ILC Report on “Fragmentation” is Out and Favors Close WTO-International Law Link

Those interested in how WTO law interacts with the rest of international law may know that in 2002 the UN International Law Commission (ILC) set up a Study Group to examine the “fragmentation” of international law. Well, the final report and conclusions of this Study Group are now published. In July-August of this year, the entire ILC “took note” of these conclusions and “commended” them to the UN General Assembly.

As Panels and the WTO Appellate Body have frequently referred to the ILC in their past decisions (remember the ILC Articles on State Responsibility and the ILC preparatory work on the Vienna Convention on the Law of Treaties), this report might find its way in Appellate Body case law. This is all the more likely as the report deals extensively with the WTO.

There are really two documents. First, a 250 page analytical study “finalized” by the Chairman of the ILC Study Group (Prof. Martti Koskenniemi of Finland). Second, a set of 42 conclusions collectively adopted by the Study Group and “taken note of” and “commended” by the entire ILC. The Study Group “emphasized that these conclusions had to be read in connection with the analytical study … on which they are based”.

For WTO purposes, several findings are interesting and all go in the direction of more, rather than less, international law in WTO dispute settlement.

First, the ILC confirms a distinction that some of us have long advocated, namely that between jurisdiction and applicable law, and explicitly applies it to the WTO:

“A limited jurisdiction does not, however, imply a limitation of the scope of the law applicable in the interpretation and application of those treaties … While the WTO Dispute Settlement Understanding limits the jurisdiction to claims which arise under the WTO covered agreements only, there is no explicit provision identifying the scope of applicable law”. (para. 45)

“Even as it is clear that the competence of WTO bodies is limited to consideration of claims under the covered agreements (and not, for example, under environmental or human rights treaties), when elucidating the content of the relevant rights and obligations, WTO bodies must situate those rights and obligations within the overall context of general international law (including the relevant environmental and human rights treaties)”. (para. 170)

Second, the ILC addresses a long-standing debate of whether international law is “in” unless contracted out from in the WTO, or whether international law is “in” only if explicitly incorporated by the WTO treaty. It concludes as follows:

“There seems, thus, little reason of principle to depart from the view that general international law supplements WTO law unless it has been specifically excluded and that so do other treaties which should, preferably, be read in harmony with the WTO covered treaties”. (para. 169)

“It is in the nature of ‘general law to apply generally’ - namely inasmuch as it has not been specifically excluded”. (para. 185)

“It is sometimes suggested that international tribunals or law-applying (treaty) bodies are not entitled to apply the law that goes “beyond” the four corners of the constituting instrument or that when arbitral bodies deliberate the award, they ought not to take into account rules or principles that are not incorporated in the treaty under dispute or the relevant compromis. But if … all international law exists in systemic relationship with other law, no such application can take place without situating the relevant jurisdiction endowing instrument in its normative environment. This means that although a tribunal may only have jurisdiction in regard to a particular instrument, it must always interpret and apply that instrument in its relationship to its normative environment - that is to say “other” international”. (para. 423, emphasis in the original)

Third, the ILC accepts the legality of inter se agreements restricting trade as between some WTO members only, in deviation of WTO rules, for as long as third party rights are respected:

“the GATT treaty contains no rules that would apply should two or more members wish to conclude an inter se agreement to restrict trade between themselves. In the absence of such rules, there appears to be nothing to prevent members from concluding an inter se agreement to the effect that in their dealings with each other they will not invoke, say, articles III and XI of the GATT with respect to what they feel to be justified trade restrictions. Such an agreement would affect the rights and obligations of the other members of WTO but as it would do so beneficially, the condition set in article 41 [Vienna Convention] would be satisfied”. (para. 306)

Fourth, when it comes to treaty interpretation that takes account of “any relevant rules of international law applicable in the relations between the parties” (as directed in Article 31.3(c) of the Vienna Convention), the ILC openly disagrees with the Panel on EC – Biotech. In what is quite a remarkable critique (remember, this is only an interim report, supposed to be confidential), the ILC does not think that such outside treaties need to be adopted by, or express the common intentions of, all WTO members:

“A better solution is to permit reference to another treaty provided that the parties in dispute are also parties to that other treaty. Although this creates the possibility of eventually divergent interpretations (depending on which States parties are also parties to the dispute), that would simply reflect the need to respect (inherently divergent) party will as elucidated by reference to those other treaties as well as the bilateralist character of most treaties underpinned by the practices regarding reservations, inter se modification and successive treaties, for example”. (para. 472)

The ILC Conclusions (nr. 22) are more ambiguous:

“Article 31 (3) (c) also requires the interpreter to consider other treaty-based rules so as to arrive at a consistent meaning. Such other rules are of particular relevance where parties to the treaty under interpretation are also parties to the other treaty, where the treaty rule has passed into or expresses customary international law or where they provide evidence of the common understanding of the parties as to the object and purpose of the treaty under interpretation or as to the meaning of a particular term”. (emphasis added)

Finally, the ILC defines “conflict” between norms broadly. Unlike some WTO decisions (Indonesia – Autos and, to some extent, US – Cotton), it holds that the criterion of “same subject matter” leads to a reductio ad absurdum (para. 22). To hold that there is no conflict between, say, Article III and the SCM Agreement or between a trade and an environmental agreement because the two do not exactly deal with the same subject matter is unsatisfactory. Rather, what the ILC calls for is to see, more specifically, whether the two rules in question “seem to point to different directions in their application by a party” (para. 23). The ILC “adopts a wide notion of conflict as a situation where two rules or principles suggest different ways of dealing with a problem” (para. 25).

ILC Conclusion Nr. 2 defines conflict as follows:

“Relationships of conflict. This is the case where two norms that are both valid and applicable point to incompatible decisions so that a choice must be made between them. The basic rules concerning the resolution of normative conflicts are to be found in the VCLT”.

Comments

[jalal Alavi]

I have seen the draft of the Report. One of the main resources of the Study Group is the famous book of Joost Pauwelyn (Conflict of Norms in Public Internationl Law). Congratulations to Professor Pauwelyn and thanks to him for this thought provoking post.

[Joel P. Trachtman]

It will take some time to understand this remarkable document, and the ILC is to be commended for taking on this important and difficult subject. Indeed, it appears that Joost's work has been influential.

In the meantime, let me ask a couple of questions. Joost, do you take para 423 to say that a tribunal should ignore an explicit and explicitly exclusive limited mandate and apply the instrument it is authorized to apply "in relationship to its normative environment." And if I may ask, what do you think they mean by that language? Does it simply mean that they are to apply other international law that is explicitly excluded from their mandate?

One other thing. Do you agree with the committee's assumption (para. 306)that an inter se modification of GATT (why do they only refer to GATT?) would always beneficially affect other states? I immediately thought of the semiconductor case where the EC complained about an inter se modification between the U.S. and Japan.

[Joel P. Trachtman]

Joost, I do not believe it is correct to frame either the analytical study or the conclusions, prepared by the study group, as the work of the ILC. The ILC has not adopted these documents, and has not adopted principles in response to these documents, and so it has not taken responsibility for any of the statements made in them. This is an important distinction, and I am surprised that you characterized these statements as statements of the ILC. At this moment the statements that you identify as statements of the ILC are merely statements contained in the analytical report and are not even subscribed by the entire study group. The analytical report states that "Nevertheless, the contents of this report - including any opinions therein - remain the sole responsibility of its author." So, what you cite as the words of the ILC are really the words of specific individuals, and are entitled to no deference beyond that which we would accord the work of other colleagues.

At its 2911th meeting, on 9 August 2006, the Commission adopted the following resolution by acclamation:
“The International Law Commission, Having taken note of report and the conclusions of the Study Group on Fragmentation of International Law: Difficulties arising from the diversification and expansion of International Law,
Expresses to the Study Group and its Chairman, Mr. Martti Koskenniemmi, its
deep appreciation and warm congratulations for the outstanding ontribution made in the preparation of the report on fragmentation of international law and for the results achieved in the elaboration conclusions, as well as the accompanying study on Fragmentation of International Law: Difficulties arising from the diversification and
expansion of International Law finalized by the Chairman.”

[Joost Pauwelyn]

On substance: para. 423, especially when read together with paras. 45 ff. and paras. 166 ff. is, in my view, pretty clear on the question of applicable law: silence means international law is "in"; to be "out" it needs to be explicitly excluded; according to the ILC Study, in WTO dispute settlement, there is no such explicit exclusion (para. 45) (the exclusion applies only in respect of the question of "jurisdiction"); hence, other international law is "in" unless deviated from in the WTO treaty. Note that reference is made not only to general international law but also to non-WTO treaties (para. 169-170).

On para. 306, you are right, Joel. I personally agree with the principle that inter se deviations should be possible (notwithstanding WTO provisions on amendment). But they must be subject to (1) explicit prohibitions in the WTO treaty, and (2) respect for third party rights. On (1), I think the ILC Study may have overlooked the explicit prohibition on so-called voluntary export restrictions (VERs) in Article 11 of the Safeguards Agreement (does Art. 11 make, for example, the recent US-Canada lumber settlement illegal?). This may be why the report refers to GATT only ... On (2), yes, one better looks at this case-by-case.

On procedure: I really think that I was careful enough to explain the difference between (1) the analytical study, (2) the Study Group conclusions, and (3) the affirmation of these documents by the ILC as a whole.

Again, the analytical study is a "Report of the Study Group ... Finalized by Martti Koskenniemi", (2) the conclusions are collective conclusions by the entire Study Group, (3) those conclusions were "taken note of" by the entire ILC (2902th meeting) and "commended" by the entire ILC to the UN General Assembly (2912th meeting), and explicitly stated to be read "in connection with the analytical study ... on which they are based" (see paras. 237-239 of the ILC's Report on its 58th Session now available on the ILC website).

So, yes, the ILC Study was not formally adopted by the ILC as a whole, but there was some form of approval.

[Joel P. Trachtman]

Thanks, Joost.

On para 423, you did not answer the question that I asked. I would still value your response.

The conclusions, which were approved by the entire study group, were far more modest than the analytical report. Most of your references were to the analytical report, which was prepared on specific individuals' separate responsibilities. I don't know what it means to be finalized by the chairman, and I don't think the word "finalized" was selected in order to clarify responsibility. In any event, the study group as a whole has no special claim to authority. Their analytical report will no doubt, however, merit extensive review and critique.

To ascribe the statements of the analytical report and the conclusions to the ILC is simply an error, and if I made such an error I would simply retract it. Of course others can decide for themselves whether they agree with you that it is correct to say "the ILC confirms, accepts, addresses, defines" when it did no such things. "Commending to the attention" of the General Assembly is not by any means "approval" or "adoption." The statements you ascribe to the ILC are simply not its statements. And most of the statements you describe as statements of the ILC are not even statements of the study group, as they are not part of the conclusions!

[Joost Pauwelyn]

Yes, right, no formal adoption by the ILC (X3). So, readers, pls add "Analytical Report" after "ILC" in my original post. I thought that this was clear enough from my original intro. If not, I formally and humbly retract, apologize and repent.

On your para. 423 question, Joel: I thought you were asking what happens if there is explicit exclusion or contracting out from a tribunal's applicable law (mandate?). If so, then, yes, the law contracted out cannot be applied.

[Joel P. Trachtman]

Thanks. And no informal adoption or explicit or implicit approval by the ILC either. And it is not an ILC Analytical Report, but an amalgam of individual reports prepared under the auspices of the ILC. I am not trying to be difficult, but I want to make sure the true responsibilities do not get lost, and I am concerned that your original post lost them.

So I think we basically agree on the response to my para 423 question. But I would not characterize the exclusion from dispute settlement as "contracting out." This confuses applicable law in dispute settlement with the broader question of the law that applies to a particular matter. But my follow-up is this. Exclusion of a particular law from WTO dispute settlement does not make that law inapplicable to the conduct of persons bound thereby.

If the treaty that constitutes the dispute settlement tribunal explicitly excludes consideration of other international law, I think we agree that the tribunal cannot consider it. But for me the next step is one of interpretation: what if the constitutive treaty almost explicitly excludes consideration of other international law? Where I think you and I probably disagree, and where the author of that part of the analytical report (but not the study group as a whole, and certainly not the ILC) agrees with you, is that there is a presumption of plenary authority of a tribunal to apply all international law. You seem to agree with me that explicit treaty language could rebut this presumption, but I am not sure the analytical report agrees with us. But more importantly, I have been presented with no basis for such a presumption, and simply revert to the normal rules of treaty interpretation as to what is intended. (I think Lorand is engaged in a research project to determine whether such a presumption has any basis, and I look forward to seeing his results.) These rules of treaty interpretation do not normally require an explicit exclusion, but would be open to an implicit exclusion. As you know, Joost, I believe there is a basis for finding an implicit exclusion in the mandate of WTO dispute settlement.

[Steve Charnovitz]

This is a fascinating report that merits close study. Today, I simply want to make two points. First, the obvious one, that the Report reflects a great deal of Prof. Paulwelyn's work, and he is to be congratulated on that achievement. Second, one the causes of fragmentation in international law is the lack of transparency in international institutions. For a long time, the GATT/WTO system was a champion in operating without public participation and scrutiny, but that has gotten better in recent years. For example, the Continued Hormones SCOO panel held public hearings. (Even Canada went along with that!) The performance of the ILC, however, does not appear to be on a similar path for openness. As far as I know, the ILC did not seek public comments on this overall project or on its draft Report, and did not hold a public hearing on it. Doing so would have provided a possibility to vet the big issues being discussed, as well as the small details, for example, the embarrassing reference in Footnote 15 to the AB report on Korea Government Procurement. As Prof. Trachtman has noted, and I hope I am taking him in context, there is something troubling about a process where an important study like this gets "finalized," "taken note of," and "commended" without any careful review by the ILC. The General Assembly presumably is not going to do anything with the Report (for example, like send it to the WTO), so what has occurred would seem to be the final administrative action. I believe that the international law community should get tougher on the ILC to upgrade its practices for the 21st century.

[Joel P. Trachtman]

Thanks, Steve. My point was that as long as we don't confuse the authorship, I don't mind that studies are undertaken under the auspices of the ILC. "Finalized," "taken note of" and "commended to the attention of" are words of denial of responsibility, and neither the study nor the conclusions should be bootstrapped into authoritative statements of the ILC. I agree with you that to come up with agreed principles or articles in this area would take a careful and transparent process. I applaud the ILC for beginning work on this topic, and do not mind that the members of its study group have produced something provocative. Sometimes that is the best way to begin a dialogue. But mislabelling the study and conclusions as the product of the ILC, or as approved by the ILC, would have a tendency to end the dialogue.

[Cesare Romano]

Very interesting discussion. However, I would not obsess too much about the actual value of this work. If it is persuasive enough it will have its impact felt, regardless of whether it is ILS, a sub-group report, or the speculation of Koskenniemi alone. The nature of international law (and the arcana of its making) is such that formal categories are often eschewed.

Having said that, I regret that the ILC decided early on to focus on international law and its fragmentation and not the institutional and substantive problems. The Commission brushed aside the thorny issue by noting that: “At the outset, the Commission recognized that fragmentation raises both institutional and substantive problems. The former have to do with the jurisdiction and competence of various institutions applying international legal rules and their hierarchical relations inter se. The Commission has decided to leave this question aside. The issue of institutional competencies is best dealt with by the institutions themselves”. (Report, para 8).
Alas, the institutions in question (particularly judicial institutions), time and again, have shown to be utterly incapable of resolving themselves the problem of competing institutional competencies. They are inherently, and understandably, parochial in defending their own jurisdictional turf.
Besides, most courts, when ruling on separate issues of the same dispute reach correct conclusions, but the answer is correct only as far as that particular aspect of the dispute is concerned and only if seen from the perspective of that particular legal regime. Another court using different laws (or even the same laws) can reach different (and again entirely correct) conclusions. For the most recent example of this dilemma, see the judgment by the ECJ in the saga of the MOX dispute (May 30, 2006), and stay tuned for the reply by the LOS Convention arbitral tribunal.
One can only hope that, having explored half of the planet, the ILC will still find the time and energy to explore the other half. If it does not, this report will unlikely produce the practical results it strives to. The report, as such, does not help decrease the chances that the WTO AB might step into the turf of, say, the ICJ, or vice-versa (or the ECJ into that of a Law of the Sea Convention Arbitral Tribunal, etc.). It just hopes that, if the underlying rules are clear enough, (reasonable) courts will reach the same conclusions, hence eliminating conflict at the root. To me, unfortunately, that is rather wishful thinking.

[James Mathis]

Thanks to Joost for posting this summation. On para 306, if the inter se restriction is incorporated in an RTA, then there is a ´rule´, that being GATT Art. XXIV.8. Here members to a customs union or free-trade area are required eliminate duties and other restrictive regulations of commerce (ORRC) with respect to substantially all trade (SAT). Since a failure to eliminate ORRC by regional members would not likely abridge any third party rights directly (in the sense of a GATT Article violation), it would seem that this particular ´rule´ has been imposed to serve a systemic interest of the larger body of members (to reduce proliferation). If so, the internal trade requirement of Article XXIV (and Article V GATS) appear to fall within the scope of VCLT 41.1(a) rather than 41.1(b), at least to the extent that VCLT 41 applies at all to this situation.