A New Presumption for the Autonomous Concept of ‘Court or Tribunal’ in Article 267 TFEU

Introduction

The preliminary reference procedure is, according to the famous words of the European Court of Justice in Opinion 2/13, a keystone of the judicial system:

which, by setting up a dialogue between one court and another, specifically between the Court of Justice and the courts and tribunals of the Member States, has the object of securing uniform interpretation of EU law thereby serving to ensure its consistency, its full effect and its autonomy as well as, ultimately, the particular nature of the law established by the Treaties.¹

One of the main objectives of the dialogue between the European Court of Justice and national courts and tribunals is, therefore, to ensure the uniform interpretation of EU law among member states. But are there also any uniform criteria regarding what a ‘court or tribunal’ is for the purpose of engaging in a dialogue with the European Court of Justice? Or is the application of Article 267 TFEU left completely to what individual member states label in their national orders as a ‘court or tribunal’?

As we shall see in the first part of this study, this problem is not new, but it is clear from the case law of the past six decades that the European Court of Justice had chosen to develop its own uniform concept of what a ‘court or tribunal’ is for the purposes of Article 267 TFEU, instead of leaving it to the member states. It has been commonly understood among the scholarship that we are looking at ‘an autonomous concept’.²

In the past few years, however, the rule of law backsliding in the EU has led to a complete revolution in the case law of the European Court of Justice, which has developed more demanding requirements regarding judicial independence in the context of Article 19 TEU and Article 47 of the Charter. The ‘spill-over effect’ of such case law into the autonomous concept of ‘court or tribunal’ under Article 267 TFEU³ was welcomed as a new step towards a better protection of the rule of law in the EU.⁴ But it has also raised, as we shall see, concerns about the risk of excluding relevant courts from the judicial dialogue,⁵ concerns that have been addressed by the European Court of Justice in the judgment which is the subject of this commentary.

In the case of Getin Noble Bank, the European Court of Justice was confronted with questions regarding whether the requirements to be a ‘court or tribunal’ had been fulfilled both in the admissibility stage regarding the referring court, and in the merits stage regarding the Polish court (which – according to the referring judge – does not comply with such requirements). Considering the limits of this commentary and the different complexities of the case, in the following pages I will focus exclusively on the impact that this judgment has on the state of the case law regarding Article 267 TFEU. For this purpose, and after briefly presenting the facts of the case, the Opinion of the Advocate General and the decision of the European Court of Justice, I will analyse the content of the new presumption of judicial independence established in Getin Noble Bank and its consequences for the autonomous nature of the concept of ‘court or tribunal’ in Article 267 TFEU.

The facts of the case: a tale of (in)admissibility and (dis)appointment

At the national level, the factual background that led to the preliminary reference in Getin Noble Bank began with the classical David versus Goliath battle. On 3 March 2017, a group of individuals went to the Regional Court of Świdnica (Poland) and claimed that Getin Noble Bank should be ordered to pay them jointly the approximate amount of €40,000, together with statutory default interests, due to the unfair nature of the loan indexation mechanism included in their mortgage agreements.⁶

A year and a half later, the Regional Court accepted the claim, though only partially, as it understood that the indexation system was not unfair in its entirety. Therefore, the Regional Court ordered Noble Bank to pay the sum of €3,634 in a judgment that was appealed by the applicants in the main proceedings and confirmed later by the Court of Appeal of Wroclaw.⁷ This last judgment was appealed before Supreme Court under the claim that the Appeal Court infringed Article 385 of the Law of the Civil Code, in that it did not render the entire indexation system inapplicable to the parties, as it should have done as result of the declaration of unfairness of the indexation clause.

The Supreme Court decided to bring a preliminary reference before the European Court of Justice. Up to this point, one might think that the question would be related to EU consumer law. But, as is usual in the Polish jurisdiction in recent times, the case has turned into a conflict about judicial independence, though with one peculiarity: while previous preliminary references had been referred by ‘old judges’ in relation to the amended rules for the appointment of ‘new judges’ in Poland, this time is was the other way round.⁸ As we will see below, it is the independence of the ‘old judges’ of the Court of Appeal of Wroclaw under EU law that is called into question by a judge who was quite recently – and, as we will also see, quite controversially – appointed to the Supreme Court.

In Getin Noble Bank, the admissibility of the appeal before the Supreme Court was examined by a panel formed by a single judge, who noted that, according to Articles 7(1) and (2) of Directive 93/13 and Polish law, the state must provide for the possibility of bringing judicial proceedings against unfair contractual terms. Hence, the relevant question for the Supreme Court was whether the panel of judges of the Appeal Court of Wroclaw that rendered the challenged decision was a ‘court or tribunal’ within the meaning of EU law.⁹

According to the referring judge, there were strong reasons to doubt the judicial nature of the Appeal Court, particularly because of the potential lack of independence of three of these judges. One of them, FO, was appointed judge under the communist government of Poland, and was chosen later for the Appeal Court by decision of the President of Poland, adopted on a proposal of the National Council of the Judiciary (KRS) – an organ that, according to the referring court, was neither democratic nor impartial.¹⁰ The other two judges, GP and HK, were appointed to the Appeal Court by the National Council of the Judiciary in a period during which, according to a judgment of the Constitutional Court of 20 June 2017, such organ did not operate transparently, and its composition was contrary to the Polish Constitution.¹¹

To assess the independence of the panel of the Appeal Court, the Supreme Court considered necessary to examine the independence in concreto, assessing how the individual judges of the panel were appointed and the possible impact on the case that it may have, as this would be the only way to safeguard the trust of the citizenship in judicial institutions.¹² However, the referring court had doubts about whether, as a result of the judgment of the European Court of Justice in A.K. and Others,¹³ the assessment of the independence of a court or tribunal can be done only in abstracto.¹⁴

In brief, the European Court of Justice was asked whether a court with such irregularities in the procedure of appointment of judges in its panel could be considered a ‘court or tribunal’ withing the meaning of EU law. Here, the referring court had clearly studied the recent case law of the European Court of Justice regarding judicial independence, as it included in its question all potentially applicable legal bases: Article 2, Article 4(3), Article 6(1) and (3), and Article 19(1) of the TEU, in conjunction with Article 47(1) and (2) of the Charter and the third paragraph of Article 267 TFEU, Article 38 of the Charter and Article 7(1) and (2) of Directive 93/13.

The single judge sitting at the Supreme Court considered that the answers to the previous questions were crucially relevant to determining the admissibility of the appeal against the judgment of the Regional Court. However, it was in the preliminary reference procedure where the second tale of admissibility emerged. The Polish Ombudsman submitted to the European Court of Justice that the request for a preliminary ruling should be declared inadmissible, as the procedure for the appointment of the referring judge from the Supreme Court was flawed and created legitimates doubts about his independence and impartiality, up to a point that he should not be considered as a ‘court or tribunal’ in the sense of Article 267 TFEU, Article 19(1) TEU and Article 47 of the Charter.¹⁵ In particular, the requirements of ‘independence’ and ‘established by law’ would not be complied with, as he was appointed by a resolution of the National Council of the Judiciary that was suspended by the Supreme Administrative Court, and he was only in office because of a presidential order, as the European Commission also had pointed out.¹⁶ We are, therefore, looking at a case that involves two questions of admissibility: one at the national level that requires the interpretation of EU law, and one at the European Court of Justice level, because of different flaws in the procedure of appointment of judges in Poland.

The Opinion of Advocate General Bobek

The preliminary reference mentioned three legal bases for judicial independence in primary EU law: Article 267 TFEU, Article 47 of the Charter and Article 19(1) TEU. Advocate General Bobek stated that this multiplicity of legal bases did not necessarily mean that there were different categories of judicial independence.¹⁷ However, considering that those provisions have different functions and objectives, he proposed that a review of whether there was compliance with the requirement for judicial independence should differentiate between the three provisions in terms of intensity.

First, regarding Article 267 TFEU, the case law of the Court is clear in that a ‘court or tribunal’ must comply, inter alia, with the requirements of ‘independence’ and being ‘established by law’. The Advocate General proposed that such analysis should be done in relation to the organ as such exclusively, and not in relation to the specific individuals who sit in it.¹⁸ The reason for this relatively low threshold of intensity lies in the very function of Article 267 TFEU, which is to identify the interlocutors with the European Court of Justice through the preliminary reference procedure.¹⁹ The purpose of the concept of ‘court or tribunal’ in this context would be to distinguish national judicial bodies from non-judicial authorities. Therefore, the analysis should focus on institutional and structural issues, and not on whether each individual sitting in the organ complies with the Dorsch requirements, even if the body is composed by just one person.

In relation to the element of ‘established by law’, this requirement is meant to exclude bodies established by contracts, particularly courts of arbitration.²⁰ Considering that, according to the Advocate General, the review should focus on institutional and structural issues only, the requirement of ‘established by law’ should mean that the body must have been established by virtue of national legislation, and not that the members of the body in the case at hand have been appointed according to national law.²¹

The Advocate General stressed the difference between the purpose of the requirement of ‘established by law’ in Article 267 TFEU, on the one hand, and in Article 6 of the ECHR and in its replication in Article 47 of the Charter, on the other: ‘identifying the appropriate judicial interlocutors in terms of bodies in a Member State which may refer a question to the Court of Justice is different to detecting breaches of the lawful composition of the bench in each individual case in order to protect individual EU law-based rights’.²² This means that, while in the framework of Article 47 of the Charter the examination must obviously be done at the level of each individual composing the panel, this is not necessarily the case with Article 267 TFEU.

In relation to the requirement for ‘independence’, this institutional and structural approach would be also applicable. The Advocate General defended the fact that, according to previous case law, the review of the requirement of independence of Article 267 TFEU was focused on the national rules that protect the independence of the body as such from external and internal pressures, but ‘without inquiring into the specific position of the inquiring judge’.²³

The application of this perspective – of how to assess what a ‘court or tribunal’ is in the sense of Article 267 TFEU – to the referring judge of the Supreme Court of Poland led the Advocate General to conclude that the preliminary reference should be admitted. The arguments presented by the Ombudsman were related to the personal and professional capacities of the referring judge, but they did not affect the judicial nature of the Supreme Court from an institutional and structural perspective.²⁴ Finally, and though the Advocate General considered that it was not the case here, he also wanted to stress that an accumulation of individual issues that leads to an overall malfunctioning of the body or the hijack of the judicial institutions could, in such extreme circumstances, be assessed at the admissibility stage of the preliminary reference procedure.²⁵ If the case at hand did not involve such extreme circumstances, the verification of the impartiality and independence of specific judges should take not take place at the admissibility stage, as the European Court of Justice is at this point ill-equipped to make an in-depth analysis – which would include interpretations of national laws potentially disputed by the parties – and a re-examination of the arguments at the merits stage would make the procedure ‘somewhat circular’.²⁶

A second dimension of the principle of judicial independence would be that of Article 47 of the Charter, which protects the right to an effective remedy and a fair trial. For the Advocate General, the assessment should focus here on the particular elements of the case in question, including institutional or structural issues only when they may have an impact on the individual proceedings.²⁷ Here, the intensity of the review of the European Court of Justice should, in words of the Advocate General, be ‘moderate’, as not all breaches of law would be a violation of Article 47 of the Charter, only those who met a certain standard of gravity.²⁸

Finally, the third dimension of judicial independence is found in Article 19(1) TEU, which requires member states to ensure effective legal protection. This provision is concerned with the structural and institutional framework of the national judiciary, being therefore connected to the question of whether or not a national judicial system complies with the principle of the rule of law included in Article 2 TEU.²⁹ For the Advocate General, the threshold to breach this provision was rather high, as it should consider only ‘breaches of a certain seriousness and/or of a systemic nature’.³⁰ Unlike the review under Article 47 of the Charter, the specific elements of the case only come into play when they are an illustration of a systemic deficiency.³¹

Regarding the merits in the Getin Noble Bank case, Advocate General Bobek considered that the Appeal Court of Wroclaw should also be considered a ‘court or tribunal’ under Article 19(1) TEU and Article 47 of the Charter. First, because the fact that one of the judges was appointed during the communist era did not necessarily mean that he was not independent under the optic of Article 47 of the Charter or that the body was not independent in the sense of Article 19(1) TEU, nor did the Ombudsman give any argument to explain how the judge could be subject to external influences today.³² In addition, compliance with the Copenhagen criteria – which include, inter alia, the respect of democracy, rule of law and human rights – was met when Poland entered the EU.³³ Second, because the judgment of the Constitutional Court that declared unconstitutional the composition of the National Council of the Judiciary that appointed the judges of the Appeal Court was based on technical reasons – the length of office – that did not have an impact on their judicial independence. Following the previous case law of the European Court of Human Rights and the European Court of Justice in this area, breaches of a purely technical nature do not create doubts about the ‘imperviousness of the body in question to external factors’ when ‘observed through the prism of Article 19(1) TEU and 47 of the Charter’.³⁴

The decision of the Court

In its decision in the Getin Noble Bank case, the European Court of Justice started the admissibility test by reiterating that:

the meaning of court or tribunal is, according to settled case law, a question governed by EU alone. Such question must be analysed taking into account, among other factors, whether the body is established by law, whether it is permanent, whether its jurisdiction is compulsory, whether its procedure is inter partes, whether it applies rules of law and whether it is independent.³⁵

After this initial remark, the Court, following its previous judgment in A.B. and Others, pointed out that the purpose of Article 267 TFEU was to establish a procedure for dialogues between the Luxembourg court and the national courts, in order to ensure the consistency, full effect and autonomy of EU law among all member states.³⁶ In the context of this dialogue, it was not the function of the European Court of Justice ‘to determine whether the order for reference was made in accordance with the rules of national law governing the organisation of the courts and their procedure’.³⁷

The European Court of Justice made an innovative contribution to the case law in this field, stating that ‘in so far as a request for a preliminary ruling emanates from a national court or tribunal, it must be presumed that it satisfies those requirements, referred to in paragraph 66 above, irrespective of its actual composition’.³⁸ This new presumption, however, only applies in the framework of Article 267 TFEU, so it does not extend to the assessment of what is a ‘court or tribunal’ under Article 19(1) TEU or Article 47 of the Charter.³⁹

The European Court of Justice also added that this presumption was iuris tantum, as it may be rebutted by a final decision of a national or international court that:

leads to the conclusion that the judge constituting the referring court is not an independent and impartial tribunal previously established by law for the purposes of the second subparagraph of Article 19(1) TEU, read in the light of the second paragraph of Article 47 of the Charter.⁴⁰

In the specific case of the single judge formation of the Polish Supreme Court, the European Court of Justice considered that no information had been provided by the parties, at the time of closing the oral part of the procedure, that rebutted such presumption. The Court concluded, therefore, that the preliminary reference was admissible.⁴¹

In relation to the merits, the European Court of Justice reviewed whether the circumstances surrounding the Appeal Court were compatible with Article 19(1) TEU and Article 47 of the Charter. The Court began by stating that, under Article 19(1) TEU, member states have an obligation to establish a system of legal remedies and procedures that guarantee the effective judicial protection of individuals in all fields covered by EU law, regardless of whether in that specific case member states were implementing EU law or not.⁴² Any court or tribunal which was called to interpret and apply provisions of EU law must, therefore, comply with the requirement of judicial protection, as safeguarded by Article 6 of the ECHR and Article 47 of the Charter.⁴³ According to settled case law, effective legal protection requires that courts must comply with the principle of judicial independence, a principle that presupposes certain rules:

particularly as regards the composition of the body and the appointment, length of service and grounds for abstention, rejection and dismissal of its members, that are such as to dispel any reasonable doubt in the minds of individuals as to the imperviousness of that body to external factors and its neutrality with respect to the interests before it.⁴⁴

In the case at hand, it was clear to the European Court of Justice that the Court of Appeal of Wroclaw was a court that was called upon to rule on the application and interpretation of a provision of EU law, namely Directive 93/13, and so the Wroclaw court must comply with the requirements of effective judicial protection under Article 19(1) TEU and Article 47 of the Charter.⁴⁵ That said, the substantive analysis of the European Court of Justice closely followed the Opinion of Advocate General Bobek. Regarding the allegation that one of the judges of the panel was appointed during the communist regime in Poland, the European Court of Justice pointed out that the judicial system complied with the Copenhagen criteria, and that no information had been provided about how the circumstances surrounding the initial appointment would enable any external agent to exercise any undue influence today.⁴⁶ Therefore, for the European Court of Justice, the initial appointment during the communist regime could not:

per se be regarded as capable of giving rise to legitimate and serious doubts, in the minds of individuals, as to the independence and impartiality of that judge in the exercise of subsequent judicial functions.⁴⁷

Finally, in relation to the appointment of the judges by a resolution of the National Council of the Judiciary, in a composition that was declared unconstitutional by the Polish Constitutional Court, the European Court of Justice also followed a formal and functional reasoning to determine whether the requirement of ‘established by law’ was fulfilled. The Luxembourg court connected Article 47 of the Charter with the case law of the European Court of Human Rights about Article 6 ECHR, via Article 52(3) of the Charter.⁴⁸ According to the case law on this matter, this right to a tribunal established by law was closely connected to the requirement of independence, as both requirements have in common the fact that they seek to guarantee the separation of powers and maintain public confidence in the judiciary.⁴⁹

Furthermore, the case law of the European Court of Human Rights and the judgment of the European Court of Justice in Simpson established that the objective of the requirement to be ‘established by law’ was that the organisation of the judiciary does not depend on the executive, but on a law approved by the legislature.⁵⁰ A breach of the legal procedure for the appointment of judges can, therefore, entail an infringement of the requirement that a tribunal be ‘established by law’.⁵¹ However, not every error may lead to such a conclusion – only those:

of such a nature as to cast doubts on the independence and impartiality of that judge and, accordingly, on whether a formation which includes that judge can be considered to be an ‘independent and impartial tribunal previously established by law’ within the meaning of EU law.⁵²

Regarding the declaration of unconstitutionality of the relevant formation of the National Council of the Judiciary, the European Court of Justice pointed out that the Constitutional Court based its decision on the term of office of the members and the system of distribution of appointments among Polish courts.⁵³ Unlike other cases in which the unconstitutionality was based on the extreme influence of the executive and the legislative on the selection of members of the National Council of the Judiciary,⁵⁴ the European Court of Justice here, following the Advocate General, considered that the breach was not serious enough to call ‘into question the independence of that body or, consequently, of giving rise to doubts, in the minds of individuals, as to the independence of the judges concerned with regard to external factors’.⁵⁵

Comments

Addressing the concern of ‘blind spots’ in the preliminary reference procedure after Banco Santander

In Getin Noble Bank, the European Court of Justice was confronted in the admissibility stage with a difficult dilemma due to the particularities of the Polish context. On the one hand, the referring judge was surrounded by serious accusations regarding his independence and impartiality, including whether he was appointed according to Polish and EU standards. For some authors, it was clear that one of the goals of the referring judge was to obtain some kind of legitimation from the European Court of Justice if the preliminary reference was admitted.⁵⁶ On the other hand, accepting the Ombudsman’s request on inadmissibility, by applying strictly the criteria developed in previous case law for Article 267 TFEU (discussed futher below), could potentially lead to a blocking of all preliminary references coming from the entire Polish judiciary, whose independence was impaired not only because of the system of appointments but also because of a new disciplinary regime that introduced sanctions because of the content of the judicial decisions, even those just requesting a preliminary ruling.⁵⁷

The legal argument to challenge the admissibility of the preliminary reference made by the single judge of the Supreme Court was clearly based in the case law of the European Court of Justice about what a ‘court or tribunal’ is for the purposes of Article 267 TFEU.⁵⁸ As we know, this line of cases started with the early Vaasen Göbbels case, which questioned whether the Scheidsgerecht (Arbitration Tribunal of the Fund for non-manual workers employed in the mining industry) could be considered a court or tribunal for the purposes of Article 267 TFEU.⁵⁹ In this case, the European Court of Justice established that the preliminary reference was admissible, despite the fact that the organ did not fall within the judicial power under Dutch law. The Court made its own assessment focused on elements such as the constitution of the organ under Netherlands law, its permanent nature, the application of adversarial rules of procedure similar to those used by ordinary courts of law, its compulsory jurisdiction and its obligation to apply rules of law.⁶⁰

Since Vaasen Göbbels, the European Court of Justice has introduced and developed the elements that should be considered for the assessment of whether a national body is a ‘court or tribunal’ under Article 267 TFEU. The famous Dorsch criteria pool the elements mentioned in previous case law and establish several requirements that must be complied with by an organ which wants to engage in a dialogue with the European Court of Justice: it must be established by law; it must be permanent; its jurisdiction must be compulsory; the procedure must be inter partes; it must apply rules of law; and it must be independent.⁶¹ These criteria are, to a greater or lesser extent, present in cases in which the European Court of Justice had to determine whether arbitration courts,⁶² economic and administrative courts,⁶³ competition authorities,⁶⁴ administrative bodies,⁶⁵ courts of auditors,⁶⁶ professional bodies,⁶⁷ ombudsmen,⁶⁸ appeal committees,⁶⁹ patent courts,⁷⁰ registrars⁷¹ or giudice di pace ⁷² were ‘courts or tribunals’ with the ability to engage in a dialogue through the Article 267 TFEU procedure.⁷³

If we examine the list of organs analysed under the lens of Article 267 TFEU, we can easily see that they are not strictly courts or tribunals under national law. But, as they intervene in the application of EU law, the application of the Dorsch criteria – and overall, that of independence – has been usually quite flexible, ‘allowing a broad variety of Member State entities to submit preliminary references, thereby furthering Article 267’s underlying objective of ensuring a uniform interpretation and application of EU law throughout the Member States’.⁷⁴

This flexible approach, however, has been the subject of strong criticism, including the famous words of Advocate General Ruiz-Jarabo Colomer, who considered the flexible interpretation of the European Court of Justice in Gabalfrisa as the culmination of ‘the gradual relaxation observed in the case-law of the Court of Justice in relation to the requirement of independence’.⁷⁵ Challenges to the rule of law over the past decade have led the European Court of Justice to adapt somewhat to this criticism, a movement that started outside the framework of Article 267 TFEU, but which has finally affected the interpretation of the Dorsch criteria.

As we know, it was in the Wilson case that the European Court of Justice began to depart from a flexible view of independence towards a stricter one. Following the case law of the European Court of Human Rights on Article 6(1) of the ECHR, the European Court of Justice explained that judicial independence has both an external dimension – which implies that the body is protected against external intervention or pressure – and an internal one – which implies, in brief, impartiality regarding the interests of the parties in conflict.⁷⁶ While the judgment in Wilson only concerned the Directive on the free movement of lawyers, the rule of law backsliding in the EU motivated the European Court of Justice to continue developing a substantive approach to the concept of ‘independence’ in the context of Article 19 TEU and Article 47 of the Charter in the well-known cases of ASJP ⁷⁷ and Commission v Poland.⁷⁸

This approach, however, did not end the rule of law saga. In the case of Banco de Santander, it was asked whether the Tribunal Económico-Administrativo Central (Spanish Central Tax Tribunal) was a ‘court or tribunal’ in the sense of Article 267 TFEU.⁷⁹ The question had already been answered positively two decades earlier in Gabalfrisa, when the European Court of Justice – despite to the Opinion of Advocate General Saggio⁸⁰ and the Spanish scholarship pointing out that these organs were not of a judicial nature and had no internal independence⁸¹ – considered that the Spanish law regulating tax tribunals ensured:

a separation of functions between, on the one hand, the departments of the tax authority responsible for management, clearance and recovery and, on the other hand, the Tribunales Económico-Administrativos which rule on complaints lodged against the decisions of those departments without receiving any instruction from the tax authority.⁸²

In Banco de Santander, due to new developments in the case law within the framework of Article 19 TEU and Article 47 of the Charter, the European Court of Justice expressly decided to ‘re-examine’ its previous approach to the element of ‘independence’ for the purposes of Article 267 TFEU.⁸³ Referring, among others, to the previous cases in ASJP and Commission v Poland, the European Court of Justice carried out a substantive analysis of the internal and external dimensions of the independence of the Tribunales Económico-Administrativos, concluding that they do not fulfil the criteria of independence in either of these dimensions.⁸⁴ This hardening of the criteria of independence in the framework of Article 267 TFEU was later confirmed in Anesco, where the European Court of Justice, contrary to what had been stated almost 30 years before,⁸⁵ concluded that the Spanish national competition authority did not fulfil the requirements of independence either.⁸⁶

With Banco de Santander, the European Court of Justice restricted the interpretation of the concept of ‘court or tribunal’ which had been, according to the scholarship, quite ambiguous.⁸⁷ Some authors have even considered that, with this change of course towards a stricter review of what a ‘court or tribunal’ is in accordance with the requirements of Article 19 TEU, the European Court of Justice had finally put ‘things in place’.⁸⁸ By contrast, authors like Reyns believed that the Court had tied ‘a dangerous knot’ giving the same content to the notion of ‘independence’ in the admissibility of preliminary references and in questions regarding rule of law infringements.⁸⁹ The consequence of this hardening of the admissibility review could be a ’structural inadmissibility’ of the preliminary references coming from the judiciaries of some member states and, therefore, ‘a blind spot on the radar of the Court of Justice’.⁹⁰ To avoid the risks of structural inadmissibility coming from the alignment of independence under Article 267 TFEU and Article 19 TEU, Reyns proposed that each autonomous concept of ‘court or tribunal’, and its attached element of independence, should be assessed in its own context.⁹¹

The previous proposal is mostly based in the Opinion of Advocate General Wahl in Torresi, who warned that an ‘overly strict application’ of the case law resulting from Wilson to the framework of Article 267 TFEU would produce a result opposite to those intended by Article 6 ECHR and Article 47 of the Charter, as a rigid interpretation could deprive individuals of the possibility of having their cases heard before the European Court of Justice.⁹² This line of reasoning was followed by Advocate General Bobek in Pula Parking and again, as we have already seen, in Getin Noble Bank, where he defended that the different purposes of the provisions of EU law referring to ‘court or tribunals’ should lead to different approaches to the application of the same criteria of independence.⁹³ Even more emphatic was Advocate General Tanchev in A.K., where he pointed out that the assessment of a ‘court or tribunal’ under Article 267 TFEU should be a qualitatively ‘different exercise’ than the assessment under Article 19(1) TEU and Article 47 of the Charter, considering that the purpose of the preliminary reference was to establish a judicial dialogue among the European Court of Justice and national courts to ensure the uniform interpretation of EU law.⁹⁴

Naturally, we could argue that the judgment in Banco Santander did not create any risk of ‘blind spots’ in the sense argued by Reyns and the Advocates General, as it was clear not only from that case, but also from previous case law on Article 267 TFEU, that the test of what is a ‘court or tribunal’ was only applied to organs that were not ‘courts or tribunals’ under national law, including the Tribunales Económico-Administrativos. However, it is also clear that the European Court of Justice had never specifically excluded the application of the Dorsch criteria to organs with such qualification under national law. The proof that such danger was not merely hypothetical is obvious from the admissibility test that the European Court of Justice had to carry out in Getin Noble Bank due to the allegations presented by the Polish Ombudsman.

As mentioned above, the European Court of Justice, instead of carrying out an examination of the Dorsch criteria according to the functional criteria proposed by Advocate General Bobek, decided to solve the risk of cutting the dialogue with certain national courts, clarifying that ‘in so far as a request for a preliminary ruling emanates from a national court or tribunal, it must be presumed that it satisfies those requirements’.⁹⁵ This presumption has been fiercely criticised in a recent article by Kochenov and Bárd, who consider that the ‘high standard’ established in Banco Santander ‘came to an abrupt end with Getin Noble Bank’, imposing a standard which is ‘a denial of what Article 6 ECHR and Article 19 TEU requires’ and is contrary to the core of the rule of law as established by the European Court of Justice and the European Court of Human Rights in previous case law.⁹⁶ From my point of view, however, the presumption does not necessarily constitute a break with the rule of law jurisprudence of the European Court of Human Rights and the high standard introduced in Banco Santander. The rebuttal of such presumption, as I will explain, is key to interpreting the presumption in Getin Noble Bank in line with the previous case law.

According to the European Court of Justice, the presumption derives from the very limits of its competence, considering that ‘it is not the function of the European Court of Justice to determine whether the order for reference was made in accordance with the rules of national law governing the organisation of the courts and their procedure’.⁹⁷ This acknowledgement, which allows the introduction of the new presumption clarifying the case law regarding Article 267 TFEU, is completed with an additional acknowledgment: that other national and international courts do have the function of determining whether or not national courts or tribunals, even if qualified as such under national law, are independent and impartial tribunals. As we have seen above, the European Court of Justice has added that the presumption may be rebutted by a final decision of a national or international court that:

leads to the conclusion that the judge constituting the referring court is not an independent and impartial tribunal previously established by law for the purposes of the second subparagraph of Article 19(1) TEU, read in the light of the second paragraph of Article 47 of the Charter.⁹⁸

Apart from final decisions that may be taken by the competent national court regarding judicial independence, and focusing on the part of international courts, the only one that I can think of that has the competence to decide that a national court is not independent, impartial, or established by law is the European Court of Human Rights.⁹⁹ That court has declared that the referring formation of the Polish Supreme Court does not fulfil the requirement of ‘established by law’ of Article 6 ECHR,¹⁰⁰ so we could say – following the rebuttal of the presumption of the European Court of Justice in Getin Noble Bank – that preliminary references from such formation of the Polish Supreme Court will not be admissible in the future.¹⁰¹

It may be also possible that by ‘international court’, the European Court of Justice is also indirectly referring to its own assessments, not in the admissibility stage but at the merits stage. Imagine, for instance, that the European Court of Justice had declared that the Appeal Court of Wroclaw did not fulfil the requirements of Article 19(1) TEU and Article 47 of the Charter.¹⁰² Then, under Getin Noble Bank, no more preliminary references would be admitted from that court, until the deficiencies detected under Article 19(1) TEU and Article 47 of the Charter had been repaired. This leads us also to conclude that, when an institution is labelled as a ‘court or tribunal’ in a member state, a negative outcome resulting from an analysis under Article 19(1) TEU and Article 47 of the Charter will allow the European Court of Justice to refuse, in future cases, requests from such organs to engage in a judicial dialogue through the preliminary reference procedure.

Therefore, instead of following the ‘three levels of intensity’ review of the one and only principle of judicial independence proposed by Advocate General Bobek, the European Court of Justice closed the circle started in Banco Santander, in such a way that the assessment of what is a ‘court or tribunal’ in the framework of Article 267 TFEU will involve an examination of Article 19(1) TEU and Article 47 of the Charter in two scenarios. On the one hand, if the referring body is not a court or tribunal under national law, the examination of the Dorsch criteria in the stage of admissibility will include, after the judgment in Banco Santander, an assessment of judicial independence under Article 19(1) TEU and Article 47 of the Charter. On the other hand, if the referring body is a court or tribunal under national law, the European Court of Justice will not carry out, in principle, an examination of the Dorsch criteria. Only if a final decision has been taken by a national or international court – here we also consider that a decision on the merits by the European Court of Justice should be included – that points out that the referring court is not an independent and impartial tribunal previously established by law, will the assessment be done under ‘the second subparagraph of Article 19(1) TEU, read in the light of the second paragraph of Article 47 of the Charter’, thereby closing the circle of a single understanding and assessment of the principle of judicial independence.

In my view, the contextual interpretation in three levels offered by Advocate General Bobek would, in the long term, have brought more clarity than the path taken by the European Court of Justice, as it directly explained which criteria a national authority had to comply with in order to submit a preliminary reference and pass the admissibility test. By contrast, the solution of the European Court of Justice has, in the short term, the advantage of maintaining and facilitating dialogue with organs that are part of the judicial power of a member state without the need to apply, in most cases, an admissibility test regarding independence. However, in those cases where there is a national or international judicial decision questioning the independence of the referring body, the European Court of Justice would have to carry out an examination of whether such decision ‘leads to the conclusion’ – thereby such external decision does not have an automatic effect – that the referring authority is not independent for the purposes of EU law. Such an assessment may become difficult in scenarios such as in Poland, where there are ‘battles of judges’ accusing each other of lack of independence. In addition, this path still poses the risk – like the proposal of Advocate General Bobek – that courts which can make a preliminary reference today will not be able to do so in the future.¹⁰³

Nevertheless, maintaining such a dialogue, even temporarily, helps the purpose of ‘uniformity’ intended by the preliminary reference procedure. At the same time, the fact that an organ – even if we have doubts about its independence – can ask the European Court of Justice about the interpretation or validity of EU law is not necessarily contrary to the right to an effective judicial remedy. As brilliantly explained by Advocate General Wahl in Torresi:

[a] strict application of the requirements under Article 6 of the ECHR and Article 47 of the Charter is necessary to strengthen the protection of individuals and ensure a high standard of protection of fundamental rights. However, an overly strict application of the criteria laid down in the Court’s case-law on the admissibility of references under Article 267 TFEU would risk producing the opposite result: individuals would be deprived of the possibility to have the ‘natural judge’ (the Court of Justice) hear their claims based on EU law and, as a consequence, the effectiveness of EU law throughout the European Union would be weakened.¹⁰⁴

Of course, we should be careful not to equate the presumption of Getin Noble Bank applied in the admissibility stage with a ‘validation’ of the referring national court under the optics of Article 19 TEU and Article 47 of the Charter, as some authors fear.¹⁰⁵ In this sense, the European Court of Justice was quite clear in establishing, apart from the analysed rebuttal, that the presumption:

applies solely for the purposes of assessing the admissibility of references for a preliminary ruling under Article 267 TFEU. It cannot be inferred from this that the conditions for appointment of the judges that make up the referring court necessarily satisfy the guarantees of access to an independent and impartial tribunal previously established by law, for the purposes of the second subparagraph of Article 19(1) TEU or Article 47 of the Charter.¹⁰⁶

To sum up, since the decision in Getin Noble Bank, it seems that the European Court of Justice will admit preliminary references coming from courts or tribunals labelled as such under domestic law, unless a final judicial decision from a national or international court can lead to the conclusion that the referring court is not independent, impartial and established by law for the purposes of Article 19(1) TEU, read in light of Article 47 of the Charter. Nevertheless, while a solution based on the national labelling of what is a court or tribunal serves the European Court of Justice in reducing the risks of creating ‘blind spots’ after the judgment in Banco Santander – at least while a decision from a national or international court has not been taken – it also has an impact on the nature of what has traditionally been known as the autonomous concept of ‘court or tribunal’ in Article 267 TFEU.

Still an autonomous concept after Getin Noble Bank?

In his analysis about whether the referring court is an independent body established by law, Advocate General Bobek suggests that ‘[t]he concept of “court or tribunal” within the meaning of Article 267 TFEU is an autonomous concept of EU law and must be defined independently of denominations and qualifications under national law’, a meaning which must be determined according to the Dorsch criteria.¹⁰⁷ This qualification of ‘court or tribunal’ as an autonomous concept is in line with the recommendations of the European Court of Justice to national courts and tribunals in relation to the initiation of preliminary ruling proceedings, in which it is expressly indicated that the ‘status as a court or tribunal is interpreted by the Court as an autonomous concept of EU law’.¹⁰⁸

From my perspective, one crucial point must be made in this regard. Does it still make sense to talk about the autonomous concept of ‘court or tribunal’ under Article 267 TFEU, when we now have a presumption according to which a court under national law is also a court for the purposes of Article 267 TFEU? We could of course argue that the question is irrelevant, as this autonomous concept has traditionally been applied in examining organs which were not courts under national law. However, as we have explained in the previous section, such exclusion of national courts from the Dorsch criteria was never specific, and we had to wait until the controversy generated after Banco Santander to have a clear pronouncement on the matter by the European Court of Justice in Getin Noble Bank. Therefore, it is relevant to re-examine the image of ‘court or tribunal’ under Article 267 TFEU as an autonomous concept in light of this new case law, starting with what an autonomous concept is.

While it is true that we do not have an absolute and comprehensive definition of what an autonomous concept is, academics who have analysed this topic in different normative regimes all agree that the purpose of the autonomous interpretation of a term is to ensure the independence and effectiveness of the provision of international law vis-à-vis national law.¹⁰⁹ If we go to general international law, authors like Brems point out that autonomous interpretation fits in the general rule of Article 31(1) of the Vienna Convention on the Law of Treaties, as it gives meaning to a term in the context of its own treaty, a meaning that might be different from the one that the same term has in domestic law.¹¹⁰ This essence of autonomous interpretation is much clearer if we go to the instruments of unification of private international law, an area in which the scholarship has repeatedly warned that the interpretation of rules of international law via the lens of national law leads to the nationalisation, or even deformation at the judicial level, of the rights and obligations assumed by states in the ratification of international conventions.¹¹¹

Similar conclusions can be drawn from the framework of the ECHR. In the famous Engel case, the European Court of Human Rights considered that:

if the Contracting States were able at their discretion to classify an offence as disciplinary instead of criminal, or to prosecute the author of a “mixed” offence on the disciplinary rather than on the criminal plane, the operation of the fundamental clauses of Articles 6 and 7 (art. 6, art. 7) would be subordinated to their sovereign will.¹¹²

In Chassagnou, the European Court of Human Rights was even clearer, stating that:

if Contracting States were able, at their discretion, by classifying an association as ‘public’ or ‘para-administrative’, to remove it from the scope of Article 11, that would give them such latitude that it might lead to results incompatible with the object and purpose of the Convention, which is to protect rights that are not theoretical or illusory but practical and effective […].

The term ‘association’ therefore possesses an autonomous meaning; the classification in national law has only relative value and constitutes no more than a starting-point.¹¹³

After analysing the case law of the European Court of Human Rights on this matter, Letsas concludes that ‘domestic law classification is relevant but not decisive for the meaning of the concepts of the Convention. This is what the adjective “autonomous” stands for: the autonomous concepts of the Convention enjoy a status of semantic independence – their meaning is not to be equated with the meaning that these very same concepts possess in domestic law’.¹¹⁴ Other authors arrive at the same conclusion, considering that the interpretation of concepts independently from national laws is required to ensure the effectiveness of any international instrument, and not only of the ECHR.¹¹⁵

Finally, the independence of autonomous concepts towards national law is even more evident at the EU level. For Charbonneau, these are ‘terms which are the subject of a uniform definition, distinct from that given by the various national laws, and which meets the objectives of the treaties’.¹¹⁶ In Unger, the first case on autonomous interpretation, the European Court of Justice stated that certain key concepts, such as that of ‘worker’, cannot be interpreted in light of national law, as it would deprive EU law of all its effects.¹¹⁷ In Ekro, a case frequently quoted by the European Court of Justice when it is going to proceed to an autonomous interpretation, the Luxembourg Court developed this approach, stating that:

the need for a uniform application of Community law and the principle of equality require that the terms of a provision of Community law which makes no express reference to the law of the Member States for the purpose of determining its meaning and scope must normally be given an independent and uniform interpretation throughout the Community; that interpretation must take into account the context of the provision and the purpose of the relevant regulations.¹¹⁸

In the development of this formula by the European Court of Justice we can see that, unlike in the context of the European Court of Human Rights, national law is not considered here even as a ‘starting point’ of interpretation. This greater aspiration of autonomy of the EU system is also present in the methods of interpretation chosen to develop autonomous concepts. While the European Court of Human Rights frequently makes use of the comparative method to look for a common denominator or consensus among the states on the content of a concept,¹¹⁹ comparative studies of the law of the EU member states are rarely made at the European Court of Justice level – and, if it is done, it is not revealed in the judgments – as the court prefers to focus on teleological methods of interpretation.¹²⁰

Considering that the essence of an autonomous concept, both at the international and the EU level, is independence vis-à-vis national law, I think that it is difficult to maintain, after the presumption in Getin Noble Bank, that the concept of ‘court or tribunal’ in Article 267 TFEU has an autonomous nature in all circumstances. In the case of court or tribunals considered as such under national law, the autonomous nature of the concept might be now excluded for two reasons: first, because the label of ‘court or tribunal’ in national law activates the presumption that the national authority is able to make a preliminary reference; and second, because the only way to break such a presumption is through a final decision of a national or an international court, therefore leaving the question of whether a national court fulfils the requirements of what is a ‘court or tribunal’ under Article 267 TFEU to a national court that will not necessarily respect the principle of uniformity or to an international court which is not part of the EU system.¹²¹

Final remarks

The assessment made by the European Court of Justice in Banco Santander of the requirement of independence to determine what is a ‘court or tribunal’ for the purposes of the preliminary reference, under the stricter optics of Article 19 TEU and Article 47 of the Charter, was considered by many an improvement to the protection of the rule of law, as opposed to the previous multiplication of standards of judicial independence. For others, however, the homogenisation of the assessment in the different frameworks of Article 267 TFEU, Article 19 TEU and Article 47 of the Charter dangerously ignored the different contexts and purposes of each of these provisions, creating the risk of ‘blind spots’ on the radar of the European Court of Justice. From this point of view, such decontextualised interpretation of Article 267 TFEU could lead to a situation in which the national courts and tribunals of certain member states, whose independence is threatened by the executive but nevertheless apply EU law on a daily basis, could no longer engage in a dialogue with the European Court of Justice.

In Getin Noble Bank, the European Court of Justice seems to have heard this last concern, establishing the presumption that the Dorsch criteria are fulfilled when the request for a preliminary ruling emanates from a national court or tribunal. This new presumption has been quickly criticised by those who previously applauded the approach started in Banco Santander, qualifying the Getin Noble Bank presumption as an attack against the principles protected by Article 19 TEU and Article 47 of the Charter and against the rule of law jurisprudence of both the European Court of Human Rights and the European Court of Justice, helping ‘fake judges’ to legitimise their positions.

In this analysis, by contrast, I have maintained that a comprehensive analysis of the presumption in its context helps in understanding how the Dorsch criteria should be applied, and why it is not a step back from the standard established in Banco Santander, but rather an additional step in the same direction. Regarding the first part, the European Court of Justice never expressly stated whether the Dorsch criteria had to be assessed not only with respect to organs that are not courts or tribunals under national law, but also to courts or tribunals qualified as such under national law. Now, it is clear that these criteria – which, after Banco Santander, will be examined in the light of Article 19 TEU and Article 47 of the Charter – will always be applicable to references presented by non-judicial organs under national law, and that they will be applicable to national courts or tribunal only when the mentioned presumption is rebutted.

Concerning the presumption, it is extremely important to point out that it is composed of three parts, all equally important. The first one is the presumption itself, according to which it must be presumed that national courts comply with the criteria to present preliminary references under Article 267 TFEU. The second part is the field of application of the presumption, which specifically excludes its extension to Article 19 TEU and Article 47 of the Charter, thereby addressing the concern about the legitimisation of ‘fake judges’. Finally, the third part is the rebuttal, which renders the iuris tantum presumption inapplicable when a final decision of a national or international court – and also, from my point of view, a previous decision on the merits by the European Court of Justice – leads to the conclusion that the national courts do not respect Article 19 TEU and Article 47 of the Charter. This presumption, in its three parts, at the same time: allows the dialogue with national courts to be maintained – at least until a potential rebuttal; avoids the legitimisation of ‘fake judges’; and permits the analysis of the admissibility of preliminary references made by national courts or tribunals under the higher standard of Banco Santander when the conditions for the rebuttal are met.

Additionally, this commentary has also examined the consequences of the Getin Noble Bank presumption on the autonomous nature of the concept of ‘court or tribunal’ in Article 267 TFEU. In my opinion, this presumption clarifies that it is an autonomous concept but only in relative terms, as it is greatly dependent on the application of Article 267 TFEU of the classification of the organ under domestic law, which is precisely the opposite of what autonomous interpretation intends to achieve. The elements of the autonomous concept of ‘court or tribunal’ developed in the traditional case law on Article 267 TFEU will remain directly applicable to preliminary references coming from organs that are not labelled as national ‘courts or tribunals’, but they will not be applied to those organs qualified as courts or tribunals under national law, unless there is a rebuttal. It is only in relation to non-judicial bodies where we can still see the nature of the autonomous concepts to its full extent, as it will serve to guarantee the admission of preliminary references of some of these bodies, even if they are not labelled as courts or tribunals under domestic law.