The EU and International Law

From IVR encyclopedie

Jump to: navigation, search

by Jan Klabbers

The relationship between the EU and international law has many dimensions, and it hardly bears emphasizing that the EU is by far the most externally active international organization. Other international organizations tend to have far less external activities, and rarely need to maintain contacts with the outside world (barring a headquarters agreement, perhaps). But the EU is different: ever since its creation it was conceived of as having an external dimension, originally most clearly visible in the presence of provisions concerning a common customs tariff, and a common external trade policy (‘commercial policy’, in jargon). Since the early 1990s, moreover, the EU even has a foreign policy ambition and takes action in the field of migration.

In order to understand the relationship between EU and international law, two broad questions need to be discussed here. First, does international law allow for action by the EU? And second, what is the effect of international law in the EU’s legal order?

The first of these questions itself divides into various different issues. First, there is the issue of the legal personality of the EU under international law, and its capacity to act internationally; second, there is the scope of the powers of the EU. As will be noted, discussing such issues unearths some complicated (and unresolved) underlying jurisprudential debates into such topics as the precise nature and function of legal personality and the concept of legal powers. By the same token, the relationship between international law and EU law provokes reflections on the co-existence of distinct legal orders. In what follows, the terms EU and EC will be used interchangeably, except where the context dictates otherwise.

Personality, Capacity, Powers

The three original European Communities were set up in the 1950s (ECSC in 1951, EEC and Euratom in 1957), and all three included a grant of legal personality. In other words, the founding fathers endowed these creatures with legal personality, albeit in very non-specific terms. Thus, article 281 TEC simply reads ‘The Community shall have legal personality’. The ECJ, however, made clear that this included legal personality under international law, and did so relatively early (ERTA case, 1971). As a result, the international legal personality of the three communities has never given rise to serious debate.

The EU, by contrast, newly created under the Maastricht Treaty in 1992 and grafted onto the existing three communities, was not explicitly granted personality. This gave rise to some speculation as to whether it could be deemed to possess such personality. Some theorists held that this was excluded. The drafters’ apparent intention not to grant personality had to be respected. Others argued that when acting externally, the EU could ‘borrow’ the legal personality of the EC by acting through the EC. Yet others still felt that the intentions of the drafters were not all that decisive, and claimed instead that what mattered was whether international law attached any consequences to the existence of such entities as the EU (Klabbers 1998). Arguably, practice following the entry into force of the Maastricht Treaty has aligned itself most closely to this latter group , and perhaps that should come as no surprise: to create an entity with ambitions in the field of foreign policy and migration issues yet withhold international legal personality from it, was always going to be an awkward construction.

This debate on the legal personality of the EU sparked reflections on the functions of legal personality more generally: whereas a majority seems to hold that personality functions as a threshold without which one simply cannot act internationally, others suggest that personality is not such a threshold: entities can and do act without legal personality, or so it seems. If so, then it would seem to follow that the function of personality is a different one, perhaps best phrased as the right to set internal rules, in relative autonomy from general international law (Klabbers 2005).

While personality may not be all that relevant, the precise capacity of international organizations (including the EC) to act under international law has never seriously been doubted. Such capacity derives from general international law, as is recognized in the 1986 Vienna Convention on the Law of Treaties with or between International Organizations. And under this Convention, it extends to all international organizations, whether they are considered legal persons or not.

Where capacity is granted by international law, specific powers are, according to most accepted theories, granted by the founding fathers. Over the years, the EC and EU have been granted specific external powers in quite a few issue areas. Originally the most eye-catching were the commercial powers: the EC was characterized, since the early 1970s, by having a common trade policy. Other obvious issue areas with external dimensions include development and the conclusion of association agreements with third states: the latter typically allows for action with respect to states willing to join the EU but not yet deemed ready (or deemed excluded for other reasons, such as geography). In addition, the case law makes clear that the EC can also boast external powers in fields where it has internal legislative powers. Such areas include transport and environmental protection, to name just two of the more relevant ones (Koutrakos 2006).

How those powers are granted is subject to debate, however, both generally (Sarooshi 2005) and within the specific context of the EU. Some have an express basis in the constituent treaties, and therewith create few (practical) problems. Others are regarded as implied powers, for instance in accordance with the doctrine of parallelism: an internal power, on this reading, would almost inevitably have to be matched by an external power. And there is room for the somewhat less orthodox view that in some cases, powers are created not by implication, but as a result of the perceived need to protect EU law. On this view, the unity of EU law could be undermined if member states would still be able to conduct their own external relations. Hence, the member states must be deemed pre-empted from doing so which, in turn, suggests that the power to act ultimately must reside with the EU. While this is difficult to reconcile with the view of legal powers as being intentionally delegated or transferred by member states, the ECJ has made intensive use here of what is sometimes referred to as the principle of fidelity as laid down in article 10 TEC (‘Gemeinschaftstreue’): the member states owe it to the EU not to jeopardize the European project.

International Law in EU Law

An important question about international law generally concerns its effects in other (typically: domestic) legal orders: can individuals or other legal persons directly invoke provisions of international law before local courts or administrative agencies, or is some act of transformation of these provisions required before local authorities can apply them? Usually, the answer is thought to reside in fundamental constitutional law doctrine: either monism, or dualism.This section discusses, first, how the EC distanced itself from its international legal foundation, and second, the implications this has for the very relationship between the international legal order and the EC.

The original treaties were undeniably binding documents under international law, giving rise to rights and obligations between the (then) six member states. In the original conception, there was nothing all that special about the EC: it was set up as creative of obligations binding under international law, but without any specific reference to its impact on the legal orders of the member states.

Still, it would not take long before the special nature of the EC would be created through the case-law of the ECJ. In Van Gend & Loos, in 1963, the Court claimed that EC law would be self-executing in the domestic law of its member states, regardless what the constitutions of those member states would have to say about that. A year later, in Costa v ENEL, the Court added a doctrine of supremacy: EC law would prevail over incompatible law of the member states. These cases suggested a strong integrative role for the ECJ, but this has been nuanced in the literature: the ECJ can act freely as long as it leaves the political decision-making powers of the member states intact (Weiler 1999).

Van Gend & Loos and Costa v ENEL were enough to elevate the EC from a regular intergovernmental organization to a more rarefied position, and to create some doubts about its legal nature. In both cases, the main justification for the ECJ had been that the EC were created as a ‘new’ legal order, creating not just rights or obligations between member states but also, and predominantly, for those within the jurisdiction of those member states: companies, individuals, administrative agencies, therewith distancing it from traditional international law. Such was manifested, so the ECJ suggested, by the existence of an Assembly and other organs which entailed some role for social actors, as well as by the reference in the preambles to the peoples of Europe, and the existence of an intimate link between EU law and member state law, expressed most visibly perhaps in the preliminary ruling procedure: the courts of the member states and the ECJ were designed to work closely together. On this basis, later commentators could conclude that the Court, from the early 1960s onwards, had started a process of constitutionalization (Stein 1981).

This meant that slowly, the EC had lost its moorings in public international law. While international law can, to a greater or lesser extent, provide a framework for analysis of regular intergovernmental organizations, it has little aptitude when it comes to an entity such as the EU. Indeed, it soon became doubtful whether the EC could even still be classified as an international organization – many settled for the telling (yet meaningless) compromise that the EC was an organization in a category of it own: sui generis.

All this had many implications, but one of them stands out for present purposes: if the EC can no longer meaningfully be regarded as a creature of international law, then how does it relate to that same branch of law? The EC itself utilizes a standard story which holds that since the EC was created under international law, it could not be anything else but friendly disposed towards international law: the EC, according to this narrative, embodies a principle of ‘Völkerrechtsfreundliche Integration’. (Timmermans 1999) Admittedly, so its proponents suggest, sometimes the EC has to protect itself from international law, but that is because it is still young and fragile – the legal counterpart to the ‘infant industry’ argument known from trade economists. And this protection then typically takes place if and when the EU is most vulnerable, i.e. when confronted with international legal demands concerning its main external policy area: trade policy. Typically, in cases coming before the ECJ involving the EU’s obligations under the WTO (of which it is a founding member), the ECJ has proved markedly reluctant to apply WTO law. In a string of cases the ECJ has consistently held that WTO law shall not be directly effective in the EU’s legal order (Klabbers 2002).

Where exactly this leaves the EU as far as other emanations of international law are concerned is less than clear. Typically, the ECJ sees no problems in allowing subjects to rely on provisions in association agreements – but this may be because those tend to be concluded in situations where the EU is usually the stronger party. With treaties concluded within the framework of the Council of Europe (numerically dominated by EU member states), the EU is often able to secure a so-called ‘disconnection clause’: de-activating the operation of such treaties in relations between EU member states inter se (Economidès and Kolliopoulou 2006). Such clauses typically provide that while the parties (including the EU’s member states) are bound to behave in accordance with the terms of the treaty at issue, in their mutual relations the EC member states are bound to give priority to EC law. While the ECJ has consistently held that the EU is bound to respect customary international law, it sometimes ends up applying customary rules in ways which few international lawyers would recognize – or recommend. A leading example here is the Racke case, in which the ECJ accepted the EU’s reliance on the controversial rebus sic stantibus doctrine (the notion that a fundamental change of circumstances may influence the sanctity of treaties) as justification for the suspension of a trade agreement.

The European Convention on Human Rights has, until recently, received only a lukewarm embrace by the ECJ, and it is arguable that the recently developed warmer embrace is the result of the political decision by the member states that the EU ought to accede to the Convention, as envisaged in the Lisbon Treaty. Finally, the ECJ has been most ambivalent when it comes to UN law. While UN law, according to the UN Charter itself, prevails over any other obligations that UN member states may have, the ECJ has usually seen fit nonetheless to make clear that it applies UN law by virtue of EU law. And should UN law be in conflict with fundamental EU law, then UN law should be set aside (Kadi case).


The role of the EU within the international legal is still shrouded in myth and mystery. While there are excellent studies of the details of EU external relations law, most of these focus on the constitutional dimension or on concrete policies, rather than on the place of the EU in the broader scheme of things (Eeckhout 2004, Cremona 2008). It is difficult, moreover, to identify serious jurisprudential work on the EU’s position in international law (Klabbers 2009): arguably academic specialization has made this all but impossible, as the EU lawyers would feel uncomfortable with the jurisprudential aspects and the international law, and vice versa. Yet, it is by no means eccentric to suggest that such serious work ought to be undertaken: the more the EU matures, the more it acts as a global power, the more fundamental issues concerning its position in international law will be raised.


Marise Cremona (ed.), Developments in EU External Relations Law (OUP 2008)

Constantin P. Economidès & Alexandros G. Kolliopoulou, ‘La clause de deconnexion en faveur du droit communautaire: une pratique critiquable’, 110 Revue Générale de Droit International Public (2006) 273-302

Piet Eeckhout, External Relations of the European Union: Legal and Constitutional Foundations (OUP 2004)

Jan Klabbers, ‘The Concept of Legal Personality’, 11 Ius Gentium (2005) 35-66

Jan Klabbers, ‘International Law in Community Law: The Law and Politics of Direct Effect’, 21 Yearbook of European Law (2002) 263-298

Jan Klabbers, ‘Presumptive Personality: The European Union in International Law’, in Martti Koskenniemi (ed.), International Law Aspects of the European Union (Martinus Nijhoff 1998) 231-253

Jan Klabbers, Treaty Conflict and the European Union (CUP 2009)

Panos Koutrakos, EU External Relations Law (Hart 2006)

Dan Sarooshi, International Organizations and their Exercise of Sovereign Powers (OUP 2005)

Eric Stein, ‘Lawyers, Judges, and the Making of a Transnational Constitution’, 75 American Journal of International Law (1981) 1-27

Christiaan Timmermans, ‘The EU and Public International Law’, 4 European Foreign Affairs Review (1999) 181-194

J.H.H. Weiler, The Constitution of Europe (CUP 1999)

Personal tools