International Economic Law and Policy Blog

Multiple Permissible Interpretations under Chevron and AD Agreement Article 17.6(ii)

Here's an excerpt from the majority opinion in the recent Loper Bright U.S. Supreme Court decision that puts an end to Chevron deference, in which the majority says that it does not make sense to talk about "permissible" interpretations because there is always a "single, best" interpretation:

Courts, after all, routinely confront statutory ambiguities in cases having nothing to do with Chevron—cases that do not involve agency interpretations or delegations of authority. Of course, when faced with a statutory ambiguity in such a case, the ambiguity is not a delegation to anybody, and a court is not somehow relieved of its obligation to independently interpret the statute. Courts in that situation do not throw up their hands because “Congress’s instructions have” supposedly “run out,” leaving a statutory “gap.” Post, at 2 (opinion of KAGAN, J.). Courts instead understand that such statutes, no matter how impenetrable, do—in fact, must—have a single, best meaning. That is the whole point of having written statutes; “every statute’s meaning is fixed at the time of enactment.” Wisconsin Central Ltd. v. United States, 585 U. S. 274, 284 (2018) (emphasis deleted). So instead of declaring a particular party’s reading “permissible” in such a case, courts use every tool at their disposal to determine the best reading of the statute and resolve the ambiguity.

In an agency case as in any other, though, even if some judges might (or might not) consider the statute ambiguous, there is a best reading all the same—“the reading the court would have reached” if no agency were involved. Chevron, 467 U. S., at 843, n. 11. It therefore makes no sense to speak of a “permissible” interpretation that is not the one the court, after applying all relevant interpretive tools, concludes is best. In the business of statutory interpretation, if it is not the best, it is not permissible.

In the WTO law context, we are all familiar with the idea of permissible interpretations as set out in AD Agreement Article 17.6(ii):

In examining the matter referred to in paragraph 5:

...

(ii) the panel shall interpret the relevant provisions of the Agreement in accordance with customary rules of interpretation of public international law. Where the panel finds that a relevant provision of the Agreement admits of more than one permissible interpretation, the panel shall find the authorities' measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations.

In Colombia - Frozen Fries, the MPIA arbitrators, when discussing Article 17.6(ii), emphasized the possibility of there being multiple permissible interpretations, rather than having to reach a single "final" or "correct" interpretation in every case:

4.13. Reading these provisions together, we aim to integrate the elements of interpretation under Article 17.6(ii). As a result, we will begin by asking ourselves whether Colombia's proposed interpretation of the phrase "where appropriate" in Article 5.2(iii) – reflected in MINCIT's decision to initiate its investigation, in part, on the basis of third-country sales prices – is a "permissible" one. As a yardstick for "permissibility", the first sentence of Article 17.6(ii) refers us to the customary rules of treaty interpretation. However, we will not engage in our own, de novo interpretation of the terms "where appropriate" so as to arrive at what we consider to be the "final" or "correct" application of Articles 31 and 32 of the Vienna Convention. Instead, we will ask whether a treaty interpreter, using the method for treaty interpretation set out in the Vienna Convention – that is, an interpretation "in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose" and, where appropriate, "supplementary means of interpretation" – could have reached Colombia's interpretation. And this even though we, as de novo treaty interpreters, might have reached a different conclusion.

4.14. Our approach assumes, as the second sentence does, that different treaty interpreters applying the same tools of the Vienna Convention may, in good faith and with solid arguments in support, reach different conclusions on the "correct" interpretation of a treaty provision.43 This may be particularly true for the Anti-Dumping Agreement, which was drafted with the understanding that investigating authorities employ different methodologies and approaches. Treaty interpretation is not an exact science and applying the Vienna Convention's method does not magically and inevitably lead to a single result. In most cases, treaty interpretation involves weighing, balancing, and choice.

4.15. Thus, the ultimate question for us when testing a proposed interpretation is to draw a line beyond which an interpretation is no longer "permissible" under the Vienna Convention method for treaty interpretation. Dictionary meanings support the idea that the search for "permissible" interpretations differs from an attempt to find one's own – "final" and "correct" – interpretation.44 Rather, the question is whether someone else's interpretation is "permitted", "allowable", "acceptable", or "admissible" as an outcome resulting from a proper application of the interpretative process called for under the Vienna Convention. Obviously, not just any interpretation put forward by an authority can be accepted as "permissible". The interpretative process under the Vienna Convention sets out an outer range beyond which meanings cannot be accepted. Just as permissible interpretations cannot be limited to a single "final" and "correct" answer as determined by a given tribunal, not all interpretations have the required degree of solidness or analytical support for them to be given deference as "permissible" within the bounds of the Vienna Convention method for treaty interpretation.

The approach set out by the MPIA above can be distinguished from earlier WTO decisions that seemed wary of a standard that was open to multiple permissible interpretations.

(Jackson and Croley laid out the connection between Chevron and the Article 17.6(ii) standard here).

I'm not sure I understand the reluctance to accept that statutes or treaties may be vague or ambiguous enough that there are multiple permissible interpretations. In some instances, the choice between one interpretation or another is a very close call. If a court is considering multiple interpretations, and one is slightly better than another (say, 55% to 45%, to put concrete figures on this), I'm not convinced that declaring the 55% interpretation to be the "best" one, which therefore must be followed, is always the right approach. At some degree of uncertainty about which interpretation is best, I would have no problem saying, in the circumstance where an agency is crafting an implementing regulation based on its interpretation of the statute, that there are multiple permissible interpretations.

Of course, if you do adopt a "best" interpretation approach, the statute or treaty drafters can step in to correct the interpretation if they don't like how the court came out. But in practice, that's often difficult for practical political reasons, and the court's interpretation is likely to govern. It's worth noting in this context that there are policy implications from these interpretive choices, and we should think about which government actors are best placed to take responsibility for these policy decisions.

I suspect that many people will consider this suggestion to be a bit off the wall, but to deal with the political and policy-making concerns, I would be interested in thinking about a process through which courts could say to the legislators/treaty drafters something to the effect of: "We're not sure what you meant with regard to how this ambiguous provision should be interpreted in the context at hand. Could you please clarify?" I haven't given any deep thought to how this might work in practice, but perhaps something along the lines of this, this, this, or this in the U.S. domestic context.

The MPIA: What’s New? (Part I)

On 21 December 2022, more than two and a half years after its creation, the World Trade Organization’s (WTO’s) Multi-Party Interim Appeal Arbitration Arrangement (MPIA) finally produced its first appellate award.[1]

The inaugural MPIA award was issued in a dispute filed by the EU in November 2019 against anti-dumping duties imposed by Colombia on frozen fries from Belgium, Germany and The Netherlands. The three MPIA appeal arbitrators reversed one panel finding of violation, and confirmed three other panel findings that were appealed by Colombia.[2] 

In this series of three posts, the first post explains what the MPIA is; a second post will ask whether the MPIA worked; a third and final post will describe some of the novelties that can be found in the first MPIA procedure.

Part I: What is the MPIA?

The MPIA was created in April 2020, by a sub-set of WTO Members, originally 19, today 25, out of a total of 136 (counting the EU and its 27 member states as one).[3] The MPIA was set up as an interim response to the demise of the WTO’s Appellate Body in December 2019, brought about by a US block on appointments to the Appellate Body.

Pending negotiations to fix the situation relating to the Appellate Body, MPIA participants wanted, above all, to preserve the functioning of WTO dispute settlement. More specifically, the main goal was to prevent “appeals into the void”, that is, blocking of the dispute resolution process by appealing a panel report to an Appellate Body that no longer functions. Instead, MPIA participants committed ex ante “not [to] pursue appeals under Articles 16.4 and 17 of the DSU [Dispute Settlement Understanding]”[4] and to use the alternative of (appeal) arbitration, explicitly foreseen in Article 25 of the DSU, to complete possible appeals in disputes between MPIA participants.[5]

The MPIA’s primary objective is, therefore, to preserve the system’s “binding character and two levels of adjudication”[6]: a panel stage, which remains the same as before (pre-2019), and an appellate stage, now conducted through Article 25 arbitration instead of the old Appellate Body.

Importantly, as the MPIA is, in effect, a form of arbitration implemented within a regular WTO dispute, albeit only at the appeal stage (see Part II), the MPIA is not a plurilateral agreement. It is nested within the multilateral WTO and explicitly foreseen and allowed for under DSU Article 25 (entitled “Arbitration”).  Illustrating the multilateral character of the MPIA, MPIA arbitrators hearing an appeal are paid out of the regular WTO dispute settlement budget and assisted by regular Secretariat staff.  

The second post in this series will examine whether the MPIA, in its first dispute, has achieved the goal of preserving the functioning of WTO dispute settlement.

A secondary goal of the MPIA was to test the waters with possible innovations to enhance “the procedural efficiency of appeal proceedings”.[7] This includes a mandate for MPIA appeal arbitrators to “take appropriate organizational measures to streamline the proceedings” including “decisions on page limits, time limits and deadlines as well as on the length and number of hearings required”.[8] MPIA participants also allowed arbitrators to “propose substantive measures to the parties, such as an exclusion of claims based on the alleged lack of an objective assessment of the facts pursuant to Article 11 of the DSU”[9] and explicitly limited the scope of appellate review to “only address those issues that are necessary for the resolution of the dispute” and “only those issues that have been raised by the parties”.[10]

The third post in this series will describe to what extent the first MPIA award brought about any such innovations.

 

(*) Full disclosure: the author was one of the three arbitrators in the first MPIA appeal. This contribution is written in the author’s personal capacity, conveys only publicly available information, and is limited to providing factual information about (not a critical or substantive assessment of) the MPIA and its process in order to enrich the public debate about WTO dispute settlement, its state-of-play and reform process.

[1] One earlier appeal arbitration award was issued under DSU Article 25, in a dispute between the EU and Türkiye, but this was not formally under the MPIA (as Türkiye is not an MPIA participant) but as a result of an ad hoc appeal arbitration agreement between the parties which incorporated some but not all elements of the MPIA. See Turkey — Certain Measures concerning the Production, Importation and Marketing of Pharmaceutical Products, Arbitration under Article 25 of the DSU, Award of the Arbitrators, WT/DS583/ARB25, 25 July 2022.

[2] Colombia – Anti-Dumping Duties on Frozen Fries From Belgium, Germany and The Netherlands, Arbitration under Article 25 of the DSU, Award of the Arbitrators, WT/DS591/ARB25, 21 December 2022.

[3] For updated information on MPIA participants and cases, see https://wtoplurilaterals.info/plural_initiative/the-mpia/.

[4] MPIA Pursuant to Article 25 of the DSU, 30 April 2020, JOB/DSB/1/Add.12 (hereafter “MPIA”), para. 2.

[5] DSU Article 25 provides, in crucial parts, as follows:

1. Expeditious arbitration within the WTO as an alternative means of dispute settlement can facilitate the solution of certain disputes that concern issues that are clearly defined by both parties.
2. Except as otherwise provided in this Understanding, resort to arbitration shall be subject to mutual agreement of the parties which shall agree on the procedures to be followed …
3. … The parties to the proceeding shall agree to abide by the arbitration award.  Arbitration awards shall be notified to the DSB … where any Member may raise any point relating thereto.
4. Articles 21 and 22 of this Understanding [on implementation and enforcement of WTO panel/Appellate Body rulings] shall applymutatis mutandis to arbitration awards.

[6] MPIA, 4^th preambular paragraph.

[7] Ibid., Annex 1, para. 12.

[8] Ibid.

[9] Ibid., Annex 1, para. 13.

[10] Ibid., Annex 1, para. 10.

U.S. Views on MPIA Arbitrators' Consideration of AD Agreement Legal Standard of Review

At last Friday's DSB meeting, the U.S. had some positive things to say about how the MPIA arbitrators in the Colombia - Frozen Fries dispute handled the issue of the AD Agreement Article 17.6(ii) legal standard of review:

- The United States intervenes to comment on the Arbitrator’s findings with respect to Article 17.6(ii) of the Anti-Dumping Agreement (“AD Agreement”).

- Article 17.6(ii) was key to the acceptance of the other provisions of the AD Agreement. The existence of such a provision confirms that Members were aware that the text of the AD Agreement could pose particular interpretive challenges, at least in part because it was drafted to cover varying and complex anti-dumping regimes and long-standing differences concerning methodology. WTO Members agreed, therefore, that it would be a legal error not to respect a permissible interpretation of the AD Agreement.

- The question under Article 17.6(ii) is whether an investigating authority’s interpretation of the AD Agreement is a permissible interpretation. As the United States has explained for years, “permissible” means just that: a meaning that could be reached under the Vienna Convention.1

- Article 17.6(ii) itself confirms that provisions of the AD Agreement may “admit[] of more than one permissible interpretation.”

- Where that is the case, and where the investigating authority has relied upon one such interpretation, a panel must find the measure to be in conformity with the AD Agreement. As one panel report stated, “[I]n accordance with Article 17.6(ii) of the AD Agreement, if an interpretation is ‘permissible’, then we are compelled to accept it.”2

- The Arbitrator seriously engaged with the text of Article 17.6(ii). The Arbitrator appropriately recognized that the subparagraphs of Article 17.6 “must be understood in a manner granting special deference to investigating authorities under the Anti-Dumping Agreement.”3

- The Arbitrator undertook its textual analysis in a manner that gave effect to the terms of the provision.

- The analysis addressed directly the nature of treaty interpretation under the Vienna Convention. The Arbitrator explained that its “approach assumes, as the second sentence does, that different treaty interpreters applying the same tools of the Vienna Convention may, in good faith and with solid arguments in support, reach different conclusions on the ‘correct’ interpretation of a treaty provision.”4

- The Arbitrator went on to note favorably the observation that the Vienna Convention rules “are facilitative not disciplinary and do not ‘instruct the treaty interpreter to find a single meaning of the treaty’ as a former Appellate Body member has written.”5

- We agree with this observation and find the Arbitrator’s analysis of Article 17.6(ii) to be persuasive.

- The Arbitrator’s analysis differs from – and implicitly rejects – the past reports of the Appellate Body that expressed disregard for the meaning and importance of Article 17.6(ii).

- In no WTO dispute did the Appellate Body ever find that there was a permissible interpretation of the AD Agreement other than the interpretation favored by the Appellate Body. Instead, the Appellate Body simply asserted “the rules and principles of the Vienna Convention cannot contemplate interpretations with mutually contradictory results”6

- That assertion was unsupported. The ordinary meaning of “permissible” is “allowable” or “permitted” – that is, an interpretation that could be reached under customary rules of interpretation.

- We welcome the Arbitrator’s willingness to adopt an interpretation that is consistent with the text agreed by Members, even and especially if the interpretation differs from the Appellate Body’s erroneous views.7

- We recall that some Members appear to promote “security and predictability” as an independent goal for dispute settlement, and from that derive that appellate interpretations must simply be followed. But this Arbitrator has developed an interpretation that is more consistent with the text agreed by Members. Adherence to the texts as negotiated and agreed by Members – rather than consistency for consistency’s sake – assists in providing security and predictability to the multilateral trading system – per the precise words of Article 3.2 of the DSU.

- We look forward to engaging further with Members on the important issues raised in the Arbitrator’s Award.

This U.S. statement raises a few questions for me related to some of the big systemic questions hovering out there on WTO dispute settlement.

First, does the MPIA arbitrators' finding here make the U.S. more sympathetic to the MPIA, and would it considering using the MPIA, or DSU Article 25 on an ad hoc basis, as an appeals mechanism in the future?

Second, how does the U.S. see the precedential value of this finding? Would the U.S. want to see this MPIA finding treated by future WTO panels as a precedent with some degree of persuasive value? In the U.S. view, to what extent should the issue be reargued in future disputes based on the agreement text?

Third, how do other governments view the MPIA arbitrators' findings here? Were they persuaded by the reasoning, so there is now a consensus around this approach? Or do some of them prefer the Appellate Body's approach? (If anyone knows if other governments spoke on this issue at the DSB meeting or elsewhere, feel free to mention it in the comments).

The First MPIA Award Is Out, and It Discusses AD Agreement Article 17.6(ii) "Permissible" Interpretations

The first MPIA Award, in the Colombia - Frozen Fries disputes, was circulated today. In a previous blog post on this dispute, I had some uncertainty about whether the arbitrators would say anything about the AD Agreement Article 17.6(ii) legal standard of review, but it turns out they did:

4.11. The Panel stated that whether a provision admits of more than one "permissible" interpretation (under the second sentence) depends on whether more than one such interpretation emerges after the Panel has examined the relevant provision under customary rules of interpretation of public international law (under the first sentence). Accordingly, the Panel stated that the "starting point" of its interpretative analysis required an interpretation of the relevant provision under treaty interpretation rules, including Articles 31 and 32 of the Vienna Convention on the Law of Treaties (Vienna Convention).41

4.12. Our approach to the interpretation claims in this appeal differs. We do not begin the interpretative exercise by focusing solely on the first sentence of Article 17.6(ii), as this in our view pays insufficient regard to the immediate context of this sentence, namely Article 17.6(i) and the second sentence of Article 17.6(ii). Each of these provisions must be understood in a manner granting special deference to investigating authorities under the Anti-Dumping Agreement.42 The second sentence of Article 17.6(ii) mandates panels to defer to and accept an authority's measure as soon as it "rests upon" a "permissible" interpretation. As we have noted, Article 17.6(i) prevents a panel from conducting a de novo assessment of the facts on record; an authority's establishment and evaluation of facts must be allowed to stand so long as it is "proper" and "unbiased and objective", and this is the case "even though the panel might have reached a different conclusion".

4.13. Reading these provisions together, we aim to integrate the elements of interpretation under Article 17.6(ii). As a result, we will begin by asking ourselves whether Colombia's proposed interpretation of the phrase "where appropriate" in Article 5.2(iii) – reflected in MINCIT's decision to initiate its investigation, in part, on the basis of third-country sales prices – is a "permissible" one. As a yardstick for "permissibility", the first sentence of Article 17.6(ii) refers us to the customary rules of treaty interpretation. However, we will not engage in our own, de novo interpretation of the terms "where appropriate" so as to arrive at what we consider to be the "final" or "correct" application of Articles 31 and 32 of the Vienna Convention. Instead, we will ask whether a treaty interpreter, using the method for treaty interpretation set out in the Vienna Convention – that is, an interpretation "in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose" and, where appropriate, "supplementary means of interpretation" – could have reached Colombia's interpretation. And this even though we, as de novo treaty interpreters, might have reached a different conclusion.

4.14. Our approach assumes, as the second sentence does, that different treaty interpreters applying the same tools of the Vienna Convention may, in good faith and with solid arguments in support, reach different conclusions on the "correct" interpretation of a treaty provision.43 This may be particularly true for the Anti-Dumping Agreement, which was drafted with the understanding that investigating authorities employ different methodologies and approaches. Treaty interpretation is not an exact science and applying the Vienna Convention's method does not magically and inevitably lead to a single result. In most cases, treaty interpretation involves weighing, balancing, and choice.

4.15. Thus, the ultimate question for us when testing a proposed interpretation is to draw a line beyond which an interpretation is no longer "permissible" under the Vienna Convention method for treaty interpretation. Dictionary meanings support the idea that the search for "permissible" interpretations differs from an attempt to find one's own – "final" and "correct" – interpretation.44 Rather, the question is whether someone else's interpretation is "permitted", "allowable", "acceptable", or "admissible" as an outcome resulting from a proper application of the interpretative process called for under the Vienna Convention. Obviously, not just any interpretation put forward by an authority can be accepted as "permissible". The interpretative process under the Vienna Convention sets out an outer range beyond which meanings cannot be accepted. Just as permissible interpretations cannot be limited to a single "final" and "correct" answer as determined by a given tribunal, not all interpretations have the required degree of solidness or analytical support for them to be given deference as "permissible" within the bounds of the Vienna Convention method for treaty interpretation.

At first glance, I like this approach in the abstract. It seems like a clear and reasonable articulation of a legal provision that many people have found confusing. It will take a few instances of application to get a better sense of how it will work in practice though. (The Arbitrators in this dispute rejected the two arguments by Colombia that the interpretations by its investigating authorities were permissible).

Of course, what I think is not as important here as what governments think. In this regard, I wonder if this approach is one that the U.S. would support in the context of its objections to Appellate Body overreach.

The Frozen Fries Appeal: Permissible Interpretations under AD Agreement Article 17.6(ii)

When the Colombia - Frozen Fries MPIA appeal started up, I noted that "most of the issues are pretty basic anti-dumping ones, but there is a possibility of an important systemic issue coming up," as "[t]he panel report gets into the issue of the AD Agreement Article 17.6(ii) legal standard of review a bit."

Recall that the Article 17.6(ii) legal standard of review states:

the panel shall interpret the relevant provisions of the Agreement in accordance with customary rules of interpretation of public international law. Where the panel finds that a relevant provision of the Agreement admits of more than one permissible interpretation, the panel shall find the authorities' measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations.

My sense is that many people find the "more than one permissible interpretation" language in the second sentence to be confusing. It clearly suggests that there may be multiple permissible interpretations, but when exactly are there multiple permissible interpretations? Treaty language is often unclear, and parties put forward competing interpretations, but how do we determine if there are multiple permissible interpretations in a given interpretive exercise?

The EU has now posted its written submission and oral statement at the hearing in the Frozen Fries appeal. In its written submission, it offers the following argument about the application of 17.6(ii):

19. The European Union disagrees with Colombia’s suggestion that the application of the legal standard under Article 17.6(ii) of the Anti-Dumping Agreement requires a panel to determine whether there are two or more permissible interpretations of a legal provision. In most cases, the conscientious application of the rules of treaty interpretation in Articles 31 and 32 of the Vienna Convention yields a single interpretation of the provision in question. It is only in cases where the application of those rules leads to two or more interpretations of the same provision of the Anti-Dumping Agreement that the second sentence of Article 17.6(ii) applies.

20. In those cases where the application of the rules of treaty interpretation in Articles 31 and 32 lead to two or more interpretations of the same provision of the Anti-Dumping Agreement, however, the European Union considers that the application of the second sentence of Article 17.6(ii) should not require a panel to choose between mutually contradictory interpretations, as Colombia seems to assume. The permissible interpretations are only those falling within a range that fits harmoniously with the terms, context, and object and purpose of the treaty. As stressed by the Appellate Body:

“the second sentence [of Article 17.6(ii)] allows for the possibility that the application of the rules of the Vienna Convention may give rise to an interpretative range and, if it does, an interpretation falling within that range is permissible and must be given effect by holding the measure to be in conformity with the covered agreement. The function of the second sentence is thus to give effect to the interpretative range rather than to require the interpreter to pursue further the interpretative exercise to the point where only one interpretation within that range may prevail. […] the enterprise of interpretation is intended to ascertain the proper meaning of a provision; one that fits harmoniously with the terms, context, and object and purpose of the treaty. The purpose of such an exercise is therefore to narrow the range of interpretations, not to generate conflicting, competing interpretations.”

21. The Panel in this dispute correctly applied the legal standard under Article 17.6(ii), whose starting point is the requirement in the first sentence to interpret the relevant provisions of the Anti-Dumping Agreement in accordance with customary rules of interpretation of public international law. Contrary to Colombia’s contentions, the legal standard under Article 17.6(ii) does not consist primarily of applying the second sentence and examining whether the authority’s determination rests on a permissible interpretation.

It is not clear whether the MPIA arbitrators in this appeal will address the Article 17.6(ii) "permissible interpretation" standard. Note that the EU argues: "The European Union must stress that Colombia’s Notice of Appeal contains no reference whatsoever to a claim of error in the interpretation or application of Article 17.6(ii) of the Anti-Dumping Agreement. Therefore, it is clear that the issue of the Panel's interpretation or compliance with Article 17.6(ii) of the Anti-Dumping Agreement is not properly before the Arbitrators in this appeal." Thus, there appears to be a procedural way out of weighing in on this difficult issue if the arbitrators want to take it.

However, if they were to address this issue, one possible approach to Article 17.6(ii) would be the following. The EU doesn't set forth its view of the specific circumstances in which there may be more than one permissible interpretation, but its reference to Articles 31 and 32 of the Vienna Convention on the Law of Treaties is helpful. Recall that Article 32 of the Vienna Convention says:  

Article 32: Supplementary means of interpretation

Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:

(a) leaves the meaning ambiguous or obscure; or

(b) leads to a result which is manifestly absurd or unreasonable.

Based on this provision, we know that there can be situations where the Article 31 treaty interpretation process "leaves the meaning ambiguous or obscure." Perhaps, then, adjudicators who are interpreting Article 17.6(ii) could use this language to inform the meaning of "permissible interpretation." In this regard, if the Article 31 interpretation process on its own "leaves the meaning ambiguous or obscure," there could be a presumption -- either rebuttable or irrebuttable -- that multiple permissible interpretations exist. In practice, what this could mean is that where there is a high enough degree of uncertainty about the meaning of a particular provision after going through the Article 31 interpretive process, the adjudicator could find that the meaning is "ambiguous or obscure," and that there are multiple permissible interpretations of the provisions within the meaning of Article 17.6(ii). It might take a few instances of applying this standard to get a sense of when the meaning of a term or phrase or provision is "ambiguous or obscure," but over time I think we would get some clarity. 

Let me make two quick additional points.

First, in theory the Article 32 interpretive process could resolve the ambiguity or obscurity, but based on what I've seen with recourse to Article 32, that is unlikely in practice.

And second, the "leads to a result which is manifestly absurd or unreasonable" language could also be used in the same way, but to me that standard seems a little harder to demonstrate.

Clarifying AD Agreement Article 17.6(ii)

The United States has expressed concerns about various Appellate Body interpretations of WTO obligations, in particular relating to trade remedies, but as far as I know it has not offered any specific proposals to address the flaws it sees in the Appellate Body's approach to treaty interpretation in these cases. There was a proposal many years ago that involved, among other things, "partial adoption" of Appellate Body reports and interim review of those reports, but that doesn't directly address the interpretive process. I have wondered if it has some proposals coming on interpretation, but I'm getting a little antsy so I thought I'd offer one of my own.

The United States -- and others as well -- has suggested that the AD Agreement Article 17.6(ii) legal standard is not working properly. Here is the text of that provision:

(ii) the panel shall interpret the relevant provisions of the Agreement in accordance with customary rules of interpretation of public international law. Where the panel finds that a relevant provision of the Agreement admits of more than one permissible interpretation, the panel shall find the authorities' measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations.

The concern is that panels and the Appellate Body have been too reluctant to find that multiple permissible interpretations exist, and in this way are applying the provision with too little deference to investigating authorities.

Are there drafting changes that could ensure more deference to investigating authorities? Here's one idea I had. Article 32 of the Vienna Convention on the Law of Treaties says:  

Article 32: Supplementary means of interpretation

Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:

(a) leaves the meaning ambiguous or obscure; or

(b) leads to a result which is manifestly absurd or unreasonable.

Based on this provision, we know that there are situations where the Article 31 treaty interpretation process "leaves the meaning ambiguous or obscure." Throughout the history of treaty interpretation, there must be a fair amount of experience with this and some good examples to guide the way. Perhaps, then, we could use this to inform the Article 17.6(ii) standard. Specifically, if the Article 31 interpretation process "leaves the meaning ambiguous or obscure," there could be a presumption that multiple permissible interpretations exist. In this way, Article 17.6(ii) could be given a little more substance, and panels and the Appellate Body would have a better sense of how to apply it with the kind of deference the United States is looking for. (The "leads to a result which is manifestly absurd or unreasonable" language could also be used in the same way, but that seems a little harder to demonstrate).

The redrafted Article 17.6(ii) text could look something like this:

(ii) the panel shall interpret the relevant provisions of the Agreement in accordance with customary rules of interpretation of public international law. Where the panel finds that a relevant provision of the Agreement admits of more than one permissible interpretation, the panel shall find the authorities' measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations. As an illustration of a situation where more than one permissible interpretation exists, if the panel concludes that the interpretive approach set out in Article 31 of the Vienna Convention on the Law of Treaties leaves the meaning of a provision ambiguous or obscure, there shall be a presumption that the provision admits of more than one permissible interpretation, and the panel shall find the authorities' measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations.

We could take the same idea even further. Here is the full text of Article 31:

Article 31: General rule of interpretation

1. A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.

2. The context for the purpose of the interpretation of a treaty shall comprise, in addition to the text, including its preamble and annexes:

(a) any agreement relating to the treaty which was made between all the parties in connection with the conclusion of the treaty;

(b) any instrument which was made by one or more parties in connection with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty.

3. There shall be taken into account, together with the context:

(a) any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions;

(b) any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation;

(c) any relevant rules of international law applicable in the relations between the parties.

4. A special meaning shall be given to a term if it is established that the parties so intended.

The revised Article 17.6(ii) text could say that if the meaning is "ambiguous or obscure" after Article 31(1) alone is applied, then the presumption of multiple permissible interpretations would apply.

This is just one suggestion. There are probably many others along the same lines.

I have no idea how other Members would react to a proposal such as this. But at least it might get the discussion moving in a productive direction.