THE DICHOTOMY BETWEEN JURISDICTION AND ADMISSIBILITY IN INTERNATIONAL ARBITRATION

Abstract

The dichotomy between jurisdiction and admissibility developed in public international law has drawn much attention from arbitrators and judges in recent years. Inspired by Paulsson's ‘tribunal versus claim’ lodestar, attempts have been made to transpose the distinction from public international law to investment treaty arbitration, yielding a mixed reception from tribunals. Remarkably, a second leap of transposition has found firmer footing in commercial arbitration, culminating in the prevailing view of the common law courts in England, Singapore and Hong Kong that arbitral decisions on admissibility are non-reviewable. However, this double transposition from international law to commercial arbitration is misguided. First, admissibility is a concept peculiar to international law and not embodied in domestic arbitral statutes. Second, its importation into commercial arbitration risks undermining the fundamental notion of jurisdiction grounded upon the consent of parties. Third, the duality of ‘night and day’ postulated by Paulsson to distinguish between reviewable and non-reviewable arbitral rulings is best reserved to represent the basic dichotomy between jurisdiction and merits.

I. INTRODUCTION

Over the last century, the universe of international arbitration has rapidly expanded in scope and prominence. Due to the experience of operating across different dispute resolution regimes (sequentially or even concurrently), many arbitrators, judges, lawyers and scholars are naturally inclined to find commonalities with an eye for harmonisation. Nevertheless, transposition or ‘shoehorning’ of principles from one unique sui generis regime to another ought to be treated with extreme caution.¹

This article seizes on a recurring issue straddling three independent separate systems (or ‘spheres’) and their corresponding adjudicative bodies. The first sphere of public international law falls under the vigilance of the International Court of Justice (ICJ).² The second sphere of international investment law encompasses arbitral tribunals of investor–State dispute settlement (ISDS) instituted via multilateral or bilateral investment treaties (BITs), particularly under the auspices of the International Centre for Settlement of Investment Disputes (ICSID).³ The third sphere of national laws focuses on recent landmark rulings by national courts in common law systems on the reviewability of arbitral awards (in particular, England, Singapore and Hong Kong).⁴

Over two decades, the duality of jurisdiction and admissibility has made two giant leaps of transposition: first, from the first sphere (ICJ) to the second sphere (ISDS); and second, from the second sphere (ISDS) to the third sphere (national courts). Despite the wealth of arbitral awards and academic literature,⁵ the first phase of transposition has elicited both resistance and reception in equal measure. Yet the second phase of transposition has rapidly escalated, culminating in the prevailing view in commercial arbitration that only issues of jurisdiction, but not of admissibility, are reviewable.

Instead of riding the wave, this article boldly swims against the tide of transposing the ‘jurisdiction versus admissibility’ dichotomy into the sphere of international commercial arbitration.⁶ Section II explores the origins and practice of the ICJ's usage of the dichotomy. Section III surveys the sprawling ISDS landscape marked by stark divergence among arbitrators in the doctrinal recognition and practical application of the dichotomy. Section IV examines the recent rulings of the common law courts endorsing the dichotomy. Section V deconstructs the dogma behind the dichotomy (ie Paulsson's lodestar) and exhorts re-interpreting the duality of ‘night and day’ postulated by Paulsson for the purpose of distinguishing between reviewable and non-reviewable arbitral rulings to instead represent the fundamental dichotomy between jurisdiction versus merits.

II. PUBLIC INTERNATIONAL LAW (ICJ)

The duality of jurisdiction and admissibility originates from the ICJ's practice. Over many decades, the two concepts have grown in importance to advance a singular but significant purpose—the characterisation of preliminary objections prior to the determination of merits.

A. Jurisdiction, Competence and Admissibility

Historically, the term ‘admissibility’ drew little attention among public international lawyers.⁷ The earliest seeds were planted by Daxner's dissenting opinion in the ICJ's maiden decision of Corfu Channel.⁸ According to Daxner (ad hoc judge of Albania), ‘jurisdiction’ carries two meanings under international law: (1) ‘to recognize the Court as an organ instituted for the purpose jus dicere and in order to acquire the ability to appear before it’; and (2) ‘to determine the competence of the Court, i.e., to invest the Court with the right to solve concrete cases’.⁹

Daxner's ideas were nurtured by Fitzmaurice's scholarly writings in the 1950s. Fitzmaurice characterised jurisdiction as a ‘field’ from the aspects of ratione materiae, ratione personae and ratione temporis ¹⁰ in contrast to the narrower ‘competence of a court to hear and determine a particular case belonging to the category to which its jurisdiction relates’.¹¹ Modern scholars are divided on the merit of Fitzmaurice's treatise, calling it ‘most sophisticated’¹² but also ‘less impressive’.¹³. Upon elevation to the bench of the ICJ a decade later, Fitzmaurice seized the opportunity in Northern Cameroons to expound on the blurry line between questions of jurisdiction (‘competence of the Court to act at all’) and admissibility, receivability or examinability (‘nature of the claim, or to particular circumstances connected with it’).¹⁴ After stepping down from the bench, Fitzmaurice continued to refine his theory prolifically:

[T]here is a clear jurisprudential distinction between an objection to the jurisdiction of the tribunal, and an objection to the substantive admissibility of the claim. The latter is a plea that the tribunal should rule the claim to be inadmissible on some ground other than its ultimate merits; the former is a plea that the tribunal itself is incompetent to give any ruling at all whether as to the merits or as to the admissibility of the claim.¹⁵

Fitzmaurice's views left a profound legacy on the ICJ's rules and jurisprudence.¹⁶ Article 79(1) of the ICJ's Rules of Court concerning ‘Preliminary Objections’ stipulates that the ICJ ‘may decide, if the circumstances so warrant, that questions concerning its jurisdiction or the admissibility of the application shall be determined separately’.¹⁷ In Oil Platforms, the ICJ observed that ‘[o]bjections to admissibility normally take the form of an assertion that, even if the Court has jurisdiction and the facts stated by the applicant State are assumed to be correct, nonetheless there are reasons why the Court should not proceed to an examination of the merits’.¹⁸ In Croatia/Serbia, the ICJ reaffirmed that the distinction between preliminary objections based on admissibility and those based on jurisdiction is ‘well recognized in the practice’ of the ICJ.¹⁹

B. Taxonomy of Admissibility and Test of Jurisdiction

It is axiomatic that the jurisdiction of the ICJ is derived from the consent of States.²⁰ It is less certain, however, what matters relate to admissibility. In Croatia/Serbia, the ICJ clarified that an objection of admissibility essentially postulates ‘a legal reason, even when there is jurisdiction, why the Court should decline to hear the case, or more usually, a specific claim’ and ‘often of such a nature that the matter should be resolved in limine litis’.²¹ Such objection ‘covers a more disparate range of possibilities’ than a jurisdictional objection.²² From a survey of the ICJ's long line of precedents, a non-exhaustive list of admissibility objections includes:

• • Dispute without object or purpose (mootness).²³

• • Non-exhaustion of local remedies.²⁴

• • Absence of bond of nationality between State and individual.²⁵

• • Absence of indispensable third party in proceedings.²⁶

• • Absence of legal right or interest.²⁷

• • Delay in submission of claim.²⁸

The ICJ's stance of leaving the categories of admissibility open-ended is to be juxtaposed with its examination of jurisdictional objections solely focused upon the scope of jurisdictional instruments as a matter of textual interpretation.²⁹ In Congo/Rwanda, the ICJ affirmed that any conditions ‘expressed in a compromissory clause in an international agreement’ impose limits upon the State's consent to the ICJ's jurisdiction and therefore ‘the examination of such conditions relates to its jurisdiction and not to the admissibility of the application’.³⁰ Recently, in Georgia/Russia ³¹ and Ukraine/Russia,³² the ICJ deemed referral to negotiations as a jurisdictional ‘precondition to its seisin’ under the compromissory clause of the International Convention on the Elimination of All Forms of Racial Discrimination.³³ As expounded in Fitzmaurice's oft-cited separate opinion in Northern Cameroons:

A given preliminary objection may on occasion be partly one of jurisdiction and partly of receivability, but the real distinction and test would seem to be whether or not the objection is based on, or arises from, the jurisdictional clause or clauses under which the jurisdiction of the tribunal is said to exist. If so, the objection is basically one of jurisdiction. If it is founded on considerations lying outside the ambit of any jurisdictional clause, and not involving the interpretation or application of such a provision, then it will normally be an objection to the receivability of the claim.³⁴

In sum, the ICJ only considers a preliminary objection relating to the scope of States’ consent to its jurisdiction as being a jurisdictional objection. All other types of non-consent-based preliminary objections fall under the broad umbrella of admissibility.³⁵ The common theme in decisions on admissibility is the ICJ's exercise of discretion to resolve disputes with ‘procedural fairness and efficiency’ in light of its ‘judicial function’ at the preliminary phase.³⁶

C. Practical Significance of Dichotomy

The ICJ's effort to distinguish the two sets of preliminary objections is reflected in the ordering of procedure.³⁷ As opined in Croatia/Serbia, ‘the effect of a preliminary objection to a particular claim is that, if upheld, it brings the proceedings in respect of that claim to an end; so that the Court will not go on to consider the merits of the claim’.³⁸ When faced with a preliminary objection, the first step is to determine whether the ICJ's jurisdiction to adjudicate the claim is challenged;³⁹ and if not, the second step is to ascertain whether the non-jurisdictional issue ought to be joined with the merits.⁴⁰ The ICJ is not bound by parties’ classification of issues and may re-arrange their ordering.⁴¹

That said, this procedure is more of a practice direction rather than an iron-clad rule. The ICJ maintains discretion on whether to bifurcate proceedings to two phases (preliminary objections and merits).⁴² As observed in Croatia/Serbia, ‘[c]hallenges either to jurisdiction or to admissibility are sometimes in fact presented along with arguments on the merits, and argued and determined at that stage’.⁴³ Exceptionally, the ICJ may even leapfrog a jurisdictional issue to dismiss a claim for inadmissibility.⁴⁴ Such procedural flexibility may account for the ICJ's reluctance to define ‘admissibility’ with minute exactitude.⁴⁵

Notably, the ICJ is a court of first instance and last resort.⁴⁶ Its judgment is final and without appeal.⁴⁷ Accordingly, there is little at stake when parties or judges characterise preliminary objections in a particular case.⁴⁸ This is to be juxtaposed with the emerging practice of national courts in resorting to the characterisation when determining the reviewability of arbitral awards—a purpose never envisaged by the ICJ (Section IV).

III. INTERNATIONAL INVESTMENT LAW (ICSID)

The transposition of the ‘jurisdiction versus admissibility’ dichotomy from public international law to international investment law remains mired in controversy. Tribunals regularly disagree with each other. Arbitrators bicker within the same panel and leave a trail of strong dissents. Scholars do not fully see eye to eye. After almost two decades in development, uncertainty still shrouds the doctrinal basis and practical utility of the dichotomy.

A. Reception or Resistance?

Initially, admissibility as a concept distinct from jurisdiction generated a lukewarm reception from ISDS tribunals. In 2004, the Enron v Argentina tribunal deemed the distinction ‘unnecessary’ since the ICSID Convention ‘deals only with jurisdiction and competence’.⁴⁹ In 2006, the Pan American v Argentina tribunal opined ‘that there is no need to go into the possible—and somewhat controversial—distinction between jurisdiction and admissibility’.⁵⁰ Similar sentiments followed in CMS v Argentina,⁵¹ Bayindir v Pakistan ⁵² and LESI v Algeria.⁵³ In Methanex v USA, the tribunal went as far as finding that it lacked any ‘express or implied power to reject claims based on inadmissibility’.⁵⁴ In academia, Schreuer⁵⁵ and Zeiler⁵⁶ rank among the notable sceptics.

Such scepticism was essentially driven by the absence of the term ‘admissibility’ in the relevant rules of procedure. Article 41(2) of the ICSID Convention stipulates that an objection that a ‘dispute is not within the jurisdiction of the Centre, or for other reasons is not within the competence of the tribunal, shall be considered by the tribunal which shall determine whether to deal with it as a preliminary question or to join it to the merits of the dispute’.⁵⁷ Likewise, Rule 43(1) of the ICSID Arbitration Rules on ‘Preliminary Objections’ permits parties to ‘file a preliminary objection that the dispute … is not within the jurisdiction of the Centre or for other reasons is not within the competence of the Tribunal’.⁵⁸ The duality of ‘jurisdiction’ and ‘competence’ harks back to Fitzmaurice's formative ideas on ICJ practice. The latter term is scrutinised even less frequently than admissibility by ISDS tribunals.⁵⁹ It is conceivable that admissibility falls within the ambit of the ‘for other reasons is not within the competence of the tribunal’ limb of preliminary objections.

Despite the early wave of scepticism, proponents of transposition gradually grew in numbers and strength. An early trailblazer was Highet whose passionate dissent that ‘lack of jurisdiction refers to the jurisdiction of the Tribunal and inadmissibility refers to the admissibility of the case’ in Waste Management v Mexico ⁶⁰ left an indelible mark in scholarly circles.⁶¹ In SGS v Philippines, the majority deemed the giving of effect to an exclusive jurisdiction clause in an investment contract as ‘concerning admissibility of the claim, not jurisdiction in the strict sense’.⁶² This approach was followed in Bureau Veritas v Paraguay.⁶³

The breakthrough arrived in 2005 when Paulsson delivered a thought-provoking article that turned the tide almost instantly.⁶⁴ Aside from garnering countless citations by tribunals, courts and scholars, Paulsson's article is notable for its creative theories and analogies, which are deserving of an entire section of appraisal (Section V). At this juncture, it suffices to mention a few key recurring themes. First, Paulsson's core premise is that the ‘jurisdiction versus admissibility’ dichotomy is of ‘considerable concrete importance’ in the reviewability of arbitral awards.⁶⁵ Second, he aimed to enhance harmonisation in the treatment of this dichotomy by international arbitral tribunals and national courts.⁶⁶ Third, Paulsson vividly analogised the concepts of jurisdiction and admissibility to ‘night and day’ with a ‘twilight zone’ in between.⁶⁷ Fourth, he dismissed the ICJ's views as ‘pure abstractions’, deconstructed Fitzmaurice's theories and discarded the conventional treatment of the two concepts under public international law.⁶⁸

The climax is a compass to guide wanderers navigating through the ‘twilight zone’. ‘Our lodestar’, Paulsson declared, ‘takes the form of a question: is the objecting party taking aim at the tribunal or at the claim?’⁶⁹ He concluded with a guidance note on determining ‘whether a challenge pertains to jurisdiction or admissibility’:

If the reason for such an outcome would be that the claim could not be brought to the particular forum seized, the issue is ordinarily one of jurisdiction and subject to further recourse.

If the reason would be that the claim should not be heard at all (or at least not yet), the issue is ordinarily one of admissibility and the tribunal's decision is final.

Within a few years, Paulsson's influence seeped into the consciousness of arbitrators.⁷⁰ In 2008, echoing Paulsson almost verbatim, the Micula v Romania tribunal stated that ‘an objection to jurisdiction goes to the ability of a tribunal to hear a case while an objection to admissibility aims at the claim itself and presupposes that the tribunal has jurisdiction’.⁷¹ This set the stage for a titanic clash of ideas that persists today.

B. Typology of Preliminary Objection

As in the ICJ, there is no exhaustive checklist of issues flagged as pertaining to jurisdiction or admissibility in ISDS arbitration. Instead, parties typically present preliminary objections under the general heads of jurisdiction: ratione materiae, ratione personae, ratione temporis and ratione voluntatis. Viewed from this typology, a substantial convergence—if not conflation—can be discerned between some (but not all) preliminary objections typically characterised as jurisdiction and admissibility.⁷²

1. Prematurity of arbitration (ratione voluntatis and ratione temporis)

Preconditions to arbitration are a common feature in the ISDS regime. The standard dispute resolution clause in BITs takes the form of a multi-tier framework requiring procedural steps to be taken by an investor prior to submission to arbitration (eg waiting period, exhaustion of local remedies, fork-in-the-road and waiver).⁷³ Concomitantly, an investor's premature institution of arbitration may be challenged on the basis that the ‘claim is not yet ripe for international jurisdiction’, whether as a matter of ratione temporis ⁷⁴ or ratione voluntatis.⁷⁵

First, concerning a waiting period requirement, tribunals are evenly split as to whether objections of non-compliance with preconditions to arbitration go to the root of a tribunal's jurisdiction. In Ethyl v Canada, a six-month ‘cooling off period’ was construed as not depriving the tribunal of jurisdiction since the claimant could resubmit the same claim by the time of the hearing.⁷⁶ The same outcome was reached in Lauder v Czech Republic,⁷⁷ Bayindir v Pakistan ⁷⁸ and Western NIS v Ukraine.⁷⁹ In SGS v Pakistan, the tribunal ventured a step further by treating ‘consultation periods as directory and procedural rather than as mandatory and jurisdictional in nature’.⁸⁰ However, the tide quickly turned with a wave of awards advocating a more restrictive approach.⁸¹ In 2010, the Burlington v Ecuador tribunal regarded non-compliance with a six-month waiting period requirement as resulting in both the claim being inadmissible and the tribunal lacking jurisdiction.⁸² That same year, the Murphy v Ecuador tribunal explained that such preconditions are ‘a key component of the legal framework established in the BIT’ to allow parties to ‘attempt to resolve their disputes amicably, without resorting to arbitration or litigation, which generally makes future business relationships difficult’.⁸³

Second, on exhaustion of local remedies, the line is blurrier. Some BITs permit an investor to commence arbitration only after instituting proceedings in the host State's courts for a certain duration (typically, 18 months).⁸⁴ Such a domestic litigation requirement was deemed by the Hochtief v Argentina majority ‘as a provision going to the admissibility of the claim rather than the jurisdiction of the Tribunal’.⁸⁵ The same approach was taken by the majority in Abaclat v Argentina ⁸⁶ and İçkale v Turkmenistan.⁸⁷ However, most arbitral awards lean in the opposite direction. The BIT's domestic litigation condition clause was deemed a ‘jurisdictional prerequisite’ in the Dede v Romania award.⁸⁸ In ICS v Argentina, the tribunal considered that such a clause imposed jurisdictional limits since the ‘choice of forum for the submission of disputes in the first instance’ being the local courts goes beyond ‘a question of ripeness’ of the claim.⁸⁹ Upon determining that the claimant's non-exhaustion of local remedies deprived the tribunal of having ‘competence to entertain the claim’, the Wintershall v Argentina tribunal remarked that ‘[e]ven a requirement of prior-negotiation may therefore qualify as a jurisdictional requirement (or it may not), depending on the language and the context’.⁹⁰ In Impreglio v Argentina, the tribunal held that such clause ‘provides a mandatory—but limited in time—jurisdictional requirement before a right to bring a case to ICSID can be exercised’.⁹¹ The same stance was taken by the majority in Daimler v Argentina ⁹² and Kiliç v Turkmenistan.⁹³ Lastly, there is a third wave of agnostic awards that refrains from committing to any characterisation altogether.⁹⁴ In sum, the bulk of ISDS jurisprudence relating to preconditions to arbitration reveals a stark divergence in reception of the ‘jurisdiction versus admissibility’ dichotomy.

2. Ineligibility of claimant (ratione personae)

An ISDS tribunal's jurisdiction ratione personae turns upon whether the claimant qualifies as an ‘investor’ as defined by the BIT's subjective criterion of nationality,⁹⁵ and additionally for ICSID arbitrations, the ICSID Convention's objective outer limits (ie the Broches test)⁹⁶. Since BITs predominantly adopt the ‘incorporation test’ rather than the ‘control test’ in determining the nationality of corporations, such a broad jurisdictional gateway easily dispenses with any objection taking aim at the nationality of its shareholders.⁹⁷

First, consider the recovery of shareholder reflective loss. Despite such loss generally being deemed unrecoverable under national laws,⁹⁸ it is well-settled in the ISDS regime that a claimant may qualify as an ‘investor’ by virtue of owning or controlling shares in a company incorporated in the host State.⁹⁹ Accordingly, the persistent attempts by host States to challenge claimants’ jus standi in the capacity of a majority,¹⁰⁰ minority¹⁰¹ and or indirect¹⁰² shareholder have consistently met with failure. Notably, tribunals remain impervious to cunning attempts to characterise the objection as a matter of inadmissibility to escape scrutiny from the jurisdictional limits embodied in the treaty text. As lucidly put in Enron v Argentina, ‘the essential question’ of ‘whether the claimant invoking the benefit of its provisions qualifies as a protected investor … needs to be discussed under the Treaty irrespective of whether it is labelled a question of admissibility or otherwise’.¹⁰³

Second, consider the permissibility of mass claims. In Abaclat v Argentina, the ICSID tribunal opined that the issue of its conduct ‘in the form of collective proceedings is an issue of admissibility and not of consent’ and adapted the ICSID procedural rules to accommodate the collective claims of 60,000 Italian sovereign debt bondholders.¹⁰⁴ However, in the second mass bondholder claim of Ambiente v Argentina, the ICSID tribunal found it unnecessary ‘to give an answer as to whether the legal issues at stake are to be classified as questions of jurisdiction or admissibility’.¹⁰⁵ Likewise, the tribunals in Alemanni and Argentina ¹⁰⁶ and Adamakopoulos and Cyprus ¹⁰⁷ remained non-committal. In sum, any challenge directed towards a claimant's standing invariably is resolved with reference to the jurisdictional framework of the BIT and the ICSID Convention. This essentially renders the classification of the challenge superfluous.¹⁰⁸

3. Illegality of investment (ratione materiae)

For jurisdiction ratione materiae, the inquiry focuses on the existence of a dispute relating to an ‘investment’ falling within a BIT's subjective definition,¹⁰⁹ and additionally for ICSID arbitrations, the ICSID Convention's objective outer limits (ie the Salini test).¹¹⁰ One notable type of preliminary objection takes aim at the legality of the claimant's investment due to the ‘in accordance with host State's law’ clause embodied in certain BITs.¹¹¹

In Fraport v Philippines, the claimant's deliberate circumvention of Philippine law culminated in a finding of inexistence of any ‘investment in accordance with law’ and the tribunal lacking jurisdiction ratione materiae.¹¹² In the second arbitration, the tribunal restated that the ‘illegality of the investment at the time it is made goes to the root of the host State's offer of arbitration under the treaty’.¹¹³ Put simply, an illegal investment falls outside the scope of the host State's consent to arbitration¹¹⁴—more so for investments tainted by corruption.¹¹⁵ Even in the absence of a ‘legality’ clause in the BIT, conformity with the host State's laws is an implicit requirement since host States ‘cannot be deemed to offer access to the ICSID dispute settlement mechanism’ to illegal investments.¹¹⁶ To extend a BIT's substantive protection to illegal investments would run counter to international public policy,¹¹⁷ and the fundamental principle of nemo auditur pro priam turpitudinem allegans (no party can benefit from their own wrong) under international law.¹¹⁸

To some scholars, an illegality objection in the absence of a legality clause ought to impair the admissibility of the claim rather than the tribunal's jurisdiction.¹¹⁹ The infamous World Duty Free Company award involving the bribery of a former Kenyan president, which held that ‘claims based on contracts of corruption or on contracts obtained by corruption cannot be upheld by this Arbitral Tribunal’,¹²⁰ is postulated as an example.¹²¹ In Churchill Mining v Indonesia, an investment tainted by forgery and fraud during its performance resulted in the claim being held inadmissible.¹²² The entire illegality analysis was conducted through the lens of admissibility because the applicable BITs ‘only contain admission requirements applying at the time of establishment of an investment, which are jurisdictional in nature’.¹²³

Is the classification purely dependent on the presence or absence of the legality clause? Not necessarily so. According to the Minnotte v Poland tribunal, the relevant factors include procedural practicalities (whether the illegality allegations are closely connected to the merits) and gravity of the alleged illegality (whether manifest and connected to the basis of the Tribunal's jurisdiction, ie the making of the investment).¹²⁴ In sum, the reasoning behind the classification of illegality objections lacks consistency.

C. Practical Significance of Dichotomy

Unlike in public international law, voices in ISDS arbitral circles have been assertive and assured in postulating reasons why the ‘jurisdiction versus admissibility’ dichotomy is of paramount importance. Upon closer examination, these reasons lack universal utility, or ring true only in specific circumstances. An analysis of Paulsson's treatise (focusing on reviewability of awards) is reserved for last.

1. Curability and waiver

According to one school of thought, a ruling of jurisdiction is inviolable and unaffected by parties’ conduct.¹²⁵ As the Hochtief v Argentina majority stated: ‘defects in admissibility can be waived or cured by acquiescence: defects in jurisdiction cannot’.¹²⁶ In Daimler v Argentina, the majority opined that all BIT dispute resolution clauses ‘are by their very nature jurisdictional’ due to being ‘a treaty-based pre-condition to the Host State's consent to arbitrate’ which ‘cannot be bypassed or otherwise waived by the Tribunal as a mere “procedural” or “admissibility-related” matter’.¹²⁷ The danger lies in tribunals unconsciously treating curability of non-compliance with mandatory preconditions to arbitration as the cause rather than the effect of classifying an objection. Indeed, such reverse logic may tempt sympathetic arbitrators to construe a precondition to arbitration embedded in the BIT dispute settlement clause as non-jurisdictional because non-compliance is curable. However, jurisdiction is blind to justice. It would be wrong, no matter how well-intentioned, for a tribunal to dismiss defects going to the root of parties’ consent to arbitration as a matter of admissibility in a valiant effort to dispense justice.¹²⁸ In any event, the sweeping proposition that jurisdictional defects are incurable is a fallacy (see Section III.C.2).

A corollary proposition is that a tribunal is empowered to examine jurisdictional issues proprio motu.¹²⁹ A prime example is the legality of an investment. In Infinito Gold v Costa Rica, the tribunal heard corruption allegations based on information furnished by an amicus curiae despite both parties’ denial at the jurisdiction phase¹³⁰ and Costa Rica withdrawing its initial objection at the merits phase.¹³¹ Indeed, Article 43(3) of the ICSID Rules permits a tribunal to ‘on its own initiative consider whether the dispute … is within the jurisdiction of the Centre or within its own competence’.¹³² But are issues of admissibility excluded from the scope of such inherent power? Not necessarily so. A combined reading with the limb ‘for other reasons is not within the competence of the Tribunal’ in Article 43(1) suggests that the term ‘competence’ in the context of preliminary objections is broad enough to encompass admissibility.¹³³

Further, the universal utility of the proposition is doubtful. Aside from illegality, what other circumstances would prompt a tribunal to insist on examining a jurisdictional issue not argued by the parties? In practice, when a party withdraws an objection aimed at the scope of the jurisdictional clause, tribunals accept the concession without further examination. International tribunals are free to consider jurisdictional objections in any particular order¹³⁴ and are inclined to address first whichever objection might directly and conclusively resolve the entire claim.¹³⁵ In short, tribunals are vested with a margin of flexibility in resolving preliminary objections, whether relating to jurisdiction or admissibility.

2. Timing of assessment

The dichotomy is said to be essential to determine the time of assessment of a preliminary objection.¹³⁶ It is trite that the critical date for determining jurisdiction is the date of institution of proceedings—and jurisdiction once established cannot be defeated by subsequent events.¹³⁷ This fundamental rule of international law is significant in the analysis of jurisdiction ratione personae and ratione materiae,¹³⁸ especially to disregard post-arbitration changes in the nationality of an investor¹³⁹ or ownership of an investment.¹⁴⁰ Stopping the jurisdictional clock allows the transfer of investment assets to avoid their indefinite freezing pending arbitration which may drag on for many years—and ultimately facilitates the free flow of investment and maximation of capital.¹⁴¹

Essentially, determining the critical date aims to prevent respondent States from challenging jurisdiction already seized and perfected. However, as Fitzmaurice clarified many decades ago, the ‘converse is not necessarily true’—for example, ‘the principle of forum prorogatum may enable a jurisdiction incomplete at the date of seizure, to be perfected by a subsequent acceptance of, or submission to, the jurisdiction’.¹⁴² In short, a tribunal is permitted to look upon subsequent events to cure jurisdictional defects. Further, determining the critical date is not absolute. The ISDS regime recognises a myriad of other dates relevant in determining the jurisdictional scope of tribunals.¹⁴³ For example, illegality objections are more concerned with dates antecedent to the institution of arbitration, ie whether the illegality impaired the making or performance of the investment (Section III.B.3).

One possible tangible utility in classification is the power vested upon tribunals to order a stay of proceedings to allow an admissibility defect to be ‘cured’.¹⁴⁴ Such an order was granted in SGS v Philippines over a claim deemed inadmissible due to the investment contract containing an exclusive jurisdiction clause referring disputes to the Philippine court.¹⁴⁵ The logic of such relief was rightly criticised by the Bureau Veritas v Paraguay tribunal since a finding of inadmissibility must mean that ‘it is exclusively for that forum to resolve all aspects of the dispute under the exclusive jurisdiction clause’ and therefore the ‘normal course would be to dismiss the claim’.¹⁴⁶ Indeed, a stay may be appropriate where the defect is trivial, such as the precondition of a short ‘cooling-off’ period. However, this further serves to demonstrate that both the determination of the critical date and moulding of appropriate relief are ultimately dependent on the nature of the specific preliminary objection, rather than any pre-conceived classification of the objection as jurisdictional or admissibility.

3. Joinder with merits

Another conventional belief holds that issues of admissibility ‘are more likely to be addressed with the merits’.¹⁴⁷ However, this is scarcely borne out in actual practice. The ‘merits of a dispute’, as aptly put by Read in Anglo-Iranian Oil, ‘consist of the issues of fact and law which give rise to a cause of action, and which an applicant State must establish in order to be entitled to the relief claimed’.¹⁴⁸ Even before the ICJ, the lines between preliminary objections (jurisdiction or admissibility) and merits can be rather blurry. In Ambatielos, the substantive issue posed was whether the UK had an obligation to arbitrate under the UK–Greece Treaty of 1886.¹⁴⁹ The ruling is further complicated by proceedings being split into two phases (the first phase was to resolve whether the ICJ had jurisdiction to decide the claim on the merits, and if not, whether it had jurisdiction to decide the issue of whether the UK is under an obligation to submit the claim to arbitration)¹⁵⁰ and its combined rulings essentially deciding the jurisdiction of another tribunal (Commission of Arbitration).¹⁵¹

Similarly, preliminary objections in ISDS may be closely connected to the merits, especially where the claimant's nationality is fiercely contested (jurisdiction ratione personae)¹⁵² or the claimant's acquisition of an investment impaired by the host State's measure is potentially an abuse of process (admissibility).¹⁵³ In Vannessa Ventures v Venezuela, the assignment of a joint venture contract to the claimant from the original investor without the government's consent became a pivotal point of contention concerning jurisdiction and merits that eventually proved decisive in resolving the claim.¹⁵⁴

Moreover, tribunals occasionally defer illegality allegations, regardless of classification, to the merits phase.¹⁵⁵ In Unglaube v Costa Rica, the decision to join Costa Rica's admissibility objection on the prematurity of the expropriation claim was largely driven by the tribunal's inability at the preliminary phase ‘to master the complex history of the dispute’ and to ascertain the applicable local rules.¹⁵⁶ Hence, a tribunal's decision to join a preliminary objection to the merits is heavily fact-sensitive.

4. Reviewability of award

Finally, the essence of Paulsson's lodestar is addressed—only rulings on jurisdiction, and not admissibility, are reviewable.¹⁵⁷ Some scholars evangelise this proposition as the gospel truth,¹⁵⁸ whilst some remain agnostic¹⁵⁹ or disbelievers.¹⁶⁰ Similarly, its reception in arbitral tribunals is mixed. According to the Abaclat v Argentina majority, one of the ‘different consequences’ arising from the dichotomy is that decisions on lack of arbitral jurisdiction are ‘usually subject to review by another body’ whereas decisions on lack of admissibility are usually not.¹⁶¹ The same view was echoed by the Supervision v Costa Rica tribunal.¹⁶² However, in Urbaser v Argentina, this proposition came under heavy fire for being ‘wrong in theory and useless in practice’ since a decision of inadmissibility can fall within the ambit of ‘manifestly exceeding its powers’ under ICSID's review mechanism.¹⁶³

ICSID awards are reviewable by ad hoc annulment committees. Out of the six grounds of annulment in Article 52(1) of the ICSID Convention, the second limb (‘that the Tribunal has manifestly exceeded its powers’) is most relevant to this analysis.¹⁶⁴ The majority of committees do not consider ‘power’ synonymous with ‘jurisdiction’. In CMS Gas v Argentina, the committee emphatically stated that ‘[i]t is well established that the ground of manifest excess of powers is not limited to jurisdictional error’.¹⁶⁵ In 2016, the Kiliç v Turkmenistan committee rejected the argument that decisions on jurisdiction engage a lower standard of review because ‘the same threshold applies to matters of jurisdiction and the merits in order for the Committee to find that an excess of powers is manifest’.¹⁶⁶ A year later, the TECO v Guatemala committee agreed that ‘there is no textual basis within the ICSID Convention to support such a difference in treatment between excesses of jurisdiction and other excesses of power’.¹⁶⁷

The traffic is not entirely one-way. In 2017, the annulment committee in Venezuela Holdings v Venezuela saw force in the argument ‘that matters of jurisdiction may call for a more rigorous approach than other grounds for annulment, simply because a tribunal ought not to be allowed to exercise a judicial power it does not have (or vice versa)’.¹⁶⁸ The committee further observed that the ‘excess of power’ limb ‘fits most naturally into the context of jurisdiction’ and is ‘less easy to apply’ to a tribunal's ‘discretionary assessment of whether … it is proper in the particular circumstances for that power to be exercised’.¹⁶⁹ However, the committee stopped short of drawing any legal conclusion as the jurisdictional issues were intertwined with the merits.¹⁷⁰ In sum, the core postulation of Paulsson's lodestar that only arbitral rulings on jurisdiction (and not admissibility) are reviewable at the annulment phase has not come to fruition in practice.

IV. REVIEWABILITY OF ARBITRAL AWARDS (NATIONAL LAWS)

Despite facing challenges in adapting to investment treaty arbitration, the ‘jurisdiction versus admissibility’ dichotomy has made a bold leap into the sphere of international commercial arbitration. Remarkably, this second phase of transposition has been smooth, facilitated by an interconnected chain of rulings by the common law courts of England, Singapore and Hong Kong in reviewing ISDS and commercial arbitral awards.¹⁷¹ Regrettably, pleas for commercial arbitrators to ‘resist the temptation to adopt this label’¹⁷² have gone unheeded.

A. ISDS Arbitral Awards

Since 2018, national courts have overseen a second phase of transposition flowing from ISDS jurisprudence. Naturally, the earliest decisions stem from applications to set aside or resist enforcement of ISDS arbitral awards.¹⁷³ In Tatneft v Ukraine, the English High Court was urged to review the ‘abuse of process’ objection that the claimant acquired the investment after a dispute became ‘reasonably foreseeable’ with the ulterior motive of gaining BIT protection.¹⁷⁴ However, Butcher J deemed the objection an issue of admissibility rather than jurisdiction as a matter of textual interpretation,¹⁷⁵ and in light of the objection being grounded upon ‘an imprecise principle of international law’ intertwined heavily with facts, evidence and procedure.¹⁷⁶ A year later, in Korea v Dayyani, Butcher J faced an application to set aside an award for lack of ‘substantive jurisdiction’.¹⁷⁷ For the first three jurisdictional questions, Butcher J found that the claimant made an ‘investment’ under the BIT (ratione materiae).¹⁷⁸ The fourth question challenged the claimant's standing as a shareholder of the company owning the investment assets.¹⁷⁹ After carefully considering the ISDS jurisprudence on shareholder reflective loss, Butcher J found the claimant qualified as an ‘investor’ and thus was not jurisdictionally barred from making a claim to the company's assets.¹⁸⁰ In passing, Butcher J observed that ISDS awards occasionally treat such challenges as an issue of admissibility or damages but stopped short of giving his opinion on such possible arguments.¹⁸¹

In Swissbourgh v Lesotho, an application to set aside an ISDS award came before the Singaporean Court of Appeal (SGCA).¹⁸² One ground of challenge revolved around the BIT's exhaustion of local remedies clause.¹⁸³ Menon CJ acknowledged that the dichotomy of jurisdiction and admissibility ‘has significant practical import in investment treaty arbitration because a decision of the tribunal in respect of jurisdiction is reviewable by the supervisory courts at the seat of the arbitration … or before an ICSID ad hoc committee … whereas a decision of the tribunal on admissibility is not reviewable’.¹⁸⁴ Despite international courts traditionally treating exhaustion of local remedies as an issue of admissibility, Menon CJ deemed its express inclusion in the BIT as a precondition to State parties’ consent to arbitration and therefore a ‘jurisdictional requirement’.¹⁸⁵

It is perfectly understandable for judges to turn to international law when reviewing an ISDS arbitral award. The concern, however, is that their analysis merely scratches the surface of ISDS jurisprudence and scholarly writings. In truth, transposition of the ‘jurisdiction versus admissibility’ dichotomy from the ICJ to ICSID is less definitive and more nuanced than their judgments have suggested (Section III). Regrettably, the courts simply took the dichotomy for granted as settled law.

B. Commercial Arbitral Awards

Within two years, the judgments reviewing ISDS arbitral awards had prompted a second wave of transposition into the sphere of commercial arbitration. In 2020, the Singaporean Court of Appeal in BBA et al v BAZ had to consider the scope of the seat court's powers to undertake a de novo review of whether a claim is time-barred.¹⁸⁶ Reliance was placed, first and foremost, upon its earlier ruling in Swissbourgh v Lesotho.¹⁸⁷ Next, Paulsson's ‘tribunal versus claim’ test was endorsed.¹⁸⁸ Loh J opined that ‘arguments as to the existence, scope and validity of the arbitration agreement are invariably regarded as jurisdictional’¹⁸⁹ whereas the ‘nature of the claim, or to particular circumstances connected with it’, relates to admissibility.¹⁹⁰ Concomitantly, the Court of Appeal declined to undertake a de novo review since ‘a plea of statutory time bar goes towards admissibility as it attacks the claim’ for being ‘stale’ and ‘defective’.¹⁹¹ Later that same year, the Court of Appeal in BTN v BTP endorsed Paulsson's test in reliance of Swissbourgh v Lesotho and BBA et al v BAZ and ruled that ‘a tribunal's decision on the res judicata effect of a prior decision is not a decision on jurisdiction’ susceptible to de novo review.¹⁹²

This trilogy of Singaporean apex rulings reverberated across the common law system. In early 2021, the English High Court in Sierra Leone v SL Mining Ltd heard a jurisdictional challenge against an ICC award on the ground of non-compliance with preconditions in a multi-tier arbitration agreement (ie negotiations).¹⁹³ The critical issue was whether the prematurity of the request for arbitration related to jurisdiction or admissibility.¹⁹⁴ Burton J held that the ‘international authorities are plainly overwhelmingly in support’ of the proposition that such a challenge ‘does not go to jurisdiction’.¹⁹⁵ Paulsson's test and BBA et al v BAZ were cited with approval.¹⁹⁶ Accordingly, a premature claim to arbitration does not engage the supervisory court's ‘substantive jurisdiction’ review mechanism.¹⁹⁷ Burton J's analysis was later approved by the High Court in NWA and others v NVF and others which ruled that pre-arbitration procedural requirements (ie mediation) are not jurisdictional.¹⁹⁸

Later in 2021, the Hong Kong courts aligned themselves with this trajectory. In C v D, the Hong Kong Court of First Instance (HKCFI) held that a condition precedent requiring parties to request negotiations prior to commencing arbitration was an issue of admissibility rather than jurisdiction.¹⁹⁹ The judicial precedents and authorities surveyed included BBA et al v BAZ, BTN v BTP, Sierra Leone v SL Mining and Tatneft v Ukraine.²⁰⁰ The ruling of C v D (endorsed in Kinli Civil Engineering v Geotech Engineering)²⁰¹ was pursued on appeal all the way to the apex court. In 2022, the appeal was unanimously dismissed by a three-member coram of the Hong Kong Court of Appeal (HKCA),²⁰² and in 2023, the appeal was again dismissed by a five-member coram of the Hong Kong Court of Final Appeal (HKCFA).²⁰³ Intriguingly, the judgment ended with an unexpected twist. Four judges deemed the dichotomy ‘helpful’²⁰⁴ and ‘useful’²⁰⁵ as a presumptive aid to determine the reviewability of a preliminary objection. However, Gummow NPJ (non-permanent judge) penned a solitary but stirring separate opinion criticising their reliance on a dichotomy that post-dated the adoption of the United Nations Commission on International Trade Law (UNCITRAL) Model Law, originated from the ICJ and ICSID treaty-based regimes, and displaced or distorted the express language of the domestic statute on reviewability of awards.²⁰⁶

A few months later, Prakash J of the Singaporean Court of Appeal expressed enthusiastic approval of the Hong Kong Court of Final Appeal's adoption of the dichotomy in C v D, aligning with Singaporean jurisprudence.²⁰⁷ It is notable that the Singaporean, English and Hong Kong courts have mainly recycled the same small sample of scholarly writings (which, in turn, cite a selective sample of ISDS authorities).²⁰⁸ Hence, the courts’ characterisation of the views of ‘leading academic writers’ as pointing ‘all one way’²⁰⁹ is an unfortunate overstatement—especially when almost all authorities, directly or indirectly, lead back to a single source, ie Paulsson's lodestar.

V. THREE LESSONS RECONSTRUCTED FROM PAULSSON'S LODESTAR

Despite Paulsson's laudable efforts to break new legal frontiers, transposing the dichotomy between jurisdiction and admissibility from public international law to commercial arbitration is a bridge too far. However, all is not lost from this bold endeavour. Three valuable lessons can be drawn from three critical flaws in Paulsson's analysis.

A. Intractability of Double Transposition

Transposing a principle between two legal regimes is a formidable challenge, let alone attempting to achieve it between three. Even at phase one of transposition, Paulsson had to jettison Fitzmaurice's ideas and refine the concept of admissibility to fit the mould of international arbitration.²¹⁰ However, Paulsson glossed over a fundamental distinction between the regimes. The ICJ is a permanent court accessible to almost all States by virtue of membership in the United Nations,²¹¹ whose jurisdiction over contentious cases is seized by way of special agreement, compromissory clauses, declarations recognising its compulsory jurisdiction, and forum prorogatum.²¹² In contrast, arbitral tribunals are constituted by the actions of parties submitting a dispute to arbitration and nominating arbitrators.²¹³ There is minimal commonality in the types of preliminary objections characterised as admissibility by the ICJ (Section II.B) and ISDS tribunals (Section III.B.1–3).

The second leap of transposition is even more problematic as ISDS arbitration is a sui generis regime. The host State's consent to arbitration is formulated in a complex web of terminologies (eg ‘investment’ and ‘investor’) and the ‘arbitration agreement’ is sealed upon an investor accepting the host State's ‘open offer to arbitrate’.²¹⁴ Hence, BITs create a unique ‘jurisdictional framework’ imbued with multiple dimensions (ratione materiae, ratione personae, ratione temporis and ratione voluntatis).²¹⁵ This is worlds apart from commercial contracts with self-contained and broadly formulated arbitration clauses which delineate the scope of an arbitral tribunal's jurisdiction in more concrete terms. As Gummow NPJ rightly observed, there is ‘no need to import’ the dichotomy between jurisdiction and admissibility into commercial arbitration.²¹⁶

On the reviewability of awards, the divergence between regimes remains stark. There is no universal standard of review for jurisdictional awards: various formulations are found in the ICSID Convention (‘manifestly exceeded its powers’),²¹⁷ UNCITRAL Model Law (‘not falling within the terms of the submission to arbitration’)²¹⁸ and UK Arbitration Act (‘tribunal did not have substantive jurisdiction’).²¹⁹ It is rather telling that all the judicial decisions endorsing the dichotomy of ‘jurisdiction versus admissibility’ essentially dismissed a challenge against a tribunal's positive finding of admissibility that left the award intact (Section IV). Will the courts apply the dichotomy with the same rigour towards a negative finding of inadmissibility that effectively renders a tribunal's refusal to hear a claim on the merits immune from review? It is unlikely. Instead, the courts will either contort the dichotomy to re-characterise the tribunal's finding as jurisdictional in nature—if not abandon the dichotomy altogether—to justify reviewing the negative award. There is therefore no principled basis to presume that issues of admissibility—a word that does even not appear in most domestic arbitral statutes²²⁰—fall outside the scope of review.

It is thus evident that the concept of ‘admissibility’ should not lightly be transposed from public international law into commercial arbitration without critically assessing its necessity and suitability.

B. Fallacy of the ‘Tribunal versus Claim’ Test

To ascertain the scope of parties’ consent to arbitration, all that is generally required is to examine the dispute settlement clause (BIT) or arbitration agreement (contract).²²¹ An adjudicator ought to eschew unnecessary classification when determining jurisdictional challenges.²²² Even Paulsson denounced reliance on ‘labels or metaphors’.²²³ The simplicity of the ‘tribunal versus claim’ test is deceptive. If utilised, an adjudicator risks stretching the distinction between jurisdiction and admissibility ‘too far’ and substituting the essential element of consent with the test (instead of deploying it merely as an ‘analytical tool’).²²⁴ Such error is exemplified by the English High Court's refusal to consider that the question of whether a time-bar constitutes a jurisdictional precondition ‘depends upon the precise wording of the clause’.²²⁵ This is to be juxtaposed with the Singapore Court of Appeal's measured dictum that a time-bar may exceptionally constitute a jurisdictional issue where the arbitration agreement expressly stipulates that time-barred claims fall outside parties’ consent to arbitration.²²⁶

For ISDS arbitration, Paulsson's proponents may feel vindicated by the express recognition of ‘admissibility’ in some BITs.²²⁷ Further, tribunals are empowered under the ICSID Rules²²⁸ and modern BITs²²⁹ to dismiss a claim summarily that is ‘manifestly without legal merit’. However, it is one thing to recognise admissibility objections as non-jurisdictional in nature; it is quite another to treat jurisdictional and admissibility objections as binary opposites, especially for the purposes of moulding relief (dismissal or suspension) or undertaking review (yes or no).

In truth, many issues labelled as ‘admissibility’ are jurisdictional in nature.²³⁰ As the Urbaser v Argentina tribunal bluntly remarked, the ‘distinction contributes more to the confusion than to any elicitation of the issue’.²³¹ Paulsson's lodestar is ‘inadequate’ because ‘it does not tell us how to identify if an objection is targeting the tribunal or the claim’.²³² For instance, the effect of non-compliance with preconditions to arbitration can be viewed in two ways: a claim not ripe for arbitration (admissibility); or conversely, the tribunal incompetent to hear an unripe claim (jurisdiction). For time-bars, even Paulsson concedes that a question of jurisdiction can be framed in terms of whether parties consented to only arbitrate on timely claims.²³³ Ultimately, whether an objection takes aim at the tribunal or claim is a matter of perspective—two sides of the same coin. It is therefore clear that there is no magical formula to determine the scope of parties’ consent to arbitration—the process is all a matter of textual interpretation.

C. Primacy of the Dichotomy between Jurisdiction and Merits

Underpinning Paulsson's lodestar is the duality of what he describes as ‘night and day’.²³⁴ He correctly observes that a ‘twilight zone’ lies in between which renders the dividing line ‘difficult to establish’ and ‘only a fool would argue that the existence of a twilight zone is proof that night and day do not exist’.²³⁵ His error, however, lies in analogising the duality with jurisdiction and admissibility. Instead, the more concrete duality is jurisdiction and merits. In between, lies the twilight zone of admissibility encompassing all other issues (eg illegality, time-bar and abuse). This reformulated dichotomy substitutes Paulsson's lodestar with a simple twist—the power of the tribunal (jurisdiction) versus the substance of the claim (merits). The justifications for this are compelling.

First, jurisdiction has a concrete independent existence. It can be easily discerned. A ‘jurisdictional objection’, as lucidly put in Micula v Romania, ‘relates to a requirement contained in the text on which consent is based’.²³⁶ In contrast, admissibility is an abstract and amorphous concept susceptible to asserting itself with the same unfettered authority as Humpty Dumpty: ‘When I use a word, it means just what I choose it to mean—neither more nor less’.²³⁷ Abuse and arbitrariness are exacerbated if admissibility issues are deemed non-reviewable. Adjudicators will be perversely incentivised to characterise preliminary objections as admissibility: for arbitrators, as a shield to immunise their awards from review; for courts, as a convenient excuse to avoid making difficult decisions.²³⁸

Second, jurisdiction must be defined and determined on its own terms. There is no need for admissibility to exist for the mere sake of proving the existence of jurisdiction. As aptly put by the Daimler v Argentina majority: ‘One cannot use the principle to prove the non-existence of apples based upon the existence of oranges’.²³⁹ Indeed, overstating the concept of admissibility risks creating a false binary (jurisdiction and admissibility) and, concomitantly, unnecessary new hoops and hurdles (tribunal versus claim). Ultimately, to determine whether an objection is jurisdictional or otherwise, it suffices to fall back on the ‘first principle that jurisdiction is grounded upon the consent of the parties’.²⁴⁰

Third, jurisdiction in the wider ‘field’ sense (as traditionally envisaged by Fitzmaurice) encapsulates subsidiary concepts, such as competence and power. It is true that the jurisdiction of international tribunals is grounded upon consent. It is equally true that even when jurisdiction is established, a tribunal may decline to exercise its jurisdiction to hear a particular case due to reasons unrelated to the merits (Section II.A–B). Freed from legal jargon, any ruling on a preliminary objection can be expressed in these simple terms: whether the tribunal can or cannot hear the claim on the merits. Since the outcome is the same, does it matter whether the objection is characterised as relating to the jurisdiction, competence or power of the tribunal, or the receivability or admissibility of the claim? Why should the reviewability of an arbitral award turn upon whether its ruling on a preliminary objection relates to jurisdiction, competence, power, receivability or admissibility? Why should admissibility objections be immune from review when the courts are tasked to ensure a tribunal does not become the sole and final arbiter of its own capacity to hear a claim on the merits?²⁴¹ To echo Prakash J's recent extrajudicial remark, ‘courts play a significant role in sifting through objections that purport to be challenges to the arbitral tribunal's jurisdiction when they are, in fact, challenges to the tribunal's decisions on the merits’.²⁴² Hence, in commercial arbitration, the critical ‘night and day’ dichotomy deserving of judicial scrutiny is the more concrete dichotomy between jurisdiction and merits.

VI. CONCLUSION

Since 2005, Paulsson's lodestar has paved the way for admissibility—a concept originating from the ICJ—to enter the realm of international arbitration. Regrettably, this giant leap of transposition into commercial arbitration is misguided.

In public international law, the ICJ's recognition of the dichotomy between jurisdiction and admissibility arises from the peculiarities of its adjudicative regime. Jurisdiction is defined by the scope of States’ consent expressed in international instruments, whilst admissibility is a broad umbrella encompassing non-jurisdictional preliminary matters.

In international investment law, ISDS tribunals are sharply divided on the dichotomy's doctrinal basis and practical utility. Paulsson's lodestar is applied sparingly and selectively. There is growing concern that the distinction is overstated to the extent of undermining the fundamental concept of jurisdiction anchored on consent. Despite it being the lodestar's core aim, ICSID annulment committees remain unpersuaded of its utility in distinguishing reviewable and non-reviewable preliminary decisions.

The dichotomy and Paulsson's lodestar have, however, received overwhelming endorsement in the English, Singaporean and Hong Kong courts concerning the de novo review of arbitral awards. It has been impressed upon judges—and accepted almost unquestioningly—that arbitrators and scholars endorse its reception in one voice. This has culminated in the troubling trajectory of the ‘tribunal versus claim’ test displacing textual interpretation of arbitral statutes as the primary method to determine the reviewability of arbitral awards on preliminary objections.

Paulsson deserves a place in legal history alongside Fitzmaurice for enriching our understanding of the jurisdiction of international courts and tribunals. Nevertheless, it is Fitzmaurice's theory that is more robust, as Paulsson's theory has three flaws. The first is the intractability of double transposition—the dichotomy is ill-suited to being imported into the sphere of commercial arbitration, especially by national courts when determining the reviewability of arbitral awards. The second is the fallacy of the ‘tribunal versus claim’ test. Issues of jurisdiction and admissibility are more alike than widely (and wrongly) believed, simply because a preliminary objection taking aim at a claim often also challenges a tribunal's jurisdiction to hear the claim. Finally, the duality of ‘night and day’ is better expressed in commercial arbitration as the more fundamental dichotomy between jurisdiction (the power of the tribunal) and merits (the substance of the claim).

There is no compelling principled reason for the dichotomy between admissibility and jurisdiction to be transposed from public international law into commercial arbitration. Rather than being the much-vaunted lodestar guiding lost wanderers, the ‘tribunal versus claim’ test is more akin to a mirage leading even the most seasoned explorers astray. To some, the plea to resist judicial activism may seem rather anti-climactic, yet certain legal axioms deserve recurring restatements to restore focus to fundamental principles. Just as night follows day in life, there is no truer axiom in arbitration than merits follow jurisdiction.