Article 31 of the VCLT and ‘textualism’

The initial post introducing our discussion of Treaty Interpretation raises various issues of interest to the study and practice of treaty interpretation. In this first reply, I focus on the first question relating to the tragedy of ‘insistent emphasis upon an impossible, conformity-imposing textuality’. My responses are informed by my study of treaty interpretation by the WTO Appellate Body.

I believe that the practice of international courts and tribunals has demonstrated that the general principle in  Article 31(1) VCLT does not emphasize ‘impossible, conformity-imposing textuality’. The text of Article 31(1) itself reads: ‘A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’. The text itself thus emphasizes that the starting point is the ‘contextualized ordinary meaning’, a notion that was actually already introduced before the ILC started its work on the law of treaties (eg, writings by Stone and Huber, Judge Anzilotti in Interpretation of the Convention of 1919 Concerning Employment of Women During the Night, the ICJ in Constitution of the Maritime Safety Committee of the Inter-Governmental Maritime Consultative Organization).

In the context of the WTO, the claim has been made that the Appellate Body has used Article 31 VCLT to the extent of, indeed, something alike ‘impossible, conformity-imposing textuality’. The Appellate Body’s apparently excessive attention to the words of the WTO treaty language might seem overdone, but it is correct nevertheless. However, such excessive attention to the words of the treaty does not justify the claim of strict textualism. Much has been said about its use of dictionaries, but in fact it has hardly ever solely relied on dictionary meanings, and already a substantive amount of ‘contextualization’ takes place when choosing the relevant dictionary meanings. The Appellate Body itself has explicitly recognized the limits of dictionary definitions and the need to contextualize them, through various methods. Of course, the fact that it consults dictionaries is not problematic, it is a common technique. But the Appellate Body was unique in the way that it openly acknowledged its use of dictionaries, in great detail. Its initial apparently excessive use of dictionaries and references to Article 31 VCLT was probably instigated by the need to assert its judicial function against the backdrop of a not fully developed institutional model and under-developed procedural rules in the DSU. To a large extent, this has been a successful strategy. In response, the explanation of its interpretative process has become less formalistic in recent cases.

If the general thread in an interpretative practice is to use the contextualized ordinary meaning as the keystone upon which the interpretation is built, this raises the question of how to define context? Article 31 VCLT gives the impression that it defines context in abstracto in an exhaustive manner. But when studying the interpretative practices and techniques used by one or more courts and tribunals, it becomes easily visible that a broader notion of contextualism applies. The ILC did not explain what ‘ordinary meaning’ entails. The Appellate Body’s jurisprudence suggests that the ‘ordinary’ meaning already implies a considerable amount of contextualization of the treaty language as part of what is commonly perceived as literal interpretation. Even if the VCLT drafters realized that context has a broader meaning and impact in treaty interpretation, the language in Article 31(2) mirrors the ultimate consensus that could be reached. The consensus among States on the final text of the VCLT was grounded on the shared understanding that consensualism was the basis for entering into treaty obligations. This helps to understand and appreciate how the VCLT reflects to what States ultimately could (only) agree. Also, at the time of the VCLT negotiations there was little practice of courts and tribunals like WTO panels and the Appellate Body, who continuously revisit the same treaties and within a particular institutional context. This might point to a certain disjuncture between the treaty language in Articles 31 to 33 VCLT and the interpretive practice of international courts and tribunals, without calling the text of Article 31 to 33 VCLT into question.

In essence, I believe that the usefulness and value of any principle of treaty interpretation, including the use negotiating history, is relative and depends on its relationship to other principles of treaty interpretation, whether codified or not.