No need to look, trust me! Mutual trust and distrust in the European arrest warrant system

Abstract

No cooperative scheme in EU law has displayed bigger tensions between mutual trust and fundamental rights protection than the EAW system. Despite the requirement developed by the CJEU for national courts to trust each other and recognise each other’s arrest warrants, the reality on the ground has shown high levels of distrust between national courts regarding Member States’ alignment with core EU values. In this contribution, we analyze how the CJEU has managed such tensions in the EAW system. To that effect, we first put the Court’s EAW case law into context by examining the broader language of mutual trust used by the Court in other fields of EU law. In doing so, we point out how the Court has espoused different levels of lawful distrust to be exercised in different circumstances under the scope of application of mutual trust. Given that broader context, it is contradictory for the Court to mainly view mutual trust as a requirement rather than a reality in need of permanent and continuing justification between national authorities. The latter conception of mutual trust is more apt to be the basis of EU horizontal cooperation, which must be value-based and sincere according to the Treaties. Therefore, we propose a bidimensional account of mutual trust as a legal principle, one that accommodates both trust and distrust as tools for managing the uncertainty and dynamic nature of trust-based cooperation. Finally, we explore how such account of mutual (dis)trust can be concretised by the Court and other political institutions.

1. Introduction

In its Opinion 2/13 on the European Union’s accession to the European Convention on Human Rights (ECHR), the Court of Justice of the European Union (CJEU, or ‘the Court’) famously described mutual trust as a principle ‘of fundamental importance in EU law’. ¹ While many criticised the reasoning and conclusions of the Court in the Opinion, ² its acknowledgement of the importance of the principle of mutual trust is uncontroversial. The principle plays a key role in the European Union (EU) legal order today: notably, it works as the basis of multiple schemes of mutual recognition in the Area of Freedom, Security and Justice (AFSJ), but as we will explore later, ³ the Court of Justice and the EU legislature have referred to and operationalised mutual trust in various other policy fields as well.

If the importance of mutual trust is undeniable, its precise meaning and, most importantly, its concrete legal consequences remain contested. The concept is not even mentioned, let alone defined, in the EU Treaties. We find references to mutual trust in secondary legislation, ⁴ most often in legislative preambles, but those do not provide full clarity either. It was the Court of Justice itself that, on the basis of these references to the concept of mutual trust in legislation and policy documents, introduced mutual trust as a legal principle and then developed it gradually by fleshing out its foundations, meaning, and legal effects. The Court has been helped, in this respect, by national courts in the context of the preliminary reference procedure, where the latter have often interrogated the Court of Justice on the precise implications of, and limits to, mutual trust.

In particular, national courts have asked how to balance mutual trust with fundamental rights and rule of law obligations, given the inherent tensions between the automaticity imposed by mutual trust and mutual recognition, on the one hand, and the individual checks required by fundamental rights law on the other. ⁵ These tensions between mutual trust and fundamental rights, and the consequent imposition of possible limits to mutual trust have, as Wendel put it, a distinct ‘federal’ impact, ⁶ as they affect the structure of cooperation in the AFSJ and determine the horizontal distribution of fundamental rights responsibilities among Member States. In deciding whether to require mutual trust and automatic mutual recognition from national authorities, or, conversely, to allow them to impose certain limits to horizontal cooperation, the EU system decides how to allocate the responsibility to uphold fundamental rights and values: in the first scenario, responsibilities lie exclusively with the Member State where, for example, a judicial decision has been produced; in the latter, they are shared with the Member State that is asked to recognise that judicial decision, whose national authorities can, and in fact should, verify the compatibility of that decision with fundamental rights standards, at least in certain occasions.

The dialogue on these tensions between national courts and the CJEU has been particularly fervid in the context of the European Arrest Warrant (EAW) scheme, an instrument that has, since its very inception, ⁷ generated several concerns at the national and EU level as to its compatibility with national and European constitutional principles. ⁸ The European Arrest Warrant Framework Decision (EAWFD) brought a radical change to the system of criminal cooperation replacing extradition procedures with a system of automatic mutual recognition of judicial decisions, which only offers limited possibilities for national judicial authorities to refuse the execution of a EAW on the basis of an exhaustive list of mandatory or optional grounds for refusal. ⁹ Fundamental rights concerns do not explicitly feature in that list, even if Article 1(3) states that the EAWFD ‘shall not have the effect of modifying the obligation to respect fundamental rights and fundamental legal principles as enshrined in Article 6 of the Treaty on European Union’. In Radu, one of the earliest decisions of the CJEU on mutual trust in the EAWFD, the Court rejected the view that fundamental rights considerations could be considered as additional – implicit – grounds for non-execution of EAWs. ¹⁰

The decision in Radu did not, however, end the discussion on the limits to mutual trust and on the tension between mutual recognition and respect for fundamental rights that is at the basis of the EAW system. In the years thereafter, national courts have continued to call upon the Court to define the relationship between mutual trust, mutual recognition, and EU values. This has become especially important in the context of the progressive erosion of fundamental rights and rule of law guarantees caused by constitutional backsliding processes in different Member States, ¹¹ which has created new tensions in the system and brought to light additional difficulties for mutual trust, within and beyond the EAW. ¹² The existence of mutual trust is implied and justified by the premiss ‘that each Member State shares with all the other Member States, and recognises that they share with it, a set of common values on which the EU is founded, as stated in Article 2 TEU’. ¹³ But what if it becomes clear that these values are not adequately protected and guaranteed in all Member States? Can national authorities be forced and required to trust each other when distrust seems to prevail in practice? And, at a more fundamental level, is it really trust what underlies and informs the reciprocal execution of EAWs and judicial decisions?

In recent years, national courts have thus repeatedly asked the CJEU for guidance on how to manage the tensions that constitutional backsliding puts on the EAW and other mutual trust and mutual recognition schemes. What is more, they have continued to pose other fundamental questions on how to balance mutual trust, fundamental rights, and the rule of law. The Court has responded by establishing a rather rigid two-step test for the suspension and ultimate refusal of the execution of a EAW, ¹⁴ and has since then always maintained it. ¹⁵ The exercise by Member States of ‘lawful distrust’ – which we define as the suspension of the automaticity of mutual recognition, followed by a set of actions and practices of further monitoring of compliance with fundamental rights and rule of law standards in the issuing Member State - is only possible ‘in exceptional circumstances’. ¹⁶

This article seeks to assess the Court of Justice’s responses to these national courts’ concerns in the context of the EAW system, focusing in particular on the cases related to judicial independence. We have chosen this system and this area of law as our starting points precisely because of the evident tensions that have emerged in this area between mutual trust as a governing principle of the EAW scheme – required and imposed by the CJEU - and trust as a reality between Member States’ authorities – with several noticeable instances of factual distrust between national courts, notably in the context of rule of law backsliding, but also beyond it. ¹⁷ In the area of judicial independence, the Court has so far been particularly careful in maintaining the rigid two-step test established in LM, while it has, at the same time, offered concrete clarifications that can help national courts with exercising lawful distrust in practice. In our article, we strive to identify and critically assess possible contradictions emerging from the Court’s approach when faced with this mismatch between required trust and factual distrust. We then suggest ways in which the Court – but also other institutional players in the EU system – may try to resolve those contradictions and reconcile the legal obligation of mutual trust with the factual reality on the ground.

We have structured our argument in the following way. We start from the ‘particular’ in Section 2, zooming in on the evolution of the case law of the CJEU on exceptions to mutual trust in the EAW system, with a special emphasis on judicial independence issues. In Section 3, we start to zoom out and show how, despite the construction and careful extension of a system of exceptions to mutual trust, the overall framework developed by the Court remains dominated by the obligation of mutual trust. Namely, the CJEU continues to consistently require Member States’ authorities to assume that all Member States share and recognise a set of foundational EU values – those of Article 2 TEU. In Section 4, we criticise the Court’s approach by zooming out even more: we show that the construction of mutual trust proposed by the CJEU in EAW cases is but one of several models of this principle developed in the jurisprudence of the Court. It is therefore far from evident that mutual trust as presently concretised in EAW cases cannot allow for more lawful distrust to be exercised by national courts. On the contrary, such an increase in the possibility for national courts to distrust would, crucially, be more in line with the very idea of trust in ‘mutual trust’. After criticising the current CJEU construction of mutual trust in the EAW system, we propose, in Section 5, an alternative account of mutual trust that is more in line with its central value-sharing premise between Member States. That proposed account of mutual trust contemplates not only the requirement for Member States to trust each other, but also bigger discretion for Member States to value actual inter-institutional trust between each other and require that such trust be concretely justified. It is important to clarify that this proposed account of mutual trust is still a legal account of a legal principle. We do not argue for the impossibility of trust as a legal concept but, differently, sustain that mutual trust, as a legal principle, must have its content and legal effects adapt to the varying level of factual trust between Member States authorities. In Section 6, we explain that concretising this different vision of mutual trust is a shared responsibility of the Court and other EU institutions, and we zoom back in on the EAW system to trace possible trajectories of change.

In terms of methodological approach, the present article subscribes to the scholarly sentiment that, even if mutual trust is a legal obligation, enriching it with insights about how trust relationships are developed in practice between institutions and individuals is key for this principle to be workable. ¹⁸ Specifically, examining trust as a sociological construct is valuable for comprehending how trust relationships established in EU cooperation between Member State authorities work in practice. ¹⁹ Some scholars have already ventured in achieving a more comprehensive understanding of mutual trust through sociological accounts thereof, including in the specific context of EU judicial cooperation. ²⁰ We aim to build upon those socio-legal constructions of mutual trust to conceive of and flesh out a dimension of this legal principle that accounts for and adapts to factual instances of both trust and distrust between national authorities involved in EU cooperation. Furthermore, we argue that it is possible to ‘connect’ that sociological dimension of trust to the law. In particular, we provide (i) a comprehensive doctrinal outlook of the broader language of the CJEU on mutual trust and the different models of this principle present in its case law; as well as (ii) an analysis of the connections of actual trust in EU cooperation with the text of the Treaties to justify that a sociologically dependent conception of mutual trust can be solidified as a principle of EU law.

Understood in this way, i.e., as a legal principle whose legal content and consequences are in part determined by the existent level of trust between Member States, mutual trust can arguably be better aligned with its own foundational premise: that Member States share and abide by the EU’s core values. ²¹ In addition and relatedly, such a conception of mutual trust can better adapt to those instances where, factually, these values are not shared or adequately guaranteed, as for example during constitutional backsliding crises. This is the ultimate aim of this piece: to propose and densify a bidimensional account of mutual trust, i.e., one where the legal requirement to trust must be complemented and informed by the varying levels of inter-state factual trust and distrust.

2. The EAW and judicial independence: the evolution of the case law of the Court of Justice

The common thread of the Court of Justice’s case law on limits to mutual trust in the EAW system is the construction of a strictly framed opportunity to exercise ‘lawful distrust’. Much has been written on how the Court came to accept limitations to mutual trust in the EAW and the AFSJ more broadly, ²² and here we will only summarise the main points. In earlier cases, the Court had adopted what seemed to be a system of blind or fully automatic trust. Judgments such as Radu or Melloni ²³ denied that national judicial authorities could rely on fundamental rights considerations to suspend or refuse the execution of arrest warrants beyond the list of mandatory and optional grounds provided by the EAWFD.

This extremely rigid position has then shifted also thanks to the pressure of national courts ²⁴ and the European Court of Human Rights (ECtHR). ²⁵ It was in Aranyosi and Câldâraru, a case concerning systemic deficiencies in prison conditions in Hungary and Romania, which created a real risk of violation of the prohibition of torture and inhuman and degrading treatment in Article 4 of the Charter, that the Court acknowledged for the first time the possibility to suspend the execution of a EAW, and thus to limit mutual trust and mutual recognition in case of systemic fundamental rights deficiencies in the issuing Member State. According to the CJEU, despite the lack of an explicit fundamental rights ground for refusing the execution of EAWs in the Framework Decision, national authorities are in exceptional circumstances authorised to postpone and ultimately refuse a EAW due to fundamental rights concerns. They can do so following a two-step test that requires them to first assess the existence of systemic or generalised deficiencies in the issuing Member State, and then verify whether, in the individual case, the person concerned will be exposed to ‘a real risk’ of a violation of their fundamental rights because of those deficiencies. ²⁶ In reaching these conclusions, the Court relied on the general fundamental rights clause of Article 1(3) EAWFD. ²⁷

The Aranyosi and Câldâraru judgement was seen as signalling that mutual trust between national authorities was not blind, and comprised (carefully constructed) exceptions, ²⁸ as in fact Opinion 2/13 ²⁹ and the N.S. ³⁰ judgement in the context of the Dublin asylum system had already anticipated. In relation to detention conditions, the Court confirmed its approach and offered further clarifications to national courts in follow-up cases including Generalstaatsanwaltschaft (Conditions of detention in Hungary) ³¹ and Dorobantu. ³²

In the next paragraphs, we analyze how the Court has translated this line of case law to the field of judicial independence, where national courts have questioned whether they should suspend or refuse the execution of EAWs because of concerns with the independence of judicial authorities in the issuing Member State. These cases crucially related to the process of rule of law backsliding in Poland but, as the Puig Gordi judgement makes clear, doubts on the independence of domestic courts may also arise in fundamentally different contexts and show the need to develop a comprehensive approach to the underlying question of the limits to the automatic execution of EAWs in light of fundamental rights and rule of law concerns. A horizontal analysis of the relevant cases shows two trends in the Court’s case law: on the one hand, the Court has consistently maintained its demanding two-step test, and thus strictly framed the opportunities to exercise lawful distrust; on the other, it has progressively offered important clarifications on how to apply the test, making it more workable and thus in practice offering additional opportunities for national courts to concretely exercise lawful distrust.

As is well known, in the LM ruling the Court followed and adapted the Aranyosi and Câldâraru logic of exceptions to mutual trust in the EAW system to the area of judicial independence. It established, again relying on Article 1(3) EAWFD, ³³ the demanding two-step test that the Court has since then always upheld. One of the crucial novelties of LM was that it made clear that exceptions to mutual trust are possible also when non-absolute rights – such as Article 47 of the Charter on the right to a fair trial – are at stake, if there is a possible interference with the ‘essence’ of the fundamental right in question. ³⁴ For the execution of a EAW to be suspended on account of judicial independence concerns, the two-step test requires national courts to demonstrate, first, that there are systemic rule of law deficiencies in the issuing Member State, which create a ‘real risk’ that the fundamental right to a fair trial protected by Article 47 of the Charter is violated. ³⁵ The assessment must be based on ‘objective, reliable, specific and properly updated’ evidence that concerns the operation of the justice system in the issuing Member State. ³⁶ Second, the national court must ‘specifically and precisely’ show that, in the specific case, there are substantial grounds for concluding that the individual subject to the EAW request will concretely run that risk. ³⁷ The assessment of the national court should be based on the ‘personal situation’ of the individual, the ’nature of the offence’ for which the person is being prosecuted, and the ‘factual context’ that forms the basis of the EAW. ³⁸ Despite the calls of the referring court to do away with the second step of the Aranyosi and Câldâraru test for cases of judicial independence, ³⁹ the Court of Justice confirmed therefore that an individual assessment of the specific situation of the application is needed after establishing that the general situation in the issuing Member State amounts to a systemic rule of law deficiency.

The judgement of the Court in LM disappointed many. Some argued that the Court should have taken a stronger stance on the rule of law deficiencies in Poland, ⁴⁰ while others focused on the fact that the test developed by the Court was difficult, if not impossible, to be applied in practice. ⁴¹ Regardless of one’s position on LM, it was evident that the judgement left much to be clarified, especially in the context of the progressive erosion of rule of law standards in Poland. ⁴² Shortly after LM, the Court of Justice was naturally called to provide further clarifications by another national court.

In July and September 2020, the Rechtbank Amsterdam sent two preliminary references to the CJEU. The Dutch court noted, also with reference to CJEU case law, ⁴³ the worsening of the judicial independence situation in Poland, and wondered whether, in light of those systemic deficiencies, it could do away with the second step of the test. ⁴⁴ While acknowledging the ‘increase’ of the systemic deficiencies, which should prompt national authorities to exercise ‘vigilance’, ⁴⁵ the Court of Justice however decided to confirm the two-step test approach adopted in LM. ⁴⁶ It clarified that the two steps – the general and the individual - cannot overlap, as each of them requires the analysis of different pieces of evidence. ⁴⁷ According to the Court, accepting that the existence of systemic judicial independence deficiencies leads to the presumption that the individual subject to the EAW will run a real risk of a breach of their right to a fair trial would ultimately amount to an automatic refusal of the EAWs issued by the Member State in question. That is a decision that the EAWFD leaves to the European Council and Council, not to the CJEU itself. ⁴⁸ The Court justified the need for a second step also in light of the EAWFD’s objective to combat impunity. ⁴⁹ It considered that such an objective precludes an interpretation of the EAWFD and in particular of its Article 1(3) according to which the existence of systemic deficiencies is sufficient by itself to justify a refusal. ⁵⁰ In short, an individual assessment remains always necessary.

Building on LM, the Court then offered additional clarifications on what needs to be assessed in the second step of the test. The Court added, for example, that ‘statements by public authorities which are liable to interfere with the way in which an individual case is handled’ are one of the factors that national courts should consider in the individual assessment ⁵¹ and pointed again at the importance of dialogue between the executing and the issuing authority under Article 15(2) EAWFD, which can provide meaningful information for the conclusion of the assessment.

The dialogue between the Amsterdam Rechtbank and the Court of Justice did not stop there. In two new parallel preliminary references issued in 2022, the Dutch Court noted the further degradation of judicial independence standards in Poland. In particular, the situation now raised questions related to respect for the fundamental right to a tribunal previously established by law, in light of the systemic deficiencies in the system of judicial appointments. ⁵² The Dutch court therefore asked the CJEU what test would it need to apply in those circumstances. In response, the Court confirmed the approach followed in previous cases. The individual step of the test remains necessary, ⁵³ even when there is information on the increase of systemic deficiencies ⁵⁴ and even when what is at stake is the right to a tribunal established by law. ⁵⁵ The ‘mere’ ⁵⁶ fact that a body (the Polish Council for the Judiciary) involved in the appointment of judges in the issuing Member State is composed mostly by politically appointed members does not warrant the creation of a different test, or a reversal of the Court of Justice’s position. ⁵⁷

Then, the Court of Justice provided further clarity on the two steps of the examination that national courts should conduct. The national court must, in the first step of the examination, conduct ‘an overall assessment’ of the judicial system of the issuing Member State, on the basis of ‘objective, reliable, specific and properly updated’ evidence, taking into account in particular ‘the general context of appointment of judges in that Member State’. ⁵⁸ As the Court had already clarified in LM, the evidence may include a reasoned proposal by the Commission under Article 7(1) TEU, but also CJEU and ECtHR case law, ⁵⁹ as well as national constitutional case law ‘which challenges the primacy of EU law … as well as the binding force of judgments of the Court of Justice … relating to compliance with EU law’. ⁶⁰

As for the second step, which remains necessary, the Court first clarified that the burden of proof lies with the person in respect of whom a EAW request has been issued: they must prove that the systemic deficiencies have ‘a tangible influence on the handling of his or her criminal case’. ⁶¹ The Court however introduces here an important addition to its line of case law: when the individual is unable ‘to demonstrate the existence […] of a real risk of breach of the fundamental right’ but has nonetheless put forward evidence ‘suggesting that […] systemic and generalised deficiencies have had, or are liable to have, a tangible influence in that person’s particular case’, the executing judicial authority ‘must’ request supplementary information to the issuing authority following the procedures of Article 15(2) EAWFD. ⁶² In other words, the Court demands that the two authorities start a process of dialogue, and stressed then that the issuing judicial authority has an obligation to provide the information requested by the executing authority. A lack of sincere cooperation by the issuing authority in this process of information exchange may be one of the factors considered in establishing whether the individual runs a real risk of a breach of their right under Article 47. ⁶³

With ample references to the Dutch cases, the Court later confirmed ⁶⁴ the two-step approach and test in Puig Gordi. The case arose in a fundamentally different context, ⁶⁵ namely the refusal of Belgian courts to execute EAWs issued by the Spanish Supreme Court in criminal proceedings concerning Catalan public officials. Here it was the Spanish Court (the issuing authority), whose jurisdiction to hear the relevant criminal proceedings was questioned by the Belgian courts refusing the EAWs, to submit a preliminary reference to the CJEU. ⁶⁶ With a long series of not perfectly clear questions, ⁶⁷ the Spanish Supreme Court essentially asked the Court of Justice to explain when exactly the execution of a warrant could be refused or stopped due to fundamental rights concerns, with the ultimate goal of forcing the Belgian courts to execute the EAWs they had rejected.

Sidestepping most of the more complex political questions underpinning the preliminary reference, the Court of Justice mostly focused on the question of whether and under what circumstances a EAW can be refused by the executing judicial authority when it considers that the individual subject to the warrant would be at risk of being tried, following their surrender, by a court lacking jurisdiction for that purpose. ⁶⁸ This would expose the individual to an infringement of Article 47 of the Charter and notably of the requirement to be judged by a ‘tribunal established by law’. ⁶⁹ Following the approach adopted in the Polish cases, the Court of Justice confirmed once again the two-step test approach. ⁷⁰ It also recalled that the two steps of such examination cannot overlap with one another and relate to the analysis of different factors. ⁷¹

The Court then added clarifications on both steps of the test. The Court stated that the first step could be considered as fulfilled not only when there are systemic and generalised deficiencies in the issuing state, but also when there are ‘deficiencies affecting an objectively identifiable group of persons to which the person concerned belongs’. ⁷² The Court thus broadened the scope of situations that can lead national authorities to consider the first step as fulfilled. As for the second prong of the test, the Court opted for a high threshold in case of doubts on the jurisdiction of the court issuing the warrant: the existence of an individual risk can only be determined if the court hearing the proceedings ‘manifestly lacks jurisdiction’. ⁷³

When reading all these cases together, a complex picture emerges. If we look at the main features of the scheme originally constructed in LM and earlier in Aranyosi and Câldâraru, it is evident that those have not been fundamentally modified. The possibilities for lawful distrust remain limited. The two-step test continues to be necessary, even when systemic rule of law problems become more acute, as in the Polish cases. And it continues to be a demanding one. In L and P, X and Y, and then Puig Gordi the Court has repeatedly stressed how the second step of the assessment always requires a consideration of the individual circumstances of the case. This implies that the executing authority cannot just rely on general factors indicating systemic deficiencies to justify a decision not to execute a EAW.

Looking more closely, however, we can notice that the CJEU has progressively offered important clarifications on the test. It has fleshed out what factors are to be taken into account both in the first general step, and in the second individual one. In doing so, the Court of Justice is giving more instruments to national courts not only to better navigate the test – making it more workable in practice – but also to concretely exercise distrust if they wish to do so. ⁷⁴ When conducting the individual test, for example, national courts are invited to take into consideration different elements, which include the personal situation of the individual, the nature of the offence, and the factual context, including statements by the public authorities that may interfere with how the case is handled. These are not cumulative conditions, but rather some of the key factors that could show the existence of a real risk of a fundamental rights violation for the individual subject to the EAW. In that context, the Court has also referred to the principle of sincere cooperation, which should guide the exchange of information between executing and issuing authorities on the basis of Article 15(2) EAWFD. This dialogue must start, the Court says, when the person concerned is able to show relevant, but inconclusive evidence that there may be a real risk of fundamental rights breach. A lack of cooperation from the issuing authority can then be another factor used to justify a decision not to execute. ⁷⁵

In conclusion, although the theoretical possibility for national courts not to execute a EAW based on fundamental rights concerns already existed since Aranyosi and LM, in a first phase national courts might have struggled to understand how to use such discretion. ⁷⁶ Now with more indications and guidelines from the CJEU, it has become easier, to a degree, for national courts to explain and motivate the reasons for their distrust. ⁷⁷

3. The Court of Justice’s approach to trust: not blind but required

Let us now start zooming out and look not only at the concrete test developed by the Court of Justice, but also at the broader language and use of mutual trust in the strand of decisions just described. As we have seen, the Court has, from Aranyosi and Câldâraru to LM and its follow-up cases, opened up limited possibilities for Member States’ courts to exercise lawful distrust and clarified the test through which this can be done. In this respect, we certainly agree with the many – including the Court of Justice’s President Koen Lenaerts – who have noted that ‘mutual trust is not blind’. ⁷⁸

However, even if the Court has progressively constructed opportunities for lawful distrust, its broader language of mutual trust in EAW cases has always remained the same. Namely, the Court always departs from the core assumption that systems based on mutual recognition – such as the EAW system - have as their basis a high level of trust between Member States. ⁷⁹ Furthermore, it stresses the foundational importance of mutual trust for the construction and maintenance of an area of freedom, security and justice without borders. ⁸⁰ As such, exceptions to this principle are and should remain ‘exceptional’. ⁸¹

The same construction was used even in the E.D.L. case, where for the first time the Court accepted, in a highly specific context, the possibility to suspend and ultimately refuse to execute a EAW outside the narrow confines of the two-step test. While broadening the possibilities for national courts to rely on fundamental rights considerations in order to limit mutual trust, the Court still stressed the ‘fundamental importance’ of mutual trust, ⁸² the requirements it imposes, ⁸³ the closed nature of the list of grounds for refusal in the EAWFD, ⁸⁴ and confirmed the typical formula that execution of EAWs is ‘the rule’, while ‘refusal to execute is intended to be an exception which must be interpreted strictly’. ⁸⁵ The ruling in E.D.L. should thus be seen as a narrow exception closely linked to the specific facts of the case ⁸⁶ – where the risk for fundamental rights, and in particular for Article 4 of the Charter, would derive from the surrender itself, due to a serious illness of the person subject to the EAW – rather than a fundamental re-thinking of the Court’s case law on exceptions to and limitations of mutual trust. ⁸⁷ The following decision in GN, where the Court returned to the classical two-step test in case a concerning possible non-execution of a EAW in order to uphold the principle of the ‘best interest of the child’, is further proof of that. ⁸⁸

In order to understand the foundational importance the Court ascribes to mutual trust, one must zoom out further and look at the first categorical formulation of this principle in Opinion 2/13, which the CJEU mentions in most its EAW decisions since Aranyosi and Câldâraru. ⁸⁹ In the Opinion, the Court presented mutual trust as a principle based on the presumption that ‘each Member State shares with all other Member States, and recognises that they share with it, a set of common values on which the EU is founded’. ⁹⁰ Such principle requires every Member State ‘to consider all the other Member States to be complying with EU law and particularly with the fundamental rights recognised by EU law’. ⁹¹ In the Court’s construction, from the requirement of mutual trust stem two more requirements, which Lenaerts described as the ‘two negative obligations’ imposed by mutual trust: Member States cannot, for one, ‘demand a higher level of national protection of fundamental rights from another Member State than that provided by EU law’, nor can they check in concrete cases whether a Member State ‘observed the fundamental rights guaranteed by the EU’. ⁹²

As can be drawn from such construction of mutual trust, even if trust represents, in its everyday linguistic meaning, an ontological reality which cannot as such be imposed on someone, the Court insists that mutual trust between Member States ‘must’ exist. ⁹³ Trust is exclusively presented as a legal obligation: as a matter of principle, Member States are required to trust that all other Member States comply with EU law and respect EU fundamental rights and values. ⁹⁴ When applied to specific cooperative schemes of the AFSJ, mutual trust is put at the service of mutual recognition. In fact, mutual trust makes mutual recognition possible by systematically requiring Member States to presume that the legal order of each Member State abides by the EU’s foundational values and fundamental rights standards. As a consequence of that required trust, Member States are further obliged to recognise each other’s legal outcomes (e.g. arrest warrants, judgments, professional certifications, etc.) as valid and, therefore, able to produce effects in all the EU. ⁹⁵

In the EAW context, the Court has added that the obligation of mutual trust ⁹⁶ is also a ‘corollary’ of the principle that ‘the guarantee of observance of the fundamental rights of a person for whom a European arrest warrant has been issued falls primarily within the responsibility of that Member State’. ⁹⁷ Therefore, these requirements of mutual trust and mutual recognition shape the horizontal relationship between Member States, by prescribing a certain normativity for their cooperative engagement. ⁹⁸ Applying this specifically to the EAW system, even where there are certain doubts as to the judicial independence of the national court that will deal with the surrendered individual’s procedure once that individual is transferred in execution of a given EAW, Member States’ courts must nevertheless, as a matter of principle, recognise and execute the arrest warrant in question and transfer the individual concerned. They must trust that the criminal procedure in question will conform to EU core values and fundamental rights, and maxime to Article 47 of the Charter. ⁹⁹ The fact that mutual trust is a requirement, and that that requirement is the rule, explains the high threshold set by the Court for mutual trust to be rebutted and for a EAW not to be executed. This is to say that the language of mutual trust influences its use, i.e., its practical consequences for a given cooperative scheme.

To conclude, in the decisions we have explored, the Court has gradually recognised possibilities for lawful distrust while at the same time refusing to revisit any of the foundational premises of the mutual trust system as implemented in the AFSJ. Moreover, the Court seemingly suggests that such a construction of mutual trust (something that must exist legally, even when in reality it does not, in order to enable borderless cooperation between Member States) with such consequences (requiring mutual recognition and precluding Member States from checking each other’s fundamental rights and rule of law standards, save in exceptional circumstances) is the only possible formulation of this principle in the AFSJ. We see this well expressed in paragraphs 115 and 116 of the ruling in Puig Gordi: the Court states that trust ‘must’ be conferred on the courts of the issuing Member State, and that in the absence of such trust, the effectiveness of the entire system of judicial cooperation between Member States is ultimately called into question. ¹⁰⁰ Yet, as we discuss in the next section, this is far from the full picture: a look at the use of mutual trust in other fields of EU law shows that the Court has promoted different approaches to mutual trust depending on the context in which this problem surfaces.

4. Problematising the Court’s account of mutual trust: making sense of the clash between the requirement and the practice of trust

Let us zoom out even further and put the CJEU’s construction of mutual trust in the EAW system in a broader context, by pinning it against the conceptualisation of mutual trust by the CJEU in other areas of EU law. Mutual trust (or in earlier stages, ‘mutual confidence’) has been brought to the fore by the Court of Justice in many cooperative fields, such as internal market law, ¹⁰¹ civil ¹⁰² and criminal judicial cooperation, ¹⁰³ competition law, ¹⁰⁴ the Common European Asylum System ¹⁰⁵ and, more recently, the execution of the EU budget. ¹⁰⁶ Given the lack of a consolidated understanding of mutual trust as a legal principle of EU law, ¹⁰⁷ it is useful to locate our analysis of mutual trust in the EAW into the broader span of applications and formulations of this principle. It is, however, important to clarify the focus and objective of this comparative exercise.

Firstly, our analysis is not focused on a field-by-field dissection of all the interests and actors involved in trust-based cooperative schemes in EU law. We do acknowledge that various differences exist in how mutual trust is translated into different policy fields. The institutions involved in trust-based relationships differ depending on the field, as do the interests and goals pursued in those relationships by each individual actor and policy maker. ¹⁰⁸ To give one simple example: in internal market law, mutual trust allows for customs authorities to accept (in principle) with no checks the circulation of certain imported Member State products; but it also allows in the AFSJ for national courts to accept e.g. foreign custodial sentences, arrest warrants or other judgments as valid in their national legal orders. Evidently, the relationships of trust in this example are established between starkly different institutions and pursue equally different interests. ¹⁰⁹ Adding to this, the theorisation of mutual trust as a principle of EU law has been the product of various efforts of jurisprudential and scholarly conceptual reconstruction from the various instruments in which this principle is applied. ¹¹⁰ Put simply, mutual trust as a legal concept emerged from several attempts to understand the different trust-based cooperative schemes in EU law, with the Court of Justice playing a key role in the development of a general horizontal principle of mutual trust.

Given the highly different contexts where mutual trust is used, we recognise that the direct transposition of normative propositions between different fields of EU law where mutual trust is featured is hardly desirable (if not plainly impossible). Our aim is in any event not to advocate for a direct export of solutions between different mutual trust-governed fields. Differently, we start our analysis from the fact that, despite all the abovementioned specificities, the CJEU has given to mutual trust the status of a foundational EU law principle, and asserted that such principle stands upon a core premise irrespective of the context in which it is applied: that all Member States share and recognise a set of EU’s core values and fundamental rights. ¹¹¹ Despite what Aasa has qualified as mutual trust‘s context-dependent nature and casuistic construction, ¹¹² the fact still remains that mutual trust is now framed as a general horizontal principle of EU law, ¹¹³ governing cooperation between national authorities, ¹¹⁴ and enabling in that context the functional equivalence of different legal rules, judgments, certificates or products stemming from different national legal orders. ¹¹⁵

As such, we want to follow the scholarly trend of examining mutual trust at a ‘higher level of abstraction’, ¹¹⁶ and analyse how it is conceptualised and used in different legal fields. Namely, given that, in one way or another, mutual trust entails an in-built requirement to trust, as stated in Opinion 2/13, ¹¹⁷ we aim to answer two questions. First, how automatic or negotiable is that requirement to trust in a given trust-based scheme of EU law? And second, to what extent does the principle of mutual trust allow for the balancing of conflicting interests or values against the objectives of furthering trust and recognition-based cooperation in a given field? In so doing, we aim to identify in sub-section A different constructions of mutual trust by the CJEU depending on their degree of automaticity. Thereafter, in sub-section B, we use the prior analysis to challenge the Court’s assertion that the current model of required mutual trust in the EAW cannot be fundamentally modified.

A. A comparative outlook of the CJEU’s approaches to mutual trust

One could approach in different ways the task of grouping the different uses of mutual trust by the Court. ¹¹⁸ We have opted to base our categorisation on the degree of automaticity with which mutual trust is applied in different fields of EU law. Mutual trust’s degree of automaticity is defined here as the extent to which the Court allows for distrust to be exercised by a given national authority as an admissible hurdle to be cleared before the trust-based recognition of any given domestic product or legal outcome. According to this criterion, the CJEU’s formulations of mutual trust can be grouped in two main categories: one of so-called ‘quasi-automatic and imposed mutual trust’; and another one of so-called ‘negotiable mutual trust’.

First, in certain cases, the CJEU requires a quasi-absolute application of mutual trust, often connected to a requirement of mutual recognition which can only be set aside, beyond the situations in which the relevant secondary legislation instruments authorise it explicitly, in exceptional circumstances. ¹¹⁹ For mutual recognition to be required from national authorities, it is immaterial whether or not there is actual trust between them. We designated this form of mutual trust as ‘quasi-automatic and imposed’. It is mainly visible in cases concerning civil and criminal judicial cooperation. In civil matters mutual trust generally dictates an almost absolute obligation of mutual recognition, to the point where national courts cannot even review judgments if those ‘are vitiated by a serious infringement of fundamental rights’. ¹²⁰ Mutual trust in the EAW system also falls in this category. ¹²¹ There, national courts cannot suspend or refuse to execute an arrest warrant solely because of evidence of systemic flaws of a Member State in protecting fundamental rights or securing judicial independence, even where such systemic flaws have been acknowledged and criticised by several EU institutions. ¹²² In short, mutual trust requires Member States, even in the presence of systemic deficiencies in a Member State, to view as admissible and equivalent the legal acts issued by their judicial authorities, unless both steps of the Aranyosi/LM test are fulfilled. As noted by Brouwer, mutual trust takes in these cases a more active role of creating cooperation. ¹²³ The very language of the CJEU emphasises this cooperation-furthering role of mutual trust: this principle is aimed at the elimination of barriers to judicial cooperation in civil and criminal matters thereby allowing for the creation and maintenance of an area of freedom, security and justice without borders. With that objective in mind, the CJEU states that mutual trust must exist between Member States. ¹²⁴

In a second category of cases, the use and stated function of mutual trust differs. Indeed, the CJEU can also be seen conceiving of a more ‘negotiable’ form of mutual trust, i.e. allowing national authorities more leeway to negotiate the limits of their obligation to trust their counterparts. Specifically, they are allowed to act upon their factual distrust and verify the adequacy of their counterpart’s Member State legal order with EU fundamental rights and other core values and, if need be, suspend or stop cooperation in a given scheme.

This can be seen, for example, in asylum law cases, where national courts and administrative authorities are given more discretion to ensure that the authorities receiving asylum seekers respect fundamental rights (being ultimately required to suspend the transfer of an individual until obtaining such assurances). ¹²⁵ Even though the Court departs from the idea that mutual trust is key for the functioning of asylum cooperation, it gives more preponderance in these cases to the need to protect fundamental rights. This is, in part, attributable to the very design of the Common European Asylum System when compared with the EAW system, in the sense that under the CEAS, a state authority that decides not to trust another Member State may decide to directly process the asylum request itself. ¹²⁶ The same opportunity is not always given under the EAW, as the executing authority often will not be able to prosecute the case itself because of jurisdictional issues, something that leads the Court to emphasise the risks of impunity in the EAW system. But the fact does stand that where state authorities involved in asylum cooperation do not trust that an asylum seeker will be guaranteed adequate conditions in the destination Member State, they must suspend cooperation, namely refrain from transferring that individual. ¹²⁷ Such consequence of mutual trust is in stark contrast with the pressure put by the Court on state authorities to execute EAWs and surrender individuals in the name of ensuring the efficiency of the system. In asylum cases, the CJEU has even directed state authorities to take proactive steps to protect the fundamental rights of asylum seekers (e.g. to ensure that the mental health of an individual is not compromised by offering them appropriate care) during and after the transfer instead of merely trusting the authorities of the host Member States to do so themselves. ¹²⁸ This is ultimately in line with the language of trust present in EU secondary legislation regulating the Common European Asylum System, where emphasis is put on the need to develop trust between Member States in order to make cooperation work, instead of simply requiring that trust exist. ¹²⁹

A similar form of negotiable mutual trust was endorsed also in the EAW context in the E.D.L. case described above, where the risks to fundamental rights derive not from systemic deficiencies in the issuing Member State, but from the surrender itself and the highly individualised risks it would impose on the surrendered person. ¹³⁰ This does not alter the fact that quasi-automatic and imposed mutual trust is prevalent in the EAW system. And soon after E.D.L., the Court in fact returned to the same construction in the GN case, following the Opinion of AG Ćapeta that E.D.L. should not be read too broadly: the requirement of the two-step test has not been abandoned by the Court in EAW cases; it remains the rule alongside required mutual trust. ¹³¹ Differently, cases such as E.D.L. constitute an exception as the highly individualised nature of the risk at stake is vastly different from situations like LM.

In the same vein, in internal market law – where mutual trust or mutual confidence was first used and formulated – this principle is not used to actively ensure that cooperation happens. It does not require that obstacles to free movement of goods, services, certificates and diplomas be removed at all times; but rather that they be reasonably justified and proportionate to the aim of securing other EU’s fundamental interests and values. Indeed, since Cassis de Dijon, ¹³² the CJEU imposed on Member States, in Snell’s words, a qualified duty of mutual recognition based on qualified mutual trust. ¹³³ That is, Member States could not simply reject each other’s products as that would constitute a violation of Article 34 TFEU (prohibiting quantitative restrictions to free movement of goods) but, decisively, were also not obliged to automatically recognise and accept them. Rather, they were given the possibility to distrust each other’s products, meaning to present reasonable and proportionate justifications to reject a certain imported product. ¹³⁴ Equivalence of market regulatory standards must still be presumed by Member States as a consequence of mutual trust, but the Court does offer them ample opportunity for lawful distrust through the exercise of host state control over a certain imported product. ¹³⁵ Furthermore, when one looks at the language of mutual trust in these cases, it is noteworthy that the Court, instead of starting its judgments by highlighting the importance of mutual confidence between Member States in the internal market, chooses to highlight the fact that host Member States’ distrust and control may be recognised as necessary in order to satisfy other mandatory requirements of EU law, such as, for example, public safety, protection of consumers, or environmental protection. ¹³⁶ What is more, such control seems indispensable to maintain trust between Member States in internal market cooperation. As Cramér argues, a pre-condition for market trust is that its limits remain negotiable, i.e., that a possibility to distrust and exercise control over imported products exists. ¹³⁷

Similarly to the language of trust used by the Court in internal market cases, in the domain of the EU budget mutual trust is seen – together with the principle of solidarity – as a pre-condition for cooperation between Member States. ¹³⁸ Indeed, in the Budgetary Conditionality Regulation, mutual trust is framed as one of the bases for the design of the conditionality mechanism that allows budgetary cooperation to be suspended if it is threatened by a Member State’s lack of adherence to the rule of law. ¹³⁹ When called upon to decide on the Regulation’s validity, the Court highlighted the link of trust established between Member States and an acceding state once the latter clears pre-accession conditionality and gains EU membership. Pre-accession conditionality justifies that Member States trust each other’s adherence to the EU’s core values expressed in Article 2 TEU. Breaches of the rule of law compromise that mutual trust and generate distrust, and those negative consequences justify the development and implementation of a conditionality mechanism that allows the suspension of budgetary cooperation, following the procedures provided by the Regulation. ¹⁴⁰

One can naturally find some outliers in a spectrum between the two forms of mutual trust just presented. We illustrate this point with one example: that of mutual trust in competition law cases. There, the General Court has stated that mutual trust should exist as the overwhelming rule between the Commission and national competition authorities. As such, the General Court applies in these cases the two-step test developed in LM, ¹⁴¹ based on an underlying requirement for national competition authorities to trust the adequacy and compliance with the rule of law of their respective procedures. ¹⁴² Therefore, one can notice the language of quasi-automatic and imposed mutual trust. However, and interestingly, the General Court also states that, before a case is allocated to a national authority, the Commission must ensure that such authority fulfils the independence and impartiality requirements necessary for it to secure the complainant’s rights. ¹⁴³ Mutual trust is therefore used to create this obligation for the Commission to check potential allegations of a risk of a violation of Article 47 of the Charter. National authorities’ independence is thus presented as a pre-condition for mutual trust and sincere cooperation to exist within the European Competition Network. ¹⁴⁴ Even if not between national authorities, the negotiation of the limits of mutual trust is allowed and entrusted to a supranational institution.

B. A critical analysis of the Court’s mutual trust(s)

This sort of mutual trust à la carte – a principle with different impacts on cooperation depending on the field of EU law at stake – makes its meaning rather elusive for national authorities and, ultimately, for the national courts entrusted with applying the Court of Justice’s judgments on mutual trust. Those rulings do not contain any settled criteria as to the specific form of mutual trust to be adopted in a certain case. Such elusive nature of mutual trust has been extensively pointed out in the literature. ¹⁴⁵ But, crucially, it also shows that, in certain fields of EU law, the Court has not taken the premise of value-sharing between Member States expressed in Opinion 2/13 at face value. Instead, it used such premise as the justification for negotiating the limits of the requirement to trust, thereby making it less automatic.

However, that is not the predominant construction of mutual trust in EAW cases. We have already established that, where there are signs of systemic deficiencies in a Member State whose courts are involved in a specific EAW case, the Court decides to follow a quasi-automatic and imposed form of mutual trust: mutual trust requires that EAWs must always be executed unless the stringent conditions of the two-step test are fulfilled, regardless of whether actual trust exists. ¹⁴⁶ We find this definition problematic in light of its unidimensionality: mutual trust is perceived and used as nothing more than a duty. As a legal principle, trust is a fiction that founds a requirement for Member States to recognise each other’s arrest warrants. In this context, the Court does not seem to acknowledge any other role for trust in EAW cooperation. Specifically, it does not seem to recognise any possibility for actual trust, i.e., trust as a reality that needs to pre-date and inform EAW cooperation, to dictate the legal content and meaning of mutual trust as a principle of EU law. To put it more concretely, national courts might not actually trust each other but are nonetheless required to mutually recognise each other’s EAWs, save in exceptional circumstances and following a very high threshold of proof.

And naturally, there are cases where national courts refuse to trust each other despite being required to do so. ¹⁴⁷ Constitutional backsliding processes, as well as generalised evidence of deficient fundamental rights standards of the asylum and criminal justice systems of Member States have led courts and other national authorities to distrust each other’s commitment and capacity to uphold the EU’s core values. It is thus evident that the presumption and requirement of trust created by the Court of Justice are detached from reality. One can qualify those situations where required mutual trust does not align with the existence of actual trust between cooperating actors as ‘pathologies of trust’. As the Court has consistently held since Aranyosi and Câldâraru and LM, these pathologies of trust are not reason per se to suspend the execution of a EAW. ¹⁴⁸ Mutual trust’s role as an enabler of cooperation leads the Court to prioritise that the mutual recognition prescribed by the EAWFD takes place as a rule. The Court explicitly acknowledges that while the EAWFD needs to be interpreted in light of the fundamental rights protected by the Charter, ¹⁴⁹ this should not lead to principled distrust between national courts involved in EAW. ¹⁵⁰ On the contrary, the EAWFD – holds the Court – can never be interpreted ‘in such a way as to call into question the effectiveness of the system of judicial cooperation between the Member States’. ¹⁵¹ Therefore, the execution of an EAW is not the consequence of the executing court’s trust on its counterpart. ¹⁵² It will rather be grounded on the executing court’s obedience to a secondary EU law obligation to recognise an arrest warrant.

All in all, the Court’s EAW case law on mutual trust hardly factors in the need for a pre-existent degree of trust between national courts as to each other’s independence. As many scholars have pointed out, it quite paradoxically requires it. ¹⁵³ But trust cannot be imposed or decreed. Nor is it innate. ¹⁵⁴ Trust is an ever-developing reality, something that exists and should be cultivated and organically maintained. ¹⁵⁵ That does not stop the Court from, even where trust falters, requiring that it be fictionalised to ensure that cooperation goes on. Required mutual trust as applied by the Court for EAW cooperation is inadequately unidimensional, as it fails to capture the complexity of the trust relationship established between national authorities of Member States. ¹⁵⁶ It is not interested in reality and existing pathologies of trust: national courts must trust each other, even when they do not.

It is in this sense puzzling that required mutual trust continues to be the prescribed model for this principle in EAW cooperation even when, as explored in the beginning of this section, there are other possible models of mutual trust already used by the Court. Again, our aim is not to argue for a ‘copy-pasting’ of legislative or judicial solutions from other fields of EU law governed by mutual trust to the EAW. Our argument locates itself at a more theoretical and general level: the Court should recognise mutual trust in the EAW as necessarily ‘bidimensional’ or ‘Janus-faced’: it entails not only the requirement to trust that leads to mutual recognition, but it also allows for distrust as a way to negotiate its limits and address the eventual presence of pathologies of trust in EAW cooperation. Mutual trust should, therefore, also contain a blueprint for how horizontal Member State relationships are to be governed in the presence of evidence that erodes the shared presumption of value-sharing between them. ¹⁵⁷ In this respect, we follow the argument that the choice of the EU legislature for a mutual recognition system in a specific cooperative scheme – like the EAW system – implies that mutual trust as a legal principle comprises a requirement to engage in cooperation but that, in addition, the extent and content of that requirement is determined by the existence of a certain degree of actual trust in the legal systems of Member States and their alignment with the EU’s core values. ¹⁵⁸ The CJEU should pay heed to the political institutions’ mention of the need to foster actual trust. ¹⁵⁹ The next section further justifies the existence of a bidimensional principle of mutual trust in EU law, with the help of trust theories from other social sciences, the political processes of EU accession, and the EU Treaties themselves.

5. Densifying and justifying bidimensional mutual trust

A. What is trust? A sociological account of trust relationships

The first element to be determined in formulating an account of mutual trust is the type of relationship established through mutual trust. Cramér defines ‘social’ trust as the ideal situation where all actors in a society are able to make a stable prognosis of the benign behavior of all other actors. ¹⁶⁰ The first element of this definition we seek to highlight is reciprocity. ¹⁶¹ Parties to a trust-based cooperative relationship are interdependent, in that their interests cannot be achieved without reliance upon one another. ¹⁶² Therefore, if one is part of a cooperative relationship, one trusts all other cooperative partners and is trusted by all of them as well. In essence, each cooperation actor is both trustor and trustee. Mutual trust establishes in this way a reciprocal system whereby each actor expects from the other certain modes of behavior.

And why does trust exist in cooperative relationships? In any multilateral social relationship, trust is required by the contingency of the future actions of one’s counterpart, about which there is current lack of knowledge. ¹⁶³ In the presence of such contingency, trust represents, as Simmel puts it, ‘a hypothesis of future conduct, which is sure enough to become the basis of practical action’ and become the ‘mediate condition between knowing and not-knowing another person’. ¹⁶⁴ In simpler words, the lack of knowledge about our partner’s future conduct inevitably generates social risks for any cooperative endeavor of our everyday life. In this context, trust constitutes not only the willingness to assume such risks, but also a way to manage and incorporate them in complex social relations. ¹⁶⁵ When observing trust from the vantage point of its social function of managing social risks, two conceptual clarifications can be achieved. Firstly, distrust is a functional equivalent of trust, and not its antithesis. When faced with the social risks inherent to imperfect knowledge of future events, one can either completely abstain from cooperative action or, conversely, seek to manage those risks by either trusting one’s partner or resorting to the ‘practice of distrust’ by ‘monitoring, controlling, constraining and sanctioning’ that partner. ¹⁶⁶ Trust and distrust are therefore mutually reinforcing in their common function of reducing social complexity. Applying this to EU cooperation in the AFSJ, the possibility to exercise lawful distrust – that is to say, the existence of mechanisms of control by national authorities of their counterparts - should not per se be seen as something negative or impeditive of EU integration and the good functioning of its cooperative systems. Distrust should instead be conceived as a condition for actual trust to be generated between cooperating national authorities in a context where they do not innately trust each other’s behavior. ¹⁶⁷

Secondly, one must distinguish trust from what rationally justifies it: an actor’s ‘trustworthiness’. ¹⁶⁸ Trust and trustworthiness are not one and the same. The fact that we trust someone does not necessarily make that someone trustworthy. ¹⁶⁹ However, trust will only be rational if the trustee possesses a certain degree of trustworthiness. Trustworthiness is furthermore a dynamic reality, as it can increase or decrease depending on the information that is continuously obtained about the trustee. ¹⁷⁰ In that sense, as trustworthiness varies, the trustor will rely on it as a measure to decide not just whether to trust someone, but also until when it should continue trusting them. That is, trustworthiness and its evolution ultimately determine the decision to trust.

As with any other situation of risk management, no trust-based relationship can ever be one where risk is completely eliminated. ¹⁷¹ Trust can mitigate risk but can never eliminate it fully. In that sense, trust also entails a unique sense of vulnerability towards the trustee. More than simple self-interest in the advantages of a given cooperative relationship, trust is motivated by the rational acceptance of the trustee’s good faith. ¹⁷² In this sense, Baier defines trust as the ‘accepted vulnerability to another’s possible but not expected ill will’. ¹⁷³ Because the trustor is vulnerable towards the trustee, there is a need to justify trust through certain actions and on such reasonable grounds that make the trustor accept that situation of vulnerability. ¹⁷⁴ This conception of trust as something that needs to be justified aligns well with the models of negotiable trust of the CJEU presented in the previous section: distrust and a possible suspension of cooperation when national courts or authorities prove untrustworthy with regard to fundamental rights protection is a legitimate way to conduct a trust-based relationship and can be conducive to remedying eventual pathologies of trust.

However, having said of all this, trust is still an essentially interpersonal phenomenon. By way of contrast, in the framework of EU cooperation, mutual trust is, to a large extent, impersonal as it is established between Member States and their institutions. ¹⁷⁵ This fact requires the adaptation of trust from an interpersonal concept to an impersonal one. In this respect, it has been extensively pointed out that trust between institutions is a complex and multilevel concept as its object is influenced in different ways across different levels of analysis. ¹⁷⁶ Taking the EAW system as an example, trust between national courts encompasses more ‘macro level’ or ‘general’ trust on courts and the broader legal system in which they operate, as well as more ‘particular’ trust on the specific judicial acts they produce – in our case, arrest warrants - and on the actual court officials, judges and prosecutors offices of the different Member States. ¹⁷⁷

Granted, the trustworthiness of national courts involved in EAW cooperation – i.e. their perceived ability to ensure an adequate treatment of the individual surrendered through the EAW procedure – will undoubtedly depend, to a certain extent, on the personal traits of judges or prosecutors involved in EAW cooperation and relationships developed between them. ¹⁷⁸ But we are more interested here in analyzing how the CJEU perceives the influence of other more general factors on the relationship of mutual trust to be maintained between national courts: mainly, the independence of the judiciary of a certain Member State, and evidence of systemic threats to it; and its necessary influence on role-specific characteristics of national courts, such as professional values and standards, and on the trustworthiness of their output, such as judgments and arrest warrants. ¹⁷⁹ Specifically in the EAW system, the Court seems to disregard the necessary connections between the more general trust of national courts on a Member State’s alignment with EU values and fundamental rights and its consequences on particular trust relating to the adequateness of specific arrest warrants. Even when a court issuing an arrest warrant is untrustworthy because of its Member State’s systemic deficiencies regarding the protection of EU values and fundamental rights, the Court still sees it possible that a specific EAW procedure continues on the basis of trust, untarnished by the existence of that macro-level untrustworthiness of one of the Member States involved. On the contrary, it (paradoxically) requires that a more particular trust between courts must exist to further EAW execution, despite the lack of actual trust between them.

We therefore argue that situations of general untrustworthiness in the EAW system should not be addressed through the principled and quasi-automatic requirement to trust. Differently, a bidimensional account of mutual trust would allow for lawful distrust to be exercised in those cases where actual trust between national courts is not enough for trust-based execution of arrest warrants . This would be a means to enhance the dialogue between courts, and to increase the scepticism regarding compliance with EU values and fundamental rights. Such scepticism is, we argue, a necessary condition for national courts to effectively protect fundamental rights. ¹⁸⁰

But does EU law support this bidimensional version of mutual trust as we endorse it here? We think that it does, through pre-accession conditionality and in the text of the TEU itself, as we demonstrate in the next sub-section.

B. The content of trust in EU cooperation: an inextricable link to accession

We have so far examined the type of relationship which is established in practice through mutual trust between national authorities involved in EU cooperation and, specifically, between national courts involved in the EAW system. We have justified why a bidimensional account of mutual trust – one that encompasses both a requirement to trust and the possibility to distrust as a means to negotiate mutual trust’s limits – should be considered by the CJEU when formulating and concretely applying this principle. But it still remains to be demonstrated whether such account of mutual trust can be accommodated by the EU legal order.

Departing from the CJEU’s formulation of mutual trust, Member States expect each other’s adherence to a set of foundational values expressed in Article 2 TEU. ¹⁸¹ But what does such adherence entail in concrete? To arrive at an answer, it is important to trace mutual trust’s content to the moment where it is established. According to both the CJEU and the EU legislature, that moment is, for each Member State, where their EU membership is established. ¹⁸² There, each Member State commits to certain modes of compliant behavior of their institutions in the framework of EU cooperation. Correspondingly, all other Member States trust that the new Member State will uphold its obligations established in and from the moment of accession. ¹⁸³ That accepted vulnerability to a new Member State’s performance of its EU law obligations is typical of a relationship of actual mutual trust, as explained in Baier‘s work addressed in the previous sub-section. Such accepted vulnerability is rationally justified by the pre-accession procedure of Member States to the Union (Article 49 TEU), whereby the acceding State must show its alignment with membership requirements including crucially its alignment with EU values.

The last decades have however shown that such alignment with EU’s core values by a Member State at the moment of accession cannot be taken as a guarantee that, post-accession, those values will forever be respected by Member States. Mutual trust, in this sense, is a necessary tool to manage the real risk that one Member State might cease to respect the requirements and values that it committed itself to when it gained EU membership. The CJEU has, in this regard, recognised that it is therefore legitimate to use conditionality mechanisms (based on mutual trust) to suspend the enjoyment of certain membership rights of those Member States that cease to respect the values they committed to when becoming EU members. ¹⁸⁴

Applying this specifically to cooperation between national courts in the EAW system, the level of trustworthiness of each issuing national court will in large measure depend, at a more general level, on the alignment of the national court’s Member State with the EU core values. Such alignment is demanded of Member States as an outcome and since the moment of the conclusion of their accession procedure. Consequently, evidence of systematic deficiencies in the alignment of a Member State with the EU’s core values and fundamental rights standards will undoubtedly have an impact on the trustworthiness of any court of that Member State issuing EAWs. Such untrustworthiness at a more macro level has ripple effects on the more particular trustworthiness related to a national court’s ability to guarantee a fair trial and a criminal procedure which respects the surrendered individual’s fundamental rights. Mutual trust between those courts and courts of other Member States will be de facto impacted. Should not the principle of mutual trust as a legal requirement also adapt to that?

Before providing an answer to that question, ¹⁸⁵ one should still highlight one other point of contact of this bidimensional account of mutual trust with the Treaties. In particular, the TEU – through the principle of sincere cooperation enshrined in Article 4(3) TEU - prescribes that EU cooperation should be carried out in a certain, sincere way. The Court itself has interpreted sincere cooperation as requiring cooperation between Member States to be carried out in good faith.+ ¹⁸⁶ And, as pointed out by Prechal, Member States cannot cooperate sincerely or in good faith if they do not (actually) trust each other. ¹⁸⁷ The CJEU has already explicitly recognised that connection between mutual trust and sincere cooperation in the cooperative schemes of the EAW and competition law: the close cooperation between courts and competition national authorities is to be based on the principle of sincere cooperation which, in turn, is based on the principle of mutual trust. ¹⁸⁸

Despite those mentions of sincere cooperation, the Court’s view on EAW cooperation can hardly be said to be akin to ‘sincere cooperation’. Article 4(3) TEU prescribes that Member States shall cooperate ‘in full mutual respect’; thus prescribing a certain ethos to horizontal cooperation in the EU. However, Member States are in principle required to recognise each other’s arrest warrants even when they do not trust the adequacy of each other’s legal systems with tegard to EU law. They are required to trust that all other Member States share and abide by the EU’s core values, even when they do not. In the latter case, mutual trust ceases to necessarily align with its underlying core values enshrined in Article 2 TEU. ¹⁸⁹

The problem here resides in the fact that the CJEU exclusively views mutual trust in EAW cooperation as a requirement to engage in such cooperation. To reduce mutual trust to that sole dimension is to conceive it unidimensionally and thus, we argue, in a reductive way. Specifically, the CJEU ignores the fact that mutual trust also entails the permanent pursuit of its own justification. This is to say that mutual trust, as a legal principle, also requires the practice of distrust, i.e., of monitoring, controlling and corrective actions aimed at ensuring Member States’ alignment with the content of mutual trust. It is highly unlikely that Member States’ institutions will trust that their counterparts will, at all times, abide by the EU’s core values. Accordingly, the principle of mutual trust should reflect that, an inherent risk of EU cooperation. In order to be meaningful and workable as a principle of EU law, mutual trust should not be just about Member States’ requirement to trust each other. In the concrete, uncertain, and imperfect reality of cooperation, mutual trust should be bidimensional: it should not simply be automatically imposed on Member States, but it should also allow for distrust as a way to continuously negotiate that requirement to trust. ¹⁹⁰ Distrust should not be viewed as something negative which hampers cooperation and calls into question the effectiveness of cooperation in the AFSJ. ¹⁹¹ On the contrary, it should be viewed as a pre-condition for genuine trust, as it justifies the trustworthiness of all cooperative actors, thereby ensuring that cooperation is carried out sincerely, as prescribed by the Treaties.

6. Concretising bidimensional mutual trust: mapping trajectories of change

Let us now zoom back in and identify different trajectories through which this different construction of mutual trust we have advanced could be concretised in the EAW system. It is to be acknowledged from the outset that the responsibility for developing a bidimensional view of mutual trust, one that is line with the pursuit of its own justification, cannot and does not lie solely with the Court of Justice. On the contrary, it is a shared responsibility of the Court and the Union’s political institutions.

After all, the conception of mutual trust advanced by the Court, which sees trust as the rule and distrust as the exception, mirrors the way in which the EU legislature has shaped the system of the EAWFD itself. Article 1(2) EAWFD firmly states that Member States ‘shall execute any European arrest warrant on the basis of the principle of mutual recognition’. ¹⁹² Fundamental rights or rule of law arguments do not feature as a possible ground for refusal of a EAW: as already recalled earlier, we only see a reference to fundamental rights in Article 1(3) EAWFD – on the basis of which the Court of Justice has created the Aranyosi and Câldâraru and LM lines of exception – but no explicit possibility in the legislative framework for national authorities to refuse execution in case of fundamental rights concerns. Furthermore, it is the EAWFD that assigns to the European Council and the Council the power to generally suspend the application of the EAW system in a Member State. ¹⁹³

Therefore, while we have been critical of the Court’s approach in previous sections, we have to acknowledge that it acted within this controversial legislative framework. In that context, the Court was called to exercise a difficult balancing exercise between a variety of different interests, from the need to uphold fundamental rights to that of fighting impunity in the EAW system, from contributing to the protection of the rule of law in cases of constitutional backsliding to ensuring the smooth functioning of that cooperative system. All of this in the context, again, of a legislative framework that tipped the scale in favor of effectiveness and cooperation over fundamental rights and rule of law guarantees. ¹⁹⁴ In that sense, we agree with those who have argued that the evolution of the Court’s case law has already allowed for a better balancing of those different interests. ¹⁹⁵

In the next paragraphs we thus strive to map possible trajectories of change that would contribute to addressing the contradictions we have identified. While acknowledging the steps already taken, we are convinced the Court could still develop a less unidimensional conception of mutual trust and better accommodate the practice of distrust by national courts and individuals as a way to justify and generate actual mutual trust in the system. Next to the Court, the Union’s political institutions have a fundamental role to play. First, they can acknowledge mechanisms of distrust in the legislative framework and support mutual trust via trust-enhancing legislation. Second, they have the fundamental responsibility to ensure that the premise of mutual trust – the recognition of a common set of values – remain grounded in reality.

A. The role of the Court: accepting and governing distrust

The Court, we argue, should more easily accept mutual distrust, moving beyond the mere dichotomy between trust and distrust that the current case law suggests, where the two are seen as incompatible alternatives. ¹⁹⁶ Lawful distrust – as the practice of information exchange between national courts in the EAW system and the establishment of consequences for EAW execution in case such information is unsatisfactory or not provided at all ¹⁹⁷ - does not necessarily undermine mutual trust but is a condition for actual trust to be (re-)generated between national authorities in the presence of pathologies of trust.

Doing so requires first of all a shift in the language of trust used by the Court. Such a shift would mean less emphasis on the foundational importance of mutual trust for AFSJ cooperation and more ample mention of the need to make mutual trust compatible with other important EU interests and crucially with the values of Article 2 TEU, which are the very foundations of mutual trust. ¹⁹⁸ Such construction is common, for example, in internal market cases, where the possibility for distrust regarding certain products and services can be justified in light of fundamental values and interests of EU law. ¹⁹⁹ If mutual trust is ‘based on the commitment of each Member State to comply with its obligations under EU law and to continue to comply […] with the values contained in Article 2 TEU’, ²⁰⁰ as the Court itself has stated, then it is evident that when that commitment is not fulfilled, consequences must follow. Cooperation between Member States cannot and should not continue ‘as is’, and automatic mutual recognition should give way to a process of lawful distrust where national authorities are able to balance different interests. This implies that the Court should accept mutual distrust not as something negative which impacts the effectiveness of the EAW system, but rather as something necessary for managing the risks of such cooperation stemming from systemic fundamental rights and rule of law problems, and for justifying the trustworthiness of all actors involved in the EAW system.

The acceptance of mutual distrust requires then also a shift in the use of mutual trust, which should reflect and govern the practice of distrust. By governing and proceduralising mechanisms of distrust, the Court of Justice can allow executing authorities to ultimately justify mutual recognition on the basis of existing, actual trust, instead of trust that is only imposed or required.

To make the argument more concrete: governing and proceduralising distrust would require rethinking the two-step test that the Court has constructed. In the context of the first step, governing distrust could crucially mean centralising the assessment of when systemic deficiencies create a real risk of a breach of fundamental rights. Rather than essentially leaving the assessment to national courts, ²⁰¹ only providing rough indications on the sources that can be consulted, ²⁰² the Court could be clearer on the cases when a national situation qualifies as a systemic deficiency. ²⁰³ The Polish case, at least before the 2023 elections and the formation of a government that has made the restoration of judicial independence and the rule of law one of its key priorities, was the obvious example. By relying on its own case law but also on the documents developed by the Commission under Article 7 TEU or in other contexts, ²⁰⁴ the CJEU could have in clearer terms concluded that the situation in Poland fulfilled the first step of the LM test. This would have allowed the EU and national courts to draw adequate conclusions from the backsliding of the rule of law in Poland, recognising that the situation qualified as ‘an exceptional circumstance’ in which the usual premise of mutual trust was systemically called into question. In trust jargon, the Court would have acknowledged the presence of a pathology of trust fundamentally affecting EAW cooperation.

This finding would then trigger the second step of the test. Despite some critical voices on this second step, ²⁰⁵ we do not deny that maintaining a two-step test allows to balance different interests at stake and is in line with the system designed by the EAWFD as it stands, a system that inter alia leaves the automatic suspension of EAW in the hands of the Union’s political institutions. We also acknowledge the importance of the ‘systemic’ criterion as a ‘federal safeguard’ between different constitutional objectives – the protection of rights and the rule of law, on the one hand, and the effectiveness of EU law and the AFSJ, on the other – and as a way to distribute fundamental rights responsibilities among Member States, where national courts are only asked ‘in exceptional cases’ to oversee other Member States’ courts’ respect for fundamental rights. ²⁰⁶

The second step of the test should be more fundamentally reconsidered, though. As it is currently designed, the second step places exclusively on the individuals subject to the EAW the onerous burden of proving – ‘specifically and precisely’ ²⁰⁷ – that they run a risk of fundamental rights violations in the issuing Member State. We envision instead a more dialogic process. We are not calling for a full reversal of the burden of proof, which would amount almost to a probatio diabolica for the issuing authority. We argue on the other hand that, in a way similar to the system in place in EU non-discrimination directives, ²⁰⁸ individuals could only be asked to provide prima facie evidence from which it can be presumed that the risk could materialise in the individual case. ²⁰⁹ The Court could in this sense build on the findings of X and Y. There, the Court has already stated that, when the individual subject to the EAW can put forward evidence that suggests that systemic deficiencies have had or are liable to have a ‘tangible influence’ on the particular case, even if that evidence is not sufficient to ‘demonstrate the existence’ of the real risk, the executing authority must request further supplementary information to the issuing authority. ²¹⁰ We claim that the Court should take an additional step: when that prima facie evidence is present, ²¹¹ it is then for the issuing authority to prove that the risk does not materialise in the individual case.

The finding of prima facie evidence should therefore open a process of exchange and conversation between the two judicial authorities, based on Article 15 EAWFD. This process can ultimately lead to mutual recognition (meaning the execution of a EAW) as an outcome if trust can be re-built through this information exchange (if the national authorities can show that the risk will not materialise in a given case). Or in contrast, the exchange could lead to suspend mutual recognition in that given case if trust between national authorities cannot be re-generated and re-established. ²¹² A crucial task of the Court of Justice is to contribute to fleshing out the procedure of exchange, further clarifying the elements of the individual assessment to be taken into account by national courts. L and P and X and Y have already offered significant clarifications, as noted in Section 2. Yet it remains unclear for example how the different elements and interests have to be balanced against each other. ²¹³ The Court should step in and offer more precise guidance.

Crucially, the Court should emphasise not only the duties of the executing authority, but also of the issuing one. Again, the remarks included in X and Y ²¹⁴ on the need for sincere cooperation between national authorities – and the fact that the lack of sincere cooperation of the issuing authority is a factor to be taken into account when conducting the second part of the test – are an important opening in that direction. ²¹⁵ But the Court could be bolder: a lack of (sincere) cooperation in a system where there are already systemic pathologies of trust may become the fundamental factor for the conclusion to refuse recognition of a given EAW, as it evidently makes it impossible to re-build genuine trust between the two systems.

B. The role of EU legislation: governing distrust and supporting trust

The Court’s effort of governing distrust should be accompanied by parallel efforts of the Union’s legislative institutions. As noted above, the Court has developed its two-step test in the context of a legislative framework that puts mutual trust exclusively at the service of mutual recognition, requiring quasi-automatic trust between state authorities. It is thus a responsibility also of the legislative institutions to update the legislative framework in a way that reflects the steps already taken by the Court and then incorporates and regulates practices of distrust. A similar step was taken, for example, in the context of the Dublin system, where the legislature translated into the revised Dublin III Regulation the N.S. line of case law of the Court of Justice. ²¹⁶

At a minimum, a revision of the EAWFD should incorporate the ‘exceptional circumstances’ doctrine developed by the Court and its two-step approach. Furthermore, one could also envisage a revision of the recital of the preamble that assigns the responsibility for a generalised suspension of the EAW to the European Council and the Council, stating that such a decision can be taken only after a determination of a serious and persistent breach of the values under Article 7(2) TEU. Different possibilities could be explored, such as anticipating the possibility of suspension to a Council decision on a threat to the values under Article 7(1) TEU, or even to the start of the procedure against a Member State. After all, a threat to the values already calls into question the premise of mutual trust that those values are shared.

A more ambitious reform could entail redrawing the balance between mutual recognition and effectiveness, on the one hand, and fundamental rights and rule of law concerns, on the other. Following the examples of other more recent instruments of the AFSJ, most notably the European Investigation Order, ²¹⁷ a revised legislative framework could explicitly acknowledge (some) fundamental rights concerns as a ground for non-execution of a EAW even beyond the two-step approach proposed by the Court. ²¹⁸ This would entail a deeper shift in the balance between automaticity and values-based considerations, which only the legislature would be able to realise. Whereas effectiveness of the cooperation system was the mantra at the time of the adoption of the EAWFD, a re-thinking of the overall framework that is tilted more heavily towards fundamental rights concerns could better align the Framework Decision with the foundations of the EU structure, also given the growing centrality of fundamental rights and values in the EU constitutional framework. ²¹⁹ Mutual trust would still remain a central feature of such a scheme, but operationalised in a way that is more firmly aligned with what we called the permanent pursuit of its own justification.

There is then a second dimension to the contribution of EU legislation to the construction of a bidimensional approach to mutual trust: the development of what has been defined ‘mutual-trust-enhancing’ legislation. ²²⁰ That refers to the development of common procedural standards that generate and support trust between Member States. ²²¹ This does not mean abandoning a system of mutual recognition: the common standards are, and should remain, minimum ones, in order to guarantee and protect the diversity of choices operated by the Member States and in accordance with the limitations prescribed by the legal bases of Articles 82 and 83 TFEU. ²²² But harmonisation serves here as a tool to generate trust, which instead of being simply imposed and based on a general presumption of compliance with a set of values, becomes grounded in reality and in objective standards developed in common legislation. An additional benefit of mutual trust-enhancing legislation is that it opens possibilities for monitoring and enforcement by the Commission under the infringement procedure. ²²³ The Commission could therefore point out reasons for distrust and then contribute to addressing them in dialogue with the responsible Member State. Another possible mutual trust-enhancing reform would be the creation of a follow-up mechanism whereby the Commission would monitor the treatment of surrendered individuals in the issuing state after a EAW is executed. This would enable a check of whether the mutual trust and recognition underlying a specific EAW execution were indeed justified, i.e., whether the guarantees of judicial independence and fundamental rights protection provided to the executing court during the process of information exchange hold true after an individual is surrendered to the executing court’s jurisdiction.

Examples of mutual trust-enhancing legislation already exist – the EU has adopted several minimum standards directives on the rights of individuals in criminal procedure ²²⁴ - and further initiatives could be pursued to realise the account of mutual trust we have proposed. Where the EU institutions would lack explicit legal basis, it might be possible to replace binding legislation with soft law documents prepared by the Commission. An important example in that sense is the Commission Communication on material detention conditions issued in December 2022. ²²⁵

C. Keeping the premise real: better efforts to uphold EU values

We conclude our mapping of possible trajectories of change by going back to the very premise and foundation of mutual trust: the commonality of the values of Article 2 TEU. It is after all first and foremost the worsening of constitutional backsliding processes in Hungary and Poland and their spillover effects on Member States’ cooperation that forced the Court of Justice to address the implications of those processes over mutual trust between national authorities. Constitutional backsliding not only calls into question mutual recognition and horizontal cooperation in individual cases. ²²⁶ Crucially, it also undermines the very premise on which mutual trust is based, namely that Member States share a common set of values. ²²⁷

The final proposed trajectory of change is thus at the same time rather straightforward and highly complex: the EU institutions – need to ensure adequate protection of the values of Article 2 TEU. And this too is a shared responsibility of the Court and of the political institutions. To link this to the language we have adopted earlier, EU institutions need to prevent the occurrence of pathologies of trust, and tackle those promptly and effectively when they emerge. This allows for the realignment of the reality of trust with its underlying premises. We agree with Halberstam that ‘for a federal union to survive, any legal obligation of mutual trust must be grounded in social reality, not judicial fiat’. ²²⁸

How to do it is a matter for another contribution. What can be said however is that the effort of the Union should focus both on the monitoring of Member States’ behaviors and on addressing concrete pathologies of trust where they emerge. When the latter happens, the key objective of the institutions should be that of bringing them to an end. But if that proves impossible, the responsibility of the institutions is also to activate the relevant mechanisms that signal the presence of ‘exceptional circumstances’ and that distrust is therefore necessary. The Court in this sense has already reminded us of the responsibilities of the political institutions, and in particular of the European Council and the Council, on the possible suspension of the EAW after a determination under Article 7 TEU. ²²⁹ If conditions so require, the institutions should not shy away from exercising those powers and signaling the necessity of distrust and its consequent impacts on EAW cooperation.

7. Conclusion

The suggestions offered in the previous section are by no means a complete package of reforms. Rather, they are possible trajectories of change that could be followed in order to concretise an alternative account of mutual trust as a principle of EU law. That was the goal of our contribution: to propose an account of mutual trust where the obligation for Member States to engage in trust-based cooperation is aligned with the factual reality on the ground, in the Member States. That ultimately means that mutual trust’s legal content must be consistent with its own justification and raison d’être, i.e., that Member States share a set of common values.

The argument we want to advance is that, at a time in which pathologies of trust are becoming more and more evident, not least because of the significant deterioration of rule of law and fundamental rights standards in several Member States, mutual trust cannot continue as is. It cannot be imposed on Member States and their judicial authorities tout court, all the while the latter are only able to exercise mutual distrust ‘in exceptional circumstances’. As Section 2 of the article has shown, we acknowledge that the Court of Justice’s position on exceptions to mutual trust in the EAW has evolved and has offered more practical room for national courts to exercise distrust. Yet this still occurs in a context dominated by the obligation and requirement of mutual trust, as discussed in Section 3. However, this is not the only possible way to conceive of mutual trust as a legal principle: in Section 4 we showed that different accounts of mutual trust are possible, and that in fact the Court has already endorsed and developed them in other fields of EU law. Moreover, the Court has shown different degrees of tolerance of lawful distrust, by accepting in certain cases – namely in internal market law, asylum law, or execution of the EU budget – that mutual trust be used to generate cooperation that accommodates other EU interests beyond the effectiveness of a certain cooperative scheme.

This has led us, in Section 5, to propose a bidimensional principle of mutual trust, arguing that mutual trust should capture more than just a requirement for national authorities to trust, recognise, and engage in cooperation acritically. It should also, as a legal principle, be able to react to pathologies of trust and guide national authorities in managing cooperation under those circumstances. Simply put, it should include more ample possibilities for distrust. The latter should not be seen as something that jeopardises EU cooperation. Conversely, distrust is a functional equivalent to the requirement to trust, i.e. another possible path for allowing cooperation to take place in a context of uncertainty regarding national authorities’ behavior. Seen in this way, distrust can act as a vehicle for the rational justification of the trustworthiness of national authorities involved in cooperative schemes.

And here is where the trajectories of change we have presented come into the picture: zooming back in on the EAW, we strived to offer suggestions on how to concretise that different vision of mutual trust in the field where constitutional tensions have emerged more clearly, and where national courts have continuously asked for the Court of Justice’s guidance in managing the effects of and limits to mutual trust. Ultimately, we believe that for mutual trust to remain a valuable tool of EU integration – in the EAW and beyond – during and after the Union’s ‘rule of law crisis’, a thicker account of it must be developed. One that does not only rely on premises and imposed obligations, but that is grounded in reality.