Australian Year Book of International Law
This paper is offered as a token acknowledgment of the immense debt that I owe to Don Greig, who gave me my first teaching job at a time when he had more confidence in me than was prudent. For that, and for his friendship, I shall always be most grateful.
In this paper I address the question of ‘concurrent’ or ‘overlapping’ jurisdiction in situations where dispute settlement provisions in two or more treaties each appear to give jurisdiction over a single dispute to a designated tribunal, and the designated tribunals differ. For example, state A might complain of seizure of a merchant ship flying its flag by state B. That dispute might fall within the terms of acceptances of the jurisdiction of the International Court of Justice (ICJ) made by states A and B under the Optional Clause, Article 36(2) of the Statute of the Court. It might also fall within the jurisdiction of the International Tribunal on the Law of the Seas (ITLOS) by virtue of the terms of the dispute settlement provisions in Part XV of the 1982 United Nations (UN) Convention on the Law of the Sea. It might also fall within the dispute settlement provisions of a bilateral treaty of Friendship, Commerce and Navigation, which might give jurisdiction over such disputes to a bilateral Mixed Commission. And the dispute might be said to involve a violation by state B of state A’s trading rights under the 1994 General Agreement on Tariffs and Trade (GATT), and therefore be subject to the World Trade Organisation (WTO) dispute settlement system. Which tribunal has, or which tribunals have, jurisdiction? What if A and B disagree? What if proceedings are initiated in more than one tribunal by state A? or if state B itself initiates proceedings in a tribunal other than that in which state A has instituted proceedings?
This is by no means a purely theoretical problem. As international tribunals proliferate, it is inevitable that the forum for the litigation of a international dispute will often be chosen because of the perceived advantages conferred upon the applicant state by the particular powers and procedures of the selected tribunal, regardless of what one might regard as the ‘essential nature’ of the dispute. How should tribunals react? How should international law approach such problems?
The question of overlapping jurisdictions is complex and it is necessary to cut it down to a manageable size for the purposes of this short paper. I shall, therefore, not deal with multi-party disputes; nor with cases where the validity, or the applicability, of one or more of the treaties said to confer jurisdiction upon a tribunal is in doubt; nor with cases involving issues of succession to jurisdictional obligations. I exclude cases, civil and criminal, involving an overlap between international and municipal tribunals. Similarly, I exclude cases where the overlap concerns not only ‘judicial’ tribunals but also dispute settlement mechanisms of other kinds, such as compulsory mediation or conciliation. Again, I exclude cases where a single tribunal has jurisdiction under two or more distinct legal instruments, and cases where an applicant goes to different tribunals at different stages of a single dispute. I also exclude cases where jurisdiction is conferred on a tribunal otherwise than by treaty, for instance by the operation of the doctrine of forum prorogatum. Each of these excluded aspects of the question introduces particular considerations of fundamental importance, demanding more than a simple qualification to the argument advanced here. I shall focus, therefore, on the simple case of a dispute between two states (I also exclude mixed, state/non-state, disputes) concerning a single situation that falls within the scope of the dispute settlement provisions of two or more treaties by which the disputing states are bound. (I shall, for convenience, refer to ‘first’ and ‘second’ treaties, intending no qualitative or other judgment upon their priority but only to distinguish between them.)
It is possible to distinguish many different kinds of overlap between jurisdictions, but for the purposes of this paper it is sufficient to distinguish three main types.
First there are cases where one tribunal has jurisdiction in respect of a wide category of potential disputes, and a second tribunal has jurisdiction in respect of a narrower category of disputes within a specific field. There are many examples of this kind of overlap. For instance, many specialised tribunals have jurisdiction over disputes that would also fall within the jurisdiction of the ICJ.
Second, there are cases where two tribunals each have jurisdiction in respect of wide and largely overlapping categories of potential disputes. Perhaps surprisingly, there are instances of this type of overlap. As will be explained, one is the potential overlap between the ICJ and arbitral tribunals established under the Organisation for Security and Co-operation in Europe (OSCE) Convention on Conciliation and Arbitration.
To these cases may be assimilated another category of overlap, those where one tribunal has jurisdiction in respect of a relatively narrow category of potential disputes within a specific field, and a second tribunal has jurisdiction in respect of much the same category of disputes. There are instances of this type in the area of the Law of the Sea.
Third, there are cases where one tribunal has jurisdiction in respect of a relatively narrow category of potential disputes within a specific field, and a second tribunal has jurisdiction in respect of a different and relatively narrow category of disputes, but a single set of facts raises issues that fall within both fields. The first paragraph of this paper sketched an example of this type, with the overlap between the Law of the Sea and GATT aspects of the example.
These types are neither pure nor exhaustive; nor is the classification that they suggest unproblematic. They will, however, serve for present purposes. The essential argument of this paper is that each type of overlap gives rise to distinct issues, coloured by the particular characteristics of the international legal system, and that each type demands a different approach.
In cases of the first type (General-Specific) the overlap may be easily dealt with: indeed, the overlap is more apparent than real. It is almost axiomatic that the jurisdiction of international tribunals derives from the consent of each state party; or, to put it another way, no state can be obliged to submit to the jurisdiction of an international tribunal unless that state has at some point consented so to submit and its consent remains, as a matter of law, effective. A corollary of this principle is that in order to establish the scope of a tribunal’s jurisdiction it is necessary to determine what the precise nature and scope of the agreement between the parties might be. Matters such as forum prorogatum having been put to one side in this paper, that consent is to be found in agreements — treaties — made by the parties, or in unilateral acts (such as declarations made under the Optional Clause in the ICJ Statute) that have an equivalent effect. This first case of overlap therefore represents situations where there are two treaties, one general and one specific, each of which ex hypothesi provides settlement procedures that are applicable to the dispute in question, and each of which confers jurisdiction on a different tribunal.
In such a case the legal relations between the parties cannot be determined by construing either one of the agreements in isolation from the other. It must be determined whether one agreement has modified the application of the other, and if so, exactly how this has been done. The question of jurisdiction is thus one that may be approached via the Law of Treaties. It is assumed that both of the treaties (in which category I include comparable unilateral acts) are valid and that they therefore are prima facie effective to confer jurisdiction upon the tribunals in question.
If the respondent state simply challenges the jurisdiction of the tribunal in which an action is first begun (the ‘first tribunal’), the matter will be dealt with as a straightforward challenge within the scope of a single tribunal. The question is simply whether or not that tribunal has jurisdiction; and in this example the answer is presumed to be, yes. If, however, the respondent challenges the jurisdiction of the first tribunal by claiming that the matter ought properly to be heard by the other tribunal (the ‘second tribunal’), the problem of overlapping jurisdiction is raised.
There is an initial question whether there is a problem here at all. May not the first tribunal take the view that if its own jurisdiction is once established it has no concern with the possible existence of concurrent jurisdiction in other tribunals? The answer to that question in this context must be, no. Either on the basis of argument by the objecting respondent, or in discharge of its own duty to satisfy itself of its jurisdiction, the tribunal must consider whether the effect of the jurisdictional provisions in the treaty that appear to establish its jurisdiction may have been modified by the treaty that purports to bestow jurisdiction upon the other tribunal. There are two main ways in which such a modification might have occurred.
First, it is in theory possible that the second treaty might have modified the first in accordance with the principle, set out in Article 30 of the Vienna Convention on the Law of Treaties, that the earlier treaty continues to apply only to the extent that its provisions are compatible with those of the later treaty. In practice, however, it is unlikely that there will be any true incompatibility. Provisions conferring jurisdiction upon a number of separate tribunals can co-exist without incompatibility, unless one or more of the provisions confers, or purports to confer, exclusive jurisdiction upon one or other tribunal.
If the first treaty confers exclusive jurisdiction on one tribunal, and the second treaty purports to confer exclusive jurisdiction upon a different tribunal, there would be a genuine incompatibility, and the principle set out in Article 30 of the Vienna Convention might come into play. Such a conflict would be unusual, and almost certainly the result of an oversight: the expectation would be that one of the treaties (probably the later in time) would deal explicitly with the potential conflict and resolve it.
Much more likely to arise is the situation in which a non-exclusive jurisdiction clause co-exists with a clause conferring exclusive jurisdiction upon a different tribunal. This is, for example, the case in relation to states such as Finland and the United Kingdom that have both made declarations under the Optional Clause in the ICJ Statute and bound themselves by Article 292 of the Treaty Establishing the European Community (EC) not to submit a dispute concerning the interpretation or application of that Treaty to any method of settlement other than those provided for in the Treaty. The provision is couched in terms of a duty upon each state party not to submit the dispute to another, non-EC, tribunal, and is in that sense expressed as a constraint upon the freedom of action of the state rather than an explicit and direct constraint upon the jurisdiction of the tribunal. There can, however, be no real doubt that a provision such as Article 292 of the EC Treaty operates so as to limit the jurisdiction of the ‘other’ (non-EC) tribunal. It does so by modifying the effect of, in this example, the acceptances of the jurisdiction of the ICJ, in what is the second way in which modifications may operate: by the application of the principle lex specialis generalibus derogat. That is to say, the more specific provision prevails over the general provision.
That principle is widely recognised to be a general principle of international law appropriate for application within the international legal system; and it reflects the fundamental importance of the consent of the parties as the basis of the jurisdiction of international tribunals. In circumstances where the parties have made special provision for a certain category of disputes, in the absence of any indication to the contrary it must be supposed that they intended that it is this special provision, and not some more general acceptance of the jurisdiction of another tribunal, that they intended should be applied to disputes in that category. If we ask what the parties have bound themselves to do in relation to those disputes, the answer is that they have bound themselves to follow the ‘special’ procedure; and either party may insist that the other fulfils this obligation.
Applications of the lex specialis principle may readily be found in the jurisprudence of international tribunals. The result of the application of the principle is, it should be emphasised, not that the tribunal of general jurisdiction is merely entitled to decide not to exercise its jurisdiction over the dispute, but rather that the tribunal of general jurisdiction must decline to accept the case, because the parties are legally bound to refer the case to another tribunal.
Thus far the discussion has been concerned with cases in which an objection is raised to the hearing of the case by the tribunal first seised (although the lex specialis principle retains its validity regardless of the tribunal in which it is raised). If the objection is raised in the second tribunal rather than the first the matter may be handled differently, by the application of the lis alibi pendens doctrine, which is discussed in the following section.
Jurisdictional overlaps of types 2 and 3 give rise to different considerations. In Type 2 situations, the premise is that the lex specialis principle cannot be applied. Considered in isolation, each of the two tribunals ex hypothesi may exercise its jurisdiction in respect of the dispute. The question is whether, given the existence of concurrent jurisdiction in the two tribunals, there might exist circumstances in which one of them should decline to exercise its jurisdiction.
One clear example of this type is the overlap that would result if a state were to make a declaration, without significant reservations, under the Optional Clause in the ICJ Statute and were also to make a similar declaration under the equivalent provision (Article 26) in the OSCE Convention on Conciliation and Arbitration, thus also accepting the jurisdiction of an ‘OSCE’ arbitral tribunal. In such cases the jurisdiction both of the ICJ and of the arbitral tribunal might extend to all legal disputes of whatever nature.
Similar problems arise in the slightly different context of overlapping acceptances of jurisdiction over a narrower category of disputes. This situation is not common, not least because the usual practice is for specialised dispute settlement provisions to be stipulated to relate specifically to disputes arising from the interpretation or application of the particular treaty in which they are found. Examples are, however, not unknown. For instance, under the 1982 UN Convention on the Law of the Sea, states may agree in advance to submit disputes arising from the 1982 Convention to one or more of a range of tribunals specified in Part XV of that Convention; and some states have nominated more than one such tribunal. Some such states have indicated an order of preference among their nominations: Austria, Chile and Germany are examples. Others, such as Belgium, Finland, and Oman, all of which nominated both the ICJ and the ITLOS, have not expressed any preference between the two; and Italy, which nominated the same two tribunals, explicitly stated that it did so ‘without specifying that one has precedence over the other.’
In all of these cases no evident priority is established, either by virtue of the choice of the parties or by application of the lex specialis principle. What, then, is the position if one party initiates proceedings in one tribunal and the other party either argues that the proceedings should be conducted in the other tribunal, or itself initiates proceedings in that other tribunal? There are certainly reasons why a party might wish to move the matter into another tribunal. For example, a party might be reluctant to endure the waiting period that now commonly attends proceedings in the ICJ; or it might wish to avail itself of procedural advantages available in the other tribunal.
Indications by international tribunals as to the proper approach in these circumstances are scant. One consideration was spelled out by the PCIJ in the Chorzów Factory (Jurisdiction) Case, where it said that
… the Court, when it has to define its jurisdiction in relation to that of another tribunal, cannot allow its own competence to give way unless confronted with a [jurisdictional] clause which it considers sufficiently clear to prevent the possibility of a negative conflict of jurisdiction involving the danger of a denial of justice.
In other words, the Permanent Court would not consider declining to exercise its own jurisdiction unless it was ‘sufficiently clear’ that there existed another tribunal with jurisdiction over the dispute. In practice it may not be clear that there does exist another tribunal that has jurisdiction in respect of the dispute. The parties may disagree on this question, and the first tribunal may be uncertain whether jurisdiction exists elsewhere. In such a case there should be no question of the first tribunal declining to exercise its jurisdiction as a matter of discretion, if that tribunal has jurisdiction and the claim is admissible. If there is any doubt as to the jurisdiction of the second tribunal, an obvious response would be for the first tribunal to suspend its proceedings until the second tribunal has decided whether or not it does have jurisdiction — a task that falls to the second tribunal by virtue of the principle that every tribunal has the authority to determine the scope of its own jurisdiction.
(It is arguable whether the second tribunal should have to determine not only that it has jurisdiction but also that the claim is admissible if, as the Permanent Court suggested, a ‘denial of justice’ is really the issue. This is not, of course, a ‘denial of justice’ in the sense in which the term is used in the context of indirect international wrongs: rather, it is an analogous principle that may be applied by international tribunals. The basis of the principle is a matter of some interest. There are two justifications that might be offered for it. First, it might be said that there is a presumption that it is desirable that there should be a forum in which disputes may be adjudicated; and accordingly, no tribunal should decline to exercise jurisdiction unless there is an alternative tribunal in which the matter might be heard. Secondly, it might be said that if there is no other tribunal with jurisdiction over the dispute, there is no conflict of jurisdictions or jurisdictional overlap, and that in these circumstances there is no legal basis upon which the first tribunal, once properly seised of the dispute, could refuse to exercise its jurisdiction. The first basis is a policy guiding the discretion of a tribunal deciding whether or not to exercise its jurisdiction: the second basis denies that any such discretion exists where only one tribunal has jurisdiction. But the explanations lead to the same result.)
The situation with which we are left, then, is one where two tribunals of equally broad jurisdiction (so that the lex specialis principle does not apply) each appears to have jurisdiction over the dispute. The problem of overlapping jurisdictions is, of course, neither novel nor confined to public international law. It has received a great deal of discussion in the context of private international law and municipal law generally, where the overlapping jurisdiction of courts is approached through concepts such as forum non conveniens, lis alibi pendens, abuse of process, and res judicata. It might be expected that these concepts offer solutions to the parallel problems in international tribunals. The context in which those concepts operate is, however, essentially different from the international legal context, with which this paper is concerned. Because these differences preclude, in my view, the straightforward translation of concepts such as forum non conveniens onto the international plane, it may be helpful to say a little more about them now. Although I shall refer to characteristics of ‘a court’, this is no more than a convenient way of referring to an entire system of courts whose domestic rules on venue will direct litigants to one particular court within the system.
In municipal systems there is a legal duty upon persons subject to the jurisdiction of a court to submit to its jurisdiction. Unlike the position in international tribunals, that duty is independent of the will of the litigants: the plaintiff has a right to institute proceedings in the municipal court. This is, in part, a reflection of the interest of the state in which the court is established in ensuring that disputes within its territory and its jurisdiction are settled in a manner conducive to the application of the rule of law and public order. It is also a requirement of international law: as is clear from the principles concerning human rights and the treatment of aliens, states are obliged to provide individuals with access to a system of justice which enables them to vindicate their legal rights. This immediately signals a difference between municipal and international tribunals. In municipal courts, the defendant will often not have consented at any stage or in any way to appear before that court. The court will obviously be concerned to ensure that an unwilling defendant is compelled to defend itself before the court only where it is fair and appropriate for the court to do so. And the court may consider that fairness and procedural propriety suggest that the plaintiff and defendant be directed to litigate in a court elsewhere.
Subjection to the jurisdiction of a municipal court is, moreover, a question upon which there are some universally accepted principles, such as the subjection of all persons within the territorial jurisdiction of the state in which the courts are established to the jurisdiction of those courts. Municipal courts are, furthermore, permanent institutions, prepared at all times to accept new litigants, whereas with the exception of the few permanent institutions such as the ICJ and the ITLOS, international tribunals have to be established by the parties when a dispute arises. It follows that if the first municipal court decides that it is appropriate to divert a case before it to another municipal court, there will be many circumstances in which it can confidently be predicted that the court to which the litigants are directed will indeed have jurisdiction over the matter. As was noted above, it may be less easy for one international tribunal to be confident that another possesses concurrent jurisdiction in respect of the dispute before it.
Furthermore, the flexibility of jurisdiction is enhanced by the ability of all municipal legal systems to direct their courts to decide certain cases by the application of a foreign law; and the determination that a foreign law is the lex causae is, of course, a routine matter in courts around the world. The choice of a court may, accordingly, be detached from the question of the substantive law that the court will apply. While there is probably more room for the parties to choose the lex causae in international tribunals than is sometimes supposed, it is true that at least in certain international tribunals claims must be couched in terms of specific bodies of law. This may be a very general body of law, such as the set of ‘sources’ of law that the ICJ is obliged by Article 38(1) of its Statute to apply to cases before it; or it may be more specific, as for example in the ITLOS and the WTO dispute settlement procedure. This contrasts with the greater flexibility in municipal tribunals, which tends to facilitate the transfer of proceedings from one jurisdiction to another.
The obligation of some tribunals to apply specific bodies of law leads to another important distinction. Municipal courts have plenary jurisdiction, in the sense that somewhere within any given legal system there will be one court or another before which can be brought every claim that can be formulated under the laws of the state in question. In municipal law it is, broadly speaking, possible to send litigants to another jurisdiction without depriving the plaintiff of its general cause of action. The plaintiff dispatched to another municipal court may, certainly, be deprived of important ancillary procedural advantages, and may be subject to equally important limitations upon the remedies available to it; but it is unlikely to have no cause of action whatever of the kind that it seeks to maintain before the first court. Moreover, to the extent that such matters do differ between the first and second courts, the differences are regarded as a corollary of the principle that these are matters that should be regulated by the lex fori, as elements of the public policy of the forum state. Again, these characteristics make it much easier for a municipal court to send a plaintiff elsewhere than for an international tribunal to do so. In the municipal context, either plaintiffs have more or less the same rights in the second court, or the differences are attributable to a generally accepted public policy of distributing competence that stands on a higher level than the interests of individual litigants. In international law there are many co-existing tribunals, or microsystems of tribunals and appellate bodies, whose jurisdiction is by no means plenary; and while it may be possible in some cases to divert proceedings to another tribunal without depriving the applicant of its central cause of action, in other cases it will not in practice be possible to do so at all.
Given these differences, to what extent may municipal law concepts such as forum non conveniens appropriately be applied by international tribunals? Forum non conveniens itself exists in a number of slightly different forms. In its Americo-British incarnation, it began as an expression of a concern for the interests of the litigants. It is formulated in Dicey and Morris in the following terms:
Rule 31—(1) English courts have jurisdiction, whenever it is necessary to prevent injustice, to stay or strike out an action or other proceeding in England, or to restrain the institution or continuance of proceedings in foreign courts… (2) As a general rule, in order to justify a stay of English proceedings on the basis that England is an inappropriate forum (forum non conveniens), there must be another forum to whose jurisdiction the defendant is amenable, which is clearly or distinctly more appropriate than the English forum, i.e. is a forum in which the case may be tried more suitably for the interests of all the parties and the ends of justice.
Claims of hardship to the defendant generated the plea. If an action that could have been brought elsewhere was thought vexatious or oppressive to the defendant, it would be stayed by the court. To force a foreign individual, or even a corporation, to defend an action might be unduly burdensome, and the chosen forum in that sense ‘inappropriate’. Evidence might have to be moved; witnesses transported; experts in foreign law hired; duplicate legal teams maintained to manage actions in different jurisdictions, and so on. The plaintiff’s interests were also important. A court would not stay the action unless it was satisfied that the plaintiff could obtain justice in a foreign court, a determination that might involve an assessment of the foreign court’s jurisdiction, of procedural obstacles such as time limitations, and (a matter of great delicacy) of the competence and impartiality of the foreign legal system. The doctrine has, however, subsequently developed in American and British courts (but not, for example, in Australian courts) so that it became less focused upon the interests of the parties and more explicitly concerned with the general interests of justice, and the efficient distribution of tasks between jurisdictions.
Things are different in inter-state actions in international tribunals. There is a strong tradition of holding inter-state litigation in ‘neutral’ locations, chosen for the convenience of the judges and the parties rather than for any link that the place might have with the facts underlying the action. The location of evidence and convenience of witnesses is usually of minimal importance. Unlike municipal litigation a move in the location of an inter-state action before an international tribunal does not entail the need to employ experts to handle questions of foreign law, or the need to maintain separate legal teams in different countries in order to handle connected cases in different fora. The notion of difficulties arising from the venue of international litigation simply does not apply in the same way that it does in private (non-state) disputes. In inter-state disputes, the lawyers prepare the case; they get on the plane; they get off, and they litigate. Where they do it is a matter of little legal significance.
More significantly, respondent states are not in the same position as unwilling defendants. As has already been noted, the municipal law defendant will be before the municipal court because it has to be — because the defendant was within the personal jurisdiction of the court and was obliged to respond to a summons from it. The respondent will be before the international tribunal, in the circumstances under discussion here, because it had at some point agreed to submit to the tribunal’s jurisdiction. And, equally important, there is not the same need to maintain a harmonious relationship between international tribunals as there is between national courts. In the municipal context, relations between courts are an aspect of relations between national legal orders. However distinct they might be at the level of precise legal analysis, the overbearing assertion of jurisdiction by courts (or its Bhopal-style mirror-image) is in much the same category as overbearing assertions of legislative jurisdiction or of economic or political power, as far as international relations are concerned. While those with sensitive antennae may from time to time detect hints of rivalry among judges on international tribunals that are actual or potential competitors for judicial business, the situation there is of a quite different order to the real and pressing need to establish a pragmatic and mutually respectful relationship between the courts that sit each day in the two hundred or so countries of the world.
For all these reasons, I do not believe that the doctrine of forum non conveniens can at the present time be translated into the international legal system. Criteria developed in the context of a proper concern for the interests of private litigants make little sense in the context of inter-state proceedings. The one possible exception is the opportunity presented by the later and still developing rationale for forum non conveniens: the notion that the demands of efficiency in the administration of justice may indicate that the court should decline to exercise its jurisdiction. The crucial characteristic of this rationale, however, is that the court making this finding must regard itself as a part of a wider system of courts which also includes the ‘appropriate’ forum, and within which rational decisions on the allocation of tasks can be taken. The features of the array of municipal courts that were noted above — permanence, predictable bases of personal jurisdiction, plenary subject-matter jurisdiction, and so on — make such a premise and such decisions possible. The time will no doubt come when the same can be said of the array of international tribunals; but I do not think that this time has yet arrived.
What of the other doctrines applied by municipal courts? What of lis alibi pendens, abuse of process, and res judicata? Here matters are again different. These are doctrines that do not derive their persuasive force from the (municipal) litigation context within which they operate. Their rationale is independent of the convenience and interests of the litigating parties. Rather, it proceeds from requirements of good order that are applicable to each and every judicial system.
Lis alibi pendens was said by the Permanent Court of International Justice in the Polish Upper Silesia Case to have as its object the prevention of the possibility of conflicting judgments. It might have added that it also avoids the danger of a race to judgment between two tribunals, which is itself inimical to good order in judicial proceedings. The doctrine indicates that if a substantially identical case is already pending before a competent tribunal, the forum may decline to exercise its own jurisdiction. Similarly, the doctrine of res judicata indicates that if legal claims have already been put in issue and decided by a competent tribunal, that decision is dispositive and the same claim may not be raised again in another tribunal in a substantially identical action between the same parties. Again, the forum may apply the doctrine with the result that it declines to exercise its jurisdiction. Plainly, the opportunity to apply these doctrines will not be available to the tribunal first seised with a case; but they may permit the second tribunal to yield jurisdiction to the first.
The doctrine of abuse of process is equally well established, though occasions for its application are likely to be very rare. It indicates that a tribunal should decline to exercise jurisdiction in a range of circumstances where the action is rendered vexatious. These include cases where the purpose of the litigation is to harass the defendant, or the claim is frivolous or manifestly groundless, or the claim is one which could and should have been raised in earlier proceedings. This doctrine arises not from the fact of multiple proceedings but rather from the inherently vexatious nature of the particular proceedings in the forum. It could be applied by either the first or second tribunal.
Each of these doctrines relates to the good order of judicial proceedings. Each is common to all the major legal systems, and may properly be applied by a tribunal in any legal system, including the international legal system, in the exercise of the tribunal’s competence to regulate its own proceedings. The result is that in cases of overlap between two general jurisdictions, or between two specialised jurisdictions in the same field, where the lex specialis principle offers no solution, a tribunal may decline to exercise jurisdiction because it decides that the doctrine of lis alibi pendens, abuse of process, or res judicata is applicable. In other cases a tribunal faced with such an overlap of jurisdictions has, in my view, no right to refuse on the ground of the overlap of jurisdiction to hear and decide the case before it.
The position of overlaps between two jurisdictions in different specialised fields is a short move from the situation just discussed. The application of the doctrines of lis alibi pendens and res judicata depend upon the substantial identity of the parallel actions. If the actions are quite different, the doctrines have no application. If a dispute were to arise concerning, say, the regulation by state A of ships from state B passing through state A’s waters in an international strait, state B might complain that state A’s conduct violated both the 1982 UN Convention on the Law of the Sea, Articles 37-44 (Transit Passage) and also GATT Article V (Freedom of Transit). Complaints might be made to the ITLOS and also to the WTO, each being couched in terms of the relevant treaty. In such a case the two jurisdictions would not overlap — and any argument that they might overlap would fall in the face of the lex specialis principle, which plainly requires the GATT disputes to go to the WTO and the Law of the Sea dispute to go to the ITLOS. The claims would not overlap, even though they spring from the same facts. And for the reasons advanced above, it would not be appropriate to apply the doctrine of forum non conveniens.
There is no reason why both sets of proceedings should not take place in parallel. This is evident if it is considered that the applicant state may be seeking as a remedy no more than a declaration of its rights under the treaty in question.[26 ]The fact that a state has sought such a declaration under one treaty cannot deprive it of the right to seek a declaration in respect of another treaty, and to go to a different tribunal for that purpose. If the tribunal has jurisdiction, it should exercise it.
Only in the case of an abuse of process would a tribunal be justified in declining to exercise jurisdiction in a case concerning two states which were litigating elsewhere over some other aspect of the dispute. The concept of abuse of process might be more potent in this area than in other types of overlap. It might, for instance, be deployed to defeat attempts by states to manufacture entirely artificial disputes in order to avail themselves of the jurisdiction of a convenient tribunal, to which the respondent happens to have submitted. It would be surprising if, from time to time, states were not tempted to initiate proceedings in, for instance, the WTO, even though the underlying quarrel with the respondent state does not arise from trade relations but from some quite different cause. It may well be that the WTO is the only tribunal whose jurisdiction both the applicant and respondent have submitted. In such cases if the application does not show, in the words of the ICJ in the ILO Administrative Tribunal Case , a ‘substantial and not merely an artificial connexion’[27 ] between the facts and the provisions upon which the applicant relies to establish the jurisdiction of the tribunal, the tribunal would be justified in dismissing it as an abuse of process.
This paper has addressed only the core issues of the simplest cases of overlapping jurisdiction. The topic is one that is rich in variations and complications, and there is a need to devise a coherent approach to it. Individual cases before particular tribunals will generate instances of the problem, and result in the gradual, case-by-case development of a strategy for dealing with it. But the problem arises from the fragmentary nature of the international judicial system, and it demands a comprehensive solution and not piecemeal answers. It is in just such areas that international legal scholars, freed from the need to decide a particular case, have a vital role to play. A pragmatic, balanced, and principled solution, drawing on their knowledge of the operation of international law and their grasp of its structure and processes: that is what good academics can contribute. That is the kind of contribution that Don Greig has now been making to international law for more than 30years.
Case Concerning Certain German Interests in Polish Upper Silesia (Preliminary Objections)  PCIJ (ser A), No 6 12
Chemin de fer Zeltweg (Austria v Yugoslavia) (1934) 3 RIAA 1795 5
Du Pont v Nemours  2 Lloyd’s Rep 585 12
ILO Administrative Tribunal Case  ICJ Rep 89 14
In re Union Carbide Corp. Gas Plant Disaster at Bhopal India in December  USCA3 92; 1984, 634 F Supp 842 (1986) 11
Islamic Republic of Iran v Pahlavi, 464 NYS 2d 487 (1983) 7
Lockerbie Case (Preliminary Objections)  ICJ Rep 102 14
Lockerbie Case (Provisional Measures)  ICJ Rep 66 14
Mavrommatis Palestine Concessions (Jurisdiction)  PCIJ (ser A), No 2 5
Oceanic Sun Line Special Shipping Co Inc v Fay  HCA 32; (1988) 165 CLR 197 10
Rights of Passage over Indian Territory Case (Portugal v India) (Merits)  ICJ Rep 6 5
Southern Bluefin Tuna Case (1999) 38 ILM 1624 2
Spiliada Maritime Corp v Cansulex Ltd  AC 460 10
The Abidin Daver  AC 398 12
Tunisia/Libya Continental Shelf Case  ICJ Rep 18 9
General Agreement on Tariffs and Trade 1947 1, 3, 13
Treaty Establishing the European Community 1957 5
[*] Chichele Professor of Public International Law and Fellow of All Souls College, University of Oxford.
 See, eg, the position of the US courts, the ICJ, and the Iran-US Claims Tribunal: G Wegen, ‘Discontinuance of International Proceedings: The Hostages Case’ (1982) 76 American Journal of International Law 717.
 See, eg, the Southern Bluefin Tuna Case (1999) 38 ILM 1624.
 See also the Statute of the Central American Court of Justice, ch II <http://www.ccj.org.ni/> .
 See V Rodríguez-Cedeño, First Report on Unilateral Acts of States, UN Doc A/CN 4/486 (1998) -.
 See B Cheng, General Principles of International Law (1987) 25-26; G G Fitzmaurice, ‘The Law and Procedure of the International Court of Justice 1951-4: Treaty Interpretation and other Treaty Points’ (1957) 33 British Yearbook of International Law 236; Lord McNair, The Law of Treaties (1961) 219; Sir R Jennings and Sir A Watts, Oppenheim’s International Law (9th ed, 1992) 1280; D P O’Connell, International Law (1970) 12-13.
 See, eg, Chemin de fer Zeltweg (Austria v Yugoslavia) (1934) 3 RIAA 1795, 1803; Mavrommatis Palestine Concessions (Jurisdiction) PCIJ  (ser A), No 2, 30-31; Chorzów Factory (Jurisdiction) PCIJ (ser A), No 9, 30; European Commission of the Danube, PCIJ (ser B), No 14, 23; Rights of Passage Case (Merits)  ICJ Rep 6. Cf J L Simpson and H Fox, International Arbitration (1959) 75.
 This leaves open the question whether the tribunal has jurisdiction which it may not exercise, or does not have jurisdiction at all.
 It is not clear that the premise is necessarily sound. It might be argued that, for example, a certain dispute that arises between two OSCE states has a particular inherently ‘OSCE’ character that renders the OSCE Convention a lex specialis in relation to that dispute, prevailing over the jurisdiction of the ICJ. I do not pursue that possibility here.
 (1993) 32 ILM 560.
[10 ] See the declarations set out at <http://www.un.org/Depts/los/los_decl.htm> . The details given are as at 14 March 2000. Comparable overlaps may arise in other fields, such as human rights, where both international bodies such as the UN Human Rights Committee, and regional bodies such as the Inter-American Commission and Court of Human Rights, may both have competence in respect of a given dispute.
 PCIJ (ser A), No 9, 30.
 There are, it is true, instances in which a tribunal has turned away litigants in cases in which it appeared to have both personal and subject-matter jurisdiction, despite the evident lack of an alternative forum: see, eg, Islamic Republic of Iran v Pahlavi, 464 NYS 2d 487 (1983); 81 International Law Reports 557. But this is rare and plainly undesirable — and, by definition, not a case that falls within the category of instances of overlapping jurisdiction.
 And it is in any event usual to insist that a defendant objecting to the hearing of a case in the forum court establish that it is amenable to the jurisdiction of a more appropriate tribunal elsewhere.
 See the comments of the ICJ in the Tunisia/Libya Continental Shelf Case  ICJ Rep 18, 21, 38.
 L Collins (ed), Dicey and Morris on the Conflict of Laws (12th ed, 1993) 395 (fns omitted).
 See Spiliada Maritime Corp v Cansulex Ltd  AC 460.
 See Oceanic Sun Line Special Shipping Co Inc v Fay  HCA 32; (1988) 165 CLR 197.
 P Prince, ‘Bhopal, Bougainville and Ok Tedi: Why Australia’s Forum non Conveniens Approach is Better’ (1998) 47 International and Comparative Law Quarterly 573; J E S Fawcett, ‘Trial in England or Abroad: The Underlying Policy Considerations’ (1989) 9 Oxford Journal of Legal Studies 205.
 There are exceptional cases. The Bosnian Genocide Cases in the ICJ hold the prospect of several hundred witnesses being called.
 In re Union Carbide Corp. Gas Plant Disaster at Bhopal India in December USCA3 92; , 1984, 634 F Supp 842 (1986); (1986) 25 ILM 771.
 PCIJ (ser A), No 6, 20.
 See the comments of Lord Brandon in The Abidin Daver  AC 398, 423.
 The question of what ‘substantial identity’ might be is important and difficult. There is much relevant practice and analysis to be found in municipal courts, not only in civil cases but also in contexts such as the application of the non bis in idem principle in criminal law and the ‘double criminality’ requirement in extradition law. See, eg, C Van den Wyngaert and G Stessens, ‘The International Non Bis in Idem Principle: Resolving Some Unanswered Questions’ (1999) 48 International and Comparative Law Quarterly 779.
 ‘The policy of the law must … be to favour the litigation of issues only once, in the most appropriate forum,’ as Bingham LJ put it in Du Pont v Nemours  2 Lloyd’s Rep 585, 589. The reference to the ‘appropriate forum’ invites the application of the doctrine of forum non conveniens to the municipal cases with which he was concerned.
 The application of the doctrine is by no means unproblematic in the context of international tribunals: see V Lowe, ‘Res Judicata and the Rule of Law in International Arbitration’ (1996) 8 African Journal of International Law 38-50.
[26 ] The question of double recovery in damages is not considered in this paper, but does not in any event touch the jurisdiction of the tribunals to hear the multiple claims.
[27 ]  ICJ Rep 89. Cf Jennings J in the Lockerbie Case (Preliminary Objections)  ICJ Rep 102.