Constitutionalism without principle: Article 2 TEU and the doctrinal construction of EU values

Abstract

Half a century ago, Martin Shapiro discovered that European constitutional scholarship was stuck at a stage of ‘constitutional law without politics’, presenting the EU ‘as a juristic idea; the written constitution as a sacred text; the professional commentary as a legal truth; the case law as the inevitable working out of the … constitutional text; and the constitutional court as the disembodied voice of right reason’. Recent efforts to doctrinally construct the values in Article 2 TEU show that the mistakes identified by Shapiro persist to this day. This article examines the proposals for constructing Article 2 TEU by Armin von Bogdandy and Dimitri Spieker, and offers three criticisms: First, their doctrinal constructivism is ill-suited to constitutional reasoning. Second, their reasoning has evolved in a way that their work has become self-contradictory. Third and most importantly, especially due to the evolution of their thinking, their scholarship amounts to constitutionalism without principle. Not only does their work lack consistent principled foundations, it also neglects the principled choices underlying EU constitutional law. Moreover, as they do not maintain any critical distance to the CJEU, their work does not offer a principled basis for evaluating its case law. Finally, and most critically, the principles informing their arguments produce an increasingly undemocratic and unconstitutional view of EU law. To address these deficits, I revisit the call for a more contextualised study of European law made by its earliest proponents four decades ago, and demonstrate its continued relevance. Rather than pursuing institution building, EU constitutional scholarship should examine and critically interrogate the socio-political choices underlying the evolution of EU constitutional law.

1. Introduction

Half a century ago, Martin Shapiro discovered that European constitutional scholarship was still stuck at a stage ‘out of which American constitutional law must have passed about seventy years ago’. ¹ The European approach was one of ‘constitutional law without politics’, ² an approach that presented

The Community as a juristic idea; the written constitution as a sacred text; the professional commentary as a legal truth; the case law as the inevitable working out of the correct implications of the constitutional text; and the constitutional court as the disembodied voice of right reason and constitutional teleology. ³

As an experienced American and comparative constitutional lawyer, Shapiro knew that such an approach ‘proved fundamentally arid in the study of individual constitutions’, reducing the study of constitutional law ‘to something like that early stage of archeology that resembled the collection of antiquities. Picking up bits of this and that, oblivious to their context or living matrix, the antiquarian stumbled upon an occasional serendipitous discovery’. ⁴

Times have clearly changed. Until the late 1980s, almost all literature on the law of the European Communities ‘had two essential characteristics: first, it was essentially sympathetic to the integration project; second, it was traditional in character, “based on the exposition of legal doctrine and the analysis of judicial decisions”’. ⁵ In the 1990s, the study of EU law slowly became more contextual, pluralistic, and interdisciplinary. ⁶ This development coincided with a growing appreciation of the limitations of legal-dogmatic research, an acceptance of the need for an open and honest appraisal of the political, constitutional, and societal limits the case law of the Court of Justice of the European Union (Court or CJEU) should recognise, and an increasing criticism of mainstream constitutional orthodoxies. European lawyers of this period experienced these changes as if the field of EU law was being ‘shaken up’. ⁷ While ‘early converts to the European idea in the legal academy … embraced their subject with something approaching a missionary zeal’, ⁸ by the early 2000s it seemed as if ‘criticizing the Court has become all the trend’. ⁹ One could almost think that Shapiro’s analysis since then offers merely a historical lens on constitutional studies in the early days of European integration.

However, some law scholars have been slow to embrace the trends started in the 1990s or still actively resist them. Their work remains traditional in character and is at times still informed by the belief that EU law is special because the European project is ‘an indisputably good cause, a triumph of rationality over the passions, of common interest over national insularity, and … of law over politics’. ¹⁰ And because ‘faith is never particularly conducive to a spirit of sceptical inquiry and to a mode or scale of theoretical reflection that might undermine its founding certainties’, ¹¹ it is perhaps no surprise that from time to time, when reading contemporary scholarship from reputed scholars, it can feel as if we have returned to the 1970s and 1980s, the time when the study of European constitutional law was characterised by a narrow focus on the Court’s case law and an uncritical and acontextual stance toward it. On such occasions, Shapiro’s remarks seem to offer more than just a historical lens on European constitutional studies. It then seems that, over a century after US constitutional studies escaped the stage of constitutional law without politics, European constitutional law remains firmly stuck in the middle of it.

This sense of a return to the constitutional era so aptly summarised by Shapiro has repeatedly struck me in recent years when reading some of the most influential and daring proposals on the construction of Article 2 TEU by Armin von Bogdandy and Dimitri Spieker. Both individually and jointly, and sometimes writing with colleagues, they have developed an approach to Article 2 TEU that has been hard to miss for anyone working in the area of EU constitutional law, especially for those with the slightest of interest in the EU democratic and rule of law crisis. In this article, I critically examine their work and the doctrinal constructivist intellectual tradition laying behind it. The core argument I develop in what follows is that their approach to EU constitutional law is fundamentally unsound, repeating many of the mistakes that Shapiro warned against in the 1980s.

A perfect illustration is an article that von Bogdandy recently published with Jürgen Bast, which seeks to offer a blueprint of what they call a ‘new and principled’ form of European constitutionalism. ¹² Picking up bits and pieces of the Treaties and seemingly unaware of their wider context, they stumbled upon a serendipitous discovery that, if true, would be monumental. Although the Treaties officially ‘have the same legal value’, ¹³ they discovered that ‘the authors of the Treaties … reorganized EU primary law’ on a three-layered model. The first, outer layer consists of all Treaty provisions, but nested inside that outer lawyer we find a second layer – what they call the Treaties’ ‘basic part’ (Articles 1–19 TEU), which ranks above all other provisions. Nested inside the basic part, we find a third layer, which they describe as ‘the EU’s constitutional core’ (Articles 1–3(1) TEU). The constitutional core again ranks above the basic part. ¹⁴

Von Bogdandy and Bast did not make this discovery by finding a hitherto unknown document stating that the drafters wanted the Treaties to be modelled this way. As they are aware, ‘the documents preparing the Intergovernmental Conference are not publicly available’. ¹⁵ No, they made their discovery just as Shapiro predicted. They present Europe as a juristic idea, the Treaties as a sacred text, their professional commentary as the legal truth, the case law as the inevitable working out of the correct implications of the Lisbon Treaties, and the Court as the disembodied voice of right reason. As we will later see, their stipulation that the EU Treaties have a basic part and a constitutional core is without any evidence or reason, and they approach constitutional law in a way that recalls European constitutional scholarship from the 1970s and 80s. Faith in the integrationist potential of constitutional adjudication seems unlimited, the judiciary is treated as the untouchable guardian of the European constitution, and constitutional reasoning is treated as being morally superior to and categorically different from political reasoning.

A similar picture emerges when we read other publications in which von Bogdandy and Spieker set out their vision for Article 2 TEU, especially when we probe beneath the surface of their reasoning. To illustrate this, I will analyse how their views on Article 2 TEU have evolved over time, while highlighting key characteristics their work has in common: inter alia, a commitment to the idea that constitutional reasoning may have recourse only to legal-doctrinal and not extra-legal considerations (or so they say), the support for a highly expansive reading of Article 2 TEU and a seemingly boundless faith in its legal and political potential, and an uncritical belief in the promises of constitutional adjudication, especially by the EU’s judiciary. What emerges, I will argue, is an approach to European constitutional law that, rather than new and principled, is dated and without principle.

Although their layered constitutional approach, with the values in Article 2 TEU in the top layer, should be seen as an attempt to provide a principled (re-)conceptualisation of EU constitutional law, I use the designation constitutionalism without principle for four reasons. First, by claiming that constitutional law may only be interpreted by legal-doctrinal means, von Bogdandy and Spieker neglect the principled choices underlying its interpretation. Second, in their desire to expand the scope of Article 2 TEU, they have repeatedly betrayed the principles they established in their earlier work, resulting, somewhat ironically, in a principled reading of EU law without consistent principled foundations. Third, by failing to maintain any critical distance from the CJEU, their work does not provide a principled basis upon which to evaluate its case law. Fourth, the principles that by now inform their arguments produce a view of EU law that, as we will see, is increasingly unconstitutional and undemocratic in character.

To illustrate these points, my analysis departs from von Bogdandy’s research from the late 2000s, the time when he articulated his legal methodology and applied it for the first time to EU values. Where appropriate, I discuss the Reverse Solange doctrine, first proposed by von Bogdandy and his colleagues at the Max Planck Institute in Heidelberg, and subsequently expanded upon by him and Spieker, to demonstrate the evolution in their thinking on Article 2 TEU. Most of all, my analysis focuses on three publications that best articulate their current views on Article 2 TEU’s position and potential: the article by von Bogdandy and Bast; von Bogdandy’s monograph The Emergence of European Society through Public Law (Oxford University Press 2024); and Spieker’s monograph EU Values Before the Court of Justice (Oxford University Press 2023). The first half of the article is almost exclusively concerned with von Bogdandy’s work, while the second half explores the growing convergence between his work and that of Spieker, which has coincided with a radicalisation of their approach to Article 2 TEU. While some of their work has been co-authored with others (such as Bast), these authors and their other academic contributions fall outside the scope of my analysis.

The criticism I offer is of a threefold nature. The first is methodological. I show that von Bogdandy and Spieker’s reasoning is inconsistent with their commitment to doctrinal constructivism, but also, more importantly, that the method, as interpreted by them, is ill-suited to constitutional reasoning. Second, in substantive terms, they have gotten carried away since they first set out their views on the justiciability of Article 2 TEU (more so von Bogdandy than Spieker). In response to the rule of law crisis, their claims have evolved to the extent that the more recent versions are squarely at odds with their earlier ones. Third, while changing one’s mind can be beneficial, not only are the changes never acknowledged, but they are also undesirable. Von Bogdandy’s original agenda rested on three valuable principles that likely command broad support among constitutional lawyers: namely, constitutional reasoning should (1) not be presented as the dogmatic truth, (2) not over-determine the political process, and (3) not violate constitutional constraints. Each of these principles has been sacrificed by von Bogdandy and Spieker in their aspiration to enable and expand Article 2 TEU’s justiciability.

This might have been less concerning if their ideas had not found fertile soil in Luxembourg. Advocate Generals have cited their work in disputes concerning the EU’s rule of law crisis, ¹⁶ and Koen Lenaerts, the President of the Court, frequently references von Bogdandy’s work in his public lectures on EU values. According to von Bogdandy, the ‘foundational status’ of Article 2 TEU is affirmed by that fact that:

The provision lists 12 principles. Twelve represents the most symbolic number of all. It designates the closed circle; Israel was made up of 12 tribes; Christ had 12 disciples; celestial Jerusalem has 12 gates; and 12 stars, arranged in the shape of a wreath, form the crown of the woman of the Apocalypse. Article 2 TEU thus takes up the European flag’s symbolism, whose 12 golden stars, arranged in a circle against a blue background, promise salvation. ¹⁷

The reasoning is surely imaginative, but more biblical than constitutional. Leaving aside the question of whether the current Union, with its increasing tendency to exclude and victimise the most vulnerable, can appropriately be placed in the Christian tradition, it just makes no sense to affirm the importance of the values in Article 2 TEU by reference to their number and the fact that this number is the same as that of the disciples of Christ or the gates of Jerusalem. Apparently, however, the reasoning makes sense to the Court’s President: in his public lectures, Lenaerts typically cites part of this passage, as if 12 is the magical number, the new mantra from which EU values derive their legitimacy. ¹⁸

Given the intellectual closeness between Heidelberg and Luxembourg, it is in my view important to closely examine the ideas promoted by von Bogdandy and Spieker, as well as the likely effects of their implementation. One effect could be a European law that is profoundly German in character, in terms of both mindset and organisation. Ignoring his former worries about a ‘German legal hegemony’, ¹⁹ von Bogdandy engages in a kind of scholarship that seemingly tries to replicate the ‘value formalism’ of the postwar German Constitutional Court at the European level. ²⁰ This may be unintentional, resulting merely from an aspiration to tie the European and German constitutional systems more closely together. He and Spieker may hope that, by framing the EU value debate in German terms, their work can be a transmission belt from the European to the German constitutional system. However, even if Germanising European law would be a good way to convince German constitutional lawyers of a more Europeanised German law, I doubt it will appeal to the other national constitutional systems.

The article is structured as follows. Part 2 outlines von Bogdandy’s original agenda and its methodological basis. Part 3 critically examines the aforementioned article by von Bogdandy and Bast. I explain that their claim that EU primary law rests on a three-layered model lacks evidence and that their vision is undemocratic and elitist by von Bogdandy’s original standards. In the next sections, I expand the scope of my analysis to demonstrate that von Bogdandy and Spieker are increasingly trying to over-determine the EU political process (part 4) and flout the constitutional constraints laid down in the Treaties (part 5). In the concluding section, I revisit the call for a contextual approach to the study of European law by its earliest proponents four decades ago and demonstrate its continued relevance. EU constitutional scholarship, rather than being an exercise in institution building, should examine and critically interrogate the effects of and the socio-political choices underlying the constitutional decisions made by EU institutions.

2. The doctrinal constructivist research agenda

The intellectual foundations of von Bogdandy’s academic agenda go back a long way, to the establishment of public law as a separate discipline at universities in continental Europe. Yet for our purposes, the analysis can start with von Bogdandy’s outline of ‘the past and promise of doctrinal constructivism’ in 2009, ²¹ and his first explicit application of this legal methodology to ‘the founding principles of EU law’ a year later. ²²

The 2009 article offers a fascinating analysis of the development of legal doctrine in different national systems over the course of the 20^th century, synthesising the results of a much larger collaborative research project on constitutional scholarship in Europe. ²³ Inspired by the continental legal tradition, he proposes ‘a more fully evolved version of the traditional understanding of doctrine’, ²⁴ which he labels ‘doctrinal constructivism’, to address the challenges facing European constitutional scholarship, including changes to national constitutional law and legal scholarship due to the growing role of European law. Without making strong predictions about how the study of law in Europe will evolve, von Bogdandy expects a European legal area to emerge in which the strands of European and national constitutional law will be simultaneously ‘interwoven and independent’. ²⁵ He worries, however, that the study of public law is too fragmented to answer the Europe’s constitutional challenges, and therefore proposes repositioning public law scholarship in Europe around doctrinal constructivism, as a common grammar, as it were.

So what is doctrinal constructivism? It is an interpretive approach that originates in 19^th-century ‘legal science’. The latter was characterised by a ‘struggle for rationality, systematic coherence, logical consistency, building on first principles, obsession with taxonomy, abstractness, precision and clarity of concepts’. ²⁶ Odd as this may sound, 19^th-century legal science was both ‘non-positivist’ and ‘black-letterish’ in spirit. ²⁷ This spirit remains in its reincarnation as doctrinal constructivism, at least in the version initially practised by von Bogdandy. The non-positivist pillar supports the constructivist side of doctrinal constructivism: legal scholars play an active role in developing the law by devising novel legal interpretations for new problems, or even in anticipation of problems that could one day arise. The black-letterish pillar supports the doctrinal side of doctrinal constructivism: scholarly creations are ‘often cloaked as doctrinal “discoveries”’ in order to create the impression that the answers to all legal problems straightforwardly emerge from the ‘seamless web’ of the law. ²⁸

German lawyers who identify with the doctrinal constructivist method today may no longer adhere to 19^th-century legal science. Due to the rise of critical and contextual legal theory, ‘the “scientific character” of legal learning has been seriously questioned’. ²⁹ The approach of many contemporary doctrinal constructivists may be characterised by the effort to interpret the law in a systematic manner and proposing frameworks for how to best structure and conceptualise the law. Rather than claiming scientific objectivity or interpretive exclusivity, these scholars may simply aim to understand how the legal order is constructed through the application of law in legal practice and scholarship. They may criticise the law or suggest improvements to it, without assuming that every suggestion can be magically pulled out of the legal system by doctrinal means. ³⁰ I wish to emphasise that such versions of doctrinal constructivism are not the subject of my critique.

Rather, I want to challenge the more radical, that is, scientific and non-political version of doctrinal constructivism. Von Bogdandy’s doctrinal constructivist approach to law is one that takes an ‘internal scientific perspective, combined with an interest in the establishment of disciplinary autonomy and functional legitimacy’. ³¹ Its core aim is to structure ‘law using autonomous concepts’, whereby ‘law is detached from social reality and tied to legal instruments that flow from the sources of law’. ³² It offers an internal perspective on law since the structuring of law is done ‘not by way of political, historical, or philosophical reflection’ but through concepts that are ‘specifically legal and … fall under the exclusive competence of legal science’. ³³ This is how he seeks to establish public law as an autonomous discipline distinct from disciplines such as ‘philosophy, history, and the empirically oriented political sciences and sociology’. ³⁴ It presents itself as scientific, that is, a quasi-objective and essentially non-political form of constitutional thought, which, to fulfil its role, must keep ‘distance from philosophical and ideological discourses’. ³⁵

Yet von Bogdandy also wants to interpret the law constructively, ie, participate ‘in the development of the law to keep it in line with changing social relationships, interests and beliefs’. ³⁶ Therefore, ‘doctrinal work should not be restricted to the analysis of the positive law but also aim at its propositive development’. ³⁷ While this certainly sounds political, von Bogdandy insists that doctrinal constructivism offers an internal criticism of law that ‘differs from general political criticism since it is phrased in legal terms, is closely connected to the previous operation of the law and can thus be absorbed by the law more easily’. ³⁸ Despite constitutional law being concerned with the political sphere at heart, his doctrinal constructivism seals off constitutional reasoning hermetically from the political domain to preserve constitutional law’s status as autonomous, functional, and objective.

The result is an insoluble tension between the ‘doctrinal’ and the ‘constructivism’. Legal scholars should be actively involved in updating the law so it stays in line with social interests and beliefs, but their interpretation and application of law should be detached from social reality. Constitutional scholarship must take an internal perspective on law, yet develop the law in response to external developments. Legal principles should be reactive to public discourse and societal change, but political or philosophical reflection, indeed any form of interdisciplinary engagement, is outside the proper remit of the study of constitutional law. This tries taking the political out of constitutional law and thereby misses what is special about constitutional law. This version of doctrinal constructivism is unsuitable for the task doctrinal constructivists have set themselves: reconstructing law in response to social change.

Leaving such methodological issues aside for now, von Bogdandy initially pursued an intellectual agenda wedded to the ideal of democratic constitutionalism: it was open to contestation, able to accommodate pluralism, keen not to foreclose the democratic process, and aware of constitutional constraints. First, he was humble in his ambitions, aware of the challenges of his agenda. Criticism of his earlier work made him worried that he would be perceived as ‘undemocratic and elitist’, so he stressed that no systemisation of positive law can be ‘dogmatically advanced as the single truth’. ³⁹ Constitutional lawyers can ‘only propose one and not the system of positive law’, ⁴⁰ and EU lawyers must not present their systemisation as reflecting ‘the one philosophy of integration’. ⁴¹ Second, his initial agenda valued the democratic process. As he argued, ‘a constitutional doctrine must (…) take account of the danger of over-determining the political process’. ⁴² In line with this, he held that the legislative process was the primary institution to be used ‘for the development of a doctrine of EU principles’. ⁴³ Third, von Bogdandy initially accepted that any effort to structure EU law was subject to established constitutional constraints. He claimed that a doctrine of principles could not ‘expand the competences of the EU at the expense of the Member States or to override specific rules [or] downplay sectoral rules which follow different rationales’. ⁴⁴ As he said it, ‘an argumentation based on principles uncoupled from the concrete provisions of the Treaties would misunderstand essential elements of the EU’s constitutional law’. ⁴⁵

In sum, while from the outset there was a mismatch between his methodological tools and his academic ambitions, the ambitions were humble in scope and based on principled constitutional commitments. As we will see, a decade and a half later, little remains of these commitments. Von Bogdandy nowadays presents his organisation of EU law as the one and only one, explicitly wished for by the drafters of the Treaties. Gone are the days, moreover, that the development of EU constitutional principles should not undermine democratic politics and be respectful of constitutional constraints. Instead, as is discussed in what follows, Article 2 TEU is being used to justify the expansion of the CJEU’s authority at the expense of democratic and constitutional norms.

3. Undemocratic and elitist: the ‘new and principled’ constitutionalism

To see what has become of the original agenda, it is best to move straight from the early 2010s to the recently published blueprint of a ‘new and principled’ constitutionalism by von Bogdandy and Bast. Alongside his monograph, which will be discussed later, the article provides von Bogdandy’s most evolved and articulate account to date of European constitutionalism, and Article 2 TEU in particular, and shows the transformation his ideas have undergone since he offered his doctrinal sketch of the EU’s founding principles a decade and a half ago. The humble origins have been discarded: his current views on EU constitutionalism bears little resemblance to his perception of principled constitutional scholarship in 2010.

In their article, von Bogdandy and Bast draw our attention to what they say are ‘two understudied choices of the authors of the Treaties’. ⁴⁶ The first choice was to position Articles 1–19 TEU as the basic part of the Treaties, and the second was to position Articles 1, 2 and 3(1) TEU as the EU’s constitutional core. While they say understudied, in reality these changes must have gone unnoticed. ⁴⁷ Indeed, they implicitly argue that EU lawyers, including they themselves, have consistently misread the choices of the Treaty drafters for almost two decades. This would be astonishing, so it is worth examining their analysis more closely. What evidence do they offer to support their claim that the Treaties have a basic part and a constitutional core, despite the fact that both ‘Treaties have the same legal value’ according to Article 1 TEU? Do they offer any justification?

It is safe to say that a surface reading of the Treaties gives us plenty of reasons to doubt their claims. It is plausible to think that some Treaty provisions are more constitutional in nature, but it is highly debatable that we only find those principles in the TEU or that they overlap exactly with Articles 1–19 TEU or, more narrowly, Articles 1–3(1) TEU. ⁴⁸ Their argument raises manifold questions. To name a few: Why is the constitutional core limited to Articles 1–3(1) TEU and does it not include Article 4 TEU – the principle of conferral, the national identity clause, and sincere cooperation – or Articles 49 and 50 TEU, nota bene the two provisions governing the modalities as to which Member States constitute the Union? Why does the Charter of Fundamental Rights or EU citizenship not belong to the basic part or constitutional core? Why are the basic part and constitutional core not reflected in the Treaty revision procedures set out in Article 48 TEU? How can the constitutional core, Articles 1–3(1) TEU, be demarcated from all provisions in primary or secondary law that interact with and give more concrete effect to this core? And why does the constitutional core not include the two ‘essential characteristics’ of EU constitutional law: primacy and direct effect? ⁴⁹

More questions could be asked, but there is no need to. After all, there is no reason to believe that the EU Treaties have a basic part and constitutional core whose position is different from the rest of the Treaties. Von Bogdandy and Bast support their claim that the Treaty authors reorganised primary law accordingly based on a historic–contextualist reading of the Lisbon treaty. They argue that the Lisbon Treaty effectively implemented the Constitutional Convention’s ambition to better systematise Treaty law, overhauling the first provisions of the former EU Treaty and transforming the first 19 provisions of the TEU into the basic part of the Treaties. Yet, contrary to what they claim, the changes in the first few Titles of the Lisbon Treaty are modest and certainly not of such a kind as to justify the conclusion that Articles 1–19 TEU distinguish themselves in substantive quality from the rest of the Treaties. Let’s go over these changes briefly:

Article 1 TEU mirrors Article 1 EU-Nice, with the exception that Article 1 TEU now mentions the principle of conferral. Article 2 TEU is a longer version of Article 6(1) EU-Nice, which already mentioned most of the EU’s founding principles. The objectives in Article 3 TEU largely replicate the objectives laid down in Article 2 EU–Nice, except for the objective to promote peace and the well-being of its peoples in Article 3(1) TEU. Articles 4 and 5 TEU offer nothing new: the requirement to respect the member states’ national identities was formerly laid down in Article 6(3) EU–Nice, while the principles of sincere cooperation and conferral date back to the earliest days of European integration. Article 6 TEU establishes the first genuine constitutional changes: the declaration that the Charter of Fundamental Rights and Treaties have the same legal value, as well as the obligation to accede to the European Convention of Human Rights. The Article 7 TEU procedure was also part of the Nice Treaty, as was Article 9 TEU in language and/or effect (on Member State nationals being EU citizens and the principle of equality between citizens). The other Provisions on Democratic Principles in Title II TEU indeed offer some important innovations, especially the European Citizens’ Initiative and the recognition that national parliaments play a vital role in the European integration process (Article 12). However, the Nice Treaty made ample reference to democratic values and most of the requirements and aspirations in Articles 10 and 11 TEU were already part of the EU’s constitutional provisions, principles, or traditions prior to Lisbon. Finally, as for the Provisions on the Institutions in Title III, the Lisbon Treaty undoubtedly ushered in some important institutional and procedural innovations, but in language and spirit, Articles 13–19 TEU are in many ways alike the institutional provisions of both the EU–Nice and EC–Nice Treaties.

So what proper constitutional changes have been introduced by Articles 1–19 TEU, apart from Article 6 TEU and the democratic innovations just mentioned? The principle of good neighbourly relations in Article 8 TEU. But that’s it. Are these innovations together with the reshuffling of pre-existing provisions sufficient to conclude that Articles 1–19 TEU stand out from the rest of the Treaties as its basic part? Most certainly not. Moreover, what about other, more important constitutional changes the TEU introduced, such as Article 50 TEU, which are not even mentioned by von Bogdandy and Bast? It remains unclear why the scope of the basis part is drawn as it is, and the evidence presented can by no stretch of imagination justify their claims.

The analysis gets even shakier as we get to the claim that Articles 1–3(1) are the EU’s constitutional core. Von Bogdandy and Bast present their professional commentary as the legal truth: they stipulate that the EU has a constitutional core without feeling the need to substantiate their claims. Their analysis is almost Biblical, reminiscent of the first verses of the book Genesis, where a new reality is created out of nothing, simply because it is so willed. They say, ‘Articles 1–3(1) TEU form Europe’s constitutional core’; and these provisions are Europe’s constitutional core, or so we are to believe. No reasons are given, no evidence is offered, neither about whether the Treaties have a constitutional core in the first place, nor about why the constitutional core ends at Article 3(1), not elsewhere. Why are Articles 4–5 TEU not part of the core, despite that their constitutional significance exceeds that of many of the principles listed in Articles 1–3(1) TEU? It all remains unclear.

Perhaps, and this is admittedly somewhat speculative, we may not know because von Bogdandy and Bast want to keep up the appearance of a purely doctrinal approach to constitutional law. But as there are no legal arguments to justify their definition of the constitutional core, certain political judgments must inform the definition if it is not be an entirely arbitrary one. For example, why does the constitutional core exclude the principle of conferral or national identity clause? That is, the assertion that Article 4 TEU is not part of the constitutional core must come from somewhere (assuming that Article 4 TEU is not excluded because it conveniently supports their backing ‘for the continuous expansion, centralisation and perfection of the EU legal order’ ⁵⁰ ). In sum, the problem for Bast and von Bogdandy is that there is no doctrinal basis to support their idea of the constitutional core and, due to their self-imposed methodological constraints, the extra-legal political reasons cannot be given. The result is an argumentative void: an approach that Shapiro would say is fundamentally arid in the study of constitutions. Unintentionally, then, their article offers the perfect illustration of the shortcomings of practicing constitutional law without politics.

For them, the answer to this problem may lie in their constitutional approach to EU law. They devote a remarkable number of pages to defending the view that appraising the EU legal order in constitutional – as opposed to administrative or international – terms is justified. ⁵¹ They then use constitutionalism as an ‘interpretative approach’ to flesh out the EU’s constitutional core. ⁵² However, the concept of constitutionalism alone does not answer the question of how meaning must be ascribed to the Treaties. After all, von Bast and von Bogdandy’s systemisation of EU law does follow necessarily or even logically from a constitutional conception of EU law. Many EU lawyers accept that a constitutional approach to the EU is appropriate, perhaps vital, without sharing the view that Articles 1–19 TEU are the Treaties’ basic part and Articles 1–3(1) TEU the constitutional core, or even that the Treaties have a basic part or constitutional core. Indeed, accepting that the EU Treaties are a proper constitutional framework does not require us to also accept von Bogdandy and Bast’s systemisation of the Treaties.

It is impossible to escape the conclusion that von Bogdandy and Bast’s principled constitutionalism lacks any constitutional anchoring in the Treaties. It is also difficult to escape the conclusion that their arguments are motivated by a hidden political agenda: to broaden the space for legitimate judicial decision-making, if not the complete judicial overhaul of the Treaties. The clearest indication of this we find in the discussion of Article 2 TEU, the keystone provision of their principled constitutionalism. They ‘identify Article 2 TEU as part of the constitutional core, because that role justifies the CJEU’s principled constitutional approach’. ⁵³ Their admission is honest, but if their enterprise is ultimately an exercise in legitimation, motivated by the desire to justify what the CJEU has decided, their constitutionalism offers no principled basis for appraising the CJEU’s behaviour. ⁵⁴

But not just is their analysis unconvincing as a matter of constitutional reasoning, it is also constitutional analysis with an unfortunate anti-democratic sting. Back in 2010, von Bogdandy said that it would be ‘undemocratic and elitist’ for a systematisation of law to be ‘dogmatically advanced as the single truth’. ⁵⁵ Nowadays, he asserts that his and Bast’s systematisation is the one intended by ‘the authors of the Treaties’. ⁵⁶ Indeed, they present their analysis as the one correct interpretation. Moreover, to see just how far von Bogdandy has drifted from his original views, back in time he argued that legal provisions should not be seen as ‘the higher truth of the volonté générale’. ⁵⁷ Now, he and Bast claim to speak for ‘European society’. In their view, their ‘manifesto shows what the Union and European society are not yet, but want to be – and thus guide and direct the process of transformation, including by legal interpretation’. ⁵⁸ The rise of Euroscepticism and the growing contestation over Europe’s future, also in circles sympathetic to the European project, is apparently irrelevant to understanding what members of European society want. As a result, at a time when ‘the wider literature is characterised by an ongoing search for new perspectives … with the aim of bringing “EU law back down to earth”’, von Bogdandy is moving in the opposite direction. He places his faith in law ‘as a top-down and institutionally driven project of social engineering at a time when such an understanding of the role of law in European integration is increasingly problematised’. ⁵⁹ A European society exists and it shares his conception of morality. In a previous era, von Bogdandy might have said that imposing a uniform conception of morality on a pluralistic society signifies a terribly undemocratic and elitist attitude towards constitutional law.

4. Over-determining the political process

Any principled account of EU constitutionalism must have a coherent and plausible view of the limits of the Court’s mandate, not only in relation to the Treaties, but also in relation to the Member States and EU institutions. As I will argue, von Bogdandy and Spieker do not have such a view. The problem is not only their approach to the study of constitutional law – the methodology and increasingly undemocratic and elitist style of reasoning – but also that their propositions are no longer compatible with the Treaties, upending the EU’s vertical division of authority and upsetting the horizontal division of authority between EU institution. Examining the latter, this section offers a twofold critique. First, von Bogdandy and Spieker attempt to settle the horizontal allocation of authority in a methodologically unsound manner. Second, their proposals over-determine the EU legislative process.

Von Bogdandy and Spieker assume that questions of EU institutional balance can be answered using the hierarchy of norms: ie, ‘doctrines of formal order [that] structure relationships between legal acts. ⁶⁰ These norms ‘determine the relationship between norms within a legal order’, ⁶¹ including the relationship between primary and secondary law, and hence, between the EU legislature and judiciary. Of course, they are not alone in assuming this; this belief is firmly entrenched in German legal scholarship and supports both Dieter Grimm’s over-constitutionalisation thesis and Fritz Scharp’s theory on the asymmetry between positive and negative integration. ⁶² Furthermore, given the influence of their work beyond the German debate, the idea that the hierarchy of norms determines institutional balance appears to be widespread in EU legal studies. ⁶³

However, while the hierarchy of norms formally is the correct point of departure, it provides little guidance in the search for correct division between legislative and judicial authority. Of course, primary law ranks above secondary law, so the latter must be struck down if violating the former. However, the meaning of most provisions of primary is open-ended, certainly that of the provisions that are typically in dispute. The crucial question, therefore, is how much discretion the legislative process must have in fleshing out such open-ended provisions. How much discretion must it have in regulating, say, the internal market or fundamental rights? At this point, we no longer deal with a legal question about the interaction of legal sources, but with a political question about the proper allocation of authority. Indeed, as Martin Loughlin has rightly noted, many legal disputes

Invariably entail claims about the relative authority of case law versus legislation, of legislative will versus constitutional principles, of constitutional principles versus judicial precedents. These conflicts … cannot be resolved by legal method; they require the exercise of political judgment. ⁶⁴

Therefore, as I have argued, ‘it is, for most purposes, not the interaction of legal sources that structures the allocation of political authority but the allocation of political authority that structures the interaction of legal sources’. ⁶⁵ From this, it follows that the doctrinal (constructivist) method is not suitable for defining the EU’s inter-institutional allocation of authority.

The Court has understood this rather well. Its case law is by no means consistent and sometimes invokes the hierarchy of primary versus secondary law to circumvent the legislature’s choices. ⁶⁶ Often, however, the Court respects these choices, for example, because the legislature is better placed to make complex technical or socio-economic judgments. ⁶⁷ This deferential stance has led to a practice known as ‘the legislative priority rule’. ⁶⁸ Accordingly, in the event of (full) harmonisation, national measures are reviewed solely in the light of the applicable legislative provisions and not on the basis of primary law. ⁶⁹ In other words, the hierarchy of norms is a less important device for structuring the division of institutional authority than many EU lawyers assume and critics of the Court fear.

In an attempt to allay such fears, Spieker has sought to ‘encourage legislation’ by limiting the Court’s influence on the legislative process. ⁷⁰ Before analysing his proposal, however, it is necessary to explain his approach to EU constitutional law and identify the similarities and differences between his approach and that of von Bogdandy. Like him, Spieker claims that EU constitutional law is based on a three-layered model. Article 2 TEU forms the ‘summit’; the next layer below consists of ‘provisions that give expression to the EU’s founding values’. Other provisions of EU primary law form the third layer, or base of the constitutional mountain. ⁷¹ However, while Spieker also adopts the doctrinal constructivist method, ⁷² his findings differ significantly from von Bogdandy’s. According to Spieker, the constitutional core consists exclusively of Article 2 TEU (excluding Articles 1 and 3(1) TEU). The second layer, in his view, includes the Charter, the principles of primacy and direct effect, and several TFEU provisions on EU institutions – all of which are absent from von Bogdandy’s conception of the basic part. These differences alone are illustrative of the necessarily arbitrary nature of choosing what belongs in which layer.

Unlike von Bogdandy, Spieker attempts to justify his decisions on what comprises the second layer (the basic part). However, it is at this stage that his constitutional edifice quickly unravels, exposing the limitations of a purely doctrinal approach to constitutional law. The second layer includes the ‘provisions that give expression to a value enshrined in Article 2 TEU’. ⁷³ Using this approach, it makes sense that the Charter is included in the second layer. However, as Spieker concedes, other core principles such as ‘autonomy, primacy, and direct effect’ and ‘the Union’s competences seem difficult to square with this scheme’. ⁷⁴ Nevertheless, he finds a way to include these principles in the second layer, arguing that ‘even these issues feature links to Article 2 TEU’. ⁷⁵ EU competences are linked to Article 2 TEU since ‘Jürgen Bast and Armin von Bogdandy have demonstrated that competences can be understood as a manifestation of the rule of law’. Autonomy, primacy, and direct effect also give expression to Article 2 TEU because these principles ‘might be derived from the rule of law and the value of equality’. ⁷⁶

This is quite the way to devalue core constitutional principles. Three shortcomings are particularly evident.

First, his justification fails to capture the constitutional value of primacy, direct effect, and autonomy, as well as the division of competences. It is precisely because of primacy, autonomy, and direct effect, that the EU legal order is considered constitutional, so deriving their ‘constitutional status’ from Article 2 TEU is putting the cart before the horse. The political and constitutional significance of primacy and direct effect lies not in the value of equality, but in how they reorganised – many would say constituted – the EU legal order, enabling EU law’s enforcement in the national legal orders. ⁷⁷ Likewise, the constitutional significance of the principles that govern the division of competences is multifaceted: it limits centralisation, enhances pluralism, allows for national democratic self-expression, and protects national autonomy and identity. We do not capture the deeper value of these principles by connecting them to the rule of law. ⁷⁸

Second, what if these constitutional principles do not give expression to an Article 2 TEU value? In his Opinion in the Van Gend & Loos case, Advocate General Roemer took the view that the EU should not enjoy direct effect because the effect would be ‘an uneven development of the law involved’. ⁷⁹ More recently, Justin Lindeboom has argued that it would be ‘either wrong or tautological’ to justify primacy by reference to equality. ⁸⁰ What if Roemer or Lindeboom are right? Should primacy and direct effect then no longer be regarded as EU constitutional principles? This conclusion would be absurd; it shows why the constitutional status of specific Treaty provisions or principles of EU law cannot be made conditional on a link with Article 2 TEU.

Third, and most importantly, Spieker’s approach does not allow us to differentiate between provisions of primary law that most EU lawyers (including Spieker) believe to be of greater constitutional significance than others. If all that is needed for a provision to be recognised as belonging to the core of EU constitutional law is a link with an Article 2 TEU value, then EU non-discrimination law belongs to this core: it features more obvious links to equality than primacy or direct effect. According to this approach, also EU citizenship, the free movement of workers and service providers, and arguably the entire economic constitution should be recognised as having equal constitutional status as primacy and direct effect: again, these areas of law, by virtue of the prohibition of non-discrimination, feature closer links to equality than primacy or direct effect. Indeed, Spieker’s approach cannot even sustain his own reorganisation of primary law (he places the economic constitution in the third layer), ⁸¹ and it leads to a vision of the EU constitution that hardly any EU lawyer would agree with.

Having explained this, let us return to Spieker’s attempt to create more room for the legislative process. He argues that the Court should ‘retain… strict scrutiny for those provisions that give expression to an Article 2 TEU value’, which would allow it to ‘relax the constraints imposed on the EU legislature by provisions that do not feature such a link’. ⁸² Spieker proposes subjecting the latter provisions to ‘interpretative pluralism’, whereby ‘both the legislature and constitutional courts share the responsibility for constitutional interpretation’. ⁸³ However, contrary to what he argues, his proposal would only reduce space for the legislative process. The legislature already exercises interpretative authority over Article 2 TEU, including values such as the rule of law, equality, and fundamental rights. In recent years, for example, legislation has been enacted on election integrity, media freedom, and corporate due diligence. ⁸⁴ Therefore, by defending interpretative pluralism only with regard to Treaty provisions that do not feature a link with Article 2 TEU, Spieker is effectively proposing to limit the legislature’s discretion. Second, the Court often does not exercise strict scrutiny over provisions that, according to Spieker, feature links with Article 2 TEU. For example, in rulings concerning the limits on EU competence, especially Article 114 TFEU, it has been ‘extraordinarily permissive of legislative discretion’. ⁸⁵ So this aspect of his proposal would also expand the power of the judiciary at the expense of the legislature. ⁸⁶

Of course, this does not settle whether interpretative authority over Article 2 TEU – or any other Treaty provision – must be shared between EU institutions or exercised only by the judiciary. But the question of how authority must be shared and the extent to which legislative discretion is justified cannot be answered using legal-dogmatic reasoning. Nor can it be settled by elevating Article 2 TEU to the summit of EU constitutional law. As said earlier, it requires the exercise of political judgement, especially a judgement on the relative merits of different institutions. Which is most legitimate or best positioned to decide this or that issue of EU constitutional law? This is not the place to discuss how authority must be divided between the legislature and judiciary, but in my view, reasons of institutional legitimacy and capacity would in many situations justify a large legislative discretion over primary law. ⁸⁷ I would also argue that interpretative pluralism should be allowed and even welcomed for Article 2 TEU. The effective protection of EU values cannot be realised by courts alone: it requires collaboration between different institutions at both the domestic and EU level. ⁸⁸ If, however, interpretative pluralism should be ruled out for Article 2 TEU, it certainly cannot be ruled out using traditional legal methods.

Despite the critical nature of my remarks, I share Spieker’s view that the legislative process must not be unduly constrained by the judicial process. Von Bogdandy (and Bast) take a different view. While Spieker at least proposes the deconstitutionalisation of internal market law, ⁸⁹ von Bogdandy suggests that Article 2 TEU be used to reconstruct ‘the Union’s economic constitution’, eradicating ‘socio-economic assumptions of the 1980s and 1990s that … have become outdated’. ⁹⁰ Which assumptions he has in mind he does not clarify, ⁹¹ and since the internal market of the 2020s is not the internal market of the 1980s and 1990s, ⁹² it is unclear if von Bogdandy’s assumptions of how the internal market works have kept up with developments in the previous decades. He shows no awareness of the fact that, unlike in the 1980s and 1990s, the bulk of internal market law today is legislated. ⁹³ Therefore, if the CJEU would use Article 2 TEU to reconstruct the economic constitution, it would displace the legislative process as the primary forum for determining the course and content of internal market law.

Indeed, this kind of constitutional thinking is precisely what von Bogdandy warned against in 2010: constitutional thought that does not ‘take account of the danger of over-determining the political process’. ⁹⁴ This danger no longer seems to worry Bogdandy. His conception of democracy has changed: it is no longer concerned ‘primarily with process, but rather with outcome’, ⁹⁵ as illustrated by his depiction of the notoriously untransparent trilogue procedure as democratic: if the outcomes of trilogues meet ‘the principles of Article 2 TEU, we should celebrate it as a sign of European democracy’. ⁹⁶ This same outcome-oriented conception of democracy seemingly underlies his claim that the doctrinal reconstruction of the EU’s economic and monetary constitution in the light of Article 2 TEU would make it ‘more responsive to European democracy’. ⁹⁷ It is irrelevant if, as a result, the political process is displaced: what matters – democratically – is whether the values of Article 2 TEU are realised.

The irony is that, in his devotion of Article 2 TEU, von Bogdandy adopts a position that contradicts his own. Outcome-oriented conceptions of legitimacy are traditionally used by those who believe that the EU cannot be democratic. For them, the EU can only be justified in output-oriented, not input-oriented (ie, democratic) terms. ⁹⁸ Evidently, this is not von Bogdandy’s position. He passionately defends the democratic legitimacy of the EU legislative process. ⁹⁹ Yet since his ultimate concern is not democracy but realising Article 2 TEU values (and no, these concerns are not necessarily compatible), he abandons the legislative process almost as soon as he is done defending it. At no point in his book does von Bogdandy reflect on the implications of his belief in the democratic nature of the EU legislative process for its role vis-à-vis the EU judiciary. So, at a time when EU constitutional studies is finally beginning to study the EU judiciary in its institutional context, ¹⁰⁰ von Bogdandy travels in the opposite direction. This is especially unfortunate since he implores public law scholars to study ‘societal institutions and their operations’ to gain ‘concrete insights’ into concepts and principles. ¹⁰¹ But rather than trying to gain concrete insights about the Court’s operation in its institutional environment, he studies the Court in isolation. ¹⁰² This betrays a defunct approach to the study of EU constitutional law. It also reflects the fact that, in his view, institutions and procedures – even the value of democratic legitimacy as such – are entirely secondary to the goal of reconstructing European public law in the light of Article 2 TEU.

5. Sacrificing constitutional constraints on the altar of EU values

The shortcomings of von Bogdandy and Spieker’s approach to EU constitutional law run deeper still. This section shows that their proposals cannot plausibly be aligned anymore with the Treaties, especially the vertical division of authority between the EU and its Member States. At times, their value-based reasoning clearly runs counter to the rule of law itself. Before showing this, I first discuss how their views on the justiciability of Article 2 TEU have evolved.

In 2012, von Bogdandy and colleagues at the Max Planck Institute in Heidelberg published the pioneering Reverse Solange proposal. Their ambition was to safeguard ‘the essence of fundamental rights enshrined in Article 2 TEU’. ¹⁰³ However, the proposal’s legal basis was not Article 2 TEU but EU citizenship, more specifically, the ‘substance of rights’ doctrine laid down in the Ruiz Zambrano judgement one year prior. ¹⁰⁴ They turned to EU citizenship believing that Article 2 TEU could not be the basis for their proposal. They argued that ‘Treaty objectives and values as such [have no] direct effect’. ¹⁰⁵ Yet not even a decade later, after the Court had construed Ruiz Zambrano so narrowly that it could no longer support the proposal, ¹⁰⁶ von Bogdandy, now writing with Spieker, took the opposite view: Article 2 TEU suddenly could be directly effective and thus the ‘hook’, ie, doctrinal basis, for the Reverse Solange doctrine. ¹⁰⁷ The caveat was that Article 2 TEU could be applied ‘in combination with other Treaty provisions’ but not autonomously. ¹⁰⁸ As Spieker argued in his book, Article 2 TEU ‘requires concretization and is difficult (if not impossible) to apply as a free-standing provision to specific cases’. ¹⁰⁹ As you can perhaps guess, this restriction has also been dropped: Spieker and von Bogdandy (together with Bast) now take the view that Article 2 TEU can be invoked autonomously. ¹¹⁰

The limits of the doctrinal constructivist method are on full display. Von Bogdandy and Spieker typically make sure to emphasise that their proposals are consistent with existing doctrine; ¹¹¹ and yet, their analysis of the same doctrinal material using the same doctrinal method has resulted in a complete reversal on the justiciability of Article 2 TEU: from a non-directly applicable provision that could only inform the interpretation of EU law to one that can autonomously be relied on against Member States. Not only is this U-turn impossible to justify from a purely internal doctrinal perspective, but it is also not acknowledged or defended by von Bogdandy. Spieker, as we saw, has tried to find a doctrinal argument for Article 2 TEU’s applicability, but he abandons his commitment to doctrinal constructivism when it suits him. For example, he has urged an end to the ‘doctrinal contortions’, claiming that the autonomous application of Article 2 TEU would be the right legal approach given ‘the gravity of the situation’ in Hungary. ¹¹² Not only is this a surprising position from someone who routinely argues that law may only be developed ‘by recourse to legal methodology’ and not ‘extra-legal’ considerations, ¹¹³ but it may also be a telling one. Spieker may also realise that it is difficult to make a sound case for Article 2 TEU’s autonomous application using traditional legal methods alone. His doctrinal arguments in any event do not make the case. ¹¹⁴

I do not rule out the possibility that decisions on the applicability of Article 2 TEU may be partly based on extra-legal considerations. As I argued in Section 4, recourse to such considerations may be necessary to resolve constitutional issues. However, this requires a different perspective on constitutional law to that taken by von Bogdandy and Spieker, and it does not mean that there are no constitutional constraints. Article 2 TEU is certainly not a trump card that can be played to circumvent such constraints, even if von Bogdandy and Spieker at times suggest otherwise. They always emphasise that their proposals respect the Treaties, but as their views on the justiciability of Article 2 TEU have relaxed, so have, as we now see, their views on where the limits of enforcing Article 2 TEU lie.

When the Reverse Solange doctrine was first formulated, it was foreseen to be of limited scope: it could be triggered only in ‘the extreme constellation that a violation is to be seen as systemic’. ¹¹⁵ Von Bogdandy and Spieker developed the doctrine, basing it on Article 2 TEU rather than EU citizenship, but they maintained similar scope limitations: ‘Article 2 TEU should be interpreted narrowly’ and ‘be read as negatively determining what is not allowed, without positively determining how it should be instead’. ¹¹⁶ Concretely, as regards fundamental rights, Article 2 TEU should only be used for ‘demarcating red lines which are transgressed … only when their “essence” is concerned’. ¹¹⁷ As we will see, von Bogdandy in his own work still stresses that national deficiencies in the light of Article 2 TEU must be systemic before this provision can be triggered, ¹¹⁸ but his work with Bast and Spieker is no longer consistent with this view. Spieker even seems willing to sacrifice initial scope limitations altogether, favouring an application of Article 2 TEU unmoored by legal constraints and considerations.

Let me first explain how Spieker’s thinking on Article 2 TEU has evolved. He has discarded the Reverse Solange doctrine as it ‘does not reflect the current jurisprudential outlook anymore’ and ‘linking citizenship and fundamental rights seems conceptually problematic’. ¹¹⁹ He now favours ‘operationalizing Article 2 TEU itself’. ¹²⁰ But of course, he also preferred this when he supported Reverse Solange: remember, his legal basis for the doctrine was Article 2 TEU, not EU citizenship. So what has changed? Not his view that Article 2 TEU can be utilised but the conditions under which it can be. First, while he, as we saw, used to oppose a positive reading of Article 2 TEU, he now finds that the Court must define the ‘positive content of the values enshrined in Article 2 TEU’, ¹²¹ including by imposing democratic requirements beyond the Charter’s scope on issues such as ‘the curtailing of opposition rights, unfair electoral laws, gerrymandering, party financing, and campaigning rules’. ¹²² Second, and more importantly, the requirement that Article 2 TEU’s enforcement must be limited to exceptional situations involving systemic deficiencies in the observance of EU values has also gone overboard. This is clearly shown by his vision on the legality of citizenship by investment (CBI) schemes under EU law. In Commission v Malta, the Court cited Article 2 TEU to justify its conclusion that the Maltese CBI scheme violated EU law, but without clearly specifying which EU value had been breached. ¹²³ Writing with Ferdinand Weber, Spieker already invoked Article 2 TEU in favour of finding a violation before the judgement. In their view, the conferral of nationality without a genuine link can put values such as solidarity, democracy, and equality at risk: the Maltese CBI scheme would therefore be contrary to Article 2 TEU. ¹²⁴ Since then, Spieker has on his own again argued that the judgment can be justified in light of Article 2 TEU. ¹²⁵

Before I address this argument on its merits, let it be clear that it entails a serious weakening of scope restrictions on Article 2 TEU. Think about the Maltese CBI scheme all you want, but it would be far-fetched to call it a systemic deficiency in the values laid down in Article 2 TEU. This is certainly so if we use the definition of a systemic deficiency Spieker uses himself. Violations are systemic if they ‘affect the entire system (or an entire area of a system) and occur in a wide-spread manner’ and when they occur with ‘regularity and persistence’. ¹²⁶ It is just not plausible to suggest that the situation in Malta is comparable with that in Hungary or that the Maltese CBI scheme is affecting an entire system in way even remotely comparable to Hungary’s persistent undermining of the independence of the judiciary. Favouring the application of Article 2 TEU in the context of Commission v Malta logically entails accepting that a deficiency need not be systemic for Article 2 TEU to be triggered. ¹²⁷

On the merits, moreover, the argument is dubious, especially if we follow Spieker in applying the usual ‘toolkit of legal methodology’ to constitutional law. ¹²⁸ The wording of the Treaties cannot support Commission v Malta. According to Article 20 TFEU, EU citizenship ‘shall be additional to and not replace national citizenship’ and Declaration No 2 on nationality of a Member State says explicitly that nationality ‘shall be settled solely by reference to the national law of the Member States’. The Treaties’ drafting history cannot support the judgement either. Declaration No 2 intended to offer Denmark guarantees so that it would ratify the Treaties. ¹²⁹ Moreover, the sentence saying that EU citizenship ‘shall be additional to and not replace national citizenship’ was added with the Amsterdam Treaty to stress EU citizenship’s subordinate nature. A systematic reading of the Treaties, following which Article 20 TFEU is read in conjunction with Article 2 TEU, would then be the only way to justify Commission v Malta, but a joint reading of these provisions cannot justify the judgement either. Many national citizenship policies infringe on the values of democracy, solidarity, or equality, so if non-systemic violations of these values would automatically amount to a violation of EU law, especially given how vaguely defined these values are, Article 2 TEU could effectively become a trump card with which the Court could rewrite the Treaties.

To drive this point home, consider the following example: third-country nationals with a strong bond with a Member State might be able to acquire nationality and hence EU citizenship. This constitutes a democratic wrong, ¹³⁰ but the EU lacks the competence to address it. Yet using the line of reasoning used by Spieker, one could argue that the situation amounts to a violation of the values of democracy and equality and that the CJEU is therefore licensed to intervene on the basis of Article 2 TEU. Or consider this example: EU citizens living in a Member State other than their own cannot vote in non-municipal national elections. This might be a democratic injustice, but it seems mandated by Article 22 TFEU. Yet following Spieker, one could argue that Article 22 TFEU is ambiguous: it does not explicitly say that EU citizens have no right to vote in non-municipal national elections. ¹³¹ Therefore, as a way of protecting the values in Article 2 TEU, EU citizens could be given a right to vote in all national elections of the state where they reside. Other examples could be given, but I hope that the point that the application of Article 2 TEU as favoured here is hard to square with the constraints imposed by the Treaties is clear.

Von Bogdandy is not too concerned with these constraints either. The implications of the argument he develops with Bast are potentially even more extreme. I say potentially because a problematic feature of von Bogdandy’s work is that it ‘remains remarkably silent about the concrete doctrinal reconstructions that Article 2 TEU demands’. ¹³² He (and Bast) offer ample suggestions, but none are elaborated upon; and they articulate some scope limitations on their argument without ever clarifying their practical effect. That said, let me try to offer a reconstruction of their argument based on the information available to us.

In their view, Articles 1–19 TEU must be used to ‘align and reorganize the entirety of EU law’. ¹³³ Although Treaty provisions have ‘the same legal value’, ¹³⁴ they submit that ‘the formal equality of norms does not imply their substantive equivalence’ ¹³⁵ and that Articles 1–19 TEU and ‘the constitutional principles of Article 2 TEU take precedence’ over other Treaty norms. Although von Bogdandy and Bast accept that these norms cannot be nullified, ¹³⁶ they argue that they must, if necessary, be reorganised using ‘principled’ and ‘transformative’ interpretation. ¹³⁷ The former amounts to the interpretation of Treaty law ‘in a manner consistent with the basic provisions’; the latter amounts to the realisation of ‘structural change’ by ‘transforming’ primary law that violates the basic part. The only condition under which both types of interpretation can be applied is that it cannot provide ‘an interpretation contra legem’. ¹³⁸

What does this imply? As I said, von Bogdandy and Bast do not properly spell out to what ends principled and transformative interpretation should be used. Yet the nature of their argument and the examples discussed suggest that the contra legem condition is a thin one that would not prevent a radical reinterpretation of the Treaties. Two examples illustrate this. First, von Bogdandy and Bast argue that Articles 2 and 10 TEU should be read as imposing a legal obligation on the European Central Bank ‘to use its monetary policy instruments to combat climate change’. ¹³⁹ However, it is not only heavily debated whether climate change mitigation falls within the ECB’s mandate, ¹⁴⁰ but if it does, it can combat climate change through monetary policy without relying on Articles 2 and 10 TEU. If not, their preferred usage of these provisions would simply violate the Treaties. Second, von Bogdandy and Bast find that the democratic principle laid down ‘in Article 2 TEU and concretized … in Articles 9–12 TEU is not fully realized … in the field of competition law’. ¹⁴¹ Their objection, it seems, is that the European Parliament’s role in making competition law policy limited, ¹⁴² and they favour, it seems, transformative interpretation to address this democratic deficit. The position is remarkable, especially as von Bogdandy formerly held the view that a doctrine of principles cannot ‘override specific rules [or] downplay sectoral rules which follow different rationales’? ¹⁴³ At a minimum, their claim downplays competition law rules that follow a different rationale. If they also want to circumvent the rationale in competition law by means of transformative interpretation, they possibly also favours an override of specific Treaty rules.

Recently, von Bogdandy and Spieker pushed their ideas even further in an attempt to overcome the possible Hungarian veto on prolonging EU sanctions against Russia. ¹⁴⁴ In their view, a Hungarian veto against extending sanctions would violate the value of solidarity, and because ‘the Russian war of aggression is … an exceptional case’, Article 2 TEU can be applied to defeat the explicit requirement in Article 31(1) TEU that such decisions be taken by the Council acting unanimously. The proposal very much affirms the view that von Bogdandy and Spieker propagate an ever-expanding universe of Article 2 TEU guided by ever-fewer standards. ¹⁴⁵ Formerly, they proposed limiting the application of Article 2 TEU to situations of systemic deficiencies in the observance of EU values. Now the Russian invasion being an exceptional case is for them enough to apply Article 2 TEU. Formerly, they proposed applying Article 2 TEU with the aim of reviewing existing legal acts. Now they want it to be used to amend the procedure by which such acts are adopted. Formerly, von Bogdandy advocated the transformative interpretation of EU law subject to the requirement that it could not be an interpretation contra legem. Now he wants to use Article 2 TEU to overcome the explicit unanimity requirement in Article 31(1) TEU – a contra legem interpretation if there is any. All in all, the proposal affirms that their preferred usage of Article 2 TEU is indifferent to constitutional constraints. It invokes EU values to override a clear constitutional rule, thereby directly challenging the EU’s own rule of law.

This should trouble someone like von Bogdandy, and not even only because of his past belief that a principled-based argument ‘uncoupled from the concrete provisions of the Treaties would misunderstand essential elements of the EU’s constitutional law’. ¹⁴⁶ His views also run counter to his supposed commitment to the rule of law – a value that, he says, takes ‘primacy over the other principles of Article 2 TEU’. ¹⁴⁷ Even EU values must apparently be ranked in order of importance – and the apex value is not democracy or equality, not even human dignity! No, it is the rule of law because it embeds ‘the principle of legality’. This principle ‘ensures that norms can fulfil their social function’ …, ultimately paving ‘the way for social trust’. ¹⁴⁸ It takes quite a lawyerly mindset to place the value of legality above democracy or human dignity, and I very much doubt that current levels of social distrust can be explained by reference to the value of legality.

However, if legality is such an important value and lies at the basis of social trust, we would not expect von Bogdandy to favour expanding Article 2 TEU beyond the limits of legality. His work reflects a worrying pattern in EU legal scholarship whereby propositions to protect EU values are made in direct contravention of basic tenets of the rule of law. ¹⁴⁹ This is illustrated by the example he offers to highlight the transformative potential of the European constitutional core. Of all the noble ends for which Article 2 TEU can be used, he proposes using it to make national judges who violate the EU’s constitutional core criminally liable. ¹⁵⁰ From the perspective of the rule of law and the values this principle is supposed to serve, his proposal is simply self-defeating. Indeed, as van de Beeten rightly remarks, ‘when protecting the EU rule of law requires a disregard for the principle of legality (nulle crimen sine lege), the limits of law as a self-legitimating normative force have surely been surpassed’. ¹⁵¹

To von Bogdandy, his proposals seem to be self-evident. He presents his views as if they are but ‘the inevitable working out of the correct implications of the constitutional text’. ¹⁵² This may explain why he dismisses those who criticise the Article 2 TEU case law as illegitimate for blurring ‘the boundaries between the legal and the extra-legal’, ¹⁵³ even though extra-legal considerations inform von Bogdandy’s reasoning at almost every turn. For him, however, what is at stake is not ‘a good order, but what is … the legal order’. ¹⁵⁴ Any hint that the measures taken to ‘enforce Article 2 TEU are ideological’ must therefore be avoided. ¹⁵⁵ So, even if his reasoning is purportedly informed by Hegelian philosophical thought and even if, as we saw in the introduction, it occasionally lapses into ‘political theology’, ¹⁵⁶ his writing suggests that his exercise is a purely legal one that finds support in the Court’s case law.

It really takes a peculiar German perspective on law to try to impose a systematic order on the entire EU primary law, with all its oddities and disparities, by reconstructing all legal provisions in light of the values in Article 2 TEU and to be convinced that this is a legal exercise free from any conception of what is a good legal order. ¹⁵⁷ Reminiscent of 19^th-century legal science, it pretends that ‘metaphysical speculation as well as moral judgments’ have no place in legal reasoning. ¹⁵⁸ What remains is an attempt to develop EU law in completely new directions, all the while disguising even the most radical proposals as natural doctrinal discoveries. At the same time, however, von Bogdandy has distanced himself from the 19^th-century legal scientists that inspired his doctrinal constructivism. For them, a true science of law would be rooted in historical fact, including legal rules found in the ‘historically given material’. ¹⁵⁹ This ‘historicist’ understanding of law enabled those legal scientists to dismiss natural law as metaphysical speculation. ¹⁶⁰ They would indeed not attempt to alter the substance of the historically given material (in this case the Lisbon Treaty). They would presumably also view it as an ahistorical error to read a legal text (again the Lisbon Treaty) in a way that it is patently contrary to its language and spirit, and then claim that this interpretation was intended by its authors. In other words, while von Bogdandy maintains that his work is doctrinal constructivist, it now resembles more the metaphysical speculations of natural lawyers whom his intellectual forbears so adamantly opposed.

Finally, it is truly remarkable that criticism of the Court is nowhere to be found, either in von Bogdandy’s work or in that of Spieker. ¹⁶¹ Judgments showing the slightest value-based reasoning are interpreted so as to fit the overarching narrative: judgments pointing another way are never mentioned. What motivates this kind of thinking is a matter of speculation, but I presume it serves a double legitimation function: in Shapiro’s words, it allows a representation of the Court ‘as the disembodied voice of right reason’ and of ‘professional commentary as a legal truth’. ¹⁶² But what value has a constitutional scholarship that is unable to see or admit what constitutional reasoning entails? Is the myth of legal neutrality – the possibility to present political choices as objective legal truths – so valuable that we should blind ourselves to the political dimensions of constitutional law? What is gained by a constitutional scholarship that cannot keep a minimum of critical distance to the legal powers that be? What, indeed, is the value of constitutional scholarship that does not offer ‘a basis for critical, democratic debate’ and only acts as ‘a legitimating force for institutional action’? ¹⁶³ At a superficial level the Court might be served by it, but it offers no principled basis for a debate about the present and future of European constitutionalism.

6. Concluding remarks: Is it time for European law in context, finally?

I opened the article by citing Shapiro’s critical remarks on the state of EU constitutional law during the 1970s and 1980s. It transpires that a single quote by a US constitutional scholar from that era, expressing his frustration with EU lawyers’ non-political approach to constitutional law, accurately reflects a particular strand of European constitutional scholarship today. If we return to Shapiro’s article, it also turns out that he had a plausible explanation for what makes this approach to EU constitutional law so appealing to some EU lawyers. It is worth citing him in full:

In short, the real political and economic world has not worked out in the simple, unidirectional way hoped for by the founders of the European Community. There is no hint of this in Professor Barav’s Article (the author whose work Shapiro criticised). Rather, his Article captures nicely the brave intellectual climate of the founding years.

But from the point of view, not of comparative scholarship, but of European political action, is not what Professor Barav has done the best thing to do? In spite of the revival of national loyalties, Euro-Communism, Gaullism, and German industrialism, the Community has been working. Under the circumstances, it may be best to preserve the myth of the founding years, to deal with juristic developments as if they were autonomous, and to speak as little as possible about economic and political threats. After all, legal realities are realities too. Because the law of the European Community has been growing successfully, the Community itself has a better chance of growing successfully. To treat the law as autonomous is to accentuate the positive; that sort of accentuation is important to institution building. ¹⁶⁴

If we adapt the paragraphs to the contemporary context and substitute Professor Barav’s name for that of von Bogdandy, these paragraphs offer a remarkably accurate description of his approach to EU constitutional law.

Once again, the real world is not working out in the simple, unidirectional way that many EU lawyers had hoped. Although EU lawyers certainly speak of threats to European unity, such as the continuing revival of national loyalties, the rise of authoritarianism, and now the decline of German industrialism, their response can be similar to that of their colleagues five decades ago. They might perpetuate the myths of the founding years, and when these are no longer compelling, as von Bogdandy finds, ¹⁶⁵ they may invent new ones – in his case, the myth of a European Society built around the values of Article 2 TEU – and continue as before, ignoring the real world around them. After all, legal realities are realities too, and maybe these legal realities will one day become social realities. ¹⁶⁶ From the point of view, not of constitutional scholarship, but of institution building, this might be the best thing to do.

This is perhaps the best way to understand the scholarship analysed in this article: as an effort to enable and legitimise European political action. EU values are solely seen as a means to justify the extension of EU law’s authority over Member States, never as a yardstick by which to assess or constrain the conduct of EU institutions. This tendency to legitimise is almost inherent in the methodology applied: the law is developed in the light of a selective reading of relevant facts, namely those that support a particular agenda, excluding those that could portray the EU in a less favourable light (eg, focusing on the rule of law crisis rather than the refugee crisis). And so, instead of asking whether, for example, the principles of supremacy, direct effect, and autonomy can undermine EU values and what ought to be done in such situations, these principles are presented as an expression or, in a more extreme case, extension of EU values. ¹⁶⁷ As a way of institution building, such approaches make perfect sense.

But we are not living in the 1980s anymore. It may have been viable to prefer myth over reality at a time when EU law was still ‘blessed … with benign neglect by the powers that be’, ¹⁶⁸ but not anymore at a time when the regulatory reach of European law is being so deeply contested. This attitude poses ‘the danger that a European public might one day reject its lawyers along with its political élites’. ¹⁶⁹ Many European citizens have long done so, and if our best response is to offer myths instead of solutions that may help to address their often legitimate concerns, they will not be won back. Neither the credibility of EU constitutional law nor that of constitutional scholarship is served by an approach that does not acknowledge the socio-political context in which the law operates and to which it contributes. It no longer works as a way to legitimise the EU legal order.

It is also a legitimising strategy the EU no longer needs. When its constitutional foundations were fragile, institution building perhaps served its purpose, but now that these foundations have more or less been solidified (high-profile challenges to them are best seen as an exception to the norm), EU constitutional lawyers really ought to invest more of their energy in examining the actual justification and limitation of EU authority. To repeat Michelle Everson’s criticism of von Bogdandy and Bast from 20 years ago, we should not want a constitutional scholarship that ‘fails adequately to address the exact mechanisms of possible interaction between European law and the society it seemingly serves’. ¹⁷⁰ No credible constitutional debate can turn a blind eye to the consequences of the EU’s actions, not only for those who are part of its society, but certainly also for those who live and try to survive at its margins. A more confident EU constitutional scholarship would take EU values seriously, even if that means questioning constitutional dogmas or their concrete application in individual situations.

The point is that we need a constitutional scholarship that is fit for the current era. This requires us, at the very least, to embrace the basic principles and ambitions of the European Law in Context movement. This movement has always argued for a shift away from purely doctrinal approaches to the study of European law, partly as a way to rectify the shortcomings identified by Shapiro. Leading voices in the movement believe that contextual approaches have become ‘mainstream in EU law’ ¹⁷¹ and even ‘joke that the new frontier, the really cool, edgy stuff in EU law these days would consist of good old-fashioned doctrinal work’. ¹⁷² However, while venturing beyond previously acceptable frontiers, the work of von Bogdandy and Spieker serves as a stark reminder that old-fashioned doctrinal reasoning (to the extent von Bogdandy’s work can still be labelled as such) is still not a plausible approach to EU constitutional law. In what remains, the call for a contextual approach to European law, made by its earliest proponents four decades ago, will be revisited to show its continued relevance.

The movement’s unofficial date of birth was probably the day when Francis Snyder published his article proposing ‘new directions’ for the study of EU law. One of these was to stop focusing ‘especially on the courts’ and ‘extend our concern not simply to other institutions but more broadly to the interrelationship of institutions, rules, ideologies, and processes’. In his view, this required lawyers to use ‘the methods, theoretical insights and findings of other social sciences’. ¹⁷³ These observations are echoed in the criticism I laid out in this article. As a result of the court-centred nature of EU constitutional studies, the inherently institutional dimension of EU constitutional law often goes unnoticed. ¹⁷⁴ Much of the articulation of EU constitutional law occurs outside the courtroom. Because legal scholars often fail to recognise the range of actors (legislative, administrative, etc.) that participate in this process, they are unable to address the crucial constitutional question of which of these actors can legitimately partake in the evolution of EU constitutional law. Moreover, to address this issue, the essentially political dimension of EU constitutional law must be acknowledged. This area of the law ‘blends political reason with legal reason in ways that make it difficult to differentiate politics with law’. ¹⁷⁵ This is why doctrinal constructivism of the kind advocated by von Bogdandy is unsuitable for constitutional reasoning, and as Snyder grasped better than most of his contemporaries, we must draw on the insights of other disciplines. ¹⁷⁶

One of his contemporaries who had an astute view on these issues was Jo Shaw. She was intrigued by ‘the reverential attitude of many EC lawyers, one sustained, rather surprisingly, right up to the present day in a significant body of academic commentary’. ¹⁷⁷ She was also suspicious of this attitude, since it carries ‘only an internal self-legitimating reference focused on the “rule of law” and the almost mythic qualities of law’. ¹⁷⁸ Despite these remarks being made 30 years ago, they apply with equal force to the body of scholarship analysed in this article. It is not only too reverential toward the Court and believes too firmly in the mythic qualities of EU law, it also sometimes fails to draw the crucial distinction between moral and legal reasoning. Von Bogdandy and Spieker have a tendency to deflect moral criticism of their propositions by pointing out that support for them can be found in the Court’s case law. ¹⁷⁹ While it is undoubtedly advantageous for the practical realisability of a proposition to find support in case law, an evaluation of its moral merits cannot depend on this fact alone. The vital distinction between analysing the law as it is and its merits would get lost. While there is much value in scholarship that describes and systematises the standards embedded in EU constitutional law, there is less in scholarship that tries to evaluate or develop the law by these very same standards.

Instead of treating law as a ‘politically neutral instrument of social control’, Snyder argued for making ‘explicit the underlying assumptions about law’ and understanding it ‘in its social, economic and political context’. ¹⁸⁰ Contextual approaches to European law may have become mainstream, but the work of von Bogdandy and Spieker shows that the lessons of the movement’s pioneers remain as relevant as ever. Rather than ‘asserting, reifying, and sanctioning’, our constitutional scholarship must ‘interrogate, question, and unsettle’. ¹⁸¹ It must investigate and critically examine the socio-economic and political choices that underlie the constitutional decisions made by EU actors, and evaluate EU law in the light of social realities, rather than attempting to justify it based on a selective reading of such realities. This requires rejecting value formalism – transformative and maximalist constitutional interpretation presented in technical, non-political language – as it inevitably produces ‘an interpretive monopoly of lawyers shutting out other voices from constitutional interpretation’. ¹⁸² Otherwise, EU constitutional lawyers risk damaging democracy in a manner not too dissimilar to how challengers to the EU would do it.