The community of jurists and philosophers of law recently lost an unusual figure in the person of Neil MacCormick. The longtime Regius Professor of public law at the University of Edinburgh and member of the European Parliament belongs to the ranks of those who have made profound contributions to renewing our approach to European law and, in particular, to the relationship between the European Union and member states. Among his many contributions to the general theory of law was his view that relations between the EU legal order and national legal orders can be explained in terms of “pluralism.” The term does not refer strictly speaking to a doctrine. It is used today by many magistrates to account for the complexity of relationships between legal orders, which appear to be simultaneously connected and independent of one another —relationships about which it is said that, “as with living organisms, the separation and integration of tasks is coordinated.”  The point is to identify a process whereby legal orders are harmonized without being fused. This position offers a third way to the well-known alternative in international law between dualism (in which the international legal order and national law are independent and do not interact with one another) and monism (in which international and national law form a single legal order). Pluralism is particularly useful in explaining the relationship between orders in European law, given the latter’s aspiration to be integrated into national legal orders. This theoretical position would appear to have even received an official blessing in French positive law. Initially, the Constitutional Council, invoking article 88-1 of the French constitution, required the legislature to transpose precise and unconditional EU directives, except in instances expressly forbidden by the constitution or when such a transposition would threaten “rules or principles inherent to France’s constitutional identity.”  With this decision, the Constitutional Council in effect determined that it was authorized to interpret EU directives. This decision was highly innovative for at least two reasons: first, it demonstrated the Council’s desire to return, as much as possible, to the jurisprudential position laid out in 1975, according to which the constitutionality of laws must be determined exclusively by their conformity to the constitution, rather than to treaties (despite the fact that article 55 of the constitution explicitly states that the authority of treaties exceeds that of the laws); second, it accorded European law a special position vis-à-vis international law.  Following the decisions of the Constitutional Council, the French Council of State recently determined that matters relating to the constitutionality  or conventionality  of precise and unconditional directives could be referred to it and that henceforth, before a question could be referred to the Court of Justice for a preliminary ruling, it was incumbent on it to verify that the directive was not incompatible with internal constitutional provisions or with the European Convention on Human Rights. To justify this radically new situation, the public magistrate in the case (now known as the “public rapporteur”) defined his position as “ordering pluralism.” In short, pluralism refers to a quasi-ideal situation in which the irreconcilable has been reconciled —in which several separate and distinct legal orders coexist peaceably, even harmoniously, while managing to establish cooperative relationships.
Yet at the risk of being a spoilsport, there is reason to be skeptical of such enthusiasm. To begin with, this description of the relationship between legal orders rests on several concepts that are poorly defined, so one is never quite sure that the terms used to describe the relationships between legal orders are being used in the same way. Moreover, this description of how things are often looks more like a prescription or a justification of how things should be. In short, the equitable cooperation between judges in the production of norms is something that is more desired than observed.
Too many words
The point is not to reiterate the fact that the word “pluralism” has multiple meanings in the realm of politics, anthropology, sociology, and law. The so-called pluralist thesis, as it relates to the coordination of internal and the EU legal orders, does not really suffer from this variety of meanings. One might even argue that the term is well served by this fact, given that all meanings tend to give a positive connotation to the idea of “plurality” in general (to the point of making it an absolute desideratum as such).
We will dwell here on two concepts: legal order and interlocking. Reduced to its simplest form, the pluralist thesis contends that legal orders are at present deeply interlocked.
As used by the proponents of pluralism, the term “legal order” would appear to refer to a set of norms. Yet in fact the term is ambiguous. It can refer both to texts and to the meaning of texts. There is a fairly traditional and formal understanding inherited from the nineteenth century that holds that meaning lies in texts like a pit in a piece of fruit. Yet in law, texts only acquire legal meaning when they have been granted it by a duly authorized power. This is what we mean by the interpretation of legal texts. It entails the exercise of considerable power beyond the simple discovery of truth.  On the basis, one can distinguish texts—the formal sources—from the meaning of texts, i.e., the norms themselves. Consequently, a legal order is a set of meanings that interpreters attribute to particular texts.
These interpreters are first and foremost judges. Their discretion is all the greater in that they find themselves at the summit of positive law. However clear or precise the texts may be, no one can force judges of the last resort to choose one interpretation over another (since no interpretation derives more from one text than another and the question of whether an interpretation ever “derives” from a text at all is largely irrelevant). Consequently, it is judges who, through their interpretations, are the true source of law and the elements that constitute a legal order.  It follows that the harmonization of legal orders really means the harmonization of the meanings that judges attribute to texts. Therefore, the harmonization of legal orders is not, as pluralism’s proponents would have it, a spontaneous process, but rather the result of voluntary choices that judges make.
These choices can be explained in two different ways. For some, they result from the fact that interpreters share the same values and are thus led, on this basis, to convergent solutions. This is the view that is broadly shared by pluralism’s proponents, who do not hesitate to claim that the harmonization of legal orders represents the emergence of a distinct value system. This is a way of expressing their view that this order has arisen independently of them—that it results, if not from the nature of things, at least from an autonomous historical and progressive process.
This thesis presupposes that values can be known and thus that agreement on values determines agreement on meanings. Yet it results in an internal contradiction: it holds that interpreters are free to determine the meaning of texts when producing norms, but that this very freedom is limited by values, which imply specific interpretive choices. Judges are thus simultaneously free and just, but more just than free.
Another explanation is conceivable. Less exposed to metaphysics and moral cognitivism, it holds that these choices are shaped less by values than by constraints related to the circumstances in which they are made. It would indeed be absurd to believe that because they enjoy considerable interpretive freedom, the top judges in various legal orders are sealed off from their environment, or, worse, that there liberty is so great that they are under no obligation to consider other interpretations. Quite the contrary: their immense freedom requires them to consider other positions, precisely to show that their decisions are rational and not arbitrary. But this in no way implies that they must conform to positions endorsed by other judges; nothing and no one can make them do so. In these circumstances, it can be very useful, if not absolutely necessary, to appeal to values—whether to feign a convergence of perspectives, to highlight agreement, to conceal the real reasons behind interpretative choices that are incompatible with other norms, or even to attempt to limit the discretion of other interpreters by pressuring them to justify their decisions in terms of these values, even when the latter cannot be forced to do so. Because values—freedom, democracy, security, the rule of law, etc.—are always expressed in very general terms, agreement can always be found at a formal level, even when, at a substantive level, it is elusive. Consequently, the idea that judicial decisions are based on values that judges allegedly share in common amounts to confusing the effect with the cause and creates the illusion that one has explained the very thing one has precluded oneself from understanding.
Yet with regards to the relationship between European and internal law, a judge’s interpretive choices are determined by an alternative that is as simple as it is radical. On the one hand, the CJEU has, since the judgment Van Gend en Loos (1963)—followed by the judgments Costa c/ Enel (1964), Internationale Handelsgesellschaft (1971) and Simmenthal (1978)—repeatedly affirmed the principle of the primacy of European law over that of member states, which is accompanied by the uniform application of European law over the European Union under the CJEU’s exclusive control. On the other hand, member states have repeatedly asserted their sovereignty and mastery over their internal affairs. Their internal jurisdictions have no choice, if they are to appear to adhere to this principle, but to insist that the constitution is the supreme norm in internal matters, so that no one can ask a judge to rule on the basis of a different norm, even an international one aimed at protecting fundamental rights.  The result is that if judges are to lend credence to the thesis that they are not at war with one another, but striving for the same goals, they have no other solution than to appeal to values and to pretend to speak the same language: consequently, not only does the law they create appear objective (if not predictable) and correspond to accepted notions of what constitutes “true” law, but it also allows judges to assert that they are in “dialogue” with one another.
The thesis of ordering pluralism rests on a second claim: legal orders are closely interlocked and interact with one another with no possibility of establishing a hierarchy between them. But the phrase “the interlocking of legal orders” is far from clear. On the face of it, this phrase seems to refer to two very different situations.
On the one hand, one can speak of interlocking or interaction to describe formal relations between legal orders. The term thus describes not so much the content of norms as their form or mode of production. For instance, orders A and B can be considered to be in a relationship of interlocking or interaction if an authority in order A can determine the validity conditions of a norm decreed by an authority in order B. But can one explain the relationship between the EU and member state orders in these terms? The answer is far from clear. Undoubtedly, norms have direct effects on internal law and these norms replace internal norms. When they conflict, EU norms often trump internal norms. Yet the fact remains that the organisms that apply EU norms are subject to internal law, and no order can determine the validity of another order’s norms. Thus neither EU nor internal judges can, on their own authority, invalidate or even repeal the norms of another order. In short, neither member-state judges nor the CJEU have the authority to declare invalid a norm belonging to a legal order than the one whose laws it is their responsibility to apply.Clearly again, the Simmenthal case—yet again—requires a specific conduct of legislators as well as judges. One could say that on its own, it achieves a formal interlocking of legal orders, in so far as it presumes to validate internal norms. Yet even if one grants the Simmenthal case this much scope, this presumption cannot, from an internal perspective, be implemented unless member-state jurisdictions recognize it. And if these jurisdictions placed European law above internal law, on the grounds that the former has invalidated the latter, interlocking would indeed consist in a hierarchy of norms and thus of orders. The alternative is clear: either legal orders are in effect interlocked, but because one prevails over the other, so that interlocking consists in a hierarchy between orders; or, they are not formally i interlocked, but independent of one another, in which case describing them as interlocked is simply false. Furthermore—and it is primarily in this sense that the expression is used—one can speak of the interlocking of legal orders not to refer to a formal relationship between orders but to emphasize the material (i.e., substantive) proximity of the norms that constitute different orders.  In other words, interlocking pertains to content, not form. In this respect, the situation is hardly new: for years, member-state jurisdictions have been involved in a kind of race, as they invoke their standard norms in order to have access to the same legal instruments as the EU or conventional order. It would be pointless to mention all the often comical cases in which uncertainty and hesitation have called into question solutions that were long taken for granted. One could also distinguish the various adaptation strategies by which each jurisdiction attempts to preserve its own autonomy, whether by appropriating a category in order to give it a different content or by giving an identical content to different category. One could even point out that in many domains (competition law, public contracts, etc.), the question of whether texts originate in European or internal law is largely meaningless, as the weight that judges give to a European as opposed to internal reading of sources depends largely on their personal preferences.
Even so, it is one thing to point out that two norms belonging to two different legal orders have the same content, and quite another to say they form a whole. In the first instance, one is pointing out a fact. In the second, one is making a judgment. Strangely, those who like to emphasize that the norms of the EU and member-state legal orders are “interlocked” often refrain from explaining why they feel justified in making the leap from observing an overlap to positing a necessity. Without such a norm, one ends up reasoning as if law could be defined by its content and as if knowing the content of norms were sufficient to deem them legal. But this amounts to confusing law with morality and taking one’s personal values for objective law. We can now take stock of the theoretical presupposition upon which pluralism is based: its definition of law holds that the form of norms is as important as their content. Yet—and here lies the difficulty—the pluralist thesis never completely abandons formal criteria: its proponents do not say that all “identical” norms form a coherent whole as such and on this basis alone. On the other hand, if norms with identical content are produced by authorities invested with the same power and seeking the same ends, one must conclude that—in some circumstances—these norms amount to a common law. Yet each time it is used, this material criterion raises many difficulties. How does one recognize the proximity of two norms? What rule for recognizing this proximity should be used and where does one find it? In short, what conceptual criteria will allow us to affirm with any certainty that two norms that have the same content belong to the same legal order? Or must one conclude that even if they do not belong to the same legal order, they fit in to the same value system? But would one not then be forced to admit that there exists a supra-legal value system that allows us to identify norms which, simply because they conform to these values, we are obliged to consider as legal?
One can thus see the extent to which the thesis of the interlocking of legal orders contains within it a number of value-laden claims: it is not only a description of the interlocking of orders, it is also a value judgment, which holds that they must be considered as interlocked because the values that they protect provide ample justification for it. In other words, it is important to understand that cooperation between judges is as prescriptive as it is descriptive.
The Obscure Object of Desire: Cooperation between Judges
If one admits that cooperation between judges is a fact, what form does it currently take? An examination of recent cases by the French Council of State can help us understand. Cooperation presupposes that member state judges resort to two interpretive techniques: conform interpretation and analogy.
After the Constitutional Council determined that legislators have a constitutional obligation to transpose directives (under certain conditions and within certain limits),  the Council of State declared that it was within its jurisdiction to determine the conformity of directives with the Constitution or the European Convention of Human Rights (1950) when a challenge to a regulative action seeking to transpose this directive was brought before it.  Beyond the question of legal technique, the political implications of such a determination are easy to grasp: it amounts to the Council of State placing the interpretation of European law within its jurisdiction—and thus interpreting the meaning of texts that it alone is in a position to apply.
Yet it is well known that the CJEU possesses—as it has, invoking Article 267 of the Treaty on the Functioning of the European Union (ex TCE article 234),  frequently recalled—the sole power to interpret the EU order and that this power requires member state judges to abstain from interference. Consequently, once member-state judges compare the ultimate norms of their legal order with those of the EU, they have no choice but to make them as compatible with one another as possible, lest they be forced to conclude the existence of a conflict of norms that they could never be guaranteed to overcome, as the Court of Justice has, for its part, a power to interpret these norms and to organize their relationship differently. One way of doing so is to use the technique that is commonly known in law as “constructive,” “harmonizing” or “conform” interpretation, which consists in reading a text so that its meaning comes as close as possible to the meaning of another text, to the point of conforming with it. This “interpretive method” comes no closer than any other to attaining a text’s true meaning. It is chosen from a range of other techniques not because it alone offers access to the truth, but because it is the product of institutional constraints from which judges cannot extract themselves, given the power configurations in which they operate.
It is clear that that power of interpretation that member-state judges believe that they possess when they are weighing constitutionality or conventionality of EU directives is a power to participate in the production of the meaning of European law, and not a power to nullify an EU action by invoking a national constitution or the European Convention on Human Rights. But this authority is not as extensive as the CJEU’s: it is, at best, a power to offer interpretations of European law from the standpoint of the EU’s or national law’s ultimate norms; at worst, it is a power to adapt internal law to the exigencies of European law with the assistance of conform interpretation.
It is thus not the pursuit of common values, but the systemic constraints to which the power of interpreting formal sources is subject that explains why internal judges resort to conform interpretation.  To put it differently, even when the values that judges seek to protect require them to choose the technique of conform interpretation, these values are in any case of far less weight than the imperative of organizing and distributing the power of interpretation within the European Union.
One could add that the implementation of the technique of conform interpretation requires judges to have frequent recourse to analogies and broad interpretations, as it on these grounds that judges can claim that the content of laws is identical. Naturally, these analogies and broad interpretations cannot be verified and obey no logical principles. But they are very useful, as they justify situations in which judges move from one order to another, yet without imposing one order on another or fusing two orders, at least not formally.
It is precisely to this end that analogy is used in Mattias Guyomar’s conclusions regarding the Arcelor judgment, notably regarding the “translation” that he requests of the Council of State when it hears arguments that an EU directive has overlooked a constitutional disposition or principle.  Thanks to analogies, one can create the illusion of a normative order that exists prior to a judge’s intervention and that the latter simply organizes. In short, affirming the existence of a community of values is all the more necessary for judges, as it allows them simply to dismiss the question of what order the ultimate norm that underpins the norms they create belongs to. This argument follows the same strategy as the moral cognitivism that inspires it: in affirming that two orders share the same values, it presupposes that these values exist and that judges simply have to recognize their identity to grasp their objectivity.
Thus, ordering pluralism implies recognizing the superiority of European law as a value if not as a reality, and, in the name of ordering pluralism, internal judges delegate to the CJEU the task of brushing aside internal law when conflicts arise, even while claiming to cooperate with it. Moreover, when speaking of cooperation, judges may also be hoping to require the CJEU if not to accept, at least to listen to their suggested interpretations. In this respect, member-state judges, because they have the most to lose from the federalization of the European Union, have every interest in convincing the court that pluralism can be ordered as much by national as supranational judges. The “judges’ dialogue” upon which the principal parties congratulate themselves can thus be analyzed in a way that contradicts the latter’s views. The “dialogue” metaphor, which is reassuringly ecumenical—there is a judges’ dialogue like there is religious or cultural dialogue and the world is both plural and singular—is as prescriptive as it is descriptive. Even when it is used simply to describe the exchanges and influence-peddling that occurs between national and supranational courts, its imprecision is a key that can open any lock. Yet it at least has the undeniable political virtue of reconciling the primacy of European law, which is at last recognized, with the primacy of internal constitutional law, which is still preserved. It is understandable that judges so readily endorse it. Careful consideration suggests there exists neither a dialogue—from which a coherent and peaceful legal order, capable of creating a new common law, would arise—nor a war, but only interpretive choices that jurisdictions make according to a hierarchy that is not formal but value-laden. Its ultimate norm, which allows national judges to justify their actions, is much more assuredly the primacy of European law and the principle of uniform application than that of national sovereignty.
Though it is fashionable, the historical dimension of ordering pluralism cannot be overlooked: it is a doctrine that corresponds to the way power is organized at a given moment. Its success reflects the fact that the question of the European Union’s fundamental nature is on the way to being resolved. It is understandable that national and supranational judges are attached to it: given the lack of a European constitution and prior to the existence of a federal state, it allows internal judges to preserve the nominal sovereignty of their states while letting the CJEU establish the European Union’s authority. As for the citizens: how could they blame judges for protecting fundamental rights while enriching, at the same time, the rule of law?
First published in www.laviedesidees.fr. Translated from French by Michael Behrent with the support of the Institut Français.