On the relations between the European Court of Justice and the international ‘authorities’

From an article first published as ‘Sur les relations entre le juge communautaire et les « autorités » internationales’ (2008) Annuaire français de droit international.


One cannot but be struck by the growing number of references made to "international law" in Community (practice) case law. But there is nothing startling about it really. It is not that the Union (Community) is itself a construction of international law insofar as it is based on an agreement among the states that put it in place and those which subsequently joined them. While undeniable, this rooting in treaty cannot conceal the ever greater liberty the Community (Union) has taken with the rules normally applicable to the interpretation and application of international treaties. It has probably not entirely freed itself of a founding agreement, assuming that it ever can. Even so it is no longer saying much to underscore simply that the Community (Union) rests upon a treaty among states as its foundation. There should be no need to labour this point.


One cannot but be struck by the growing number of references made to ‘international law’ in Community (practice) case law. But there is nothing startling about it really. It is not that the Union (Community) is itself a construction of international law insofar as it is based on an agreement among the states that put it in place and those which subsequently joined them. While undeniable, this rooting in treaty cannot conceal the ever greater liberty the Community (Union) has taken with the rules normally applicable to the interpretation and application of international treaties. It has probably not entirely freed itself of a founding agreement, assuming that it ever can. Even so it is no longer saying much to underscore simply that the Community (Union) rests upon a treaty among states as its foundation. There should be no need to labour this point.

If international law is growing in importance, it is primarily because the litigation being brought before the ECJ increasingly, directly or indirectly, implicates states or organizations that are not members of the Union (Community). It does not necessarily follow that the relations that have been or will be established with the Union (Community) are necessarily subject to rules other than those of Community law. However, it can readily be conceived that a priori the rules come under ‘classical’ international law when the relationship with the Community (Union) does not result from some form of submission to its law accepted by the third party, whether a state or an international organization. This is the case, notably, when the Community enters into an agreement with third states or when such an agreement is entered into by one of its member states with third states in at least partly ‘community-governed’ subject areas. However, there are not just treaty-based relations; it is also international law that would be activated, as it were, if the Community were to be sued in the national courts of a third state from whose jurisdiction it might purport to exempt itself by claiming immunity from legal proceedings.

What is fundamentally at issue under these assumptions is the status of a norm and the problems arising form it: On what condition is it applicable? What becomes of it when it is in conflict with a Community norm? To what extent can any violation of it be sanctioned by the ECJ? And so forth. The problems of themselves are quite classical ones. One might waver over the solutions to them, especially if one refused to systematically apply those solutions that would be applicable if equivalent rules of international law alone were at issue. Most certainly there is nothing to prevent quite different solutions from being developed if they are useful or at least preferable, which may (might) be commended without too much difficulty by a sort of specificity—and heaven knows it has often been lauded!—of the Community and so, by correlation, of its law. Whatever one might think of the fate thus attending ‘rules’, it is decidedly something else entirely to question the relations of the Community courts with an international ‘authority’, whatever the status reserved for the norm produced, at least when there is no systematic refusal to take that norm into consideration. It is a mutual collaboration, with multiple potentialities, that is then virtually at issue.

Some commentators may contest the very existence of an authority, at least an authority over states, in the egalitarian world of the international community. It is true there is no power that can impose itself on states if those subject to it have not so consented. Would that suffice to preclude any authority from being exercised? One can but hesitate to accept as much. It is certainly not Community precedent that will belie it! Whatever the case, there is no cause to dally on this point. The important thing is in any event that there are ‘institutions’ with their own freedom of action in international relations, which allows them in particular to take ‘decisions’. These ‘institutions’ are manifold, so numerous are the agreements which, for their specific purposes, have conferred particular powers on them. There is no need here to make an inventory of them, and even less to contemplate the various forms of collaboration that the Community court does or did or should maintain with each of them. In the light of (relatively) recent case law, for two of them it is worth dwelling briefly on the conditions of their co-existence with the European Community (Union) courts. These are the ‘international’ court and the Security Council. The Community courts have long taken into consideration decisions by jurisdictions established by international treaties for various ends. In the often cited Mox Plant ruling, it has however more explicitly than elsewhere, albeit implicitly, ruled on the scope of and presuppositions about its collaboration with other international courts. And in its Kadi and Yusuf rulings, it has—much less clearly—adjudicated on the somewhat different conditions of its collaboration with a genuinely political authority, the UN Security Council.

This paper does not set out to comment in detail on the grounds for these generally familiar decisions, which have already been largely evoked in this Yearbook. One can be less than fully convinced of the well-foundedness of a ground which, behind an apparent ‘logic’, sometimes brushes aside far more problems than it solves. It is true that the ECJ does not seem to have truly found its own ‘style’ for drafting its rulings, which does not always facilitate the reading of them. It ordinarily seems content to stack up supposedly ‘probative’ points, as if bricks alone were enough to build a house satisfying the basic requirements of comfort and attractiveness. Which does not really simplify the understanding of its judgments ... even if it is true that in these areas appreciations are even more subjective than they are naturally.


It is a fact that international jurisdictions are multiplying, which naturally raises questions about the circumstances in which they mutually co-habit, which it is hoped are as harmonious as possible. Even so agreement must be reached about what an ‘international’ jurisdiction is. The question is less straightforward than it looks. Received wisdom has it that an ‘international’ jurisdiction is one that owes its existence to an instrument of international law, be it an agreement between states or the decision of an international organization, as evidenced in particular by the tribunals recently set up by the Security Council (ICTY, ICTR). There is no contesting the conclusion when the dispute that this jurisdiction is called upon to decide comes intrinsically under international law, that is, it falls under the relations directly governed by international law. Is the conclusion equally obvious when this is not so, for instance when the issue is between private individuals when their rights or obligations are determined in part or in full by some treaty? There is room for doubt, especially when the disputes the jurisdiction is called upon to adjudicate escape entirely from any determination by the national courts, which is in certain respects the case of the Community courts. It would be hard to understand, for example, that in a case of the sort the rule of exhaustion of domestic remedies should not require the matter to be brought before the international jurisdiction so-called because established by a treaty, for the injury incurred by a foreigner, before the state of which he is a national can exercise diplomatic protection. Which would make the Community court, in the ‘national’ matters within its exclusive jurisdiction, a ‘domestic’ court whose collaboration with the other ‘national’ or ‘international’ jurisdictions should, at first sight at any rate, obey a different logic from that which attends its cohabiting with those jurisdictions as an international court.

It is pointless emphasizing that the question would inevitably be complicated by the involvement of two courts that were ‘truly’ international in that they were both established by an instrument of international law but ‘falsely’ international in that they are not called upon to adjudicate a dispute falling substantially within international law. This might have been the case, for example, of the ECJ and the Benelux court had the latter’s jurisdiction not been limited to interpreting the Benelux agreements. The hypothesis is passably theoretical. Will it remain so? That is not sure, especially if, with the development of movements towards economic integration aiding, the ECJ were to be multiplied by cloning as it were.

For international jurisdictions, harmonious co-existence seems to imply fundamentally that the judgments of each of them can be recognized by the others—failing which one would run the risk of trials that might be constantly re-opened—and that they can work together where necessary before judgment is passed so as to meet the requirements of ‘good’ administration of justice so that it becomes ‘better’. The same goes, for that matter, for national jurisdictions when they are ‘foreign’ to one another. This is what makes up the most part of the subject matter of ‘conflict of jurisdiction’ in private international law. The solution to these conflicts proves decisive for reducing the risks weighing on private individuals when they have to cope with a plethora of ‘sovereign’ courts. That the court is ‘international’ changes nothing in this respect for individuals when they do have access to it. However, co-existence may be complicated by the difference in standing between the national court and the international court, at least when involvement in a ‘community of courts’[1], to take up an expression that had its hour of glory, does not invalidate beforehand any conclusion that one might claim to derive from their unequal dignity.

Anyway it seems that these complications have to be set aside when states alone are concerned, as subjects of international law. It would only be otherwise if a hierarchy could be or were to be established among international jurisdictions according to contemporary international law. Apart from what might result from application of UN Charter article 103, such a hierarchy is however lacking as at present. Does it ensure that the position of states before international jurisdictions is wholly comparable with the position of individuals before national courts? Probably not. The difference is at least that the jurisdiction of (international) courts among which cohabitation must be organized invariably relies—at least to date—on the consent of the states concerned. Any ‘dysfunctioning’, in international law, therefore comes down to a problem of compatibility among the various treaties to which those states are parties. Such problems are not generally easy to solve. That, nobody can deny. But they remain very different from those which, for individuals, arise from their forced subjection to the laws of several states over which, directly or indirectly, they have no control.

Still, it plainly has to be accepted that another authority can claim to ‘try’ the matter. That seems self-evident. Without radically contesting it, the Community courts appear to want to substantially restrict the space that the ‘international’ courts can use.

1.1 ‘Accepting’

A court certainly only ever holds its powers from the law of the state that instituted it. And it must exercise those powers on the terms laid down by that law. It cannot in this be subject to any foreign law nor rely on any, except insofar as its national law accepts it or authorizes it explicitly to do so. That much is self-evident.

It is another thing to determine to what extent litigants may or may not address themselves in certain matters to ‘foreign’ arbitrators or courts. It is plain, once again, that each court is subject in this matter to its national law alone, which leaves any foreign court entirely free to do whatever its own law allows. The field of such exclusive jurisdiction seems to be shrinking today, as does the field of non-amenability to arbitration of certain disputes. That is probably one of the effects of globalization on legal practice. The European Union (Community) cannot be seen on this point as displaying excessive ‘nationalism’ of another age. However, article 292 (ex-219) CE does stipulate that ‘Member States undertake not to submit a dispute concerning the interpretation or application of this Treaty to any method of settlement other than those provided for therein (sic)’. In itself the provision is commonplace enough, attributing primacy to one treaty over all others, at least inasmuch as that primacy is not explicitly refuted by a later treaty. Still, agreement must be reached as to its scope. In the interpretation given to it by Community case law, that scope is maximum as evidenced by the now well-known Mox Plant ruling. With regard to disputes among member states, it is evidence of an inclination not just to deny any authority to an international judgment about a question of interpretation of Community law, but also to bring an action against any member state that addresses itself to this end to any ‘international’ court other than the Court of First Instance or the European Court of Justice. That is, not only not to recognize, but also to ‘punish’.

The Mox Plant case involved Ireland’s right to have an arbitration tribunal or the International Tribunal for the Law of the Sea (ITLOS) decide on a point of law raised in a dispute over the lawfulness of the construction of a nuclear fuel production plant in the United Kingdom under circumstances that Ireland felt were incompatible with the provisions of the Convention for the Protection of the Marine Environment of the North-East Atlantic (OSPAR) and the United Nations Convention on the Law of the Sea (UNCLOS/Montego Bay Convention) and which Ireland feared would have catastrophic consequences on the environment. The Commission brought an action against Ireland for failure to fulfil an obligation based on article 226 EC (169 EAEC) before the ECJ which ruled there had been a breach of Community law since the Community was also party to UNCLOS, even if it only has ‘shared’ jurisdiction within that framework. No mention was made, though, of the OSPAR Convention, other than very marginally, and of the arbitration award made based on its provisions, whereas the Community is a party to that Convention and ‘Community law largely inspired the drafting of that international instrument’[2].

There are two bases to the finding: a failure to comply with the duty of ‘loyal cooperation’ with the ‘Community institutions’ devolving on its member states and violation of article 292 EC (219 EAEC) cited earlier.

1.1.1 ‘Loyalty’

It is rather difficult to understand the failure to comply with the ‘general duty of loyalty resulting from Article 10 EC’. It is not that the duty can be contested. Besides, there is no need for article 10 in this respect. Can it be imagined that in the absence thereof a ‘right of disloyalty’ would be asserted? It would be curious to say the least.

What challenges understanding is that it could be found ‘disloyal’ to resort to a tribunal whose jurisdiction, by the terms of the convention instituting it, is incontrovertible. It is a little like criticizing individuals for taking a case to a foreign court on the sole ground that the action could also have been brought before their national court. Which has never been argued… Besides this is not what the Court says, pointing out that the breach of article 292 was enough in itself without there being any need in making that finding to have recourse to any more ‘specific expression of Member States’ more general duty of loyalty resulting from Article 10 EC’[3]. What is incriminated, by contrast, is that Ireland instituted extra-Community proceedings without hesitation; this was supposedly a failure to comply with ‘a duty to inform and consult the competent Community institutions’[4] beforehand that is incumbent upon member states because of ‘an obligation of close cooperation in fulfilling the commitments undertaken by them under joint competence when they conclude a mixed agreement’[5]. In the case in point, the institution involved is the Commission. And it is probably because it was at least unofficially consulted that it did not bring action against Belgium and the Netherlands for resorting to ad hoc arbitration in the Iron Rhine case, to settle their dispute over the conditions for renewed operating of a railway line involving (Community) protection of the environment[6]. Had any authorization been necessary, that failure would not have given rise to any contestation. In its absence, it is a little thin even so to settle for an absence of ‘consultation’ to find there had been a failing, at least in the absence of any (particularly) serious imminent danger… especially as it is hard to see how, the Community being party to the Conventions since jurisdiction was ‘shared’, the Commission could not be informed in some way or other. At any rate, it would have required somewhat extraordinary circumstances for the mere fact that is was not ‘informed’ or ‘consulted’ to be held to be prejudicial to the Community, even if it is considered that it would have been ‘civil’ to do so.

1.1.2 Article 292 EC

Any point of loyalty apart, the ECJ holds that recourse to international arbitration in the case at hand is a breach of article 292 EC since ‘the [UNCLOS] Convention provisions on the prevention of marine pollution relied on by Ireland, which clearly cover a significant part of the dispute relating to the MOX plant, come within the scope of Community competence which the Community has elected to exercise by becoming a party to the Convention’[7]. Of itself, one cannot see why it follows necessarily, as it were, that the ‘Community’ mechanism for dispute settlement should be preferred to the international apparatus. Apparently, it would seem to be UNCLOS itself that decides so in its article 282 on ‘Obligations under general, regional or bilateral agreements’[8]. This article is far from being entirely clear. If its scope is indeed that attributed to it by the ECJ, it would have to be inferred that what Ireland is criticized for in the case at hand is less its failure to comply with article 292 EC than the absence of an agreement by the UK on the resort to international arbitration since UNCLOS article 282 only gives preference to the Community mechanism ‘unless the parties to the dispute otherwise agree’, which is hardly compatible with the grounds in the decision.

Beyond all the questions on the subtle/complicated distinctions made between exclusive, shared jurisdiction, etc., it is primarily the interpretation of the concept of dispute that is fundamentally at issue. In the case in point, the litigation is about the lawfulness of constructing and operating a nuclear fuel reprocessing plant in the UK by a specialized UK public company, both activities having been duly authorized by the authorities[9]. Of itself, the matter does not call into question the relations between two states in their capacities as subjects of public international law. It only comes into the arena of public international law because the Irish authorities, out of concern about the potential consequences for the quality of the waters of the Irish Sea, asked in vain for further information from the UK authorities. Hence the OSPAR and UNCLOS conventions cited above… In other words, the inter-state dispute is nothing more than an ‘incident’ arising within the context of legal proceedings whose specific object relates to the circumstances of the construction and operation by a public company of a plant whose activity it is feared might have very detrimental effects on the quality of the environment. That in itself has nothing to do with public international law. And it must be observed that to date the dispute persists, the judgment of 30 May 2006 having left unanswered the important questions raised.

Can a disagreement about the scope of a duty of information that must allow all the interested parties to appraise the lawfulness of the construction and operation of a nuclear fuel reprocessing plant be ‘a dispute concerning the interpretation or application of Community law’, on the sole ground that the Community is party to conventions or has unilaterally adopted provisions the scope of which may be debated on this occasion? In Iron Rhine, the arbitral tribunal implicitly answered no to the question. If we answer yes, there are surely not many disputes that could ever be submitted to an international court so substantial are the developments of Community law and so numerous are the conventions to which the Community is a party, either alone or alongside its member states. It would then have to be acknowledged that the international court is in some sense condemned to growing unemployment; and the Community court to ever more Herculean labours since it is the exclusive court in the final analysis for any interpretation to be made of Community law.

It would probably be more reasonable to consider that a dispute about Community law within the meaning of article 292 EC is one whose exclusive or at least main subject matter pertains directly to the interpretation or application of Community law. And it is inferred from this that disputes whose primary object comes under public international law must be submitted primarily, in the very interests of the international community where the need for its own case law is felt, to the settlement mechanism it provides. In Iron Rhine opposing Belgium and the Netherlands, the arbitral decision referred explicitly to article 292 of the Treaty establishing the European Union. The arbitral tribunal inferred that it was in an ‘analogous’[10] situation to a domestic court, and that it could or could not adjudicate on the point of Community law before it depending on whether the interpretation of which it was seized was decisive or not for solving the dispute, which it was not in the case at hand[11]. Which remains somewhat approximate… It is of little matter in this respect that it is two member states of the (European) Community that are involved. It is hard to see what could really explain a conclusion to the contrary. After all, the Community will plainly not be bound by res judicata, which is determined by the subject matter of the claim in question much more than by the arguments used to support it. And any res judicata apart, it has sufficient legislative, judicial and political resources to defend itself if perchance its rights are threatened or its interests in jeopardy… including resort for failure to comply with an obligation and the financial sanctions to which it may lead when such a failure is not rectified after due finding by the courts. Or at any rate, should one not concur in this conclusion, the danger justifying that it be set aside would need to be clearly identified. The idea, that haunts the Court’s case law, is probably that any ‘risk that a judicial forum other than the Court will rule on the scope of obligations imposed on the Member States pursuant to Community law’[12] must be categorically prevented, whenever the Community might find itself weakened by it. Whereas it is commonplace for a jurisdiction to interpret a foreign law… In any event, the risk would only be real if its institutions could be bound by res judicata, but one fails to see how they ever could be if the Community is not a party to proceedings. This is somewhat reminiscent of the controversies surrounding the possible review of constitutionality of a foreign law by a court of the forum in private international law. Mainstream opinion considered that such review was inexpedient. No one, though, has ever argued that the constitutional edifice of the state in question might seriously be shaken by it.

Were the logic of the judgment to be followed through, we would inevitably come to construe article 292 as prohibiting not only two member states from calling on diplomatic techniques to settle their dispute[13], but also as prohibiting one of them from going to an international court to settle a dispute opposing it to a third state on the sole ground that it challenges the interpretation or application of the EC Treaty, which plainly cannot be ruled out. That would be a genuinely schizophrenic logic. It is hard to see at a time when, like it or not, the world is ‘opening up’, the Community court restoring a form of enclosure from which domestic courts have barely just escaped, by shutting itself up in what has been termed its ‘jurisdictional territory’[14]. That would be both to lack confidence in others and to over-rate oneself, which seem compatible neither with the ‘good administration’ of justice nor with the ‘good order’ of international relations. The community of courts is admittedly not yet fully fledged. Even if it were doomed to remain so, it would nonetheless be right and proper that no court should disregard as a general rule either the freedom of the litigants to choose their court when they have access to it nor the advantages they might have in taking their case to it, which is categorically excluded by the ECJ in the Mox Plant judgment[15]. And it does not matter in this respect that some have judged it would have been preferable, in the plaintiff’s own interest, for it to have gone to the Community court rather than to an arbitral tribunal[16]. That is a matter of expediency and not of law.

1.2 ‘Recognizing’

The recognition of the judgment of one international court by another international court fundamentally implies that the latter sanctions the former’s decision as res judicata, that is, it does not question what has been definitively adjudicated. The parties may admittedly have rights and obligations that the first court recognized or conferred, but if they have not validly employed them it is not for the second court to contest them, even if it may, as the case may be, infer from them consequences other than those that both parties or either of them might wish. It is true that it is not always easy to fix the bounds of res judicata; but the principle of it goes uncontested.

It is common knowledge that in domestic (national) legal systems such recognition is fairly broadly granted to ‘foreign’ judgments, provided certain elementary requirements are satisfied, especially with respect to the observance of the rights of the defence without which one can scarcely conceive of the fair trial to which everyone is entitled. One fails to see, a priori, what could explain things being different when international courts are involved. It may be that the conditions to which the recognition of the ‘international’ judgment is subordinated are not identical in all particulars to the conditions on which the recognition of a ‘foreign’ judgment depends. But with this proviso, such recognition should be secured here as elsewhere. Even so, for the question to have any point, the second court must have a comparable if not identical power of adjudication to the first court. Otherwise, it may or may not refuse to sanction the new legal position created by the judgment, but anyway it has no capacity to re-try what has been tried by another court. Just as it may refuse to sanction a foreign administrative act even if it is powerless to alter it in any way.

In private international relations the claim to re-try matters is ‘natural’ enough whenever the terms in which each state defines the jurisdiction of its courts are substantially very close if not absolutely identical. Were res judicata not to be recognized, there would then be scarcely any obstacle to a new judgment should the initial plaintiff or defendant so wish.

Things are different in relations between states whenever the jurisdiction of the court still depends on an agreement between the parties (or on the decision of an international organization, the binding force of which is also based in treaty). The problem can only then arise if two or more states have, in two or more separate conventions, given the power to two or more courts to hear virtually the same dispute. In itself there is nothing very exceptional about this hypothesis. It is borne out, for example, when a state’s very broad recognition of a non-specialized court’s jurisdiction (say because of unreserved adhesion to the optional clause of mandatory jurisdiction in article 36(2) of the ICJ statute) exists alongside an attribution of jurisdiction in a given matter to a specialized court (say the European Court of Justice and the Court of First Instance). Should this possibility be deemed inopportune? There is no reason why. Why condemn a priori the choice available to parties between several formulas, which results not only as in the previous example from them adhering in succession to two or more separate instruments but may also have been explicitly provided for in a single instrument, as is the case for example in UNCLOS article 287? Even so, for the second court to be seized of the case decided by the first, either the parties must agree to submit the dispute again, res judicata notwithstanding, or the court must refuse in the absence of any such agreement to recognize the initial decision as res judicata. The first hypothesis, which amounts to organizing a form of appeal, raises no difficulties per se. It is true that nothing in contemporary practice explicitly condemns the second, in the absence of any precedent. It is hard, though, to see that it should not be deemed contrary to a general principle of elementary law; what is not accepted in private international law in relations between legal systems that are ‘foreign’ to each other seems even more difficult to accept within a ‘single’ legal order of which public international law is the embodiment, even if the court is still, to date, based on treaty.

It seems the issue has never been raised in Community practice, but there is nothing to suggest if it were that any different solution would be forthcoming. It is true that the Luxembourg court has more than once refused to sanction the obligations resulting for one or all member states from decisions made by another international court. It has, though, never claimed to re-try, assuming it had the power to do so, matters that have been tried before by another court or tribunal, whether of human rights, of the WTO, the EFTA, or any other. Generally it has sufficed for it to give or in most cases refrain from giving effect to the obligation arising for the interested parties from a decision taken elsewhere[17]. It is not that it has necessarily adopted on the merits provisions that are substantially different from those that were ‘imposed’ by the international court. But it has given them at least an exclusively Community basis, that is foreign to public international law as such.

The best example is provided by human rights. In the wake of the Strasbourg Court’s Matthews case law, the Luxembourg court did not fail to closely review compliance with human rights by the Community authorities. But it did so on the sole basis of Community provisions that it holds to be analogous to those of the Rome Convention, which in the case in point is only one of the ‘references’ to be relied on in establishing what these fundamental rights are. It is true that the Community is not party to the Convention and to its supplementary protocols. Does that alone exclude violation of the Convention from being directly sanctioned as such? This can be doubted from the moment all the members states are. It may be deemed politically inopportune; but technically it does not seem to raise insuperable difficulties.

It is remarkable that this approach, that is not without reason according to its own logic, almost entirely eschews the specifically judicial dimension of the question. Whether it is an international treaty, a Security Council decision or a ruling of the International Court of Justice that is at issue, the conclusion is identical. The Community court gives effect or not to the ‘international’ obligation depending on whether or not it seems opportune to it, in line with the not always readily comprehensible view it forms of what European integration is or ought to be and of what is good for it. In such a context, it can be readily understood that it is of no matter whether the ‘other’ (international) court is a ‘real’ court or not. Whether it is the (real) court of human rights or the (false) court of the WTO for example that is at issue, the important thing is merely to check whether or not there is any obligation for the Community authority, and if so whether or not there is reason to review compliance with it or oversee its performance.

Is such a policy beneficial for the Community, of which the Court is a component part and on which substantial freedom is thus indirectly conferred? It is surely to be hoped so. Is that policy (always) compatible with the best interest of (each of) its member states or their nationals? That is less certain, although it is true that it is not for the court to specifically defend that interest.

1.3 ‘Collaborating’

Collaboration between courts across borders does not lose all interest simply because the judgment made cannot be recognized in any event by the court requesting such collaboration. It may be useful for a national forum to ask for the support of a foreign forum so as to properly (better) adjudicate, even if the national court will refuse to recognize the foreign court’s judgment. It is clear, nonetheless, that such collaboration is all the more warranted if such recognition is not refused as a matter of principle.

In Mox Plant, three international jurisdictions were seized in succession: two arbitral tribunals respectively formed pursuant to the OSPAR convention and to UNCLOS (Annex VII) and the International Tribunal for the Law of the Sea (ITLOS).

The OSPAR tribunal and ITLOS—the latter was only seized with a request for provisional suspension of the authorization to construct the plant in dispute[18]—found they had jurisdiction but ruled the application unfounded; neither evoked the questions of Community law that the matter raised, directly or indirectly. Initially the Annex VII tribunal ruled it had jurisdiction despite the ‘parallel’ action under the OSPAR convention. But it subsequently decided to suspend proceedings pending an answer to the issue of Community law brought before it[19]. The reasons for the suspension do not come out clearly in truth. No application challenging the grounds for Ireland’s claim had been made at the time to the Community Court by the UK or by the Commission on any grounds whatsoever. Consequently there was no reason to enforce the lis alibi pendens rule. The Annex VII tribunal simply referred to requirements of ‘comity’ and of the mutual respect two international courts owe one another. However respectable they may be, mutual respect or courtesy towards a ‘virtual’ court (here the ECJ) alone cannot justify the ‘real’ court (here the arbitral tribunal) not adjudicating one way or another on the matter before it[20]. It may be that it took the application to be ‘impetuous’ given the complications to which it might lead. But it is for the plaintiff to face up to its responsibility in the matter; it is not for the court to substitute for the plaintiff. The ‘frivolous or vexatious’ character of an application means the plaintiff may be liable to the defendant for financial compensation if the defendant makes a claim; it does not give the court the right to refuse to hear the matter. One fails to see how an appreciation by the court of what is in the best interest of the plaintiff might ever authorize the court not to adjudicate, that is, for the plaintiff to obtain a decision on the substance of its claim when there is no (longer any) doubt as to its admissibility. That, however, is not what the Annex VII tribunal thought. The sole outcome of its ‘wisdom’ was the Commission’s introduction of an action against Ireland… which is all the more peculiar as the outcome of a legal action in that the plaintiff’s legitimate concerns about the serious threat from the construction of a nuclear fuel reprocessing plant for the environment have remained unanswered. It may be that the Irish plaintiff ‘bungled’ the case’. Its question deserved better, though, than to see the Community court falling back on its exclusive power to hear disputes involving, even indirectly, the interpretation or application of Community law.

That said, it would not be easy to understand why, in the event of litispendance, the withdrawal of one international court for the benefit of another international court, normally the first before which the action was brought, should not be accepted when the case has the same subject matter, the same cause and the same parties, as a traditional rule requires. The concept of ‘cause’ can alone raise certain difficulties in this respect. A widespread opinion seems to consider that the cause is constituted by the legal rule the violation of which must, according to the plaintiff, allow him to see his application admitted on the merits, whereas it seems rather to rest on the set of facts that the plaintiff invokes in support of his claim… whatever the rule that will enable the plaintiff to ‘see his cause prevail’. That at any rate is what must allow the court to base its decision on a legal rule other than the one the plaintiff asks to be enforced, without disregarding a traditional distribution of roles between the parties in proceedings: to the judge the law, to the parties the facts.

It is self-evident that withdrawal for litispendance raises no difficulty when it is explicitly provided for (organized) by an agreement. It does not seem that this is the case in Community practice, apart from some rather vague provisions analogous to those of UNCLOS articles 281 and 282 that have indirectly become ‘Community-governed’ further to the Community’s adhesion. Should it therefore be excluded? There seems to be no reason to do so. On the contrary, it is a general principle of law that seems to command acceptance to avoid the parties finding themselves in the untenable position of having to comply with two contradictory judgments[21]. But it is a good bet that standing down in favour of the first court seized will only be done easily if the second does not esteem its ‘dignity’ or its ‘knowledge’—whatever these terms might mean exactly—too highly to give up to it the responsibility of adjudicating alone. This is probably what was evoked in the rather Sibylline terms[22] used by the PCIJ in Certain German interests in Polish Upper Silesia (question of jurisdiction) when called upon to examine a problem of litispendance. At the risk of things turning out for the worse on the pretext of wanting them to turn out for the better.

It shall be said that, in private international relations, the objection based on litispendance cannot be entertained as a general rule unless a treaty provides for it, which might prompt one to subordinate it to the same requirement in public international law[23]. That is to forget that, even if to date they necessarily have a treaty-based foundation, all the ‘international’ courts and tribunals are the mouthpieces of a ‘single’ legal order, even if it is still a stuttering one in many respects. Except if the convention to which they owe their power to adjudicate explicitly excludes it, withdrawal should probably be admitted, then, as a matter of principle. Which has nothing to do with the coexistence among organs of sovereignties ‘foreign’ to one another, across boundaries that limit the exercise thereof.


In internal (national) orders, the exercise of justice is in the hands of what is called ‘judicial’ power. Its fundamental attributions are normally laid down by the constitution and the conditions for exercising them are specified by the legislator, who shares with the executive the responsibility for the proper working of the judicial apparatus. The precise configuration of collaboration thus established among the three traditional powers may vary from one state to another. It does not, however, call into question the principle of independence of each of them and especially of the courts when ruling on the decisions made by the legislator or the executive in the exercise of their specific responsibilities. There are necessarily ‘interactions’ among the ‘constitutional’ powers the specifics of which are normally determined by statute in compliance with the constitution but that it is for the court itself to establish when they do not explicitly transpire from one or the other. It is hard, even so, not to be struck often by the reluctance of the courts to extend their hold over the other two traditional powers beyond what the law explicitly orders, at the risk for that matter of sometimes disregarding the rights of individuals. National traditions are decidedly far from being identical on this point, and the techniques used to this effect are disparate (act of government, political questions doctrine, appeal against a decision contravening a statutory provision, and so on). It may be that some courts appear too timid and others particularly audacious. That is a question of appreciation. In substance, though, it is hard to see how one could radically challenge these expressions of natural solidarity among the three powers that assume responsibility for the exercise of sovereign power.

This ‘rationale’ is hardly applicable in international (inter-state) relations. It is not that the idea of an international ‘community’ can be dismissed, however faltering it remains in many respects. It is just that it does not to date have its own organs among which the exercise of power might be shared, unless it is considered that it is expressed with the aid of the organs of the United Nations… which remains, for the time being, a bold claim. It is only ever within the bounds of the convention that put them in place and in respect of the specific objectives it pursues, and not of the requirements of some sort of common good that might transcend them, that international ‘authorities’, including the courts, are called upon to collaborate with each other. That of itself does not rule out any adjustments comparable to those attested in national case law. The role of the courts, however, is small by the very fact that the ‘power’ they are generally recognized to hold is limited; the International Court of Justice, for example, has carefully avoided to date ruling on the validity/lawfulness of Security Council resolutions although it could without too much difficulty have done so had it wished. And dissenting opinions sharply reminded it at the time that it supposedly did not have power to do so[24]. It is not that it has been sought to preclude the courts from defending a form of general interest, especially when the United Nations is at issue. It is simply that one seeks to make it known the courts that the best way to defend such an interest would be to refuse any implication of the Security Council. Which probably has only full convinced the major powers that for the time being control the Council. The game is probably not entirely different from that which is played in the municipal (national) orders. But its horizon is bounded in international law by the founding convention, even if its ambitions are particularly high in the case of the San Francisco Charter, and the weight of the courts is generally (very) limited, making it difficult for them to escape from a form of forced submission. This is certainly not always the case, as singularly attested by the European Union (Community). But the example remains isolated, outside of the other undertakings of ‘integration’ that plan to follow the model and which—whatever the initial intent—largely end up substituting for much rather than serving the states that put them in place.

Whatever the (internal) practices followed within a treaty-based group entity, it is a separate question to decide on the (external) relations that its organs must or may entertain with another treaty-based group entity. It is self-evident that any relation is excluded if the parties to the treaty have excluded in advance any relation of the international organization—to take the most usual hypothesis—with third parties. The possibility can only be exceptional, or the very idea of international ‘personality’ would be meaningless. That said, the states parties to the conventions instituting these group entities (organizations) can never settle anything other than what concerns them themselves. It is for them to make relations with third parties possible or not, especially for organizing essential collaboration, but they alone cannot determine their legal regime. The simplest thing would be for rules of general international law to be developed to this end. They remain pretty much non-existent to date. In the absence of such rules, one has to settle for rather vague ‘principles’ at least until such time as specific agreements are made by the interested parties.

There is a void with which the (passive) co-existence among authorities can cope without too much difficulty but that hardly facilities any (active) collaboration among them, at least when they have no specific jurisdiction to decide on it. Which brings us back to the international courts. General principles may, as said, warrant them recognizing the decisions of other international courts or even of collaborating with them for the good (better) administration of justice especially by accepting to relinquish jurisdiction in their favour within the bounds of what is governed by litispendance. But it is quite another thing to collaborate with a ‘political’ authority in the absence of any agreement to that effect.

That said, one fails to see why it should be prohibited for the court to collaborate ‘actively’ in the absence in its statute or in any other regime applicable to it of any such explicit provision, provided of course that it does not overstep its judicial functions. The question is not so much as to whether active collaboration may be prohibited; it is much more to determine whether the court can be compelled to so collaborate. Absent any agreement explicitly providing for this, one fails to see how this could be, on the basis of general international law. There is probably just one exception that should be admitted; that available to the Security Council whenever its decision-making power cannot be restricted by the effect of any specific convention, under article 103 of the UN Charter. This is a little of what is at issue in the decision in Yasin Abdullah Kadi and Al Barakaat International Foundation v. Council of the European Union of 3 September 2008 by the European Court of Justice[25].

The case is a familiar one involving ‘intelligent’ sanctions by the Security Council in the wake of the attacks of 11 September against certain persons suspected of having ties with Usama bin Laden, Al-Qaeda and the Taliban. Those sanctions took the form of restrictions on people’s movements and the freezing of funds and other financial resources they held. Community regulations were passed to give effect to them materially, pursuant to the normally applicable procedure involving the European Council, Commission and Parliament. Since the Maastricht Treaty, however, a common position had first to have been adopted by the EU Council.

Initially each of the parties took the matter to the EU Court of First Instance arguing that the Security Council decision that the Community was seeking to enforce was unlawful in that it violated certain fundamental human rights and particularly the right to defend oneself against any accusation made against one or the right to respect for property. In a somewhat confused judgment which has been commented upon in this Yearbook[26], the Court found it had the power to review the compatibility of the measures decided by the Security Council with jus cogens, even if it was not for the Court in principle to review the lawfulness of the Council’s resolutions. On the merits, though, it dismissed the plaintiffs’ claims. Harsh as they were, the sanctions did not appear to the Court to overstep the restrictions that might lawfully surround the exercise of their rights ‘measured by the standard of universal protection of the fundamental rights of the human person covered by jus cogens’[27], to use the Court’s terms to summarize this point of the CFI’s decision which was appealed before the ECJ.

In its decision of 3 September 2008, the ECJ decided the case on its merits, without referring it back to the CFI considering that it was in a position to give a final judgment on the claims. It annulled the contested regulations inasmuch as they concerned the plaintiffs, considering that neither their fundamental right to an effective resort to the courts nor their defence rights had been respected, no more than Mr Kadi’s fundamental right to respect for his property. However, that did not prevent the ECJ from maintaining the effects of the annulled regulation for a period not exceeding three months. That provisional enforcement, which was a priori rather odd, was justified by the circumstance that any ‘annulment […] with immediate effect would be capable of seriously and irreversibly prejudicing the effectiveness of the restrictive measures’[28] imposed by the Security Council and that ‘it cannot be excluded that, on the merits of the case, the imposition of those measures on the appellants may for all that prove to be justified’[29] insofar as it was fundamentally the procedure rather than the substance of the measures adopted that came in for criticism. In other words, three months were left to correct the failings observed… This was perhaps rather vain if it were to be counted that the Security Council might accept to fundamentally revise its opinion in ninety days. It was less so if such revision was expected of the Community performer of the obligation alone… at the risk, which it is true is far more uncertain than a new plea to the Community courts, that the member states might have to face the Security Council’s wrath.

The central issue in the case in point is to determine whether the plaintiffs’ fundamental rights have been ignored. If those rights have been respected, there is no call to dwell on the powers of the Community courts when Security Council decisions are at issue. If those rights have been ignored, as the ECJ ruled in the case in point, the question becomes one of determining whether the Security Council is entitled, under Chapter VII of the UN Charter, to disregard the rights the violation of which has been established. And if it is held that the Council is not so entitled, it remains to be confirmed whether or not the primacy that the Charter is recognized to have over any other convention, as its article 103 asserts, prohibits the courts of a UN member state—even if common to several of them as with the Community courts—from refusing to apply a Security Council decision even if the Security Council is not entitled to ignore the fundamental rights of persons.

This is plainly not the approach taken in the case at hand by the courts. All other questions of pure Community law apart—such as the basis on which measures to enforce the Security Council decision may be taken—the ECJ settles purely and simply for reviewing that the measures are consistent with Community law, emphasizing that, in a Community ‘based on the rule of law’ like the European Community[30] ‘the obligations imposed by an international agreement cannot have the effect of prejudicing the constitutional principles of the EC Treaty, which include the principle that all Community acts must respect fundamental rights’[31], which inevitably entails their annulment when, as in the case at hand, they seriously disregard those rights. If need be, the ECJ specifies that it is not for it ‘under the exclusive jurisdiction provided for by Article 220 EC, to review the lawfulness’[32] of the resolution the performance of which is at issue, it being understood that any censorship of such performance ‘would not entail any challenge to the primacy of that resolution in international law’[33]. The solution might have been different had the Charter imposed ‘the choice of a particular model for the implementation of resolutions adopted by the Security Council under Chapter VII’;[34] as this is not so, one cannot refuse the Community authority the right either to decide alone on the arrangements for performance or to ensure that they comply with the EC Treaty requirements.

It is as simple as that. One may well wonder, besides, what explains the Court’s delay in justifying what it calls ‘immunity from jurisdiction for the contested regulation’[35], at the risk of making its line of thought somewhat obscure at times. It is probably that it had to reply to the argument brought before it; it is perhaps also that it wished to preclude in some way the criticisms that any apparent ‘disdain’ for the United Nations might prompt. Besides, it explicitly speaks of ‘the deference required of the Community institutions vis-à-vis the institutions of the United Nations’[36], without contesting the principle of it but refusing—not without reason—to sanction the limitation of its jurisdiction that it seems to induce. But it says nothing of article 103, which goes beyond deference.

This is a very foreign hypothesis to that before the Luxembourg court in Mox Plant. A common point of both decisions, though, is that the disputes were adjudicated by taking into consideration Community law alone, without finding any direct support in international law. However, it is specific to the Kadi decision that it substantially sets out why the solution is not contrary to international law and in particular to UN law, whereas the Court in Mox Plant cares not a whit for the implications that might arise for the implementation of other international conventions (rules) out of the provisions of Community law whose interpretation was controversial. The outcome in terms of collaboration has been said to be the same. It is not so sure… even if it is true that collaboration with a political authority like the Security Council is certainly not identical to the collaboration a court should, or at least might employ for the needs of the proper administration of justice.

The interesting point is that in Kadi there was apparently no international court that could have jurisdiction whereas in Mox Plant there were too many; and that the question was particularly intricate in that it involved a Security Council decision possibly being void for breach of jus cogens. But there is no UN court before which matters can be brought. The International Court of Justice cannot be seized either by individuals or by or against an association of states, whether or not it takes the form of a classical international organization. And had two EU member states come to an arrangement as it were to submit the question to it by feigning a dispute, they would have exposed themselves to being condemned by the ECJ for failure to comply with article 292 as happened to Ireland in Mox Plant since such an action would have challenged the lawfulness of measures taken by the Community in the exercise of its jurisdiction to execute sanctions decided on by the Security Council… There is nothing for it then but to put oneself exclusively in the hands of the Community courts… or of the national courts if by chance the measures imposed by way of sanctions fall within their exclusive jurisdiction, which was not so in Kadi. At the very least when people’s basic rights are at issue, it is hard to see how they could refuse without disregarding the fundamental requirements of the rule of law which must be observed whether it is a state or any other community organized autonomously that is at issue. There can for sure be diverging opinions as to the compatibility with basic rights of sanctions decided by the Security Council, especially depending on one’s judgement of the balancing point that must be found in each specific instance between collective needs and individual prerogatives. Some may not be convinced in every respect in the case at hand of the soundness of the grounds for the annulment of the provisions which were deemed incompatible with the observance of the right of each person to have access to a court or to his right of property. But the annulment could not in any way be held to be groundless.

It is true that this does not dispense one from questioning the scope that should be given in such cases to Charter article 103. But it can perfectly well be argued that the article only entails the primacy of UN decisions over measures taken under contrary conventions on the proviso that those UN decisions are not void. There would be some logic in that, especially in the absence within the UN of any mechanisms for ensuring the validity/nullity of decisions the implications of which can be very serious for people’s lives. To which others will answer that this might excessively weaken a collective security endeavour; which would not be a totally unfounded fear.

That is not at all the road the ECJ has gone down. Quite the contrary, it discharges the courts of any review of the measures decided by the Security Council. It simply imposes upon the them ‘the review, in principle the full review, of the lawfulness of all Community acts in the light of the fundamental rights forming an integral part of the general principles of Community law, including review of Community measures which […] are designed to give effect to the resolutions adopted by the Security Council under Chapter VII of the Charter of the United Nations.’[37] The solution is an effective one, safeguarding as it does a review without challenging a certain natural precedence of the UN that is not evoked as such. However, it is only effective insofar as a decision on sanctions is not judged directly applicable within the meaning of international law, which might be defended per se but is unlikely ever to be defended especially in view of the Kadi case law. Where is the ‘collaboration’ in all that? It is certainly not very ‘active’. The fact is simply that the solution discharges the United Nations of what should be its responsibility: to assume responsibility for sanctions decided upon by the Security Council not only politically in the UN but also legally if need be in the courts. The technique is not specific to the ECJ. It is also used by the ECHR when measures taken by member states of the Council of Europe to conform with obligations required of it within the framework of the UN or of the European Union are contested before it. It is not that in the absence of any UN or European review it would have been prohibited for a national court or the ECJ to rule on the conformity of such measures with ‘fundamental’ rights and on the implications for the case of UN Charter article 103. It is simply that any direct review might have weakened associate bodies that, at least as far as the UN is concerned, are not yet fully consolidated. And perhaps such a claim—although fairly natural if one accepts the requirements of (democratic) rule of law—might have verged on arrogance in the view of those who either are not convinced of its soundness or who resist the court’s claim to be the ultimate mouthpiece of the law on all matters.


There is surely no call to conclude these few reflections on the solutions provided by judgments based on reasons that were surely far more complex, or at any rate more complicated, than the foregoing comments suggest. No one will contest, besides, that the situations were tricky and there was no obvious solution for deciding the disputes before the ECJ. It is no less striking that in both instances the ECJ only considered, that is, was only attentive to, Community law, that is, to the internal law of the group entity of which it is part. Which amounts to saying that it does not apply any law other than its own, and does not accept that a dispute involving its meaning can be submitted to any other instance than itself, at least when the ‘obligations’ of EU member states may be at issue. And in truth one cannot easily see when, because of their capacity as members, they could ever not be… There is no reason to be overly surprised as such about this special attachment of an authority to its own law. This is fairly natural, as national case laws have long confirmed… and supposedly beneficial, unless it is believed that the court’s national law is necessarily less good than some ‘other’, which would be odd to say the least. Even if there is no blameworthy servility in that, it hardly justifies its seeming to fairly systematically want to remain ignorant of the rules in force abroad. And such ignorance is in any case hard to understand when the requirements of an ‘international order’ are at issue of which the state, international organizations and the original entity that the EU constitutes are components of it. It is hardly possible to put international law and foreign law on the same footing from this standpoint. If this is how it is, one ought probably to identify whoever such ‘ignorance’ is seeking to protect, for it is difficult to believe that it is not, even misguidedly, ‘well’ intentioned.

It is probable that, in internal orders, reluctance to take international rules into consideration directly, and even more to apply them, primarily reflects the relative weakness of the balance achieved among its powers. And it is probably a certain ‘sovereignty’ of parliament, that is, of statute law, that the courts seek primarily to safeguard, from fear of seeing the undertakings given by an executive roving very feely in international relations compromise observance of rules that are deemed to be elementary. It is probably not that the courts favour one power rather than another. It is rather that they are concerned for the rights of individuals who do not usually have the means of defence or of pressure with respect to the executive that they may bring to bear when it is the legislator who purports to infringe their rights. That may explain in particular why the courts have for so long refused to recognize that a treaty might have a direct effect, that is, confer rights and obligations on individuals, in the absence of any (legislative) measure of execution. And the same goes for customs, excluding exceptionally those which, in matters such as immunities that necessarily concern individuals, are based on centuries-old practices, and are besides often based originally on written ‘rules and regulations’ whatever the specific legal force thereof.

Does the same go within the European Union (Community)? Surely not as far as observance of a certain parliamentary pre-eminence is concerned. Even if its powers have increased markedly, the European Parliament cannot yet seriously claim to make the ‘law’ alone. That does not necessarily exclude the Luxembourg court from displaying special attention with respect to individuals. This is manifest, for example, in the domain of liability, whether it is state authorities or Community institutions that are involved. But is it really a concern to protect people that in the final analysis can explain a certain lack of interest for any law that is not strictly ‘Community’ law? One can but hesitate to admit as much. For it is probably more distrust of states than any will to protect what is deemed a particularly vulnerable category of parties that largely explains a degree of sidelining of international law. This is patent with respect to dispute settlement. One might say that article 292 is clear. It is anything but if it is given the scope that Mox Plant suggests. Perhaps for that matter it will be regretted that the court in that case seemed indifferent to the particularly serious risks that construction of plant held for people… Besides there is not just dispute settlement. ‘Disconnection’ clauses are multiplying in international treaties to which EU member states are parties, subjecting their mutual relations to the provisions of Community law alone, that is, excluding any application of the provisions concluded by treaty. Which does not attest to any great faith in the virtues of international law even if one holds that such clauses are fully consistent with it[38]. It will be said that things are different for the questions of direct effect evoked above. It is true that Community case law tends to submit individuals ‘immediately’ to international rules, not necessarily for that matter to derive any advantage from that but sometimes also so as to be subject to specific duties. Yet it should be observed that this direct effect is largely foreign to international law itself. It seems indeed not to depend in any way on the will of the contracting parties, or by extension of the rule makers, at the risk of making the rule produce effects that were not sought in common and that consequently vary from one state to another. Once again the international rule as such is left aside; all that counts is the effect that it seems useful to confer on it ‘directly’ in the internal order. The direct effect thus understood becomes simply one way among others of executing an agreement or more broadly a rule. Recent Community case law seems increasingly clear on this point. The interesting thing is nonetheless that, according to the Community court, there seems to be in that a power specific to the national court and not a prerogative of any ‘executive’ whatsoever[39]. We again find the distrust of the state. From the Community to the complicity of the courts… The courts as saviour, the justice system that consoles or restores, ... it is somehow part of the current climate.

It is probably wrong to push images too far. It is hard nonetheless not to be struck by the mistrust that the Community court, prompted by the Commission no doubt, constantly displays towards its members, although ‘constitutent’ members… more than fifty years after the founding of a ‘Community’ that seems however to be firmly anchored in European realities. One need not get unduly upset by this. The worrying thing even so is that one sometimes has the impression of dealing with a shepherd of whom one no longer knows whether he is afraid of the wolf or of his sheep; which is not a good thing for any of the three, and not for the passer-by either.

[1] See especially A.M. Slaughter, ‘A global community of courts’ (2003) Harv. ILJ 191 ff.

[2] P. Weckel, ‘Chronique de jurisprudence’ (2003) RGDIP 991.

[3] Paras. 169, 171.

[4] Para. 179.

[5] Para. 175

[6] See V. Barral, ‘La sentence du Rhin de fer, une nouvelle étape dans la prise en compte du droit de l’environnement par la justice internationale’ (2006) RGDIP 652, note 22.

[7] Para. 120.

[8] See para. 125.

[9] On the particulars of the case, see S. Maljean-Dubois and J.-C. Martin, ‘L’affaire de l’usine Mox devant les tribunaux internationaux’ (2007) Clunet 438 ff.

[10] Para. 103.

[11] Award of 24 May 2005, para. 97 ff;  for a general commentary, see P. d’Argent, ‘De la fragmentation à la cohésion systémique : la sentence arbitrale du 24 mai 2005 relative au ”Rhin de fer“’, Mélanges Jean Salmon (2007), p. 1113 ff.

[12] Mox Plant, para. 177.

[13] See Y. Kerbrat and P. Maddalon, note on Mox Plant (2007) Rev. trim. dr. europ. 176.

[14] D. Simon, ‘Chronique de jurisprudence’ (2006) AFDI 734.

[15] See paras. 136 and 137.

[16] See, for example, R. Churchill and J. Scott, ‘The Mox plant litigation: the first half-life’ (2004) ICLQ 672-5.

[17] See generally M. Bronckers, ‘The relationship of the EC courts with other international tribunals’ (2007) Com. MLR 602 ff.

[18] Without entertaining the Irish claim for suspending construction the Tribunal nonetheless required the parties of its own motion, as an interim measure, to consult each other immediately without waiting for the aribtral tribunal to be put in place. Some have seen in this the ‘manifestation of a form of judicial demagogy’ (P. Weckel, ‘Chronique de jurisprudence’ (2002) RGDIP 198).

[19] Order of 24 June 2003, (2003) ILM 1387 ff.

[20] See note U. Röben, ‘The order of the UNCLOS Annex VII arbitral tribunal to suspend proceedings in the case of the Mox Plant at Sellafield: how much jurisdictional subsidiarity?’ (2004) Nord. JIL 244; N. Lavranos, ‘Das Rechtspregungs- monopol des EuGH im Lichte der Proliferation internationaler Gerichte’, (2007) Europa Recht 446-8.

[21] For doubts about there being such a general principle see especially A. Gattini, ‘Un regard procédural sur la fragmentation du droit international’ (2006) RGDIP 313.

[22] See 25 August 1925, Series A, no 6, p. 20.

[23] This is notably the objection that, as early as 1929, G. Ténékidès made to any transposition of private international law solutions into public international law (G. Ténékidès, ‘L’exception de litispendance devant les organismes internationaux’ (1929) RGDIP, 503). Whereas it is ‘national’ practices that might provide inspiration... at least in this day and age.

[24] See the dissenting opinions of President Schwebel and of Judge Jennings in Questions of Interpretation and Application of the 1971 Montreal Convention arising from the Lockerbie Aerial Incident (Libyan Arab Jamahiriya v. United Kingdom), decision, ICJ, Collected Cases 1998, p. 64 ff and 99 ff.

[25] C-402/05 P and C-415/05 P.

[26] See A. Vandepoorter, ‘L’application communautaire des décisions du Conseil de sécurité’ (2006) AFDI 102 ff.

[27] Para. 90 of the Court’s decision.

[28] Para. 373.

[29] Para. 374.

[30] Para. 281.

[31] Para. 285.

[32] Para. 287. Article 220 states ‘The Court of Justice and the Court of First Instance, each within its jurisdiction, shall ensure that in the interpretation and application of this Treaty the law is observed’.

[33] Para. 288.

[34] Para.298.

[35] Even if the formula seems to be spreading in Strasbourg and Luxembourg, it must be recalled that immunity is the prerogative of a person and not the quality attaching to a thing.

[36] Para. 318.

[37] Para. 326.

[38] For hesitations on this point, see especially C. Economidès and A. Kolliopoulos, ‘La clause de déconnexion en faveur du droit communautaire : une pratique critiquable’ (2006) RGDIP 273 ff.

[39] See note on Merck Genericos of 11 September 2007, case C-431/05; cf. Intertanko of 3 June 2008, case C-308/06 in which the ECJ refuses to assess the validity of a Community directive relative to pollution caused by ships in respect of UNCLOS.