This paper offers two propositions. Firstly, WTO obligations are bilateral/reciprocal in nature. They are not of the multilateral, integral or erga omnes partes type (Section I). Secondly, WTO obligations are of a continuing nature. They cannot be equated to an expression of state consent limited to 15 April 1994, the date of conclusion of the Marrakesh Agreement (Section II).
The WTO Agreement is, obviously, a multilateral agreement. It will soon have 144 signatories. But what is the nature of WTO obligations? Are they of the bilateral (or reciprocal) type, in that WTO obligations can be reduced to a compilation of bilateral treaty relations, each of them detachable one from the other. Or are they of the multilateral (erga omnes partes or integral) type, in the sense that their binding effect is collective and the different relationships between WTO members cannot be separated into bilateral components? This is the first question that this paper tries to answer. Classifying WTO obligations in either of those two categories has major legal consequences: Firstly, for the permissibility of inter se modifications to the WTO Agreement. Secondly, in terms of the acceptability of suspension of WTO obligations in response to breach. Thirdly, for the rules on standing to bring a complaint before a WTO panel. Generally speaking, in case WTO obligations were of the multilateral type, inter se modifications to the WTO treaty and the suspension of WTO obligations as against a wrongdoing state would not be acceptable, whereas standing to bring a WTO complaint would be granted to all WTO members, irrespective of the breach. In contrast, if WTO obligations were seen as bilateral obligations, inter se modifications and suspension in response to breach would be permissible, whereas standing would be limited to only those WTO members at the other end of the (compilation of) bilateral relationship(s) allegedly breached.
The second issue addressed in this paper is whether it is feasible to put a fixed time-label on the WTO treaty. This is important, firstly, for the interpretation of WTO provisions. Should they be interpreted with reference to intentions and law contemporary with the formal conclusion of the WTO treaty, or in an evolutionary manner taking account of new developments at the time of application? The issue of timing is crucial also when WTO provisions are in conflict with other rules of international law and the question arises as to which of these two norms is later in time, i.e., the lex posterior which, in principle, prevails. For purposes of applying the lex posterior rule to WTO obligations, is 15 April 1994 really the critical date or should one rather consider the WTO treaty as a regulatory instrument which evolves and is continuously monitored, confirmed and re-concluded in the context of an international organisation? If the latter were true, a proposition defended here, in case of conflict between WTO provisions and other what I term "continuing" norms (such as, arguably, a number of multilateral environmental or human rights treaties), there would no longer be "successive treaties", but rather "parallel treaties". As a result, the lex posterior principle would not apply in the first place. Rather, the conflict ought then be resolved by means of other tools, in particular the lex specialis rule pursuant to which the more specific norm prevails over the more general one.
Both questions examined in this paper are crucial for the relationship and hierarchy between WTO law and other norms of international law. More profoundly, taking sides on these questions offers a prominent yardstick when setting out the WTO's object and underlying purpose as well as its place in the wider context of other, non-trade-related aspects of international relations.
© 2002 Joost Pauwelyn