¹ Office of the United States Trade Representative (USTR), 2018 Trade Policy Agenda and 2017 Annual Report of the President of the United States on the Trade Agreements Program (2018) 22.

² ibid, 22–8.

³ These criticisms are by no means new, dating back at least to the US–FSC dispute in 2000. See for a brief overview: T Paysova, G Hufbauer and JJ Schott, ‘The Dispute Settlement Crisis in the World Trade Organization: Causes and Cures’ (2018) 18(5) Peterson Institute for International Economics Policy Brief. cf Greenwald, J, ‘WTO Dispute Settlement: An Exercise in Trade Law Legislation?’ (2003) 6(1) JIEL 113CrossRefGoogle Scholar.

⁴ On the ways such changing understandings shape international economic governance over time: Ruggie, J, ‘International Regimes, Transactions and Change: Embedded Liberalism in the Postwar Economic Order’ (1982) 36(2) IntlOrg 379Google Scholar; Howse, R, ‘From Politics to Technocracy – and Back Again: The Fate of the Multilateral Trading Regime’ (2002) 96(1) AJIL 94CrossRefGoogle Scholar; Lang, A, World Trade Law after Neoliberalism: Re-imagining the Global Economic Order (Oxford University Press 2011)CrossRefGoogle Scholar.

⁵ See for such proposals: European Commission, WK 8239/2018 INIT WTO – EU's Proposals for WTO Modernisation (5 July 2018); J Bacchus, ‘How to Solve the WTO Judicial Crisis’ (CATO at Liberty, 6 August 2018) available at <https://www.cato.org/blog/how-solve-wto-judicial-crisis>.

⁶ A Mitchell, Legal Principles in WTO Disputes (Cambridge University Press 2008).

⁷ I Van Damme, Treaty Interpretation by the WTO Appellate Body (Oxford University Press 2009) Ch 4.

⁸ A Qureshi, Interpreting WTO Agreements: Problems and Perspectives (2nd edn, Cambridge University Press 2015).

⁹ See eg Howse (n 4); Lang (n 4). For studies that move somewhat closer to normative theorizing, while remaining predominantly sociological: R Howse, ‘Adjudicative Legitimacy and Treaty Interpretation in International Trade Law: The Early Years of WTO Jurisprudence’ in JHH Weiler (ed), The EU, the WTO and the NAFTA: Towards a Common Law of International Trade (Oxford University Press 2000); S Cho, ‘Global Constitutional Lawmaking’ 31 UPaJIntlL 621.

¹⁰ eg D Regan, ‘What are Trade Agreements for? Two Conflicting Stories Told by Economists, with a Lesson for Lawyers’ (2006) 9(4) JIEL 951.

¹¹ eg R Howse and K Nicolaidis, ‘Legitimacy and Global Governance: Why Constitutionalizing the WTO Is a Step Too Far’ in RB Porter, P Sauve, A Subramanian and AB Zampetti (eds), Efficiency, Equity and Legitimacy: The Multilateral Trading System at the Millennium (Brookings Institution Press 2001).

¹² eg FM Abbott, C Breining-Kaufman and T Cottier (eds), International Trade and Human Rights: Foundations and Conceptual Issues (University of Michigan Press 2006).

¹³ eg F Garcia, Global Justice and International Economic Law: Three Takes (Cambridge University Press (2013); O Suttle, Distributive Justice and World Trade Law: A Political Theory of International Trade Regulation (Cambridge University Press 2018).

¹⁴ eg SE Rolland, Development at the World Trade Organization (Oxford University Press 2012).

¹⁵ eg T Cottier, O Nartova and SZ Bigdeli (eds), International Trade Regulation and the Mitigation of Climate Change (Cambridge University Press 2009).

¹⁶ For a similarly capacious conception of moral reasoning, J Raz, Ethics in the Public Domain (Clarendon Press1994) 328–30.

¹⁷ This is true as much for positivists as anti-positivists. See eg HLA Hart, The Concept of Law (2nd edn, Oxford University Press 1994) 130–33; J Raz, ‘The Relevance of Coherence’ (1992) 72(2) BULRev 273; S Shapiro, Legality (Harvard University Press, 2011) 251; and for the leading anti-positivist view, R Dworkin, Law's Empire (Harvard University Press 1986) passim.

¹⁸ One set of arguments not considered here is those grounded in democratic legitimacy, as in eg Howse and Nicolaidis (n 11). In endorsing some interpretations (those affording policy space to States to realize particular goals) over others on the ground that they better realize the value of democracy, such approaches necessarily endorse a role for values in adjudication. The question of whether values play a role is distinct from the question of which values are appropriate. It is on the former question that this first section concentrates, and on this question, advocates of democratic deference are committed to an affirmative answer.

¹⁹ Understanding on Rules and Procedures Governing the Settlement of Disputes (DSU) art 1 and Annex 1. On the relation between WTO law and public international law more generally: J Pauwelyn, Conflict of Norms in Public International Law: How WTO Law Relates to Other Rules of International Law (Cambridge University Press 2003). cf M Trebilcock, R Howse, and A Eliason, The Regulation of International Trade (4th edn, Routledge 2013) 190–2, 198–203.

²⁰ DSU art 3.4.

²¹ For this view of the relation between WTO law and general international law: D Steger, ‘The WTO in Public International Law: Jurisdiction, Interpretation and Accommodation’ in Ten Years of WTO Dispute Settlement (International Bar Association 2007). Although contrast Pauwelyn (n 19). This does not mean that general international law is irrelevant to WTO adjudication. The Appellate Body regularly looks beyond the covered agreements to aid to interpretation, but causes of action and defences are limited to those in the agreements.

²² G Marceau, ‘WTO Dispute Settlement and Human Rights’ (2002) 13(4) EJIL 753, 762–3.

²³ For a prominent positivist view highlighting the extent to which the law may fail to morally obligate: J Raz, ‘The Obligation to Obey the Law’ in The Authority of Law (2nd edn, Oxford University Press 2009) 233. Greenberg emphasizes this gap in criticizing positivist views generally: M Greenberg, ‘The Moral Impact Theory of Law’ (2014) 123 Yale LJ 1288, 1304.

²⁴ On the merits of making this distinction, Hart (n 17) 207–12.

²⁵ This is as true of Raz, with his understanding of pre-emption, as it is of Hart. While Raz characterizes the directives of legitimate authorities as constituting content-independent and exclusionary reasons, he denies that the law has the kind of extensive authority that it claims, and recognizes that reasons other than dependent reasons may defeat the directives of legitimate authorities in appropriate circumstances. J Raz, The Morality of Freedom (Oxford University Press 1986) 46, 62, 74–80; J Raz, ‘The Problem of Authority: Revisiting the Service Conception’ (2006) 90 MinnLRev 1003, 1022.

²⁶ By way of illustration, Davey puts the successful implementation rate for WTO disputes in the first ten years at 83 per cent. WJ Davey, ‘WTO Dispute Settlement: The First Ten Years’ (2005) 8(1) JIEL 17. For a sceptical review of the extent to which WTO dispute settlement accords due respect to non-WTO international law: A Lindroos and M Mehling, ‘Dispelling the Chimera of ‘‘Self-Contained Regimes’’: International Law and the WTO’ (2006) 16(5) EJIL 857.

²⁷ See for example on the ways a direct conflict between trade and environmental regimes would play out in dispute settlement: GR Winham, ‘International Regime Conflict in Trade and Environment: The Biosafety Protocol and the WTO’ (2003) 2(2) World Trade Review 131, 147–50. On the lack of mechanisms for resolving conflicts between WTO adjudication and dispute settlement under regional trade agreements, J Hillman, (2009), ‘Conflicts Between Dispute Settlement Mechanisms in Regional Trade Agreements and the WTO – What Should the WTO Do?’ 42 CornellIntlLJ 193. Another worry here might be that admitting values in WTO adjudication will further encourage States to treat WTO decisions as stating their all-things-considered obligations. Hart celebrates the separation of validity and morality as enabling a clear-eyed assessment of how far laws in fact merit obedience: Hart (n 17) 207–12. The more values are integrated within WTO law, the harder it is to separate these issues. This may be especially troubling if particular values are likely, in the hands of trade adjudicators, to acquire a distinctly pro-trade interpretation. See eg P Alston, ‘Resisting the Merger and Acquisition of Human Rights by Trade Law: A Reply to Petersmann’ (2002) 13(4) EJIL 815.

²⁸ C Fried, Contract as Promise (Harvard University Press 1981).

²⁹ PS Atiyah, The Rise and Fall of Freedom of Contract (Clarendon Press 1979).

³⁰ R Posner, Economic Analysis of Law (9th edn, Wolters Kluwer 2014) Ch 4.

³¹ For detailed arguments for such trade-specific values: Garcia (n 13); A James, Fairness in Practice: A Social Contract for a Global Economy (Oxford University Press 2012); Suttle (n 13).

³² For this view of treaties generally, T Nagel, ‘The Problem of Global Justice’ (2005) 33(2) Philosophy & Public Affairs 113, 141. Howse suggests that the WTO treaties are appropriately read in more contractarian terms than, for example, human rights treaties: R Howse, ‘The World Trade Organization 20 Years on: Global Governance by Judiciary’ (2016) 27(1) EJIL 9, 44–5. On some of the challenges that arise in applying ideas developed in the context of contractual treaties to ‘law-making’ treaties that do not fully fit that mould: J Klabbers, ‘How to Defeat a Treaty's Object and Purpose Pending Entry into Force: Toward Manifest Intent’ (2001) 34 Vanderbilt Journal of Transnational Law 283. For a more general discussion of the ways the turn to law-making treaties might suggest rethinking treaty interpretation: Qureshi, (n 8) 5–9. cf Mitchell (n 6) 85–8.

³³ This view seems implicitly engaged in AB Member Hernández's analysis of the appointment crisis as in part reflecting ‘conceptual differences among the Membership as to the nature of the WTO. That is, is the WTO a contract or a constitution?’. Farewell Speech of Appellate Body Member Ricardo Ramírez-Hernández (28 May 2018) available at <https://www.wto.org/english/tratop_e/dispu_e/ricardoramirezfarwellspeech_e.htm>.

³⁴ See sources (n 31).

³⁵ ‘Natural’ here is used in the sense in which Hart distinguishes natural rights from special rights, referring to duties that do not arise from particular transactions or special relationships into which States have entered: HLA Hart, ‘Are There Any Natural Rights?’ (1955) 64(2) The Philosophical Review 175.

³⁶ Something like this view is implicit in the discussion of the WTO's institutional role in R Howse, J Langille and K Sykes, ‘Pluralism in Practice: Moral Legislation and the Law of the WTO after Seal Products’ (2015) 48 GeoWashLRev 81, 89–91.

³⁷ On the latter point, and its significance in distinguishing liberal and libertarian views: S Freeman, ‘Illiberal Libertarians: Why Libertarianism Is Not a Liberal View’ (2005) 30 Philosophy & Public Affairs 105.

³⁸ For a similar point, P Maffetone, ‘The WTO and the Limits of Distributive Justice’ (2009) 35(3) Philosophy and Social Criticism 243.

³⁹ See eg E McKendrick, Contract Law (12th edn, Palgrave 2017) 171.

⁴⁰ Raz (n 23) Ch. 2, ‘The Claims of Law’.

⁴¹ J Bello, ‘The WTO Dispute Settlement Understanding: Less Is More’ (1996) 90 AJIL 416; WF Schwartz and AO Sykes, ‘The Economic Structure of Renegotiation and Dispute Resolution in the World Trade Organization’ (2002) 31 JLS 179. And, for the opposite view, JH Jackson, ‘The WTO Dispute Settlement Understanding: Misunderstandings on the Nature of Legal Obligation’ (1997) 91 AJIL 60; JH Jackson, ‘International Law Status of WTO Dispute Settlement Reports: Obligations to Comply or Option to “Buy Out”?’ (2005) 98 AJIL 109.

⁴² See Jackson (n 41).

⁴³ On the ways the domains of law and morality overlap, Hart (n 17) 167ff. Marmor suggests that the core claims of legal positivism are about distinguishing legal from other forms of normativity. A Marmor, ‘What's Left of General Jurisprudence? On Law's Ontology and Content’ (2018) Cornell Legal Studies Research Paper No 18–26.

⁴⁴ Indeed, in the context of dispute settlement, complainants tailor retaliation to impact politically influential constituencies, to maximize political pressure for compliance. They thus harm some agents instrumentally, in order to change the behaviour of others, which Kantians in particular will find morally troubling.

⁴⁵ Raz (n 16) 328–9; R Alexy, A Theory of Legal Argumentation (1978) (R Adler and N MacCormick trans, Clarendon Press 1989) 9.

⁴⁶ For a similar point, Raz (n 16) 334–5.

⁴⁷ This may not be the only relevant consideration. If we each have an interest in the justice of the institutions in which we participate then we will find value for just States in a just WTO law, even if this does not change unjust States’ behaviour.

⁴⁸ The objection here links to arguments for normative minimalism in public international law more generally. See most prominently P Weil, ‘Towards Relative Normativity in International Law?’ (1983) 77(3) AJIL 413.

⁴⁹ B Kingsbury, ‘The Concept of Compliance as a Function of Competing Conceptions of International Law’ (1998) 19 MichJIntlL 344, 350–1. In the trade context, Krasner gives an account of the determinants of trade openness in these terms: S Krasner, ‘State Power and the Structure of International Trade’ (1976) 28(3) World Politics 317, 322–3.

⁵⁰ As Fuller pointed out, the logic of the legal form and the requirements of effectiveness impose a limited ‘internal morality’, including a requirement of generality, regardless of the ultimate political motivation of lawgivers; L Fuller, The Morality of Law (Yale University Press 1964) 33.

⁵¹ See eg Press Release, President Trump Announces Strong Actions to Address China's Unfair Trade (March 2018) Office of the United States Trade Representative <https://ustr.gov/about-us/policy-offices/press-office/press-releases/2018/march/president-trump-announces-strong>. Indeed, various features of the Uruguay Round Agreements are explained as attempting to bring WTO law sufficiently close to US expectations that it would be willing to restrain its previous unilateralism: Howse (n 32) 18; D McRae, ‘Measuring the Effectiveness of the WTO Dispute Settlement System’ (2008) 3 Asian Journal of WTO and International Health Law and Policy 1, 4–6.

⁵² Much recent commentary on problems adjudicating the GATT National Security exemption reflects similar concerns.

⁵³ See generally A Hasenclever, P Mayer and V Rittberger, Theories of International Regimes (Cambridge University Press 1997) 33–6.

⁵⁴ For this challenge to the Appellate Body, Greenwald (n 3) 114–15. For a recent discussion emphasizing the specificity of relevant reputation in institutionalist accounts: R Brewster, ‘Exit from Trade Agreements: A Reputational Analysis of Cooperation and Fairness’ (2018) 21(2) JIEL 379.

⁵⁵ Kingsbury (n 49) 358–60. See generally A Wendt, Social Theory of International Politics (Cambridge University Press 1999).

⁵⁶ The issue here is one of ‘social legitimacy’— how far relevant constituencies, whether those affected or those required to implement, regard decisions as being legitimate. See generally Howse (n 9).

⁵⁷ As Ruggie observes, ‘political authority represents the fusion of power with legitimate social purpose’. Ruggie (n 4) 382.

⁵⁸ See generally R Keohane, After Hegemony: Cooperation and Discord in the World Political Economy (Princeton University Press 1984). This echoes an insight of critical and Marxist thinkers about the emancipatory potential latent in hegemonic ideologies.

⁵⁹ On the ways existing understandings condition ethical reasoning and its limits in constructivist thought, R Price, Moral Limit and Possibility in World Politics (Cambridge University Press 2008) 9–12.

⁶⁰ Indeed, Shaffer and Trachtman suggest that Article 3.4 DSU mandates Panels and the Appellate Body to make acceptability to members a consideration in crafting decisions: G Shaffer and J Trachtman, ‘Interpretation and Institutional Choice at the WTO’ (2011) 52 VaJIntlL 103, 120. Howse reads the Appellate Body's evolving interpretive approach as reflecting this kind of political sensitivity, given the changing fortunes of the WTO as an institution and the neoliberal consensus on which he sees it being built: Howse (n 32) 29–30.

⁶¹ This reflects the oft-highlighted difference between the diplomatic dispute settlement of the GATT, particularly in its earlier years, and the legalized approach under the WTO. See generally JHH Weiler (2001) ‘The Rule of Lawyers and the Ethos of Diplomats: Reflections onf the Internal and External Legitimacy of WTO Dispute Settlement’ 35 JWT 191; Lang (n 4).

⁶² For a discussion of the problems that result from the conflation of a diverse tradition with a specific thesis or theses: J Gardner, ‘Legal Positivism: 5 ½ Myths’ (2001) 46(1) AmJJuris 199.

⁶³ Raz (n 23) 37.

⁶⁴ The extent to which moral reasoning can constitute a subsidiary element in legal reasoning distinguishes ‘exclusive’ and ‘inclusive’ positivist views. See generally KE Himma, ‘Inclusive Legal Positivism’ in J Coleman and S Shapiro (eds), The Oxford Handbook of Jurisprudence and Philosophy of Law (Oxford University Press 2002). Coleman suggests that the core disagreement between inclusive positivism and Dworkin's anti-positivism is about whether the role of morality in legal reasoning is necessary or contingent/conventional: J Coleman, The Practice of Principle: In Defence of a Pragmatist Approach to Legal Theory (Oxford University Press 2001) 108. To the extent this is the case, the argument I advance below may be compatible with inclusive positivism, insofar as my claims relate to a particular legal system, rather than addressing law in general. However, the arguments about customary interpretation and object and purpose are likely too general to be compatible with Coleman's conventionalism. Regardless, nothing of substance in my argument turns on this point of classification.

⁶⁵ Gardner (n 60) 201 This does not commit positivists to a view of law as simply the semantic content of legislative acts; rather, the necessary implications, presuppositions etc of those texts as communicative acts can similarly fall under the positivist account of law: Marmor (n 43).

⁶⁶ See eg R Dworkin, ‘The Model of Rules’ (1967) 35 UChiLRev 14, 17; Hart (n 17) 127–8; Shapiro (n 17) 251.

⁶⁷ See generally Hart (n 17) 124–8. This point requires slight modification in the international context, where State behaviour is an important source for customary international law which may in particular cases have little or no linguistic content. In such cases, the issue is not one of identifying whether new cases fall under old words, but rather whether the example constituted by past conduct extends to the facts of a new case. However, the same problem of uncertainty at the margins remains, and is probably more significant in the case of such non-linguistic sources.

⁶⁸ See generally Hart (n 17) 128–33; Shapiro (n 17) 198–200. On the efficiency of legislative incompleteness in the WTO context, J Trachtman, ‘The Domain of WTO Dispute Resolution’ (1999) 40 HarvIntlLJ 333.

⁶⁹ For this point in the WTO context, Van Damme (n 7) 141–6.

⁷⁰ For this point in the international context, albeit using the language of formalism rather than positivism: J D'Aspremont, Formalism and the Sources of International Law (Oxford University Press 2011) 18–21.

⁷¹ Raz (n 16) 330–5.

⁷² Hart (n 17) 136.

⁷³ Although, for the argument that judicial law-making is nonetheless not ‘legislative’, Gardner (n 60) 214–18.

⁷⁴ AV Lowe, International Law (Oxford University Press 2007) 25–7; B Simma and A Paulus, ‘The Responsibility of Individuals for Human Rights Abuses in Internal Conflicts: A Positivist View’ (1999) 93(2) AJIL 302, 303–4. The latter authors recognize a second aspect of positivism as the separation between ‘law in force’ and ‘nonlegal factors’, approximating the analytical positivism described above.

⁷⁵ See eg I Brownlie, Principles of Public International Law (7th edn, Oxford University Press 2008) 4; S Besson, ‘Theorizing the Sources of International Law’ in S Besson and J Tasioulas (eds), The Philosophy of International Law (Oxford University Press 2010) 165–6.

⁷⁶ As to which see: Brownlie (n 75) 19–24; Lowe (n 74) 90–7.

⁷⁷ This point is highlighted, albeit as a criticism of the internal consistency of international legal positivists, in M Koskenniemi, From Apology to Utopia: The Structure of International Legal Argument, Reissue with Epilogue (Cambridge University Press 2005) 131–2.

⁷⁸ D'Aspremont (n 70) 21–4. Gardner draws a further distinction, between analytical positivists, who endorse positivism as conceptually true, and normative positivists, who endorse positivism because of perceived benefits from doing so. Nothing in my argument turns on this distinction. For clarity, I use ‘analytical positivism’ to refer to those who endorse the positivist thesis outlined above, regardless of reasons for doing so. Gardner (n 60) 204–5. Many of the arguments that I rejected in Section II would fall under Gardner's normative positivism.

⁷⁹ Strictly speaking positivism, as an analytical claim about the truth conditions of legal propositions, only has implications for adjudication where it is combined with a further normative claim that the task of judges is—at least in part—to find and apply the law. An analytical positivist might consistently be a normative anarchist, holding that judges and legal subjects should generally ignore the law. As Gardner observes, analytical positivism strictly construed is ‘normatively inert’. However, insofar as we are concerned with WTO adjudication, and with the approach and self-presentation of the WTO Appellate Body, we can largely discount this caveat. For this point generally: Gardner (n 60) 202–3.

⁸⁰ Raz distinguishes between a narrower sense of legal reasoning (‘reasoning about law’), which is limited to source-based considerations, and a wider sense (‘reasoning in accordance with law’), which looks to non-source-based considerations to fill gaps and resolve ambiguities: Raz (n 16) 332–3. cf Gardner (n 60) 215–17.

⁸¹ Whether its reasoning and conclusions can in fact be supported in these terms is a separate question: as Hudec has emphasized, there can be a gap between the the Appellate Body's self-presentation and what is going on beneath the surface: R Hudec, ‘GATT/WTO Constraints on National Regulation: Requiem for an “Aim and Effects” Test’ (1998) 32(3) The International Lawyer 619. For a more recent observation to this effect, J Trachtman, ‘WTO Trade and Environment Jurisprudence: Avoiding Environmental Catastrophe’ (2017) 58(2) HarvIntlLJ 273.

⁸² WT/DS8/AB/R Japan–Alcoholic Beverages, Appellate Body Report (4 October 1996) 15.

⁸³ See eg WT/DS26/AB/R EC–Hormones, Appellate Body Report, para 65; WT/DS103/AB/RW2 Canada–Dairy (2nd Article 21.5 Reference, New Zealand), Appellate Body Report, para 94. In other cases the AB has emphasized the importance of respecting those sovereign rights that States have not agreed to restrict, including most prominently their right to tax (DS87/AB/R Chile–Alcohol, Appellate Body Report, para 60; WT/DS108/AB/R US–FSC, Appellate Body Report, para 90; WT/DS108/AB/RW US–FSC (Art 21.5), Appellate Body Report, para 139, 148; WT/DS316/AB/R EC–Aircraft, Appellate Body Report, para 1130; WT/DS353/AB/R US–Aircraft, Appellate Body Report, para 811 and to choose regulatory goals (DS58/AB/R US–Shrimp, Appellate Body Report, para 87).

⁸⁴ On the importance of text in the AB's approach, see eg: WT/DS8/AB/R Japan–Alcoholic Beverages, Appellate Body Report, 17–18; WT/DS50/AB/R India–Patents (US), Appellate Body Report, paras 45–48; WT/DS58/AB/R, US–Shrimp, Appellate Body Report, para 114. Qureshi suggests that the AB's approach to subsequent practice similarly reflects a prominent concern for State consent: Qureshi (n 8) 40–5.

⁸⁵ On the AB's response to art 3.2, A Mitchell, ‘The Legal Basis for Using Principles in WTO Disputes’ (2007) 10(4) JIEL 795, 809.

⁸⁶ As noted above, this does not mean that the agreements are read independent from general international law. Indeed, in its first decision, the AB emphasized that WTO law ‘is not to be read in clinical isolation from public international law’: WT/DS2/AB/R United States–Gasoline, Report of the Appellate Body. In doctrinal terms, non-WTO law may be relevant under art 31.1 VCLT as context, or under art 31.3, as subsequent agreements or practice, or relevant rules of international law.

⁸⁷ Howse (n 32).

⁸⁸ In practice, the Appellate Body applies an informal system of precedent, and expects panels to follow previous opinions. Further, in the substance of its reasoning, as Venzke explains, it is frequently the meaning of previous Appellate Body opinions, more than the agreements that are the focus of argument: I Venzke, How Interpretation Makes International Law (Oxford University Press 2012) Ch IV. As noted earlier, this practice of informal precedent is among the United States’ current complaints about the Appellate Body. However criticism of the AB's ‘gap-filling’ is by no means a new phenomenon: Qureshi (n 8) 15.

⁸⁹ In the logic of such iterated games, R Axelrod The Evolution of Cooperation (Basic Books 1984).

⁹⁰ This point is well made in Howse (n 4).

⁹¹ As Henkin notes, the role of legal advisers, and the significant proportion of legally trained officials and politicians in non-legal roles in foreign offices and international organizations allows international law and courts to benefit from some of this socialization effect. L Henkin, How Nations Behave: Law and Foreign Policy (Council on Foreign Relations 1968) 61. However, the remoteness of international law from the educational experience and daily practice of most legal professionals means that this effect is substantially weaker, and can play little role as regards wider (non-trade) legitimacy constituencies.

⁹² JHH Weiler, ‘The Transformation of Europe’ (1991) 100(8) YaleLJ 2403.

⁹³ Further, the direct effect of WTO law and dispute settlement has been excluded in key jurisdictions, whether through legislative or judicial decision. See eg (United States) Uruguay Round Agreements Act 1994 sec. 102, 19 USC 3512; (European Union) Joined cases C-120/06 P and 121/06 P Fabbrica italiana accumulatori motocarri Montecchio SpA (FIAMM), Fedon & Figli and others v Council and Commission [2008] ECR I-6513.

⁹⁴ ADA Art 17.6.ii.

⁹⁵ WT/DS184/AB/R US–Hot Rolled Steel, Appellate Body Report, paras 57–62; WT/DS350/AB/R US–Continued Zeroing, Appellate Body Report, paras 267–273, 317. See also eg WT/DS322/AB/R US–Zeroing (Japan), Appellate Body Report, paras 188–189; WT/DS264/AB/R US–Softwood Lumber V, Appellate Body Report, paras 113–116; WT/DS344/AB/R US–Stainless Steel (Mexico), Appellate Body Report, para 136. The difficult relationship between art 17.6(ii) and the requirements of art 3.2 and the customary principles of interpretation in the VCLT, is highlighted in Qureshi (n 8) 321–4.

⁹⁶ Van Damme explains this by reference to the need for the Appellate Body to offer ‘judicial finality’, albeit expressing scepticism that there will always be a right answer, ‘given the complexities of language and context and changing circumstances, often unforeseen’: I Van Damme, ‘Treaty Interpretation by the WTO Appellate Body’ (2010) 21(3) EJIL 605, 610. Howse argues the AB's difficulties with zeroing reflect a departure from an otherwise commendable judicial minimalism: Howse (n 32) 71.

⁹⁷ Strictly speaking, adjudicators faced with legal indeterminacy might simply declare a null result. However, in the context of an adversarial system, the result of such a rule is that the respondent ‘wins’—the need for a legal decision is not avoided: Dworkin (n 17) 142–3. Such a pro-respondent presumption seems particularly inappropriate in the context of a dense multilateral regime like the WTO, with its complex balance of rights and obligations: Mitchell (n 6) 53. Further, such a rule does not eliminate indeterminacy in adjudication. Rather (as the art 17.6 cases illustrate), it simply changes the question, from what the law is, to whether it is clear what the law is.

⁹⁸ The point to take from this discussion is a relatively modest one: not that the indeterminacy of sources undermines legal positivism tout court, but simply that it renders it incompatible with the formal mandate, political position and judicial self-presentation of the Appellate Body, as outlined above.

⁹⁹ Van Damme explains the choice to refer to ‘customary principles’, rather than the VCLT, as reflecting the fact that not all WTO members are or can be parties to the VCLT. Van Damme (n 96) 608.

¹⁰⁰ The possibility that provisions that appear clear when examined textually become problematic once their purpose is considered is emphasized by Fuller in criticizing Hart's core/penumbra distinction: L Fuller, ‘Positivism and Fidelity to Law: A Reply to Professor Hart’ (1958) 71(4) HarvLRev 630, 662–5. cf N MacCormick, Legal Reasoning and Legal Theory (Clarendon Press 1979) Ch VIII.

¹⁰¹ Van Damme (n 96) 616–20. Further, as Van Damme observes, the three approaches mentioned in art 31.1 are by no means the only principles relevant to treaty interpretation: ibid, 621.

¹⁰² Qureshi (n 8) 24–6.

¹⁰³ For this point in respect of principles of interpretation generally, Van Damme (n 96) 616–17. Domestically the same point applies to the multiplicity of competing canons of interpretation: JR Macey, ‘Promoting Public-Regarding Legislation through Statutory Interpretation: An Interest Group Model’ (1986) 86 ColumLRev 223, 264. The recognition of competing interpretive approaches points towards what Dworkin labels ‘theoretical disagreement’, about the criteria for identifying law, which he argues is particularly challenging for positivism: Dworkin (n 17) 4–6.

¹⁰⁴ While my focus in this article is interpretation, object and purpose plays various roles in the VCLT, and its exact content may vary across these. See generally: DS Jonas and TN Saunders, ‘The Object and Purpose of a Treaty: Three Interpretive Methods’ (2010) 43 Vanderbild Journal of Transnational Law 565; J Klabbers, ‘Some Problems Regarding the Object and Purpose of Treaties’ (1997) 8 FYBIL 138; U Linderfalk, ‘On the Meaning of the ‘‘Object and Purpose’’ criterion in the Context of the Vienna Convention on the Law of Treaties, Article 19’ (2003) 72 NordicJIntlL 429.

¹⁰⁵ J Raz, Between Authority and Interpretation (Oxford University Press 2009) 285–8. This is not to say that anti-positivist views will have any greater interest in the ‘originalist’ intent of legal drafters. However, as discussed further below, anti-positivists can at least make identify a plausible role for object and purpose in legal interpretation. On the minimal role of legislative intention in anti-postivist views, R Dworkin, ‘Originalism and Fidelity’ in Justice in Robes (Harvard University Press 2006).

¹⁰⁶ This possibility is acknowledged, albeit with reservations, in Raz (n 105) 285, 291–4. The limited role accorded to preparatory work in art 32 VCLT suggests such positivist object and purpose should play a minimal role in international law.

¹⁰⁷ Howse (n 32) 70. For this reason, other sources sometimes used to identify object and purpose, including in particular negotiating history, may offer little assistance in the context of contested WTO provisions, a point the Appellate Body has acknowledged in the context of zeroing: WT/DS464/AR/R US–Washing Machines, Appellate Body Report, para 5.168.

¹⁰⁸ On the conflicting goals of the Safeguards Agreement: AO Sykes, ‘The Safeguards Mess: A Critique of WTO Jurisprudence’ (2003) 2(3) World Trade Review 261; K Jones, ‘The Safeguards Mess Revisited: The Fundamental Problem’, (2004), 3:1 World Trade Review 83; AO Sykes, ‘The ‘‘Safeguards Mess’’ Revisited – A Reply to Professor Jones’ (2004) 3(1) World Trade Review 93.

¹⁰⁹ WT/DS213/AB/R US–Carbon Steel, Appellate Body Report, para 73.

¹¹⁰ Greenwald (n 3) 117. On the challenges of identifying an object and purpose behind the trade remedies agreements: Qureshi (n 8) 276–94.

¹¹¹ As Van Damme observes, ‘Interpretation is about finding the intentions of the parties; that is undisputed. But this gives little or no answer to questions such as whose intention, what was intended, and at what time that intention matters.’ Van Damme (n 96) 618. For discussion of the extent to which enquiries into object and purpose are frequently quite far removed from seeking the actual objects of actual agents: V Crnic-Grotic, ‘Object and Purpose of Treaties in the Vienna Convention on the Law of Treaties’ (1997) 7 Asian Yearbook of International Law 141.

¹¹² For this point more generally, M Greenberg, ‘How Facts Make Law’ (2004) 10 Legal Theory 157, 184–5.

¹¹³ On the iterative nature of this inquiry, Jonas and Saunders (n 103) 581–2.

¹¹⁴ The point here is perhaps clearest in respect of object and purpose, but also applies—albeit less visibly—when seeking the ‘ordinary meaning’ of a text. Words can carry different meanings, and we know which is the relevant ‘ordinary meaning’ by asking—explicitly or implicitly—what would be a sensible meaning given what we know about the speaker, the context in which they are speaking, and their relevant goals. On the context-sensitive nature of ordinary meaning in respect of likeness, R Howse and D Regan, ‘The Product/Process Distinction – An Illusory Basis for Disciplining ‘Unilateralism’ in Trade Policy’ (2000) 11(2) EJIL 249, 260–2. cf U Linderfalk, ‘Is Treaty Interpretation an Art of a Science? International Law and Rational Decision Making’ (2015) 26(1) EJIL 169, 172–3; Greenberg (n 112) 175; Dworkin (n 105).

¹¹⁵ For a detailed discussion of the difficulties of attributing intentions to collective agents: R Ekins, The Nature of Legislative Intent (Oxford University Press 2016).

¹¹⁶ This is a point the Appellate Body has emphasized. See WT/DS62/AB/R EC–Computer Equipment, Appellate Body Report, paras 80–82. cf WT/DS464/AR/R US–Washing Machines, Appellate Body Report, para 5.168.

¹¹⁷ On the problems of attributing shared intentions across individuals with diverse purposes: Dworkin (n 17) 313–27; J Waldron, Law and Disagreement (Oxford University Press 1999) Ch 6. cf the discussion of political constraints, and their limits as guides, in Section IID.

¹¹⁸ See eg Howse (n 4) 98–101.

¹¹⁹ Although Regan notes that the attribution of negotiators’ goals to States will, to be normatively plausible, require an element of filtering, to ensure that ‘whatever we consider as a possible purpose of the agreement must be something that could in principle be regarded as the countries’ purpose’: Regan (n 10) 965.

¹²⁰ Howse (n 32).

¹²¹ For example, as Mavroidis observes, ‘the overwhelming majority of historical accounts [of the original GATT] focus on the US and UK negotiating positions’, which were substantially determinative of the outcome. If our goal was simply the explanatory one of understanding why the agreements are as they are, such a strategy might seem reasonable. However, given that the relevant object and purpose is to be attributed to the members as a whole, and to form a basis for interpreting each of their obligations, such a narrow focus seems unjustifiable. P Mavroidis, Trade in Goods (2nd edn, Oxford University Press 2012) 9.

¹²² See eg Regan (n 10). Integrating political economy concerns lead to somewhat different interpretations: K Bagwell and R Staiger, The Economics of the World Trading System (MIT Press 2002).

¹²³ This is the perspective adopted by contemporary exponents of the embedded liberalism perspective. eg Howse and Nicolaidis (n 11); Howse (n 32) 44–5.

¹²⁴ There are elements of this agenda in Rawls’ characterization of States as concerned to secure their ‘proper self-respect of themselves as a people’: J Rawls, The Law of Peoples (Harvard University Press 1999) 34. For an example of this approach at the level of persons: J Tomasi, Free Market Fairness (Princeton University Press 2012).

¹²⁵ The priority of relative gains for positionalists is highlighted in realist scholarship on international cooperation. eg J Mearsheimer, The Tragedy of Great Power Politics, Updated Edition (WW Norton & Co 2014) 51–3.

¹²⁶ K Waltz, Theory of International Politics (McGraw-Hill 1979) 5–6.

¹²⁷ The moral imperative towards national self-interest is expressed in much realist thought. See eg GF Kennan, ‘Morality and Foreign Policy’ (1985) 64(2) Foreign Affairs 205.

¹²⁸ For this approach generally: J Rawls, ‘Justice as Fairness: Political not Metaphysical’ (1985) 14(3) Philosophy & Public Affairs 223. For its application in the international context: Rawls (n 124) passim.

¹²⁹ See generally Dworkin (n 17). The approach sketched in these subsequent sections draws on Dworkin's insights and methods. However, it is not strictly Dworkin's approach, both because the specific version of interpretivism that Dworkin advances, law as integrity, is itself a poor fit in the international context, and because Dworkin himself in a posthumously published paper advanced a somewhat different theory of international law: R Dworkin, ‘A New Philosophy of International Law’ (2013) 41(1) Philosophy and Public Affairs 2. My goal in this article is to illuminate WTO law, rather than to engage in Dworkinian exegesis, so I will set aside these points for another day.

¹³⁰ On the ways concerns for formal justice motivate such concerns: MacCormick (n 100) Ch IV.

¹³¹ See generally Raz (n 17) 307–9.

¹³² For a positivist model of reasoning along these lines, including in particular the distinction between rule-bound and discretionary/moral/consequential reasoning, MacCormick (n 100).

¹³³ cf Fuller (n 100).

¹³⁴ It is thus not the case that we can simplistically interpret the law to be ‘more moral’. Rather, as Greenberg observes, there are good moral reasons why one might want to respect the existing practices through which a society makes law, and the laws that it has sought to make. Greenberg (n 112) 193ff. Contrast, from a positivist perspective: Raz (n 16) 332–3.

¹³⁵ For an effort to read WTO law from a strongly egalitarian perspective: F Garcia, Trade, Inequality and Justice: Towards a Liberal Theory of Just Trade Law (Brill 2003).

¹³⁶ See eg D Moellendorf, ‘The World Trade Organization and Egalitarian Justice’ (2005) 36(1-2) Metaphilosophy 145; T Broude, ‘The Rule(s) of Trade and the Rhetos of Development: Reflections on the Functional and Aspirational Legitimacy of the WTO’ (2006) 45 Columbia Journal of Transnational Law 221.

¹³⁷ While not strictly egoist, views that recognize minimal anti-poverty duties will also struggle here.

¹³⁸ A prominent example of such a view is Nagel (n 32). cf Kennan (127).

¹³⁹ Given the prominence of such egoist assumptions, and their relation to the international legal positivism which the AB has endorsed, it is perhaps unsurprising that the AB commonly invokes object and purpose to motivate such anti-avoidance interpretations. See eg WT/DS98 Korea–Dairy, Appellate Body Report, para 111; WT/DS139 Canada–Autos, Appellate Body Report, para 142; WT/DS257 US–Softwood Lumber IV, Appellate Body Report, para 64; WT/DS320 US–Continued Suspension, Appellate Body Report, para 308.

¹⁴⁰ The concern here relates to the interpretive contribution, as opposed to the legitimizing capacity, of State consent. We need not deny that State consent provides an adequate justification for WTO law to identify its defects as an interpretive guide. Contrast: E-U Petersmann, ‘Between ‘‘Member-Driven’’ WTO Governance and ‘‘Constitutional Justice’’: Judicial Dilemmas in GATT/WTO Dispute Settlement’ (2018), 21 JIEL 103, 114; ‘Why Treaty Interpretation and Adjudication Require ‘‘Constitutional Mind-Sets’’’ (2016) 19 JIEL 389.

¹⁴¹ Regan (n 10). Embedded liberalism explanations are similarly built on broadly egoistic premises, but emphasizing concerns for regulatory sovereignty and social protection over wealth maximization.

¹⁴² On the weaknesses of economic accounts of the trade remedies disciplines: Suttle (n 13) Ch 8. cf Howse (n 32) 69–70.

¹⁴³ JM Finger, ‘The Origins and Evolution of Antidumping Regulation’ in JM Finger and NT Artis (eds), Antidumping: How it Works and Who Gets Hurt (University of Michigan Press 1993).

¹⁴⁴ See eg WT/DS27/AB/R, EC–Bananas, Appellate Body Report, para 190; WT/DS56/AB/R, Argentina–Textiles and Apparel, Appellate Body Report, para 47; WT/DS75/AB/R, Korea–Alcohol, Appellate Body Report, para 432ff. WT/DS139/AB/R, Canada–Autos, Appellate Body Report, para 84. This willingness to engage in purposive reasoning has not prevented the Appellate Body recognizing the need to balance competing interests—a principled interpretation can also recognize its own limits. See eg WT/DS58/AB/R US–Shrimp, Appellate Body Report, para 116; WT/DS27/AB/RW2/ECU EC–Bananas (Ecuador, 2nd Art 21.5 Reference), Appellate Body Report, paras 431–435; WT/DS62/AB/R EC–Computer Equipment, Appellate Body Report, paras 80–84; WT/DS103/AD/R Canada–Dairy, Appellate Body Report, para 137; WT/DS276/AB/R Canada–Wheat Exports, Appellate Body Report, paras 95–97.

¹⁴⁵ WT/DS400/AB/R EC–Seal Products, Appellate Body Report, para 5.129; WT/DS406/AB/R US–Clove Cigarettes, Appellate Body Report, passim, particularly paras 84–102. Object and purpose also plays a significant role in a number of key domestic regulation cases, motivating reflection on the balancing of competing values, including trade liberalization, human health, transparency and regulatory autonomy. eg WT/DS26/AB/R EC–Hormones, Appellate Body Report, paras 177, 189–190, 205–206. cf WT/DS231/AB/R, EC–Sardines, Appellate Body Report paras 214–215; WT/DS381/AB/R US–Tuna II, Appellate Body Report, para 379.

¹⁴⁶ WT/DS246/AB/R EC–Tariffs Preferences, Appellate Body Report, paras 91–97, 152–176 (recognizing the competing goals of promoting developing country trade, eliminating discrimination and rationalizing preference schemes, but disagreeing with the panel on the appropriate balance between them).

¹⁴⁷ This point is clear from efforts by the Appellate Body to articulate the object and purpose of the relevant agreements, which rarely extends beyond ‘having rules about the relevant issue’. See eg WT/DS473/AB/R EU–Biodiesel, Appellate Body Report, para 6.25; WT/DS213/AB/R US–Carbon Steel, Appellate Body Report, para 73; WT/DS257/AB/R US–Softwood Lumber IV, Appellate Body Report, para 95. cf WT/DS46/AB/R Brazil–Aircraft, Appellate Body Report, para 173. Although contrast WT/DS202/AB/R, US–Line Pipe, Appellate Body Report, paras 82–83. The Appellate Body has invoked object and purpose on a number of occasions to emphasize the differing standards applicable to remedies for ‘fair’ (safeguards) and ‘unfair’ (subsidies and dumping) trade. A different balance is, we are told, expressed in the different agreements. However, unlike the balancing of competing objectives under the GATT and Regulation Agreements, we get little guidance on what exactly the values being balanced are, or how the relevant balance is to be struck. See eg WT/DS98/AB/R Korea–Dairy, Appellate Body, paras 87–88; WT/DS121/AB/R Argentina–Footwear, Appellate Body Report, paras 93–95; WT/DS178/AB/R US–Lamb, Appellate Body Report, para 124. cf WT/DS202/AB/R US–Line Pipe, Appellate Body Report, para 257; WT/DS296/AB/R US–Countervailing Duty Investigation on DRAMs, Appellate Body Report, para 115; WT/DS399/AB/R US–Tyres (China), Appellate Body Report, para 183–5.

¹⁴⁸ See eg WT/DS234/AB/R United States – Offset Act (Byrd Amendment), paras 281–294; WT/DS213 US–Carbon Steel, Appellate Body Report, para 83.

¹⁴⁹ This formalism extends to the Appellate Body's identification of parallels between the trade remedies agreements, reflecting their use of similar techniques (injury, causation), without serious reflection on the different purposes to which these techniques are put. See eg WT/DS184 US—Hot Rolled Steel, Appellate Body Report, paras 229–230; WT/DS202 US–Line Pipe, Appellate Body Report, paras 212–214; WT/DS414 China–GOES, Appellate Body Report, 133–154, esp 153.

¹⁵⁰ WT/DS121/AB/R Argentina–Footwear, Appellate Body Report, paras 91–98, 131. For the problems this interpretation poses, Sykes (n 102). For the challenge that it is unmotivated: Greenwald (n 3).

¹⁵¹ For discussion, Alford, RP, ‘Reflections on US–Zeroing: A Study in Judicial Overreaching by the WTO Appellate Body’ (2006) 45 Columbia Journal of Transnational Law 196Google Scholar; Voon, T, ‘The End of Zeroing? Reflections Following the WTO Appellate Body's Latest Missive’ (2007) 34(3) Legal Issues in Economic Integration 211Google Scholar; Bown, C and Sykes, AO ‘The Zeroing Issue: A Critical Analysis of Softwood V’ in Horn, H and Mavroidis, P (eds), The WTO Case Law of 2004–5: Legal and Economic Analysis (Cambridge University Press (2008)Google Scholar.

¹⁵² While the Appellate Body has placed weight on the requirement of fair comparison, it has failed plausibly to articulate what is unfair about zeroing. Zeroing, it has argued, leads to unfair inflated margins. WT/DS264/AB/RW US–Softwood Lumber V (Art 21.5), Appellate Body Report, paras 138–42. Yet, as United States argued and the panel agreed in that case, this argument is question-begging unless accompanied by an explanation of why one standard or another should have priority. The Appellate Body only avoids this objection by invoking the incompatibility of zeroing with other provisions of the agreement, effectively draining the fair comparison requirement of any independent content. ibid, paras 143–146.

¹⁵³ On the limits of economic approaches in guiding trade remedies adjudication: Suttle (n 13) Ch 8. It is worth noting that the specific examples cited by the United States in criticizing the Appellate Body's interpretive approach are almost universally drawn from trade remedies cases. See USTR (n 1).

¹⁵⁴ See in particular Suttle (n 13).

¹⁵⁵ See, for a promising development in this direction, Tomasi (n 124). In other ongoing work I seek to develop such a view with specific reference to international economic law: ‘The Puzzle of Competitive Fairness’, unpublished manuscript on file with the author. Views emphasizing economic liberty rights might also fit well here. See eg Petersmann, E-U, ‘Time for a United Nations “Global Compact” for Integrating Human Rights into the Law of Worldwide Organizations: Lessons from European Integration’ (2002) 13(3) EJIL 621CrossRefGoogle Scholar.