Why Finality Demands Attention

There are few principles in dispute resolution that carry the weight of res judicata. The idea that a matter, once finally decided, should not be relitigated is foundational to every legal system on earth. It shields parties from the burden and expense of perpetual litigation, conserves the resources of adjudicators, and underpins the basic social compact that disputes must, at some point, come to rest. In international commercial arbitration, where parties from different legal traditions and different jurisdictions submit their disputes to private tribunals under different substantive laws, the principle takes on even greater significance. And yet, in my experience, it is one of those areas where the practice has long lagged behind the aspiration.

The IBA Arbitration Committee’s Task Force on Res Judicata in International Arbitration released its report in early 2025. I will be frank: though I was aware of it at the time, it took me several months before I could sit down with the full document and give it the close reading it deserves. Life in international practice has a way of pushing “important but not urgent” items down the queue. Having now spent considerable time with the report, and having reflected on it against the backdrop of disputes I have personally handled where the preclusive effects of prior awards were fiercely contested, I feel compelled to share my thoughts. The report is excellent. It is also overdue, and I do not mean that as a criticism of the Task Force but rather as a comment on the state of the field. We should have been having this conversation more loudly, and earlier.

What follows is my attempt to engage with the report’s findings, to test its conclusions against the realities of practice, and to offer some additional reflections on where the international arbitration community should go from here.

The Problem: A Patchwork Without a Pattern

The difficulty at the heart of the report is one that every experienced arbitration practitioner will recognise. When a party invokes a prior arbitral award as a bar to claims or issues raised in a subsequent arbitration, the second tribunal faces a threshold question that ought to be straightforward but almost never is: what standard governs the preclusive effect of the earlier award?

The Choice of Law Labyrinth

In practice, most tribunals reach for a domestic law standard identified through a conflict of laws analysis. The most common candidate is the law of the seat of the second arbitration, on the reasoning that res judicata is procedural and the lex arbitri governs procedure. A smaller number of tribunals apply the substantive law of the contract, typically when the parties have framed their arguments accordingly. Still others apply what the Task Force aptly calls a “hybrid” approach, blending domestic law with transnational principles, including the 2006 ILA Recommendations on Res Judicata and Arbitration.

The result, as the report demonstrates through a truly impressive survey of arbitral case law, is inconsistency that borders on incoherence. Tribunals do not agree on which domestic law to apply, often fail to explain their reasoning, and sometimes apply a test that bears little resemblance to the law they have nominally selected. The Grupo Unidos cases are a striking example. There, ICC tribunals applied Panamanian law to reject issue preclusion that would plainly have been available under the US federal law of the seat. The choice of domestic law proved outcome determinative, and the parties had no reliable way to predict which law the tribunal would choose. That is a deeply uncomfortable state of affairs.

Divergent Scope Across Legal Traditions

Even when the choice of law question is settled, the scope of preclusion varies enormously depending on the system selected. Under English law, issue estoppel and the extended doctrine in Henderson v Henderson can preclude not only the re-litigation of claims and issues actually decided, but also matters that could and should have been raised in the earlier proceeding. Under the traditional civil law approach in jurisdictions such as Japan or pre-reform France, res judicata was confined to the dispositive part of the judgment and required strict identity of parties, object, and cause of action, leaving the reasoning and any unraised claims entirely open for future litigation.

Let me put this in concrete terms. A party that prevails in a first arbitration seated in London may reasonably believe that the dispute is comprehensively settled. If the losing party then initiates a second arbitration seated in a civil law jurisdiction with a narrower res judicata standard, the winner of the first arbitration may discover, to its dismay, that the second tribunal considers itself free to reopen issues that the first tribunal decided in its reasoning but did not include in the dispositive part. This is not hypothetical. The Swiss Federal Tribunal’s decision in the law firm partnership case described in the report illustrates exactly this scenario: it held that a German seated award’s reasoning carried no res judicata weight in a subsequent Swiss seated arbitration. The practical consequences for the winning party are severe.

Domestic Rules That Do Not Fit

The report’s most compelling observation, to my mind, is that domestic res judicata standards were designed for domestic court judgments and reflect the procedural architecture of their home systems. International commercial arbitration is a fundamentally different creature. Parties choose arbitration precisely to escape the peculiarities of any single national court system. They expect a neutral, efficient, and final resolution of their dispute. Applying the procedural law of a seat that was often selected for its neutrality and pro-arbitration courts, rather than for its views on res judicata, is an exercise in formalism that serves nobody’s legitimate expectations.

I have witnessed tribunals spending entire hearing days on the preliminary question of which domestic law governs the preclusive effect of a prior award. These hearings involved expert testimony on foreign procedural law and generated hundreds of pages of briefing, all before anyone actually addressed whether the prior award precluded the claims at hand. This is a poor use of the arbitral process, and an affront to the efficiency that parties expect and pay handsomely for when they choose international arbitration.

The Convergence the Report Identifies, and Why It Matters

One of the report’s most valuable contributions is its careful demonstration that the gap between common law and civil law approaches to res judicata is narrower than conventional wisdom suggests, and is continuing to narrow.

The French Cour de cassation’s 2006 Cesareo decision, which required claimants to raise all grounds supporting their claim in initial proceedings, moved French law decisively toward a fact based conception of res judicata that is not terribly far from the common law “transaction” test. Belgium’s 2015 amendment to Article 23 of its Judicial Code followed a similar path. Italian courts have progressively extended res judicata to cover “what was raised and what could have been raised,” and the Corte di Cassazione’s 2014 en banc decisions on contract nullity mark a significant expansion of preclusive effects beyond the dispositive part. Germany’s focus on operative facts (Lebenssachverhalt) as the touchstone for identity of claims, and Switzerland’s insistence on a substantive rather than grammatical comparison of claims, point in the same direction.

This convergence is not coincidental. It reflects a shared recognition, across legal traditions, that narrow res judicata standards create perverse incentives for tactical re-litigation and undermine the legitimacy of adjudicative processes. The fact that this evolution has occurred organically within domestic systems makes the case for an autonomous arbitration specific standard considerably stronger. If the most sophisticated civil law jurisdictions are already moving toward broader preclusion, there is less reason to fear that an autonomous standard reflecting those broader principles would be alien to any major legal tradition. The ground, so to speak, has been prepared.

The Autonomous Approach: Legal Foundations and Practical Merits

Party Autonomy as the Primary Anchor

The report correctly identifies party autonomy as the most secure legal foundation for an autonomous res judicata standard. In the absence of overriding mandatory rules or public policy, parties are free to agree on the procedural and substantive framework of their arbitration. If the parties agree, whether explicitly in their arbitration agreement, through the adoption of institutional rules that incorporate autonomous res judicata principles, or through terms of reference, that the preclusive effects of any prior award shall be assessed according to an autonomous standard, no tribunal has reason to look elsewhere.

The LCIA’s Article 26.8 offers an instructive precedent here. It provides that every award “(including reasons for such award) shall be final and binding on the parties.” By incorporating this language, parties who select LCIA arbitration have effectively opted into a broader conception of res judicata that extends beyond the dispositive part to the reasoning. This is a real world demonstration, not a theoretical possibility, that party autonomy can deliver meaningful standardisation without anyone needing to pass a law.

Inherent Powers as a Complementary Basis

Where the parties have not made an express choice, the report argues, and I think persuasively, that tribunals can apply an autonomous standard on the basis of their inherent powers. The ILA’s 2014 Report on Inherent Powers in International Arbitration recognised that tribunals possess implied powers to give effect to the parties’ agreement and to protect the integrity of the arbitral process. Applying a res judicata standard tailored to the specific features of international arbitration, rather than mechanically importing a domestic law that no party chose for its preclusion rules, falls well within this mandate.

I would go a step further than the report on one point. The principle of finality is not merely a procedural convenience. It is a constitutive feature of the arbitral mandate itself. When parties agree to arbitrate, they agree that the tribunal’s decision will be final and binding. A tribunal that applies a res judicata standard so narrow as to leave the door wide open for re-litigation of the same dispute under a different label is, in a very real sense, failing to honour the mandate the parties conferred. Conversely, a tribunal that applies an unduly broad standard may deprive a party of its right to have genuinely distinct claims adjudicated. The autonomous standard must navigate between these poles. It should be calibrated to serve the finality that the parties bargained for, while preserving the fairness that legitimises the arbitral process in the first place.

The Public Policy Obstacle: Significant But Not Insurmountable

The most serious challenge to an autonomous standard comes from jurisdictions whose courts regard their domestic res judicata rules as part of public policy, and which would therefore review, and potentially set aside, an award that applies a different standard. This is not a theoretical concern. It has teeth.

The report’s survey identifies Switzerland and Germany as the most prominent examples. The Swiss Federal Tribunal has been explicit: international arbitral tribunals seated in Switzerland must apply the same res judicata principles as Swiss state courts, and any departure constitutes a violation of Swiss public policy. The German Federal Court of Justice has taken a comparable position. Courts in Spain, Argentina, and Brazil have signalled similar views, though the case law is less developed.

These positions deserve respectful attention. But having thought about this at length, I do not believe they should be treated as a veto on the entire project. Here is why.

First, the jurisdictions that regard res judicata as public policy represent a minority, an important one, of the world’s arbitration seats. The predominant position, as the report shows, is that a tribunal’s determination on res judicata will not be reviewed on public policy grounds. Courts in England, Singapore, the United States, France, and common law Canada have all declined to set aside awards on this basis. Belgium, Italy, Japan, Romania, and Sweden appear likely to follow suit.

Second, an autonomous standard need not be a maximalist expansion of preclusive effects. The Swiss concern, expressed in the law firm partnership case, was that a tribunal seated in Switzerland might consider itself bound by the reasoning and factual findings of a prior foreign award, and thereby decline to decide a claim that, under Swiss law, had a different subject matter. An autonomous standard that extends res judicata to the necessary reasoning of a prior award, while preserving the tribunal’s obligation to decide genuinely distinct claims, need not conflict with Swiss or German public policy. The key, as always, is calibration.

Third, and this is the point I find most reassuring, the history of international arbitration is full of examples of domestic courts eventually accommodating autonomous arbitral standards that were initially viewed with suspicion. The IBA Guidelines on Conflicts of Interest, the IBA Rules on the Taking of Evidence, and the IBA Guidelines on Party Representation all addressed areas touching on fundamental procedural rights and public policy in many jurisdictions: arbitrator independence, document disclosure, witness preparation. None of these instruments has led to a wave of set asides. Instead, they have contributed to a broader judicial acceptance that arbitration operates by its own logic, provided that basic fairness is preserved. I see no reason why res judicata should be different.

Practical Considerations for the Autonomous Standard

What the Standard Should Cover

The Task Force’s recommendation to limit the initial scope of any guidelines to objective res judicata, meaning the preclusive effect of a prior award on the same or related claims and issues, without addressing the identity of parties question, strikes the right balance between ambition and prudence. I think this is wise. Trying to do everything at once is a reliable way to accomplish nothing.

On the objective side, the guidelines should address at least three questions that recur constantly in practice. First, they should confirm that the preclusive effects of a prior award extend not only to the dispositive part but also to the determinations in the reasoning that are necessary to it. This is the position taken by the ILA Recommendations, and it reflects the functional reality that an award’s reasoning is often inseparable from its operative conclusions. Second, the guidelines should recognise some form of issue preclusion: where a factual or legal issue has been actually litigated and decided, and the determination was essential to the prior award, the issue should not be open to re-litigation between the same parties. Third, the guidelines should address the preclusion of claims that could and should have been raised in the prior proceeding, subject to appropriate safeguards against procedural unfairness.

What the Standard Should Not Do

Equally important is what the autonomous standard should not attempt. It should not purport to override the parties’ express agreement on the applicable res judicata standard. It should not displace mandatory rules of the seat or of the place of enforcement, where such rules genuinely form part of the public policy of those jurisdictions. And it should not extend to investment arbitration, where the identity of parties question and the treaty specific legal bases for claims create complexities that require their own separate treatment.

The Task Force is right to recommend against addressing the subjective element of res judicata at the outset. The identity of parties question raises profoundly different issues in commercial and investment arbitration. Any attempt to address it in the first iteration of the guidelines risks politicising the entire project, and I have seen enough promising initiatives stall under the weight of overreach to take that risk seriously.

Flexibility and Due Process

Any autonomous standard must preserve sufficient flexibility to account for the circumstances of individual cases. A rigid rule that mechanically precludes all claims arising from the same factual matrix would be as unsatisfactory as the current patchwork, just unsatisfactory in a different direction. The standard should allow tribunals to consider whether a party had a full and fair opportunity to present its case in the prior proceeding; whether the procedural framework of the prior arbitration was adequate to the claims now said to be precluded; and whether applying preclusion in the specific case would work a fundamental injustice.

The ILA Recommendations’ concept of “procedural unfairness or abuse” as a qualification on the preclusion of unraised claims provides a useful model. It acknowledges what every experienced arbitration lawyer knows: that preclusion must serve justice, not frustrate it.

Lessons from Practice

In my experience, the most contentious res judicata disputes in commercial arbitration arise not from crude attempts to relitigate an identical claim, but from situations where the relationship between the first and second proceedings is genuinely ambiguous. Let me offer a typical scenario. Consider a construction dispute in which the first arbitration determines that the contractor is entitled to an extension of time for certain delay events. In a second arbitration, the employer claims liquidated damages for late completion, arguing that different delay events caused the delay. The contractor invokes the first award, arguing that the time extension already determined the completion date and that the employer’s claim is precluded.

Whether the employer’s claim is barred depends entirely on the scope of the preclusion standard. Under a narrow civil law approach, the employer may be free to relitigate the completion date under a different cause of action. Under a broad common law approach, the employer may be estopped from contesting factual findings that were essential to the first award. I have lived through disputes like this, and the frustration on both sides is palpable. The parties cannot plan their litigation strategy. They cannot assess the finality of the first award. And they face the prospect of spending significant sums on a threshold question that, in a well functioning system, should be readily determinable.

An autonomous standard that provides clear, context sensitive guidance on these recurring scenarios would represent an enormous practical improvement. It would not just benefit lawyers; it would benefit the commercial parties who hire us and who, at the end of the day, simply want to know whether a dispute is over.

The Road Ahead: From Report to Guidelines

The Task Force’s recommendation that the IBA Arbitration Committee establish an Expanded Task Force to draft guidelines on an autonomous res judicata standard deserves strong support from across the arbitration community. The composition of that Expanded Task Force will be critical. It must include not only practitioners and academics from both common law and civil law traditions, but also in-house counsel who can speak to the expectations of commercial parties, former judges who can address enforcement concerns, and representatives from key jurisdictions where res judicata is regarded as public policy. If the guidelines are developed by a narrow group, they will be perceived as reflecting a narrow perspective, and they will fail.

The drafting process should be iterative and transparent, with opportunities for public comment. The IBA’s experience with its Rules on Evidence, Guidelines on Conflicts of Interest, and Guidelines on Party Representation provides a proven methodology. The resulting guidelines should be accompanied by explanatory commentary that addresses the policy choices made, the alternatives considered, and the relationship between the autonomous standard and domestic law. Arbitrators need to understand not just what the standard says but why it says it, especially when they face arguments from parties invoking domestic law.

Institutional adoption will be key to the guidelines’ success. If major arbitral institutions (the ICC, LCIA, SIAC, HKIAC, SCC, and others) incorporate the autonomous standard into their rules or recommend it in their practice notes, the standard will gain traction rapidly. The LCIA’s existing Article 26.8 demonstrates that institutional action in this area is both feasible and effective. I would also encourage arbitral institutions to host events, publish model clauses, and issue practice guidance that makes the autonomous standard accessible and easy to adopt.

Conclusion

Looking back on the year since the Task Force released its report, I am struck by how timely the work proves to be. The arbitration community has grown increasingly aware that the current approach to res judicata is unsustainable. The convergence of civil law and common law standards has created a more hospitable environment for an autonomous approach than existed when the ILA issued its Recommendations nearly two decades ago. The proliferation of complex, multi contract disputes has made the preclusion question more frequent and more consequential. And the broader trend toward harmonisation in international arbitration, reflected in the growing influence of soft law instruments, the standardisation of procedural practices, and the professionalisation of the arbitral community, provides a receptive context for new guidance.

An autonomous res judicata standard will not eliminate all uncertainty. It will not prevent creative litigants from testing its boundaries. And it will not bind the courts of jurisdictions that regard their domestic standards as mandatory public policy. But it will do something that the current patchwork cannot: it will give tribunals, parties, and counsel a shared reference point, a common language for discussing the preclusive effects of arbitral awards, grounded in the specific needs and features of international arbitration rather than the historical accidents of any single domestic system.

That, by itself, would be a significant achievement. The IBA Arbitration Committee has the credibility, the institutional capacity, and the track record to deliver it. I, for one, hope they move forward without delay.\ \ Av. Dr. Ömer KESİKLİ