THE EXTENT AND LEGITIMACY OF THE JUDICIAL FUNCTION IN UNCLOS DISPUTE SETTLEMENT

Abstract

This article examines reactions to the South China Sea and Chagos Marine Protection Area arbitrations under the United Nations Convention on the Law of the Sea (UNCLOS), in particular concerns about the potential widening of Part XV jurisdiction and its impact on the dispute resolution system's consent basis. It argues that assessing the impact of such cases involves a characterization of both the function of Part XV and of international judges. Ultimately, it suggests that the best test of whether UNCLOS case law has gone too far is the reaction of States in designing dispute settlement under the new Agreement under UNCLOS on the Conservation and Sustainable Use of Marine Biological Diversity of Areas Beyond National Jurisdiction.

International law allows States not to submit to judicial settlement. But without such submission the dispute remains and it cannot simply be wished away, even by the most powerful. ¹

I. INTRODUCTION

Have judges and arbitrators of the law of the sea undermined confidence in the system of compulsory dispute settlement under Part XV of the United Nations Convention on the Law of the Sea (UNCLOS)? This article reflects on the criticism offered by commentators, and occasionally by States, that recent cases under the UNCLOS dispute settlement system have expanded Part XV jurisdiction beyond the drafters’ intentions. If true, this could call into question the consent basis of that system, embedded as it is in treaty law. This article first attempts to tease out some of the assumptions underlying the debate in the law of the sea scholarship about Part XV jurisdiction and the nature of the judicial function. There is a tendency in much of the literature to assume that Part XV dispute settlement was intended to be a less than comprehensive system, and that the legitimate function of law of the sea judges and arbitrators does not, generally, include progressively developing the law. It is suggested that such conclusions are easily overdrawn in light of both the negotiating history of the Convention and the general literature on the international judicial function. Further, it is argued that the literature—and to some extent the case law—has been distracted by the unproductive question of whether the ‘real dispute’ between the parties fits within the UNCLOS frame.

This implies that courts and tribunals should engage in a process of characterizing the ‘real dispute’ before them and assessing whether they have jurisdiction over that dispute before answering any questions that may be within jurisdiction on the pleadings. The authors consider such an approach untenable and contrary to the majority of the case law. Finally, the justification for alarm at the supposed expansion of UNCLOS jurisdiction is its potential to vitiate State consent and even, perhaps, drive some States to leave the Convention. Evidence of such concern among States, as opposed to scholars, is muted at best. In particular, in the final section, the article focuses on the negotiation of the dispute settlement provisions of the Agreement under the United Nations Convention on the Law of the Sea on the Conservation and Sustainable Use of Marine Biological Diversity of Areas beyond National Jurisdiction (BBNJ Agreement).² If serious concerns existed about the scope of Part XV dispute settlement, along with a general consensus that jurisdiction had expanded too far, some reflection of this could be expected to be seen in these provisions.

That said, such concerns do not arise in a vacuum. There has indeed been a detectable ‘preference to exercise jurisdiction’³ among UNCLOS dispute settlement bodies, despite the Convention's complex jurisdiction-limiting provisions.⁴ An early, narrow, interpretation of jurisdiction in the Southern Bluefin Tuna arbitration⁵ was followed by a more expansive approach in subsequent decisions. Indeed, there has been a notable trend in the jurisprudence of allowing certain rights or obligations arising prima facie outside the Convention to be adjudicated as part of disputes concerning the interpretation of or application of the Convention.⁶

As noted, this has led to a range of assessments, often negative, from commentators. At one extreme it has been suggested that

States have concerns that there is a trend of ‘judicial activism’, and the original consent of States Parties when they joined UNCLOS has often been violated by means of ‘evolutionary interpretation’ on provisions such as Articles 281, 283 and 298,⁷

and even that such

awards contravene the law, set an ill precedent, and threaten to undermine the international rule of law … [and by interpretation have] usurped, in effect, the law-making power which belongs to the States parties to the Convention.⁸

More measured concerns have been expressed that the incorporation of legal subject matters seemingly ‘beyond the four corners of the Convention’ under Part XV could ‘lead States to reconsider their commitments to Part XV procedures, and hence to the Convention itself’.⁹

Evidence of such concerns actually being held by States is somewhat harder to come by. China has certainly stated that the finding of jurisdiction in the South China Sea case ‘eroded the integrity and authority of the UNCLOS’.¹⁰ In the course of litigating Chagos Marine Protected Area, the United Kingdom (UK) similarly claimed that taking an expansive approach to jurisdiction ‘would call into question the whole system of dispute settlement under the Convention, and with it, the Convention itself’.¹¹ In its maritime delimitation dispute with Mauritius, Maldives argued before the International Tribunal for the Law of the Sea (ITLOS) that ‘States Parties did not envisage that ITLOS … would be exploited to settle territorial disputes, let alone without the consent of indispensable third parties’ and further that ‘UNCLOS States Parties did not sign up for unrestrained judicial activism. They consented to the Part XV procedures to achieve predictable and stable results.’¹² Otherwise, such statements in actual State practice are rare.

The broad approach to jurisdiction taken by tribunals in recent times has included controversial cases where applicant States ‘have sought’—or have been perceived as seeking—‘to find ways to have territorial sovereignty [or maritime delimitation] disputes resolved, or at least influenced, through UNCLOS proceedings’.¹³ It has been suggested that tribunals have overreached in taking on such cases, and as noted, that this may risk undermining the UNCLOS dispute settlement system or even States’ commitment to the Convention.¹⁴ The cases usually mentioned in such a context are two arbitral awards of 2015: the award on the merits in the Mauritius/UK Chagos Marine Protected Area dispute¹⁵ and the award on jurisdiction in the Philippines/China South China Sea dispute.¹⁶

In assessing the systemic impact such cases may have had on UNCLOS dispute settlement two matters are key. First is the debate as to the proper role of judges and arbitrators under the UNCLOS dispute settlement system. That debate cannot be understood without considering the nature of the judicial function in international dispute settlement. The increasing ‘judicialization of international relations and diminishing government control over how international legal agreements are understood’ is not unique to, nor a unique concern of, the law of the sea.¹⁷ Second, evidence will be sought as to whether States consider UNCLOS dispute settlement has gone too far. The most obvious evidence would be the worst-case scenario posited by critics: withdrawal from the Convention, refusal to participate in Part XV proceedings, or deliberate efforts to stymie the dispute settlement system. Such actions are not unthinkable as a matter of international law. Four States have threatened withdrawal from the International Criminal Court following dissatisfaction with judicial or prosecutorial decisions, and two have in fact withdrawn.¹⁸ Refusal by major powers to participate in dispute settlement validly consented to in advance is, unfortunately, a not-infrequent occurrence in international law.¹⁹ Notoriously, the United States (US) has prevented the effective functioning of the World Trade Organization (WTO) Appellate Body by blocking the appointment of new members. Nothing comparable has occurred in relation to UNCLOS and Part XV.²⁰ What other concrete evidence of dissatisfaction might State practice provide? One of the better sources, it is suggested, may be the approach taken to the subject incorporated in the final text of the BBNJ Agreement.

The remainder of the article proceeds as follows. First, it outlines the debate between the ‘narrow’ and ‘comprehensive’ approaches to Part XV dispute settlement. In doing so, it considers in more detail the role and function of the judge or arbitrator under the UNCLOS dispute settlement system. Second, it considers how the judicial function plays out as regards jurisdiction: when may an international judicial officer or arbitrator find they have jurisdiction? Here some of the doctrinal critiques of the findings on jurisdiction made by the arbitral tribunals in Chagos Marine Protected Area and South China Sea are examined. Finally, the theory that existing jurisprudence on jurisdiction under Part XV has ‘gone too far’ is tested by looking for evidence supporting it in the provisions of the BBNJ Agreement as both the most recent UNCLOS implementing agreement and the first which could be negotiated in the light of such concerns. That said, as discussed below, the circumstances of its conclusion may mean any inferences regarding the future of UNCLOS dispute settlement must be drawn with a degree of caution.

II. THE DISPUTE SETTLEMENT SYSTEM UNDER UNCLOS

A. Background

Given that UNCLOS was negotiated as one grand bargain, any system which allowed States to adopt unilateral interpretations risked undermining the integrity of compromises reached both on specific issues and the Convention as a whole. A compulsory dispute settlement system was thus seen by many as providing a useful brake on the possibility of fragmented and diverse interpretations of the Convention's provisions.²¹ Further, given the scale of UNCLOS and manner of its negotiation, ‘many of the 1982 Convention rules contain lacunae or, inevitably, are generally phrased, and since some State practice is of doubtful consistency with them, the availability of judicial or arbitral proceedings offers an exceptional [and necessary] opportunity to clarify the law and to resolve disputes’ between State parties.²² The incorporation of compulsory dispute settlement was useful for powerful States, who wished to protect their navigational rights peacefully, as well as developing States, who saw an advantage in holding larger powers to account.²³

As noted, the dispute settlement system established in UNCLOS is contained in Part XV. It is, in principle, compulsory and comprehensive but contains, nonetheless, a number of wide exceptions.²⁴ In the first instance, Part XV provides a number of options allowing States to opt out of compulsory dispute settlement in some circumstances, either in relation to specific subject matters, or as part of treaty regimes.²⁵ However, at the core of Part XV is a complex system²⁶ that is intrinsically ‘mandatory but limited’.²⁷ In essence, the starting principle of Part XV is that ‘any dispute concerning the interpretation or application’ of UNCLOS is subject to compulsory dispute settlement, unless the parties agree to another mechanism, or an exception applies.²⁸ The major exceptions are found in Articles 297(2) and (3) of the Convention. They expressly exclude from compulsory dispute settlement disputes over marine scientific research (as an exclusive economic zone or continental shelf activity) and fisheries regulation (within the exclusive economic zone) but provide for a system of compulsory but non-binding conciliation in such cases. In addition, States may also choose to make a declaration that they do not accept the application of Part XV to certain sensitive subject matters, principally: maritime boundary disputes (which are again subjected to an alternative and mandatory conciliation procedure), military activities, and exclusive economic zone law enforcement activities in respect of marine scientific research or fisheries management.²⁹ Nonetheless, certain subject matters specified in Article 297(1) are always covered by Part XV and cannot be excluded by the exceptions in Articles 297(2) and (3),³⁰ notably including freedom of navigation and disputes concerning ‘specified international rules and standards for the protection and preservation of the marine environment’.

B. Varying Approaches to Interpretation

Commentators and arbitrators have differed on how Part XV, given its structure, should be interpreted and applied in practice. As mentioned above, a number of commentators have been critical of the perceived expansionist approach to jurisdiction employed by Part XV tribunals.³¹ This assessment may seem justified if one takes the view that the complex structure of, and numerous exceptions to, jurisdiction under UNCLOS dispute settlement evidence a dispute resolution system that was always intended to be limited and narrow.³² The view could be taken, based on these aspects of the treaty text, that this is all the political consensus at the time would bear and the drafters of UNCLOS therefore deliberately chose not to establish ‘a comprehensive dispute settlement system’.³³ The cardinal text usually quoted in support of this conclusion is the assessment of Part XV delivered by a majority of arbitrators in the Southern Bluefin Tuna arbitration that:

UNCLOS falls significantly short of establishing a truly comprehensive regime of compulsory jurisdiction entailing binding decisions.³⁴

However, wider approaches to the scope of Part XV jurisdiction usually commence from a different starting point. Here the touchstone text is the statement of the first President of the Conference that negotiated UNCLOS, Ambassador Amerasinghe, that:

[d]ispute settlement procedures will be the pivot upon which the delicate equilibrium of the compromise [reached] must be balanced.³⁵

On this basis, an expansive approach to jurisdiction may be seen as a good thing, prioritizing the idea that UNCLOS dispute settlement was always intended to be both compulsory and as wide as possible, applying in principle to all disputes ‘concerning the interpretation or application of [the] Convention’.³⁶ Such an approach is consistent with a reading of the Convention as a comprehensive package deal, where all exceptions were the result of hard-fought compromises, and in which compulsory dispute settlement was seen as a necessary bulwark against the balance of rights and obligations achieved in negotiations being unpicked by unilateral interpretations.³⁷ On such an approach, UNCLOS tribunals are right to read down exceptions to jurisdiction precisely because the parties intended a comprehensive and wide-ranging dispute settlement system subject only to limited exceptions.³⁸

The practical implications of such differences in approach are exemplified by two cases that examined the operation of Article 281(1) of UNCLOS. That article provides that parties can choose to settle disputes by a peaceful means of their own choice, but if no settlement is reached then compulsory jurisdiction applies to the dispute as long as ‘the agreement between the parties does not exclude any further procedure’.

In 2000, the arbitral tribunal in Southern Bluefin Tuna relied on Article 281(1) in concluding that Article 16 of the Convention for the Conservation of Southern Bluefin Tuna (CCSBT) implicitly or effectively excluded UNCLOS dispute resolution. The CCSBT did so, in the majority view of the tribunal, by providing that, if consultations and negotiations failed, a dispute under the Convention could be referred to arbitration or the International Court of Justice (ICJ) if all parties to the dispute consented. The tribunal held that this was in line with the structure of Part XV, because the purpose of Article 281(1) was to ‘confine the applicability of [the] compulsory procedures of … Part XV to cases where all parties to the dispute have agreed upon submission of their dispute to such compulsory procedures’.³⁹ Notably, nothing on the face of Article 281 contains such a requirement. The effect of this logic was to replace a compulsory procedure (under Part XV) with a voluntary one (under the CCSBT), and to do so by implication rather than express words.⁴⁰ As noted, the tribunal arrived at this conclusion, in part, by characterizing Part XV as a whole and finding that it fell ‘significantly short of establishing a truly comprehensive regime of compulsory jurisdiction entailing binding decisions’.⁴¹ The tribunal thus took a consent-maximizing approach, and found the CCSBT dispute settlement provisions reflected an overall preference in UNCLOS for consent-based dispute resolution.

In contrast, the arbitral tribunal in the South China Sea arbitration took an expansive view of Part XV. It observed that:

Although the Convention specifies certain limitations and exceptions to the subject matter of the disputes that may be submitted to compulsory settlement, it does not permit other reservations, and a State may not except itself generally from the Convention's mechanism for the resolution of disputes.⁴²

Earlier in its Award on Jurisdiction and Admissibility the tribunal had emphasized that the essence of Part XV is encapsulated in UNCLOS Article 286,⁴³ which provides:

Subject to section 3 [on exceptions], any dispute concerning the interpretation or application of this Convention shall, where no settlement has been reached by recourse to section 1 [on preliminary negotiations and alternative mechanisms], be submitted at the request of any party to the dispute to the court or tribunal having jurisdiction under this section.

Both statements of the tribunal emphasize the binding nature of Part XV and, despite its exceptions, the central role of compulsory dispute settlement system in the Convention regime.

The South China Sea arbitral tribunal declined to follow the majority in Southern Bluefin Tuna, considering the minority opinion of Sir Kenneth Keith in that case and the relevant provisional measures jurisprudence of ITLOS more persuasive.⁴⁴ It thus held that opting out of Part XV required either express treaty language excluding the application of any other procedure under Article 281, or the provision of an alternative compulsory dispute resolution mechanism under Article 282. This may accord with a different reading of the Convention, one emphasizing its nature as a package deal that does not allow parties to pick and choose their obligations.⁴⁵ Thus, in respect of Part XV, parties cannot ‘remove a pivotal part of the Convention without clearly expressing an intention to do so’.⁴⁶ This reading of UNCLOS has not found favour with authors who consider the principle of consent in international dispute settlement to require that States be given wide latitude to ‘opt out’ in the interpretation of Part XV.⁴⁷ In any event, this conflict in the case law reveals ‘a lack of clarity’ about the effect of dispute settlement clauses in treaties where, due to the subject matter, a dispute might also be brought under UNCLOS.⁴⁸ On one view, recourse to Part XV might be excluded by the existence of any alternative dispute settlement procedure; on a stricter view, it might only be capable of exclusion by express words.

This section has attempted to highlight that there can be legitimate differences of view on the question of how comprehensive UNCLOS's system of compulsory dispute resolution is intended to be. The perspective taken on this question is likely to affect how the scope of various express limitations on the jurisdiction of an UNCLOS dispute settlement body is interpreted. The point is discussed further below in relation to ‘procedural’ and ‘substantive’ models of judging. The stance taken may also influence the approach to a more fundamental underlying question—which disputes are disputes ‘concerning the interpretation and application of the Convention’?

III. THE JUDICIAL FUNCTION IN INTERNATIONAL LAW

A. Judging as Governance

The debate about the appropriate approach to interpreting Part XV is also influenced by commentators’ assumptions about the role of judges and the judicial function in international law. As Scott has put it:

[a] constitution does not distribute rights and responsibilities simply for the sake of doing so, but in order to facilitate harmonious interactions amongst members of the society.⁴⁹

In their excellent recent monograph, Klein and Parlett usefully contrast two views of the role of judges and arbitrators under UNCLOS: a substantive and a procedural model.⁵⁰ The substantive model ‘envisages a vigorous role [for judges and arbitrators] whereby tribunals have a dynamic responsibility through their engagement with the balance of rights and interests between States party to the United Nations Convention on the Law of the Sea’ to maintain ‘the public order of the oceans’.⁵¹ Public order, in this sense, involves a degree of active judicial governance through progressive interpretation and development of the law. The second, procedural, model of judging:

recognises the enduring importance of State consent as a signature feature of inter-State litigation. Without State consent to arbitration or adjudication, these modes of international dispute settlement are simply not available. With this consideration firmly in place, the role for international tribunals may be more circumscribed, or a more circumspect approach to the role of dispute settlement in relation to the dispute at hand may be warranted.⁵²

The authors’ sympathies appear firmly in the procedural camp, encouraging judicial modesty. This is particularly apparent in their adoption of stakeholder theory to explain the role of UNCLOS judges. Stakeholder theory is ordinarily used to explain how ‘managers … determine who are the salient stakeholders in making decisions in the interests of the organisation’.⁵³ Klein and Parlett expressly identify judges with managers, and the ‘organisation’ with the UNCLOS regime. The role of judges/managers is to ‘achieve the organisation's objectives’, and to do so successfully they must be responsive to stakeholders (who may possess varying claims to power, legitimacy and urgency, all of which should be weighed in making managerial decisions).⁵⁴ At least implicitly, then, the stakeholders in this equation are State parties.

Overall, Klein and Parlett's preferred vision of judges is one of technocratic managers of a regime, administrative servants of a given stakeholder community. It may also correspond with a ‘contractual’ or State-sovereignty maximizing reading of UNCLOS dispute resolution provisions which starts from the premises that, given the limitations and exceptions to Part XV:

(1) the entire system was designed to be less than comprehensive; and (2) one should therefore construe the applicability of the system as a whole narrowly, in order not to exceed the bargain struck.⁵⁵

Klein and Parlett are not the only supporters of the procedural approach. For example, in commenting on the exercise of jurisdiction rationae materiae in the Chagos Archipelago and South China Sea cases, Proelss argued:

in light of the crucial importance of State consent regarding the legal basis of jurisdiction … adjudicating bodies are generally obliged to carefully observe and respect the limits of their powers in relation to the States parties as masters of the treaty.⁵⁶

While this approach undoubtedly maximizes the role of State consent in interpreting the UNCLOS dispute settlement regime, it assigns a curiously circumscribed role to judicial officers.

A potential difficulty with the procedural conception is that negotiating States presumably knew what they were doing when they expressly empowered judicial and arbitral officers under UNCLOS, given that in building judges into a dispute settlement system, it necessarily becomes a judicial system of dispute resolution. This is a different thing to conciliation, mediation, and so on—which must have been apparent to negotiating parties at the time. Choosing in treaty design to delegate ‘interpretive authority to [international courts] is politically significant because it introduces an independent outside actor with the legal authority to say what international law means’.⁵⁷ As Weiler has aptly observed in the WTO context:

Juridification is a package deal. It includes the rule of law and the rule of lawyers. It does not affect only the power relations between members, the compliance pull of the agreements, the ability to settle disputes definitively, and the prospect of authoritative interpretations of opaque provisions. It imports the norms, practices, and habits – some noble, some self-serving, some helpful, some disastrous, some with a concern for justice, some arcane and procedural – of legal culture. It would be nice if one could take the rule of law without the rule of lawyers. But with one, you get the other.⁵⁸

And while some suggest the consequences of a transition from a diplomatic to a legal culture of dispute settlement caught some by surprise in the trade environment,⁵⁹ it is less obvious that there was such naiveté as regards the law of the sea.⁶⁰ ‘Indeed, for less powerful States the [inclusion of a] dispute settlement system was often a sine qua non of their consent to the whole.’⁶¹ On the other hand, a degree of hesitation about the potential power of judicial settlement to develop the law is precisely why the UNCLOS dispute settlement system first prioritizes consensual dispute settlement and, in the event of third-party settlement, defaults to arbitration.⁶² Overall, however, negotiators expected UNCLOS tribunals’ role to ‘go beyond settling bilateral disputes’ and ‘expected [them] to provide normative guidance to States in implementing the Convention in order to safeguard [its] uniformity and integrity’.⁶³ As Klein has put it:

the willingness to create a regime that includes mandatory dispute settlement procedures could be an indication that [UNCLOS parties saw] adjudication or arbitration … [as] a vital component in protecting rights, as well as for interpreting and amplifying the meaning of the provisions of the treaty to facilitate the regulation of State conduct under the terms of UNCLOS.⁶⁴

It is hard to disagree.

If investigating whether a system of judicial dispute resolution is functioning properly, the first question should presumably be: what is expected of judges? As Crawford and McIntyre note, despite differences between national and international systems of law, when examining judges in either system ‘there appears to be a fundamentally common judicial function’.⁶⁵ While dispute settlement at the international level generally lacks appellate review and is inevitably grounded in consent ‘[i]t is unclear why these differences call for a lower level of judicial integrity [or independence], or to what extent they permit or demand different [judicial] standards’ be applied.⁶⁶ That is, the core concepts and principles of the judicial role do not vary between the national and international level, though what those principles demand in context might.⁶⁷ A judiciary contributes to governance through its role in dispute resolution. Ultimately, the resolution of disputes is fundamental to maintaining ‘the basic degree of order’ (in the sense of predictability and certainty) necessary in any society,⁶⁸ including international society. But, in addition to dispute resolution, judges have a duty ‘to maintain the vitality of the law as a dynamic and effective normative system’⁶⁹ which may promote a society's ‘flourishing over the longer term’.⁷⁰ None of this is particularly shocking. It should be uncontroversial to suggest that a ‘judiciary not only determines and develops the law, but helps to maintain the system of governance by law’.⁷¹

In conceptualizing the role of international judges, the question is not one of prioritizing either a narrow conception of the judicial function focused on dispute resolution (providing ‘good order’) or a vision of judges as functionaries tasked with the further development of a normative system of governance (‘a public order’).⁷² Rather, ‘the judicial function uniquely blends distinct forms of third-party merit-based dispute resolution and social (legal) normative governance’.⁷³ Certainly, much of what is expected of judges appears contradictory, demanding ‘both responsive justice and predictable order’.⁷⁴ However, it can be seen that the ‘genius of the judicial function is [precisely] located in [this] dualistic tension between dispute resolution and general normative governance; both aspects must be valued for the function to flourish’.⁷⁵ These tensions both ‘constrain and liberate’ the judicial function.⁷⁶

In addition, there are, even on a narrower view of the legitimate role of dispute settlement in UNCLOS, provisions of the Convention that were left ‘deliberately obscure or ambiguous’⁷⁷ (to avoid irresolvable debates) or which deliberately used open-ended or ‘ambulatory’ language.⁷⁸ In the former case the parties actively anticipated judicial gap-filling, in the latter a judicial role in updating the Convention through interpretation.⁷⁹ As one Canadian delegate put it during negotiations ‘a [dispute settlement] system would in the long run provide an important means of elucidation and interpretation of the text of the convention’.⁸⁰ Such progressive judicial development of the law is clearly a form of governance, and it was clearly intended by the parties at least in relation to some provisions of the Convention.⁸¹

The point is that in judging judges and arbitrators there is no easy choice between proceduralism and substantive governance. Judging necessarily involves both at the same time. In fairness, while Klein and Parlett exhibit the strongest sympathy for the procedural model—and little detectable enthusiasm for any form of judicial decision-making as governance—they are nonetheless quite clear that the coin has a second side. UNCLOS judges and arbitrators are, in their view, both guardians of a ‘consent-based treaty regime … where State actors’ should be able to ‘make rational decisions on their level of involvement in the regime’,⁸² and consider it important that binding decisions arrived at under Part XV also ‘contribut[e] to the overall good order of the oceans’.⁸³

Of course, ‘order’ is a slippery concept and what makes it ‘good’ is debatable. As discussed above,⁸⁴ ‘good order’ is generally considered to be associated with the legal virtues of stability and certainty in resolving disputes. But ‘public order’ may also be referred to in the sense that judges also have a dynamic institutional mandate to develop the law progressively to meet the needs of the wider society it serves. In these senses, Klein and Parlett appear to see the role of judges as being to promote both the overall ‘good order of the oceans’ and the ‘public order of the oceans’.⁸⁵ Thus they appear to acknowledge that both types of order play a role in dispute settlement.

Given this, is there really a debate over the proper role of law of the sea judges and arbitrators? Most scholars accept both that judges and arbitrators within the UNCLOS regime are there to provide third-party dispute settlement within stable and predictable lines, and that they have also been empowered to uphold, interpret and develop the law. The debate is only as to the balance between these two functions and the limits on each. That this limited debate can seem stark (even over-heated) is perhaps due to lawyers’ tendency to downplay the ‘creative aspect of the judicial role’ for fear that community confidence in judicial institutions ‘will be affected’ if the judicial law-making function is presented too strongly.⁸⁶ When confronted with the issue the tendency is, perhaps, to cry ‘Pay no attention to that man behind the curtain!’.⁸⁷

Assessing how well or badly judges have carried out their two inextricably linked functions of dispute resolution and normative governance is the difficult part. Klein and Parlett reasonably suggest that factors that could be looked at in assessing judicial performance in a given case could include:

the immediate reaction of the parties to the decision; the shorter, medium, and longer term reactions of the parties to the decision; the responses of other relevant States in the immediate aftermath of the decision, taking into account the relative importance of those States; commentators’ assessments of the judgment; whether the judges concerned are re-elected or appointed to another court or tribunal again; and, over time, subsequent judicial decision-making or other international law making activities … that reflect the decisions of the judges.⁸⁸

Section IV analyses the BBNJ Agreement dispute settlement provisions as the most obviously relevant example of subsequent multilateral ‘international law making activities’. A discussion of the remaining possible factors in relation to Chagos Marine Protected Area and South China Sea is not possible in an article of this length. Nonetheless, a few observations may be ventured here, at least as regards South China Sea. As already noted, scholarly reactions on the question of jurisdiction have been mixed. China and Chinese scholars have consistently rejected the validity of the award.⁸⁹ As the successful applicant, the Philippines has unsurprisingly continued to support it, saying that the Tribunal ‘conclusively settled the issue of historic rights and maritime entitlements in the South China Sea’.⁹⁰ The reaction of ‘other relevant States’ may be thought to have been telling—at least on questions of substance. A number of important coastal States have since endorsed the tribunal's ruling as a correct statement of the law of historic rights and/or the maritime entitlements of rocks and islands including Indonesia, the US, Australia, Japan and New Zealand.⁹¹ At best though, these endorsements of the substance of the award only implicitly accept the tribunal had jurisdiction in the first place.

One issue where the appropriate vision of the role of dispute settlement is most acute is in determining the subject-matter ambit of the UNCLOS dispute settlement system—that is, in questions of jurisdiction. This is addressed in the next section.

B. Out of Bounds? The Problem of Jurisdictional Determinations

[Part XV contains] important exceptions, and important uncertainties. Some States have been pushing to expand the exceptions; others have sought to contract them. Whether these exceptions can or should remain stable – or be changed only by a consensus of all the parties – has emerged as one of the most contested contemporary issues facing UNCLOS courts and tribunals.⁹²

A key criticism of awards such as the South China Sea award has been that they have extended the jurisdictional ambit of Part XV of UNCLOS. How can the fairness of this criticism be assessed? Determining whether a dispute falls within the scope of Part XV of UNCLOS is a question of characterization, as indeed is deciding whether a particular interpretative approach to jurisdiction expands or contracts the scope of Part XV.⁹³ In McIntyre's terms, part of the role of a judge is to discern the legal norm which governs a dispute (the ‘dispute-norm’). To do this a ‘judge must engage in a process of identification, interpretation, clarification and assessment’⁹⁴ in order to articulate why, in the particular case, the dispute falls within the genus of disputes ‘about the interpretation and application of UNCLOS’ and which species of dispute it implicates therein (eg freedom of navigation, protection of the marine environment, etc).

The argument has been made that the South China Sea arbitration effectively determined two critical questions which were on their face beyond the tribunal's subject-matter jurisdiction: questions of title to territory and of maritime boundary delimitation. The line of argument runs that the tribunal erred in taking jurisdiction because these questions were the real dispute, and that resolving Philippines’ cleverly framed UNCLOS questions could not be done without also trespassing on these forbidden questions.⁹⁵ This argument goes directly to the judicial function under the Convention because it frames the question as one of strict fidelity to State party consent (staying within the bounds of the Convention) versus judging as governance (expanding the reach of the normative order).

How is a treaty-empowered international court or tribunal to determine properly whether a dispute before it falls within its jurisdiction? Should it content itself with finding there are legally opposed claims between the parties under the relevant convention, or is there a broader duty to exclude cases brought with some ‘ulterior’ purpose? The argument here is that international tribunals should only be concerned with the former, and that in general the case law supports that approach. However, in the specific domain of the law of the sea there has been a tendency to misread authorities supporting the first proposition as giving rise to a broader duty of exclusion. This is easy enough to do. Decisions of the ICJ on jurisdiction in cases such as Fisheries Jurisdiction and Nuclear Tests use language such as ‘[t]he Court will itself determine the real dispute that has been submitted to it’⁹⁶ or ‘it is the Court's duty to isolate the real issue in the case and to identify the object of the claim’.⁹⁷ But in context these statements only went to the duty of a court to satisfy itself of the existence of a legal dispute (in the sense of opposed claims) or to identify the nature of the relief sought by a party. It was only in the Chagos Marine Protected Area case that an UNCLOS tribunal first discerned a need to engage in broader characterization of the dispute before deciding the case was one concerning the interpretation and application of the Convention.⁹⁸ Here there was a question as to whether the tribunal had jurisdiction over Mauritius’ first claim that the UK was not relevantly ‘the coastal State’ under UNCLOS. The UK replied that this question could not be resolved without determining a question of sovereignty (in the sense of title to territory), a question which it submitted was outside the Convention.⁹⁹ In finding it lacked jurisdiction to answer this question the tribunal engaged in a process of characterization, holding it ‘must evaluate where the relative weight of the dispute lies’.¹⁰⁰ Nonetheless, in respect of other claims it found that the ‘issue of sovereignty’ formed only ‘one aspect of a larger question’ and therefore the tribunal had jurisdiction.¹⁰¹

A similar question arose in the South China Sea award on jurisdiction and admissibility. In the face of Chinese objections that the ‘“essence of the subject-matter of the arbitration” was territorial sovereignty over … maritime features’, the tribunal decided it had to engage in a process of characterization to ask if the dispute could ‘fairly’ be said to be about the interpretation and application of the Convention.¹⁰² This involved asking whether the ‘actual objective’ of the Philippines’ case was to advance its position in a dispute over sovereignty arising outside UNCLOS.¹⁰³

As Harris points out, however, the characterization approach adopted in Chagos Marine Protected Area and South China Sea involves a ‘fundamental departure’ from the usual rules determining when a dispute is one concerning the interpretation and application of a Convention.¹⁰⁴ These cases involve a re-purposing of the language of ‘real dispute’ or ‘object of the claim’, usually used to describe a narrow inquiry into the case as pled in legal terms by the parties, to justify a broad-ranging inquiry into the surrounding circumstances of a case in the name of ‘characterization’. Such an approach is attractive to respondents, as it provides an opportunity to argue that the case brought against them is somehow non-legal (ie political) or non-justiciable (falling outside UNCLOS).¹⁰⁵

The characterization approach also involves a misreading of the judicial function. The rule that it is for a court or tribunal to determine the ‘real dispute’ at hand does not imply the existence of a category of ‘false’ or ‘confected’ disputes which should be thrown out—other than perhaps genuine cases of bad faith or abuse of right.¹⁰⁶ The rule is simply a restatement of the duty to determine the existence (or not) of relevant jurisdictional facts: simply, that a tribunal must be satisfied that all necessary legal elements are present for jurisdiction to exist. Properly understood, then, the ‘real dispute’ rule is nothing more than a statement of Kompetenz-Kompetenz.

As noted above, a common criticism of these decisions on jurisdiction is that they undermine good order by potentially extending the reach of UNCLOS dispute settlement. What such a critique fails to take into account is the possibility that good order (or public order) is also undermined by a sense of unaddressed injustice. It is not obvious that the legitimacy of UNCLOS and the ‘good order of the oceans’ would have been promoted by the South China Sea arbitral tribunal declining jurisdiction. All international dispute settlement bodies have to be wary of the ‘South West Africa risk’ that if they are seen to decline to do justice they become an irrelevance.¹⁰⁷ As discussed in Section III(A), that a tribunal has to navigate between these competing priorities is not unique to the international system.

There is always a tension in treaty-based dispute settlement systems. States frequently wish to have such a system as a check on the self-serving behaviour of others but dislike the prospect of it being invoked against them.¹⁰⁸ The design of dispute settlement systems must reassure States that a comprehensive system has been established, but also that any threat to their decision-making autonomy is limited.¹⁰⁹ It is in the nature of the complex structure of Part XV that, despite its seemingly wide exclusions, there is considerable scope to find that things fall within jurisdiction. Parties will seek to persuade judges of the justiciability of the dispute through careful characterization of the case being put.¹¹⁰ There is ultimately nothing new in this.

IV. THE DISPUTE RESOLUTION CLAUSES IN THE BBNJ AGREEMENT

The BBNJ Agreement¹¹¹ is significant as it is the first ‘implementing agreement’ concluded under UNCLOS since the divergent jurisdictional decisions in Southern Bluefin Tuna and South China Sea. If States are truly disgruntled with the operation of Part XV, it would be expected that the text adopted for dispute settlement under the Agreement would reflect this through, for example, incorporating more restrictive rules on jurisdiction. Indeed, the debate over the appropriate approach to dispute settlement did attract considerable disagreement in the final meetings of the Intergovernmental Conference (IGC). Some provisions arguably reflect concerns of States about the limits of UNCLOS dispute resolution, although ultimately they are limited in scope. Nonetheless, any conclusions drawn from the BBNJ Agreement must be tempered both by the circumstances of its conclusion and the nature of the documentary record.

At the outset it must be noted that there is no official record for the BBNJ negotiations or debates in plenary session. Scholars thus have limited access to travaux preparatoires, beyond documents produced by the preparatory commission (PrepCom) or President of the IGC as aids to negotiation, plus the written proposals of negotiating States (conference room papers or ‘CRPs’) and short oral reports of the facilitators of informal working groups at the five IGC meetings (ICG1–5). The limited documentary record indicates that the question of dispute resolution was ‘very complex and contentious’ as early as 2017.¹¹² The early meetings of the IGC did not have time to discuss dispute settlement, although the early drafts of the text contained articles modelled on provisions of the UN Fish Stocks Convention (UNFSA), itself another UNCLOS implementing agreement (this version became known as Option I).¹¹³ These provisions applied Part XV to the Agreement mutatis mutandis, adopting the formulation in Article 30 of UNFSA (‘the mutatis mutandis approach’).¹¹⁴ Later drafts of the Agreement included additional provisions, some of which are relevant to this discussion.¹¹⁵

In adopting dispute resolution provisions, a variety of models were open to negotiators.¹¹⁶ At one end of the spectrum, States could have favoured a strong consent-based model rejecting compulsory and binding dispute resolution. Arguably, this would be the plainest repudiation of ‘expansionist’ interpretations of Part XV, and a signal of States’ potential concern that ‘the original consent of States Parties when they joined the Convention’ regarding the scope of Part XV had been violated ‘by means of “evolutionary interpretation”’.¹¹⁷ An example of text which could have been adopted is furnished by Article 22(3) of the Port State Measures Agreement which permits recourse to binding judicial or arbitral settlement only ‘with the consent of all Parties to the dispute’.¹¹⁸

In fact, two options were discussed during the IGC that would have provided for a consent-based dispute settlement process. The Further Revised Text, which was issued in June 2022, contained an Option II for dispute settlement.¹¹⁹ This Option allowed Parties to opt in to compulsory jurisdiction over disputes. If two Parties did not choose the same forum, the dispute would be submitted to conciliation. Thus, Parties that did not desire compulsory dispute settlement could avoid it altogether. This option was primarily supported by non-Parties to UNCLOS and a few others.

A third option was proposed by China at the final session of the negotiations.¹²⁰ Option III allowed for third-party dispute settlement only if all parties to the dispute agreed.¹²¹ It had a provision excluding from jurisdiction a wide range of matters: disputes ‘concerning the land territory, sovereignty, sovereign rights or jurisdiction’. It also applied the mutatis mutandis approach for Part XV, but subject to the severe restrictions imposed in the previous paragraphs.

Thus, the negotiations established a clear choice between consent-based or compulsory dispute settlement. A clear majority of delegations preferred Option I, the mutatis mutandis approach modelled on UNFSA.¹²² However, as the President of the IGC had stressed that the Agreement should be adopted by consensus, a concerted effort was made to meet the concerns of the States reluctant to adopt this approach. One group, non-Parties to UNCLOS, was appeased through the incorporation of Part XV into the text of the Agreement through Articles 60(2)–(7), as discussed below.

Of all the delegations, it seems that China was most concerned about the approach of Part XV tribunals. Although the case was never expressly mentioned, undoubtedly China's support for consent-based dispute settlement was influenced by its views about the South China Sea arbitration. Ultimately, however, China abandoned its preference for consent-based dispute settlement in return for restrictions on jurisdiction for disputes arising under the Agreement.

Due in part to the rushed circumstances of its conclusion and the need to find compromises, the final dispute settlement provision contained in Article 60 is not a masterpiece of clarity. It is also lengthy, running to ten paragraphs. Article 60(1) provides that disputes under the BBNJ treaty shall be resolved in accordance with Part XV; while Article 60(2) provides that Part XV and related annexes ‘shall be deemed to be replicated for the purpose of the settlement of disputes involving a Party to this Agreement’ that is not an UNCLOS member. This latter provision is intended to meet the misgivings of some States at applying UNCLOS provisions directly to non-parties. Articles 60(3) and 60(4) provide that any UNCLOS party's choice as to preferred forum for dispute settlement (under Article 287 of UNCLOS) or any declarations made as to optional subject-matter exclusions (under Article 298 of UNCLOS) apply equally under the BBNJ agreement unless otherwise specified. Articles 60(5)–(7) essentially make provision for non-UNCLOS parties to be able to make Article 287 and 298 elections for the purposes of the BBNJ agreement. All these provisions might be taken to exhibit a degree of comfort with Part XV as it stands.

Article 60(8) concerns the interaction of BBNJ agreement dispute settlement provisions and those under other legal ‘instruments and frameworks’. It provides that Article 60 is without prejudice to the settlement of disputes under other relevant instruments. This provision could, perhaps, have been drafted to express a clear preference for either a Southern Bluefin Tuna or South China Sea approach to this important question,¹²³ but instead it is drafted as a ‘without prejudice’ clause and thus takes no position on the issue.

Aspects of Article 60's drafting that could have been influenced by States’ reactions to both the South China Sea and Chagos cases are the paragraphs that limit the jurisdiction of courts and tribunals in relation to BBNJ disputes. Paragraphs 9 and 10 read:

9. Nothing in this Agreement shall be interpreted as conferring jurisdiction upon a court or tribunal over any dispute that concerns or necessarily involves the concurrent consideration of the legal status of an area as within national jurisdiction, nor over any dispute concerning sovereignty or other rights over continental or insular land territory or a claim thereto of a Party to this Agreement, provided that nothing in this paragraph shall be interpreted as limiting the jurisdiction of a court or tribunal under Part XV, section 2, of the Convention.

10. For the avoidance of doubt, nothing in this Agreement shall be relied upon as a basis for asserting or denying any claims to sovereignty, sovereign rights or jurisdiction over land or maritime areas, including in respect to any disputes relating thereto.

These paragraphs are based on several proposals. Paragraph 9 began as an Australian proposal that appeared as draft Article 55(8) in the Further Refreshed Text of December 2022.¹²⁴ That text was drawn from Article 298(1)(a)(i) of UNCLOS which excludes disputes over land territory from the jurisdiction of compulsory conciliation. The inclusion in Article 60 makes it a far more consequential exclusion than the UNCLOS paragraph it draws from, however, as it applies to all disputes arising under the Agreement. There was no opposition to this proposal, which attracted little discussion during the negotiations. One might speculate that this proposal could have been in response to the Chagos arbitration, although again there was no express reference to the case during the discussions. It may also reflect Australia's experience of compulsory maritime boundary conciliation with Timor Leste under Article 298(1)(a)(i) of UNCLOS.¹²⁵

The words ‘concerns or necessarily involves the concurrent consideration of the legal status of an area as within national jurisdiction’ in paragraph 9 were negotiated in response to the Chinese proposal. China had wanted to exclude issues concerning ‘sovereignty, sovereign rights or jurisdiction’ of a State Party from disputes. This was seen as far too wide. After all, what actions or activities of a State do not implicate its sovereignty, sovereign rights or jurisdiction? Even marine scientific research on the high seas occurs under flag State jurisdiction and, if done by a State, is arguably an action within the exercise of its sovereignty. This language was not accepted by other States for that reason. In relation to the wording that is now found in paragraph 9, the language ensures that no tribunal formed under the Agreement could, for example, consider whether the South China Sea includes an area of high seas.

The final part of paragraph 9 stipulates that: ‘nothing in this paragraph shall be interpreted as limiting the jurisdiction of a court or tribunal under Part XV’. This appears to prevent any argument that paragraph 9 could be used to interpret the jurisdiction of a Part XV tribunal constituted directly under UNCLOS. That is, the opening words of paragraph 9 cannot be taken to be evidence of a subsequent agreement as to the correct interpretation of UNCLOS itself.¹²⁶ Any implicit criticism in paragraph 9 of the approach taken by various tribunals to their jurisdiction under Part XV is therefore only prospective in operation and confined to disputes arising under the Agreement.

Paragraph 10 also reflects some of the concerns expressed by China and seen in paragraph 3 of Option III. Indeed, it mirrors to some extent Article 6 of the BBNJ Agreement, which states that the Agreement:

shall be without prejudice to, and shall not be relied upon as a basis for asserting or denying any claims to, sovereignty, sovereign rights or jurisdiction, including in respect of any disputes relating thereto.

The words ‘for the avoidance of doubt’ were seen as a reassurance that the principle expressed in Article 6 would also apply in regard to any disputes.

The negotiating history provides some insight. To the extent that the Chinese proposals may be read as a clear expression of China's disillusionment with its experience of the South China Sea arbitration, it is equally clear that concern was not universally shared. In particular, there was a strong commitment by most delegations that Part XV would apply to BBNJ disputes,¹²⁷ including its compulsory and binding processes. Much of the jurisdiction-limiting language proposed by China was not accepted, although of course some restrictions were introduced in paragraph 9 that go beyond the model of the UNFSA.

V. CONCLUSIONS

This article has sought to re-examine the scholarly reaction to the South China Sea and Chagos Marine Protection Area arbitrations, including in the light of subsequent law-making efforts. In particular, it has addressed criticism that these cases extended Part XV jurisdiction in a manner that risked undermining confidence in, or the consent basis of, the UNCLOS dispute resolution system. The argument has proceeded in several steps. First, there is the question of whether UNCLOS contains a compulsory and comprehensive dispute settlement system, albeit with a number of exceptions—or whether that system was deliberately designed to be less than comprehensive and compulsory. This question of characterization is consequential, as it may lead to a view that the various exceptions to or limitations upon jurisdiction should be read either narrowly (to give effect to an overriding intention that the system be compulsory and comprehensive) or expansively (to maximize the consent basis of jurisdiction and therefore State parties’ options to opt out).

The view taken on this first question will likely influence the approach to the second key issue, the correct role of judges and arbitrators under UNCLOS. A not-uncommon view is that there are two possible characterizations of the role of the law of the sea judge. Under a procedural model, the judge is to remember that the overriding basis of any international dispute settlement system is State party consent, without which there will be no dispute settlement. On this view, diplomatic and political solutions are generally preferable to compulsory measures and optimally the role of a judge is to be a modest, technocratic manager of a narrow legal regime aimed at preserving ‘the good order of the oceans’. This is contrasted with a substantive vision of judging, in which judges have a more dynamic role in maintaining the balance of rights and duties in the Convention and in pursuing wider community interest in the ‘public order of the oceans’.

Third, this article has argued that these two visions of the judicial function are not really in competition. Judging inherently involves both an attempt to maintain the predictability and stability of a legal regime, while also being a broader system of social governance empowered both to deliver flexible justice to parties and to further develop the law itself. Indeed, most commentators on the judicial role under UNCLOS dispute resolution accept that both of these aspects are involved. The basis of disagreement is usually as to the proper limits of these different functions and the balance to be struck between them.

The negotiators of UNCLOS would have been well-acquainted with the idea that if judges exist, they will not only apply but also develop the law. Further, any suggestion that there was universal agreement that judges were merely meant to be functional technocrats is implausible. While this may have been the view of some States, is not widely reflected in the negotiating record. Many developing States, still smarting from the South-West Africa Cases and actively pursuing their New International Economic Order agenda, hoped for a wider judicial role. This was, after all, one of the justifications for having a specialist court established under the convention in the form of ITLOS, the ICJ being considered narrow and unrepresentative.

This leads to the fourth issue: the unhelpful tendency in scholarship on international litigation, furthered by unfortunate jurisprudence, to go in search of ‘the real’ dispute. Harris is right to suggest that this test (which has seldom actually been applied to deny jurisdiction) should be abandoned for a simpler test of whether there is an actual dispute as to a justiciable UNCLOS issue between the parties. It is a good thing to resolve according to law part of a dispute, as it may allow progress to be made on wider aspects of the dispute. For example, in the South China Sea arbitration, clarifying ‘the interpretation and application of various articles to rocks, reefs and low-tide elevations’ could ideally ‘narrow the matters in dispute between parties and allow negotiated settlement of other issues such as maritime boundaries’.¹²⁸ It may also allow States normally lacking the power to enforce international law on their own to avail themselves of the ‘normative power’ of binding decisions created by treaty-based compulsory dispute settlement systems.¹²⁹ Given that dispute settlement under UNCLOS was intended to be both a sword and a shield—especially for the interests of developing States—there should be reluctance to find that justiciable aspects of a dispute should be ruled out of bounds because they arise in a wider factual context implicating legal questions falling either wholly outside of the Convention or within one of its exceptions to jurisdiction.

Finally, the question of how to judge the judges arises. The argument here is that the views of commentators are of limited assistance. How judicial performance is assessed in relation to questions such as whether jurisdiction should have been taken in a case such as the South China Sea arbitration turns too much on prior commitments: especially the tendency to formalism in law of the sea scholarship, with its commensurate emphasis on the procedural model of judging and tendency to downplay the social governance function of judging. A scholar committed to the procedural approach will find much to fault in the South China Sea decision. A scholar more sympathetic to a substantive model will find much to praise.

A better guide to the impact of such decisions on the legal order of the oceans is the actual reaction of States. In the first place, it is applicant States that have put arguments to arbitral tribunals that have been seen by some as jurisdiction-expanding. Plainly some States do favour a wider jurisdictional ambit for UNCLOS dispute settlement. In terms of the reaction of other States, there has not been any stampede to the exits. There have been no withdrawals from UNCLOS in the manner of the actual and threatened withdrawals from the International Criminal Court, nor any efforts to stymie the operation of the dispute settlement system as have been seen at the WTO.¹³⁰ The drafting of the BBNJ dispute settlement clause does, however, appear to reflect a degree of nervousness about the extent to which UNCLOS dispute settlement has infringed upon territorial and sovereignty disputes. Nevertheless, the final shape of Article 60 of the Agreement does not reveal a wholesale retreat from compulsory dispute settlement jurisdiction. The limitations on jurisdiction in paragraph 9 apply only to disputes arising under the Agreement, and by its own terms the Agreement does not constitute subsequent practice evidencing the parties’ interpretation of jurisdiction under Part XV generally.

On the whole, then, States seem perhaps more nervous than alarmed about the future of dispute settlement under UNCLOS. In general, States wish to be seen as being in favour of compulsory dispute settlement—they would just rather it was not applied to them.

ACKNOWLEDGEMENTS

The research for this article was supported in part by an Australian Research Council Future Fellowship. The authors are grateful to the anonymous reviewers for their helpful comments.