Hostname: page-component-848d4c4894-m9kch Total loading time: 0 Render date: 2024-05-02T09:53:17.322Z Has data issue: false hasContentIssue false

THE REBELLION OF CONSTITUTIONAL COURTS AND THE NORMATIVE CHARACTER OF EUROPEAN UNION LAW

Published online by Cambridge University Press:  18 December 2023

Csongor Istvan Nagy*
Affiliation:
University of Szeged, Szeged, Hungary and HUN-REN Centre for Social Sciences, Budapest, Hungary
Rights & Permissions [Opens in a new window]

Abstract

This article offers a reconstruction and assessment of the emerging rebellion of European constitutional courts against the exceptionless supremacy of European Union (EU) law. It presents the ontological theories of supremacy and how the Court of Justice of the EU (CJEU) overcame the first two major challenges of its history: the existential challenge of canonizing the general doctrine of supremacy and the Solange challenge of national fundamental rights. It provides an account of the emerging ultra vires challenge, including its root cause and evolvement, and provides an assessment and sets out proposals. The article demonstrates that the crux of the matter is not the primacy of EU law but the interpretive primacy of the CJEU. It argues that the rebellion was triggered by the perception that the CJEU case law features a declining normative and an increasing policy character. The debate about the CJEU's evolutionary interpretation, in a certain sense, parallels US constitutional law's debate between originalism and the living constitution, with the difference that the EU is a pluralist legal order.

Type
Articles
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
Copyright © The Author(s), 2023. Published by Cambridge University Press on behalf of British Institute of International and Comparative Law

I. INTRODUCTION

In the last few years, European Union (EU) law has seen the re-emergence of a debate that had been thought long-gone. National constitutional courts have a long history of denying the exceptionless supremacy of EU law over national law and have developed a set of limitations, reservations and qualifications. Despite this, until recently few had actually refused to apply an EU law provision or a ruling of the Court of Justice of the EU (CJEU).Footnote 1 Although these qualifications appeared to have developed into a mantra with little practical relevance, in recent years the plot began to follow Chekhovian dramaturgy.Footnote 2 In the last three years three national constitutional courts (the German, the Polish and the Romanian) have blocked the implementation of a CJEU ruling.Footnote 3 This ‘rebellion’ by constitutional courts, which, in terms of population, has already spread to one third of the EU, was triggered by various CJEU rulings viewed as ultra vires, and has ushered in an historic challenge to EU law.

This article offers a reconstruction of this rebellion and demonstrates that the crux of the matter is not the primacy of EU law but rather the interpretive primacy of the CJEU. It does not take sides on the cause of the rebellion but is guided by the principle that the sociological fact of the rebellion is independent of whether one applauds or reprehends it, and it brings forth a serious challenge for European integration.Footnote 4 The article does not argue that there is necessarily a direct link between an individual CJEU ruling and a defiant constitutional court decision. On the contrary, it argues that the rebellion has been triggered by national courts’ perception of a declining normative and increasing policy character of the CJEU's case law, and the tipping point cannot always be identified. The individual defiant national judgments were triggered by CJEU rulings that construed EU law in a way that arguably went beyond what was contemplated by the founding constitutional provisions. Although the national courts may not have been conscious of the collective nature of the process and addressed different aspects of the CJEU case law, their reactions are connected by the same root cause.Footnote 5 The debate about the CJEU's evolutionary interpretation, in a certain sense, parallels United States (US) constitutional law's debate between originalism and the living constitution, with the difference that the EU is a pluralist legal order and, in a sociological sense, the CJEU may arguably not enjoy the unquestionable authority of the US Supreme Court.

Section II presents the two ontological theories of supremacy: the CJEU's ‘explanation’, which originates EU law's supremacy in EU law itself; and national courts’ ‘constitutional authorization’ theory, which derives supremacy from the national constitution(s). Section III discusses how EU law and the CJEU overcame the first two major challenges of their history: the canonization of the general doctrine of supremacy (existential challenge); and the notion that national courts do not need to review EU law under the national constitution's fundamental rights provisions (Solange challenge). Section IV provides an account of the emerging third challenge (ultra vires challenge), including its root cause and evolution. Section V concludes.

II. WHY DOES EU LAW HAVE SUPREMACY?

The axiomatic truth that EU law has supremacy over national law has overshadowed the search for the doctrine's ontology. Arguably, European discourse has unduly focused on the CJEU's conceptualization and constitutional mythology, and downplayed (though not ignored) the heretical precept harboured in some of the Member States. Indeed, for a very long time, this asymmetric presentation of the heretical movements was of little practical relevance. After all, supremacy was accepted as an ecumenical faith by both the canon and the heretics; they differed only as to the derivation, and the ‘why’ remained simply an academic question as long as national courts did not refuse to apply a particular rule or ruling. The last couple of years have, however, shown that the ‘why’ does matter in cases of overload and it is the pivot in defining the ultimate limits of the interpretive primacy of the CJEU based mainly on Article 19 of the Treaty on European Union (TEU), among other provisions.Footnote 6

In an extraordinary display of self-sufficiency (perhaps even ‘bootstrapping’) within the EU's legal structure, the CJEU originated the supremacy of EU law in EU law itself, without the need for external validation or recognition. According to the Court, EU law is a ‘new legal order’Footnote 7 and although the sovereign powers enjoyed by the EU were conferred on it by the Member States, from the moment of conferral, EU law took on a life of its own. The EU is a quasi-federation by aggregation,Footnote 8 where the Member States created a self-contained, self-dependent and self-sustaining polity with an autonomous legal order. EU law is thus not a compilation of a set of concurring origin myths but has a single canonized origin myth. With this, the Court tried to secure the uniform status of EU law, instead of giving room to what were then six (currently 27) national constitutional mythologies. This was the only way for the CJEU to ensure equal status and uniformity in a constructed border-free market and to secure effective rights. The CJEU's conceptualization also imports that it is the prerogative of the CJEU to interpret the purview and the ultimate boundaries of EU law and, hence, its interpretation, by definition, cannot be ultra vires under any circumstances.

Interestingly, contrary to constitutional identity, which has a clear textual basis (Article 4(2) of the TEU), the doctrine of supremacy has never been codified in the EU Treaties.Footnote 9 The Constitutional Treaty aimed to solve the issue by providing in Article I-6 that ‘[t]he Constitution and law adopted by the institutions of the Union in exercising competences conferred on it shall have primacy over the law of the Member States’, but the Treaty was not adopted.Footnote 10 When it was converted into the Lisbon Treaty, this provision was specifically rejected as having the very constitutional character that was disallowed by the European pouvoir constituant.Footnote 11 This rejection might potentially give rise to a contrario arguments.Footnote 12 Article I-6 was replaced with Declaration 17 on Primacy attached to the Lisbon Treaty. Although the Declaration may be interpreted in such a way that the Member States signed up for the CJEU's narrative of supremacy, it is ambiguous how much legal weight national constitutional courts will give to its ‘recall[ing]’ the CJEU's case law. More importantly for the present analysis, however, the Declaration gives no hint as to interpretive primacy. After all, the crux of the matter is not that EU law has supremacy, which is a generally accepted fact in the Member States (it is not questioned even by the rebellious constitutional courts), but whether the allegedly ultra vires rulings of the CJEU, in the sense that a given national court considers that the CJEU has gone beyond the lawful scope of the Treaties in its interpretation, should equally have supremacy on account of the CJEU's interpretive primacy.

The conception of constitutional authorization, espoused by national constitutional courts, derives the status of EU law from the national constitution.Footnote 13 Accordingly, the bearers of sovereignty are the Member States, which conferred some of their powers on the EU. Some constitutions even conceive of the authorization as a ‘joint exercise’ of sovereign powers.Footnote 14 In this conceptualization, the Member States conferred no powers on the EU but exercise their sovereign powers jointly through the EU. In any case, in the narrative of constitutional authorization, the sovereignty of the EU is not separate, but rather it relies on the Member States’ systems and is supported by their national sovereignty. This conceptualization has important implications. The ultimate boundaries of EU law are set by the national constitutions, which are, in turn, interpreted by the national constitutional courts. Furthermore, given that national constitutional authorizations may differ from one another, these boundaries are not uniform and vary from Member State to Member State, and these different limits create fragmentation across the EU and EU law. National courts have the power to inspect whether the EU, including the CJEU, acts within the limits of the conferred powers. This conceptualization acknowledges the supremacy of EU law. It also acknowledges the CJEU's interpretive primacy, but only as long as it remains within the precinct of the conferred powers as interpreted by the national judiciary, including the national constitutional court.

The theory of constitutional authorization creates three major (and partially overlapping) limitations of EU law. First, the very core of national constitutionalism, the State's constitutional identity,Footnote 15 is inalienable; thus, the conferral of powers on the EU cannot be regarded as relinquishing that national constitutional identity.Footnote 16 An EU act that is counter to national constitutional identity lacks binding character within the given Member State(s). Second, Member States cannot confer more power on the EU than they have, and their power is limited by constitutionally protected fundamental rights. This implies that, to be considered binding, EU acts must comply with the fundamental rights protected by the national constitution.Footnote 17 Third, and most importantly for the present analysis, as the conferral of powers was carried out by the national constitution, it is the latter that defines the substance and the ultimate boundaries of the conferred powers. Only those EU acts that fall within the EU's conferred powers can benefit from legal status in that national legal order. EU acts outside the scope of the conferred powers are ultra vires and, as such, lack any binding force.Footnote 18

Of course, the question of ultra vires ultimately turns on the simple procedural question of interpretive primacy. The CJEU is not the exclusive interpreter of EU law, quite the contrary—the CJEU and national courts together make up the EU's pluralist judicial architecture. The CJEU has, however, by virtue of Article 19 of the TEU, interpretive primacy concerning EU law. Bluntly put, EU law is what the CJEU says it is. If the ultimate boundaries of EU law are set out in EU law itself, they come under the CJEU's interpretive primacy. On the other hand, if they are set out in the national constitutions, national constitutional courts have the interpretive primacy on this matter.

From a purely legal positivist perspective, the meaning of the law is what the body authorized to interpret it attributes to it.Footnote 19 The CJEU and national constitutional courts have, however, competing notions about authorization. For decades, the existence of these parallel universes was an academic question of little practical importance, as, at the end of the day, national constitutional courts, with the exception of the odd decision of the Czech Constitutional Court (Ústavní soud) in Holubec, had approved of the CJEU's interpretation. Using private international law terminology, these cases featured a ‘false conflict’.Footnote 20 Nonetheless, the constitutional controversies of the last few years have radically changed this landscape and called for the replacement of the current tacit balance with a clear solution.

III. THE FIRST TWO CHALLENGES TO SUPREMACY: EXISTENCE AND SOLANGE

The CJEU's interpretative primacy has faced three major challenges throughout its history. First, the Court had to sell the general doctrine of supremacy to national courts and make them accept it as a natural part of the law. Second, it had to remove EU law from the purview of national human rights review and convince national courts that they could safely relinquish this power as the CJEU expressly ensures that EU law complies with all national constitutional standards applicable to the exercise of public power. Third, the challenge currently faced by the Court is to entrench its interpretive primacy as the final arbiter: that EU law is what the CJEU says, however unreasonable, irrational or excessive the interpretation may be. This principle implies that although the Court must not act ultra vires, it is the Court which decides whether it itself acts ultra vires.

The first two of these challenges have already been overcome by the Court very successfully. The doctrine of supremacy is accepted throughout Europe without exception. Although national constitutional courts have made it clear that they retain the dormant right to review the acts of the EU under the human rights provisions of their national constitution, this power remains dormant as long as EU law has a sound system of human rights protection to control the operations of the EU. The third challenge is, however, currently unfolding and some of the controversial judgments of the CJEU have encountered blocking constitutional court decisions in some Member States, which have accused the Court of acting ultra vires. This challenge might prove to be the most difficult one to overcome. In a practical sense, the authority of the CJEU rests on the national judiciaries. The first two challenges were overcome in partnership with national courts, by convincing them, rather than opposing them, whereas the third challenge sees the national courts opposing the CJEU.

A. The Gospel of the New Legal Order

The very first challenge the CJEU faced was to sell the idea of the ‘new legal order’ to national courts. Although now a matter of history, at the time it amounted to a revolutionary change. The CJEU had to overcome very significant conceptual issues. When the European Economic Community (EEC) was established, as today, many Member States, if not the majority of them, followed a dualist approach to the relationship between international and domestic law, and most of those following a monist approach applied this to customary international law but not treaty law.Footnote 21 It thus amounted to a paradigm-shift that EU law (EEC law as it then was) operated in a purely monist fashion. A similarly important conceptual issue was to overcome national particularities and create a uniform and uniformly accepted system where national courts refrained from handling the issue under their national law. The CJEU successfully removed this issue from the realm of national laws and made it subject to a uniform regime.

At the end of the day, the supremacy of EU law was a sociological issue. The CJEU laid down important principles, but these remained an outstretched hand until national courts grabbed it. In reality, it took several decades for the Court to ensure that the doctrine was accepted throughout Europe. This was a well-documented process, where various national courts accepted EU law's supremacy at different points in time. In the narrative of the CJEU, the doctrine of supremacy was laid down in Costa v ENEL in 1964,Footnote 22 but was part of EU law from the outset. This is, however, only one of the narratives. Supremacy in fact became a reality when it was accepted by national courts, and different national courts accepted it at different times, with uniform acceptance not being achieved until the early 1990s.Footnote 23

B. The Solange Challenge: Escaping National Human Rights Review

While, as part of the doctrine of supremacy, the CJEU required national courts to put aside the human rights protected by the national constitution when applying EU law, the original text of the founding treaties contained no human rights provisions. This resulted in an untenable situation. The presence of human rights boundaries is essential for the proper functioning of public power, and the CJEU could not expect national courts to abstain from reviewing EU acts under the national constitution without ensuring that EU law secured the same level of constitutional protection.Footnote 24 It is the very essence of human rights that ‘the people are entitled to [them] against every government on earth, general or particular, and … no just government should refuse [them]’.Footnote 25

The CJEU overcame this problem by reading human rights into EU law as general principles of law. In 2000, these general principles of lawFootnote 26 were codified in the EU Charter of Fundamental Rights (the Charter)Footnote 27 and became a cornerstone of the European constitutional architecture. When first introduced, however, the idea was highly revolutionary, given the complete lack of a textual basis. The original text of the founding treaties made no reference to the rule of law or human rights and the CJEU, when first confronting the question in Stork in 1959,Footnote 28 refused to read such requirements into EU law. It ruled out the application of the constitutional principles of German law but offered no alternative in EU law. A decade later, however, the Court realized that comparable constitutional protection must be available in EU law in order to expect national courts to abstain from applying their own standards. In Stauder Footnote 29 the CJEU held that EU law encompasses a set of general principles of law, which, although not provided for explicitly, are part of EU law, and fundamental rights are included in these general principles. The CJEU understood that the recognition of human rights (and the rule of law more generally) was, as a precondition of EU law's supremacy, a constitutional necessity.Footnote 30 One year after Stauder, the CJEU pronounced the supremacy of EU law over national constitutions in Internationale Handelsgesellschaft.Footnote 31 This constitutional trade off can be seen as a successful strategy of the CJEU.Footnote 32 Following the preliminary ruling in Internationale Handelsgesellschaft, the German administrative court referred the case to the German Federal Constitutional Court (Bundesverfassungsgericht), which handed down its famous Solange I judgment.Footnote 33 This was, however, followed by Solange II, in which the Bundesverfassungsgericht accepted the CJEU's outstretched hand. It held that as long as EU law provides for a sufficiently high level of human rights protection, the Bundesverfassungsgericht would refrain from reviewing EU legal acts under the Grundgesetz, thus accepting the supremacy of EU law over the national constitution.Footnote 34

Subsequently, the constitutional requirements governing the EU's operations were extended to the Member States acting in place of the EU. In reality, the chief administrators and enforcers of EU law are not the EU institutions but national authorities. Member States regularly act as the EU's ‘agents’Footnote 35 and, thereby, their actions are attributable to the EU.Footnote 36 Later, 30 years after Stauder, in Wachauf Footnote 37 the CJEU held that the Member States must respect the general principles of law when implementing EU law. The case featured a contradiction similar to Stauder. The German authorities were required to apply EU law, which, in their reading, breached fundamental rights under the Grundgesetz. Denying fundamental rights protection was inconceivable for the same reasons applicable in cases concerning actions of the EU. At the same time, subjecting German authorities to the requirements of German constitutional law would have gone against the doctrine of supremacy, since that would have subjected EU law to German constitutional law. The only possible way out was to shield the national application of EU law from German constitutional requirements and to subject it to the corresponding EU requirements.Footnote 38 The result was the emergence of two distinct human rights regimes: EU fundamental rights governing EU actions and domestic fundamental rights governing domestic actions.Footnote 39

IV. THE ULTRA VIRES CHALLENGE OF THE CJEU'S INTERPRETIVE PRIMACY

The last decade has seen a number of cases involving EU law that have puzzled national courts. Some of these resulted in gentle defiance, others in head-on conflict. In Taricco, the Italian Constitutional Court (Corte costituzionale) invited the CJEU to ‘re-interpret’ its ruling and indicated that otherwise it would be compelled to pronounce it controlimiti, following the Italian version of the Solange principle. The Bundesverfassungsgericht, the Polish Constitutional Court (Trybunał Konstytucyjny) and the Romanian Constitutional Court (Curtea Constituțională) have each expressly pronounced a CJEU ruling non-binding for being ultra vires.Footnote 40 It seems that, unfortunately, Daniel Sarmiento's 2017 warning of an emerging judicial rebellion proved to be a Cassandra prophecy.Footnote 41

These were not the first cases where the CJEU adopted an ‘artistic’ interpretation and rendered decisions motivated more by policy considerations than normative arguments. In fact, the Court's jurisprudence is replete with judgments that circumvented express limits on EU competence but did not provoke national defiance.Footnote 42 However, whether justifiably or not, it seems that the above cases finally exceeded the constitutional threshold. They may have bitten into the core of national sovereignty by substantially affecting the division of competences within the EU and interfering with non-conferred sovereign powers, or they may simply have been the straw that broke the camel's back.

Below, an illustrative sample of recent cases that addressed sensitive aspects of national sovereignty is presented using the following taxonomy: first, two cases are addressed where the national constitutional court's reaction was provoked by the CJEU's extension of the EU's powers beyond what was originally constitutionally contemplated and, at times, even in defiance of the very language of the founding treaties; second, the tensions resulting from the suppression of national fundamental rights by EU law are addressed through the example of the Taricco saga; and third, some controversial rule-of-law judgments are presented, which have heightened the constitutional tension by provoking allegations of ultra vires.

A. Extending the EU's Powers: Is There Always a Way?

Two of the three defiant constitutional court judgments mentioned above were provoked by the perception that the CJEU had unjustifiably extended the EU's powers and changed the contemplated balance of competences between the EU and the Member States. While the CJEU might have seen a need to extend the EU's competences to make the EU effective or even operational, constitutional courts viewed the decisions as competence creep and as an extension of EU powers in breach of the principle of conferral.

1. The PSPP saga

The Weiss caseFootnote 43 concerned a challenge to behaviour of the European Central Bank (ECB) and the European System of Central Banks (ESCB) (hereafter jointly referred to as ‘ECB’) which arguably amounted to monetary financing. The ECB's competence is limited to monetary policy and Article 123 of the Treaty on the Functioning of the European Union (TFEU) expressly prohibits the ECB and national central banks from financing the Member States, for instance, by means of ‘overdraft facilities or any other type of credit facility’ or directly purchasing ‘debt instruments’ from them. The ECB formally complied with this provision but set up a programme (Public Sector Asset Purchase Programme; PSPP) to purchase national ‘debt instruments’ indirectly. The bonds were to be held until maturity and not resold.

The CJEU faced a difficult dilemma. On the one hand, the ECB used its monetary powers to pursue economic policy. The primary objective of monetary policy is to maintain price stability, not the financial stability of the Member States. Furthermore, the ECB's programme circumvented the prohibition of monetary financing by setting up a scheme that committed to purchase national bonds in the secondary market to achieve essentially the same macroeconomic result. On the other hand, the programme aimed to correct a serious institutional deficiency. The EU has a uniform monetary policy, but no uniform fiscal policy. The financial crises showed that the lack of the latter undermines the effectiveness of the former. The public commitment to purchase the bonds of Member States in financial difficulty functioned as financing: the ECB assumed a financial burden to capitalize (indirectly) the Member States and to build a safety net to calm financial markets and help them obtain sustainable interest rates.

In the end, the desire of effectiveness overshadowed the notion of limited competences. The CJEU found that although the measure was more economic than monetary policy, it was still covered by the ECB's monetary competence. It purported to carry out a ‘proportionality’ analysis but, in fact, for the Court it sufficed that the measures had more than a frivolous link to monetary policy. The CJEU held that, in essence, every measure that is rationally related to monetary aims, however remote this relation may be, is covered by the monetary competence.Footnote 44 The Court also held that the ECB has an enormous margin of appreciation to decide whether such a link exists and, ultimately, whether it acts within its competences.Footnote 45 Although the fact that the ECB should have the power to finance the Member States does not imply that it does, the Court also noted that successful economic policy, which is clearly not an EU competence, is beneficial to monetary policy; hence, the ECB can purposefully engage in economic policy, if its measures ultimately further the goals of monetary policy.Footnote 46

The CJEU's application of Article 123 of the TFEU may be viewed as similarly controversial. The Court pronounced that indirect purchases fall foul of this provision if they have ‘an effect equivalent to that of a direct purchase of bonds’.Footnote 47 However, it did not judge the PSPP under the ‘equivalent effect’ test but acquitted it on the basis of a ‘constructive proxy’ test: it concluded that the private purchases in the primary market were not ‘attributable’ to the ECB.Footnote 48 The Court's main argument was that the purchase of national bonds in the secondary market could amount to monetary financing only if the private purchasers were the ECB's de facto intermediaries and, hence, their purchases could be attributed to the ECB.Footnote 49 According to the Court, this could be established only if there was a straightforward and specific link between the private purchases and the ECB's promise to buy (immediately and fully) these bonds in the secondary market.Footnote 50 Not surprisingly, the CJEU found that there was no such link, as the PSPP did not specifically promise that the ECB would purchase every bond immediately. Although the ECB introduced a functional equivalent of monetary financing, private purchasers could not be regarded as de facto intermediaries, because they had no legal guarantee that they would later be able to sell all their bonds to the ECB. The Court concluded that Article 123 of the TFEU did not prevent the ECB from spending significant financial resources in the secondary market to stimulate the purchase of national bonds in the primary market with the purpose of providing financial help to the Member States.

The Bundesverfassungsgericht foundFootnote 51 the ruling ultra vires and, hence, to be ignored. It established that it could not rely on the CJEU's judgment, as its central part was ‘simply not comprehensible’ and ‘objectively arbitrary’Footnote 52 and took the interpretation of the relevant provisions of EU law into its own hands.Footnote 53 Its ‘independent’ interpretation concluded that the ECB decisions ‘manifestly’ violated ‘the principle of proportionality’.Footnote 54 The Bundesverfassungsgericht's opinion on the ECB's competence engineering was devastating. It concluded that the ECB abused its position by purposefully pursuing economic policy under the guise of monetary policy and using its monetary competence as a pretext to achieve economic policy results.Footnote 55 It also held that the CJEU's overly vague construction of monetary competence and the replacement of proportionality with the simple requirement of ‘rational relation’ deprived the principle of conferral of its substance.Footnote 56 The purpose of Article 123 of the TFEU is to prevent the ECB from spending EU taxpayers’ money so as to help Member States acquire finance. If the purpose of the PSPP programme is to provide economic help by means of assuming a financial burden, it goes against the prohibition of monetary financing. Of course, monetary policy may have beneficial financial side-effects, but once the chief or only purpose of the programme is to create this financial effect, the ECB transgresses its powers.

The division of competences between the EU and the Member States is not structural but functional. This means that that there is no clear division between what is ‘conferred’ and what is ‘reserved’. The ECB may adopt measures that are necessary to pursue the pre-set objectives. It is thus crucial to interpret these competences in a bona fide and proportionate manner and to limit EU action to measures primarily focused on the goal of monetary policy, and not those where it is merely a secondary consideration. With the employment of the ‘rational relation’ test, the CJEU threw open the door to competence creep. The Member States vested the ECB with the power to pursue monetary policy but did not authorize it to make the taxpayers of some Members States finance the budgetary deficit of others. A similarly puzzling element of the CJEU's approach was the immense deference afforded to the ECB to define its own competence. Professional bodies may enjoy a certain margin of appreciation when making policy choices within their competence, but they should enjoy no deference when the question is whether they are within the competences conferred on them. After all, it is the court's job to patrol the ultimate boundaries of EU competence and not the ECB's role to judge its own interpretation of its own competences.

This clash between the CJEU and the Bundesverfassungsgericht threatened to plunge EU law into a real constitutional crisis. Both the CJEU and national constitutional courts had believed that they were the final arbiter, and these conflicting beliefs—euphemistically referred to as legal pluralismFootnote 57—co-existed peacefully, provided they did not collide. The clash between the two judgments thus posed an unsolvable problem, necessitating instead the feigning of its non-existence. The Commission had little choice but to pay lip service to the interpretive primacy of the CJEU and not to insist on it too much to avoid effectively admitting that ‘the king is naked’. After all, interpretive primacy works in practice due to constitutional courts’ wise restraint, and extreme cases should not be allowed to upset this delicate institutional balance. The Commission's reluctance was obvious given that it was very slow to launch but very quick to terminate the infringement procedure. It took more than a year for the Commission to send a formal notice to GermanyFootnote 58 and it closed it as soon as the German government confirmed its recognition of the CJEU's interpretive primacy.Footnote 59 It is difficult to comprehend what the relevance of the German government's formal declaration was in this matter (let alone the serious separation of powers issue it raised that the political branches made commitments concerning matters that come under the power of the Bundesverfassungsgericht). In fact, the existential problem upon which the infringement procedure was based could only have been remedied in two ways: either by the issue of a different and compliant decision of the Bundesverfassungsgericht or, failing that, by amending the Grundgesetz. Since the task of interpreting the German Constitution lies with the Bundesverfassungsgericht and not the German government, and the Constitution was not amended, the German government was bound by the judgment of the Bundesverfassungsgericht. The legal validity of the judgment was not called into question, nor did the Bundesverfassungsgericht retreat from its position. The Commission defined the aims of the infringement procedure as rectifying ‘the future practice of the German Constitutional Court itself’ and sending a message to ‘the supreme and constitutional courts and tribunals of other Member States’.Footnote 60 None of these aims was fulfilled by the German government's response.

It seems it would have been a too high price to sacrifice the practical operability of EU law just to clarify a question of principle. Nonetheless, the ECB may still take the Bundesverfassungsgericht's judgment as a friendly (or unfriendly) warning, so it may not completely be devoid of influence on the interpretation of EU law.

2. Binding non-binding recommendations?

In Asociaţia ‘Forumul Judecătorilor din România’,Footnote 61 the debate concerned the independence of the Romanian judiciary and the CJEU was invited to decide whether the Commission's non-binding recommendations were binding. When Romania joined the EU in 2007, it did not meet EU standards in terms of judicial independence and corruption. As a compromise, the country was admitted to the EU but had to make commitments to overcome these deficiencies, which have to be achieved before Romania can join the Schengen area. The Commission has regularly monitored the progress made in these fields and issued recommendations in the framework of the Cooperation and Verification Mechanism (CVM).

Romania adopted a series of judicial reforms between 2017 and 2019,Footnote 62 which contained both structural changes and stringent rules on the disciplinary, civil and criminal liability of judges. The Commission criticized the reforms in its CVM reports for impairing judicial independence and recommended the withdrawal of various elements of the reform. After the Curtea Constituțională declared the reform package constitutional, Romanian courts made several references to the CJEU and claimed that the reforms breached general EU law, the benchmarks included in the CVM DecisionFootnote 63 and the Commission's CVM reports and recommendations.

The investigative powers and the unjustifiably stringent liability rules introduced by the reform exposed judges to unacceptable pressure and undue influence from the government and were condemned by the CJEU. As to the CVM reports and recommendations, however, the CJEU reached a staggering conclusion. It held that the Commission's recommendations are de facto binding and have direct effect and, hence, national courts must set conflicting national rules aside. The Court's reasoning was constructed in the following manner. In terms of methodology, the CJEU established that ascertaining the binding nature of EU provisions must consider the context in which the EU act was adopted and the powers of the institution which adopted it.Footnote 64 On the basis of this, the Court set out two premises. First, the function of the CVM reports and recommendations is to interpret the benchmarks included in the CVM decision (that is, binding EU law).Footnote 65 Second, Member States are obliged to comply and bring their laws in conformity with EU law.Footnote 66 Proceeding from these two premises, the Court drew the conclusion that the recommendations are de facto binding: Romania not only ‘must take due account of’ the recommendations but ‘cannot adopt or maintain’ conflicting national legislation:

In those circumstances, in order to comply with the benchmarks set out in the Annex to Decision 2006/928, Romania must take due account of the requirements and recommendations formulated in the reports drawn up by the Commission under that decision. In particular, Romania cannot adopt or maintain measures in the areas covered by the benchmarks which could jeopardise the result prescribed by those requirements and recommendations. Where the Commission expresses doubts, in such a report, as to whether a national measure is compatible with one of the benchmarks, it is for Romania to cooperate in good faith with the Commission with a view to overcoming the difficulties encountered with regard to meeting the benchmarks, while at the same time fully complying with those benchmarks and the provisions of the Treaties.

… Romania is required to take the appropriate measures for the purposes of meeting those benchmarks, taking due account, under the principle of sincere cooperation laid down in Article 4(3) TEU, of the reports drawn up by the Commission on the basis of that decision, and in particular the recommendations made in those reports.Footnote 67

The Court's argument contained serious deficiencies. In terms of methodology, it adopted a questionable approach. The context of an expressly non-binding instrument can hardly make it binding. In terms of logic, the Court gave a textbook example of how unsound deduction works. A deduction is sound if the premises are true and the conclusion drawn from these is valid (ie the conclusion follows from the premises). A closer look at the CJEU's deduction shows that although the Court's premises were true, the conclusion was invalid, as it did not follow from the premises. The purpose of the Commission reports and recommendations is, indeed, to provide an interpretation of the CVM benchmarks and Member States must, indeed, comply with EU law. Nonetheless, the conclusion that the recommendations are de facto binding does not follow from these premises. The fact that the CVM reports and recommendations are meant to interpret the benchmarks does not imply that they are binding interpretations or that only the Commission has the right to interpret and Romania is bound by this. The fact that EU law is binding does not mean that a soft-law instrument interpreting these binding rules should also be binding.

Although the Court's judgment convincingly demonstrated that the reform package impaired the independence of the judiciary, it undermined its own conclusions by attributing de facto binding nature to the Commission's reports and recommendations.Footnote 68 Unfortunately, instead of reaching these substantive conclusions on its own, it twisted the question to the point where ‘non-binding’ was turned into ‘binding’ and ignored the conspicuous fact that recommendations are just that: recommendations. The fact that the policy considerations provide a more credible reconstruction of the decision than the legal considerations (or, to put it more strongly: the fact that the policy considerations suppressed the law) undermined the judgment's normative force, notwithstanding the cogency of its substantive conclusions.

The main problem with the Court's ruling, besides the fact that it may be viewed as applying policy instead of the law, is that its ‘legal juggling’ with the binding nature of recommendations clearly interferes with the division of competences between the Commission and the Member States and goes against the constitutional structure. Romania agreed to the CVM benchmarks but has never agreed to the Commission's paternalistic (or to put it more strongly: unilateral) application thereof. The CVM benchmarks represent an agreement between Romania and the EU (and the other Member States), and there is no indication in EU law that the Commission has any prerogative whatsoever to interpret them. If there is a dispute as to these benchmarks, the CJEU is supposed to settle it as a fair and unbiased arbiter. It is not supposed to subject Romania to an authority it has never accepted. Even the idea that some deference should be given to the Commission's assessment presupposes a hierarchical relationship between the Commission and Romania, where Romania has to bow before the Commission's interpretation of the agreement it concluded as an equal.

This is not the first case in the CJEU's judicial practice where a non-legislative act has been considered binding. In Fishery Agreement,Footnote 69 the Court found an inter-institutional agreement between the Commission and the Council to be binding on the parties, as ‘the two institutions [clearly] intended to enter into a binding commitment towards each other’.Footnote 70 Asociaţia ‘Forumul Judecătorilor din România’ was, however, the first case where a non-binding instrument was imposed on a Member State that had not agreed to it.Footnote 71

While it is true that, on rare occasions, the CJEU has been willing to review the legality of various non-legislative acts under Article 263 of the TFEU on account of them being ‘intended to have legal effects’, it is not clear whether the Court also attributes a binding effect to these measures. A soft-law instrument may have ‘legal effects’ in terms of persuasive authority without being binding. In the European Agreement on Road Transport (ERTA) case,Footnote 72 the Commission sought the annulment of Council proceedings regarding the negotiation and conclusion of an international treaty, whose subject matter came under EU competence. The Court found that the proceedings were ‘intended to have legal effects’ and thus reviewable.Footnote 73 In Public Undertakings Communication,Footnote 74 the Court annulled a Commission Communication on the application of Article 5(2) of Directive 80/723 on the transparency of financial relations between Member States and public undertakings. The Court found that the Communication added new obligations to those provided for by Article 5(2) of the Directive and was ‘intended to have legal effects of its own distinct from’ that provision.Footnote 75 In Pension Funds Communication,Footnote 76 the CJEU annulled a Commission Communication, as it ‘constitute[d] an act intended to have legal effects of its own, distinct from those already provided for by the [EC] Treaty’.Footnote 77 The Court annulled the Communication for lack of competence, but it is doubtful that the Court generally considered these Communications (or Commission communications in general) to be binding on the Member States. Contrary to these cases where the CJEU was inclined to review a Commission soft-law instrument for its persuasive authority, in Asociaţia ‘Forumul Judecătorilor din România’ it attributed de facto binding authority to it.

The reaction of the Curtea Constituțională to the CJEU's ruling was quick and devastating:

The CVM reports, drawn up on the basis of Decision 2006/928, by their content and effects, as established by the judgment of the CJEU of 18 May 2021, do not constitute rules of EU law, which the court should apply as a matter of priority, removing the national rule. Therefore, the national judge cannot be asked to decide that recommendations should be applied as a matter of priority to the detriment of national law, since the CVM reports do not establish legal rules and are therefore not likely to conflict with national law. This conclusion is all the truer where the national legislation has been declared to be in conformity with the Constitution by the national constitutional court in the light of the provisions of Article 148 of the Constitution.Footnote 78

It held that EU law has no supremacy over the Constitution of Romania and those Romanian laws that are pronounced constitutional.Footnote 79 The Curtea Constituțională created a wide exception to the direct effect and supremacy of EU law, but a closer look reveals that its reaction was provoked by the ultra vires character of the CJEU ruling. It held that the CVM reports and recommendations are not for national courts to apply:Footnote 80 first, because they are not binding; and, second, because they are not justiciable and, hence, national courts cannot be expected to put aside allegedly conflicting national legislation.

The Curtea Constituțională did not question Romania's obligation to comply with the benchmarks and cooperate with the Commission in this regard, but it questioned the direct effect of these very general requirements and, reading between the lines, may have also questioned whether the cooperation between the Commission and Romania is asymmetric: that is, Romania is not only subjected to the mutually accepted CVM benchmarks but also to the Commission's interpretation of these benchmarks. The Curtea Constituțională put this in such a way as if it were re-interpreting the CJEU's ruling; however, in reality it rejected it. The CJEU declared that both the CVM Decision and the CVM reports and recommendations had direct effect and supremacy. The Curtea Constituțională held that the benchmarks lack this character and compliance with them is a matter strictly between the Commission and Romania:

The Court finds that the CJEU, in declaring Decision 2006/928 to be binding, has limited its effects from a twofold perspective: on the one hand, it has established that the obligations resulting from the Decision are a matter for the Romanian authorities competent to cooperate institutionally with the European Commission (paragraph 177 of the judgment), and thus for the political institutions, the Romanian Parliament and the Government of Romania, and, secondly, that the obligations are to be exercised in accordance with the principle of sincere cooperation laid down in Article 4 TEU. From both perspectives, the obligations cannot be binding on the courts, i.e. State bodies which are not empowered to collaborate with a political institution of the European Union.Footnote 81

Although the reaction of the Curtea Constituțională may have been excessive, in particular the statements that the Constitution and national legislation pronounced constitutional are immune from the supremacy of EU law, it is clear that it was provoked by the CJEU's controversial ‘legal juggling’ where non-binding recommendations were miraculously turned into binding rules and that concerned the core of national sovereignty.

Contrary to the PSPP saga, the Romanian case has not faded away quietly. In PM and others Footnote 82 the CJEU expressly reiteratedFootnote 83 the tenets laid down in Asociaţia ‘Forumul Judecătorilor din România’ and insisted that national courts comply with them and, in defiance of the judgment of the Curtea Constituțională, accord the CVM report and recommendations direct effect and de facto supremacy.Footnote 84

B. Union Interest v National Human Rights: The ‘Solange’ Power

There has been no documented case where a national constitutional court has refused to honour a CJEU judgment for breaching nationally protected fundamental rights. The ‘Solange’ compromise has worked effectively, resulting in two distinct and parallel fundamental rights regimes. On the one hand, national constitutional courts deactivated their constitutional review powers, but reserved the right to use them, if necessary. As noted by the Bundesverfassungsgericht in the Right to be Forgotten II case,Footnote 85 ‘[i]t cannot be assumed that the Charter … corresponds with the Grundgesetz and is congruent with its guarantees in all details’,Footnote 86 but this is not required. What is required is that the protection in EU law is essentially comparable, and the current state of EU law satisfies this requirement.Footnote 87 On the other hand, the CJEU denied the national constitutional review powers but applied the greatest common denominator of the constitutional traditions common to the Member States to ensure that constitutional courts do not have to choose between the supremacy of EU law and their own constitutional values.

In Taricco, however, this balance was almost disrupted when the CJEU expected Italy to prosecute criminal offences even after the expiry of the Italian statute of limitations.Footnote 88 This case does not feature on the list of defiant constitutional court decisions because the Court bowed before the covert threat of the Corte costituzionale and revised its ruling.

The Taricco caseFootnote 89 arose when the Corte costituzionale found that the limitation period under Italian criminal law was so unreasonably short that it interfered with the effective criminal enforcement of the EU value added tax (VAT) rules and referred the case to the CJEU. The source of the problem was that although various measures, such as the order fixing the preliminary hearing, interrupted the running of the limitation period, according to Article 161 of the Italian Penal Code that interruption could not extend the limitation period by ‘more than one quarter of the maximum prescribed period’.Footnote 90 Because of the complex and time-consuming investigations required, the rules on limitation created a situation of de facto impunity for tax fraud.Footnote 91 The Court established that the principle of effectiveness required Italian courts, at least in VAT cases, to disapply this Italian rule retroactively and to enforce the criminal rules even beyond the extension of a quarter of the maximum prescribed period, that is, even if the statute of limitation had expired. The Court found that the retroactive disapplication of Article 161 of the Italian Penal Code did not contravene the EU Charter of Fundamental Rights, specifically the principles of legality and proportionality of criminal offences enshrined in Article 49. It pointed out that ‘the sole effect of the disapplication of the national provisions at issue would be to not shorten the general limitation period in the context of pending criminal proceedings’.Footnote 92

The retroactive disapplication and the idea of reopening time-barred criminal proceedings apparently perplexed the Corte costituzionale, which, a year later, re-submitted the question in M.A.S. and M.B. (also referred to as Taricco II).Footnote 93 In this case the Corte costituzionale noted that in Taricco the CJEU only examined the question of retroactivity under Article 49 of the Charter, and overlooked the requirements of legislative clarity and foreseeability. The Corte costituzionale made it clear, as far as judicial conventions permitted, that it considered these requirements to be part of the general principles of law and thus that Taricco contravened EU law.Footnote 94 The CJEU backed down and established that the national court could still refuse to disapply Article 161 of the Italian Penal Code, if it considers the disapplication to breach the requirements of clarity and foreseeability.Footnote 95 The operative part of the ruling reiterated Taricco and added an exception to it:

unless that disapplication entails a breach of the principle that offences and penalties must be defined by law because of the lack of precision of the applicable law or because of the retroactive application of legislation imposing conditions of criminal liability stricter than those in force at the time the infringement was committed.Footnote 96

This was not the first instance of the CJEU employing a lower standard than the national constitution and suppressing a nationally recognized fundamental right. In Melloni,Footnote 97 the CJEU held that a Member State executing a European Arrest Warrant cannot, with reference to its constitution, make the surrender of a person convicted in absentia conditional upon the conviction being open to review in the issuing Member State. The higher level of rights protection was forced to cede to the EU's interest in mutual recognition and cooperation in criminal matters. In this case, however, contrary to the amorphous principle of effectiveness, the EU law requirement was explicit; furthermore, the Tribunal Constitucional did not consider the suppression of the nationally recognized fundamental to be untenable. However, Melloni triggered subsequent national court responses that led the CJEU, again, to backtrack at least slightly on the strictures of Melloni.Footnote 98 In Taricco and M.A.S. and M.B., it was clear to the CJEU that the suppression of the national fundamental right might provoke the re-activation of the review powers of the Corte costituzionale and result in actual defiance of the CJEU's ruling.

C. Enforcing EU Rule-of-Law Requirements on the Member States

The ongoing rule-of-law debate has proved to be a stress test for EU law. EU institutions have experimented with the penumbra of their limited powers to protect the rule of law in the Member States. This resulted in a number of controversial CJEU judgments and a good deal of criticism and accusations of ultra vires. The root cause of this line of case law is an internal contradiction in the EU's constitutional architecture. As explained in the following paragraphs, the EU law mechanism that was created to protect the rule of law in the Member States is not workable, while the EU law mechanism that is feasible was not created to protect the rule of law in the Member States (but to protect it in respect of EU action).

1. The diagonal application of the EU rule of law

The EU rule of law and human rights (referred to as ‘rule of law’) may be applied to the Member States through two distinct methods.

The application to the Member States (termed ‘diagonal’ in this article)Footnote 99 is explicit but ineffective. Articles 2 and 7 of the TEU set out substantive requirements and a procedure specifically with the purpose of ensuring that the Member States comply with the EU's founding values. Nonetheless, this strand of the EU rule of law is rudimentary. Article 2 of the TEU announces that human rights (and other fundamental values) have to be respected, but fails to specify this requirement in any meaningful way and attaches no effective enforcement mechanism.Footnote 100 Although the author of this article proposed a limited application of the Charter through Article 2 of the TEU to the Member States by way of incorporation,Footnote 101 this provision has never been applied as an independent legal basis.Footnote 102 Article 7 of the TEU provides for the suspension of membership rights in response to a systematic violation of Article 2 of the TEU but the requirement of unanimity stifles the mechanism's political feasibility.Footnote 103 This strand of EU human rights is not the result of organic development and reflects political tokenism. Article 7 was introduced by the Treaty of Amsterdam in 1999Footnote 104 (and reformulated by the Treaty of Nice in 2001 and the Treaty of Lisbon in 2007).Footnote 105 However, Article 7 of the TEU has never been successfully appliedFootnote 106 and creates the impression that it was set up with the intention that it would never be applied.

The application to the Member States acting as the EU's agents (termed ‘horizontal’ in this article) relies on the most effective procedural toolkit that EU law has, but, at least theoretically, does not aim to ensure that Member States comply with the EU rule of law in general. This strand emerged very early from, and has been accessory to, the application of EU human rights to the EU. The initial function of EU human rights was not to enforce rule-of-law requirements against the Member States but rather to ensure that the EU exercises its powers in conformity with human rights requirements.Footnote 107 These requirements were initially read into EU law as ‘general principles of law’ and were turned into the Charter in 2000.Footnote 108 They were only subsequently extended to the Member States when acting as the EU's ‘agents’Footnote 109 and their actions were thus attributable to the EU.Footnote 110

2. Lockstep and spill-over effects

Although this strand of the EU rule of law does not apply to the Member States when acting in domestic matters, it generates significant spill-over effects by reason of the EU's idiosyncratic institutional architecture. The EU is a legal giant but an institutional dwarf. National authorities and courts apply EU law more often than EU institutions, along with provisions of national law. For instance, customs, unfair commercial practices and competition law have been federalized but are predominantly applied by national authorities; and VAT law is a blend of EU and national provisions. This gives an opportunity for the EU to widen the horizontal scope of the Charter in a way that also influences Member State action in domestic matters. If, as is often the case, it is not possible to divide the issue into EU and national law, the Charter's horizontal scope will cover the whole matter, including the national law elements. By way of a metaphor, the relationship between EU and national law is not like oil and water, which build a laminar structure, but like a marble cake, where the two batters mix but do not fuse. In most matters, the applicable rules are made up of a blend of EU and national provisions. Although EU and national norms can be identified and distinguished from each other, in most cases they make up the applicable rules jointly. Legal harmonization makes EU law pervasive and quite often applicable in some abstract sense, even if its role in the matter is marginal.

For instance, if a Member State establishes a general tax offence, this may come under the scope of the Charter, even though EU legislation is limited to VAT (sales tax). This happened, for instance, in Åkerberg Fransson.Footnote 111 Sweden established a general tax offence and the CJEU found the Charter applicable to its VAT strand. It held that this came under the scope of EU law and could not be separated from the other strands, and thus the Charter applied to the tax offence as a whole, including non-VAT violations.

The above dual character explains why the Conditionality RegulationFootnote 112 simultaneously has both a limited scope and also spill-over effects outside its scope. On the one hand, the Regulation protects the EU's founding values but only in the context of the EU's financial interests, if the breach of the rule of law endangers ‘the Union budget’.Footnote 113 It does not deal with the protection of national financial interests concerning the spending of national funds and, a breach that merely endangers the national budget falls outside its scope. On the other hand, however, the Regulation has significant spill-over effects, given that financial abuses are generally criminalized and are usually investigated by the same authorities and adjudicated by the same courts irrespective of whether they affect the national or the Union budget.Footnote 114 The EU's protection of the rule of law in the context of EU funds may also ensure a general compliance with these rule-of-law standards.

A similar lockstep effect can be perceived in relation to judicial independence. Although this concerns another line of case law (under Article 19 of the TEU),Footnote 115 the EU requirement of judicial independence has been based on the same horizontal agency rationale and has given rise to the same spill-over effects. EU law requires Member State courts to be independent when applying EU law and, with the exception of Article 2 of the TEU, does not specify any requirements for their application of national law. Nonetheless, the EU law requirement spills over since the same courts apply both EU and national law which often cannot be separated. As a corollary, the requirement of independence set out in respect to the application of EU law equally applies to national courts when applying national law, thus creating a general requirement of judicial independence.

This spill-over effect gave rise to the Trybunał Konstytucyjny's judgment of 7 October 2021 in Case K 3/21. The Trybunał Konstytucyjny reacted to a series of CJEU judgments which provided for the disapplication of various provisions of Polish law for endangering the independence of the judiciary.Footnote 116 Although, at first glance, the decision may appear to question the supremacy of particular rules of EU law, on closer examination it reveals that it was not the rules but their interpretation by the CJEU that was pronounced unconstitutional and inapplicable. It must be noted that the decision was based on an apparent misconception about EU competences. The Trybunał Konstytucyjny claimed that the division of powers between the EU and the Member States is ‘structural’ in the sense that there are some fields, such as the organization of the judiciary, that are reserved for the Member States. In reality, however, the founding treaties make it clear that the division of powers is ‘functional’, and irrespective of how powers are divided, Member States committed to ensuring the effective application of EU law by their judiciary, and this implies the requirement of independence. The spill-over of such independence into Polish domestic matters is an inevitable consequence.

The judgment of the Trybunał Konstytucyjny was an irascible reaction to spill-over effects which touched on the heart of national sovereignty in a case where the CJEU's construction of EU law was solid. However, some of the Court's recent case law, presented in the next section, arguably stretches EU rule-of-law standards beyond the scope envisaged by its constitutional foundations. These rulings provoked no open defiance, presumably because they do not directly intrude upon the core of national sovereignty, but they contribute to the current reputation of the CJEU.

3. The Charter's ever widening diagonal scope

The CJEU's increasingly wide conception of ‘agency’ as a trigger for the Charter's (and more generally the EU rule of law's) application to the Member States has given rise to rulings with questionable normative foundations, which have arguably exceeded the constitutional intention of the pouvoir constituant and could be criticized as competence creep. Unlike with the cases presented in Sections IV.A.1 and 2, this line of case law has not given rise to defiant constitutional court judgments, but arguably may have contributed to the general perception of a jurisprudence with a declining normative and increasing policy character.

The CJEU has laid down in several cases that the Charter applies to the Member States only if the matter has a ‘degree of connection’ to EU law, which exists if EU law imposes a ‘specific obligation’ on the Member States. A weak and remote connection to EU law, which is manifested by some marginal general requirement without any specific or detailed obligation, does not justify the application of the Charter. A contrary approach would overstretch the scope of the Charter and make it generally applicable, since Member States could always be regarded as pursuing the aims listed in Article 3 TEU or transforming the values listed in Article 2 TEU into reality. Nonetheless, the Court has been ready to apply the Charter in a number of cases where there was no ‘specific obligation’ set out in EU law.

One of the extreme examples is Florescu.Footnote 117 Romania adopted austerity measures with a view to meeting the targets of an EU financial assistance programme. The measures to be adopted by Romania were not specified—EU law merely provided that budgetary cuts should be made, amongst other methods, by reducing the public sector wage bill and reforming the key parameters of the pension system. To meet these targets, Romania adopted a number of measures, including a provision that restricted the net pension from being combined with income from activities carried out at public institutions. The plaintiffs were retired judges who either had to suspend their pensions or terminate their university teaching positions.

The CJEU found that the austerity measures were covered by the Charter, although the only link to EU law was Romania's commitment to cut expenses by reducing the public sector wage bill and reforming the pension system.Footnote 118 Surprisingly, the Court did not criticize the lack of specificity in the EU law framing of the national measures and described these commitments as ‘leav[ing] Romania some discretion’,Footnote 119 and thus it found that Romania ‘adopt[ed] measures in the exercise of the discretion conferred upon it by an act of EU law’.Footnote 120 The Court concluded that the requirement to reduce the public sector wage bill and reform the pension system was, in itself, ‘sufficiently detailed and precise’ to bring the above national measure within the scope of the Charter.Footnote 121

The earlier case law on minimum harmonization directives also showcases an expansive interpretation of the scope of the Charter. It is the very premise of minimum harmonization directives that while Member States are obliged to adopt the required minimum standards, beyond that their original regulatory power remains untouched and, as a corollary, they are free to decide whether or not to introduce higher standards. Despite this, for many years, the CJEU interpreted minimum harmonization clauses as if the Member States were not conferring part of their regulatory power to create harmonization but rather as if the EU had granted them the authority to regulate nationally. Although this was overruled in Hernández,Footnote 122 Spetsializiran nakazatelen sad Footnote 123 and, finally, in TSN,Footnote 124 it still represents an inexplicable line of case law.

In N.S.,Footnote 125 the CJEU compelled the United Kingdom (UK) to process an asylum application that it was not obliged to process, because the responsible Member State (Greece) was expected not to fulfil its duties under the now repealed Dublin II Regulation.Footnote 126 Under this Regulation, Member States were free to examine applications for asylum, even if they did not qualify as the Member State responsible. The question was whether the Charter governed a Member State's discretion to take over an application instead of directing it to the Member State responsible. The Court held that the Charter applied,Footnote 127 and while the Regulation itself did not limit this discretion, the Charter did. It argued that the freedom to process an asylum application was conferred on the Member States by way of re-delegating the power that was conferred on the EUFootnote 128 and held that the Charter prevented the transfer of an asylum seeker, if there were ‘substantial grounds for believing that the asylum seeker would face a real risk of being subjected to inhuman or degrading treatment within the meaning of that provision’.Footnote 129

The Court's conceptualization was quite odd. In fact, policy considerations give a more plausible explanation than normative ones. The EU did not delegate the power to process asylum applications to the Member States, but the Member States delegated a regulatory power to the EU. Article 3(2) of the Regulation was an exception to this delegation and confirmed that the Member States retained their power to process asylum applications if they wished. Furthermore, according to the CJEU's case law, the Charter applies only if EU law imposes a ‘specific obligation’ on the Member States. Article 3(2) of the Regulation set out no specific obligation (in fact, no obligation at all). A more convincing explanation for the ruling is that the opposite decision would have admitted the serious defects of the system, and the fact that the EU cannot, in a field almost entirely regulated by EU law, ensure that responsible Member States treat asylum seekers in compliance with the Charter. The Court shifted the duty of the responsible Member State to all the other Member States because it could not ensure that the former would fulfil this duty.

The ruling was similarly alarming if approached from the angle of the principle of conferral. The adoption of the Dublin II Regulation required unanimity in the Council.Footnote 130 This means that the Member States conferred this legislative power on the EU with the reservation of a veto right, and when they adopted the Regulation unanimously, they clearly allocated the relevant duties. The Court's ruling interfered with this constitutional allocation by failing to reproduce and respect that constitutional understanding.Footnote 131

The EU institutions have, at times, used tortuous reasoning to apply supportive ‘side-effects’Footnote 132 of apparently unconnected EU norms to protect the rule of law in cases where the EU had no other power to intervene.Footnote 133 The use of such side-effects is, in itself, not reprehensible, as long as it is balanced and the normative construction of the law is not replaced with policy considerations.Footnote 134 However, if over-used or the interpretation of the law is hijacked by end-driven solutions, it seriously damages the CJEU's legitimacy and authority as a judicial institution, as it should operate under ‘normative constraints’,Footnote 135 and may result in divisive political debates.Footnote 136

A notable example of the above is Commission v Hungary,Footnote 137 where the prohibition of discrimination based on age was used to protect the independence of the judiciary. In this case, Hungary reduced the mandatory retirement age for judges, public prosecutors and public notaries from 70 to 62 years. This resulted in the retirement of almost 300 judges and public prosecutors,Footnote 138 creating a potential threat to the independence of the judiciary, given that the mass recruitment of judges which followed might have given the government an opportunity to influence the composition of courts. This concern was reinforced by the fact that a large number of the judges affected by mandatory retirement were senior high court judges and supreme court justices. While the Commission was reluctant to base its claim on judicial independence, it successfully attacked the Hungarian provisions before the CJEU on the basis that they fell foul of the principle of equal treatment (as embedded in Directive 2000/78/EC), which prohibited discrimination at the workplace, inter alia, on grounds of age. The Commission doubted whether it had the power to address the primary issue directly, so it relied, successfully, on the EU prohibition of discrimination based on age. Tellingly, although the Commission's legal arguments were wrapped up in anti-discrimination law, its press release on the infringement procedure makes it perfectly clear that the problem addressed was the independence of the judiciary.Footnote 139 Although this convoluted argument appears to have been unnecessary in light of the CJEU's ruling in Portuguese judges,Footnote 140 where the Court held that the independence of national courts is protected by Article 19 of the TEU, at the relevant time the interpretation of this provision was uncertain.

In Commission v Hungary (Central European University),Footnote 141 the CJEU both made use of the supportive side-effects of an international treaty concluded by the EU and provided an arguably end-driven interpretation of the Charter's scope. The case was presented as a trade dispute, but was, in fact, about academic freedom.

In this case, Hungary imposed novel requirements on non-EU universities, which apparently aimed to expel the Central European University (CEU) from the country and thus breached academic freedom. As the CEU was a US-registered higher education institution, the rules of the internal market did not apply. The CJEU established the applicability of the Charter by means of the General Agreement on Trade in Services (GATS): it held that the GATS was EU law and as the Hungarian law breached the GATS, the Charter applied. This was the first case where the CJEU applied World Trade Organization (WTO) law not just as a tool of interpretation, but rather as purely internal EU law. At the same time, the Court limited standing to infringement procedures launched by the Commission. Unfortunately, the Court was laconic and provided only a two-sentence explanation as to why the limited invocability of the GATSFootnote 142 entailed the application of the Charter:

[T]he GATS forms part of EU law. It follows that, when the Member States are performing their obligations under that agreement, including the obligation imposed in Article XVII(1) thereof, they must be considered to be implementing EU law, within the meaning of Article 51(1) of the Charter.Footnote 143

The Court's approach raised serious issues and gave rise to the suspicion that it was influenced by a desire to protect academic freedom. First, the Court's jurisprudence has maintained that, for the Charter to apply, the mere application of an EU norm is not sufficient, but there needs to be a ‘degree of connection’ and the EU norm needs to be ‘specific’. This requirement was not met here, and the Court did not even claim a ‘degree of connection’ to exist. Second, the rationale of the apparent diagonal application (supremacy of EU law, as well as uniformity and efficacy) did not warrant the application of the Charter, the relevant question being whether a Member State complies with the international obligations assumed by the EU. Third, the CJEU even shifted the balance of mutual treaty concessions to the unilateral detriment of the EU.Footnote 144 By pronouncing the Charter applicable, the CJEU subjected Member States to extra burdens, which were neither provided for, nor contemplated by WTO law and which are not borne by the other members of the WTO. Finally, the application of the Charter was redundant for the practical outcome of the case: the Hungarian law was pronounced inapplicable through the application of the GATS. Nevertheless, the Court still applied the Charter and condemned the law by virtue of this additional legal basis.

The judgment is a good example of how the Court may overstretch the scope of the Charter. For a critical observer, this may suggest that ‘if there is a will, there is a way’ to have the Charter applied, even beyond the constitutional intentions or even the jurisprudence of the CJEU itself.

The concern over the CJEU's application to the Member States of the EU rule-of-law requirements, most notably the Charter, is demonstrated by the debate it sparked in the context of the Lisbon Treaty. Poland and the UK made a reservation in the form of Protocol No 30 attached to the TFEU. This provides that ‘[i]n particular, and for the avoidance of doubt, nothing in Title IV [Solidarity] of the Charter creates justiciable rights applicable to Poland or the United Kingdom except in so far as Poland or the United Kingdom has provided for such rights in its national law’.Footnote 145 Czech President Václav Klaus refused to finalize the ratification of the Lisbon Treaty and insisted on the extension of Protocol No 30 to the Czech Republic out of fear of the potential application of the Charter in challenging the Benes decrees, which provided for the expulsion of ethnic Germans and for the confiscation of their property after World War II. The Czech government requested the extension of Protocol No 30,Footnote 146 but, after the change of the political leadership, withdrew that request the following year.Footnote 147 The Polish and Czech objections may appear redundant. The purpose of the Charter is not to protect fundamental rights in the Member States in general but to ensure that fundamental rights are respected in the actions attributable to the EU. The Charter applies to the Member States only when they are implementing EU law.Footnote 148 Still, these objections powerfully express the Member States’ bitter experience that the CJEU's application of EU law is not always in conformity with what seems clearly to have been constitutionally contemplated and seeks to apply the Charter by any means possible.

V. CONCLUSIONS

A superficial glimpse may suggest that the rebellion of national constitutional courts questions the supremacy of EU law. A closer look reveals, however, that the heart of the matter is, in fact, the CJEU's interpretive primacy over the ultimate boundaries of EU law and competences and, as such, a Kompetenz–Kompetenz issue. The primacy of EU law is a cornerstone of the EU and has not been called into question. National courts have, however, questioned the omnipotence and interpretive primacy of the CJEU.

A mixed chamber, consisting of CJEU and national constitutional court judges and functioning as a super-court, has been proposed to ‘watch the watchers’Footnote 149 and solve the most serious debates of competence.Footnote 150 Such a mixed chamber may ease accusations of competence creep and ultra vires, and obviate defiant constitutional court judgments. It could abate legal pluralism by institutionalizing (internalizing) it. Factoring national judiciaries into the decision-making process might mean that rather than the constitutional authorization theory presenting challenges to decisions, there could instead be agreement and the chance to adopt a uniform approach. Furthermore, and equally importantly, procedure is not only about maximizing the chance of good decisions but also about legitimacy. The adoption of a procedure that visualizes and addresses the relevant questions increases the social recognition and acceptance of the final decision. It is questionable, however, whether institutional pride will allow such a reform to be made.

The rebellion may have caught the European discourse off guard.Footnote 151 The last few decades may have given the impression that the CJEU's omnipotent concept of primacy, including its theoretical underpinnings, was beyond dispute. Indeed, there was no reason to be wary of the CJEU's jurisprudence: national constitutional courts accepted the primacy of EU law, albeit via a different conceptualization, and their reservations appeared to lack clout. Until recently, constitutional courts had rarely used these reservations to refuse to apply EU law or comply with a CJEU ruling.Footnote 152 This may have led observers to overlook the fact that the CJEU's omnipotent version of primacy coexists with a traditionalist and statist conceptualization, and to view national conceptualizations as a heresy with limited practical significance. After all, it does not matter why EU law has supremacy, as long as it has. However, it has become apparent that while this fiction is effective in times of peace, it cannot survive a stress test.

The CJEU has been a pioneer of European integration by interpreting EU competences in an expansive manner. As a justificatory narrative it is often claimed that disagreements among the Member States are overcome by ambiguous statutory language and the Court is expected to interpret these provisions without the assistance of an ascertainable constitutional or legislative intent. Although this may be true in some cases, it is also true that in these cases the benefit of the doubt has arguably been given to integration and not the Member States. As an alternative narrative, the trajectory of the CJEU case law may be viewed as a subtle and piecemeal process of gradually increasing integration by interpretation where the Member States have been reluctant to do so via legislation.Footnote 153 The key to the success of this approach was that the CJEU's expansive interpretation only exceeded what had been constitutionally contemplated by the Member States by a tolerable amount.Footnote 154 This was, however, a very delicate process, and whilst national courts may raise an eyebrow to some CJEU interpretation, there is a point at which they may consider it to have gone too far and refuse to implement the judgment. Unsurprisingly, Euro-realists have characterized the CJEU's case law as always finding a way in cases of importance for integration.

This approach is not alien to courts more broadly, and the debate, in a certain sense, parallels the US debate concerning constitutional interpretation. Originalists argue that courts are expected to ascertain the original meaning of the US Constitution,Footnote 155 while the theory of the living constitution argues for an evolutionary interpretation,Footnote 156 which is, at times, detached from the original constitutional intent.Footnote 157 An originalist approach results in a narrow interpretation, which gives a wider scope to democratic choice both at the federal and state level, whilst the extensive interpretation of the living constitution limits that scope more significantly.Footnote 158 Putting philosophical questions aside, from a practical perspective, evolutive interpretation can also be conceived as a question of degree, also involving a margin of tolerance.

Ultimately, interpretive primacy is a sociological question. National courts have accorded interpretive priority to the CJEU, and the rebellion of constitutional courts does not reverse that process. However, it makes it apparent that the CJEU has not been accorded indisputable interpretive primacy and the incontestable final word.Footnote 159 In the early days, the CJEU's judicial practice was infused with the consideration that it was not sufficient for the Court to make good judgments, it also needed to convince national courts, reflecting the horizontal nature of the relationship. Since then, the power to enforce CJEU judgments has been improved,Footnote 160 and the Court has adopted a more authoritative tone, which has, at times, relied more on prerogative than on persuasion.

The rebellion of the constitutional courts suggests that the CJEU might have gone too far in imposing its will and should return to strict normativity,Footnote 161 avoiding decisions where a policy reconstruction sounds more plausible than the legal–dogmatic reconstruction. This predicament should not be portrayed as a debate between the friends and the enemies of European integration. The debate is much wider than that and rendering it in such simple terms risks not being able to identify and solve the problem. The idea that the Court seeks to establish justice (whatever ‘justice’ may mean) rather than the law is harmful for European integration. The ‘law’ has played a central role in the European project, and it should not be forgotten that the EU is an integration through law and not law through integration.

Of course, the problem is a question of degree. Member States have amended the founding treaties on many occasions, and they have never explicitly overruled the CJEU's case law on the foundations of the EU legal order.Footnote 162 This may suggest acquiescence to the CJEU's interpretative approach. However, the reservations made to the Charter suggest significant discontent, which could have been obviated if the Court had adhered to an originalist interpretation.

Normativity is the only source of judicial power. As Justice Scalia put it: ‘[t]he judge who always likes the results he reaches is a bad judge’.Footnote 163 Of course, one may name examples where the CJEU was less integrationist. For instance, in Hungary v Slovakia,Footnote 164 the CJEU was ready to curtail free movement and read a non-existent exception to the free movement of persons into the TFEU to cater to the sovereigntist sensitivity of the Member States. But here, again, the judgment's policy reconstruction sounds more plausible than its legal–dogmatic reconstruction. Justice in an individual case may inflict far-reaching harm by impairing normativity and the CJEU's reputation as a fair arbiter. The primary function of the CJEU is not to further integration but rather faithfully to reproduce the intentions of the pouvoir constituant.

Acknowledgements

The research was supported by the ICT and Societal Challenges Competence Centre of the Humanities and Social Sciences Cluster of the Centre of Excellence for Interdisciplinary Research, Development and Innovation of the University of Szeged. The author is a member of the ‘Law and Competitiveness’ research group. The author is indebted to Samantha Cheesman, András Jakab, János Martonyi, Daniel Sarmiento, Niamh Nic Shuibhne, Ernő Várnay and the anonymous reviewers for their comments. Of course, all views and any errors remain the author's own.

References

1 The two exceptions were the decision of the Czech Constitutional Court (CCC; Ústavní soud) in Holubec, Plenary Judgment of 31 January 2012 of the CCC, PL. ÚS 5/12, following the CJEU's judgment in Case C-399/09 Landtova ECLI:EU:C:2011:415, and the Danish Supreme Court's judgment in Rasmussen, Case 441/14 Dansk Industri v Rasmussen ECLI:EU:C:2016:278; Case No 15/2014 Dansk Industri (DI) acting for Ajos A/S v The Estate Left by A. Holubec was the first case in the history of the EU that a national court declared a CJEU ruling ultra vires. See Kaczorowska-Ireland, A, European Union Law (Routledge 2016) 297–9CrossRefGoogle Scholar. For an analysis of the case, see Bobek, M, ‘Landtová, Holubec, and the Problem of an Uncooperative Court: Implications for the Preliminary Ruling Procedure’ (2014) 10 EuConst 54Google Scholar; Komárek, J, ‘Czech Constitutional Court Playing with Matches: The Czech Constitutional Court Declares a Judgment of the Court of Justice of the EU Ultra Vires: Judgment of 31 January 2012, Pl. ÚS 5/12, Slovak Pensions XVII’ (2012) 8 EuConst 323, 330Google Scholar. In both cases, the national court considered the CJEU's interpretation unreasonable and refused to accept it. These cases were the precursors of the current rebellion but, probably because their subject matter did not present a significant challenge to sovereignty, they generated less attention.

2 According to a rule of Chekhovian dramaturgy (also referred to as ‘Chekhov's gun’), ‘[i]f in the first act you have hung a pistol on the wall, then in the following one it should be fired. Otherwise don't put it there.’ I Gurliand, ‘Reminiscences of A. P. Chekhov’ in Teatr i Iskusstvo, No 28, 11 July 1904, 521.

3 See Bundesverfassungsgericht (BVerfG), Judgment of the Second Senate of 5 May 2020 – 2 BvR 859/15; Trybunał Konstytucyjny, Judgment of 7 October 2021, Case K 3/21; Curtea Constituțională, Decision No 390 of 8 June 2021.

4 cf Scaccia, G, ‘The Lesson Learned from the Taricco Saga: Judicial Nationalism and the Constitutional Review of E.U. Law’ (2020) 35 AmUIntlLRev 821Google Scholar, presenting the general tendency of national constitutional courts ‘to claim a more intrusive review on E.U. law’.

5 cf Blutman, L, ‘Az uniós jog elsőbbsége: alkotmánybíróságok lázadása’ (2022) 15(1) Közjogi Szemle 1–10Google Scholar, stating that the defiant national judgments represent the emerging age of ‘practical sovereignty protection’.

6 For an overview of the CJEU's case law, see Vincze, A, ‘Zum Bedeutungswandel des Art. 19 EUV in der Rechtsprechung des EuGH’ (2023) 62(1) Staat 71–89Google Scholar.

7 Case 26/62 van Gend & Loos ECLI:EU:C:1963:1; Case 6/64 Costa v ENEL ECLI:EU:C:1964:66; Case 11/70 Internationale Handelsgesellschaft ECLI:EU:C:1970:114.

8 For general theories of federalism and their application to the EU, see Schönberger, C, ‘Die Europäische Union als Bund: Zugleich ein Beitrag zur Verabschiedung des Staatenbund-Bundesstaat-Schemas’ (2004) 129(1) AoR 81Google Scholar; Schönberger, C, Unionsbürger: Europas föderale Bürgerrecht in vergleichender Sicht (Mohr Siebeck 2005)Google Scholar; Beaud, O, Théorie de la Fédération (Presses Universitaires de Frances 2007)Google Scholar.

9 Bruggeman, R and Larik, J, ‘The Elusive Contours of Constitutional Identity: Taricco as a Missed Opportunity’ (2020) 35 UtrechtJIntlEurL 20, 24Google Scholar.

10 See Dougan, M, ‘The Treaty of Lisbon 2007: Winning Minds, not Hearts’ (2008) 45(3) CMLRev 617, 699Google Scholar.

11 Brussels European Council, 21–22 June 2007, Annex 1, para 4.

12 It has to be noted that even if accepted, art I-6 may not have obviated the circularity of the above argument. It could still be argued that the reference of art I-6 to ‘the law of the Member States’ did not embrace national constitutions. P Craig and G de Búrca, EU Law (6th edn, OUP 2015) 274. Furthermore, art I-6 referred to ‘law adopted by the institutions of the Union in exercising competences conferred on it’, which may be viewed as implying that ultra vires law is not law at all. Finally, art I-6 does not necessarily obviate the controversy about the CJEU's categorical interpretative primacy, which is not about the primacy of EU law but about the status of its ultra vires interpretation.

13 cf Bruggeman and Larik (n 9) 22–4, referring to the CJEU's conception as ‘absolute primacy’, while to national constitutional courts’ conception as ‘relative primacy’.

14 See, eg, Hungarian Constitution, art E; Romanian Constitution, art 148.

15 See SW Schill and A von Bogdandy, ‘Overcoming Absolute Primacy: Respect for National Identity under the Lisbon Treaty’ (2011) 48(5) CMLRev 1417; M Sulyok, ‘Nemzeti és alkotmányos identitás a nemzeti alkotmánybíróságok gyakorlatában’ in MA Jakó (ed), Nemzeti identitás és alkotmányos identitás az Európai Unió és a tagállamok viszonyában (Szeged 2014) 44–62; E Orbán, ‘Constitutional Identity in the Jurisprudence of the Court of Justice of the European Union’ (2022) 63(2) HungJLegStud 142; C Timmermans, ‘Mediating Conflicts between National Identities and EU Law: The Potential of Article 4(2) TEU’ (2022) 59 CMLRev 75.

16 For Belgium, see, eg, Belgian Constitutional Court's Judgment 62/2016 of 28 April 2016. For France, see, eg, Decision No 2006-543 DC of 30 November 2006; Decision No 2007-560 DC of 20 December 2007. For Germany, see, eg, BVerfG, Urteil des Zweiten Senats vom 30. Juni 2009 – 2 BvE 2/08 (BVerfGE 123, 267–437); BVerfG, Beschluss des Zweiten Senats vom 15. Dezember 2015 – 2 BvR 2735/14. For Hungary, see Decision 22/2016 (XII. 5.) and Decision X/477/2021 of the Hungarian Constitutional Court. For Italy, see, eg, Italian Constitutional Court (ICC), Judgment of 18 December 1973, Case 183/1973 Frontini ECLI:IT:COST:1973:183; ICC, Judgment of 5 June 1984, Case 170/1984 Granital ECLI:IT:COST:1984:170; ICC, Judgment of 13 April 1989, Case 232/1989 Fragd ECLI:IT:COST:1989:232; Order No 24/2017 of the ICC. Some constitutional courts went even further and denied the supremacy of EU law over the national constitution. See, eg, the judgment of the Lithuanian Constitutional Court of 14 March 2006 in Case No 17/02-24/02-06/03-22/04; Decision K 18/04 of the Constitutional Tribunal of Poland of 11 May 2005; Decision K 32/09 of the Constitutional Tribunal of Poland of 24 November 2010.

17 For Belgium, see, eg, Belgian Constitutional Court's Judgment 62/2016 of 28 April 2016. For Germany, see, eg, BVerfG, Entscheidung vom 29. Mai 1974 – 2 BvL 52/71 (BVerfGE 37, 271 – Solange I); BVerfG, Entscheidung vom 22. Oktober 1986 – 2 BvR 197/83 (BVerfGE 73, 339 – Solange II); BVerfG, Beschluss des Zweiten Senats vom 07. Juni 2000 – 2 BvL 1/97 (BVerfGE 102, 147). For Hungary, see Decision 22/2016 (XII. 5.) and Decision X/477/2021 of the Hungarian Constitutional Court. For Italy, see Frontini v Minister delle Finanze, Judgment of 27 December 1973, 18 Giur. Cost. I 2401.

18 For Denmark, see Case No 15/2014 Dansk Industri (DI) acting for Ajos A/S v The Estate Left by A. For Germany, see, eg, BVerfG, Beschluss des Zweiten Senats vom 15. April 2021 – 2 BvR 547/21 (BVerfGE 157, 332–94); BVerfG, Urteil des Zweiten Senats vom 06. Dezember 2022 – 2 BvR 547/21. For Hungary, see Decision 22/2016 (XII. 5.) of the Hungarian Constitutional Court. See N Chronowski, B Szentgáli-Tóth and A Vincze, ‘Decision 22/2016. (XII. 5.) AB – Constitutional Self-identity of Hungary’ in F Gárdos-Orosz and K Zakariás (eds), The Main Lines of the Jurisprudence of the Hungarian Constitutional Court (Nomos 2022) 441–58. In Decision X/477/2021, the Hungarian Constitutional Court supplemented the ultra vires doctrine with cases where the EU fails to exercise its non-exclusive competences in an effective manner. The Court held that, outside the purview of exclusive EU competence, Hungary is entitled to take the enforcement of the jointly exercised competences in its own hand in case the EU's actions feature ‘incomplete effectiveness’. For Poland, see, eg, Decision K 32/09 (n 16).

19 H Kelsen, Reine Rechtslehre (Franz Deuticke 1934) 90–106.

20 This term refers to cases where the court has to choose between two potentially applicable laws, but the two laws do not differ. See, eg, EF Scoles and P Hay, Conflict of Laws (2nd edn, West Group 1992) 17.

21 cf JR Crawford, Brownlie's Principles of Public International Law (8th edn, OUP 2019) 88–104; A Kaczorowska, Public International Law (Routledge 2010) 148–9; MN Shaw, International Law (CUP 2008) 138–79.

22 Costa v ENEL (n 7).

23 See KJ Alter, Establishing the Supremacy of European Law: The Making of an International Rule of Law in Europe (OUP 2003). For instance, in France it took until 1989 for all the three supreme judicial bodies to accept the supremacy of EU law. Alter ibid 124–81.

24 See GF Mancini, ‘A Constitution for Europe’ (1989) 26 CMLRev 595, 611 (‘Reading an unwritten Bill of Rights into Community law … was forced on the Court from the outside, by the German and, later, the Italian Constitutional Courts.’) See also T von Danwitz, ‘The Charter of Fundamental Rights of the European Union between Political Symbolism and Legal Realism’ (2001) 29 DenvJIntlL&Pol 289, 300–2; D Sarmiento, ‘Who's Afraid of the Charter? The Court of Justice, National Courts and the New Framework of Fundamental Rights Protection in Europe’ (2013) 50 CMLRev 1267, 1268–9.

25 ‘Letter of Thomas Jefferson to James Madison (Paris, Dec. 20, 1787)’ in MA Giunta, JD Hartgrove and M-JM Dowd (eds), The Emerging Nation: A Documentary History of the Foreign Relations of the United States Under the Articles of Confederation, 1780–1789 (National Historical Publications and Records Commission 1996) 679, 680.

26 For a comprehensive overview, see T Tridimas, The General Principles of EU Law (OUP 2006); and X Groussot, General Principles of Community Law (Europa Law Publishing 2006). The general principles of law had the same scope as the Charter. See P Eeckhout, ‘The EU Charter of Fundamental Rights and the Federal Question’ (2002) 39 CMLRev 945, 958–69.

27 Charter of Fundamental Rights of the European Union [2012] OJ C326/391. See J Dutheil de la Rochère, ‘Charter of Fundamental Rights and Beyond’ (2001) 4 CYELS 133, 136–7; C Franklin, ‘The Legal Status of the EU Charter of Fundamental Rights after the Treaty of Lisbon’ (2010) 15(2) TilburgLRev 137; T Tridimas, ‘Fundamental Rights, General Principles of EU Law, and the Charter’ (2014) 16 CYELS 361; M Dougan, ‘Judicial Review of Member State Action under the General Principles and the Charter: Defining the “Scope of Union Law”’ (2015) 52(5) CMLRev 1201, 1204–7.

28 Case 1/58 Stork ECLI:EU:C:1959:4. Here, the CJEU applied the European Coal and Steel Community (ECSC) Treaty.

29 Case 29/69 Stauder ECLI:EU:C:1969:57.

30 See T Jurczyk, ‘The Role of the Court of Justice of the European Union in Setting European Union Standards of Protection of Fundamental Rights’ in M Jabłoński, T Jurczyk and P Gutierrez (eds), Międzynarodowa ochrona praw człowieka – współczesne problemy na świecie (Faculty of Law, Administration and Economics of the University of Wrocław 2015) 141, 142–4.

31 Internationale Handelsgesellschaft (n 7).

32 See M Mahlmann, ‘1789 Renewed? Prospects of the Protection of Human Rights in Europe’ (2004) 11 CardozoJIntl&CompL 903, 905–9.

33 Solange I (n 17) 271ff.

34 Solange II (n 17) 339ff. See SA Bibas, ‘The European Court of Justice and the U.S. Supreme Court: Parallels in Fundamental Rights Jurisprudence’ (1992) 15 HastingsIntl&CompLRev 253, 260–7. Solange I (n 17) was a starting point of a long process hallmarked by a number of classical German constitutional court (GCC) judgments (such as Solange II ibid 339; Vielleicht, BVerfGE 52, 187; Maastricht, BVerfGE 89, 155). For an overview of the BVerfG's case law, see R Pracht, Residualkompetenzen des Bundesverfassungsgerichts: ultra vires, Solange II, Verfassungsidentität (Mohr Siebeck 2022).

35 JHH Weiler and NJS Lockhart, ‘“Taking Rights Seriously” Seriously: The European Court and its Fundamental Rights Jurisprudence – Part I’ (1995) 32 CMLRev 51, 63–4.

36 See D Denman, ‘The Charter of Fundamental Rights’ (2010) 4 EHRLR 349, 351; F Fontanelli, ‘Implementation of EU Law through Domestic Measures after Fransson: The Court of Justice Buys Time and “Non-Preclusion” Troubles Loom Large’ (2014) 39(5) ELR 682, 683.

37 Case 5/88 Wachauf ECLI:EU:C:1989:321.

38 A couple of years later, the CJEU made another important extension: in Elliniki Radiophonia Tiléorassi (ERT) it held that when restricting one of the four freedoms with reference to the local public interest, Member States are, in fact, applying EU law, which ‘must be interpreted in the light of the general principles of law and in particular of fundamental rights’. Case C-260/89 ERT EU:C:1991:254. This notion has been part of the case law since then. See, eg, Case C-390/12 Pfleger EU:C:2014:281, paras 31–36.

39 BVerfG, Beschluss des Ersten Senats vom 06. November 2019 – 1 BvR 276/17 (BVerfGE 152, 216–74 – Recht auf Vergessen II) para 46.

40 See n 3.

41 D Sarmiento, ‘An Instruction Manual to Stop a Judicial Rebellion (Before it is Too Late, of Course)’ (Verfassungsblog, 2 February 2017) <https://verfassungsblog.de/an-instruction-manual-to-stop-a-judicial-rebellion-before-it-is-too-late-of-course>.

42 See T Horsley, The Court of Justice of the European Union as an Institutional Actor: Judicial Lawmaking and its Limits (CUP 2018) 159–212 (providing a range of examples in the fields of employment (arts 147 and 149 of the TFEU); social policy (art 153(5) of the TFEU); education, vocational training, youth and sport (arts 165 and 166 of the TFEU); culture (art 167 of the TFEU); public health (art 168(7) of the TFEU); industry (art 173 of the TFEU); tourism (art 195 of the TFEU); and civil protection (art 196 of the TFEU)). For a more general account of competence creep, see S Garben, ‘Competence Creep Revisited’ (2019) 57(2) JCommonMktStud 205.

43 Case C-493/17 Weiss EU:C:2018:1000. For an analysis, see AAM Mooij, ‘The Weiss Judgment: The Court's Further Clarification of the ECB's Legal Framework’ (2019) 26(3) MJECL 449–65.

44 Weiss ibid, paras 59–62, 66. According to the ruling, the measures must be ‘proportionate to the objectives of [monetary] policy’, para 71.

45 ibid, para 73 (‘a broad discretion’). The Court concluded that ‘it does not appear that the ESCB's economic analysis … is vitiated by a manifest error of assessment’, ibid, para 78.

46 The CJEU held that ‘the ESCB is to support the general economic policies in the Union’ and ‘the authors of the Treaties did not intend to make an absolute separation between economic and monetary policies’, ibid, paras 51, 60. This implies that the ECB can, in fact, engage in economic policy, if its actions are indirectly also related to monetary policy.

47 ibid, para 106.

48 cf C Gerner-Beuerle and E Küçük, ‘Consistency and Coherence in Adjudicating the ECB's Unconventional Monetary Policy’ (2021) 70(4) ICLQ 859, 881, noting that ‘[w]hen the Court speaks of “equivalent effects”, it seems to have a test in mind that examines whether secondary market purchases have identical, rather than merely similar, effects to primary market purchases’.

49 Weiss (n 43) para 110.

50 ibid, paras 114–116, 118, 127, 128.

51 BVerfG (n 3) paras 1–237.

52 ibid, paras 116, 118.

53 ibid, para 164.

54 ibid, para 165.

55 ibid, paras 136–137.

56 ibid, para 123.

57 See NW Barber, ‘Legal Pluralism and the European Union’ (2006) 12 ELJ 306; G Davies and M Avbelj (eds), Research Handbook on Legal Pluralism and EU Law (Elgar 2018); G Halmai, ‘Conclusive Remarks’ (2018) 10(2) ItJPL 477. For an alternative term, see D Halberstam, ‘Constitutional Heterarchy’ in JL Dunoff and JP Trachtman (eds), Ruling the World? Constitutionalism, International Law, and Global Governance (CUP 2009) 326, 351–3.

58 European Commission, ‘June Infringements Package: Key Decisions’ (9 June 2021) <https://ec.europa.eu/commission/presscorner/detail/en/inf_21_2743>.

59 European Commission, ‘December Infringements Package: Key Decisions’ (2 December 2021) <https://ec.europa.eu/commission/presscorner/detail/en/inf_21_6201>: Primacy of EU law: Commission closes infringement procedure based on formal commitments of Germany clearly recognizing the primacy of EU law and the authority of the CJEU. See M Ruffert, ‘Verfahren eingestellt, Problem gelöst?: Die EU-Kommission und das Bundesverfassungsgericht’ (Verfassungsblog, 7 December 2021) <https://verfassungsblog.de/verfahren-eingestellt-problem-gelost/>.

60 European Commission (n 58).

61 Joined Cases C-83/19, C-127/19, C-195/19, C-291/19, C-355/19 and C-397/19 Asociaţia ‘Forumul Judecătorilor din România’ EU:C:2021:393.

62 ibid, paras 28, 48.

63 Decision 2006/928/EC establishing a mechanism for cooperation and verification of progress in Romania to address specific benchmarks in the areas of judicial reform and the fight against corruption [2006] OJ L354/56.

64 Joined Cases (n 61) para 173.

65 ibid, para 175. The reports are intended ‘to analyse and evaluate Romania's progress in the light of the benchmarks’ and the recommendations included in these reports are ‘formulated with a view to those benchmarks being met and in order to guide that Member State's reforms in that connection’.

66 ibid, para 176.

67 ibid, paras 177, 178 (emphasis added).

68 It is noteworthy that AG Bobek found the CVM reports and recommendations non-binding but having persuasive authority as soft-law instruments. Opinion of AG Bobek in Joined Cases C-83/19, C-127/19 and C-195/19 Asociaţia ‘Forumul Judecătorilor din România’ ECLI:EU:C:2020:746, paras 157–172.

69 Case C-25/94 Commission v Council ECLI:EU:C:1996:114. In this case, the Commission and the Council entered into an arrangement under which the Commission was to exercise the right to vote in the United Nations Food and Agriculture Organization. Subsequently, however, the Council made a decision that provided that the Member States should exercise the right to vote individually. The CJEU annulled this decision, because it breached the arrangement the two institutions committed to and the subject matter came under exclusive EU competence.

70 ibid, para 49.

71 See Kaczorowska-Ireland (n 1) 147, noting that ‘[t]he ECJ [European Court of Justice] has, on rare occasions, held that where the institution concerned has expressed its intention to be bound by them, such acts may produce legal effects’.

72 Case 22/70 Commission v Council ECLI:EU:C:1971:32.

73 ibid, para 42.

74 Case C-325/91 France v Commission ECLI:EU:C:1993:245.

75 ibid, para 23.

76 Case C-57/95 France v Commission ECLI:EU:C:1997:164.

77 ibid, para 23.

78 Curtea Constituțională (n 3) para 85 (emphasis in original).

79 ibid, paras 83–85. See also Comunicat de presă, 23 December 2021.

80 The CJEU held that the CVM Decision, notwithstanding the general language of the benchmarks set out in it, was clear and precise enough to have direct effect. These benchmarks required Romania to ‘[e]nsure a more transparent, and efficient judicial process notably by enhancing the capacity and accountability of the Superior Council of Magistracy’, to establish ‘an integrity agency with responsibilities for verifying assets, incompatibilities and potential conflicts of interest and for issuing mandatory decisions on the basis of which dissuasive sanctions can be taken’, to ‘continue to conduct professional, non-partisan investigations into allegations of high-level corruption’ and to ‘[t]ake further measures to prevent and fight against corruption’. Although the justiciability of these benchmarks may be dubious and it is doubtful if they can be applied in individual cases by traditional means of statutory interpretation, the CJEU concluded that these ‘benchmarks are formulated in clear and precise terms and are not subject to any conditions, they have direct effect’. Joined Cases (n 61) para 249.

81 Curtea Constituțională (n 3) para 84 (emphasis in original).

82 Joined Cases C-357/19, C-379/19, C-547/19, C-811/19 and C-840/19 PM and others EU:C:2021:1034.

83 ibid, paras 170–175.

84 ibid, para 263.

85 BVerfG (n 39).

86 ibid, para 45.

87 ibid, paras 47–48.

88 For an overview of the Taricco saga, see M Bonelli, ‘The Taricco Saga and the Consolidation of Judicial Dialogue in the European Union: CJEU, C-105/14 Ivo Taricco and Others, ECLI:EU:C:2015:555; and C-42/17 M.A.S., M.B., ECLI:EU:C:2017:936 Italian Constitutional Court, Order no. 24/2017’ (2018) 25 MJECL 357; G Piccirilli, ‘The “Taricco Saga”: The Italian Constitutional Court Continues its European Journey: Italian Constitutional Court, Order of 23 November 2016 No 24/2017; Judgment of 10 April 2018 No 115/2018 ECJ 8 September 2015, Case C-105/14, Ivo Taricco and Others; 5 December 2017, Case C-42/17, M.A.S. and M.B.’ (2018) 14 EuConst 814.

89 Case C-105/14 Taricco EU:C:2015:555. For an analysis, see M Timmerman, ‘Balancing Effective Criminal Sanctions with Effective Fundamental Rights Protection in Cases of VAT Fraud: Taricco’ (2016) 53(3) CMLRev 779.

90 Taricco ibid, para 15.

91 ibid, para 24.

92 ibid, para 55.

93 Case C-42/17 M.A.S. and M.B. EU:C:2017:936. For an analysis, see C Rauchegger, ‘National Constitutional Rights and the Primacy of EU Law: M.A.S.’ (2018) 55(5) CMLRev 1521.

94 M.A.S. and M.B. ibid, para 19.

95 ibid, paras 58–59, 61.

96 ibid, para 62.

97 Case C-399/11 Melloni EU:C:2013:107.

98 See E Xanthopoulou, ‘The European Arrest Warrant in a Context of Distrust: Is the Court Taking Rights Seriously?’ (2022) 28(4–6) ELJ 218, 223–32.

99 CI Nagy, ‘The Diagonality Problem of EU Rule of Law and Human Rights: Proposal for an Incorporation à l'Européenne’ (2020) 21(5) GermLJ 838.

100 See J Wouters, ‘Revisiting Art. 2 TEU: A True Union of Values?’ (2020) 5(1) EurPapers 255, 260 noting that ‘there is a striking asymmetry between the proclamation of the values in Art. 2 and the Union's competences to act upon these values’.

101 Nagy (n 99).

102 The CJEU has never used art 2 of the TEU by itself, but only in combination with another standalone legal basis. Art 2 of the TEU thus plays a merely interpretive role.

103 See K Lenaerts, ‘Fundamental Rights in the European Union’ (2000) 25(6) ELR 575, 586–8.

104 This was inserted in art 6 of the then-effective TEU. See Wouters (n 100) 255–77.

105 G de Búrca, ‘Beyond the Charter: How Enlargement Has Enlarged the Human Rights Policy of the European Union’ (2004) 27 FordhamIntlLJ 679, 680, 695–9; P Cramér and P Wrange, ‘The Haider Affair, Law and European Integration’ (2001) Faculty of Law, Stockholm University Research Paper No 19.

106 There are pending procedures against Hungary and Poland; however, these are stuck in the Council, where the initiative has not been put to the vote for many years, presumably because of the lack of unanimity.

107 cf Opinion of AG Villalón in Case C-617/10 Åkerberg Fransson ECLI:EU:C:2012:340, para 37 observing that: ‘[t]he effect is that the assumption by the Union of responsibility for guaranteeing fundamental rights when Member States exercise public authority in those cases must be examined in terms of a transfer, in the sense that the original responsibility of the Member States is passed to the Union as far as that guarantee is concerned’; K Lenaerts, ‘Exploring the Limits of the EU Charter of Fundamental Rights’ (2012) 8(3) EuConst 375, 377, arguing that the fact that the Charter is legally binding does not imply that ‘the EU has become a “human rights organisation” or that the ECJ has become “a second European Court on Human Rights” (ECtHR)’.

108 See Dutheil de la Rochère (n 27) 136–7; Franklin (n 27); Tridimas (n 27); Dougan (n 27) 1204–7.

109 Weiler and Lockhart (n 35) 63–4.

110 See Denman (n 36) 351; Fontanelli (n 36) 683.

111 Case C-617/10 Åkerberg Fransson ECLI:EU:C:2013:105.

112 Regulation (EU, Euratom) 2020/2092 of the European Parliament and of the Council of 16 December 2020 on a general regime of conditionality for the protection of the Union budget [2020] OJ L433I/1.

113 Conditionality Regulation, art 1.

114 For an exception, see, eg, the Hungarian Integrity Authority, whose remit is limited to matters concerning EU funds. See Act XXVII of 2022 on Controlling the Use of European Union Budget Resources.

115 See Case C-64/16 Associação Sindical dos Juízes Portugueses v Tribunal de Contas ECLI:EU:C:2018:117; Case C-619/18 Commission v Poland ECLI:EU:C:2019:531; for an in-depth analysis of the ruling, see L Pech and S Platon, ‘Court of Justice Judicial Independence under Threat: The Court of Justice to the Rescue in the ASJP Case’ (2018) 55(6) CMLRev 1827; Case C-791/19 Commission v Poland ECLI:EU:C:2021:596; Case C-896/19 Repubblika v Il-Prim Ministru ECLI:EU:C:2021:311. For a general overview of the case law, see L Pech and D Kochenov, Respect for the Rule of Law in the Case Law of the European Court of Justice: A Casebook Overview of Key Judgments since the Portuguese Judges Case, Report No 3 (SIEPS 2021) <https://www.sieps.se/globalassets/publikationer/2021/sieps-2021_3-eng-web.pdf>.

116 Case C-192/18 Commission v Poland EU:C:2019:924; Joined Cases C-558/18 and C-563/18 Miasto Łowicz and Prokurator Generalny zastępowany przez Prokuraturę Krajową and Skarb Państwa – Wojewoda Łódzki and Others EU:C:2020:234; Joined Cases C-585/18, C-624/18 and C-625/18 A. K. and Others v Sąd Najwyższy, CP v Sąd Najwyższy and DO v Sąd Najwyższy EU:C:2019:551; Case C-619/18 Commission v Poland (Independence of the Supreme Court) EU:C:2019:325; Case C-824/18 A.B. and Others v Krajowa Rada Sądownictwa and Others EU:C:2021:153 (the CJEU found the new Polish law on the appointment of judges to the Supreme Court to be contrary to EU law); Case C-204/21 R Commission v Poland EU:C:2021:878 (interim measure to suspend the new Polish disciplinary regime for judges); Case C-791/19 Commission v Poland EU:C:2021:596 (the CJEU found the Polish disciplinary regime for judges incompatible with EU law).

117 Case C-258/14 Florescu EU:C:2017:448. See M Markakis and P Dermine, ‘Bailouts, the Legal Status of Memoranda of Understanding, and the Scope of Application of the EU Charter: Florescu’ (2018) 55(2) CMLRev 643.

118 Florescu ibid, paras 46–47.

119 ibid, para 48 (emphasis added).

120 ibid, para 48.

121 ibid, para 48.

122 Case C-198/13 Hernández EU:C:2014:2055.

123 Case C-467/19 PPU Spetsializiran nakazatelen sad EU:C:2019:776.

124 Joined Cases C-609/17 and C-610/17 TSN ECLI:EU:C:2019:981. For an analysis, see M Tecqmenne, ‘Minimum Harmonisation and Fundamental Rights: A Test-Case for the Identification of the Scope of EU Law in Situations Involving National Discretion?: ECJ (Grand Chamber) 19 November 2019, Joined Cases C-609/17 and C-610/17, TSN’ (2020) 16 EuConst 493.

125 Joined Cases C-411/10 and C-493/10 N.S. EU:C:2011:865.

126 Council Regulation (EC) No 343/2003 of 18 February 2003 establishing the criteria and mechanisms for determining the Member State responsible for examining an asylum application lodged in one of the Member States by a third-country national [2003] OJ L50/1.

127 N.S. (n 125) paras 68–69.

128 ibid, para 65.

129 ibid, para 106.

130 European Community (EC) Treaty, arts 63 and 67.

131 The CJEU followed a similar approach in Case C-426/11 Alemo-Herron v Parkwood ECLI:EU:C:2013:521. For an analysis, see S Weatherill, ‘Use and Abuse of the EU's Charter of Fundamental Rights: On the Improper Veneration of “Freedom of Contract”’ (2014) 10(1) ERCL 167; M Bartl and C Leone, ‘Minimum Harmonisation after Alemo-Herron: The Janus Face of EU Fundamental Rights Review: European Court of Justice, Third Chamber Judgment of 18 July 2013, Case C-426/11, Alemo-Herron v Parkwood Leisure Ltd’ (2015) 11(1) EuConst 140.

132 CI Nagy, ‘Do European Union Member States Have to Respect Human Rights? The Application of the European Union's “Federal Bill of Rights” to Member States’ (2017) 27(1) IndIntl&CompLRev 1.

133 See, eg, CI Nagy, ‘The Commission's Al Capone Tricks: Using GATS to Protect Academic Freedom in the European Union’ (Verfassungsblog, 20 November 2020) <https://verfassungsblog.de/the-commissions-al-capone-tricks/>; CI Nagy, ‘Case C-66/18’ (2021) 115(4) AJIL 700.

134 Case C-78/18 Commission v Hungary (Transparency of Associations) EU:C:2020:476. (Hungary introduced restrictions on the cross-border funding of non-governmental organizations. The CJEU found that the measure restricted the free movement of capital.)

135 H Rasmussen, On Law and Policy in the European Court of Justice: A Comparative Study in Judicial Policymaking (Nijhoff 1986) 415–18.

136 See Horsley (n 42) 261, 266–9.

137 C-286/12 Commission v Hungary ECLI:EU:C:2012:687.

138 European Commission Press Release IP/12/24, European Commission launches accelerated infringement proceedings against Hungary over the independence of its central bank and data protection authorities as well as over measures affecting the judiciary (17 January 2012).

139 The press release section dealing with the retirement age is entitled ‘2) Independence of the judiciary’, ibid 3.

140 Associação Sindical dos Juízes Portugueses (n 115).

141 Case C-66/18 Commission v Hungary EU:C:2020:792.

142 Nagy, ‘Case C-66/18’ (n 133) 702–3.

143 Commission v Hungary (n 141) para 213 (emphasis added).

144 The preservation of the balance of treaty concessions was one of the motivations behind earlier consistent case law denying the direct effect of General Agreement on Tariffs and Trade (GATT)/WTO law in the EU legal order, with only narrow exceptions for measures incorporating WTO law and using WTO law as a means of interpretation. See Joined Cases 21 to 24/72 International Fruit Company ECLI:EU:C:1972:115, paras 21, 27; Case C-149/96 Portugal v Council ECLI:EU:C:1999:574, para 36. None of the major trading nations grant WTO law direct effect. As to the US, see 19 U.S. Code § 3512; as to Japan, see the Kyoto District Court's judgment of 29 June 1984, in Endo v Japan, 530 Hanrei Taimuzu 265, affirmed by the Osaka High Court's judgment of 25 November 1986, 634 Hantei 186, and the Japanese Supreme Court judgment of 6 February 1990, 36 Shomu Geppo 2242. Hence, as a political matter, if the EU unilaterally opened its internal legal space to WTO law, it would seriously handicap its own bargaining position in international trade disputes. Case C-149/96 Portugal v Council ibid, paras 43, 46.

145 Protocol (No 30) on the application of the Charter of Fundamental Rights of the European Union to Poland and to the United Kingdom, art 1(2).

146 European Parliament resolution of 22 May 2013 on the draft protocol on the application of the Charter of Fundamental Rights of the European Union to the Czech Republic (Article 48(3) of the Treaty on European Union) (00091/2011 — C7-0385/2011 — 2011/0817(NLE)) [2016] OJ C55/141.

147 Council of the European Union, ‘Press Release: 3313th Council Meeting’ (13 May 2014) <https://www.consilium.europa.eu/media/25589/142581.pdf>.

148 The travaux préparatoires of the Charter clearly confirm that the purpose of ‘implementing Union law’ was to incorporate the pre-Charter case law on applicability to the Member States, including Wachauf and ERT. See the explanations attached to art 51, Explanations relating to the Charter of Fundamental Rights [2007] OJ C303/17.

149 Quis custodiet ipsos custodes?

150 D Sarmiento and JHH Weiler, ‘The EU Judiciary After Weiss: Proposing A New Mixed Chamber of the Court of Justice’ (Verfassungsblog, 2 June 2020) <https://verfassungsblog.de/the-eu-judiciary-after-weiss/>; JHH Weiler and D Sarmiento, ‘The EU Judiciary After Weiss – Proposing A New Mixed Chamber of the Court of Justice. A Reply to Our Critics’ (EU Law Live, 6 July 2020) <https://eulawlive.com/the-eu-judiciary-after-weiss-proposing-a-new-mixed-chamber-of-the-court-of-justice-a-reply-to-our-critics-by-j-h-h-weiler-and-daniel-sarmiento/>.

151 Of course, the scholarship has not been completely unaware of the potential issue—the clash of EU law and national constitutional identities and the need for a carefully balanced approach have been anticipated for many years. See J Martonyi, ‘Európai jog – magyar jog’ in Tizenegyedik Magyar Jogászgyűlés, Balatonalmádi (Sixteenth Hungarian Legal Assembly, 24–26 May 2012, Budapest) 71–6.

152 For the exceptions, see n 1.

153 The interpretation of the citizenship clause (art 21) of the TFEU provides a good example in this regard. This provision was initially inserted by the Treaty of Maastricht. Its purpose was not to bring about any substantive change but to list the already existing entitlements. Accordingly, it acknowledged and summarized the rights granted to EU nationals by other treaty provisions and emphasized that these rights are subject to limitations and conditions, which in any case had already been included in the Treaty beforehand. Nonetheless, the CJEU departed from the initial constitutional intentions and even from the wording of the citizenship clause. While leaving the rules on actual entry and residence in another Member State untouched, it created piecemeal, by the extensive interpretation of the right to equal treatment, a uniform legal status for all EU citizens lawfully residing in another Member State. While the citizenship clause simply reframed the dispersed rights of the nationals of Member States without any intention to expand them, the Court used it significantly to expand those rights. Nagy, CI, ‘European Citizenship: Free Movement Rights Awarded by the ECJ’ (2005) 1(1) LLR 241–58Google Scholar. See also Horsley (n 42) 269.

154 See Martonyi (n 151) 76, noting that the supremacy of EU law and its reasonable limits, as well as its relationship to constitutional adjudication, is an inevitable question of European integration. In this regard, the European project needs the same mutual restraint and rational consideration of the reality as in all other fields.

155 See, eg, FB Cross, The Failed Promise of Originalism (Stanford University Press 2013); I Wurman, A Debt Against the Living: An Introduction to Originalism (CUP 2017); and LJ Strang, Originalism's Promise: A Natural Law Account of the American Constitution (CUP 2019).

156 See, eg, DA Strauss, The Living Constitution (OUP 2010).

157 ibid, 12–16 (listing several cases where returning to the original meaning of the Constitution would lead to absurd results).

158 See LJ Strang, ‘Incorporation Doctrine's Federalism Costs: A Cautionary Note for the European Union’ in CI Nagy (ed), The EU Bill of Rights’ Diagonal Application to Member States (Eleven International 2018) 145.

159 This kind of occasional defiance may not be unprecedented in the history of federations by aggregation. When, in 1832, in Worcester v Georgia, 31 U.S. 515 (1832), the US Supreme Court sided with the Cherokee Indians against Georgia, the US President publicly defied the judgment, tauntingly saying that ‘John Marshall has made his decision, now let him enforce it’. The judgment was never enforced and the Cherokee Indians were expelled from the territory legally given to them. A similar defiance may be unimaginable today.

160 A €1 million daily fine was imposed on Poland for not dissolving the Disciplinary Chamber of the Supreme Court; see Case C-204/21 R Commission v Poland EU:C:2021:878.

161 See Interview with Andreas Voßkuhle, the president of the GCC, ‘Erfolg ist eher kalt’, Zeit-Online (13 May 2020) <https://www.zeit.de/2020/21/andreas-vosskuhle-ezb-anleihenkaeufe-corona-krise>.

162 Although, arguably, there have been attempts of limited scope to override the implications of certain CJEU rulings, eg Denmark's Accession Treaty, in Documents concerning the accession to the European Communities of the Kingdom of Denmark, Ireland, the Kingdom of Norway and the United Kingdom of Great Britain and Northern Ireland [1972] OJ L73/5, allowed the country to maintain its restrictions on the acquisition of immovable property by foreigners. This exception is currently included in Protocol 32 of the TFEU. In the same vein, Protocol 35 of the TFEU, initially inserted by the Treaty of Maastricht, shields the application of the since-repealed abortion ban of the Irish Constitution. Protocol 33 of the TFEU was inserted by the Treaty of Maastricht in response to the CJEU's ruling in Case 262/88 Barber v Guardian Royal Exchange Assurance Group ECLI:EU:C:1990:209.

163 K Glueck, ‘Scalia: The Constitution is “Dead”’ (Politico, 29 January 2013) <https://www.politico.com/story/2013/01/scalia-the-constitution-is-dead-086853>.

164 Case C-364/10 Hungary v Slovakia EU:C:2012:630.