Show Summary Details
Page of

(p. 191) 7. EU law: is it international law? 

(p. 191) 7. EU law: is it international law?
(p. 191) 7. EU law: is it international law?

Bruno de Witte

Page of

PRINTED FROM OXFORD LAW TROVE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Law Trove for personal use (for details see Privacy Policy and Legal Notice).

date: 01 March 2021

1 Introduction

The European Communities, and later the European Union, came into being as creatures of international law, as they were established by means of international treaties concluded by States. Today still, the TEU and the TFEU form the basic documents of the EU legal order. It logically follows from this that EU law is still part—albeit a very distinctive and advanced one—of international law. There are, however, also good reasons for thinking that the EU is now so different from any other international organization in the world that it has become ‘something else’, more like the central unit of a European federal State. This discussion on the proper ‘legal nature’ of the EU, and of EU law, has been going on for decades, and it is indeed a curious characteristic of EU law that its precise place within the legal universe is not firmly established. This is an illustration of how legal reality—in this case, the way in which European integration has been shaped in legal terms—puts pressure on the established categories of legal science. This chapter will present the terms of the discussion on the legal nature of EU law, to help the reader situate the ‘place of EU law’1 within the world of legal phenomena.

So, this chapter is not centrally concerned with another question which is sometimes dealt with under the same title of ‘EU law and international law’, namely the conditions under which norms of international law that are binding on the EU (in most cases, because the EU has concluded a treaty with non-Member States) are applied within the EU legal order. That different, though related, question will be briefly considered in section 4. Before that, section 2 will present the ‘straightforward’ view that EU law is a part (or ‘sub-system’) of international law, and section 3 will discuss the ‘alternative’ view that EU law, although originating in international law, is now so distinctive that it should no longer be considered to be part of international law.

(p. 192) 2 EU law as a sub-system of international law

2.1 The EU as an international organization?

There is no disputing the fact that ‘the EU has successfully expanded its substantive mandate and institutional prerogatives to a level without parallel among international organizations’,2 and this raises the question whether it is still situated ‘among international organizations’ or has ceased to belong to that category. In fact, if one reads the scholarly literature of international law and of EU law, we find remarkably contrasting views on this question. A large part of the EU law literature takes the view that the EU, whilst certainly not yet a federal State, is also no longer an international organization but rather a legal construct that is somewhere in between those two poles, in a category all of its own (this is often expressed by saying that the EU is a sui generis entity). If, however, one reads textbooks on international law or the law of international organizations, their authors unanimously agree that the EU is still an international organization and is still situated within international law. The legal education programmes of different European countries imply similarly contrasting conceptions. In the UK, for example, the Law of the European Union was, until recently, a mandatory core subject of the undergraduate law degree, whereas international law is typically considered as an altogether different subject, usually offered as an optional course for few students. In the Netherlands, by contrast, law students are usually confronted, in the mandatory part of their undergraduate degree, with a combined introduction course to international and European Union law, which highlights the close connection that is seen to exist between the two subjects.

There are two basic reasons why it makes sense—subject to later qualifications of this view—to view EU law as part of international law. The first reason is that EU law continues to find its origins in international treaties. The rule of recognition of EU law (ie the rule that allows us to decide whether a norm is part of EU law or not)3 is whether a particular norm can be traced back, directly or indirectly, to the text of the TEU and the TFEU. We will highlight this continuing reliance on international law as the foundation of the EU legal order in the following subsections. The second reason is the extraordinary flexibility of international law. International law is made mainly by means of treaties between States, but there are no legal limits to the kinds of cooperation that States can undertake by means of treaties;4 indeed, the content of some treaties puts major limitations on the sovereign powers of the participating States. This is especially so when States create an international organization, that is, a permanent institutional structure for cooperation at the international level with its own organs possessing independent decision-making powers. There is a wide variety of such international organizations. They range from the United Nations Organization, of which almost all countries of the world are members and which deals with vital matters of peace and security, to the Office franco-allemand pour la jeunesse, a small bilateral organization created in 1963 by France and Germany in order to stimulate youth exchanges between the two countries. Each such treaty creates a particular sub-system of international law with its own special rules. Although there is some controversy about the criteria to be used in deciding whether a legal entity can be called (p. 193) an international organization, there is no doubt that the European Communities first, and the European Union now, tick all the boxes. They were created by an international Treaty between States and possess organs (which are called ‘institutions’ in EU law) with a will distinct from the will of the Member States.5

This said, the European Communities were, right from the start, a very special kind of international organization (section 2.2), and this peculiarity has become more evident with the passage of time (section 2.3). Yet, even today, the EU’s pedigree in international law remains. It reappears most forcefully when the Treaties on which the EU is based are being reformed, as they repeatedly have been in the past 30 years: those are major political events, which take the legal form of Treaty amendments, that is, acts of international law (Case study 7.1).

2.2 The international origins of the European integration process

Between 1948 and 1951, three new international organizations were created in quick succession in Western Europe. The Organisation for European Economic Co-operation (OEEC; later named OECD) was established by a Treaty signed in Paris on 16 April 1948 by 16 European States and the three Western Commanders-in-Chief of the German occupation zones. The Council of Europe was created by a statute (taking the legal form of a Treaty) signed in London on 5 May 1949 by ten European States. In addition, a Treaty signed in Paris on 18 April 1951 by six European States established the European Coal and Steel Community (ECSC). The creation of new international organizations was common practice. Dozens of multilateral international organizations had been set up since the middle of the nineteenth century.6 Most of them, though, had a universal remit, or at least an extra-European one, and the focus on creating ambitious multilateral entities responding to the ‘need of a closer unity between all like-minded countries of Europe’7 was a new development of the post-war period. All three organizations expressed, in their own way, the aspiration towards greater European unity, and they were often considered together in the legal scholarship of that period.8

In fact, during those immediate post-war years, active groups of European federalists had hoped to create a United Europe based on a federal constitution inspired by that of the US, and had sought to promote the adoption of federal solutions in the context of the various political initiatives taken in that period to reorganize the European political landscape. The European governments chose, instead, not to take the federal route and they rather shaped their new initiatives of European cooperation by using the age-old legal tool of the international Treaty. The OEEC, the Council of Europe, and the ECSC all took the form of an international organization based on a Treaty subject to ratification by the parliaments of their Member States. Both the OEEC and the Council of Europe were set up as rather old-fashioned and traditional international organizations, as their main organs were kept under the close control of the Member State governments (the European Convention on Human Rights was a legal breakthrough within the Council of Europe, but it happened (p. 194) by means of a separate Treaty which entered into force only later).9 The ECSC Treaty, by contrast, was an innovative endeavour; it created a different, functionally and politically more effective, international organization.

Indeed, whereas the ECSC was based on what seemed like an ordinary international Treaty, that Treaty was far from ordinary in terms of its content. A major, and at the time startling, innovation of the ‘Schuman Plan’,10 contrasting with most pre-war and post-war cooperation projects, was that sovereign States should agree to transfer their powers to regulate the coal and steel industries to a common body, the High Authority. By signing the ECSC Treaty, as it emerged from the Schuman Plan through negotiations in 1950 and 1951, the governments of the six founding States agreed to relinquish national control over these two sectors of the economy and to allow the supranational High Authority to exercise State-like public authority in their stead.

The term supranational, itself, appeared only once in the ECSC Treaty, and not in a very prominent place, but it had been repeatedly used during the negotiations, and was rapidly adopted by political and legal commentators as the defining characteristic not only of the High Authority but also of the entire Community of which it was an organ. The supranational character of the ECSC Treaty did not spring so much from the substantive norms it contained, but from the way in which it distributed legal authority and organized the decision-making by its organs. The ECSC had three institutions composed of persons who were not government representatives (the High Authority, a parliamentary Assembly, and a Court of Justice). Its institutions (particularly the independent High Authority) had the power to adopt binding acts, often by a majority vote; some of those decisions were directly applicable to private individuals and firms; and compliance by both the institutions and the Member States with their obligations was subject to judicial enforcement. None of those features was unprecedented in international law, so that the novelty of the ECSC did not reside in one or other of those specific characteristics but rather in their cumulative presence.11

When the EEC and the European Atomic Energy Community (EAEC) were created in 1957, as a further stage of the European integration process, they were equipped with an institutional regime that was as supranational as that of the ECSC. It is true that the Commission had a less prominent place in the decision-making system of the EEC than the High Authority had in the ECSC. Yet, the Community’s power to make directly applicable rules was now extended in its scope: broadly-based regulations could be adopted in a number of economic areas by the Council acting on a proposal by the Commission, in contrast with the limited decisions, relating solely to the coal and steel industries, which the High Authority could adopt under ECSC law. The EEC preserved all the other supranational characteristics of the earlier Treaty, with the interesting addition of the preliminary reference procedure before the European Court of Justice whose supranational potential went, however, largely unnoticed in 1957.

2.3 The later evolution of the European Communities and the European Union

The ‘Treaty path’ towards closer European integration, after having been confirmed when the EEC was established in 1957, was also followed during the later stages of the European (p. 195) integration process. Deliberate attempts to transform the European Communities into ‘something else’ (legally speaking) were made at several moments in the history of European integration, but none of those attempts resulted in a clear and explicit change of the legal nature of the organizations. Indeed, when the Maastricht Treaty, signed in February 1992, eventually created a new legal entity, called the European Union, alongside the European Communities, many considered this a step back in the European integration process. That new legal entity was created, by the then 12 Member States, to host two new forms of interstate cooperation, in common foreign and security policy (CFSP) and justice and home affairs (JHA), which were marked by a lesser degree of supranationalism than the existing European Communities. So, the creation of the EU in the early 1990s was not a jump forward towards a federal Europe, but a partial return to more traditional forms of intergovernmental cooperation—despite the fact that, in other ways, the Maastricht Treaty did indeed deepen the integration process: it created the Economic and Monetary Union, it coined the co-decision procedure for EC law-making, it introduced the concept of European citizenship, etc.

The curious ‘legal architecture’ resulting from Maastricht, consisting of two separate treaties corresponding to two separate but interconnected organizations (the EC and the EU), has now been replaced—through the Lisbon Treaty—by a new architecture. We now still have (at least12) two separate and legally equivalent Treaties (the TEU and the TFEU) but only a single organization, namely the European Union, which has now absorbed the European Community.13 That single organization is still firmly rooted in international law. In fact, the first sentence of Article 1 TEU uses the traditional language of international law:

By this Treaty, the HIGH CONTRACTING PARTIES establish among themselves a EUROPEAN UNION … on which the Member States confer competences to attain objectives they have in common.14

The evolution traced in the previous paragraphs would seem to lead to the logical conclusion that the EC, first, and the EU, now, remain creatures of international law, and therefore continue to belong to the legal category of international organizations. The European Treaties themselves do not contain express language to confirm this, but the EU Member States have accepted this qualification when operating in the broader international arena. Indeed, there are many multilateral Treaty provisions that use the terms ‘international organization’ or ‘regional economic international organization’ (REIO) where it is clear from the context that the (only) organization intended by that term is the EC or the EU.15 The EU Member States participate in drawing up those multilateral international (p. 196) conventions and could therefore object to the qualification of the EC or the EU as international organizations, but they do not.

The EU (or, previously, the EC) is also considered an international organization under the constitutional law of many of its Member States. Some national constitutions contain clauses dealing with the transfer of powers to the EU (such clauses were introduced by some ‘old’ members usually at the time of the Maastricht Treaty, and by ‘new’ members at the time they acceded to the EU). Other countries, however, continue to adopt a generic approach of allowing transfers of powers, or limitations of sovereignty, for the benefit of international organizations or international institutions without making special reference to the EU.16 It is clear that those generic references to international organizations include the EU. Therefore, the prevalent view, from the perspective of national constitutional law, seems to be that the EU is indeed a creature of international law and therefore an international organization, and that State sovereignty has not been abandoned or shared, but rather is being ‘exercised in common’ within the framework of the Union.17

When we turn to the Court of Justice, which has the principal authority to interpret and apply EU law,18 and therefore the authority to declare what the legal nature of the EU is, we find no conclusive views on the matter. It is true that the Court of Justice has regularly stated that ‘the EEC Treaty has created its own legal system’,19 or similar language, but this, in itself, is nothing special: every other international organization has its own legal sub-system as defined by its founding Treaty. The Court never added that this legal system was situated outside the scope of international law. In its famous early judgments, Van Gend en Loos and Costa, the Court sought to differentiate the EEC Treaty from ‘other’ or ‘ordinary’ international Treaties, but that otherness was not pushed to the conclusion that the EEC Treaty had created something other than an international organization.20 Much has been made of the fact that the Court held, in its Van Gend en Loos judgment of 1963, that the EEC Treaty had created ‘a new legal order of international law’, whereas it dropped the last three words one year later in Costa when it simply spoke of ‘a new legal order’. This has been interpreted, by a number of commentators, as a deliberate tearing away of Community law from its international legal moorings. Yet, the way in which the Court described the peculiar characteristics of the EEC Treaty was very similar in both those early judgments. It would therefore be very odd if that description had led, in 1963, to the conclusion that this was a special legal order still of international law, and only one year later to the opposite conclusion that it was a special legal order no longer of international law.

In the many years since its Costa judgment, the Court of Justice did not seek to develop a doctrine affirming the specific and non-international nature of the EU. Recently, the (p. 197) Court stated that ‘the founding Treaties, which constitute the basic constitutional charter of the European Union … established, unlike ordinary international treaties, a new legal order, possessing its own institutions, for the benefit of which the Member States thereof have limited their sovereign rights, in ever wider fields’.21 The Court hereby claims that the EU is not a simple sub-system within international law but a true ‘legal order’, and that the TEU and TFEU are not ‘ordinary Treaties’. Yet, the Court still does not state in so many words that the EU legal order has ceased to be part of international law. In fact, there is no need for the Court to adopt the view that the European Union is ‘something other’ than an international organization in order to affirm and protect the advanced features of European Union law since, as we saw, international law is extraordinarily flexible as to the content of the cooperation between States. On the few occasions on which the Court gives an indication of the kind of legal entity the EU might be, it tends to use tautological categories, as when it described the EU as a ‘union based on the rule of law’.22

Case study 7.1: Revision Treaties

We have seen, in the preceding discussion, that the European Communities were founded, back in the 1950s, by means of international Treaties concluded between their Member States. Since then, the deepening of the European integration process has taken place essentially by means of new Treaties between the Member States ‘revising’ (or ‘amending’, which has the same meaning) the existing Treaties. The widening of the European integration process took the form of a series of accession Treaties between the existing and the acceding States, organizing the way in which the new Member States could join the existing organizations. The main revision Treaties were the Single European Act, the Maastricht Treaty, the Treaty of Amsterdam, the Treaty of Nice, and most recently the Lisbon Treaty. Seven accession Treaties were concluded so far: with the UK, Ireland, and Denmark; with Greece; with Spain and Portugal; with Austria, Finland, and Sweden; with nine Central and European countries and Cyprus; with Bulgaria and Romania; and most recently with Croatia. Revision Treaties are, legally speaking, more interesting than accession Treaties.

Whereas the latter tend to be based on the simple principle that the newcomers must accept the existing legal order with only little room for institutional accommodation and derogations, revision Treaties have led to major changes in the institutional set-up and policy scope of the European Communities and the EU.

It frequently happens, of course, that States may want to amend a Treaty that they have previously concluded. The rules relating to such operations are set out in Articles 39 to 41 of the Vienna Convention on the Law of Treaties (1969). Article 39 contains the very simple default rule that a Treaty may be amended by an agreement between all the parties, and that the normal rules on the conclusion of Treaties apply to this amending agreement. The parties may set aside that default rule when concluding the original (to-be-amended) Treaty.

(p. 198) The international legal regime of Treaty amendment is, thus, one of utmost flexibility; the contracting parties23 are free to arrange for the later amendment of ‘their’ Treaty in the way they wish.24 Indeed, a large and increasing number of multilateral Treaties contain such a special amendment procedure, generally aimed at facilitating adaptation to changing circumstances, often by allowing for the amendment of a Treaty without the agreement of all the States.25 Article 48 EU, the amendment provision applying to changes of both the TEU and TFEU, is an example of a specific amendment clause but, contrary to most others, it does not provide more flexibility than the default rule of Article 39 Vienna Convention. It requires the agreement of all the parties (in this case, the Member States of the EU) for the valid adoption of the amendments and, in addition, it requires some involvement of the EU institutions in the preparatory phase and, in almost every case, the separate approval of the amendments by each State according to its own constitutional requirements.26

The EU’s ‘rule of change’ is thus particularly rigid, not because this rigidity is prescribed by general rules of international law, but because its Member States have committed themselves to such a rigid mechanism. The original rule that the EC Treaties could only be revised with the agreement of all the Member States, which made perfect sense when the EC counted only six States, was left untouched on each later occasion, despite the fact that membership grew from six to 28. This has rendered the unanimity requirement an ever more cumbersome obstacle in the way of Treaty change.27

As to the content of the Treaty amendments, the Member States traditionally acted, to use a famous German expression, as the Herren der Verträge (‘Masters of the Treaties’), bound by nothing else than the limits posed by their respective constitutions; they acted as ‘independent and sovereign States having freely decided … to exercise in common some of their competences’.28 The fact that the Member State governments acted as ‘masters’ of the Treaty text did not mean that they could also control what happened under the new Treaty once it had entered into force. The dynamic evolution of EU law in between Treaty revisions partly escaped their control, and they could correct unwanted evolutions only on the occasion of a further Treaty revision and on the unlikely condition that they found a consensus to overrule, say, a particular interpretation of the Treaty text adopted by the Court of Justice.

The close control exercised by the Member State governments on the revision process has now been diluted somewhat. Compared to the previous revision clauses, the new rules enacted by the Lisbon Treaty for future revisions of the European Treaties innovate in two limited (p. 199) respects: they prescribe the use of the ‘Convention method’ for future revisions, and they provide for two so-called simplified revision procedures, in addition to the ordinary revision procedure.

The Convention method consists in the fact that all important Treaty revisions must, from now on, be preceded by a deliberative phase within a ‘Convention’ comprising representatives not only of the national governments but also of the national parliaments and of the EU institutions (especially the European Parliament). The model for this new body is the Convention that was convened on an experimental basis, in 2002–3, to prepare the adoption of the Constitutional Treaty. The debate on the direction of future Treaty reforms is thus no longer confined to traditional diplomatic exchanges between national governments but is opened up to a wider constituency. Still, the Convention can only express an opinion or make proposals. The actual decision to amend the Treaties will still take the form of an agreement between Member State governments laid down in an international treaty.

The new simplified procedures, which apply for certain kinds of amendments mentioned in Article 48(6) and (7) EU, allow for the text of the Treaties to be amended, not by another international Treaty but by a unilateral decision of an organ of the EU itself (ie the European Council). However, the Articles thus amended will then take their place among the other Articles of the TEU or TFEU, thereby sharing the international legal nature of those Treaties.29

Those innovations of the Lisbon Treaty do not affect the fundamental rule of the game, namely that all Member States must give their unanimous consent to future amendments of the Treaties. The EU’s rules of change thereby continue to be much more rigid than the ones applying to national constitutions, but also more rigid than the ones applying to the founding instruments of other, less integrated, international organizations.30 They are a powerful confirmation of the fact that EU law remains anchored in international law and that the Union’s legal fate remains in the hands of the States that have created it, or joined it later on.

3 The specific features of EU law

Although the view that the EU is, and remains, a creature of international law may seem logical in view of the preceding discussion, it is, in fact, strongly contested in the legal literature. Many EU law scholars have argued, already since the 1960s, that the legal characteristics of Community law (and now Union law) are so peculiar and far removed from what one normally observes in the world of international relations that it does not make good sense to consider it to be part of international law.31 The EU is then no longer an international organization at all, but ‘something else’, a sui generis legal system that does (p. 200) not fit in the traditional dichotomy between (federal) States and international organizations. To continue to refer to it as an international organization would be ‘to try to push the toothpaste back in the tube’.32

The legal features of EU law that could support this alternative view are manifold: the broad and flexible nature of EU competences which extend into many areas of law-making; the existence of a (partially) common currency and a common (though derivative) citizenship; the decision-making regime, marked by the involvement of some institutions not controlled by the Member State governments and by recourse to majority voting as the usual rule in the State-controlled Council of Ministers; the relatively effective mechanism of State compliance; the habit of obedience by national courts to their duty to apply EU law.

This raises the question whether any of these legal features are incompatible with the notion of an international organization, thereby placing EU law as a whole outside the scope of international law. One could also argue that the incomparability of the EU results not from any one of those features but from their combination, which makes the EU a unique phenomenon and makes it seem contrived to insist on the formal view of the EU as just another international organization. We will discuss these arguments in the following two subsections.

3.1 The incomparability of the EU’s legal features

Advanced international organizations are created year after year, and display one or several of the characteristic features of the EU, mentioned previously. Many international organizations have organs possessing the power to adopt operational decisions that are binding on States, and occasionally such decisions can even be adopted by a majority vote, that is, against the wishes of single Member States of the organization.33 It is true, though, that those decisions seldom deal with very sensitive policy matters; they tend to be about technical issues, with some well-known exceptions such as the UN’s Security Council.

In the EU, by contrast, there is a constant stream of new legislation in a broad range of policy areas, typically adopted nowadays through what the Lisbon Treaty appropriately renamed the ‘ordinary legislative procedure’.34 This procedure is marked by the fact that the European Parliament and the Council must agree on the text, and that the Council can decide by a qualified majority of the represented Member States. Furthermore, in some crucial areas of EU law, such as competition law and monetary policy, supranational institutions (the European Commission and the European Central Bank, respectively) that are not controlled by the Member States exercise crucial decision-making powers. The European Parliament, which is directly elected by the European citizenry, exercises democratic control of the European Commission, and democratic input in the law-making process—this is, of course, unknown in any other international law regime.

If we move from law making to considering judicial enforcement of the law, also there we can find some examples of international organizations equipped with international (p. 201) courts having compulsory jurisdiction to settle disputes between States.35 It is rare, though, for international courts to provide for direct access by individual plaintiffs. Apart from the case of the European Court of Human Rights, we mostly find this in the framework of regional integration organizations whose political and judicial institutions were deliberately modelled on the EU.36 Examples include the Caribbean Court of Justice, and the Andean Tribunal of Justice (an organ of the Andean Community), which is engaged in an active dialogue with national courts through a preliminary reference mechanism comparable to that existing in the EU.

A feature of the EU’s judicial enforcement regime which remains fairly unique is the power of the European Commission (an institution acting independently from the Member States) to sue the Member States before the Court of Justice for infringement of their EU law obligations. This mechanism has allowed the Court to state, very early on, that the general international rule allowing States to retaliate in the event of non-compliance by other parties to the same Treaty (inadimplenti non est adimplendum)37 did not apply in the context of Union law. The Court deduced this innovative characteristic from a specific feature of the EEC Treaty itself, namely this power of the Commission to bring infringement actions against non-complying States.38 There is, thus, a specific system of State responsibility in EU law which does not allow States to adopt countermeasures in the event of a violation of EU obligations by the other States; yet, this specificity follows logically from choices made by the Member States themselves when they created the EEC and in particular from their decision to vest the Commission with an independent power to take non-compliant Member States to court. It can therefore be seen as an example of the flexibility provided by international law: when creating an international organization, States can devise operational rules for their organization that derogate from (ie go further than) the general rules of international law.

As for the preliminary reference mechanism, through which national courts engage in a dialogue with the Court when dealing with the application of EU law norms, this mechanism has been imitated elsewhere (see earlier in this subsection). Yet, it has produced, in the EU legal system, some seemingly unique consequences, in particular the formulation of an obligation for national courts to recognize the primacy of EU law. Since this obligation is often cited as the most compelling evidence that EU law is no longer part of international law, it deserves a closer look in the next subsection.

3.2 The primacy of EU law, a federal characteristic?

The feature of Union law that is most widely mentioned as showing its remoteness from international law is its primacy within the national legal orders of its Member States. (p. 202) Indeed, whereas international Treaties take precedence over national law, this precedence operates at the international level rather than within the domestic legal orders. Already in 1930, the Permanent Court of International Justice held that it was ‘a generally accepted principle of international law that in the relations between powers who are contracting parties to a treaty, the provisions of municipal law cannot prevail over those of the treaty.’39 Yet, this statement is, in fact, limited to the ‘relations between powers’ on the international plane; it does not apply to the internal workings of the national legal systems, in respect of which international law does not seem to claim a priority of Treaty norms over conflicting national norms. Indeed, the received view among international law scholars is the following:

From the standpoint of international law states are generally free as to the manner in which, domestically, they put themselves in the position to meet their international obligations; the choice between the direct reception and application of international law, or its transformation into national law by way of statute, is a matter of indifference … These are matters for each state to determine for itself according to its own constitutional practice.40

One finds this view repeated in all contemporary international law textbooks.41 In practice, though, we can observe a tendency towards greater recognition of the primacy of international law over national law in the text of national constitutions or in the case law of national constitutional and supreme courts.42 It is also possible for a particular Treaty to give specific guidance on the steps that States must take within their domestic legal order. The European Court of Justice noted this possibility in a judgment of 1999:

According to the general rules of international law there must be bona fide performance of every agreement. Although each contracting party is responsible for executing fully the commitments which it has undertaken it is nevertheless free to determine the legal means appropriate for attaining that end in its legal system, unless the agreement, interpreted in the light of its subject-matter and purpose, itself specifies those means.43

The TEU and TFEU (and their earlier pre-Lisbon versions) are agreements of the kind referred to in the sentence above. Their wording does not specify in any detail the means by which the Member States must comply with their commitments (except for the fact, recorded in what is now Article 267 TFEU, that they must allow their courts to refer preliminary questions to the Court on the interpretation of EU law), but, interpreting the (p. 203) Treaties in light of their subject matter and purpose, the Court came to the conclusion that some of their provisions should have direct effect in the domestic legal system (Van Gend en Loos and its progeny) and that all those norms that have direct effect should also have primacy over conflicting national law (Costa and its progeny).44

The principle that EU law prevails, or should prevail, over national law even in cases decided by national courts is reminiscent of the supremacy of federal law over Member State law, which is entrenched in the constitutions of prominent federal States such as the US and Germany. It is not surprising, therefore, to find many EU law scholars who saw the primacy doctrine as a hallmark of the quasi-federal and non-international nature of European law. Writing very shortly after the foundational judgments Van Gend en Loos and Costa, Peter Hay devoted a chapter of his work on Federalism and Supranational Organizations to what he termed the ‘Federal Relation of Community Law to National Law’.45 In 1991, Joseph Weiler wrote, in his Transformation of Europe, that the doctrines of direct effect and primacy rendered the relationship between Community law and national law ‘indistinguishable from analogous relationships in constitutions of federal states.’46 Many other authors have expressed similar views until the present day.

On a closer look, though, the primacy of EU law is quite different from the supremacy of federal law in countries such as Germany, Switzerland, the US, or Canada. In all those countries, the supremacy of federal law is effectively guaranteed by the fact that its enforcement is largely in the hands of federal courts. In the EU legal order, the inconsistency of a national norm with an EU law norm can be directly examined by the Court of Justice only in the framework of an infringement action brought by the Commission under Article 258 EC, where the Court can make Union law prevail as a matter of course, just like any international court would give precedence to international law over the domestic laws of the States parties to an international dispute. Usually, however, inconsistencies between national law and EU law appear through litigation before national courts and have to be solved by them, with the possible guidance of a preliminary ruling by the Court of Justice. In contrast with all federal States, there is no right of appeal to the Court of Justice against judgments of national courts that fail to recognize the primacy of EU law. Therefore, it is crucially important that the national courts should faithfully absorb and apply the primacy doctrine laid down by the Court of Justice, since it falls to them to make the doctrine a living reality.

This institutional factor has important consequences for the operation of the primacy doctrine. Since national courts conceive themselves to be organs of their State, they try to fit their ‘European mandate’ within the framework of the powers attributed to them by their national legal system. For those courts and, indeed, for most constitutional law scholars throughout Europe, the authority of EU law is rooted in their constitution, and subject to restrictions that may be imposed by the constitution.47 This situation is in stark (p. 204) contrast with the position prevailing in federal States where the primacy of federal law is based on the federal constitution whose authority as the fundamental norm of the State is uncontested. In the EU legal order, on the contrary, the hierarchical relationship between EU law and national law needs to be more heavily emphasized by the Court of Justice and its supporters because the ultimate hierarchy of norms is not settled in favour of EU law.

It is not so clear, therefore, that the doctrines of primacy and direct effect count as features of EU law that distinguish it from international law. The doctrine of primacy, as presently formulated by the Court and accepted by the Member State courts, has, no doubt, a distinct federal flavour but it could also be termed a creative development of international law. The central rule of international treaty law is pacta sunt servanda: States are bound by their treaty obligations; and whenever a conflict between a treaty obligation and a norm of national law is brought before the International Court of Justice, or any other international court, the answer is clear: the Treaty rule will prevail. The originality of the EEC Treaty, when it was concluded back in 1957, was to grant to the newly established Court, the ECJ, a jurisdictional competence which was then unique in the panorama of international law, namely that of guiding the activity of national courts while they are applying EU law by means of the preliminary rulings procedure. By means of this procedure, the ECJ had a ‘window’ through which it could intervene in pending national court cases and direct those national courts to respect the precedence of EU law.48

Seen from this angle, the direct effect and primacy doctrines do not so much signal a shift away from international law; they rather illustrate the dynamic potential of international law. The fact that, as a rule, international law leaves to States a choice among various methods of domestic enforcement of international obligations does not prevent specific Treaties or international decisions from imposing specific requirements in this respect. The TEU and TFEU may no longer be the only Treaties that do this. Given appropriate circumstances, primacy within the domestic legal orders may be required by other international Treaties as well.

3.3 Finding a name to describe the EU’s specificity

The preceding discussion does not lead to an unambiguous conclusion. There are, in fact, two credible options for explaining the legal nature of the EU. The first option consists in saying that, in view of its basis in a set of international Treaties, the EU is still an international organization and EU law is part of international law. This option must necessarily come with the recognition that the EU is an international organization unlike any other that has existed before, in view of the width and depth of the limitations imposed on the sovereignty of its Member States. However, this option allows for other experiments of international cooperation to adopt some or all of the distinctive features of EU law in order to create similarly advanced organizations elsewhere in the world. This explanatory option expresses, therefore, a belief in the flexibility and dynamic potential of international law.49

The second option consists in arguing that, because of its particular legal features, the ‘European legal order has … cut the umbilical cord with the international legal order’.50 There is no entity, anywhere in the world, that resembles the EU, and there is no realistic (p. 205) prospect for such an entity to appear any time soon. As it is not a credible option to argue that the EU has become a federal State, the authors defending this view argue that the EU finds itself on a middle ground. It is more than an international regime and not (yet) a federal State, but it possesses features stemming from both sides.51

Under both these options, there is a strong urge to find new names and concepts to describe the specific nature of the EU. Under the first option, that denomination would serve to identify the sub-category of advanced international organizations of which the EU is the leading exemplar. Under the second option, the need for a new denomination is even more urgent. Since the proponents of that option argue that the EU is neither an international organization nor a federal State, they are hard pressed to find an appealing name for this intermediate category, beyond the rather feeble solution consisting in saying that the EU is simply sui generis (meaning: it is in a category all of its own).

The term supranational organization served the name-giving purpose for many years, in the early decades of European integration, but it has now fallen into disuse, partly because of its now unfashionable hierarchical overtones, and partly because it failed to reflect the intergovernmental mode of integration that prevailed in the Second and Third Pillars of the EU after the Maastricht Treaty.52 The term confederation is occasionally proposed, but it bears the stigma of weakness and instability that derives from the historical examples of confederations, and is therefore rather unusual as a denomination for the EU.53 Alternative terminology should, in order to be attractive and true to reality, encapsulate both the international roots of the EU and its uniquely massive pooling of Member State powers and constitutional mode of operation.

No such alternative terminology has obtained Europe-wide currency so far, but influential descriptions in this vein have been proposed in Germany (the Staatenverbund) and in France (the fédération d’Etats-nations). The latter is a rather paradoxical denomination, which is particularly popular among French authors.54 It comes close to the German Staatenverbund, a term that was coined by Paul Kirchhof, and adopted in the Federal Constitutional Court judgment (the Maastricht judgment) which he helped to draft (as a member of that court) in 1993.55 It was proposed as a deliberate neologism intended to occupy an intermediate space between the Bundesstaat (meaning: federal State) and the Staatenbund (meaning: confederation of States). According to Everling, a former German judge at the European Court of Justice, the term Staatenverbund ‘stresses, albeit in a fashion that is almost impossible to convey in other languages, that the Member States are bound (p. 206) more tightly in the Union than in the traditional confederation of states.’56 Other, very similar, terms proposed in the legal literature are those of a federation of States,57 a commonwealth,58 or a constitutional order of States.59

This terminological discussion is, perhaps, not of great practical importance. One does not need to find a consensus on the best categorical denomination for the EU in order to study its law and its politics. However, the underlying discussion on the true legal nature of the EU is interesting from the perspective of the self-understanding of legal science. Legal science has traditionally been based on two crucial distinctions which were used to ‘organize the world’ of normative phenomena: between State law and non-State law (ie legal norms developed by groups in society outside the State structures); and between national law and international law. The latter dichotomous distinction, which slowly emerged through centuries of scholarship, is called into question by the ‘unruly’ phenomenon constituted by EU law. The question is whether we can continue to stick to the dichotomy (whereby EU law would be considered a sub-system within the category of international law) or whether we should recognize that the world of ‘State law’ should henceforth be divided in three: national law, international law, and EU law. However, would this not be a rather Eurocentric perspective?

3.4 The constitutional perspective on EU law

The controversy on the legal nature of the EU does not stand in the way of a widespread tendency, in the legal literature, to use constitutional terminology to describe and understand EU law. It is often claimed, also occasionally by the Court of Justice, that EU law forms a ‘constitutional’ legal order. Only a minority of scholars object to this usage, on the ground that constitutions are the basic legal instruments of sovereign States, and that constitutional terminology should therefore be reserved for States and not extended to non-State entities such as the EU.60

Around the turn of the century, the use of constitutional terminology to describe central features of the EU legal order had become commonplace in legal writing.61 This approach was transferred from legal scholarship to practical politics in 2002–3, when the Convention on the Future of the Union set out to draft a new Treaty text that was called the Treaty establishing a Constitution for Europe. Most EU legal scholars applauded this project to formalize the EU’s informal constitutional arrangements,62 but it failed. During the so-called ‘period of reflection’ that followed the negative referenda in France and the (p. 207) Netherlands and the demise of the Constitutional Treaty, the Member State governments came to the conclusion that part of the Treaty’s failure to take force was due to its overambitious constitutional design. In order to save most of the content of the reforms, they stated in June 2007 that ‘the constitutional concept, which consisted in repealing all existing Treaties and replacing them by a single text called “Constitution” is abandoned.’63 They also scrapped from the text of the Lisbon Treaty, which was subsequently agreed, some of the terminology used in the Constitutional Treaty that seemed too redolent of national constitutional texts, such as ‘law’ (instead of regulation or directive) or ‘minister of foreign affairs’ (instead of High Representative of the European Union) and ‘primacy of European Union law’ (see generally chapter 2).

Still, the constitutional perspective on EU law continues to make sense even after the demise of the Constitutional Treaty,64 and we find several textbooks on EU law that flaunt constitutional terms in their titles.65 By using constitutional language, these authors seek to emphasize that the Treaties perform many of the same functions as a national constitution: they establish public authority, and allocate powers between different political institutions; they establish a division of competences between the EU and the Member States, in a way which is reminiscent of similar delimitations of competences in federal constitutions; they set out the basic values of the EU legal order, and the fundamental rights of persons; they organize a system of judicial review for the enforcement of EU law obligations and for the review of the validity of EU action. In other words, the usage of constitutional terminology seeks to underline the close similarity (or analogy) between some of the legal issues that arise under EU law and those that arise in the constitutional law of States (especially federal States).

4 The EU as object and subject of international law

In this final section, attention is drawn to the complex situation that arises from the fact that the EU, whilst being a Treaty-based legal construct, can itself, in turn, conclude further Treaties in order to advance its policy objectives (and see further chapter 23). In other words, the EU is both an object of international law made by others (its Member States) and a subject entitled to make new international law. This is, once again, not a unique feature of the EU: other international organizations have the capacity to conclude Treaties;66 indeed, some of them have concluded Treaties with the EC or the EU.

(p. 208) The EU contributes to the development of international law in the same three main ways as States do: through its unilateral practice (which may contribute to the emergence of rules of customary international law);67 through concluding Treaties with non-EU States or international organizations; and through its activity as a member of some multilateral organizations, such as the World Trade Organization (WTO).68 Of those three, Treaty making is particularly important. The EU, in fact, has replaced its Member States as an international actor in a number of policy fields, such as international trade or international fisheries regulation. The Member States have had to accept abandoning their power to conclude Treaties in those areas that are within the EU’s exclusive competence.69 In other areas, such as environmental protection or immigration, the EU and the Member States share their Treaty-making competence.

When the EU concludes an international agreement, then international law starts acting as a constraint on the EU. This constraint is simply expressed by the rule in the TFEU stating that EU agreements are binding on the institutions of the EU and also on its Member States.70 However, the legal force of the constraint is, in fact, very variable. Increasingly, the Union seeks to insert clauses into international agreements that guarantee the prevalence of existing or future EU law over the obligations contained in the agreement. The most blatant form of this ‘reverse primacy’ is the so-called disconnection clauses that the EU manages to insert into many Council of Europe conventions. According to these clauses, EU law dealing with subject matter covered in the convention shall continue to apply between EU Member States, so that, on those matters, the convention’s provisions will only apply to non-EU Member States.71 Even in the absence of such conflict rules that preserve the integrity of the Union’s own law, the prevalence of the international agreements may be limited by the fact that their self-executing nature is denied (as is notoriously the case with WTO law).72 Finally, as highlighted in the Kadi judgment of 2008, the primacy of international agreements recognized by Article 216(2) TFEU relates to secondary EU law, but the application of international agreements within the EU legal order will be denied if they conflict with the Treaties themselves or with the unwritten principles of primary EU law.73

The Court of Justice sees the relationship between EU law and other forms of international law as a relationship between distinct legal orders. Thus, the Court held that ‘Security Council resolutions, on the one hand, and Council common positions and (p. 209) regulations, on the other hand, originate from distinct legal orders. Measures within the framework of the United Nations and the European Union are adopted by organs with autonomous powers, granted to them by their basic charters, that is to say, the treaties that created them.’74

The statement that the EU legal order is autonomous with regard to the surrounding international legal environment may, in fact, mean two rather different things: either that EU law, as a specialized international legal order, deviates from the general rules of international law on one or other point, or that the EU fails to comply with specific international obligations and gives priority instead to its own internal rules.75

The first meaning of autonomy, namely the capacity to adopt special legal rules that deviate from the general rules of international law, is unproblematic. The general rules of international law are default rules that States can set aside and replace by more suitable rules in their mutual relations. This is particularly true for treaties establishing international organizations, in which the founding States are free to equip ‘their’ organization with institutional mechanisms and operational rules of their own liking with hardly any limits to their creativity. In an Advisory Opinion of 1996, the International Court of Justice affirmed that ‘constituent instruments of international organizations are … treaties of a particular type; their object is to create new subjects of law endowed with a certain autonomy, to which the parties entrust the task of realizing common goals.’76

A second and stronger meaning of autonomy is the capacity for a particular system to give priority to its own internal rules over and above external international obligations. This form of autonomy is typical for those international organizations that are subjects of international law, and have used their capacity to conclude international agreements with States or other international organizations. When acting as subjects of international law, they may—just like States—incur obligations under international law that may appear, at some point in time, to conflict with their own internal rules. In view of the multiplication of such external legal commitments of the EU, it is quite likely that they may occasionally enter into conflict with the EU’s own domestic law.

In Kadi, the European Court of Justice thus refused the application of ‘external’ international obligations in order to preserve fundamental norms of the ‘internal’ legal order of the EU, namely the right of defence and the right to property. The wisdom of the Court’s attitude has been very widely discussed in the literature.77 From the point of view that concerns us here, namely that of the relation between the EU legal order and the surrounding international legal environment, the judgment did not imply a major change. In particular, the Court did not call into question its traditional view that international agreements of the EU form part of the EU legal order upon their ratification and entry into force. However, the Court did repeat the unsurprising view that international obligations concluded by the EU cannot prevail over the highest norms of the internal EU legal order. In defending that view, (p. 210) which is very similar to that adopted by national courts when confronted with a conflict between international and national constitutional law, the European Court of Justice appropriately highlighted the autonomy of the special legal order that the EU Member States have decided to carve out by using the instrument of an international Treaty.

5 Conclusion

The EU can be seen, as Wyatt and Dashwood aptly put it, as ‘a developed form of international organization which displays characteristics of an embryonic federation’.78 The various revision Treaties of the past decades (including the last one, the Lisbon Treaty) do not signal a change in this respect. They rather show that the Member State governments were not willing to contemplate a fundamental change of the EU’s legal nature; on the contrary, they have actively experimented, in the Lisbon Treaty, with the toolkit of international Treaty law, with generous use of Protocols, declarations, transition clauses, derogations, opt-outs, etc. Still, as we have seen, many EU law scholars disagree with this assessment and propose a different view of the EU as an unprecedented type of non-State entity.

The fact that, in this matter, legal scholars keep disagreeing has never stopped EU law from evolving through repeated Treaty revisions and through the institutional practice of its institutions and the legal interpretations by its Court of Justice. Indeed, hardly any practical consequences derive from the choice of a given qualification. In particular, qualifying EU law as international law does not imply that one should interpret the text of the Treaties from an intergovernmental perspective and that limitations of sovereignty should be narrowly construed. The Court has convincingly shown, since Van Gend en Loos, why it could construe the EC Treaty in a broad and purpose-oriented way, and other international courts have followed the same jurisprudential line in interpreting ‘their’ Treaties. Still, the recognition that EU law is an advanced species of the genus international law explains more easily some crucial characteristics of the EU legal order without hindering its autonomous development. It has the added advantage of making the ‘European way of law’79 a more amenable source of inspiration for other States, in other parts of the world, when they devise their own forms of international cooperation. The progressive development of international law would be weakened if its conceptual links with EU law were cut off.

Further reading

P D’Argent, ‘Jusqu’où y a-t-il du droit international? Considérations sur le droit dérivé des organisations internationales et sur le droit de l’Union européenne’, in Les limites du droit international—Essais en l’honneur de Joe Verhoeven (Bruxelles: Bruylant, 2015)Find this resource:

B de Witte, ‘The European Union as an International Legal Experiment’ in G de Búrca and JHH Weiler (eds), The Worlds of European Constitutionalism (Cambridge: Cambridge University Press, 2012)Find this resource:

TC Hartley, ‘International Law and the Law of the European Union’ (2001) 72 British Yearbook of International Law 1 (p. 211) Find this resource:

J Klabbers, An Introduction to International Organizations Law (3rd edn, Cambridge: Cambridge University Press, 2015)Find this resource:

J Klabbers, ‘Straddling the Fence: The EU and International Law’ in A Arnull and D Chalmers (eds), The Oxford Handbook of European Union Law (Oxford: Oxford University Press, 2015)Find this resource:

A Rosas and L Armati, EU Constitutional Law: An Introduction (3rd edn, Oxford: Hart Publishing, 2018) chs 2 and 3Find this resource:

R Schűtze, European Constitutional Law (Cambridge: Cambridge University Press, 2012) ch 2Find this resource:

D Simon and A Rigaux, ‘Les Communautés et l’Union européenne comme organisations internationales’, in E Lagrange and JM Sorel (eds), Droit des organisations internationales (Paris: LGDJ, 2013)Find this resource:

N Walker, ‘The Place of European Law’ in G de Búrca and JHH Weiler (eds), The Worlds of European Constitutionalism (Cambridge: Cambridge University Press, 2012)Find this resource:


1 N Walker, ‘The Place of European Law’ in G de Búrca and JHH Weiler (eds), The Worlds of European Constitutionalism (Cambridge: Cambridge University Press, 2012).

2 A Moravcsik, ‘The European Constitutional Settlement’ in S Meunier and K McNamara (eds), Making History: European Integration and Institutional Change at Fifty (Oxford: Oxford University Press, 2007) 23.

3 The notion of the ‘rule of recognition’ is a well-known element of the legal theory exposed by HLA Hart in his The Concept of Law (Oxford: Clarendon Press, 1972).

4 Except for one, rather nebulous, limit, namely that States are not allowed to make treaties that would conflict with jus cogens norms; see Art 53 of the Vienna Convention on the Law of Treaties (1969).

5 On the legal definition of an international organization, see J Klabbers, An Introduction to International Organizations Law (3rd edn, Cambridge: Cambridge University Press, 2015) 6–14.

6 For a synthesis of this historical development, see A Peters and S Peter, ‘International Organizations: Between Technocracy and Democracy’ in B Fassbender and A Peters (eds), The Oxford Handbook of the History of International Law (Oxford: Oxford University Press, 2012).

7 Statute of the Council of Europe, Preamble, para 4.

8 In the English language literature, see eg AH Robertson, European Institutions: Co-operation, Integration, Unification (London: Stevens, 1959).

9 On the extent to which the Statute of the Council of Europe, as agreed in 1949, constituted a dampener on the high hopes held by European federalists, see AJ Zurcher, The Struggle to Unite Europe 1940–1958 (New York: New York University Press, 1958) ch 5.

10 Thus named after the French minister of foreign affairs Robert Schuman who had launched the idea of the ECSC in a speech delivered in Paris on 9 May 1950.

11 F Capotorti, ‘Supranational Organizations’ in Encyclopedia of Public International Law, vol 5 (Amsterdam: North-Holland, 1983) 263–264.

12 ie without counting the EAEC Treaty, which continues to have a separate legal existence and is not subordinated to the TEU and TFEU; but its political importance has dwindled over the years.

13 See M Cremona, ‘The Two (or Three) Treaty Solution: The New Treaty Structure of the EU’ in A Biondi, P Eeckhout, and S Ripley (eds), EU Law after Lisbon (Oxford: Oxford University Press, 2012).

14 The capitalized words in the main text are printed in capitals in the version published by the Official Journal of the European Union (OJ [2008] C115/16).

15 See, among many other examples, the Energy Charter Treaty (1991), Art 1(2) and (3); the Convention on Biological Diversity (1992), Art 35; the UN Framework Convention on Climate Change (1992), Art 20; the UNESCO Convention on the Protection and Promotion of the Diversity of Cultural Expressions (2005), Art 27. See, on the practice of making these ‘REIO’ references, E Paasivirta and PJ Kuijper, ‘Does One Size Fit All? The European Community and the Responsibility of International Organizations’ (2005) 36 Netherlands Yearbook of International Law 169 at 206 et seq.

16 See M Claes, ‘Constitutionalising Europe at Its Source: The “European Clauses” in the National Constitutions: Evolution and Typology’ (2005) 24 Yearbook of European Law 81.

17 See, on this question of the continuing relevance of the notion of sovereignty in the context of European integration, N Walker, ‘Late Sovereignty in the European Union’ in N Walker (ed), Sovereignty in Transition (Oxford: Hart Publishing, 2003) and B de Witte, ‘Sovereignty and European Integration: The Weight of Legal Tradition’ in A-M Slaughter, A Stone Sweet, and JHH Weiler (eds), The European Court and National Courts: Doctrine and Jurisprudence (Oxford: Hart Publishing, 1998).

18 Art 19(1) TEU.

19 eg in the European Court of Justice (ECJ), Joined Cases C-6/90 and C-9/90 Francovich and Bonifaci v Italy [1991] ECR I-5357, para 31.

20 See, however, the Opinion of AG Poiares Maduro in the Kadi case who stated that the ECJ, in Van Gend en Loos, had considered the EEC Treaty to form a new legal order which was ‘beholden to, but distinct from the existing legal order of public international law’ (Opinion of 23 January 2008 in Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat v Council and Commission, para 21). The ECJ, in fact, did not quite use those words in its 1963 judgment, nor later.

21 Case C-621/18, Andy Wightman and others v Secretary of State for Exiting the European Union, EU:C:2018:999, para 44 (emphasis added). The same formula was also used, a few years earlier, in Opinion 2/13, Accession to the ECHR, EU:C:2014:2454, para 157.

22 Case C-583/11 P Inuit Tapiriit Kanatami v Parliament and Council, EU:C:2013:625, para 91 (emphasis added). Note the shift from ‘Union’ (the species) to ‘union’ (the genus). The small-u term is left undefined, and should not be read as denying that EU law belongs to international law. Curiously, the concept of ‘international union’ was of common usage in the nineteenth and early part of the twentieth centuries to describe the early international regimes for technical cooperation, eg for post, telegraphy, and copyright (A Peters and S Peter, ‘International Organizations: Between Technocracy and Democracy’ (n 6) 174–176).

23 ‘Contracting parties’ is the expression used in the practice of international law to denote the States (or international organizations) that have accepted to be bound by a particular treaty.

24 See A Aust, Modern Treaty Law and Practice (Cambridge: Cambridge University Press, 2000) 214: ‘It is wrong to think that the Vienna Convention is a rigid structure which places obstacles in the way of treaty modification: rather, it allows states to include in treaties such amendment provisions as they wish.’

25 For a survey of the wide variety of those special amendment procedures, see J Brunnée, ‘Treaty Amendments’ in DB Hollis (ed), The Oxford Guide to Treaties (Oxford: Oxford University Press, 2012).

26 On the current, post-Lisbon, EU Treaty amendment regime, see chapter 5, and for an overview of the content of the major revision Treaties, see chapter 2. For further discussion, see S Peers, ‘The Future of EU Treaty Amendments’ (2012) 31 Yearbook of European Law 17; B de Witte, ‘Treaty Revision Procedures after Lisbon’ in Biondi, Eeckhout, and Ripley, EU Law after Lisbon (n 13).

27 For an analysis of how, in particular, the requirement that each Treaty revision should be separately ratified by each State has placed obstacles in the way of EU Treaty change on many occasions, see C Closa, The Politics of Ratification of EU Treaties (Abingdon: Routledge, 2013).

28 The Member State governments used this phrase in the introductory part of the Decision on Denmark, adopted at the Edinburgh summit of 12 December 1992, but it referred back to their earlier adoption of the Maastricht Treaty.

29 The possibility for a treaty establishing an international organization to be amended through a unilateral act of one of the organs of that organization is not unheard of in the practice of international law; see Brunnée, ‘Treaty Amendments’ (n 25) 362.

30 For this comparison, see W Lehmann, ‘Federal States and International Organisations: A Short Comparison of Their Amending Rules with the European Union’, European Parliament note, December 2011 (available at

31 Among those scholars figure also some prominent former members of the European Court of Justice who defended this view in an extra-judicial capacity: P Pescatore, ‘International Law and Community Law—A Comparative Analysis’ (1970) 7 Common Market Law Review 167; GF Mancini, ‘Europe: The Case for Statehood’ (1998) 4 European Law Journal 29. As mentioned in the previous section, the Court itself, in its judgments, never expressed a clear position on this controversy.

32 JHH Weiler and UR Haltern, ‘Constitutional or International? The Foundations of the Community Legal Order and the Question of Judicial Kompetenz-Kompetenz’ in Slaughter, Stone Sweet, and Weiler, The European Courts and National Courts (n 17), 331 at 342.

33 For a general view, see J von Bernstorff, ‘Procedures of Decision-Making and the Role of Law in International Organizations’ (2008) 9 German Law Journal 1939; ND White, ‘Decision-Making’ in J Klabbers and A Wallendahl (eds), Research Handbook on the Law of International Organizations (Cheltenham: Edward Elgar, 2011) 225; and J Wouters and P De Man, ‘International Organizations as Law-Makers’ ibid 190.

34 Art 294 TFEU. See further chapter 5.

35 C Baudenbacher and MJ Clifton, ‘Courts of Regional Economic and Political Integration Agreements’ in C Romano, K Alter, and Y Shany (eds), The Oxford Handbook of International Adjudication (Oxford: Oxford University Press, 2014).

36 KJ Alter, ‘The Global Spread of European Style International Courts’ (2012) 35 West European Politics 135.

37 See N White and A Abass, ‘Countermeasures and Sanctions’ in ME Evans (ed), International Law (4th edn, Oxford: Oxford University Press, 2014).

38 Joined Cases 90/63 and 91/63 Commission v Luxembourg and Belgium [1964] ECR English Special Edition 625:

the [EEC] Treaty is not limited to creating reciprocal obligations between the different natural and legal persons to whom it is applicable, but establishes a new legal order which governs the powers, rights and obligations of the said persons, as well as the necessary procedures for taking cognizance of and penalizing any breach of it. Therefore … the basic concept of the Treaty requires that the Member States shall not take the law in their own hands. (emphasis added)

The same principle has been repeated by the Court many times since 1964; eg in Case C-111/03 Commission v Sweden [2005] ECR I-8789, para 66.

39 Permanent Court of International Justice, Greek and Bulgarian Communities, PCIJ, Series B, No 17, p 32.

40 RY Jennings and A Watts (eds), Oppenheim’s International Law, vol I (9th edn, London: Longman, 1992) 82–83.

41 For a more detailed treatment of the limited dictates of international law, and of divergent national constitutional practice, see M Mendez, The Legal Effects of EU Agreements: Maximalist Treaty Enforcement and Judicial Avoidance Techniques (Oxford: Oxford University Press, 2013) ch 1. See also E Denza, ‘The Relationship between International and National Law’ in ME Evans (ed), International Law (4th edn, Oxford: Oxford University Press, 2014) 412.

42 See A Peters and UK Preuss, ‘International Relations and International Law’ in M Tushnet, F Fleiner, and C Saunders (eds), Routledge Handbook of Constitutional Law (Abingdon: Routledge, 2013) 36–40.

43 Case C-149/96 Portugal v Council [1999] ECR I-8395, para 35 (emphasis added). Of course, the agreement to which the ECJ refers in this extract is not the EC Treaty or EU Treaty but an external agreement concluded by the EC.

44 On the legal effect of EU law in national legal systems, see generally chapter 6. For more on the development of the direct effect and primacy doctrines by the Court of Justice and their acceptance at the national level, see M Claes, The National Courts’ Mandate in the European Constitution (Oxford: Hart Publishing, 2006) and, for a shorter synthesis, B de Witte, ‘Direct Effect, Primacy, and the Nature of the Legal Order’ in P Craig and G de Búrca (eds), The Evolution of EU Law (2nd edn, Oxford: Oxford University Press, 2011).

45 P Hay, Federalism and Supranational Organizations: Patterns for New Legal Structures (Champaign, IL: University of Illinois Press, 1966).

46 JHH Weiler, ‘The Transformation of Europe’ (1991) 100 Yale Law Journal 2403 at 2413. On the concept of federalism in EU law, see further chapter 4.

47 For the evidence backing this statement, see the various ‘national reports’ in Slaughter, Stone Sweet, and Weiler, The European Courts and National Courts (n 17); and the comparative analysis by Claes, The National Courts’ Mandate in the European Constitution (n 44).

48 The importance of the preliminary rulings procedure in allowing for the emergence of the primacy doctrine was emphasized by D Wyatt, ‘New Legal Order, or Old?’ (1982) 7 European Law Review 147.

49 For a fuller articulation of this view, see B de Witte, ‘The European Union as an International Legal Experiment’ in de Búrca and Weiler, The Worlds of European Constitutionalism (n 1); and TC Hartley, ‘International Law and the Law of the European Union’ (2001) 72 British Yearbook of International Law 1.

50 R Schütze, European Constitutional Law (Cambridge: Cambridge University Press, 2012) 66.

51 Among the authors who have recently expressed this view, we can mention Schütze (see n 50), and M Avbelj, ‘Theory of European Union’ (2011) 36 European Law Review 818.

52 The term supranational organization is still used, occasionally, to describe the nature of the EU, for lack of a better term. See eg K Lenaerts and P Van Nuffel, European Union Law (3rd edn, London: Sweet & Maxwell, 2011) 16–17.

53 On the ‘stigma of confederation’, see M Burgess, Federalism and European Union: The Building of Europe, 1950–2000 (Abingdon: Routledge, 2000) 259–60. Nevertheless, Burgess proposes rehabilitating and revitalizing the old concept as an adequate description of the EU (at 265–69).

54 Among the French authors who have discussed or promoted the use of this concept in connection with the EU are: JL Quermonne, ‘La “Fédération d’Etats nations”: concept ou contradiction?’ (2010) 84 Revue française de droit constitutionnel 677; V Constantinesco, ‘Europe fédérale ou fédération d’Etats-nations’ in R Dehousse (ed), Une constitution pour l’Europe? (Paris: Presses de Sciences Po, 2002); G Ricard-Nihoul, Pour une Fédération européenne d’États-Nations: La vision de Jacques Delors revisitée (Brussels: Larcier, 2012). A related concept is that of ‘plurinational federation’, proposed by A Bailleux and H Dumont, Le pacte constitutionnel européen (Bruxelles: Larcier, 2015), 226–233.

55 P Kirchhof later presented an updated version of his views, in English, in ‘The Legal Structure of the European Union as a Union of States’ in A von Bogdandy and J Bast (eds), Principles of European Constitutional Law (Oxford: Hart Publishing, 2005).

56 U Everling, ‘The European Union between Community and National Policies and Legal Orders’ in von Bogdandy and Bast, Principles of European Constitutional Law (n 55) 719 (emphasis added).

57 Schütze, European Constitutional Law (n 50) ch 2.

58 N MacCormick, Questioning Sovereignty: Law, State, and Nation in the European Commonwealth (Oxford: Oxford University Press, 1999).

59 A Dashwood, ‘States in the European Union’ (1998) 23 European Law Review 201.

60 See eg D Grimm, ‘Does Europe Need a Constitution?’ (1995) 1 European Law Journal 282.

61 Among the many writings of that period in which the terms constitution and constitutional were considered in relation to EU law, see JHH Weiler, The Constitution of Europe (Cambridge: Cambridge University Press, 1999); K Lenaerts, P Van Nuffel, and R Bray, Constitutional Law of the European Union (London: Sweet & Maxwell, 1999); J-C Piris, ‘Does the European Union Have a Constitution? Does It Need One?’ (1999) 24 European Law Review 557; P Craig, ‘Constitutions, Constitutionalism, and the European Union’ (2001) 7 European Law Journal 125; A Peters, Elemente einer Theorie der Verfassung Europas (Berlin: Duncker & Humblot, 2001).

62 But see, for a critical voice, JHH Weiler, ‘In Defence of the Status Quo: Europe’s Constitutional Sonderweg’ in JHH Weiler and M Wind (eds), European Constitutionalism beyond the State (Cambridge: Cambridge University Press, 2003).

63 Those were the words used in the European Council’s mandate for the intergovernmental conference that was to elaborate the Lisbon Treaty (Annex 1 to the European Council Conclusions of 21–22 June 2007, p 1).

64 On the latter point there is a solid consensus in the EU law literature; see among others: G de Búrca, ‘Reflections on the EU’s Path from the Constitutional Treaty to the Lisbon Treaty’ (2008) Jean Monnet Working Paper 03/08; N Walker, ‘European Constitutionalism in the State Constitutional Tradition’ (2006) 59 Current Legal Problems 51; L Besselink, ‘The Notion and Nature of the European Constitution after the Lisbon Treaty’ in J Wouters, L Verhey, and P Kiiver (eds), European Constitutionalism beyond Lisbon (Antwerp: Intersentia, 2009); S Griller, ‘Is This a Constitution? Remarks on a Contested Concept’ in S Griller and J Ziller (eds), The Lisbon TreatyEU Constitutionalism without a Constitutional Treaty? (Vienna: Springer, 2008); K Lenaerts, ‘De Rome à Lisbonne, la Constitution européenne en marche?’ [2008] Cahiers de droit européen 229; L Pech, ‘The Fabulous Destiny of the EC Treaty: From Treaty to Constitution to Treaty Again?’ (2008) 15 Irish Journal of European Law 49.

65 A Rosas and L Armati, EU Constitutional LawAn Introduction (3rd edn, Oxford: Hart Publishing, 2018); Schütze, European Constitutional Law (n 50); A von Bogdandy and J Bast (eds), Principles of European Constitutional Law (2nd edn, Oxford: Hart Publishing, 2011): K Tuori, European Constitutionalism (Cambridge: CUP, 2015). See also the discussion of constitutionalism and EU law in chapter 4.

66 See Klabbers, An Introduction to International Organizations Law (n 5) ch 12.

67 See, on this aspect of the interaction between EU law and general international law, F Hoffmeister, ‘The Contribution of EU Practice to International Law’ in M Cremona (ed), Developments in EU External Relations Law (Oxford: Oxford University Press, 2008) 54 et seq.

68 For a collection of contributions on this theme, see B Van Vooren, S Blockmans, and J Wouters (eds), The EU’s Role in Global Governance: The Legal Dimension (Oxford: Oxford University Press, 2013).

69 This is perhaps the characteristic that is most unusual from the point of view of general international law; see discussion in B de Witte, ‘The Emergence of a European System of Public International Law: The EU and Its Member States as Strange Subjects’ in J Wouters, A Nollkaemper, and E de Wet (eds), The Europeanisation of International Law (The Hague: TMC Asser Press, 2008).

70 Art 216, second para, TFEU.

71 For a discussion of these disconnection clauses and similar devices aiming at preserving the integrity of pre-existing or future EU law against conflicting international obligations, see M Cremona, ‘Disconnection Clauses in EC Law and Practice’ in C Hillion and P Koutrakos (eds), Mixed Agreements RevisitedThe EU and Its Member States in the World (Oxford: Hart Publishing, 2010).

72 See, for a detailed discussion, M Mendez, The Legal Effects of EU Agreements: Maximalist Treaty Enforcement and Judicial Avoidance Techniques (Oxford: Oxford University Press, 2013).

73 Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat v Council and Commission [2008] ECR I-6351. For further discussion of this judgment, see chapters 8 and 9.

74 Case C-380/09 P Melli Bank v Council, EU:C:2012:137, para 54. The term autonomy has repeatedly been used in the Court’s case law. See also Opinion 2/13 Accession of the EU to the ECHR, EU:C:2014:2454, paras 170 and 183.

75 For a similar distinction between the different meanings of the autonomy of EU law, see M Klamert, ‘The Autonomy of the EU (and of EU Law): Through the Kaleidoscope’ (2017) 42 European Law Review 815.

76 ICJ, Legality of the Use by a State of Nuclear Weapons in Armed Conflict, Advisory Opinion of 8 July 1996, para 19.

77 See, for some contrasting views (among many others): T Tridimas and JA Gutierrez-Fons, ‘EU Law, International Law, and Economic Sanctions against Terrorism: The Judiciary in Distress’ (2009) Fordham International Law Journal 660; and G de Búrca, ‘The ECJ and the International Legal Order: A Re-Evaluation’ in de Búrca and Weiler, The Worlds of European Constitutionalism (n 1).

78 Wyatt and Dashwood’s European Union Law (5th edn, London: Sweet & Maxwell, 2006) 132.

79 A-M Slaughter and W Burke-White, ‘The Future of International Law is Domestic (or, the European Way of Law)’ (2006) 47 Harvard Journal of International Law 327.