The International Law Commission (ILC), during its seventy-sixth session in May 2025, provisionally adopted the first-reading draft conclusions on Subsidiary means for the determination of rules of international law. Among these, Draft Conclusion 6 [7] addresses the use of judicial decisions, provisionally stating:

Draft conclusion 6 [7] Absence of legally binding precedent in international law 
Decisions of international courts or tribunals may be followed on points of law where those decisions address the same or similar issues as those under consideration. Such decisions do not constitute legally binding precedent unless otherwise provided for in a specific instrument or rule of international law.

This formulation reflects a deliberate political compromise. By employing the permissive term “may,” the Commission seeks to respect state sovereignty and acknowledge the textual boundaries of Article 38(1)(d) of the ICJ Statute, which characterizes judicial decisions as “subsidiary means” rather than sources of law. The Drafting Committee’s caution is understandable: states remain wary of anything resembling stare decisis by stealth, particularly where diverse treaty regimes demand flexibility.

Yet this caution creates a normative gap between text and practice. The problem is not that Draft Conclusion 6 [7] rejects formal stare decisis; that rejection is doctrinally sound. Rather, the issue lies in reducing reliance on prior judicial determinations to casual discretion. By framing precedent as something courts “may” follow, the current draft obscures the strict burden of justification that international tribunals actually impose upon themselves when departing from established jurisprudence. There is a fundamental difference between respecting the absence of formal stare decisis and accurately capturing that burden because the former is a matter of structural hierarchy within sources of international law; the latter is a matter of judicial methodology. Draft Conclusion 6 [7] conflates the two.

Against this backdrop, this post argues that the ILC should replace the permissive “may” with a structured, rebuttable obligation, one that codifies the presumption of consistency already embedded in the practice of international courts and tribunals while explicitly preserving flexibility for justified departure. The post first situates this argument within the Commission’s constitutional mandate of progressive development, before demonstrating the cross-systemic operation of the presumption across the ICJ, investment arbitration, the ECHR international criminal tribunals, and the WTO. It then examines the doctrinal foundations of the “compelling reasons” standard and addresses operationalization concerns, before concluding with what such a reformulation should achieve.

The Mandate of Progressive Development

Under Article 1(1) of the ILC Statute, the Commission’s formal object is the promotion of both the “progressive development of international law and its codification.” Yet the choice of the permissive term “may” was driven by conservative anxieties regarding state sovereignty and the formal boundaries of Article 38(1)(d); anxieties that, while genuine, sit uneasily with the Commission’s institutional legacy.

Progressive development does not require waiting until a practice achieves complete consensus; it entails providing normative scaffolding where systematic practice is emerging but not yet fully crystallized. The ILC has done precisely this before when for example codifying jus cogens in VCLT Article 53, or systematizing countermeasures (ARSIWA Articles 49-54) and aggravated responsibility (Articles 40-41) when consensus was incomplete, and codifying treaty invalidity based on error or fraud despite minimal practice. By reducing the presumption of consistency to a discretionary “may”, the Commission does not merely describe existing law, it actively undercuts an evolving judicial methodology. This is not progressive development; it is normative abdication.

The Cross-Systemic “Presumption of Consistency”

The lived reality of international adjudication demonstrates that courts across various legal regimes do not treat prior jurisprudence as a matter of casual preference. Because the international legal order structurally lacks a centralized legislator, adjudicators cannot merely act as la bouche qui prononce les paroles de la loi; they inevitably become agents of legal development and transformation. This structural reality generates a de facto “presumption of consistency” rooted in institutional legitimacy, legal certainty, and predictability.

The core conceptual insight is this: the true structural test of precedent lies not in its casual replication, but in the normative basis upon which a decision to depart is made. Consequently, the presumption operates through a “negative test” (Judge Read, p. 233); the burden of justification systematically shifts onto the adjudicator wishing to diverge. Rather than asking whether to follow prior decisions, international fora frame the inquiry in reverse, asking whether there is cause not to follow them. Moreover, as von Bogdandy and Venzke observe (p. 991), in practice, disputing parties are aware of this and when a party relies on a prior decision, adversaries rarely dismiss it merely based on the absence of formal stare decisis. Instead, the legal discourse shifts to a dynamic of distinguishing the case and arguing why the factual or legal parameters of the precedent do not apply.

This methodological default is cross-systemic. The ICJ has long established that it cannot disregard the jurisprudence which has been created by its previous decisions unless there is a specific “cause not to follow” them (Cameroon v. Nigeria, 1998, para. 28) or “unless it finds very particular reasons to do so” (Croatia v. Serbia, 2008, para. 104). In international investment arbitration -where no formal hierarchy exists and tribunals are ad hoc– arbitral panels routinely articulate a functional duty to follow consistent solutions established in prior awards “subject to compelling contrary grounds” (Saipem v. Bangladesh, 2009, para. 90; Burlington v. Ecuador, 2012, para. 187; Electrabel v. Hungary, 2012, para. 4.15). The European Court of Human Rights, for instance, has required “cogent reasons” to depart from its precedents in the interests of legal certainty (Cossey v. UK, 1990, para. 35). The ICTY and ICTR Appeals Chambers have consistently ruled that they should follow their previous decisions for the sake of “consistency, certainty and predictability” unless “cogent reasons in the interests of justice” require departure (Aleksovski, 2000, paras. 97,111; Semanza, 2000, para. 92). The WTO Appellate Body has similarly anchored adjudicatory predictability in the rule that, absent cogent reasons, the same legal question must be resolved consistently (US – Stainless Steel, 2008, para. 160).

Across these diverse jurisdictions, a relevant prior decision imposes a strict burden of justification upon any tribunal wishing to depart from it; a reality that Draft Conclusion 6 [7], by reducing this cross-systemic practice to a discretionary “may,” completely fails to register.

Doctrinal Foundations and the “Compelling Reasons” Standard

Mainstream international legal doctrine has long recognized that judicial departure from established precedent requires an exceptionally high threshold of justification; and in doing so, provides the conceptual foundation for the presumption of consistency identified above. Manley O. Hudson stated that a tribunal should utilize the principles underlying its prior decisions “unless cogent reasons should appear for departing from them” (p. 627); Alain Pellet notes that the Court will not depart from jurisprudence unless it finds very particular reasons to do so (p. 56). Sir Hersch Lauterpacht provided the deepest anchor for this consensus: the World Court follows its own precedents for the exact same reasons that guide all courts across domestic and international legal systems, irrespective of whether they are formally bound by stare decisis. Prior decisions are not merely historical records, but an accumulated “repository of” authoritative legal knowledge. Once a court has determined a specific interpretation to be a correct norm, that determination becomes a foundational asset for the entire legal order. Consequently, as Lauterpacht definitively concluded, an adjudicator “is bound to adduce reasons for departing from the obligation of consistency and of observance of settled principles” (p. 13-15). To reduce this universal judicial behavior to casual discretion – as the ILC currently does with “may” – is to deny the very essence of what makes an institution a court of law.

What, then, constitutes a “compelling reason”? Conceptually, such justifications logically encompass several illustrative categories: a demonstrable error in prior reasoning or a decision rendered per incuriam; a fundamental change in the relevant circumstances; subsequent developments in treaty law or custom governing the same subject matter (e.g. Lotus or Nottebohm); or the specific requirements of the tribunal’s constituent instrument or applicable treaty regime. The last two categories deserve particular emphasis as structural realities of international law: a tribunal interpreting a specific BIT is inherently unbound by jurisprudence developed under a different treaty architecture, and a decision rendered before UNCLOS cannot logically dictate outcomes on questions that the convention explicitly resolves. These categories are not exhaustive; their value lies precisely in channeling departure through reasoned justification rather than eliminating it.

One might fear that introducing a “compelling reasons” standard will create vagueness. However, such a concern would be misplaced. All legal standards involve some level of discretion. The proposed formulation simply channels this discretion into a strict duty to justify. This ensures that tribunals remain open to scrutiny by the parties and the broader legal community. By contrast, the current “may” formulation offers no guidance and leaves inconsistencies completely unexplained. Ultimately, a tribunal forced to articulate why it departs from a precedent is much more constrained than one that can simply ignore it. Furthermore, this presumption will operate smoothly within existing procedures. It will not impose an unrealistic ex officio research obligation on adjudicators. Instead, the duty to justify will arise naturally once a party invokes a relevant precedent during adversarial proceedings. Crucially, recognizing this presumption does not mean all precedents are created equal. By integrating the burden of justification with the weight criteria already established in Draft Conclusions 4 and 7 -ensuring, for instance, that a consistent line of ICJ jurisprudence carries more weight than a single ad hoc arbitral award- the Commission can accommodate institutional diversity without abandoning the fundamental requirement of reasoned justification.

Conclusion

Draft Conclusion 6 [7], in its current form, represents a missed opportunity. By reducing reliance on judicial precedent to a discretionary “may,” the Commission has chosen political caution over normative accuracy, failing to capture the presumption of consistency that systematically governs international adjudication, obscuring the burden of justification that tribunals impose upon themselves, and undermining the progressive development mandate that defines its constitutional role.

A revised approach should retain what the current draft gets right: judicial decisions do not constitute legally binding precedent. But it must go further. Rather than asking whether prior decisions may be followed, the inquiry should be framed in reverse with a negative formulation: decisions should be followed unless there are compelling reasons not to. Critically, the burden of articulating those reasons must rest on the departing tribunal, not on those urging consistency.

Ultimately, judicial precedents are not an optional tool in the adjudicator’s discretionary toolkit. They are a binding legal compass, not in the sense of formal hierarchy, but in the sense of methodological discipline. A tribunal may depart from the compass when the terrain demands it. But it must explain why. And that explanation must be compelling.