AJIL Symposium: Szewczyk Comment on “The Travaux of the Travaux”

[Dr. Bart Szewczyk is an Associate in Law at Columbia Law]

This excellent article provides an invaluable contribution to our knowledge of the original understanding of Articles 31 and 32 of the Vienna Convention of the Law of the Treaties.  Its careful attention to the factual details, articulated in an elegant narrative, provides a vivid picture of the debates and decisions in Vienna.  And its comprehensive analysis of the historical record corrects any modern misperceptions as to what the drafters of the VCLT expected as the rules applicable to treaty interpretation.  The follow-on question, as the article notes, is “whether a regular and uncontested contrary practice has arisen—not just as a matter of what interpreters say, but of what they do—sufficient to undercut that original understanding.” (at 785).

Indeed, alongside the VCLT, there may exist several conventions (in the commonwealth, rather than international, sense of the term) governing interpretation for particular treaties, courts, or jurisdictions.  Such contemporary customs or practice may be as important in interpreting treaties as the rules of the VCLT.  For instance, judgments of the International Court of Justice are formally binding only between the parties to a particular case.  The ordinary meaning of the text of Article 59 of the Court’s Statute—the “decision of the Court has no binding force except between the parties and in respect of that particular case”—allows for no other interpretation.  Yet, any State would be highly remiss—and its advocates would border on malpractice—if it argued that an ICJ judgment on a specific legal question should be disregarded because it is not binding.  On the other hand, judicial decisions of other courts may be granted less weight in the ICJ, even though formally, they have equal status with ICJ judgments under Article 38(1)(d) of the ICJ Statute as “subsidiary means for the determination of rules of law.”

Or take Article 27 of the U.N. Charter:

Decisions of the Security Council on all other matters shall be made by an affirmative vote of nine members including the concurring votes of the permanent members.

In the Legal Consequences for States of the Continued Presence of South Africa in Namibia (South West Africa), the ICJ held that “concurring,” notwithstanding its apparent textual clarity and travaux to the contrary, included voluntary abstentions from voting.  The Court’s interpretation was based on the “consistent[] and uniform[]” practice of the Security Council.” (para. 22).  As for the U.N. Charter so too for the VCLT, subsequent practice can inform or even transform the original interpretation of a treaty provision.

The article recognizes this tension between the original understanding of the VCLT and subsequent interpretive practice of international courts.  It notes that 

[d]espite the VCLT drafters’ express rejection of PCIJ and ICJ precedent as a model for their treaty’s approach to travaux, the ICJ continued for decades its pre-VCLT practice of ‘privileging textual interpretation’ in the Vattellian sense and of ‘accord[ing] minimal space for the intention of the parties. (at 822).

The relative composition of such interpretive practice may very well differ from court to court.  And without an external normative criterion, it is impossible to choose between the uniformity of the VCLT and the plurality of practice.

For countries that have not ratified the VCLT, such as the United States and France, the dilemma dissipates as all rules of interpretation fall into custom or practice.  For other states, there may be conflict between faithful application of the original understanding of VCLT and comporting with contemporary interpretive practice.

It may have been this recognition of the potential friction between interpretive practice evolving over time and the text of the VCLT frozen in time that animated the U.S. position at the Vienna conference. And as a prediction of future trends in interpretation, rather than an assessment of the conference understandings, U.S. concerns regarding treaty interpretation seem to have been validated.  This was perhaps in part due to a self-fulfilling prophecy but also possibly due to an accurate forecast of how the VCLT text would be received and applied, notwithstanding of how it may have been written and intended.