Author: Tommaso Moneta*
| Jurisdiction: |
Topics: |
I. Introduction
The award in Zaur Leshkasheli & Rosserlane Consultants Ltd. v. Republic of Azerbaijan (“Rosserlane”), now the subject of pending annulment proceedings, underscores a procedural feature of investment arbitration that extends beyond the merits of the dispute. The choice of the governing framework determines not only how a dispute will be resolved, but also the procedural architecture through which the resulting award may subsequently be challenged. Once an award has been rendered, objections to the tribunal’s reasoning must therefore be assessed through the legal grounds recognised by that framework, potentially changing the inquiry through which the underlying concern can be advanced. The mechanisms available at the final stage of dispute resolution may therefore also inform the choice of dispute-resolution framework at its outset.
In addressing the Republic of Azerbaijan’s denial-of-benefits defence under Article 17 of the Energy Charter Treaty, the tribunal reviewed competing lines of authority before adopting a different legal approach that became the basis of the award. This article does not address the substantive implications for the denial-of-benefits doctrine, the scope of investor protection under the Energy Charter Treaty (“ECT”), or the balance between investor and state interests. Instead, Rosserlane illustrates why investors and counsel assessing available dispute-resolution avenues may consider not only the procedure through which a dispute will be decided, but also the framework governing any subsequent challenge to the resulting award.
II. The Rosserlane Award
Rosserlane arose under the ECT and concerned investments in Azerbaijan’s oil and gas sector. The procedural inquiry examined in this article originates in the tribunal’s resolution of the parties’ competing legal approaches to whether Article 17(1) ECT could be invoked after the formal commencement of arbitral proceedings. The investor relied upon the line of ECT authority beginning with Plama Consortium Limited v. Republic of Bulgaria, under which a state generally cannot invoke Article 17(1) ECT after acquiring knowledge of an investor and allowing the investment to proceed (Rosserlane, para 485). The Republic of Azerbaijan, by contrast, maintained that denial of benefits remained available notwithstanding the timing of its invocation and that the requirements of Article 17(1) ECT had been satisfied (Rosserlane, para 484).
In the reasoning that became dispositive (Rosserlane, paras 572-587), the tribunal accepted neither party’s arguments in full. Although it accepted the investor’s position that denial of benefits cannot generally be imposed retrospectively, the tribunal determined that the existing case law did not adequately address situations in which the host state lacked prior knowledge of the investor. Relying upon Littop Enterprises Limited, Bridgemont Ventures Limited, and Bordo Management Limited v. Ukraine (“Littop”), the tribunal reasoned that retrospective denial could in principle remain available where the host state did not know of the investor prior to the start of the dispute. The tribunal further concluded that the State Oil Company of Azerbaijan’s knowledge of the investor was neither attributable to the Republic of Azerbaijan nor otherwise relevant for purposes of Article 17(1) ECT.
III. From Tribunal Reasoning to Post-Award Challenge
The rendering of the award in Rosserlane marks the conclusion of the tribunal’s adjudicative function and a corresponding shift in the procedural relationship between the parties and the tribunal. At that stage, the available means of challenging the award are already determined by the procedural framework governing the arbitration. The tribunal’s legal analysis can no longer be contested on the same terms on which the underlying legal questions were argued before the tribunal. Instead, objections to that analysis become relevant only insofar as they engage the legal questions that the applicable post-award regime is competent to determine.
The significance of this procedural transition is independent of whether the tribunal’s distinction in Rosserlanerepresented the preferable interpretation of Article 17 ECT. An objection to the tribunal’s interpretation of prior authority, its legal distinctions, or its analytical framework does not, by itself, establish a basis for challenging the resulting award. Once incorporated into the award, such an objection must instead satisfy a ground of challenge recognised by the applicable post-award regime. In the ICSID framework applicable to Rosserlane, those grounds are prescribed by Article 52 of the ICSID Convention (“Article 52”). The procedural framework therefore determines not only the extent to which tribunal reasoning may subsequently be scrutinised, but the legal inquiry through which an objection to that reasoning can remain relevant after the award has been rendered.
IV. Challenging Tribunal Reasoning Under the ICSID Annulment Framework
Rosserlane illustrates how the application of Article 52 may require consideration of the reasoning through which a tribunal reached its conclusions without permitting reconsideration of those conclusions on the merits. Article 52 prescribes five exhaustive grounds for annulment, although not all bear equally upon objections originating in tribunal reasoning. Improper constitution of the tribunal under Article 52(1)(a) and corruption of a tribunal member under Article 52(1)(c) concern defects distinct from the tribunal’s legal analysis. By contrast, objections originating in the tribunal’s reasoning may, depending upon the circumstances, require consideration under manifest excess of powers under Article 52(1)(b), serious departure from a fundamental rule of procedure under Article 52(1)(d), or failure to state reasons under Article 52(1)(e). Any disagreement concerning the tribunal’s reliance upon Littop, its treatment of competing ECT jurisprudence, or its distinction based on prior knowledge cannot itself constitute a ground for annulment. Rather, an objection concerning that reasoning becomes relevant to annulment only if the circumstances giving rise to it satisfy one of the grounds prescribed by Article 52.
The transition changes the legal inquiry. An allegation that a tribunal incorrectly interpreted or applied the governing law is not equivalent to an allegation that it manifestly exceeded its powers under Article 52(1)(b). ICSID annulment jurisprudence distinguishes between an erroneous application of the applicable law and a failure to apply the proper law (see inter alia Klöckner I, para 60, Enron, paras 68, 220, and Azurix, para 47), while recognising that the line between the two may sometimes be fine. In Rosserlane, this distinction could affect how an objection to the Tribunal’s reliance on Littop is framed at the annulment stage. A contention that the Tribunal incorrectly interpreted Article 17(1) ECT by introducing prior knowledge into its analysis would ordinarily concern the application of the applicable law. A different question could arise, however, if the Tribunal’s prior-knowledge distinction were characterised as substituting a rule of its own construction for the rule contained in Article 17(1), rather than as an interpretation of that provision. Similarly, disagreement with the tribunal’s analytical framework does not establish a failure to state reasons under Article 52(1)(e). A failure to state reasons may arise where the reasons given are so inadequate that the coherence of the tribunal’s reasoning is seriously affected (see inter alia Mitchell v. Democratic Republic of the Congo, Decision on Annulment, para 21). In Rosserlane, an applicant might seek to frame the Tribunal’s departure from Plama and subsequent ECT jurisprudence as an inadequate engagement with a material line of authority rather than merely disagreement with the interpretation ultimately adopted. Such a characterisation would, however, have to account for the Tribunal’s express consideration of that jurisprudence and its explanation of why it regarded the Plama reasoning as unpersuasive. The question under Article 52(1)(e) would therefore concern the existence and adequacy of the Tribunal’s reasons, rather than whether its treatment of Plama and reliance on Littop produced the preferable interpretation of Article 17(1). Nor does concern about how a decisive issue was addressed establish a serious departure from a fundamental rule of procedure under Article 52(1)(d) without satisfying the requirements of that ground. Ad hoc committees have considered whether a tribunal’s failure to address a question submitted to it may, in appropriate circumstances, amount to a serious departure from a fundamental rule of procedure (see Amco I, para 32, and Klöckner I, para 115). In each instance, aspects of the tribunal’s reasoning may remain relevant, but they are examined for the purpose of determining whether an annulment ground has been established rather than whether the tribunal reached the preferable legal conclusion.
This distinction also identifies the limits of post-award scrutiny. Annulment is concerned with the integrity of the adjudicative process rather than the substantive correctness of the decision and does not constitute an appeal (see CDC Group plc v. Republic of Seychelles, Decision on Annulment, para 34). An objection may therefore constitute a coherent criticism of tribunal reasoning yet provide no basis for annulment because it does not satisfy any of Article 52’s exhaustive grounds. Conversely, where an objection does engage one of those grounds, the tribunal’s reasoning may require scrutiny, but only insofar as necessary to determine the distinct question posed by that ground. The ICSID Convention therefore determines not merely the legal categories through which an award may be challenged, but whether and for what purpose objections originating in the tribunal’s reasoning remain legally cognisable after the award has been rendered.
V. Implications for Forum Selection
Rosserlane underscores a dimension of forum selection that extends beyond the conduct of the original proceedings. The framework selected for resolving a dispute also determines the legal architecture applicable once an award has been rendered. As the ICSID annulment framework illustrates, this consequence concerns not merely whether post-award scrutiny is available, but the legal inquiry through which an objection to the adjudication may subsequently be considered. A criticism directed at a tribunal’s substantive reasoning may therefore require examination through a fundamentally different legal question after the award or may provide no cognisable basis for challenge at all.
This consequence is particularly relevant when more than one dispute-resolution avenue is available. An ICSID Convention award is subject to the Convention’s autonomous annulment mechanism and its corresponding recognition and enforcement regime. An award rendered outside that framework may instead be subject to national set-aside proceedings and to recognition and enforcement under the applicable domestic and international framework, including, where applicable, the New York Convention. This distinction also extends to enforcement: whereas the ICSID Convention provides an autonomous recognition and enforcement regime under Article 54, a non-ICSID award falling within the New York Convention remains subject to the grounds for refusal of recognition and enforcement under Article V, including the public-policy ground in Article V(2)(b). These mechanisms do not necessarily provide for reconsideration of the merits, but they subject the resulting award to different institutions, grounds, and legal inquiries. The choice of arbitral framework therefore carries procedural consequences that extend beyond the rendering of the award.
This does not point towards a preferable regime, nor does it suggest that post-award review should determine forum selection. It does, however, identify a downstream consequence of the initial procedural choice. Where competing dispute-resolution avenues genuinely exist, foreseeable post-award considerations are among the variables that can inform that choice. That choice determines the body of law through which counsel must formulate and advance any subsequent challenge, a consideration that may carry greater practical weight where post-award resistance is foreseeable or recognition and enforcement may need to be pursued across multiple jurisdictions. The analysis can extend beyond the procedure for obtaining an award to the legal framework that will govern what can, and cannot, be challenged once that award has been rendered.
*Tommaso Moneta is a Ph.D. candidate at the University of Innsbruck, specialising in public international law