GC IV Commentary Series – On Interpretive Methodology

by | Oct 7, 2026

Methodology

Editors’ note: The following series considers ICRC’s 2025 updated Commentary on the Fourth Geneva Convention. Rosa-Lena Lauterbach, Liisi Adamson, and Sean Watts’s introduction to the series is available here.

In its recently updated commentary of Geneva Convention IV, the International Committee of the Red Cross (ICRC) explains the interpretive methodology it has adopted as follows:

The updated Commentary applies the methodology for treaty interpretation set out in the 1969 Vienna Convention on the Law of Treaties (hereinafter ‘Vienna Convention’), in particular Articles 31-33. These rules are now generally considered to reflect customary international law. As such, they are applicable to the interpretation of the Geneva Conventions, even though the Conventions were concluded before the entry into force of the Vienna Convention (para. 98. Internal references omitted).

This methodology underpins the commentary’s frequent recourse to subsequent practice as an aid to interpretation, including its use as a supplementary means of interpretation where the practice in question falls short of establishing a general agreement among the parties as required under Article 31(3)(b) of the Vienna Convention on the Law of Treaties.

Such an interpretive approach reflects two propositions that have emerged in recent decades. First, the rules of treaty interpretation set forth in Articles 31-33 of the Vienna Convention are regarded as reflecting customary international law and, as such, are applicable to earlier treaties, including 1949 Geneva Conventions, which were adopted prior to the Vienna Convention’s entry into force in 1980. Second, the applicable rules of interpretation permit recourse to subsequent practice in the application of a treaty, even where that practice does not establish the agreement of the parties regarding the treaty’s interpretation, as a supplementary means of interpretation in the sense of Article 32 of the Vienna Convention.

These propositions are by no means unique to the ICRC’s commentaries. They are frequently assumed in legal scholarship and have increasingly found endorsement in official documents and institutional practice. However, their growing and often unqualified acceptance, without critical reflections, has long troubled me. In this post, I take the opportunity to explain my reservations concerning these two propositions and to suggest reasons for exercising caution when reading or relying on this commentary.

The Rule of Treaty Interpretation Applicable to the 1949 Geneva Conventions

The four Geneva Conventions of 1949 entered into force in 1950, three decades before the Vienna Convention on the Law of Treaties entered into force in 1980. As a technical matter, therefore, the interpretive rules codified in Articles 31-33 of the Vienna Convention are not directly applicable to the Geneva Conventions, given the principle of non-retroactivity set forth in Article 4 of the Vienna Convention.

The first of the interpretive propositions introduced above seeks to overcome this difficulty by drawing a distinction between the treaty rules embodied in the Vienna Convention and their equivalent in customary international law.  According to this view, the interpretive rules codified in Articles 31-33 did not originate with the Vienna Convention but rather reflected pre-existing customary law that was already binding when the Geneva Conventions were adopted and entered into force.

The International Court of Justice has affirmed the customary international law status of the Vienna Convention’s rules of treaty interpretation on numerous occasions, including in its 1994 judgment on the territorial dispute between Libya and Chad (para. 41). It follows that these treaty interpretation rules apply to present day treaties concluded among non-parties to the Convention on the basis of customary international law. As Don Greig has observed, however, “the tendency of the Court to employ the same justification in applying those provisions to treaties that predate the Convention is an entirely different matter” (p. 208).

What, if anything, constituted customary international law governing treaty interpretation prior to the Vienna Convention is a question that requires careful and historically informed inquiry rather than assumption. Before the Vienna Convention codified the law of treaties, treaty interpretation largely developed through judicial and arbitral practice, drawing on doctrines, case law, and maxims. Far from reflecting a settled body of customary rules, the pre-codification practice reveals considerable diversity in methodological approaches as reflected in extensive debates during the early work of the Institute of International Law in the 1950s (p. 360-406). As Jean-Marc Sorel and Valérie Boré Eveno have observed, “no method seems to have received the anointing of an infallible opinio juris” (p. 917).

Presumably mindful of this historical background, the judges of the International Court of Justice initially framed their observations about the customary law status of the Vienna Convention’s rules of treaty interpretation with a degree of caution. In its 1991 judgment concerning the arbitral award between Guinea-Bissau and Senegal, the Court did not assert that Articles 31 and 32 as a whole reflected customary international law. Instead, it merely observed that those provisions “may in many respects be considered as a codification of existing customary international law on the point” (emphasis added), specifically referring to the principle that treaty terms must be interpreted according to their natural and ordinary meaning (para. 48). Similarly, in the aforementioned 1994 judgment, the Court stated that, under customary international law, “a treaty must be interpreted in good faith in accordance with the ordinary meaning to be given to its terms in their context and in the light of its object and purpose” (para. 41). Significantly, the Court did not refer to extraneous elements of the treaty interpretation rule codified in Article 31(3), such as subsequent practice or other relevant rules of international law applicable between the parties (see also, Oil Platforms preliminary objections, para. 23 and Bosnian Genocide judgment, para. 160).

The Court never attempted to differentiate between interpretive rules that reflected pre-existing customary international law and those that did not prior to the Vienna Convention. It may well have been reasonable to find that certain core elements of the Vienna Convention’s general rule of treaty interpretation, such as good faith, ordinary meaning, context, and object and purpose, reflected pre-existing customary international law insofar as these canons were consistently invoked by States (p. 347). The other elements of Article 31, as well as the supplementary means under Article 32, would rather be considered to have crystallized into customary international law at the time of the codification (Sorel and Boré Eveno, p. 917-19; Le Bouthillier, p. 960-61).

However, the Court soon began making more sweeping statements, extending the Vienna Convention’s rules of treaty interpretation as a whole to treaties that long predated the Convention’s entry into force. In Kasikili/Sedudu Island, the Court relied on Article 31 to interpret the 1890 Anglo-German Agreement as the basis for boundary delimitation between Namibia and Botswana (para. 18), and it took a similar approach in Pulau Ligitan and Pulau Sipadan with respect to the 1891 Anglo-Dutch Convention (para. 37). In 2016, the Court applied the Vienna Convention’s interpretive rules to the 1948 Pact of Bogotá in affirming its jurisdiction over the Nicaragua-Colombia continental shelf dispute (p. 116, para. 33). Most recently, in its Advisory Opinion on the Right to Strike under ILO Convention No. 87, the Court invoked the same rules as a whole in construing the 1948 Freedom of Association and Right to Organise Convention (paras. 62-65).

Judge Oda, in his separate opinion in Kasikili/Sedudu Island, cautioned against excessive reliance on the Vienna Convention when interpreting a nineteenth-century treaty. While Richard Gardner views Judge Oda as “swimming against the tide” (p. 260), Don Greig expresses sympathy for his concern, noting that the Court failed to explain why Article 31 should be regarded as applicable in the first place (p. 211). In any event, the facts and documents examined as subsequent practice in these cases did not play a critical role, serving only to reinforce conclusions already reached on textual grounds (Kasikili/Sedudu Island judgment at para. 80; Pulau Litigan/Pulau Sipadan judgment, para. 80; Columbia v. Nicaragua Continental Shelf preliminary objections, para. 44). As discussed below, however, the Court recently went further, accepting subsequent practice by a majority of States as a supplementary means of interpretation on the same footing as the preparatory work of a convention (para. 102).

Recourse to subsequent practice as a means of interpreting pre-Vienna treaties on the basis of customary international law is particularly precarious. Under the Vienna Convention, treaty terms are open to an interpretation that departs from the meaning originally attached to them (conclusion 7, para. 1), notably where subsequent practice establishes agreement of the parties regarding their interpretation (Vienna Convention, art. 31(3)(b)). Whether such an approach formed part of the customary law of treaty interpretation when the 1949 Geneva Conventions were concluded is dubious. In U.S. Nationals in Morocco, the International Court of Justice held that treaty terms must be interpreted according to their meaning at the time of the treaty’s conclusion (p. 189). In his separate opinion in Minquiers and Ecrehos, Judge Levi Carneiro also observed that “an instrument must not be appraised in the light of concepts that are not contemporaneous with it” (p. 91). Consistent with this understanding, Sir Gerald Fitzmaurice identified the principle of contemporaneity as one of the six major principles governing treaty interpretation prevailing at that time (p. 225-27; also cited by Sir Humphrey Waldock in his Third Report on the Law of Treaties submitted to the International Law Commission, p. 55-56).

The principle of subsequent practice, another of the major principles identified by Sir Gerald, was rather understood as “a reliable guide” to the parties’ intentions, serving merely to confirm or elucidate the natural and ordinary meaning of a treaty term rather than to modify or replace that meaning (p. 223-24). This understanding was articulated clearly by the Permanent Court of International Justice, the predecessor of the International Court of Justice, when it observed that the facts subsequent to the conclusion of a treaty “can only concern the Court in so far as they are calculated to throw light on the intention of the Parties at the time of the conclusion of that Treaty” (para. 63). Indeed, the idea that subsequent practice might informally alter a treaty’s interpretation proved highly controversial during the negotiations of the Vienna Convention, with various proposals recognizing such informal modification ultimately failing to receive sufficient support for inclusion in the final text (p. 41-47).

Although explicit references to contemporaneity were ultimately dropped while subsequent practice remained in the final text, this does not necessarily mean that the commission abandoned a contemporaneous reading of treaties in favor of an evolutive approach. The principle of contemporaneity was omitted because it was considered to be inherent in the determination of a term’s natural and ordinary meaning (p. 56, para. 15), which is ordinarily ascertained through interpretation in good faith (p. 222, para. 16).

An exception is admitted where a treaty term was “intended to follow the evolution of the law and to correspond with the meaning attached to the expression by the law in force at any given time” (p. 32, para. 77; see also Navigational and Related Rights judgment, p.  242, para. 64; The Iron Rhine Railway arbitration, paras. 79-80). In such cases, the resulting evolutive interpretation rests on the common intention of contracting parties, rather than departing from it based on any subsequent agreement or practice that may have developed later (p. 76-77).

The Role of Subsequent Practice as a Supplementary Means of Interpretation

Assuming that the Vienna Convention’s rules of treaty interpretation, as reflective of customary international law, apply to the 1949 Geneva Conventions, the next question is whether recourse may be had to subsequent practice as a supplementary means of interpretation. As noted above, subsequent practice forms part of the general rule of treaty interpretation under Article 31(3)(b) of the Vienna Convention but only insofar as it establishes the agreement of the parties regarding the interpretation of a treaty. However, where the relevant State practice is fragmented, inconsistent, or otherwise insufficient to demonstrate a common understanding among the parties, the question then posed by the second proposition is whether such practice may nevertheless retain interpretive value as a supplementary means of interpretation under Article 32 of the Vienna Convention.

According to Article 32 of the Vienna Convention,

Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:

(a) leaves the meaning ambiguous or obscure; or

(b) leads to a result which is manifestly absurd or unreasonable.

The language of the text (“including”) makes clear that it does not provide an exhaustive list of supplementary means but rather refers to the preparatory work of a treaty and the circumstances of its conclusion merely as examples of the supplementary means “most often used” (p. 202, para. 50). This deliberate drafting choice affords interpreters wide discretion to draw upon additional means of interpretation as they are deemed appropriate in a particular case (p. 969, para. 24).

However, this latitude has generated divergent views regarding other materials that may qualify as supplementary means of interpretation. Anthony Aust, for example, points to various legal and linguistic techniques, such as a contrario and ejusdem generis (p. 220-21). Oliver Dörr, on the other hand, understands supplementary means as encompassing substantive materials that may assist interpretation, such as documents closely connected to the treaty’s drafting and the preparatory work of treaties that are identical or similar to the one under consideration (p. 626-27). Ulf Linderfalk goes further by arguing that elements of a treaty’s context may, in certain circumstances, be relied upon as supplementary means of interpretation, although he does not explain how this approach relates to the extraneous elements provided in Article 31(3) (p. 259-69, 272). While a few other scholars have also referred to the use of subsequent practice in this way (p. 981, n. 146), there appears to be little theoretical justification for such a “backdoor” approach, already rejected under the general rule of treaty interpretation.

Nevertheless, fifty years after its work on the Vienna Convention, the International Law Commission took a significant step toward recognizing subsequent practice as a supplementary means of interpretation in its work on subsequent agreements and subsequent practice in relation to the interpretation of treaties (conclusion 4, para. 3). The commission built on a single passing remark made during the debate on the role of subsequent practice in the course of the Vienna Convention’s preparatory work. In his capacity as Special Rapporteur, Sir Humphrey Waldock referred only in passing to the possibility that the practice of individual States could be taken into account as a supplementary means, and did so only by way of contrast to subsequent practice establishing the common understanding of all the parties as an authentic means of interpretation (p. 204, para. 13). However, as Judge Tomka noted in his dissent to the recent advisory opinion on the right to strike, this reference did not survive the commission’s subsequent deliberations (paras. 33-34) and was omitted from its final 1966 report (p. 221-22, para. 15).

Drawing on this single passing remark, the Commission sought to justify recourse to a broader range of subsequent practice by citing various decisions of international courts and other adjudicatory bodies. Yet these decisions fail to demonstrate any consistent endorsement of such an approach and, at best, offer only tenuous support, without specifically recognizing subsequent practice in the context of Article 32 (see p. 33-36). Indeed, where subsequent practice falling short of the Article 31(3)(b) standard has been considered, its role has been confined to ascertaining the common intention of the parties at the time the treaty was concluded.

Thus, in European Communities—Custom Classification of Frozen Boneless Chicken Cuts, the Appellate Body of the World Trade Organization stated, “it is possible that documents published, events occurring, or practice followed subsequent to the conclusion of the treaty may give an indication of what were, and what were not, the ‘common intention of the parties’ at the time of the conclusion” (para. 305; emphasis in original).

However, in its recent advisory opinion on the right to strike, the International Court of Justice has gone a step further by accepting subsequent practice by a majority of States, despite contrary views expressed by a number of States, to confirm the finding reached on the basis of Article 31 of the Vienna Convention, rather than the common intention of contracting parties at the time of a treaty’s conclusion (paras. 89, 102, 112-15). As Judge Xue noted, the Court’s approach “unduly expands the applicable scope of Article 32” (para. 64), introducing a kind of majoritarianism in treaty interpretation, without having regard to a critical aspect of the long-standing disagreement.

Given that the use of subsequent practice as a supplementary means was barely discussed, let alone property theorized, during the drafting of the Vienna Convention, it appears difficult to justify extending such an approach to the interpretation of the 1949 Geneva Conventions as a matter of customary international law. Even if it were to be accepted, the interpretive value of such practice would have to be restricted to serving merely as an indication of the common intention of the parties at the time of the conclusion of the Geneva Conventions. Reliance on partial, inconsistent, or otherwise inconclusive subsequent practice to resolve ambiguity in treaty terms, or to modify the meaning originally attached to them, is inadmissible and may even amount to an exercise in bad-faith interpretation.

Concluding Observations

I must admit that my research on these two propositions remains preliminary and that further study is needed to determine the state of customary international law on treaty interpretation at the time the Geneva Conventions were adopted, including the permissibility of relying on partial subsequent practice as a supplementary means of interpretation. Nevertheless, the analysis presented above, together with the materials cited in support, would be sufficient to cast doubt on the methodological soundness of the approach employed by the ICRC in its recent update of the commentaries on the 1949 Geneva Conventions.

This methodological shortcoming is not entirely the ICRC’s fault. International courts and tribunals have repeatedly relied on these propositions to circumvent the problem of retroactivity and apply a familiar interpretive framework to earlier treaties. In effect, this approach offers a convenient solution to the interpretive challenges posed by pre-Vienna treaties, notwithstanding the express agreement that the Vienna Convention should not be applied retroactively.

However, the increasingly permissive approach adopted by judicial bodies risks eroding meaningful methodological constraints, granting interpreters considerable latitude to project subjective preferences or political agendas under the guise of legal technicality. Readers of the ICRC’s updated commentaries would therefore be well advised to treat them with caution and with due awareness of the potential biases lurking behind this contentious methodological approach.

***

Hitoshi Nasu is a Professor of Law in the Department of Law and Philosophy at the United States Military Academy.

The views expressed are those of the author, and do not necessarily reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense. 

Articles of War is a forum for professionals to share opinions and cultivate ideas. Articles of War does not screen articles to fit a particular editorial agenda, nor endorse or advocate material that is published.

 

 

 

 

 

 

Photo credit: Marine Corps Cpl. Matthew Williams