reveal the challenges the Court faces in its attempt to conceptualise in an effective way the interaction between mutual trust and the ECHR. While such regimes are at the heart of the integration process within the EU, the Court’s stance highlights the difficulties when navigating through the subtleties of the Union order. Specifically, the analysis of how the Bosphorus condition of ‘strict obligations’ has fared in mutual trust cases before the Court has brought to the fore a series of inconsistencies. While hesitant at first, the Court quickly developed a principled jurisprudence on whether state discretion is present in mutual trust regimes. It, accordingly, argued that when EU law does not provide grounds for refusal of recognition or enforcement, the requested state enjoys no state discretion and, hence, the Bosphorus doctrine becomes applicable. Conversely, when the relevant EU instruments provide for refusal grounds or grant member states the ability to override mutual trust altogether, then obviously requested member states have discretion and their acts in implementing the aforementioned Union rules are subject to full review under the ECHR. Yet, this admission creates a paradox at the heart of the Court’s treatment of mutual trust regimes: When states enjoy no discretion – and thus they cannot proceed to a human rights review when implementing EU law – the Court applies the Bosporus doctrine and, consequently, it also restricts the intensity of its conventionality review vis-à-vis member states’ acts. This double presumption of human rights compliance, stemming from the principle of mutual trust (executing courts shall not review) as well as the Bosphorus doctrine (the Court shall not review except for manifest deficiencies), was doubly damaging for alleged victims of human rights violations, whose applications before the Court would have had few chances of success.

For those reasons, the Court, as soon as it stabilised its jurisprudence on mutual trust regimes, proceeded to perform a turnaround. Specifically, it argued that in cases of state discretion the Bosphorus doctrine was applicable (Avotinš or Moldovan), while in cases of no discretion the Court could proceed to full review of state acts implementing Union rules that are premised on mutual trust (Sevère, Royer and O.C.I.; Rinau being an outlier). This fiction allows the Court to exercise a more intensive control when mutual trust prevents the requested state from doing so and, conversely, apply the manifest deficiency threshold when requested states are enabled within mutual trust regimes to review judgments and decisions issued in another member state. Nevertheless, it completely distorts the logic and place of the ‘strict obligations’ condition within the Bosphorus doctrine. This raises questions about the condition’s – or the whole doctrine’s – continuing pertinence and usefulness, particularly when it comes to mutual trust regimes.

Actually, it is difficult for the Court to persuasively sustain that states enjoy no margin of discretion when implementing obligations stemming from mutual trust regimes and in the same breath suggest that such regimes cannot be enforced automatically and mechanically as there is an obligation upon member states to review other EU member states’ acts for human rights deficiencies. This undermines the effectiveness of the judicial protection it purports to offer and does not even serve its struggling relationship with EU law and the EU legal order. Ultimately, this case law constitutes a cautionary tale on the pitfalls and blind spots of the Bosphorus doctrine that have mushroomed through the years of its existence, courtesy of the Court’s piecemeal approach that makes it challenging to coherently systematise its case law.

A solution, as I have argued, would be for the Court to focus on the other condition of the doctrine, namely the inadequate equivalence of human rights protection in mutual trust regimes. Instead, it continues to apply the Bosphorus doctrine, calling on requested courts to review manifest human rights deficiencies. ¹⁵⁹ Yet, this standard should be applicable only by the Court and only because human rights protection is equivalent in the EU legal order. Equivalence cannot be satisfied when requested EU member states review manifest deficiencies only – it warrants a standard that is similar to the full ECHR standard, not a substandard one. In the end, a review regarding manifest human rights deficiencies is all it takes for the requested EU member states to evade violation of the ECHR, transforming the latter into a weak protective framework.

For all these reasons, the agreement on the EU accession to the ECHR ¹⁶⁰ represents the only way out, since it stipulates that the EU will be subject to the Court’s control along with its member states under the co-respondent mechanism (Article 3). Hence, the Bosphorus doctrine seems to have no place in the Court’s post-accession case law. ¹⁶¹ Any other outcome runs counter to the principle of equal footing, which constitutes a cornerstone of the renewed negotiations. ¹⁶² Besides, the declaration on mutual trust (Article 6) ¹⁶³ included in the said agreement mainly restates the European Court of Justice’s position thereon, and reflects the partial jurisprudential convergence of the two Courts. Thus, it primarily purports to appease the Luxembourg Court before it re-appraises the agreement’s compatibility with EU law. ¹⁶⁴ Consequently, the agreement offers a unique opportunity to put to rest, in relation to the EU, this convoluted doctrine.