Subsequent Practice in Treaty Interpretation

Subsequent practice is a primary means of treaty interpretation codified in Article 31(3)(b) of the Vienna Convention on the Law of Treaties (VCLT) of 23 May 1969, which directs that there shall be taken into account, together with the context, "any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation." The doctrine predates the VCLT and reflects customary international law, as the International Court of Justice (ICJ) has repeatedly confirmed, including in the Kasikili/Sedudu Island case (Botswana v. Namibia, 1999). Its conceptual root lies in the principle that the parties who concluded an instrument are best placed to demonstrate, through how they actually apply it, what they intended its terms to mean. The International Law Commission (ILC) revisited the subject comprehensively, adopting in 2018 its Draft Conclusions on Subsequent Agreements and Subsequent Practice in Relation to the Interpretation of Treaties, which the UN General Assembly took note of in Resolution 73/202.

Procedurally, an interpreter invoking Article 31(3)(b) must establish two cumulative elements. First, there must be conduct "in the application of the treaty" — this can consist of acts, statements, or omissions by organs that engage the international responsibility of the party, such as the conduct of foreign ministries, customs authorities, courts, or legislatures acting in the treaty's domain. Second, that practice must "establish the agreement of the parties regarding its interpretation." The agreement need not be embodied in any formal instrument and need not be expressed; it can be inferred from concordant conduct or from the silence or acquiescence of parties that were aware of the practice and in a position to react. The interpreter weighs the practice's clarity, specificity, and the degree to which it is shared, then integrates it into the holistic interpretive operation of Article 31 — practice does not operate as a free-standing override but alongside ordinary meaning, context, and object and purpose.

A crucial distinction separates the two tiers of relevant conduct. Article 31(3)(b) requires the agreement of all the parties, though that agreement may be passive. By contrast, the ILC recognized in its 2018 Draft Conclusion 4(3) a broader category of subsequent practice as a supplementary means of interpretation under Article 32 — conduct by one or more parties that does not establish the agreement of all parties yet still illuminates meaning. This second tier carries less weight: it confirms or, where Article 31 yields ambiguous or absurd results, determines meaning, but cannot by itself fix an authentic interpretation. The threshold therefore matters greatly in litigation, because characterizing conduct as Article 31(3)(b) practice gives it determinative interpretive force, whereas Article 32 relegates it to a confirmatory or gap-filling role.

Named applications abound. In Kasikili/Sedudu Island, the ICJ examined whether Bechuanaland's and Caprivi's conduct after the 1890 Anglo-German Treaty constituted subsequent practice fixing the location of the "main channel" of the Chobe River, ultimately finding no qualifying agreement. The World Trade Organization Appellate Body, in Japan — Alcoholic Beverages II (WT/DS8, 1996), held that isolated acts do not suffice and required a "concordant, common and consistent" sequence establishing a discernible pattern. The European Court of Human Rights treats the practice of Council of Europe member states as evidence of evolving consensus, as in Soering v. United Kingdom (1989). Practice of conferences of the parties — for instance under the 1992 UN Framework Convention on Climate Change or CITES (1973) — frequently generates interpretive output that the ILC's 2018 Draft Conclusion 11 addresses with care, distinguishing binding decisions from those expressing mere agreement.

Subsequent practice must be distinguished from several adjacent concepts. It is not the same as a subsequent agreement under Article 31(3)(a), which is a single express agreement on interpretation reached after conclusion; practice, by contrast, is inferred from a course of conduct. It must also be distinguished from amendment or modification under Articles 39–41: interpretation operates within the bounds of the text's permissible meaning, whereas practice that contradicts the text effects a change in the treaty itself, which the ILC's Draft Conclusion 7(3) treats with caution and many states reject. Finally, subsequent practice differs from the travaux préparatoires under Article 32, which are pre-conclusion materials; the two pull in opposite temporal directions, the former emphasizing the treaty as a living instrument.

The doctrine generates genuine controversy at its edges. The principal debate concerns whether subsequent practice can modify a treaty in derogation of its terms — a proposition advanced by some arbitral tribunals and the ICJ's dictum in the Temple of Preah Vihear case (1962) but firmly resisted by states wary of obligations evolving without formal consent. The ILC deliberately refrained from endorsing modification by practice as an established rule. A further difficulty is attributing practice in multilateral treaties with many parties, where genuine concordance is rare and silence is ambiguous. Evolutive interpretation, by which terms acquire contemporary meaning, also rests partly on subsequent practice and raises sovereignty concerns, particularly for states that emphasize the original bargain.

For the working practitioner, subsequent practice is an indispensable interpretive instrument and a litigation hazard in equal measure. A desk officer drafting a diplomatic note, a customs ruling, or a vote in a treaty body should recognize that such conduct may later be marshaled as evidence of an authentic interpretation binding the state. Counsel preparing for proceedings before the ICJ, investment tribunals, or the WTO must document concordant practice to advance favorable readings and scrutinize an opponent's reliance on isolated or contested acts. Because the line between interpreting and amending a treaty is policed jealously, professionals should calibrate official conduct deliberately, knowing that consistent practice — and even silence in the face of another party's conduct — carries lasting legal consequence.