European Legal Integration and Legitimacy

From an article published in L. Fontaine (dir.), Droit et légitimité (Bruylant, Bruxelles, 2011).


For it to be legitimate, this study must be carefully circumscribed. The problem it purports to address is not that of the legitimacy of the structures or organs of the European Union but that of the integration of EU law into the legal systems of its Member States. By what title can EU law impose its norms on national legal systems, and so much so that the conditions in which rule-making is organized within those orders are affected to the point of turmoil? On what conditions can EU law be integrated? What consequences does such normative integration hold for the way in which relations among legal systems are described? Our discussion addresses not so much the actual dynamics of legal integration but rather the discourses of justification it produces and on which it rests.


For it to be legitimate, this study must be carefully circumscribed. The problem it purports to address is not that of the legitimacy of the structures or organs of the European Union but that of the integration of EU law into the legal systems of its Member States[1]. By what title can EU law impose its norms on national legal systems, and so much so that the conditions in which rule-making is organized within those orders are affected to the point of turmoil? On what conditions can EU law be integrated? What consequences does such normative integration hold for the way in which relations among legal systems are described? Our discussion addresses not so much the actual dynamics of legal integration but rather the discourses of justification it produces and on which it rests.

What is legal integration? Achieving a better grasp of this idea is, in truth, one of the challenges of this study. For the time being, let us be content with an image. EU law is integrated into national laws in the sense that it behaves like an occupying authority on foreign soil, by making use of national procedures and by mobilizing state organs so as to directly incorporate its norms within the national jurisdiction of the EU states. Admittedly it is a peaceful occupation, it is not imposed by means of constraint, which the EU lacks. However, it is an occupation devoid of any legitimate popular basis: the EU being an international organization, instituted by treaties, its foundation lies not in the will of a people, sanctioned by a constitutional instrument, but in the series of concordant commitments given by a group of states that accept to transfer a share of their powers to it. The problem of the basis of this occupation is the problem of legal integration.


As European norms are produced within an organization and by organs that are external to the state, their place in national law should depend, a priori, on the conditions domestic law sets out for foreign or international norms. However, it appears that EU norms have some peculiar authority that is very different and certainly superior to the authority emanating from foreign or international instruments. The problem of the legitimacy of legal integration arises from this difference. There are two ways to justify it: either the authority is ascribed to some national title, given a constitutional basis, but it then becomes impossible to justify any full and effective incorporation of EU law other than by admitting a few contradictions within domestic law; or it is based on a European title, a rule of ‘internal primacy’ of the norms of EU law, but the difficulty then lies in identifying a political entity to which those norms might be ascribed[2]. In the first instance, we come to ask how different criteria of validity can co-exist within the same legal system, some applying to national and international norms, the others to European norms and the national norms implementing them; in the second instance, we wonder how foreign norms produced by an intergovernmental organization can prevail over national norms expressing the sovereign will of the people. These are the two sides of a single problem, the problem of the legitimacy of integration.

The situation of European norms in the French legal system reflects these difficulties fairly well. On the one hand, the implementing acts of EU law derive their validity from a ‘constitutional requirement’ contained in article 88-1(1) of the French Constitution.

Important consequences follow from this in terms of litigation: administrative measures implementing EU law are protected, whereas legislative norms transposing European directives enjoy immunity from review in respect of constitutional norms. Domestic courts treat the implementing acts of EU law as constitutionally protected measures. But, on the other hand, this ‘constitutional requirement’ has a special status: being dictated by the development of EU law, its source does not lie properly in the constituent power, but in the conclusion of international treaties to which the Constitution expressly refers. Hence a degree of ‘constitutional dualism’[3]. This explains the Constitutional Court may have jurisdiction to review certain domestic laws with respect to EU norms4[4] but also that to the integration of EU law is subject to the ‘more fundamental’ provisions of the Constitution[5]. Similar dualism, and the contradictions it engenders, is easily to be found albeit in different forms in the law of other Member States.

To claim to absorb the problem of integration into national law but a national law that is unable to formalize a relationship of total submission to EU law, or alternatively to seek to endow EU law with its own force but a force that is powerless to provide a basis for legitimation: that is the crux of the dilemma constantly facing European lawyers.

1.1 The contradictions in the relations between legal systems

It is not in terms of the classical opposition between monism and dualism that these contradictions can be solved. For, faced with the problems raised by European integration, the two approaches come together. In the monist approach, foreign rules are incorporated as such into the domestic legal order. Incorporation is gained by satisfying the validity criteria laid down by the system in which those rules have their source. Such direct incorporation is nevertheless the outcome of a decision internal to this system which, moreover, will decide on the place the foreign rules occupy within its hierarchy of norms. If it opts to subordinate its own domestic norms to certain international norms, this is still done, as A. Pellet writes, ‘by the grace of the provisions of the Constitution itself’[6]. The upshot is a reinforcement of the postulate that the Constitution is superior to any other norm, whatever its source, in the domestic order. Such an approach does not readily admit that certain foreign norms, such as the norms of EU law, enjoy special protection, by being shielded from the constitutional principles of the hierarchy of norms (lex superior, lex posterior, lex specialis)[7]. Therefore, the only practical solution is to reserve a specific regime for EU law. This is how French Constitutional Court recognized ‘the existence of a Community legal system integrated within the national system and distinct from the international legal system’. In so doing it recognized the existence of a new regime created by the Constitution so as to govern the relations between European norms and national norms.

The dualist approach addresses the problem in a different way. A foreign norm can only enter the national legal system by virtue of an individual act of incorporation governed by national law. Thus, the effects produced by this norm within the national system shall depend exclusively on how it has been transformed and on the state’s sovereign decision to discard any acts within its national order that are incompatible with the norm. How are we to understand, in these circumstances, the application of norms like the norms of EU law that have as such ‘force of law in all countries of the Community’?[8] It must be recognized that there is a relatively autonomous system, within the national sphere, standing apart from the system of national sources and governing the application of European norms. This is precisely the approach adopted by a dualist court like the Italian Constitutional Court: the norms of EU law hold within an area of application where contrary national norms are not invalidated but are merely considered inapplicable. However, this area can only be a zone of derogation; European norms can prosper only within the bounds consented by the constitutional order. These bounds are the‘counter-limits’ that the Constitutional Court opposes to the application of EU law[9]. Whether in a monist or a dualist system, it seems difficult to escape the contradiction engendered by the process of integration of EU law.

1.2 Authority and autonomy of EU law

Probably no-one more than the European Court of Justice itself has contributed to recognizing these contradictions. On the relations between the Community legal system and the national legal systems, its first statements showed great humility: ‘The ECSC treaty is based on the principle of a strict separation of the powers of the Community institutions and those of the authorities of the Member States. Community law does not grant to the institutions of the Community the right to annul legislative or administrative measures adopted by a Member State.’ And to a plaintiff’s claim that for Community law to be effective the ECJ had to be recognized as having the capacity to set aside illegal acts by the national administration, it replied that it was not allowed to ‘interfere directly in the legislation or administration of Member States’[10]. But, once these limits had been recognized, the ECJ was careful to define a strategy for establishing the authority of Community law. That strategy involved two essential operations: the teleological interpretation and the systemic interpretation of Community law provisions making it possible both to align the rule-making process with the objectives of integration and to impart to such rules the coherence of an autonomous and complete legal system. Basically, the whole work of the ECJ was to forge, from the material it was given to construe, an objective ‘unity of meaning and value’ protecting Community law against the dangers of duplication or of dissolution within the various national legal systems. The paradox is that this unity was constructed from a particular standpoint, that of the centralized interpreter of Community law. The Community legal system arose out of the objectivization of the standpoint taken by the ECJ on the international obligations contracted by the Member States, which it chose to read as a ‘permanent limitation of their sovereign rights’. By ‘constitutionalizing’ the EEC Treaty in this way, the ECJ sought to shield the European norms from the uncertainties of the constraints associated with the national mechanisms of reception, the risks of new political negotiations, and the bureaucratic complications that foreign provisions are liable to encounter upon entering the national sphere. Its case law had a dual effect: it endowed Community law with a formal basis of validity by creating the conditions for a ‘EC legality’ while conferring complete control of this newly acquired autonomy on the Community Court.

The Costa/Enel decision was the scene for the operation whereby Community law is produced as an independent legal system by the simple effect of the formulas produced by the ECJ. These formulas are familiar enough: ‘the law stemming from the treaty, an independent source of law, could not, because of its special and original nature, be overridden by domestic legal provisions, however framed, without being deprived of its character as Community law and without the legal basis of the Community itself being called into question’[11]. Thus, it was from ‘the very concept of Community’ — an institutional reality but above all, here, a political ambition — that the Community norms drew both their form as common rules and their force as mandatory rules. EC legality was fully aligned with a project for the economic, legal and political unification of Europe, with a ‘grand idea of order’[12]. Legal integration was conceived as a two-tier arrangement: the assumption of a political project — the concept of Community — served as the basis for legal integration, that is, to the subsuming of national legal orders within the Community legal system[13]. Hence the authority of the provisions of Community law was justified: from then on, it was understood that ‘those provisions and measures not only by their entry into force render automatically inapplicable any conflicting provision of current national law but—in so far as they are an integral part of, and take precedence in, the legal order applicable in the territory of each of the Member States—also preclude the valid adoption of new national legislative measures to the extent to which they would be incompatible with Community provisions’[14].

But these assertions of authority were not enough. It was difficult indeed to establish with the national authorities a hierarchical relation between European norms and national norms: in the course of integration, there was no shortage of examples of challenges to the authority of EU law. From one national order to the next, and sometimes even within one and the same order, legal integration prompted many instances of resistance. The ECJ reacted to this in two ways: one was to endeavour to incorporate within the EU legal order principles derived from the protection of fundamental rights and social rights protected by some national constitutions; the other was to reformulate the standards for the implementation of European norms. For the ECJ, these forms of resistance did not relate to the actual existence of the ‘law of integration’, but merely to the effects it was likely to produce. Accordingly, it acknowledged that the effectiveness of EU law could be secured by less radical means than by substituting European norms for national norms. In a decision that came twenty years after that in which these assertions of authority were first made, the ECJ judged that ‘it cannot (...) be inferred from the judgment in Simmenthal that the incompatibility with Community law of a subsequently adopted rule of national law has the effect of rendering that rule of national law non-existent’, the only obligation imposed on the national court being ‘to disapply that rule’[15]. The rule of primacy of EU law was never meant to merge EU law with the laws of the Member States, but simply to respond to their essential separation and to the resulting need for some link between them. The recent Melki case is another example of the readiness of the Court for softening the rule of primacy stated in Simmenthal[16].

1.3 The dualist conception of integration

These assertions rest on the idea that there are two kinds of legal system, one supranational and the other national, professing to be equally autonomous and sovereign, and that are in fact in a position of proximity and of interdependence. This dualism, which is not to be confused with the ‘dualist’ approach to relations between national law and international law, has become the challenge to be taken up. The autonomy called for by EU law doctrine in the legal literature is based on this belief that integration is a specific form of compromise between these orders, the different expressions of which should be examined and the relational techniques discovered. Such a representation determines the entire epistemology of the discipline. If it is accepted that European integration is dependent upon the coexistence of two types of legal system (each being autonomous and sovereign) it is inevitable their relations will be addressed from two separate standpoints: either from the uniform supervision processes that EU law projects and that produces reactions in national law depending on the specific constraints in each order; or from heterogeneous treatments that national laws apply to the provisions of EU law and that in turn produce reactions in EU law, which is concerned with preserving its autonomy. European legal scholarship arising from public law has generally chosen to adopt the first standpoint by looking at projection phenomena. Private law scholarship seems more attentive to the other standpoint. This leads it to study the phenomena of reception giving rise to what one author calls ‘the legal governance of diversity’[17]. But these are two poles of one and the same approach to integration.

The dualist vision also has important practical effects. How can the gap be filled between the formulation of objectives of integration by EU law and their transposition in the national legal systems? The European Community traditionally responded to this by a strategy of compensation: multiplication of uniform regulations, development of autonomous interpreta­ tions of European rules and, above all, creation of a framework in which national actors are in some sense forced to justify themselves with regard to the objectives of integration. I propose the term ‘cadre de comparution’ (framework of justification) to describe the set of administrative and judicial procedures of control set up or called for by EU law, because their main effect is to expose the national authorities to constant tests of justification before the European instances, before their peers and before their own control authorities[18]. The concept of ‘legitimation by procedure’, coined by Niklas Luhmann, seems well suited to the process of legal integration organized in this way: this concept consists in ‘absorbing conflicts, weakening and exhausting participants, transforming and neutralizing their motives in the course of history in which presentations and commitments are changed into presentations serving to eliminate alternatives’[19]. The fact remains that such efforts are doomed to fail so long as they develop under the dualist postulate, which recognizes that the effectiveness of European norms (including judicial norms) invariably depends in the last resort on the willingness of the national authorities who apply them. This gave rise to a recent change of orientation that could be perceived in EU strategy and consisting in preferring the creation of new systems of control to the intensification of legislative output[20].

For sure, the dualist representation of integration is not without its advantages: based on the principle of autonomy, it protects the integrity of each of the legal systems it brings into relation, it protects their normative coherence and the ideological system on the basis of which they find their legitimacy. However, it also has the effect of enclosing the integration process in an insurmountable paradox: for how can one discipline relations between systems that are recognized to be sovereign and heterogeneous? To overcome this paradox, there is no other solution but to appeal to the will to achieve economic, social and political integration that transcends the legal autonomies. The ideology of integration tends thereby to substitute for the democratic ideology that forms the basis of national legal systems. In other words, the introduction of conditions capable of ensuring the legitimation of integration ‘by procedure’ does not dispense with the need to resort to a principle of legitimacy that ends up affecting the foundations of national systems[21].


Is it possible to account for the integration process by departing from the dualist regime that shaped it in its present forms? That there are forms in EU law that do not match this representation is not in doubt: there are hybrid instruments that combine features of EU law with features of national law, collective agreements which bring together European authorities and national authorities, or incomplete measures operating by reference (renvoi) to national law[22]. However, the proliferation of such forms has in no way called into question the predominant dualist approach. The point is that it is not enough to identify original forms. One must also provide the means to enlighten in some other way the relations between legal systems that are established in the context of integration.

I propose to build a typology of these relations from a distinction drawn between two schemes: the distinction between a normative scheme and an institutional scheme. The normative scheme corresponds simply to the idea that integration aims to create stable, concrete relations between the sets of norms making up the different legal systems. The institutional scheme stems from the idea that integration is also to be understood as a set of relations between the different authorities applying European norms. By combining these two schemes in two basic relational modes (distinction/coordination of legal orders; hierarchy/independence of authorities), we get four possible integration regimes: i) the regime where legal orders are separate and authorities hierarchized, which corresponds to the ‘dualist’ representation just described; ii) the regime where legal orders are coordinated while authorities remain independent, which is the ‘pluralist’ hypothesis; iii) the regime of coordinated legal orders and hierarchized authorities, which is generally termed a ‘federal’ organization; iv) and the regime where legal orders are separate and at the same time authorities are independent, which I will call a ‘conflictualist’ type regime. By examining each of these regimes from the standpoint of European legal integration, two points are worth noting: first, they all find a place in the developments of the positive law of integration; second, the question of the legitimacy of legal integration is posed differently depending on the regime in question.

2.1 Law and politics under the dualist regime

Let me begin from what seems to me a fundamental point: this regime is based on cognitive monism. The legal systems being separate, autonomous and mutually exclusive, each legal system has its own standpoint on the relations between its fundamental norms and the external norms (norms from an outside source). Under the ‘conditions of reciprocal cognitive indeterminacy of legal orders’[23], there are as many perspectives on integration as there are sovereign legal systems. At the extremes of perspectives that can be envisaged lie, at one end, the perspective of EU law which views primacy as a rule with absolute authority and of universal scope within the scope of EU law, and at the other end, the national perspective that considers primacy a pure rule of foreign origin the scope of which can only be relative and that must find a compromise with the hierarchical principles of national law. It is obvious that, from this standpoint, any direct communication between legal systems seems to be excluded. If the rules coincide, if the evaluations concur, this is merely the effect of decisions that are internal to the systems concerned. There is strictly speaking no communication between legal orders that are separate and autonomous. As for assuming the existence of common values, dualism restricts itself to either ethical anticognitivism of positivist obedience (the world of values is unknowable) or to radical axiological relativism (each to his own values).

Under these circumstances, the problem facing dualism is how can one ensure the superiority of EU norms in domestic law when domestic legal systems have their own hierarchical requirements that are indifferent to those contained in EU law. How can the rule of primacy of European norms be internalized? How can one activate the ‘validation relationship’ that must be established between European norms and national norms[24], when the EU legal system does not have the authority to determine the validity of national law?

The answer forged by dualism consists in transforming the rule of normative primacy into an institutional and relational technique. Legal orders being separate, primacy necessarily involves the authorities. Primacy is a norm about power-conferral, by which the Union confers on the national authorities a title to act pursuant to EU law. The problem, though, is that there is no established hierarchy between the EU authorities and national authorities: their relationship corresponds to a regime of separation of powers and of functions[25]. How can a European title be created when the national authorities hold their entitlement solely from the national constitution, which the ECJ is quite prepared to recognize for that matter?[26] It is somewhat misleading, then, to speak of a conferral of power because an authority established within a legal order cannot receive its power from a foreign norm. There remains, then, the possibility of giving to the national authorities ‘European grounds for action’. This is precisely the function of the co-operation mechanisms put in place by EU law and especially the procedure and practice of preliminary reference. The preliminary reference procedure operates as a device for reminding national courts, in the very course of their domestic proceedings, of the reasons for and objectives of integration. In this procedure, proximity becomes such that the ECJ takes on the function of a higher ranking ‘national court’[27], while the national court is commonly characterized as the ‘European ordinary judge’[28]. The strength of the ‘European grounds for action’ must lead the national courts to pre-empt the grounds for action drawn from the national constitution and even, as the case may be, to interpret the norms on which their jurisdiction is based in such a way as to recognize they themselves have the necessary powers to accomplish their mission of implementing EU law[29]. By conferring on the national organs the power to adjudicate as to their own entitlement, the ECJ ‘augments’ their prerogatives, invests them with a ‘European function’, with a ‘quasi-title’ of authority. Now, enhancing the power of a third party by bestowing a title or by formulating grounds for action that substitute for the internal grounds of the actors, that is the very mark of authority, in which the legitimation of legal integration resides[30].

This whole arrangement comes up against a reality test, though. It only holds provided that the national actors adhere to the reasons given by the European authorities. As numerous works of political science have shown, such support usually relies on individual motives, voluntary submission driven by a promise of emancipation or a transfer of loyalty towards what is viewed as a new centre of power and interest[31]. But an authority is only legitimate if it can be recognized collectively and not just individually. Such recognition requires a general shared assumption. That assumption is embodied historically by the project of unification of Europe: the legitimacy of integration relies on a commitment by the actors, a ‘political choice’, a certain representation of European unity[32]. The sociological works of A. Vauchez are aimed precisely at showing that, in the history of European construction, that commitment is not only a theoretical postulate but a social reality[33]. There has gradually formed a transnational community of jurists whose judicial, administrative, political and scholarly functions have proved permutable, conveying a shared culture and constituting a relatively autonomous field. In that lies a great deal of the success of the integration project. But should that commitment weaken or deteriorate, in a changing context, and the entire construction is imperilled. And herein lies the reason for recomposing the integration process, notably around the pluralist hypothesis.

2.2 Law and values in a pluralist context

Although it is sometimes presented as the extension of dualism, pluralism is in fact the exact opposite. Firstly because it breaks with the exclusivism that characterized dualism. Pluralism accepts that legal systems may communicate, dialogue and enter into contact with each other: norms and interpretations circulate. However, the authorities tasked with applying and interpreting those norms remain strictly separate, each of them having its own view of the norms and the relations among norms that are applicable. But, just as there is a throng of viewpoints, there is the possibility of exchanging viewpoints. In the European legal area, authority is irretrievably dispersed[34], but that dispersion is supposed to develop on a bedrock of shared values.

The striking thing is how easily this approach was able to be cast within the context of classical representations of integration. On the one side, as the reach of EU law extended to sensitive areas of internal law and national policy, the absolute authority of EU law postulated by the dualist regime was ever less acceptable; but on the other, the approximation and cooperation procedures that had long been put in place by EU law finally meant that laws and their interpretations coincided increasingly. Under these circumstances, the pluralist argument of ‘the nesting of legal systems’ made it possible, under the cover of a shared foundation of norms and values, to free the national authorities from the constraints of submission, with a view precisely to giving them an incentive to join the dialogue[35]. For, if the norms of competing legal orders are no longer held as anything other than the ‘particular expression of common values’[36], it becomes feasible for every legal actor to take them into consideration and interpret them. Thus a direct confrontation among legal systems is no longer excluded: it takes the form of a simple conflict of principles where fundamental norms of different origins are ‘weighted’. The rule of the primacy of EU law then loses its reason for being; it is pushed into the background and superseded by a method of reconciliation based on the ‘reasonable’ character of the assessments made by the various parties[37]. The general principles of law are tasked with an ‘osmotic function’[38]. At this level, at least, laws are commensurable; what is still not commensurable in the idea of pluralism are the authorities tasked with adjudicating conflicts[39].

Therefore, the problem that is posed in this approach is symmetrical to the problem dealt with by dualism: how can common norms be maintained when the authorities applying those norms are totally independent and there is nothing to guarantee the convergence of their viewpoints? It is this problem that the processes of translation of one law into the other developed by the ECJ and by national courts set out to tackle. For genuine communication among orders to be established it is important not just that norms should be able circulate but also that roles can be swapped. This is what is being put in place in certain recent jurisprudential developments. For example, adopting a conciliation technique, the ECJ accepts to take into consideration the interpretation made by a national constitutional court and to integrate it into its grounds by relaxing its national law review standard[40]. Or alternatively it accepts to delegate to the national authorities a part of its power to construe the rules of EU law in a manner consistent with the principles of European law[41]. Or again, it refrains from settling the legal dispute and prefers to refer to the national court the care of completing its interpretation of the rule in relation to the specific features of its constitutional order[42]. Analogously, the French Conseil d’État accepts to modify the position its system imposes on it in order to ‘translate’ the conflict between a European directive and a constitutional norm into the sphere of review of EU law[43]. In another case, it expressly recognizes it has the power to construe a European directive in respect of treaty-based norms expressing common values of European legal systems, by complying with the court’s interpretation of EU law[44].

Be it via conciliation, delegation, abstention or translation, in all these cases, the court accepts that competing authorities are in a better position than it is to solve the conflicts of interpretation it has to deal with. Better still, it attempts to understand the perspective the authorities take up. In this sense, pluralism is a form of perspectivism: it can vary points of view. But such perspectivism is not relativism, since each standpoint takes a set of common values as its reference norms. Pluralism makes it necessary to assume common supra-legal values that allow interpretations to be ordered and exchanged. It generates a ‘new principle of legitimation’, that is substantive and no longer procedural, axiological rather than formal: pluralism is said to be ‘ordered’ in that it rests on the belief in the existence of a set of higher values and of fundamental rights[45]. Thus the commissaire du gouvernement Mattias Guyomar can declare before the Conseil d’État, that ‘the legal space we share is based on common values: (...) observance of fundamental rights protected within the European framework of which we are all together, European and national courts, the final guarantors’[46].

However, this hypothesis, which leads to equalizing the viewpoints and to removing the problem of the hierarchy of legal orders would, if taken to extremes, lead to giving any court the power to protect and define the contents and bounds of the fundamental principles of European legal area. The reality test that pluralism comes up against is the risk of fragmentation of interpretations. That is why, in the context of integration, pluralism can never be anything but apparent. Some coherence must be restored to the integration process for the benefit of the objectives it conveys. That can only be done by recognizing the ECJ enjoys a privilege in its access to the shared values and in the interpretation of the norms that are supposed to express them. As D. Ritleng recognizes in a fresh study of the principle of primacy: ‘to the full extent that constitutional homogeneity is thus ensured between the national legal order and the EU legal order, the national courts leave it up to the Court of Justice to check that EU acts comply with constitutional principles and values’[47]. The multiplicity of authentic interpreters must resolve itself to accepting there is a single interpreter of the supreme values: a supreme court. The legitimacy of legal integration lies no longer in adhering to the political project of European unity but in the moral authority of a court.

2.3 The idea of an ‘œuvre’ and European federalism

Is the federal hypothesis to be encountered in the process of European integration? This is a demanding hypothesis. It assumes a nesting of norms and interaction among the authorities of competing legal systems. The federalist approach describes legal systems as being both autonomous and interdependent. That interdependence is usually attested by the occurrence of what one might call ‘hybrid norms’. Those who use this hypothesis to study European integration suggest there is no longer any cause to draw boundaries between the national legal orders and the EU legal system as they together form a ‘compound’ legal area[48]. ‘Europe’s compound Constitution’ is materialized in a set of principles, values and even of procedures common to the legal systems[49].

The problem federalism raises is exactly the opposite of that posed by the previous two approaches: here it is not a question of looking for connections but of differentiating between legal orders that are in constant relation. This problem is solved classically by a principle of sharing of powers. Integration is brought about by the allocation of European and national competences. Relations between legal orders are not conceived as the imposition of hierarchies of norms but as the outcome of an allocation of competences. In this way, the conflicts that arise will bring about coordination solutions. A national norm will not be set aside because it is contrary to a norm of EU law that is superior to it, but because the national legislator has exercised his jurisdiction without allowing for the existence of a competing jurisdiction on the part of the European Union that was entitled to govern the situation in question. This type of formulation is found in two important doctrines developed by the ECJ: the implied powers doctrine, which has allowed the powers of the EU to be extended within the international order[50], and the doctrine of pre-emption, by which the recognition of competence reserved to Member States does not prevent the exercise of that competence from being subject to the compliance provisions of EU law[51]. In both cases, it is a matter of justifying an extension of powers by the necessity of building the EU[52].

To understand this type of reasoning, one must accept the idea that the EU system and the national systems form a ‘coherent whole’[53]. Federalism is based on a principle of justification of the globalist type: integration results from the inclusion within a global order of heterogeneous purposes produced by the various legal orders that compete on the European playing field. The integration process ceases to be fragmented among legal systems separated by bulkheads, it is an order composed of special regimes, with their separate and sometimes contradictory objectives, but among which a compromise can be sought. Such a conception at the same time justifies the existence of a mediator whose tasks is to reconcile the different finalités pursued at EU and national level: this is the role of a supreme court, or the judge of integration – the ECJ[54].

However, this hypothesis confronts EU law with a new justification test: for if the orders commune, how can it be explained that one dominates and limits the others? The ECJ itself admits, and long has, that the pursuit of integration justifies ‘intrusions of Community competence into national sovereignty’ even beyond the domains expressly transferred to the Community, and ‘wherever they are necessary’[55]. This model engenders the fear of phenomena of colonization of arrangements consolidated in domains that a priori are outside the scope of application of EU law. The extension of the reach of EU law creates what private law theorists call ‘diagonal conflicts’[56]: by this they mean conflicts of coherence that may arise from effects produced by the teleology of EU law on domains that this law is not supposed to govern (personal status, welfare protection, nationality law, etc.). To solve such a problem, the case law of the United States Supreme Court has developed a doctrine of pre-emption that relies on a certain understanding of the relationship between the part and the whole[57]. Thus, if state power must yield before the power exercised by the federation, it is because state power is entitled to govern within its sphere of competence only a fraction of the situations covered by the powers of the federation.

A similar type of justification has been thematized by some commentators analysing the process of European integration: EU law takes on a corrective function for the national decision-making processes that are trapped in necessarily exclusive and limited frameworks of representation[58]; it introduces to national systems, that are considered as parts of the whole, principles of openness that force them to co-exist; it provides an incentive for national authorities in applying the law to allow, in their assessments, for interests that are not represented in the national legislations, those that come from or are established in the other Member States. The legitimacy of integration relies here therefore entirely on the defence of common interests represented by the legal system of the EU and transcribed by its court. But where do such interests come from? In the absence of any will of the European people to which to ascribe them, it can only be supposed that they transpire from an ‘idée d’oeuvre commune’ that is manifested in the founding treaties and in each revision of them[59]. The advantage of this type of legitimacy is that it does not need to refer to any constituent power, it is enough for it to identify a constituted founding power – the High Contracting Parties of the European treaties[60].

2.4 Conflictualism and the holistic hypothesis

It is hard to imagine that the conflictualist hypothesis can find a place in the framework of the integration process. This scheme begins from a double series of distinctions between legal systems, both in normative terms and in institutional terms. Legal systems are heterogeneous, competing and incommensurable. Under this assumption, EU law does not escape from these relations of reciprocal independence. From the viewpoint of national legal systems, it is relegated to the rank of one law among others. Its position is no different from that of any other foreign law[61]. That being so, the provisions of EU law are part of a discipline similar to private international law: it is the national authorities that choose, unilaterally, to attribute validity and enforceability to these norms to govern domestic legal relations. The rule of primacy is ‘internalized’ but pays a price for its transformation: it becomes a simple rule of conflict which is not designed to impose EU law on national law but, at best, to prevent conflicts of norms by making a space for EU law that national law declines to occupy[62].

The problem facing conflictualism is that of coordination among independent orders. It is the problem Santi Romano thematized under the notion of ‘relevance of one order for another’[63]. As there are no ‘third norms’ – common values that can serve as a connection as provided for by the hypothesis of ‘ordered pluralism’ – coordination relies exclusively on arrangements for establishing relations specific to each system. Such processes are found in European law. EU law recognizes the process of renvoi by which it spontaneously refrains from settling a given matter, leaving it for national rules to apply. Thus, in its Courage decision, concerning the effect of European competition rules on compensation cases that are in principle governed by national contract law, the ECJ decided that EU law was not entitled to govern the subject matter[64]. It accepted that each national private law system could lay down its own rules of coordination with this part of EU law (a form of ‘reversed primacy’). Reciprocally, it happens that EU law opens up a ‘void’ within which the rules of national law are immediately and directly applied[65]. Moreover, there are many instances of horizontal recognition between national legal orders and of coordination among national legislations. For instance, under the ‘new approach’ in the realm of harmonization of national laws, the European rule of harmonization provides for the simultaneous and disjunctive application of European provisions and national provisions[66].

In any event, there is not strictly speaking any merger of legal systems in a global and third order but simply cohabitation within each individual order of rules of different origins. Between them, conflict is avoided, primacy does not come into play, and relations of separate applicability are established. The orders are circles that only touch in the event of infringement of the fundamental values of one of them, what private lawyers call ‘the imperative norms of national public order’. Public order is therefore the ‘counter-limits’ opposed to the limits granted by national law for incorporation of EU law. In such a regime of coordination, the problem is to strike a balance between maintaining the autonomy of the national order and incorporating foreign norms. Legitimating integration then depends on the possibility of maintaining that balance without jeopardizing the objectives of integration. To this end, EU law has multiplied the procedures of participation and deliberation of the national executive authorities: the legitimation of integration depends on proceduralization. National courts and administrations are involved in ‘complex networks’ the purpose of which is to foster the development of ‘European’ solutions to locally posed problems[67]. These networks compose what are now called ‘European governance’ (‘deliberative administrative committees’ or ‘courts diplomacy’).

Even so those networks do not protect against the risk of differentiation. Nothing here guarantees that interactions will occur; the mediators are vulnerable to the capture of national interests; national arrangements are not sufficient to ensure the overall coherence on which the defence of the common good of integration depends. It is necessary therefore to posit a principle that transcends the particular instances of judgment. What might be the ‘title of relevance’ of EU norms for national laws? Short of any transcendant instance —political project, supreme judge or ‘idée d’oeuvre’— the solution lies in a general principle of distribution that attributes to each order a specific place within the common framework. Legitimacy in a conflictualist regime is of a holistic type: it is for each order to interiorize the common good of integration, for each authority to find its place in the set of networks composing the integration process. This principle assumes that the enforcement authorities develop a broader ‘European’ vision of legal situations: of the type, say, recommended by the ECJ in the Grunkin-Paul case concerning the German authorities’ refusal to recognize a child who was a German national but born in Denmark and living in that country under the double name of ‘Grunkin-Paul’ which had been registered by the Danish authorities[68]. The political function of the enforcement authorities and especially of national courts becomes decisive in this case: it is through them and through the constraints they exert on public and private authorities that laws are Europeanized within each national system[69].


By way of conclusion, it is possible to present the process of European integration as the framework of a multiplicity of justification tests to which EU law has to face. Each test can only be understood in respect of a certain relational arrangement among legal systems that is defined within a general grammar of relations between legal orders. In the actual course of integration, the dominant dualist regime made do with the presence of other regimes that became discretely and progressively established. That may give rise, depending on the way in which similar situations are handled, to different and sometimes contradictory solutions. Moreover, it is always possible to set oneself in the perspective of one of these regimes to evaluate the justification tests instituted by the others. For example, it is possible to understand from a conflictualist standpoint the problem of grounds for action posed in the dualist regime; but then it will neither be envisaged nor solved in the same way: the dualist solution of imposing European grounds by constraint of cooperation between the national court and the EU court is contrary to the constraints of independence of the conflictualist regime (which will prefer a solution of internalization of European grounds in the jurisdiction of the national court). Likewise, the pluralist solution of translation of one law into another may be understandable from a dualist standpoint as the outcome of successful judicial cooperation, but it is excessively risky cooperation because of the independence it seems to offer national courts.

The second result of this study is that, whatever the regime under which it is contemplated, the legal integration process is subjected to a ‘legitimacy pathway’. A pathway that comprises three separate sequences: (i) to resolve the tensions engendered by its relations with national legal systems, EU law is compelled to create relational mechanisms (power-conferral, translation processes, distribution of competence, relevance processes); (ii) mechanisms that only operate properly if they rest on a principle of justification (legitimation by authority, by values, by the whole or by procedures) which is never quite adequate for the project of effective integration; (iii) this leads to postulating the existence of a general assumption having legal effects (political project, supreme judge, idée d’oeuvre, general principle of distribution of roles). This entire pathway has one addressee and a single goal: the national authorities, that must be persuaded to apply EU law. That may well be a good definition of integration: a set of duly justified, technical processes of relation between legal systems, in which the executive authorities must feel authorized to believe.

[1] This study takes on board the entry into force of the Lisbon Treaty replacing the European Community by a new European Union that inherits its competences, its powers and its acquis. It employs the expressions ‘EU law’ and ‘European’ for what were, before the treaty came into force ‘Community law’ and ‘Community’ and only keeps the latter term for the parts involving historical analysis.

[2] See M. Troper, ‘L’Europe politique et le concept de souveraineté’, in O. Beaud, A. Lechevalier, I. Pernice and S. Strudel (eds), L’Europe en voie de constitution. Pour un bilan critique des travaux de la Convention, (Brussels: Bruylant, 2004), p. 117.

[3] J. Ziller, ‘La Constitution’, in J.-B. Auby (ed.), L’influence du droit européen sur les catégories du droit public, Dalloz, Paris, 2010.

[4] Conseil const., décision no 2006-543 DC, 30 nov. 2006, Loi relative au secteur de l’énergie.

[5] The Constitutional Court thus judges that the legislation transposing European norms, and therefore the norms too, must comply with the ‘principles inherent to the constitutional identity of France’ (Conseil const., décision no 2006-540 DC, 27 juil. 2006, Loi relative au droit d’auteur et aux droits voisins dans la société de l’information).

[6] A. Pellet, ‘Vous avez dit ”monisme“ ? Quelques banalités de bon sens sur l’impossibilité du prétendu monisme constitutionnel à la française’, in L’architecture du droit. Mélanges en l’honneur de Michel Troper, (Paris: Economica, 2006) p. 828.

[7] As Michel Troper has shown, under some circumstances, such protection may even be greater than the protection afforded to constitutional norms themselves (‘La théorie constitutionnelle et le droit constitutionnel positif’ (2000) 9 Les Cahiers du Conseil constitutionnel).

[8] Italian Constitutional Court, 27 December 1973, Frontini, no 183.

[9] See G. Itzcovich, Teoria e ideologie del diritto comunitario, (Turin: Giapppichelli editore, 2006) p. 210 ff.

[10] Case 6/60, Humblet v. Belgian State [1960] ECR 559.

[11] Case 6/64, Costa v. ENEL [1964] ECR 585.

[12] P. Pescatore, ‘Les objectifs de la Communauté européenne comme principes d’interprétation dans la jurisprudence de la Cour de justice. Contribution à la doctrine de l’interprétation téléologique des traités internationaux’, in Miscellanea Ganshof Van der Meersch, t. II, (Brussels, Paris: Bruylant, LGDJ, 1972) p. 325 ff.

[13] See, on this arrangement, K. Caunes, Le principe de primauté du droit de l’union européenne. Contribution à l’étude de la nature juridique de l’Union européenne et des rapports de système européens, Thesis, European University Institute, Florence (2009).

[14] Case 106/77, Amministrazione delle Finanze dello Stato v. Simmenthal SpA [1977] ECR 629.

[15] Case C-10/97, Ministerio delle Finanze v. IN.CO.GE’90, [1998] ECR I-6307. MELKI.

[16] C-188/10, Melki [2010]: ‘In so far as national law lays down an obligation to initiate an interlocutory procedure for the review of constitutionality, which would prevent the national court from immediately disapplying a national legislative provision which it considers to be contrary to EU law, the functioning of the system established by Article 267 TFEU nevertheless requires that that court be free, first, to adopt any measure necessary to ensure the provisional judicial protection of the rights conferred under the European Union’s legal order and, second, to disapply, at the end of such an interlocutory procedure, that national legislative provision if that court holds it to be contrary to EU law’ (§ 53).

[17] J. Porta, La réalisation du droit communautaire. Essai sur le gouvernement juridique de la diversité, (Paris: LGDJ, 2007).

[18] See ‘Le rôle constitutionnel de la Cour de justice des Communautés européennes tel qu’il se dégage de sa jurisprudence’ (2008) RTDE, 29, 35-36.

[19] N. Luhmann, La légitimation par la procédure, trad. de l’allemand par L.K. Sosoe, (Paris: Les Editions du Cerf, 2001) p. XLIX.

[20] See on this, J. Dutheil de la Rochère and J.-B. Auby (eds), L’exécution du droit de l’Union, entre mécanismes communautaires et droits nationaux (Brussels: Bruylant, 2009).

[21] See also, M. Troper, ‘L’Europe et les principes’ (1992) 20/21 Droit et Société.

[22] See on this point the present author’s, ‘Autonomie et antinomie du droit communautaire : la norme communautaire à l’épreuve des intérêts et des droits nationaux’, (2004) no spécial, 199 Petites affiches, 4.

[23] This is the premise of the dualist approach in international law. See C. Santulli, Le statut international de l’ordre étatique. Etude du traitement du droit interne par le droit international (Paris: Pedone, 2001).

[24] On this report, see J. Combacau, Le droit des traités (Paris: PUF, 1991) p. 84.

[25] In Case 6/64, Costa v. ENEL [1964] ECR 585, the Court speaks of ‘a clear separation of functions).

[26] Case 106/77, Amministrazione delle Finanze dello Stato v. Simmenthal SpA [1978] ECR 629, pt 21; Case C-430/93 and C-431/93, Van Schijndel [1995] ECR I-4705, pt 22.

[27] See G. Morelli, ‘La Cour de justice des Communautés européennes en tant que juge interne’, in G. Morelli, Studi sul processo internazionale, (Milan: Giuffrè Editore, 1963) p. 93.

[28] A characterization that national courts henceforth readily make their own (most recently, Conseil d’État, Ass., 30 oct. 2009, Perreux, no 298348).

[29] See Case C-432/05, Unibet [2007]; Case C-355/04 P, Segi [2007].

[30] On the connection between the concept of authority and that of reason to act, see H.L.A. Hart, Essays on Bentham. Studies in Jurisprudence and Political Theory (Oxford: Clarendon Press, 1982); J. Raz, The Morality of Freedom (Oxford: Clarendon Press, 1986).

[31] On the definition of integration as a process of transfer of loyalty and activities towards a new power centre, with jurisdiction within nation states, see E.B. Haas, The Uniting of Europe. Political, Social and Economic Forces (1950-1957), (Stanford: Stanford University Press, 1958).

[32] See G. Itzcovich, ‘Ordinamento giuridico, pluralismo giuridico, principi fondamentali. L’Europa e il suo diritto in tre concetti’ [2009] Diritto pubblico comparato ed europeo.

[33] See A. Vauchez, ‘Droit et politique’, in C. Belot, P. Magnette and S. Saurugger (eds), Science politique de l’Union européenne (Paris: Economica, 2008).

[34] N. Walker, ‘Sovereignty and differentiated integration in the European Union’, (1988) 4 European Law Journal at 361-362.

[35] On this argument and its implications, see P. Brunet, ‘L’articulation des normes (analyse critique du pluralisme ordonné)’, in J.-B. Auby (ed), L’influence du droit européen sur les catégories du droit public, Dalloz, Paris, 2010.

[36] B. De Witte, ‘Droit communautaire et valeurs constitutionnelles nationales’ (1991) 14 Droits 87, at. 90.

[37] Case C-112/00, Schmidberger [2003] ECR I-5659.

[38] D. Simon, ‘Les principes en droit communautaire’, in S. Caudal (ed.), Les principes en droit (Paris: Economica, 2008), p. 303.

[39] See N. Walker, ‘The Idea of Constitutionalism Pluralism’, EUI Working Paper LAW, no 2002/1, Florence, 2002 (

[40] See Case C-36/02, Omega [2004] ECR I-9609.

[41] In a decision on application of the directive on the right to family reunification of minor children of third-country nationals, the Court ruled that ‘while the Directive leaves the Member States a margin of appreciation, it is sufficiently wide to enable them to apply the Directive’s rules in a manner consistent with the requirements flowing from the protection of fundamental rights’ (Case C-540-03, Parliament v. Council [2006] ECR I-5769, pt 104). See also Case C-101/01 Lindqvist [2003] ECR I-12971.

[42] An example has been given in a matter concerning the granting of a regional aid measure by the Basque autonomous community in Spain. In this case, the Court recognized that, to interpret Community law on state aid, allowance must be made for ‘institutional, procedural and economic autonomy’ enjoyed by the infra-state authority concerned, the Basque autonomous community. The Court held it was for the national court to check whether such autonomy existed in the case at hand according to the criteria laid down by national law alone (Joined Cases C-428/06 to C-434/06, UGT-Rioja et al. [2008] ECR I-6747).

[43] Conseil d’État, Ass., 8 févr. 2007, Arcelor, req. no 287110.

[44] Conseil d’État, Sect., 10 avril 2008, Conseil national des barreaux e.a., req. nos 296845, 296907.

[45] See also P. Brunet, ‘L’articulation des normes (analyse critique du pluralisme ordonné)’, in J.-B. Auby, op. cit.

[46] Conclusions sur Conseil d’État, Sect., 10 avril 2008, Conseil national des barreaux e.a., RFDA 2008, p. 575.

[47] See also, D. Ritleng, ‘De l’utilité du principe de primauté du droit de l’Union’ (2009) 4 RTDE.

[48] I. Pernice, ‘Multilevel constitutionalism in the European Union’, (2002) 27 European Law Review 511; I. Pernice and F. Mayer, ‘De la constitution composée de l’Europe’ (2000) 36(4) RTDE 623.

[49] See also H. Gaudin, ‘Révision des traités communautaires, révision des constitutions nationales : recherche sur la symétrie d’un phénomène’, in Mélanges en hommage à Guy Isaac. 50 ans de droit communautaire, tome 2, (Toulouse: Presses de l’Université des Sciences Sociales, 2004), p. 541.

[50] It has been possible to show that, to solve the question posed in the AETR ruling (Case 22/70, AETR [1971]), the Court could have chosen to apply its rule of primacy: primacy of Community norms over international obligations of Member States in the domain of Community law. It is remarkable that, to avoid formulating the conflict of rules, the Court chose to ground its ruling on a model of coordination of powers (See the study by P. Eeckhout, in M. Poiares Maduro and L. Azoulai (eds), The Past [amp ] Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Oxford: Hart Publishing, 2010).

[51] See, for example, Case C-415/93, Bosman [1995] ECR I-4921; Joined Cases C-76/05 and C-318/05, Schwarz [2007] ECR I-6849.

[52] On the idea of ‘necessity’ involved in the argument of implied powers, see G. Tusseau, Les normes d’habilitation (Paris: Dalloz, 2006).

[53] O. Beaud, Théorie de la Fédération (Paris: PUF, 2007) p. 191-192.

[54] This role of interface and mediator appears clearly, for example, in Viking Line and Laval (Case C-438/05, Viking Line [2007] ECR I-10779; Case C-341/05, Laval [2007] ECR I-11767).

[55] Case 23/59, De gezamenlijke Steenkolenmijnen in Limburg/Haute Autorité [1961] ECR 3, p. 46. See on this point P. Pescatore, ‘Fédéralisme et intégration : Remarques préliminaires’ (1973), republished in Etudes de droit communautaire européen 1962-2007 (Brussels: Bruylant, 2008) p. 450. The Law of Integration (trans. C. Dwyer (Leiden: Sijthoff, 1974) p. 27.

[56] See C. Joerges, A. Furrer and O. Gerstenberg, ‘Challenges of European Integration to Private Law’, Collected Courses of the Academy of European Law, vol. VII, Book 1 (1996), (Netherlands: Kluwer Law International, 1999), p. 281.

[57] In McCulloch v. Maryland, 17 US 316 (1819), Justice Marshall wrote of the solution to the conflict of powers: ‘The difference is that which always exists, and always must exist, between the action of the whole on a part and the action of a part on the whole—between the laws of a Government declared to be supreme and those of a Government which, when in opposition to those laws, is not supreme’. See generally on the question of pre-emption in US constitutional law, L.H. Tribe, American Constitutional Law (New York: The Foundation Press, New York, 2000) 3rd edn.

[58] The argument is found in C. Joerges and J. Neyer, ‘”Deliberative Supranationalism“ Revisited‘, EUI Working Paper Law, 2006/20; M. Poiares Maduro, ‘Reforming the Market or the State? Article 30 and the European Constitution: Economic Freedom and Political Rights’ (1997) European Law Journal, 55; S. Weatherill, ‘Pre-emption, Harmonisation and the Distribution of Competence to Regulate the Internal Market’, in C. Barnard and J. Scott (eds), The Law of the Single European Market. Unpacking the Premises (Oxford: Hart Publishing, 2002) p. 41. See also the critical examination by A. Somek, The Argument From Transnational Effects I: Representing Outsiders Through Freedom of Movement, The University of Iowa College of Law, No 09-23, May 2009 (available at

[59] P. Pescatore, Le droit de l’intégration, (Brussels: Bruylant, 2005) republished, p. 41. The Law of Integration (trans. C. Dwyer (Leiden: Sijthoff, 1974).

[60] See on this point the study by G. Marti who undertakes to transpose Maurice Hauriou’s institutionalist arguments to the study of EU law: Le pouvoir constituant européen, PhD Thesis, Université de Nancy 2 (2008).

[61] On this hypothesis see C. Joerges, ‘Rethinking European Law’s Supremacy’, EUI Working Paper Law, No 2005/12.

[62] On this model, in Italian legal theory, G. Itzcovich, Teoria e ideologie del diritto comunitario (Turin: Giapppichelli editore, 2006) p. 391.

[63] Santi Romano, L’ordre juridique (1918), trad. de l’italien par L. François et P. Gothot, (Paris: Dalloz, 2002).

[64] Case C-453/99 Courage et Crehan [2009] ECR I-6297.

[65] See for example the Community Regulation on the European Economic Interest Grouping. The Regulation’s article 2(1) provides: ‘Subject to the provisions of this Regulation, the law applicable, on the one hand, to the contract for the formation of a grouping (...) and, on the other hand, to the internal organization of a grouping shall be the internal law of the State in which the official address is situated, as laid down in the contract for the formation of the grouping’. In its European Information Technology Observatory judgment of 18 December 1997, the ECJ recognized the Regulation was incomplete on the question of the grouping’s business name (C-402/96, Rec. p. I-7515).

[66] See J.-S. Bergé, ‘Le droit d’une ”communauté de lois“ : le front européen’, in Mélanges en l’honneur de Paul Lagarde. Le droit international privé : esprit et méthodes (Paris: Dalloz, 2005) p. 113.

[67] See C. Joerges, ‘Sur la légitimité d’européaniser le droit privé. Plaidoyer pour une approche procédurale’ (2004) Revue Internationale de Droit Economique 133, at 167.

[68] Case C-353/06, Grunkin and Paul [2008] ECR I-7639.

[69] See also G. Canivet, ‘Le fédéralisme judiciaire : étude comparée du fédéralisme juridictionnel américain et de l’architecture juridictionnelle de l’Union européenne’, in Liber Amicorum Guy Horsmans (Brussels: Bruylant, 2004) p. 133.