Treaty Interpretation and Preparatory Work

In the introduction to our discussion, Richard also raised the question of: ‘Should preparatory work be investigated only to identify agreement among the negotiators or, particularly in the case of treaty provisions with a codifying aspect, to assess the state of international law leading up to the treaty and how this affected the negotiators?’

With respect to the first aspect of the question of whether preparatory work be investigated only to identify agreement among the negotiators:

The proposition of McNair that preparatory work reflects the common intentions of the contracting parties is troubling, especially when applied to multilateral treaties. Legal fictions can be helpful and necessary to move from theoretical impasses to practical solutions. It is probably the case that there ‘is [no] single legislative will behind international law’ (ILC Fragmentation Study Report 2006), even treaty law. Treaties such as the WTO covered agreements are negotiated by a heterogeneous group with particular and diverse interests. The larger this group the more unlikely it becomes that negotiators converged to a single, specific meaning of the treaty language. Preparatory work cannot, almost by definition, concern the common intention of all contracting parties. The common intention is only formed once the treaty language has been drafted and will develop over time and with the accession of new parties. Everything before the conclusion of the treaty expresses interests, values, objectives, and concerns of a single signatory or a group of contracting parties. Negotiations are as much about common intentions as about self-interests and strategy. Using preparatory work is a matter of relevance and availability but also of weighing and balancing.

A recent Appellate Body report illustrates this well. In US – Stainless Steel (Mexico), the Appellate Body found that there was no need to consider negotiating history. But it did, nevertheless, as one of the disputants extensively relied on the negotiating history of the Anti-Dumping Agreement and its predecessor Tokyo Round Anti-Dumping Code. The US contended that historical materials demonstrated that the permissibility of zeroing was left undecided by WTO Members. The silence in the agreement on this matter was not constructive ambiguity but was intentionally excluded from its scope. As a result, the Appellate Body should read the text of the agreement to include a prohibition on zeroing. The Appellate Body rejected this argument on several grounds. Ratione personae, the negotiating history only reflected the positions of a few negotiating parties. Rationae materiae, it was also not substantively relevant, as it ‘did not resolve the issue of whether the negotiators … intended to prohibit zeroing’ [para 131]. Ratione temporis, much of the invoked materials pertained to the interpretation and drafting of the plurilateral Tokyo Round Anti-Dumping Code. This was a separate treaty from the GATT 1947 and was terminated with the adoption of the 1995 Anti-Dumping Agreement [para 132]. They are two separate treaties, with different texts, membership, binding force, and mainly different temporal scopes. Their respective negotiating history cannot easily be interchanged. As a result, the Appellate Body found that the Anti-Dumping Code ‘is of little relevance for the interpretation of differently phrased or new provisions of the Anti-Dumping Agreement’ [para 132]. In response, the US strongly disagreed with this reasoning because ‘the Division’s conclusions regarding the negotiating history simply cannot be reconciled with that history’. Ratione temporis, the US argued that relevant negotiating history of treaties no longer in force, such as the Tokyo Round Anti-Dumping Code, could serve to interpret another treaty, such as the Anti-Dumping Agreement.

Any action and inaction of negotiators preceding the conclusion of the treaty is part of the history of text, but history is not applicable law. From the practice of judicial treaty interpretation, it seems that there exists a presumption that materials evidencing such conduct, statements, or lack thereof cannot inform the meaning of the treaty text. It can be rebutted but the burden of proof is considerable. The larger the group of negotiators and signatories of a treaty, the higher that burden. It involves proving that there was a common intention of that group that the treaty text or term has a particular meaning, with written evidence.

But this raises another question, namely, if this burden of proof is met how does the negotiating history proven to reflect a common intention among (not necessarily all) negotiating partners still qualify as a ‘supplementary means of interpretation’? Perhaps, it may be possible to qualify such type of negotiating history as context under Article 31(2) VCLT, or even as evidence of a special meaning under Article 31(4) VCLT? These questions illustrate the earlier point made about understanding the relationship between different principles of interpretation, and their relative value and meaning.

With respect to the second aspect of the question of whether preparatory work be investigated particularly in the case of treaty provisions with a codifying aspect, to assess the state of international law leading up to the treaty and how this affected the negotiators:

These are just some preliminary remarks: I believe this is a different matter and that here negotiating history serves a different purpose depending on the substance of, for example, statements of negotiators. Statements of negotiating partners may be relevant to establishing what the state of customary international law is at the time of negotiating the treaty and how to codify such customary norm. Such statements may inform whether the constitutive elements of opinio iuris and state practice are available, to the extent that they inform whether States belief they are bound by a norm as a customary international law.

But when it comes to subsequent interpretation of the treaty that codifies the customary norm, the same statements can only be relevant to inform the meaning of the treaty if they clarify the extent to which negotiating States intended to be bound by the same norm as treaty law – on the assumption that pure codification is possible. Each statement should thus be examined to determine their scope. To the extent that such statements only relate to the norm that will be binding as treaty law, these statements cannot be taken into account to examine the state of customary international law (though admittedly, this may lead to a paradox, especially in the event of a large multilateral treaty being negotiated). A single statement may thus simultaneously inform whether a state believes to be bound by a norm of customary international and what that norm is, and explain how a state perceives the relationship between the treaty norm to be negotiated and the pre-existing customary norm (codification, crystallization, contracting out, etc). A single statement may thus serve two functions and an examination of the substance must determine what part is relevant to informing the state of customary international law and what part serves the purpose of treaty interpretation.

Such statements might also become relevant in the event that a court or tribunal decides to interpret the treaty against the background of other rules of international law. To the extent that such statements inform that States intended to contract out of pre-existing customary international law, it signals that the interpretation of the treaty norm acknowledges this decision. In that sense, the interpretation of the treaty against the background of other rules of international law is not about ensuring ‘harmony’ but, instead, useful to respect the choice to contract out of customary international law, in whatever degree or extent.

As I mentioned, these are just a few preliminary remarks on a most intriguing question.