From Constitutional Pyramid to Constitutional Pluralism: The Transformation of the European Constitutional Imaginary in Context

This paper explores the recent history of the European constitutional imaginary. It argues that the constitutional imaginary that solidified during the early decades of European integration has been deeply challenged in the 1990s, with the emergence of a body of thought known as ‘constitutional pluralism’. In order to do so, the paper first reconstructs the original constitutional imaginary of the EEC. It then analyses the socio-historical context of the emergence of constitutional pluralism, on the one hand, and clarifies how these ideas contributed to redefining the European constitutional imaginary, on the other. 
 
The paper was written as part of a workshop organized in the framework of the Project IMAGINE, which has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 803163).

The iCourts Working Paper Series is funded by the Danish National Research Foundation Grant no. DNRF105.
iCourts -Centre of Excellence for International Courts -focuses on the ever-growing role of international courts, their place in a globalizing legal order, and their impact on politics and society at large. To understand these crucial and contemporary interplays of law, politics, and society, iCourts hosts a set of deeply integrated interdisciplinary research projects on the causes and consequences of the proliferation of international courts.
iCourts opened in March 2012. The centre is funded by a large grant from the Danish National Research Foundation (for the period 2012-18).

Introduction
hold rather different views of its meaning, and of how it relates to the older EU's constitutional imaginary, both claim to support it. In other words, constitutional pluralism has become a (possibly the) legitimate constitutional theory of the EU -and, I will argue, has conveyed a new constitutional imaginary. While constitutional pluralism has been extensively analysed in a theoretical perspective, the conditions (how it happened) and implications (what it meant) of its emergence have been less scrutinized. 7 To this aim, I will focus on the early days of this transformation. I will not be concerned with the genealogy of constitutional pluralism, nor with the detailed discussion of its theoretical underpinnings; rather, I will try to highlight its significance by placing it in the broader context of the European constitutional imaginary. The paper will analyse what pluralist ideas did to this imaginary, and how they contributed to redefining it.
Before exploring this shift in more details, though, it will be useful to define more carefully what I will mean by constitutional imaginary and imagination. The French philosopher Gaston Bachelard, in a study of poetic discourse, once drew a distinction between 'reproductive imagination', the ability to mentally re-create objects previously perceived, and 'creative imagination', the ability to distort perceptions to produce new images. 8 While the first type operates by rearranging actual perceptions in a meaningful way, the other is a process of altering these perceptions. Imagination can be a reconstruction of a given reality; it can also be regarded as a process of redefinition of this very reality. The same could perhaps be said of constitutional imagination -except that it is, usually, not to be found in the writings of poets and novelists, but in more austere sources. By constitutional imagination, I generally mean the process through which a 'collective self-representation' of any given political community and of its institutions is produced. 9 As a result of a long professionalisation, not only politicians or judges, but also legal scholars have come to play a crucial role in this process. 10 Legal theories, among others, carry such representations and contribute to shaping them. The outcome of this process will here be labelled a constitutional imaginary. 11 It takes the form of a relatively structured account of a community and its institutions, specifying in particular the boundaries, the shape and the organisation of power. It can eventually circulate beyond the narrow circles of those constitutional professionals who produce it. 12 Constitutional imagination -the process of devising such a collective representation -is crucial to the legitimation of modern 'imagined communities'. 13 Much like the reproductive imagination defined by Bachelard, it can be seen as a mental re-creation, formalising the arrangements existing in a community, and thus providing 'ways by which [they] can be 'explained' and justified'. 14 But, it should be noted, constitutional imagination can also come closer to 'creative imagination'. That is to say, it can also directly challenge the existing order. For constitutional imagination can also be the process of scrutinising existing institutional arrangements and devising solutions to actual or potential problems. Therefore, while constitutional imaginaries can play a legitimising role in a political community, they can also perform a delegitimising function. In short, they can have at the same time an ideological and a utopian dimension. 15 A fruitful approach to constitutional imaginaries is thus to combine an ideational analysis of their contents, and an analysis of the social processes through which they are produced and used in the context of actual debates. This context encompasses 'not just the representation or embodiment of ideas' carried by legal texts (the constitution or its functional equivalents), but also the broad range of interpretations produced about it by a variety of 'policy communities and entrepreneurs' 16 (politicians, judges, lawyers, etc.) claiming to possess legitimacy on constitutional matters, and struggling to define the legitimate constitutional imaginary. 17 The rest of the paper will proceed in three steps. First, I will sketch out the construction of the European constitutional imaginary until the 1990s. In a second part, I will turn to the specific conditions that made the emergence of constitutional pluralism possible in the European debates of the early 1990s: on the one hand, the fluid political and academic context of the time will be small groups of professionals dealing with constitutional matters. C. Taylor, 'Modern Social Imaginaries', (2002) 14 Public Culture, 91-124. 12 As Charles Taylor puts it, 'what start off as theories held by a few people may come to infiltrate the social imaginary, first that of elites, perhaps, and then of society as a whole'. C. Taylor, 'Modern Social Imaginaries', 106. Here, the distinction between imagination and an imaginary is thus both chronological (imagination is a process, whose outcome is an imaginary) and sociological (the process of constitutional imagination mainly involves constitutional experts, while a constitutional imaginary can be more widespread in a given society -even though constitutional matters usually remains confined to relatively narrow circles). Yet, in both regards, there is more continuity than opposition between these two terms: imaginaries are produced by imagination. highlighted; on the other, the career and work of its 'inventor', the Scottish professor and politician Neil MacCormick (1941-2009) will be analysed. In the final part, I will discuss what constitutional pluralism did to the pre-existing European constitutional imaginary. I will argue that it deeply altered it, without fundamentally breaking up with it.
The 'co-production' of the European constitutional imaginary: Between political utopia and professional ideology The very fact that one can speak of a constitutional imaginary in the EC, and later the EU, constitutes a riddle. For European integration was launched in the 1950s as co-operation under international treaties ratified by the first Member States. Nevertheless, the view that these treaties were to be regarded as functional equivalent of a constitutional text gradually gained ground. It was theorised from the 1980s by prominent legal scholars in the field of Community law. 18 It has since the late 1980s become a doxa among advocates of European integration and many specialists of Community law -even though it is still firmly contested by others. The making of this constitutional representation of Europe has been abundantly studied by law, political science and sociology. Drawing on these works, I will underline that it functioned both as a political utopia (a political project challenging the existing order of European nation-states) and as a professional ideology (meant in a non-evaluative sense, as 'the body of thought and practices through which a profession […] develops and promulgates ideas about the nature of its work and the identities of its practitioners'). 19

European Constitutionalism as Political Utopia
Establishing a European constitution was at first a utopian political project. Although it is generally identified with federalist plans drafted between the inter-war period and the early 1950s, it should not be forgotten that these plans were themselves connected to an older utopia: that of establishing the 'sovereignty of law', and, more precisely, of international law, over the turbulent and bellicose European states. 20 Throughout the second half of the XIX th and the early XX th centuries, movements advocating a better legal regulation of international relations emerged in the Western world. They promoted a law aimed at curbing the will of sovereign states, whose rivalry had extended to the world through colonisation. Law was conceived as an instrument of 'civilisation' of states. 21 The French jurist Georges Scelle for instance theorised in the 1930s a 'unified inter-social law', which could even be regarded as 'international constitutional law. 22 In this perspective, national legal orders were found to be taken in a universal law, to which, far from totally controlling it, they had to submit.
After the Second World War, this utopia was partly taken up by European federalist movements. They were generally eager to point out the intimate connection between federalism and constitutionalism, defined not only by the existence of a higher norm, but more generally by limited government under the law. 23 Yet, they no longer attempted to create a global, constitutional law. Rather, their focus was on Europe: as the case of G. Scelle illustrates, advocates of a global constitutional law increasingly reconverted into European activists. 24 Indeed, in the post-war situation, many proposals of federal European constitutions were drafted. 25 The proposed constitutions were conceived as the expression of a fundamental decision to institute a new sovereign political being. A constitution, federalists believed, was the medium by which the problem of national sovereignty could be solved in Europe: states should be submitted to a higherlevel legal rule, i.e. to a European constitution. Such plans for a European constitution at first met with relative successes. These culminated at The Hague conference in 1948, and most importantly, in 1952, when negotiations for a European Political Community were officially launched by the Six members of the ECSC. Between 1952 and 1954, hopes for a European constitution were very high, with a constitutional committee created within the ad hoc Assembly in charge of drafting the treaties creating the European Political Community and the European Defence Community. But these hopes were to be severely dampened when France rejected the proposed treaties, without even a vote: it was clear after 1954 that the prospect of a 'constituent moment' which would establish a European federation was no longer relevant. Nevertheless, far from disappearing, the constitutional project would gradually be transferred to the EEC. The most famous example of a transfer to the EEC of the European constitutional project was perhaps given by the first President of the European Commission, the convinced federalist and German professor of private law, Walter Hallstein. The making of a European constitution would not, in his view, be achieved through a great constituent 'moment' (an idea that the 'empty chair crisis' was to prove very accurate); rather, it was to result from a process in which law played the central role. As he was to famously put it, the EEC was a 'community of law' (Rechstgemeinschaft): a creature of law, it was inherently bound to the constitutionalist ideal of limited government, equated with that of federalism; 26 a creator of law, it formed an autonomous, supranational legal order, necessarily evolving towards hierarchical supremacy over national legal orders. Even though W. Hallstein was not the sole to theorise European integration as a constitutional process, he thus provides a fairly clear example of the continuities of the European constitutional imaginary: The horizon remained the construction of a hierarchical organisation, in which Community law was to gradually become the supreme norm -much like a federal constitution indeed.

Constitutionality as Professional Ideology
The interpretation of European integration as a process of constitutionalisation was crucially strengthened by the emergence and support of a disciplined body of legal knowledge on the European Communities. This performed an important ideological function: it legitimised the view that the EEC was constitutional, and not international, in nature; its legal interpreters, in turn, were not international jurists but rather professionals of a new kind of law. Yet, this professional legal knowledge at first hardly emphasised the constitutionalist element (limited government) of the European constitutional imaginary; it focused instead on a technical and functional representation of the constitutionality of EC law. Contrary to what W. Hallstein might have liked to think, the success of the constitutional interpretation of the EEC was by no means a preordained process. On the contrary, it resulted of the long-lasting mobilisation of legal scholars, European civil servants, politicians and activists. In fact, efforts to build up a European legal profession were, themselves, politically encouraged, with the hope that it would legitimise the idea of a sui generis European order, independent from the Member States and open to autonomous development. 28 Civil servants and politicians tried to foster academic research and teaching about European unification, especially in the field of law, because it was perceived as an old, respectable and neutral knowledge, whose social authority was hardly disputable. In this context, Community law emerged as a 'weak field', with porous borders between academic and political concerns and actors: 29 the European legal imaginary gained its strength precisely because it was constructed as a field of intermediation between holders of a specialised professional knowledge and political actors. A crucial moment of these early efforts took place at the first European-wide legal congress, promoted by the legal service of the European Communities and held in Stresa in 1957. 30 There, several jurists claimed that, although created according to the standard international law procedure, the ECSC had already evolved into something different from traditional international organisations. Its law was independent from, and, in case of conflict, superior to, that of the Member States. As a consequence, it was said, Community law was of a new, supranational, type, which was constitutional in all but name. But in Stresa, the outcome of the discussions and the reports presented severely disappointed the hopes of the civil servants and activists who had organised the conference. Indeed, many international law scholars systematically attacked the sui generis nature Electronic copy available at: https://ssrn.com/abstract=3647169 of Community law, acknowledging its political role but denying its legal soundness. This criticism was largely driven by the need to reassert the relevance of their discipline: extremely weakened by the events of the first half of the XX th century, international law, they expected, would gain new legitimacy if they could define the new Communities as its jurisdiction 31 . It resulted a long-lasting 'doctrinal schism' in legal conceptions of European integration: the opposition between 'internationalists', holding a rather traditional view of Community law, and 'supranationalism', holding that it was superior and autonomous to national legal orders. However, in this debate to define Community law, a new player progressively emerged, instigating a 'quiet revolution', as J. Weiler was to put it: 32 the European Court of Justice (ECJ). It had been meant to play a limited role in European integration: following the example of the ECSC -in which no court had originally been provided for by J. Monnet -it was expected to deal with technical economic issues. Yet, it decisively contributed to imposing the distinctive nature of Community law, identified with its constitutional character, as a fact. The process is well known and need not be recalled in detail here. 33 Suffice it to say that the Court, in a series of famous cases, defined the essential features of Community law: primacy, direct effect, autonomy. Far from being identified with traditional international treaties between sovereign states, the Court came to construe them as agreements between the peoples of Europe, thus granting to Community law a direct legitimacy, independent from its recognition by national legal orders. To this aim, it adopted a 'teleological' method of interpretation, reading the treaties in light of their general objective of an 'ever closer union'. Eventually, it went so far as describing the treaties as the 'constitutional charter' of the EEC, enjoying primacy over even the national constitutions. The court's judgements were widely discussed. This was the case at the European Parliament, where it found vibrant support among federalists such as Fernand Dehousse, who drafted two reports in 1965 and 1967 advocating primacy and direct effect of Community law. But the judgements were also disseminated in the European legal profession, through their reception by national courts, of course, but also through a dense network of professional associations, publications, conferences and reports. 34 As early as 1963, another conference was organised in Cologne to celebrate the first ten years of activity of the Court of the ECSC. There, the majority of the guests (academics, politicians and practitioners carefully selected among the advocates of European integration) expressed their support for the constitutional thesis. One working group was Electronic copy available at: https://ssrn.com/abstract=3647169 specifically dedicated to discussing the Court of justice as a 'constitutional court'. 35 For, they claimed, in practice it was necessary for Community law to be regarded as superior to all national laws, i.e. as a functional equivalent to a constitution. Otherwise, different rules would apply to similar situations in the Common market, and its economic viability would be seriously threatened. As Pierre Pescatore emphasised, this was not only an economic necessity: such an interpretation was a professional requirement for jurists traditionally 'in need for system' and of a clear hierarchy of norms. 36 Consequently, Community law was to be accepted as supreme. While many acknowledged that it was not yet fully constitutional, given its limited scope, they generally insisted that it was clearly its vocation to become so: Echoing W. Hallstein, the constitutionalisation of the EEC was described as an ineluctable process. Tellingly, it is also P. Pescatore who was among the firsts to point out an issue that was to become central in the debate about the constitutional status of Community law: fundamental rights. 37 For it was objected to the constitutional thesis that the limited scope of the treaties was far from equivalent to that of a constitutional text. This problem was soon taken up by courts. Famously, during the following years, the German constitutional court first endorsed the constitutional character of Community law in 1967, before adopting a much more nuanced position in the Solange decision in 1974. There, the German judges stressed that Community law didn't provide for a sufficient protection of fundamental rights, and therefore affirmed their constitutional right to review it. This was to claim that Community law fell short of meeting the material requirements of constitutional law. But, at the same time, this jurisdictional confrontation paradoxically confirmed that the discussion about Community law was constitutional in nature, since its relevant interlocutors were national highest courts. Gradually, it prompted legal and political action to extend the scope of the problems dealt with by the EEC. At first limited to mostly economic matters, the stakes of the debates rose to become materially constitutional, through the question of the guarantee of fundamental rights and the issue of the highest jurisdictional authority it triggered. This brief historical survey has highlighted two main points. The European constitutional imaginary was at first associated with a political utopia: roughly, that of creating a superior, autonomous, federal-like order. Aimed at 'civilising' European states, it directly challenged the traditional organisation of power in Europe. Through the development of a body of professional legal knowledge of European integration, originally permeated by political actors, this utopia found support in a professional ideology, a theorisation of the specific character of Community law emphasising its functional equivalence to a constitution, and shared by part of the European legal profession. Over the years, the constitutional imaginary was indeed built in EC legal scholarship: in numerous publications, textbooks and commentaries, EC law was reconstructed as a systematic, pyramidal, legal order standing above national laws and directly applicable to the citizens of the Member States. Of course, not all jurists accepted this understanding. hegemonic, it remained confined to relatively narrow circles of supranational 'entrepreneurs' 38but those were also among the most active interpreters of the EEC. It is against this background that the emergence of constitutional pluralism is to be understood.

The Conditions of Change: Crisis and Hybridisation
The 'constitutionalisation' thesis was formalised -even though not invented -in the 1980s. 39 However, it was to be severely questioned during the early 1990s. The 1990s may indeed be regarded as a 'critical juncture' in the socio-history of European legal thinking. A critical juncture is a period of crisis. Institutionalised frameworks are 'relaxed', prompting an increased 'fluidity' between usually distinct areas of social life. 40 As historical institutionalist analysis has pointed out, during such periods, the role of individual actors in producing innovative solutions to the crisis is greatly amplified. Even though this concept was developed to account for political institutional change, it can prove useful to understand innovations in the constitutional imaginary of European integration during the 1990s. In this case, the weak field of Community law was affected by political events (the fall of the USSR, the 'relaunch' of European integration and the creation of the EU) as well as academic transformations (a scientific crisis, a changed academic configuration) that fostered intellectual change. Indeed, facing new challenges, the institutionalised interpretations of Community law were put into question; in this context, individuals coming from outside the field, but possessing highly legitimate resources, were able to enter the debate and to propose new ways of thinking. It is, I will argue, precisely what made the emergence of constitutional pluralism possible. The 'weakness' of the European legal field -the porosity of its internal and external borders -operated as a facilitator of intellectual change -a change of which N. MacCormick was one of the brokers.

The early 90s as a 'critical juncture'
In spite of its constitutional claims, before the 1980s, Community law tended to be framed as a narrowly defined, essentially technical discipline. This was in part aimed at asserting its academic legitimacy. It also resulted from the fact that many of the first jurists with an academic training to Community law retained a strong link with the legal practice of European integration. This proximity can for instance be seen from the first textbooks dedicated to Community law during the 1960s-1970s: They generally tended to adopt a technical perspective, well suited to the problems encountered during the first years of the EEC. Indeed, they often limited themselves to the exegesis of the founding texts or of the Court's judgements. This narrow focus in turn contributed to strengthening the idea that the EEC was essentially a technical, unpolitical organisation. As the American scholar Martin Shapiro bluntly summarised in a review in 1980, But, from the mid-1980s, this situation was to change for political as well as academic reasons. First, politics forced its way into Community law, which became a more salient and controversial issue. 42 From the mid-1980s, efforts led by European institutions to 'revive' European integration, with the Spinelli project in 1984, the Single Act and the Maastricht Treaty, were aimed at reinforcing Community powers. Yet, they triggered a genuine 'constitutional crisis'. Indeed, while the collapse of the USSR made the legitimacy of the EU a little more difficult to express in terms of security, the discourse on the democratic deficit developed. The French President, V. Giscard d'Estaing had already criticised in the early 1980s 'illegal acts' of the ECJ, and instiled suspicions of a 'government of judges'. Shortly after, the Danes voted against the Maastricht Treaty, and the debate was heated in other countries. The politicisation of Community law also resulted from salient jurisdictional controversies. For instance, the Factortame case, and the decisions of the ECJ it generated in 1990 and 1991, raised an important public debate in the UK about the continuing sovereignty of national legislative bodies in the EC and EU. 43 However, the most visible controversy was undoubtedly the 'Maastricht judgement' delivered by the German Constitutional Court (FCC) in 1993. It basically revived the problem of sovereignty: while authorizing the ratification of the Treaty by Germany, the Court of Karlsruhe criticised the extensive interpretations produced by the ECJ and affirmed the necessity to guarantee the 'principle of democracy' in the EU. This judgement, as well as that delivered by the French Constitutional Council, was a direct challenge to the constitutional interpretation of Community law: far from acknowledging the gradual constitutionalisation of the EEC, and the normal superiority of the ECJ, it put them into question in the name of democracy and national sovereignty. On the other hand, for the advocates of the constitutional interpretation of Community law, it not only undermined the legitimacy of European integration as a whole, but it also set the ECJ and the FCC on a potential collision course. Both views appeared irreconcilable. Thus, the politicisation of European legal issues created a strong demand for fresh conceptualisations to trace a way out of the crisis. This political context met with a remarkable transformation in the academic field of Community law: Reflecting simultaneous trends towards professionalisation and opening of the discipline, a new collective of younger specialists of Community law, often with a transnational education, emerged and took up this political demand. Community law was by then a discipline taught in most European legal curricula and producing a steadily raising number of specialised researches. Electronic copy available at: https://ssrn.com/abstract=3647169 late 1980s, with regards to the broader aim of building a 'People's Europe', the European Commission decided to increase the presence of Community-related courses in the universities by launching the Jean Monnet Action. 45 This new strategy in part diverted Community funding from the older academic channels and fora built up during the early years. It also resulted in new specific resources and rewards for those academics specialising in European studies and law. However, in national universities, Community law often remained a small subfield of international, or economic law. 46 This relatively weak position contrasted with the situation of Community law in a few transnational places dedicated to its study that had emerged and gained legitimacy throughout the 1970s -the most important example of which certainly is the European University Institute (EUI), which opened in Florence in 1976. 47 Explicitly encouraging a 'denationalisation' of European legal training, the EUI Law department not only gathered researchers from different European legal traditions. It also developed strong transnational academic exchanges, in particular with North-American universities, contributing to opening Community law to new influences -such as the law in context movement that had been gaining pace in the English-speaking legal world. Finally, its unique status as a 'European university' quickly designated it to European officials as a privileged provider of the kind of constitutional expertise that was so much needed at the time. For instance, J. Weiler and his team in Florence contributed to the drafting of the Spinelli constitutional project in the mid-1980s. 48 As the case of the European Policy Unit -an academic think tank led by J. Weiler at the EUI in the 1980s -also illustrates, these new specialists were directly engaging with the new demand raised by European integration. The emergence of this new professional collective fostered a redefinition of the boundaries of the field. The new European legal scholarship was no longer satisfied with the kind of narrow exegesis described by M. Shapiro. Rather, it sought new intellectual resources in order to renew the theoretical approaches to European constitutional problems. To this aim, they favoured a broader approach to the EEC. This was illustrated by the research programme known as 'Integration through law (ITL)', which insisted on the need to develop a more comprehensive vision of Community law, borrowing to other branches of legal studies -such as comparative law -but also to social and political sciences. 49 In close co-operation with the European commission and the EUI Law department, researchers such as Mauro Cappelletti and the young J. Weiler endeavoured to build up this vision, and thus fostered an opening of the field of Community law. This was later illustrated by the creation of the  Electronic copy available at: https://ssrn.com/abstract=3647169 editors, six were then holding a position in, or associated with, the EUI Law department (including J. Weiler). But, if one looks at the advisory board, the openness of the project of this journal of 'European law in context' is striking: it included political scientists (B. Kohler-Koch, F. Scharpf, G. Majone) as well as legal philosophers (N. MacCormick, A.-J. Arnaud) -and, it should be noted, European officials (C.-D. Ehlermann of the DG IV, and E. Noël, who had retired from the Presidency of the EUI, after remaining for about 30 years the General Secretary of the European Commission). As Francis Snyder made clear in the first editorial, the aim of the journal was indeed to 'trace a new path', calling on interdisciplinary work to go beyond the 'black-letter tradition of law teaching and research' 50 -a tradition deemed unable to deal with the new problems raised by the evolution of European integration. In spite of growing professionalisation, in the early 1990s Community law thus remained a 'weak field', at the intersection of different disciplines and strongly linked to political debates. However, the legitimate academic credentials required to take part in the European legal debate had slightly changed: the new scholarship that developed was more thoroughly specialised in Community law, due to the increasing presence of the discipline in curricula; at the same time, it was borrowing less from international or economic law and more from comparative law, legal theory and social sciences -a situation which made it possible for professionals of these disciplines to step in the European debate. It was, one could say, a case of 'professionalisation through hybridisation'. 51 Gradually, a new role emerged in European legal academia -that of the European constitutional lawyer, less concerned with the exegesis of European treaties than with the discussion of their theoretical and political implications. In the following years, European constitutional law was indeed gradually constituted as a sub-discipline of Community law. This was to be manifested with the creation of institutes (the Walter Hallstein-Institut für Europäisches Verfassungsrecht created by I. Pernice in 1997 in Berlin), hybrid networks (the European Constitutional Law Network, created in 2001) or journals (The European Constitutional Law Review, created in 2005).

A Broker of Innovation
It is in this context that the basic ideas of constitutional pluralism were formulated. They were coined by a Scottish legal philosopher, N. MacCormick. I will here highlight some aspects of his career prior to this intervention, in order to outline the trajectory that led him to be interested in contributing to the European debate, and to be perceived as legitimate to do so. 52 Focusing on this period entails some limitations: in particular, his thought changed over time, especially regarding the definition of the pluralism he promoted. He evolved from an early 'radical pluralism' to a 'pluralism under international law' often regarded as more conventional. 53 But, although this evolution shouldn't be overlooked, insisting on the 'radical' version of his theory, as it was initially expressed during the early 1990s, provides, I believe, a clearer sense of the challenges he posed to the European constitutional imaginary. The first thing to be noted about the career of N. MacCormick until the 1990s is that he was an almost complete outsider to the field of Community law. After a MA in literature and philosophy obtained in Glasgow, N. MacCormick had pursued brilliant studies in jurisprudence at Oxford. There, he met with H. L. A. Hart, who was to become a major influence on his thought. He then, at an unusually early age, was appointed as Regius Chair of Public Law, and the Law of Nature and Nations at the University of Edinburgh, where he remained for the rest of his career. These early achievements provided him with the opportunity to develop his work, and to strengthen his legal philosophical profile. His works in this field were well and broadly received. By the early 1990s, his first book, for instance, had been cited by such authors as N. Luhmann, J. Habermas, R. Alexy or R. Posner. Interestingly, it was also cited by an American court of appeal as early as 1982. 54 Reflecting this success, his famous essay 'Beyond the sovereign state' was originally written as the prestigious annual Chorley lecture of The Modern Law Review, organised at the London School of Economics since its inauguration in 1972 by H. L. A. Hart. Even though he taught in Scotland for most of his professional life, this reputation rapidly provided N. MacCormick with a strong international profile, regularly invited abroad to present his work, and translated in several languages including French, German, Spanish, Italian or Chinese. Some of his works touched upon issues that would prove very relevant to European debates of the early 1990s. In particular, he engaged with the topic of 'law in context', regularly publishing in the British Journal of Law and Society and in political and social science journals, or editing a book on Lawyers in their Social Setting in 1976. Nevertheless, N. MacCormick was decidedly anchored in the field of jurisprudence, where he had secured a strong national and international legitimacy. This is confirmed if we take a look at the references he cited in his articles, and the articles who cited him. 55 In the rich bibliography of N. MacCormick, his most cited articles all belong to the series about European issues published between 1993 and 1997. 56 Unsurprisingly, then, the articles citing these works are mainly to be found in the field of European legal studies. For instance, the journal in which most references to these four articles are to be found is the European Law Journal (13 references), and the author who referred to them the most is Joseph H. H. Weiler (6 articles). Yet, a look at the references used by N. MacCormick provides a completely different picture. These articles hardly feature any reference to authors writing in the field of Community law. Electronic copy available at: https://ssrn.com/abstract=3647169 mobilisations during the years leading to the creation of a Scottish Parliament in 1998. This political experience, accumulated over the years, eventually opened to him the path of European politics. He would indeed run and be elected as a MEP under the flag of the SNP in 1999. He then was appointed an alternate member of the Convention on the future of Europe, in charge of drafting a Constitution for Europe. During the 1990s, N. MacCormick thus not only developed his theoretical interest for European integration: he also entered the game of European politics. The point of this survey of his academic and political career is to stress that N. MacCormick was, before the 1990s, an outsider in the field of Community law debates. However, he had developed a keen political interest in European integration, and possessed all the academic credentials required, at this particular moment, to take part in the European debate. A leading figure of jurisprudence, he was able to offer a much-needed fresh view on European integration by importing its intellectual tools into the European debate; a Scottish political activist, he had a strong political vision for Europe. This made him a very legitimate outsider in a context of increased porosity of the European legal field. It is in this context that the basic ideas of constitutional pluralism were formulated by N. MacCormick -and, maybe most importantly, were audible in the European debate.

The Challenge of Constitutional Pluralism: Transforming the European Constitutional Imaginary
Following Augustin J. Menendez, I will concentrate here on two main theses of constitutional pluralism: 'the differentiated but equal standpoint thesis', stating the absence of supreme legal authority in Europe, and 'the stability beyond sovereignty thesis', i.e. the idea that 'the stability of the European legal order is rooted on non-legal bases'. 61 I will take these theses as starting points to summarily reconstruct the imaginary provided for by constitutional pluralism. Both theses can be read as complementary answers to the crisis of the 1990s. The first thesis reasserted the constitutional character of Community law, but deeply altered its meaning, thus reconfiguring the professional ideology supporting the EU's constitutional imaginary. The other provided European constitutionalism with a redefined political utopia stating that, ultimately, European integration cannot be the entirely legal project of Walter Hallstein. So doing, constitutional pluralism transformed the European constitutional imaginary.

Constitutionalism Reformalised: From Pyramid to Network
The 'differentiated but equal standpoints' thesis directly confronted the professional ideology of the European legal profession, until then mostly revolving around the idea of EC-EU law as a supreme norm. Yet, it opened a path to transforming it, rather than simply deeming it irrelevant. This thesis was grounded on N. MacCormick's previous works. Since the 1970s, elaborating on the teaching of Hart, he had developed his theory of law as a social fact, as opposed to a 'brute fact' of nature. In this perspective, he had drawn a distinction between 'institutional-normative' orders, 62 the paramount example of which is given by national legal orders, and normative but noninstitutional orders, such as morality. His major move, then, was to stress that, while state law constitutes the most obvious example of such an order, not all institutional-normative orders are state law: 63 Churches, sports, etc. are organised by institutionalised norms as well. In other words, 'the institutional theory detaches law conceptually from state'. 64 It opens a path 'to escape from the idea that all law must originate in a single power source, like a sovereign, […] to discover the possibility of taking a broader, more diffuse, view of law.' 65 One need not think about law as a pyramid: it is in fact better understood as a network. Importing his philosophical conceptual apparatus into the 'black-letter' tradition of Community legal scholarship allowed N. MacCormick to answer questions raised by the legal situation of the European Communities in the early 1990s. It first provided a description of the legal situation in Europe. Its aim, reflecting on the Factortame case, was to answer a simple question: 'How does the European Community affect our own sovereignty?' 66 To do so, he attempted to move the debate beyond the traditional divide between international and constitutional interpretations of Community law. Focusing either on states or on European institutions, both views were criticised as 'monocular' 67 and thus provided an inadequate description of reality. Instead, N. MacCormick analysed the European legal practice brought about by European integration through the lens of his legal theory. In particular, in Hartian fashion, he stressed the need to take into consideration the different perspectives available on European law: the national and the Community. What he saw, then, was the coexistence in Europe of a multiplicity of legal orders without a 'single source of power' -or, more precisely, without any order being in the position of claiming ultimate, allencompassing supremacy over the others. It was not only possible, but also explained by a more general theory of law. It also suggested a normative solution to the Maastricht crisis. Reflecting on the Maastricht case, N. MacCormick concluded that, in a pluralist perspective, the German constitutional Court was right to assert that the EEC or the EU was not the ultimate authority in Europe; yet, it was wrong to believe that such an ultimate authority remained in the hands of the German state, or the German people. 68 This was an elegant but firm qualification of both claims: on the one hand, it was not wrong to regard Community law as, in effect, a legal order of its own with primacy over national orders in the appropriate context; but, on the other hand, this did not wipe out the claims made in the name of national constitutions. Or, as he put it, 'the doctrine of supremacy of Community law is not to be confused with any kind of all-purpose subordination of Member State law to Community law.' 69 The 'differentiated but equal standpoints' thesis thus provided a much-wanted fresh conceptualisation of EU law. It was possible to think of it, at the same time, as constitutional and as coexisting with equally supreme national constitutions: It confirmed its constitutional character, and thus reasserted the professional ideology of many EU legal professionals -while strictly limiting their claims. This, however, came at a price. It also involved a major change in the utopian element of the European constitutional imaginary: Since there need not be a single ultimate authority in Europe, the political horizon of European integration is that of coordination, rather than subordination.

A New Utopia for Europe: From Federation to Commonwealth
While the first thesis defined the theoretical coordinates of constitutional pluralism, the 'stability beyond law' thesis dealt with its conditions of possibility. Practical concerns were, from the early 1990s, central to the development of constitutional pluralism. The constitutional crisis was after all prompted by conflicts between courts, as well as political contestations of the democratic legitimacy of European integration. Any answer to this crisis would thus have to be able to provide a practical solution to these challenges. But, if there is no ultimate authority, no 'single source' of law and power, how could in practice such a pluralist order be stable? To answer this question, N. MacCormick suggested that we need to look beyond integration through law. Law cannot provide a solution to all legal, not to mention social, conflicts. 70 Instead, N. MacCormick recommended that judges make use of 'circumspection and [of] political as much as legal judgement'. They should not be blind to the political consequences of their decisions; rather, they should adopt a broad understanding of the matter they have to decide on. Law needs constantly be apprehended in its broader social and political context. This does not imply that law becomes, in his thought, an inherently political matter. On the contrary, he strongly insisted that law and politics are two 'distinct, but related' areas of social life. 71 This, however, emphasised that certain problems require a political rather than a legal solution: law is a tool of European integration, but its integrative power is limited. European integration cannot be a creature of law only. This limitation of the realm and powers of legal integration has a positive side complementing it: the strengthening of (democratic) politics in the EU. Indeed, N. MacCormick developed, throughout the 1990s, an elaborate political project for Europe, trying to reconcile European integration and democratic legitimacy in a pluralist setting. He insisted that democracy was not necessarily connected with the state. As he put it, 'the state-sovereignty version of popular sovereignty can be itself an enemy of other democratic rights', such as minority rights. 72 On the other hand, European integration did already possess a relative democratic legitimacy: it was a 'lawfully constituted' organisation, submitted to the control of Member States, of Courts and of a democratically elected (albeit with limited powers) Parliament. 73 This was, of course, directly engaging with the problem, raised by the German constitutional court, of the absence of a European demos, preventing, in the Such a loyalty by no means implied a loss of national feelings -something that the life-long member of the SNP would have deplored: quite to the contrary, he also pleaded for the rehabilitation of nations and nationalism, in a liberal version he theorised in parallel. 75 In his view, the democratic legitimacy of European integration was not only dependent on reforms at the European level. It was to be improved only by taking seriously a principle put forward by the Maastricht treaty: subsidiarity, and the opportunities not only for national, but also for regional entities -such as Scotland. A strong principle of subsidiarity would not only safeguard state powers, but it would also allow for a better repartition of power within the states themselves. But, contrary to federalists (also keen on insisting on subsidiarity), he warned against 'the weakness of any "grand design" approaches to constitution-making for Europe.' 76 Strengthening democracy in Europe did not imply the construction of either a unitary, or even a federal state. Europe was in fact seen as a 'mixed constitution', combining elements of several types of regimes. He proposed to label this organisation a 'Commonwealth', a concept he defined as an organisation comprising 'a group of people to whom can reasonably be imputed some consciousness that they have a 'common weal', something which really is a common good, and who are […] striving after it through some form of organised political structure, embodied in some common constitutional arrangements'. 77 Such a loose organisation was a far cry from a federation. This was not only a theoretical project. It was also deeply intertwined with his political concern for regional government. During his time at the EP, he actively pushed forward these ideas about subsidiarity and the rights of 'stateless nations' -as was named the parliamentary intergroup to which he belonged. It comes as no surprise that this intergroup explicitly called for a 'constitutional pluralism' in the EU, and its 'logical corollary': an increased subsidiarity. 78 The legal theory of constitutional pluralism and the political defence of regional government went hand in hand. This is a suggestive illustration of the porosity between academic and political discourses in the European constitutional debate: Legal theories are used as weapons in political battles, while these political battles are fought in legal terms. This cursory reading of the 'stability' thesis and its political corollaries suggests how deeply constitutional pluralism challenged the utopian element of European constitutional imaginary. First, it departed from the narrative that Community law was bound to develop into an all-encompassing legal order. It underlined that 'integration through law', too, was of limited reach. Secondly, it proposed to replace the federalist horizon of European legal integration with that of a 'Commonwealth' organised by a strong principle of subsidiarity. It thus not only offered a new formalised description of the European legal situation; it deeply altered its political horizon.

Conclusion
This paper has attempted to provide a contextualised reading of the shift in the EU's constitutional imaginary that was initiated in the early 1990s. It has argued that the mainstream European Electronic copy available at: https://ssrn.com/abstract=3647169 constitutional imaginary had until then represented Community law as evolving towards full supremacy, reflecting both a political utopia (a federal Europe) and a professional ideology (a rationalisation of law as a pyramid). Constitutional pluralism can be read as a redefinition of this European constitutional imaginary. In a critical context, it proposed a new description of the European legal situation. While the constitutional claims of EU law were not denied, hierarchy was crucially replaced by heterarchy; supremacy was acknowledged but limited. It also redefined the political horizon of European integration, replacing the federalist project with that of a European 'Commonwealth' in which no supreme, all-encompassing power was meant to emerge. In this sense, it radically challenged the pre-existing constitutional imaginary of the Community in its ideological as well as utopian dimensions. A more complete study should of course not remain limited to the work of N. MacCormick: to account for the importance of constitutional pluralism, it would be necessary to examine in detail its reception. Yet, by locating these theoretical and political moves in the context of the time, I have suggested that Community law was a 'weak field' indeed. This relative openness of Community law is precisely what made possible the emergence of pluralism in a critical moment. This is not to say that European constitutional imagination is shaped by political considerations only. Rather, the thought of N. MacCormick illustrates the continuous interdependence of political and theoretical discourses in European constitutional matters.