Those applying treaties and interpreting them must remember two salient points: (1) as international adjudicators or as members of the judiciary they must apply a treaty not on the basis of discretionary powers and a judicial hunch but on the basis of the Vienna Convention on the Law of Treaties (VCLT) and (2) lack of proper treaty interpretation can lead to a breach of that treaty by the Contracting State.
First, the primacy that the VCLT gives to the treaty language must be the starting point for all issues regarding treaty interpretation.
Second, the travaux préparatoires can aid a national court judge in ascertaining the treaty’s real object and purpose, in circumstances either where the treaty language is ambiguous, or perhaps equally as important, where the issue before the judge is not one where the treaty provides an immediate answer (thus necessitating a review of the travaux to find a way through to a reasonable interpretation and a result that most people can sensibly live with). The drafting history of a treaty can really elucidate – and give a wider meaning – as to the proper application of a treaty to the situation the judge faces.
When national court judges apply an instrument of international law
The VCLT was concluded in 1969 and in the time since its Articles of Interpretation have been declared international customary law. The latter is of particular relevance for States that are either not a Contracting State or have ratified the VCLT fairly late. Its articles of interpretation being of customary international law means adjudicators can rely on it for the purposes of the uniform interpretation of an instrument of international law.
Articles 31-33 are what I refer to as the ‘Vienna Rules on Treaty Interpretation’. They have been created by the drafters as an expression of both existing customs on treaty interpretation and desirable methods of treaty interpretation. The VCLT drafters preferred dominance of the text rather than a primary use of a teleological interpretation of the text. The latter would make the text subordinate to the purpose. This makes little sense as readers of treaties naturally look to the text first. Moreover, drafters of treaties are expected to say what they mean. Thus, the Vienna Rules are built first and foremost on a good faith-based interpretation of its text. That text, on the basis of good faith, must be read bearing in mind the purpose of that treaty and the remaining text of the entire treaty. I refer to this as the bicycle interpretation: a bicycle cannot function without both its wheels and its steering wheel; they are used at the same time. That is how text, context, and purpose are used. One does not take precedence over the other nor can they be used separately while ignoring the other.
Yet, the primary sources of the Vienna Rules, as codified in Article 31, would not do justice to the complexity of a treaty if the reader cannot take into account subsequent practice and interpretation along with other instruments of interpretation concluded by the Contracting States. It is the secondary sources described in Article 32 that have complicated matters a great deal. Yet, this source is today more important than ever. It is the drafting history, also known as the travaux préparatoires that is crucial for understanding treaties today. The drafting history of a treaty is relied upon either to confirm a meaning obtained on the basis of the primary sources or it is used if the use of the primary sources leads to an outcome that is unreasonable or ambiguous. In order to understand the use of all those sources, I developed the Snail Diagram. The challenge is how to determine whether a certain outcome would be unreasonable or ambiguous. In practice, most users of treaties keep the threshold for that determination low and resort to the drafting history.
Relying on the judicial hunch instead of Articles 31-33
Today, it is rather unclear as to whether a national court judge relies on proper treaty interpretation as prescribed by the VCLT or just discretionary power and sense of judgment. This is in part due to the fact that often treaties are implemented and thus become part of national law that is the subject to, for example, statutory rules of interpretation. The Vienna Convention itself does not resolve the conflict that ensues.
The reason for relying on the VCLT and the drafting history is clear: the drafting history of the treaty itself, not the parliamentary history of the national enactment, explain the why and how of a treaty provision and will enable the Contracting States through their national courts to comply with their obligations under international law.
For example, the 1958 New York Convention provides proper guidance on its scope: which awards may be enforced under the New York Convention? Article I of the treaty prescribes that foreign and non-domestic awards fall within its scope and so do awards that are rendered by permanent arbitral tribunals. The latter must be understood in the light of its Drafting History though. The delegates in 1958 perhaps faced different challenges than those the community faces today. Yet, the concerns were similar. In the 50s, some delegates at the United Nations working on the New York Convention, emphasized that some Eastern European countries motivated by the communist school of thought attempted to control the resolution of international disputes in international trade, not by subjecting those to their national courts but to sovereign tribunals with ‘permanent arbitral tribunal’ as its nomenclature. The tribunals were not independent, however, and the appointment of permanent members of those courts was in the hands of the State. Even though they were named arbitrators and even though they issued so-called awards, the delegates in 1958 and the drafters of the New York Convention wanted to make certain through the drafting of Article I(2) that such decisions would be excluded from the scope of the New York Convention (see also the discussion here).
In 1958, the delegates shared some concerns in this regard. Article I(2)is instructive not only for understanding the terms ‘award’ and ‘arbitration’ as discussed above, but also for appreciating the importance of party autonomy under the Convention: arbitrations that are of a mandatory nature do not fall under the Convention’s scope. (M. Paulsson, ‘The 1959 New York Convention in Action’, (Kluwer 2016), pp. 120-121.)
Today, proponents for a Multilateral Investment Court view that the New York Convention will apply. Is this an ipse dixit? If one were to predict, judges in the 161 Contracting States might rely on statutory interpretation; they might rely on precedents, they might rely on legal scholars, or they might rely on the treaty’s drafting history. Thus, a great amount of uncertainty exists as so whether semi-judicial decisions of a multilateral investment court are New York Convention awards. One does well to remember the instructions of the Vienna Rules and the intent of the drafters of the New York Convention: only decisions that truly resulted from international arbitration are admissible under the scheme of the New York Convention. The Vienna Rules and the drafting history do matter.
This post concludes our celebration of the 50th jubilee anniversary of the Vienna Convention on the Law of Treaties, to see the full series click here.
The ideas in this blog post are further elaborated in The Vienna Convention on the Law of Treaties in Investor-State Disputes: History, Evolution, and Future, edited by Esmé Shirlow & Kiran Nasir Gore (Kluwer, 2022).