Mutual Trust Under Pressure: Civil Justice Cooperation in the EU and the Rule of Law

Abstract

In recent years, widespread concerns have been expressed about some of the proposed reforms to the judicial system in Poland posing a risk to the rule of law and the independence of the judiciary in the country. The present contribution argues that the proposed reforms also threatens the mutual trust between EU Member States, which is the backbone of judicial cooperation in civil matters. It will show how mutual trust underpins the mutual recognition of judgments under various EU Regulations and discusses to what extent these regimes allow for exceptions to mutual recognition in case of fundamental rights concerns in the Member State of origin of a judgment. It is shown that the Court of Justice of the European Union (CJEU) maintains a strict approach to such cases, allowing for exceptions only in extreme cases. The case law of the European Court of Human Rights (ECtHR) is shown to be deferential to the authority of the CJEU in this regard.

1 Introduction

[L]egal certainty and trust are the preconditions of judicial co-operation and therefore the preconditions of mutual recognition and enforcement of court decisions across Europe.

These words are part of a letter from the Network of the Presidents of the Supreme Judicial Courts of the European Union (hereinafter: ‘the Network’) to the Vice-President of the European Commission, Mr. Frans Timmermans.¹ In this letter, the Network expresses—not for the first time²—its concerns regarding the intended judicial reforms and the rule of law in Poland. The proposed measures could, the letter warns, seriously undermine legal security and mutual trust within the European Union.

As the Network emphasises in its letter, judicial cooperation between EU Member States is founded on mutual trust in the adequate functioning of the legal and judicial systems in Member States. In the integrated legal order of the European Union, the independence and reliability of the judiciary are no longer solely the concern of individual Member States, but are essential to judicial cooperation. This is borne out by the fact that many EU Regulations now allow for recognition and enforcement of civil judgments across EU borders without any formal permission being required, but also without the possibility of checks in the Member State where recognition or enforcement is sought (see Sect. 2, below). The grounds for refusal (in particular the public policy exception) which a judgment debtor could traditionally invoke so as to halt enforcement of a foreign judgment against him are no longer available for all judgments.

As the letter from the Network quoted above points out, such free movement of judgments (i.e., without any checks) is only possible when it is based on a mutual trust between Member States: the confidence that judicial systems throughout the EU, though different, all comply with the essential guarantees of the right to a fair trial. A fundamental element of a fair trial is the right to adjudication by an independent and impartial tribunal. There are currently serious doubts as to whether, in the light of certain proposed reforms, the Polish judicial system will continue to comply with this guarantee.³ One controversial proposal is the intended amendment whereby the majority of the members of the National Council (whose main task is the nomination of judges) would be elected by Parliament. This would give the Polish Parliament paramount control over the election of judges and in the long run would politicise the judiciary and undermine its independence. Another is the proposed procedure by which Polish court proceedings of the past 5 years can be re-opened under vague or arbitrary conditions. A third measure is the intended lowering of the age of retirement of Supreme Court judges, which would have a significant impact on the composition of the Supreme Court, impeding its autonomy and independence.

The proposals have been met with concerned reactions from (among others)⁴ the European Commission, which issued four Rule of Law Recommendations⁵ and even took a first step towards triggering the so-called ‘nuclear option’ of Article 7(1) of the Treaty on European Union (TEU): a temporary suspension of Poland’s right to vote in the European Council.⁶ The strong response from the European Commission is understandable. Protection of the rule of law and the independence of their judiciary are not only of concern to Polish citizens. The assurance that all EU Member States equally and adequately subscribe to these values is also the backbone for judicial cooperation in the EU, both in civil and in criminal matters. As stated above, some EU Regulations have abolished cross-border checks on the basis of the presumption that the rule of law is observed throughout the EU. Using the Polish situation as a test case, the present contribution aims to show what will happen when this presumption is no longer justified.

To this end, the article provides an outline of the current framework for the cross-border recognition and enforcement of civil judgments in the European Union (Sect. 2). Section 3 delves deeper into the role of fundamental rights, specifically the right to a fair trial, in that framework, in the light of recent case law from the Court of Justice of the European Union (CJEU) and the European Court of Human Rights (ECtHR). Section 4 discusses under what conditions EU law allows for an exception to mutual recognition on the basis of fundamental rights concerns, while Sect. 5 shows what the ECtHR requires from the EU and its Member States in this regard. Section 6 offers a conclusion.

2 Cross-Border Recognition and Enforcement of Civil Judgments in the EU

2.1 Background

Since the inception of the European Union, it has been acknowledged that effective judicial cooperation and in particular effective cross-border recognition and enforcement of civil judgments, are essential to the functioning of the internal market. Economic activity within the internal market is only possible when businesses and consumers are able to effectively resolve any disputes arising from that economic activity.⁷ Furthermore, as EU citizens live and work across borders, family relationships increasingly have an international element as well, thus increasing the need, for example, for effective cross-border recognition of divorce decrees.

To this end, the 1957 Treaty on the European Economic Community (the EEC Treaty) already contained a provision requiring Member States to enter into negotiations to ‘simplify formalities governing the reciprocal recognition and enforcement of judgments’. These negotiations resulted in the 1968 Brussels Convention,⁸ which was later replaced by Regulation 44/2001 on jurisdiction and the recognition and enforcement in civil and commercial matters (the ‘Brussels I Regulation’).⁹ This Regulation combined automatic recognition of civil judgments with a simple, harmonised procedure for obtaining permission to enforce in another EU Member State than that where judgments were originally handed down: the exequatur. Both recognition and enforcement could be refused on the basis of a limited number of refusal grounds.

2.2 Impact of Mutual Recognition

In 1999, it was decided at the European Council of Tampere that judicial cooperation needed to be further intensified.¹⁰ The cornerstone of judicial cooperation was to be mutual recognition. This concept originates in EU internal market law and denotes a system where goods lawfully marketed in one Member State should, in principle, be admitted to the markets of all other Member States.¹¹ One advantage of mutual recognition is that the need for extensive harmonisation of, for instance, product standards is reduced, though in practice a certain level of harmonisation has always been found necessary to mediate the consequences of mutual recognition.¹² Translated to the field of justice, mutual recognition means that differences in Member States’ substantive or procedural law should, in principle, not stand in the way of a free movement of judgments.¹³ This means that only in exceptional circumstances, Member States would be permitted to refuse the recognition or enforcement of a judgment from another Member State.

The idea as such was not new. As early as 1968, the Brussels Convention provided for a simple procedure (the exequatur) combined with a limited number of narrowly defined grounds for refusal. The innovation of the 1999 Tampere European Council and subsequent policy documents and legislation was that the exequatur should be abolished and that grounds for refusal should be limited even further. In fact, a number of instruments have since abolished refusal grounds altogether, and replaced them with ‘minimum standards’ which needed to be checked by the Member State where judgment was rendered (i.e., the Member State of origin). Since the 1999 Tampere European Council, developments in civil justice cooperation have been characterised therefore by an increased reliance on mutual trust by Member States in each other’s legal systems. Cross-border checks on the acceptability of judgments have increasingly been reduced. As shall be explained in more detail further on, this means that the circulation of civil judgments is increasingly automatic and that in some cases it will be impossible to ‘deny entrance’ to judgments that are the product of a clearly unfair procedure.

2.3 Free Movement of Judgments under Current EU Legislation

Free movement of judgments is about two things: recognition and enforcement. Recognition means accepting the determination of rights and obligations made by the court of origin.¹⁴ In other words, accepting its legal effects. Parties may then invoke it, either by taking steps to enforce the judgment or by relying on the legal effects in another procedure. For example, a judgment granting a divorce may be used to obtain maintenance in a separate procedure. Enforcement presupposes recognition but goes one step further. It means that the judgment creditor is able to take action (for instance by means of attachment of a bank account) to ensure that a debtor obeys the judgment.¹⁵ Under EU legislation, all civil judgments rendered by Member State courts with some exceptions are automatically recognised throughout the EU. Since the abolition of exequatur under Brussels I bis,¹⁶ it is also possible to enforce a judgment throughout the EU without any kind of procedural step being required.¹⁷ Depending on the applicable Regulation, it is however still possible to deny recognition or enforcement on the basis of a limited number of grounds for refusal. Currently, the picture looks as follows.

The main instrument governing free movement of civil judgments is the recast Brussels I Regulation, or the ‘Brussels I bis Regulation’. The Brussels I bis Regulation applies to most civil and commercial judgments.¹⁸ It allows for automatic recognition¹⁹ and enforcement²⁰ of these judgments throughout the EU. Article 45 provides that recognition may be refused on the application of any interested party, provided that one of the grounds for refusal applies. Article 46 provides that enforcement may be refused on the same grounds, but only on the application of the person against whom enforcement is sought (therefore, a more limited category of persons). In practice, refusal of recognition will often be invoked incidentally, i.e., as a defence, in a procedure where the effects of a judgment are invoked. For example: a Bulgarian judgment establishing that the seizure of a large shipment of whisky in the Bulgarian harbour by a Dutch company was unlawful is later invoked before the Dutch courts in a procedure to obtain compensation from the Dutch company for damage suffered because of the unlawful seizure. The Dutch company may then argue that the Bulgarian judgment cannot be relied on because one of the refusal grounds applies.²¹ The unlawfulness of the seizure—the basis for the claim for compensation—is then no longer a given. When it comes to enforcement, the person against whom enforcement is sought will be informed by way of a standardised certificate setting out the most important features of the judgment (Articles 43 and 53 Brussels I bis).²² He or she may then apply for refusal of recognition by filing an application with the appropriate court in his or her Member State (Article 47).

So which are these grounds for refusal? Under the Brussels I bis Regulation, recognition (or enforcement) can be refused on the following grounds:

• (a) if such recognition is manifestly contrary to public policy (ordre public) in the Member State addressed;

• (b) where the judgment was given in default of appearance, if the defendant was not served with the document which instituted the proceedings or with an equivalent document in sufficient time and in such a way as to enable him to arrange for his defence, unless the defendant failed to commence proceedings to challenge the judgment when it was possible for him to do so;

• (c) if the judgment is irreconcilable with a judgment given between the same parties in the Member State addressed;

• (d) if the judgment is irreconcilable with an earlier judgment given in another Member State or in a third State, involving the same cause of action and between the same parties, provided that the earlier judgment fulfils the conditions necessary for its recognition in the Member State addressed; or

• (e) if the judgment conflicts with:

  • (i) Sections 3, 4 or 5 of Chapter II where the policyholder, the insured, a beneficiary of the insurance contract, the injured party, the consumer or the employee was the defendant; or

  • (ii) Section 6 of Chapter II.

Refusal grounds (a), the public policy exception, and (b), insufficient service on a defaulting defendant, are most important when it comes to safeguarding the right to a fair trial. Refusal grounds (c) and (d) are more practical in nature and provide a safeguard against the impossibility of enforcing two irreconcilable judgments. Refusal ground (e) has some significance with regard to the right to a fair trial, because it allows Member States to refuse recognition or enforcement with regard to judgments that were handed down by courts in violation of certain specific rules on jurisdiction aiming to protect the weaker party in some cases (e.g., consumers or employees). In the sections below, this article focuses on refusal ground (a), public policy, since this—as is explained further on—is most likely to come into play when there are doubts about the independence and impartiality of the court that handed down the judgment whose recognition or enforcement is requested.

Before discussing how the public policy exception may come into play in such situations, it is important to emphasize that not all EU instruments which allow for the free movement of civil judgments contain a public policy exception, or indeed any refusal grounds relating to fair trial whatsoever. The following briefly discusses five instruments which fall into this category. It includes two autonomous European procedures,²³ i.e., the European Small Claims Procedure (ESCP)²⁴ and the European Order for Payment Procedure (EOP).²⁵ Both these procedures result in a title which is immediately enforceable throughout the EU, without any kind of permission being required and without any possibility of reviewing whether fundamental principles of fair trial were observed.²⁶ The idea is that such a review in the Member State where enforcement is sought is unnecessary because the judgments result from procedures which are uniform in all EU Member States and which are known to conform to the requirements of a fair trial. The EOP and ESCP Regulations allow for a review of the resulting judgment before the court of origin if the debtor was not served with the necessary documents in a manner allowing him to arrange for his defence.²⁷

The European Enforcement Order (EEO) operates on the same basis, but goes one step further.²⁸ The Regulation applies to uncontested claims (i.e., most default judgments) and allows the court of origin of such judgments to issue a certificate (the EEO) which constitutes an enforceable title. This allows for a judgment to be enforced throughout the EU, but again without any further permission being required and without any possibility to review whether fundamental principles of fair trial were observed.²⁹ In the case of the EEO, conformity of the procedure which resulted in the judgment with the right to a fair trial is ensured by ‘minimum standards’, the fulfillment of which is checked by the court issuing the EEO. These minimum standards relate to the effective service of documents and provision of information about the claim to the judgment debtor. Like the EOP and the ESCP Regulations, the EEO Regulation allows for a review of the resulting judgment before the court of origin if the debtor was not served with the necessary documents in a manner allowing him to arrange for his defence.³⁰

The ‘minimum standards’ model is also part of Regulation 2201/2003 (also called the Brussels II bis Regulation) where it concerns the recognition and enforcement of judgments in the case of the return of a child after the illegal removal from his or her Member State of residence. Since this category of cases presents its own particular complexities, it will not be discussed further in this article, beyond noting that for the judgments mentioned, refusal grounds have been abolished.³¹

Finally, a very practically important instrument is the Maintenance Regulation (Regulation 4/2009) which contains the simplest approach to free movement of judgments within its scope: for these judgments, exequatur has been abolished and no refusal grounds apply.³²

The judgment debtor’s position is therefore very different depending on whether a judgment is enforced against him under the ‘normal’ Brussels I bis regime, or on the basis of the EEO Regulation or one of the harmonised procedures. The EEO, ESCP and EOP Regulations only allow for review (in the Member State of origin) where service was not effective.³³ They do not provide a remedy in case the independence or impartiality of the court which issued a maintenance judgment, an EEO certificate, an EOP or the judgment in an ESCP may be questioned. By contrast, the debtor confronted with the enforcement of a judgment falling within the Brussels I bis regime is able to file an application with a court in his own Member State, in his own language, requesting that enforcement shall be denied on the basis of the public policy exception.

It is thus clear that whether or not a judgment debtor is capable of challenging recognition or enforcement depends on the regime for enforcement that was chosen by the judgment creditor. Automatic or complete mutual recognition—no refusal of recognition or enforcement, or review of any other kind in any Member State other than that where judgment was rendered—is a reality under some of the Regulations discussed above. Although the Brussels I bis Regulation, which encompasses most civil judgments, retains its public policy exception, the political impact of the gradual introduction of automatic mutual recognition in other instruments should not be underestimated. When the European Commission proposed a reform of the public policy exception under Brussels I bis so that it would only encompass ‘fundamental principles underlying the right to a fair trial’, this was hotly debated³⁴ and explicitly contested by some.³⁵ In the end, the public policy exception was retained in Brussels I bis. The following section discusses if and how it can be used to address fair trial violations.

3 Application of the Public Policy Exception

As follows from the preceding section, some EU Regulations allow for the possibility to refuse recognition or enforcement on public policy grounds, whereas others do not. The present section discusses the interpretation and application of the public policy exception as circumscribed by the CJEU. It is shown that public policy encompasses the right to a fair trial, but that not every violation of this right to a fair trial warrants the application of the public policy exception, because it should only be applied in extreme cases.

3.1 Public Policy in the Case Law of the CJEU

Public policy, or ordre public, is a concept which refers to the norms that are at the core of a political entity, such as a state, in other words the fundamental principles underlying the legal order of that entity.³⁶ Public policy plays a role in defining relationships between states or between states and international organisations. For instance, within the EU, public policy regulates the relationship between the Union and its Member States. A threat to national public policy is one of the few grounds a Member State may invoke to block the application of free movement law.³⁷ In the context of free movement of judgments, if the recognition or enforcement of a foreign judgment or the application of foreign law is deemed contrary to public policy, the public policy exception allows the court addressed to refuse recognition or enforcement or refuse to apply the foreign law. In this way, public policy is essentially used to protect a domestic legal order against unacceptable foreign influences. This form of public policy is therefore referred to as ‘external’ public policy or ordre public externe.

It follows that public policy is a narrow concept: it encompasses only fundamental values underlying a legal order. The effects of its application are radical. Refusing enforcement to a foreign judgment negates the judgment creditor’s right to enforcement³⁸ and is contrary to the principle of res judicata. Especially in the EU, where all Member States are bound both by the European Convention on Human Rights (‘ECHR’) and by the EU Charter of Fundamental Rights (‘CFR’) to guarantee a fair trial and where cooperation is governed by mutual trust, it seems very radical for a Member State to refuse enforcement of a judgment from a fellow Member State if that judgment is not the result of a fair trial. Yet it is clear from CJEU case law that the right to a fair trial is part of the concept of public policy as included in the EU Regulations discussed above.³⁹ In its judgment in the case of Krombach v. Bamberski,⁴⁰ the CJEU ruled that application of the public policy exception would be appropriate if recognition or enforcement of the judgment would be

[…] at variance to an unacceptable degree with the legal order of the State in which enforcement is sought inasmuch as it infringes a fundamental principle. In order for the prohibition of any review of the foreign judgment as to its substance to be observed, the infringement would have to constitute a manifest breach of a rule of law regarded as essential in the legal order of the State in which enforcement is sought or of a right recognized as being fundamental within that legal order.⁴¹

The case was about Mr Krombach, a defendant in a civil trial, who did not want to attend this civil trial in France in person for fear of being arrested there. Under French law, however, this meant that he was not allowed to appoint lawyers to attend on his behalf, effectively barring him from defending himself at trial. In its judgment, the CJEU stressed the fundamental importance of the right to a defence and made clear that, as an element of the right to a fair trial, its violation might lead to a refusal of recognition or enforcement of a judgment:

With regard to the right to be defended, to which the question submitted to the Court refers, this occupies a prominent position in the organisation and conduct of a fair trial and is one of the fundamental rights deriving from the constitutional traditions common to the Member States.⁴²

The CJEU’s judgment in Krombach v. Bamberski makes clear that public policy encompasses the right to a fair trial, so that a violation of one of its elements (in this case, the right to effectively defend oneself) may trigger the application of the public policy exception. But does every fair trial violation warrant its application? The CJEU already gives some indication of a negative answer in its judgment as cited above: only a ‘manifest’ breach of a rule of law regarded as fundamental comes within the remit of the public policy exception. The CJEU has not given any indications as to when a violation should be seen as ‘manifest’, but it is clear that public policy is meant to cover only extreme cases, given the radical effects of its application. Whether a fair trial violation is sufficiently serious must of course be assessed in the context of the case at hand: empirical research⁴³ shows that the application of public policy by EU Member State courts usually involves situations where the defendant was prevented from defending himself due to practical reasons: for instance insufficient or even misleading information being given about the proceedings, prohibitively short time limits for the preparation of the defence were, or language barriers.⁴⁴

It also seems possible to identify categories of fair trial violations which would most likely not qualify. For instance, the right to a fair trial also includes a right to a trial within a reasonable time, yet it seems hardly appropriate to remedy a violation of that requirement by refusing enforcement of the judgment that was finally obtained. The importance of expeditious proceedings does not outweigh other elements of the right to a fair trial, such as equality of arms. The ECtHR has ruled that a speedy trial should not be ensured by limiting the number of documents that could be submitted for one party, but not for the other.⁴⁵ Equally, it would not be appropriate to remedy a violation of one party’s right to a trial within a reasonable time by refusing enforcement of the resulting judgment, thereby completely negating the other party’s right to enforcement of his judgment.⁴⁶ The public policy exception is an ultimum remedium: it should be applied in those cases where the debtor’s right to a fair trial was violated to such an extent that refusal of recognition or enforcement is the only way of remedying that violation.

It is submitted that cases where there is a clear absence of judicial independence or impartiality meet this criterion. Even though Article 6 ECHR nor Article 47 CFR distinguishes between more and less fundamental elements of the right to a fair trial, judicial independence or impartiality arguably is a prerequisite for all other elements of that right. If the judge deciding the case is prejudiced as to the outcome of the case, then the trial itself, elaborate though it may be, can amount to nothing more than going through the motions. It is for this reason that concerns regarding judicial impartiality, and in particular corruption in the judiciary, have been pointed out as a reason for keeping a public policy exception in EU Regulations. It has been argued that in cases of corruption, a cross-border remedy is especially necessary, firstly because corruption (especially where it is systemic) will not be detected in the Member State of origin, but more importantly, because it is not realistic to expect a judgment debtor to trust the Member State of origin to provide an adequate remedy.⁴⁷

3.2 ECtHR Case Law on Judicial Independence and the Evidentiary Standard

The ECtHR’s case law also bears out the paramount importance of judicial independence and impartiality. For instance, in its judgment in Micallef v. Malta, the Court held that even when deciding on a request for interim measures, such as injunctions, where speed is of the essence, the right to an impartial and independent tribunal must always be respected, even if other elements of Article 6(1) may only apply in so far as they are compatible with the nature and purpose of interim proceedings. Even in such cases, the right to an impartial and independent tribunal remains an indispensable and inalienable safeguard, according to the Court.⁴⁸

In most cases, the problem will probably consist in proving that there is an independence or impartiality issue in a specific case and that this lack of independence or impartiality has affected the outcome of the case. Assessing this may require a review of the internal rules of the judiciary in the Member State of origin. According to the ECtHR, the absence of procedures for requesting the withdrawal of judges suspected of bias may amount to an objective factor giving rise to legitimate doubt about the judges’ impartiality.⁴⁹ Likewise, an absence of independence may be derived from deficiencies in the judicial organisation, such as a lack of guarantees for a random assignment of cases,⁵⁰ or the absence of rules to prevent interference from executive powers.⁵¹ It seems clear that such a review requires knowledge of the procedural law and judicial organization in the Member State of origin of the judgment, which will most likely not be easily available to the judge deciding on refusal of enforcement. Assessment may be easier when a lack of independence of the judge deciding the case is immediately apparent from legislation. As an example, in the case of Brudnicka v. Poland, a violation was found in a maritime dispute because of the fact that the judges who decided were immediately accountable to the executive, and their appointment and removal was also decided on by the executive. This hierarchy was immediately obvious from the legislation governing the procedure.⁵² However, it is most likely that such cases where evidence of blatant interference is available are rare. Presumably, in most cases, the interference of the executive or other actors happens more covertly. It is submitted that in such cases, courts should be able to take into account generally available and objective information as to, for example, the presence of corruption in the judiciary.

This was the approach of the Dutch Court of Appeal on the recognition of arbitral awards in the Yukos Oil v. Rosneft case. Arbitral awards were annulled by a Russian court. Despite the annulment, Yukos Oil shareholders sought enforcement of the arbitral awards, stating that the trial in which they had been annulled was tainted by corruption. The Amsterdam Court of Appeal analysed a diversity of sources, including journalistic publications, reports by the Council of Europe and reports by Transparency International and other non-governmental organizations, which all concluded that there were severe concerns as to the Russian court’s independence from the State. It also considered that judges in other EU Member States had refused the extradition of former Yukos Oil employees on the basis that their prosecution was politically motivated. Finally, the Court of Appeal considered important that the Russian State had a vested interest in the outcome of the procedure: the State was the majority shareholder in Rosneft and its management consisted of persons who were also government employees. The close connection between the State and Rosneft was confirmed, according to the Court of Appeal, by statements from (then and again now) President Putin. On the basis of these sources, the Amsterdam Court of Appeal ruled that it was sufficiently likely that the courts who had decided on the annulment had been influenced by the State to the benefit of Rosneft, and that therefore this decision should be ignored in the proceedings on recognition.⁵³ It is submitted that, extreme though this example is, this approach is correct, given that it is unrealistic to require parties affected by corruption or partiality to provide evidence thereof.

3.3 Conclusion

Were the reforms in Poland to take effect, they would cast doubt on the independence of Poland’s judges. But would this be sufficient reason to refuse recognition or enforcement to a judgment handed down under a potential new regime?

Needless to say, the reform which most immediately affects judicial independence, i.e., the new retirement regime for Supreme Court judges, primarily casts doubt on Supreme Court judgments. Undoubtedly, the fact that the independence of the Supreme Court can rightfully be questioned casts doubt on the independence of the judiciary as a whole. It raises doubts as to the principle of irremovability of judges, which is a key element of the independence of the judiciary.⁵⁴ However, this does not automatically mean that all judgments from Polish first-instance courts are also affected to the extent that their recognition or enforcement should be called into question. Some of the proposed measures do affect ordinary judges, in particular the reform of disciplinary proceedings, rendering these disciplinary proceedings incompatible with Article 6 ECHR in the opinion of the European Commission.⁵⁵ At the moment (and as reported by the European Commission), it does not seem that disciplinary proceedings could be brought arbitrarily against ordinary judges as a way to ensure a particular outcome in a given case. However, if the new regime on disciplinary proceedings were to function in that way in practice, this might give rise to legitimate doubts as to a judge’s independence in a particular case. As ever, the question is whether there are legitimate doubts as to the judge’s impartiality or independence in a particular case, even if these doubts can be substantiated with generally available information about the organisation of the judiciary.

An interesting problem is posed by the proposed rule allowing for the arbitrary reopening of cases (the extraordinary appeal).⁵⁶ In this procedure, the Supreme Court may overturn any decision by a Polish court from the last 30 years. The extraordinary appeal can be lodged if it is necessary to ensure the rule of law and social justice and the ruling cannot be repealed or amended by way of other extraordinary remedies, and either it (1) violates the principles or the rights and freedoms of persons and citizens enshrined in the Constitution, or (2) it is a flagrant breach of the law on the grounds of misinterpretation or misapplication, or (3) there is an obvious contradiction between the court’s findings and the evidence collected.⁵⁷ This procedure has been condemned for destabilizing the judicial system and undermining legal certainty.⁵⁸ In the view of the European Commission, the extraordinary appeal does not conform to Article 6 ECHR. But does that mean that recognition or enforcement should be refused in another EU Member State? On the one hand, it would be the result of a procedure that clearly violated one of the principles that is part of EU public policy. On the other hand, refusal of recognition or enforcement negates the judgment creditor’s right to enforcement, whereas he may not have benefited from the appeal. This (potential) benefit to the creditor is, however, not a prerequisite for the application of the public policy exception. After all, its function is to protect the legal order of the Member State where recognition or enforcement is requested, not to provide a remedy to the judgment debtor. Given the strong condemnation of the extraordinary appeal procedure by the European Commission, there is a clear possibility that the recognition or enforcement of a judgment resulting from that procedure would be found to go against public policy.

An alternative route for a judgment debtor could be to invoke Article 45(1)(d) of the Brussels I bis Regulation, which provides that recognition or enforcement shall be refused if a judgment is irreconcilable with an earlier judgment given in another Member State or in a third State involving the same cause of action and between the same parties, provided that the earlier judgment fulfils the conditions necessary for its recognition in the Member State addressed. Though of course not designed for this purpose, it could be argued that it would be contrary to the principle embodied in this clause to enforce a judgment that is necessarily incompatible with an earlier judgment given in the same case between the same parties and one that had become final, before becoming subject to an extraordinary appeal.

4 Mutual Trust: A Rebuttable Presumption?

4.1 The Political Context

The preceding section discussed situations where refusal of recognition or enforcement on the basis of the public policy exception is possible. The present section discusses what can be done in cases where recognition or enforcement of a judgment is governed by a Regulation which does not allow for refusal on this basis. As discussed in Sect. 2, this category includes maintenance orders, judgments resulting from a European Small Claims Procedure, European Orders for Payment or default judgments for which a European Enforcement Order has been granted. What can a judgment debtor do if a judgment in one of these categories is enforced against him, which he suspects to be the outcome of a procedure in which the judge was not independent or impartial? The short answer is: very little.

Before giving the long answer, it must be pointed out that judgments that enjoy complete free movement under EU Regulations which have abolished refusal grounds are probably not the judgments that will be immediately affected by judicial reforms such as those in Poland. As described above, the proposed measures are primarily aimed at the Supreme Court and (in the case of an extraordinary appeal) at judgments deemed to affect the rule of law and social justice. At the moment, the types of judgment which benefit from complete free movement (i.e., maintenance judgments and money judgments of a low value) are perhaps not immediately at risk, but that does not mean that such a risk is only imaginary.

The following discussion is primarily important with regard to the political discussion about the free movement of judgments and the reality of mutual trust between the Member States, which is a prerequisite for free movement. It was discussed that, when a reform of the public policy clause under Brussels I bis was proposed, many opposed that reform, arguing that a public policy clause is necessary to protect fundamental rights, at least for now. It was perhaps at that time not foreseen that the rule of law in one EU Member State would come to be threatened in the near future, yet the possibility of something like that happening was clearly not considered to be purely theoretical. In that light, it is significant that though the Brussels I bis Regulation retains a public policy clause, it was nevertheless considered apposite to adopt Regulations which lack such a clause, such as the EEO, the ESCP, EOP and the Maintenance Regulation, which are currently in force. Even if only in specific cases, complete free movement of civil judgments, without the possibility of refusing recognition or enforcement, is currently possible in the EU, even when (as the following analysis shows) there are severe doubts concerning the impartiality or independence of the court handing it down.

4.2 Mutual Recognition and Fundamental Rights

The question whether Member States can refuse to apply EU legislation when that application would in their opinion result in a fundamental rights violation has given rise to some very significant CJEU judgments. This is particularly true where that legislation is based on the mutual recognition mechanism. Mutual recognition, as was explained in Sect. 1, is based on the presumption that the level of protection of the interest at issue is at a similar level in all Member States.⁵⁹ This presumption is, essentially, non-rebuttable. Once mutual recognition is in place, its application may only be halted in so far as this is allowed under EU legislation itself. Concerns about the current fundamental rights situation in another Member State are essentially of no importance, unless the CJEU rules otherwise.

The EU constitutional framework is as follows. When Member States apply EU law, they must respect fundamental rights.⁶⁰ To be clear, this means rights that are protected by the ECHR or CFR. Member States may under no circumstances refuse to apply EU legislation because that application would be contrary to their own conceptions of fundamental rights, which includes constitutional rights.⁶¹

If courts are to apply EU law in conformity with fundamental rights, they must have some discretion in that application.⁶² If application of EU law was to result in a fundamental rights violation, and the Member State had no discretion, the only way to avoid that violation would be to declare EU legislation invalid. It is a well-established doctrine of EU law that only the CJEU may declare Union law invalid for reasons of incompatibility with some other legal principle; national courts may not do so.⁶³ Therefore, in such cases, courts must make a preliminary reference to the CJEU to receive a ruling on the validity of EU legislation.⁶⁴

Where there are no grounds for refusal of recognition or enforcement, there is no discretion for the judge in the Member State where that recognition or enforcement is sought to review whether it should be granted. If (s)he believes that a fundamental principle is at risk, the only route is to request a preliminary ruling from the CJEU, asking whether that principle may indeed stand in the way of recognition or enforcement. This is essentially what happened in the Zarraga case, which was about the application of the Brussels II bis Regulation.⁶⁵ Under this Regulation, judgments ordering the return of children to their Member State of habitual residence must be enforced throughout the EU, without refusal on any grounds being possible. To ensure that the rights of the child are respected, the Regulation contains a minimum standard: the Member State ordering the return of the child must ensure that the child was given the opportunity to be heard, unless a hearing was considered inappropriate having regard to his or her age or degree of maturity. Only after such an opportunity was given may the return order be issued. In this case, the child had been taken from Spain to Germany by her mother. The Spanish court ordered the child’s return. The child was not heard by the Spanish court. The court refused both the mother’s application that she and her daughter be permitted to leave Spanish territory after her daughter’s hearing, and a request that the child be heard via video conference. The German court, which was asked to enforce the Spanish judgment ordering the child’s return to Spain, asked the CJEU, essentially, whether it could refuse enforcement of the Spanish judgment on the basis that the judgment contained a ‘serious infringement of fundamental rights’. The CJEU essentially said no. It ruled that the EU legislator had deliberately abolished grounds for refusal. It follows that the court of the Member State of enforcement is obliged to enforce the judgment, and it has no power to oppose either the recognition or the enforceability of that judgment.⁶⁶

Until now, the CJEU has thus shown itself reluctant to review the compatibility of mutual recognition mechanisms with fundamental rights. In the Zarraga case, the CJEU did not consider the question whether the protection granted by this provision was in fact effective, and did not grant the court in the Member State of enforcement any leeway to refuse enforcement for that reason.

4.3 The Exception to Absolute Mutual Recognition: ‘Systemic Deficiencies’

The preceding conclusion needs to be nuanced: the CJEU has shown itself to be prepared to make exceptions to the ‘near-absolute rule of mutual recognition’, namely in asylum cases and in cases concerning extradition of accused or convicted persons. EU asylum law also operates on the basis of mutual recognition. The Dublin II Regulation⁶⁷ and the recast Dublin III Regulation⁶⁸ create a Common European Asylum System (CEAS) by laying down rules that determine which Member State is responsible for examining an asylum application. The principle of mutual recognition as implemented by that Regulation entails that when asylum seekers make their way to another Member State, this state should send them back to the Member State that is responsible for examining the asylum application on the basis of the Regulation. Mutual trust in the context of the Dublin system entails that all Member States are expected to treat asylum seekers and examine their applications in conformity with the relevant rules of national, EU and international law, and that their fellow Member States should trust that they do so.⁶⁹

In the context of the CEAS, however, it has been decided that this mutual trust cannot be absolute. According to the CJEU, there should be a possibility to rebut the presumption that fundamental rights are respected in the Member State to which the asylum seeker is to be returned, when there is evidence that this is clearly not the case. This was decided in N.S. and M.E. and Others.⁷⁰ This judgment is the result of two joined cases, in which the applicants were asylum seekers who, under the CEAS, were to be transferred from the UK and Ireland back to Greece. At the time, conditions in which asylum seekers were received in Greece were very bad. The ECtHR had already ruled, in M.S.S. v. Belgium and Greece, that these conditions were incompatible with Article 3 ECHR, the prohibition on torture.⁷¹

Against this background, the CJEU could do little but to make an exception to the rule of mutual recognition. It started its reasoning by stating that the CEAS was ‘conceived in a context making it possible to assume that all participating States […] observe fundamental rights, and that Member States can have confidence in each other in that regard’. The EU legislature, according to the Court, was therefore able to use this principle of ‘mutual confidence’ in legislation, in such a way as to speed up and simplify the handling of asylum applications within the EU. Nevertheless, the CJEU stated:

[I]t is not inconceivable that that system may, in practice, experience major operational problems in a given Member State, meaning that there is a substantial risk that asylum seekers may, when transferred to that Member State, be treated in a manner incompatible with their fundamental rights.⁷²

The CJEU then stated that not every minor infringement of the Regulations by the responsible Member State should have as a consequence that the asylum seeker could not be transferred to that State, as that would deprive the Regulation of its substance and endanger the realisation of its goal. However, the CJEU stated:

By contrast, if there are substantial grounds for believing that there are systemic flaws in the asylum system and reception conditions for asylum applicants in the Member State responsible, resulting in inhuman or degrading treatment, within the meaning of article 4 of the EU CFR, of asylum seekers transferred to the territory of that Member State, the transfer would be incompatible with that provision [emphasis added].⁷³

The CJEU thus ruled that only ‘systemic flaws in the asylum system and reception conditions’, as opposed to ‘the slightest infringement’ of the Directives, could result in mutual trust not being justified and in an obligation for the transferring Member State not to transfer the asylum seeker to the State experiencing those systemic flaws. In the case of Greece, at that time, it was considered proven that such systemic flaws were present. The CJEU concluded by stating that the CEAS could not be applied on the basis of a conclusive presumption that the asylum seeker’s fundamental rights would be observed in the Member State to which he was to be transferred. Such a conclusive presumption ‘could itself be regarded as undermining the safeguards which are intended to ensure compliance with fundamental rights by the European Union and its Member States’.⁷⁴

The conclusion of the Court is clear: while mutual trust is required for the effective functioning of the regulation establishing the CEAS, it cannot result in a conclusive and non-rebuttable presumption that fundamental rights are protected everywhere. However, the threshold for rebuttal is very high. It is only justified when ‘systemic deficiencies’ give rise to ‘substantial grounds’ for believing that the asylum seeker would be exposed to a ‘real risk’. In the case of Aranyosi and Căldăraru,⁷⁵ which concerned the enforcement of a European Arrest Warrant (EAW), the CJEU nuanced the ‘systemic’ requirement somewhat by giving a number of quite specific guidelines regarding how a court should proceed in such cases. Initially, the court should determine, on the basis of ‘objective, reliable, specific and properly updated’ information, that there is a ‘real risk’ of the individual concerned being subjected to unacceptable detention conditions.⁷⁶ The court must then ‘assess whether there are substantial grounds to believe that the individual concerned will be exposed to that risk’.⁷⁷ While the ‘systemic’ requirement does not appear in this judgment, it does seem clear that the problems with detention conditions should be widespread and well reported. This is not likely to be the case where only individual and isolated incidents have occurred. Moreover, these circumstances must amount to ‘substantial grounds for believing that there is a real risk’ for the person concerned, which should be interpreted as a foreseeable and serious risk.⁷⁸

What does this mean for civil judgments that are subject to mutual recognition? Firstly, the yardstick the CJEU has developed, i.e., that deficiencies should be widespread, is understandable for fields of law where mutual trust applies prospectively. In the fields of criminal and asylum law, mutual trust exempts the judicial authority from having to make a prospective analysis of the future risks to which the person concerned is likely to be subjected. Such an analysis would require not only detailed knowledge of the situation in the Member State where he is to be sent to, but also an assessment of the risks that the person concerned will be subjected to. The lesson from the N.S. and Aranyosi and Căldăraru judgments is therefore, that a national judicial authority cannot be expected to make such an assessment unless the evidence is publicly available and there are substantial grounds for believing that the person will be subjected to serious risks.

In civil justice, however, mutual trust applies retrospectively to proceedings that have already taken place. This means that an exception to the rule of mutual trust is all the more necessary and acceptable from the viewpoint of efficiency, since the judge in a civil case is not required to prospectively assess possible risks and in so doing make a statement about the fundamental rights situation in another Member State. Rather, they can decide in individual cases and on the basis of the evidence already available in the judgment and the case file, whether an infringement has occurred. This means, in the author’s view, that in civil cases, there should not be a requirement that there should be widespread deficiencies before the presumption of trust is rebutted.

That said, this is not currently the view of the CJEU. As it stands, CJEU case law requires that there is publicly available evidence of widespread deficiencies before an exception to mutual recognition can be made. Presumably, this threshold would not be difficult to cross with regard to the current situation in Poland, in particular since the EU itself has extensively documented the situation in its Recommendations to the Polish government. It should also be shown that those deficiencies have affected, or on substantial grounds were capable of affecting, the person concerned. This is a more difficult test, especially because in civil cases there is usually not one but at least two parties involved. Finally, it should be emphasized that in no circumstances Member State courts may refuse to apply EU law on the grounds that fundamental rights are at risk (or on any other ground, for that matter): the CJEU always has the last word.

5 What Does the ECtHR Require?

5.1 The Bosphorus Test

A complex question is whether the ECtHR would ever require an EU Member State to refuse enforcement to judgments from another EU Member State if that judgment was the result of a procedure that did not conform to Article 6 ECHR. The ECtHR has ruled that ECHR rights also have an ‘indirect’ effect. States that are Contracting Parties to the ECHR have to protect the fundamental rights contained therein at all times, which means that they cannot hide behind the fact that the fundamental rights violation was (or will be) committed elsewhere and they are merely a conduit. In Drozd and Janousek, the ECtHR ruled that Contracting Parties could violate the ECHR by enforcing an arrest warrant that entailed sending accused persons to another state where they would be treated badly.⁷⁹ In Pellegrini, it ruled that Contracting Parties should review foreign judgments prior to enforcing them in order to ensure that the right to a fair trial was respected.⁸⁰ However, both these cases are about ECHR Contracting Parties enforcing decisions from non-Contracting Parties, i.e. states not bound by the ECHR. For a long time, it was therefore unclear whether the ECtHR would require Contracting Parties to review each other’s judgments. Since all EU Member States are also Contracting Parties to the ECHR, this would mean that EU legislation that removed any possibility of doing so—for example, by abolishing refusal grounds—could be found to be contrary to the ECHR. This would potentially put EU Member States in a very difficult position. On the one hand, EU law would require them to automatically enforce each other’s decisions without any kind of check, while on the other hand they could be found guilty of an ECHR violation for not performing such a check.

The ECtHR has, however, proven itself to be sensitive to this problem. In its well-known Bosphorus judgment, it essentially ruled that it would only perform a marginal review of acts committed by Contracting States in the implementation of EU law. It would only review whether there had been a ‘manifest deficiency’ in the EU’s framework for protecting fundamental rights.⁸¹ If not, the ECtHR would trust the EU’s framework—comprising control of Member State actions by the CJEU through its preliminary ruling procedure—capable of dealing with any potential fundamental rights problems. In this way, EU Member States are able to profit from a presumption that they have acted in conformity with the ECHR if they perform obligations imposed on them by EU law.

5.2 The ECtHR and Mutual Recognition

In its Povse decision and its judgment in Avotiņš, the ECtHR showed that the Bosphorus ‘test’ is also applicable to cross-border enforcement of judgments on the basis of EU legislation. The Povse case concerned the return of a child to her Member State of habitual residence under the Brussels II bis Regulation.⁸² The ECtHR ruled that, since EU Member States have no discretion in this regard—they cannot legally decide not to enforce the return order—the Bosphorus test applied. This meant that the ECtHR would review whether there had been a manifest deficiency in the protection of the rights of the child provided by the EU. Since there had been a preliminary ruling from the CJEU, the EU system for fundamental rights protection had been fully deployed. Consequently, the ECtHR ruled, there had been no manifest deficiency and the enforcement of the return order would not constitute a violation of the ECHR.

In its Avotiņš judgment, the ECtHR reached the same conclusion about the Brussels I Regulation.⁸³ In this case, Mr. Avotiņš, a national of Latvia, was confronted with the enforcement of a Cypriot judgment against him, which had been delivered by default. He argued that he had not been served with the documents instituting the proceedings in such a way and in sufficient time for him to prepare for his defence, and that no remedy in Cyprus was available, so that enforcement should be refused on the basis of Article 34(2) of the Brussels I Regulation (see Sect. 2, above). The ECtHR considered that, since the matter was governed by EU legislation, the Bosphorus test should be applied.⁸⁴ It found that there had been no manifest deficiency, but, it is submitted, its reasoning on this point was not altogether convincing. Essentially, the ECtHR concluded that Article 34(2) Brussels I requires that the enforcing court investigates whether the conditions for its application have been fulfilled, and that the enforcing (Latvian) court was able to conclude that those conditions had indeed been fulfilled, so that no violation of Article 6 ECHR was apparent. However, as judges Lemmens and Briede write in their concurring opinion to this judgment, the ECHR reached this conclusion by taking into account that Mr Avotiņš had been able to invoke a remedy in Cyprus (but had failed to do so), whereas this information was provided by the Cypriot government during the proceedings before the ECtHR and were not part of the reasoning of the Latvian court. In fact, the concurring judges write, Mr Avotiņš had not argued that he had been unable to invoke a remedy in Cyprus. Therefore the conditions of Article 34(2) were fulfilled and the Latvian court was not required to investigate whether a remedy had been available.

More importantly, however, the concurring judges point out that it was not necessary to apply the Bosphorus reasoning in the current case, because this would only be necessary if the case concerned the implementation of EU law and there had been a shortcoming in the proceedings, which in Avotiņš, there had not. It could even be put a little more strongly: application of the Bosphorus test was not appropriate in this case. This test is designed to provide a solution in cases where EU Member States have no choice but to implement EU legislation, even when this creates a risk that fundamental rights will be violated. They are then ‘absolved’ of their responsibility before the ECtHR, unless there has been a manifest deficiency. It is submitted that in the Avotins case, unlike in the Povse case, there was no complete absence of discretion: what was at issue was whether the Latvian court had adequately investigated whether the conditions for application of Article 34(2) Brussels I Regulation had been fulfilled. This is an investigation which is left to the Member States. It is unclear why, in that context, they could not be held accountable by the ECtHR in the ‘normal’ way if in this investigation they violated Article 6 ECHR. The fact that the ECtHR did apply the ‘reduced’ standard developed in Bosphorus shows that it is reluctant to find fault with (the application of) EU law, in the sensitive field of mutual recognition in particular, and that it trusts the (CJ)EU to provide adequate fundamental rights protection. This stance has caused some surprise⁸⁵ against the background of the CJEU’s negative opinion on the proposed accession of the EU to the ECHR,⁸⁶ in which it advised against accession partly because ECHR supervision may undermine mutual recognition.⁸⁷

Does this cooperative stance of the ECHR towards mutual recognition mean that it would not step in, even if, for example, it was argued that EU legislation required the automatic enforcement of a judgment from an EU Member State where there were severe doubts about judicial independence, as is currently the case in Poland? Though criticism can be levelled at the ECtHR’s treatment of the Avotiņš case (see above), the ECtHR did stress that the court of enforcement should be empowered to investigate ‘any serious allegation’ of a violation of fundamental rights occurring in the Member State of origin. It cannot escape this responsibility on the sole ground that it is applying EU law.⁸⁸ The court of enforcement should therefore take allegations of fundamental rights violations of a Member State of origin seriously so as to ensure that there was no manifest deficiency in the protection of fundamental rights which would warrant an exception to mutual recognition in the case at hand. It should also request a preliminary ruling on the matter from the CJEU in order to ensure that the mechanism for fundamental rights protection was fully deployed.⁸⁹ It is clear that the ECtHR expects the EU Member States to actively step in and investigate whether a manifest deficiency is apparent. Though the concept of manifest deficiency has never been defined,⁹⁰ the ECtHR’s own case law suggests that an obvious and systemic threat to judicial independence (see Sect. 3, above) would fit the bill. Going by the Bosphorus judgment, a failure to request a preliminary ruling from the CJEU could constitute a manifest deficiency in its own right, though the ECtHR has recently nuanced its stance on this matter. In any case, the lesson of Avotiņš is that though the ECtHR acknowledges that mutual recognition is designed to be automatic, it still expects EU Member States (and, by extension, the CJEU) to consider whether there are reasons not to apply it in specific cases when fundamental rights violations in the Member State of origin are apparent.

The question remains: what if, in a given case, the CJEU rules that the fundamental rights violations at hand were not sufficiently serious to warrant an exception to mutual recognition, and the ECtHR disagreed? Would it then still find that a manifest deficiency had occurred and rule that the Member State in question had violated the ECHR? This is—given the spirit of cooperation described above—unlikely, but perhaps not impossible. After all, in M.S.S. v. Belgium and Greece it has shown itself prepared to throw a serious (and necessary) spanner in the works of the EU asylum system. It may do so again, if inevitable, in the context of free movement of civil judgments.

6 Conclusion

The preceding sections have shown that, depending on the applicable regime, it may or may not be possible to challenge recognition or enforcement of a civil judgment on the basis that it is the result of a procedure that does not conform to the requirements of a fair trial. Although they are unlikely to be immediately affected by the reforms proposed in Poland, it is striking to think that for some categories of judgments, this type of review was completely abolished. As shown in Sect. 1, the implementation of free movement of judgments was believed possible because ‘the level of trust among Member States [had] reached a degree of maturity which [permitted] the move towards a simpler, less costly, and more automatic system of circulation of judgments’.⁹¹ Arguably, against the background of the developments discussed in this contribution, such a statement would sound very optimistic today. It must be conceded that the developments in Poland were probably difficult to foresee at the time. Yet it is a fact that when a reform of the public policy clause in the Brussels I Regulation was proposed, this triggered much spirited debate. Many commentators at the time expressed their belief that to prioritise free movement over fundamental rights protection is fundamentally wrong,⁹² or at least unwise in the current political situation, where differences in language, legal culture and national procedural law are simply too great to allow exequatur to be abolished.⁹³ Others did not oppose the reform as such, but nevertheless agreed that the application of refusal grounds sometimes remains necessary to address fundamental rights concerns.⁹⁴ Such views seem especially accurate today.