1. Introduction
The Seals case concerns an EU ban on placing seal and seal products on the market. Three exceptions to the ban apply: (i) seal products from hunts traditionally conducted by Inuit and other indigenous communities in a region with a tradition of seal hunting, contributing to their subsistence and being partially used, consumed, or processed within the community according to their traditions can be placed on the EU market (IC exception) (ii) travelers may import small consignments of seal products for personal use, and (iii) seal products resulting from hunts for the sole purpose of sustainable marine resources management may be placed on the market on a non-profit, non-systematic basis (MRM exception).Footnote 1
Most Canadian seal products are commercially hunted and are therefore excluded from the EU market. Canada's large Inuit population also hunts seals but has not exported seals to the EU under the IC exception. Most seal products from Greenland are hunted by Inuit and distributed and exported by their own processing facilities on a large scale. Most Norwegian commercial seals are hunted according to a marine resources management plan but remain excluded because they are sold to make a profit and because hunts are conducted systematically. A small number of European seals are placed on the market under the MRM exception.
At the panel stage, the main legal issues were whether or not the TBT Agreement applies to the Seals Regime and whether or not it complies with its Article 2.1, and in particular whether or not the difference in treatment of seals hunted commercially, by Inuit communities and for marine protection purposes stemmed exclusively from a legitimate regulatory distinction and was applied even handedly. Under Article 2.2, the legitimacy of the objectives, the number of objectives pursued, and the necessity of the Seals Regime were contested. More particularly, the issues as to whether the seal welfare and consumer ethics objectives were legitimate objectives, whether the purposes of the exceptions set forth additional objectives, and how the ban and exceptions together contributed to fulfilling the legitimate objective weighed against their trade-restrictiveness, taking account of the risk non-fulfillment of the objective would create. The complainants also proposed less trade-restrictive alternatives to the ban on commercially hunted seal and seal products in the form of animal welfare regulation, certification, and labels. Claims in respect of certification were also made under Articles 5.1.2 and 5.2.1 of the TBT Agreement. The main legal issues under the GATT were whether or not the Seals Regime violates Articles I:1, III:4, and XI:1, and whether or not such violation can be justified under Article XX. The complainants also made claims under Article 4.1 of the Agreement on Agriculture and the non-violation remedy. The appellants and appellee appealed panel findings on the applicability of the TBT Agreement and several findings under Articles 2.1 and 2.2. Since the Appellate Body reversed the panel's findings on the applicability of the TBT Agreement and refused to complete the analysis on this point, it did not review any of the panel findings under Articles 2.1, 2.2, 5.1.2, and 5.2.1, but instead declared them moot. In relation to the GATT, the parties appealed findings of a violation of Article I:1, the rejection of the legitimate regulatory purpose test of the TBT Agreement under GATT Articles I:1 and III:4, and Article XX findings on necessity of the Seals Regime for the protection of public morals and under the chapeau.
The case throws up interesting questions concerning the interaction of animal welfare concerns and the ethical-expressive preferences of European citizens against animal cruelty,Footnote 2 the interpretation and applicability of key provisions in the Agreement on Technical Barriers to Trade (TBT) and the General Agreement on Tariffs and Trade (GATT), the concept of public morals and the ability of WTO members to reconcile the legal obligations they have stemming from the functionally differentiated sub-systems of WTO law, international human rights law on protection of indigenous communities and minorities and emergent international law principles on animal welfare.Footnote 3
This contribution focuses on the TBT aspects of the case. The main focus is on how public morals objectives of multipurpose technical regulations can be justified under Articles 2.1 and 2.2 of the TBT Agreement. The prior question of how detrimental impacts on like products are established will not be addressed. The article first analyses the question of whether the TBT Agreement applies to the Seals Regime or whether, instead, only the GATT applies. It is argued that measures that apply to production or processing methods (PPMs) that leave no physical traces in a product can nevertheless be considered to be technical regulations as long as they pertain to production or processing aspects with an inherent relation to the product. The article then reviews the panel's non-use of other international law in its Article 2.1 analysis and discusses the distinction between the legal tests in Articles 2.1 and 2.2. Subsequently, the article examines how the concepts of ‘public morals’ and ‘legitimate objectives’ were interpreted in the case. It criticizes the heavy weight the panel attributed to public opinion in determining the content of public morals. Next, the article evaluates how the panel examined a regulatory measure embodying complex cost–benefit analysis under Articles 2.1 and 2.2. Drawing on normative political theory, Section 4 then proposes an alternative interpretation of the term ‘public morals’ that differentiates ethical-expressive values from deontological morality and explains why the former should not normally be considered to fall under the term ‘public morals’.
2. Summary of relevant findings
2.1 The analysis of whether the Seals Regime is a technical regulation
The panel found that the Seals Regime with the ban and the exceptions considered together was a technical regulation.Footnote 4 It found that the prohibition on seal products laid down product requirements in the negative form (products must not contain seal).Footnote 5 It also drew on the Appellate Body's determination in EC–Asbestos that the chrysotile exception set down mandatory provisions for products with certain objective characteristics and thus fell within the definition of a technical regulation.Footnote 6 The panel determined that the criteria of the IC, MRM, and travelers exceptions also laid down objectively definable features of products (products must meet the terms of the exceptions) and, for this reason, set down product characteristics.Footnote 7
The Appellate Body reversed the panel's finding that the Seals Regime constituted a technical regulation.Footnote 8 It faulted the panel for not having carried out a holistic assessment of the weight and relevance of all the components of the measure and for having given the most weight to the fact that the Seals Regime laid down product requirements in the negative form.Footnote 9 In the Appellate Body's assessment, the panel failed to examine the integral and essential characteristics of the measure. The panel did not take proper account of the fact that the Seals Regime also contained a prohibition on seal as such and created exceptions depending on the way seals were hunted and therefore did not prescribe product characteristics.Footnote 10 The Appellate Body also considered that the applicable administrative provisions of the Seals Regime pertained to the way seals were hunted rather than laying down product characteristics.Footnote 11 It refrained from determining whether or not the exceptions set out PPMs related to product characteristics out of concern for due process as the parties had only made limited arguments on this point.Footnote 12 As a result of having reversed the panel's finding that the Seals Regime was a technical regulation, the Appellate Body also declared moot all other legal findings under the TBT Agreement of the panel.Footnote 13
2.2 The panel's analysis of Article 2.1 TBT
The panel followed the analytical approach to Article 2.1 of prior TBT cases by examining whether any detrimental impact on foreign products stemmed exclusively from a legitimate regulatory distinction.Footnote 14 The likeness of commercial, IC, and MRM seal was not contested.Footnote 15 As the majority of Canadian products were banned while virtually all Greenlandic products qualified under the IC exception and an absolutely small number of products representing the bulk of the EU's total seal production qualified under the MRM exception, the panel found a detrimental impact.Footnote 16
Turning to the facts and the legitimacy of the regulatory distinction between commercial, IC, and MRM seals, the panel first reviewed the conditions for seal hunting in general and found certain inherent animal welfare risks.Footnote 17 Next, the panel established enhanced animal welfare risks in commercial hunts due to the competitive pressure to kill more seals.Footnote 18 Having reviewed the characteristics of IC hunts, the panel also found seal welfare concerns to be present and a lack of rational connection with the principal animal welfare objective of the Seals Regime.Footnote 19
Subsequently, the panel analyzed whether commercial and IC hunts differed in their purpose, and whether this difference justified the regulatory distinction between the products.Footnote 20 Although the panel saw a resemblance to the purpose of commercial hunts when IC seals were sold on the market to generate family income, it considered that the subsistence purpose of personal consumption and the link to IC traditions constituted a difference in purpose.Footnote 21 It found that the objective of protecting IC traditions justified the regulatory distinction drawn by the IC exception.Footnote 22 However, the EU Seals Regime was not applied in an even-handed manner. The panel suggested that the Greenlandic IC hunts exhibited commercial scale activities but benefited from the exception while for Canadian Inuit, who had relied on the commercial sealing industry for distribution and hunted a low number of seals, accessing the exception through development of their own distribution channels was not cost effective.Footnote 23
The panel conducted its analysis of the MRM exception along similar lines as the IC exception. According to the EU, the purpose of the MRM exception was protection of local fish stock from nuisance seals and population control if seals posed ecosystem threats. The panel found an absence of a connection to the seal welfare purpose of the Seals Regime inter alia due to relaxed requirements on hunting methods, and considered the commercial motives of the hunt (protection of fisheries) similar to the motives of commercial hunts.Footnote 24 It rejected the even-handedness of application, because of evidence from the legislative history that the MRM exception was designed to fit European MRM hunts.Footnote 25 The panel's overall conclusion on the Seals Regime was that it violated Article 2.1 of the TBT Agreement.Footnote 26
2.3 The panel's analysis of Article 2.2 TBT
The EU argued that the purpose of its seals regulation was to protect public morals related to seal welfare (reduction in number of animals killed inhumanely) and to prevent the participation of consumers in sustaining the market in seals and seal products.Footnote 27
The panel asked itself whether there was evidence of EU public opinion of a concern with seal welfare and their participation as consumers in sustaining trade in seals and seal products.Footnote 28 Then it looked at whether the public's concern with seal welfare was described as an issue of public morals in EU legislative instruments.Footnote 29 It affirmed both and therefore concluded that that the public's concerns about seal welfare constitute a moral issue for EU citizens.Footnote 30 Given that both GATT Article XX (a) and GATS Article XX (a) mention public morals, the panel concluded that public morals were a legitimate objective within Article 2.2 of the TBT Agreement.Footnote 31 The panel then assessed whether addressing public morals concerns on seal welfare was a legitimate objective (LO). It made a positive conclusion because the concerns of the EU public had been demonstrated to involve standards of right and wrong and because animal welfare is a globally recognized issue in other national and international legal instruments.Footnote 32
The panel considered that the objective of the Seals Regime was to protect public morals concerns in respect of seal welfare and rejected the argument that the IC, MRM, and travelers exceptions added further objectives.Footnote 33 It drew a distinction between IC, MRM, and travelers' interests accommodated in a measure and the main objective of the measure as a whole.Footnote 34 It did so because the Seals Regime was formulated as a ban with exceptions operating as a prohibition against seal products subject to the exceptions, because the legislative history referred only to ensuring that the interests of Inuit were not affected and because the interests addressed in the exceptions were not reflected in the concerns of EU citizens.Footnote 35
The panel determined the measure was trade-restrictive.Footnote 36 It then examined the degree of contribution of the ban to animal welfare.Footnote 37 It found that the exceptions reduced the contribution towards fulfilling the LO because they allowed seals to be killed inhumanely, while EU public concerns did not distinguish between the purposes of the hunt in their rejection of seal hunting.Footnote 38 The panel concluded that the measure made only some contribution to its stated objective.Footnote 39 In its evaluation of the risk non-fulfillment created, the panel reiterated the measure's limits to protecting seal welfare.Footnote 40 The panel rejected the complainants' proposed less restrictive alternatives of market access on the basis of labels and certification of seals and seal products to humane killing methods.Footnote 41 As a result, the panel concluded that the EU seal regime was not more restrictive than necessary to fulfill legitimate objectives.Footnote 42
3. Analysis of the main findings
3.1 Is the Seals Regime a technical regulation?
The EU's Seals Regime is essentially a ban on seal and any product containing seal subject to the IC and MRM exceptions, which regulate how seals are hunted.Footnote 43 The issue arose whether a measure is a technical regulation as soon as one of its elements meets the definition or whether the nature of the whole measure must be such that it can be classified as a technical regulation. The other question was whether the IC and MRM exceptions, which did not affect or change the physical characteristics of the seal products and were therefore unincorporated PPMs, set down product characteristics or related PPMs such as to fall under the definition of a technical regulation.Footnote 44
Concerning the first issue, it would be problematic if a multi-element measure could be considered as a technical regulation as soon as one of its elements meets the definition of a technical regulation because this element could be relatively unimportant within the overall regulatory scheme. The result would be that multi-element measures would have to be justified under the TBT Agreement not designed to regulate non-technical regulations in the first place. The Appellate Body's approach arguably gives more weight to all the elements of the measure and understands that the essence of the Seals Regime is a PPM related to the way seals are hunted (commercial vs. IC or MRM seals).Footnote 45
On the second issue, the panel had found that the terms of the exceptions set down objectively definable features of products containing seal, and therefore laid down product characteristics. This interpretation empties the term ‘related PPMs’ in the definition of a technical regulation of independent legal meaning because any PPM, whether related or unrelated to product characteristics, would do so.Footnote 46 Such a result goes against the principle of effective treaty interpretation according to which all the terms in a treaty must be given meaning.Footnote 47 For example, even a trade sanction imposed because WTO member X is non-democratic puts forth objectively definable features of products because it stipulates: ‘a product must not come from country X’ and would thus lay down product characteristics. Arguably, such a measure is not dealing with product characteristics but with the general political situation in the country of export because no change in the physical characteristics of the product could ever confer access to the market of the WTO member imposing the sanction. The fact that the WTO members included related PPMs in the definition of a technical regulation suggests that there must also be some category of unrelated PPMs that are excluded from the definition of a technical regulation, yet under the panel's approach such a category would have ceased to exist.
Are the exceptions of the Seals Regime excluded from the definition of a technical regulation because they contain an unincorporated PPM not affecting product composition? Howse and Levy have suggested that only PPMs unrelated to traded goods (such as PPMs applicable to trade in services) were meant to be excluded from the scope of the TBT Agreement.Footnote 48 Their broad view ignores that the term ‘their related PPMs’ does not refer simply to products but rather to product characteristics. Of course, any product has the characteristic of being tangible but this feature is arguably inherent in the term ‘product’ itself and would not require the addition of the term ‘characteristics’.
I submit that the term ‘their related PPMs’ refers to incorporated or unincorporated PPMs that relate to inherent characteristics of the product. Some unincorporated PPMs can in my view relate to product characteristics. For example, assume that there is no way to farm seals for commercial exploitation. The product seal would have the inherent characteristic of living and having to be hunted in the wild. Any PPM that allows seals hunted by Inuit also prevalent in this area but bans others is an unincorporated PPM that relates to this product characteristic. Two considerations support my interpretation. First, the TBT Agreement refers to product characteristics and not to the physical properties of the product. Second, the Appellate Body in EC–Asbestos referred to product characteristics as ‘objectively definable features … intrinsic to the product itself’.Footnote 49 In the example given, it is an objectively definable feature of the product seal that the animal lives and has to be hunted in the wild in regions where Inuit also live because this feature has an inherent link with the seal itself. In EC–Seals, the Appellate Body noted that a PPM that is connected with or has a relation to any such intrinsic or objectively definable feature relates to product characteristics.Footnote 50 In the example given, requirements pertaining to the way seals are hunted relate to an intrinsic aspect of the product and are therefore ‘related PPMs’. If my argument on related PPMs is correct, then the balance may well be tipped in favor of the definition of a technical regulation being met.
3.2 The panel's non-use of other international law in Article 2.1
Evaluation of the panel's non-use of other international law as context for the interpretation of Article 2.1 is important for the later argument that terms such as ‘public morals’ and ‘legitimate’ need to be given some normative – or moral – content. International human rights covenants are arguably a ‘positivization’ of international standards of morality.
It is submitted that the panel should have drawn on international human rights documents protecting indigenous people and minorities as context in its interpretation of Article 2.1.Footnote 51 Indigenous rights are protected in the ILO Indigenous and Tribal Peoples Convention of 1989, the non-binding Declaration on the Rights of Indigenous People, but more importantly also in the form of minority rights to enjoy their own culture in Article 27 of the International Covenant on Civil and Political Rights (ICCPR) and Articles 11 and 15 of the International Covenant on Economic, Social and Cultural Rights (ICESCR) as well as in interpretations of these provisions.Footnote 52
WTO law should be interpreted in accordance with Article 31 of the Vienna Convention on the Law of Treaties, which requires the treaty interpreter to take into account relevant rules of international law applicable in the relations between the parties in interpreting the treaty provision at issue.Footnote 53 The ICCPR and ICESCR enjoy near universal support and bind almost all WTO members.Footnote 54 In EC–Aircraft, the Appellate Body has relaxed the conditions under which other international law documents not strictly binding on all WTO members can be taken into account, and effectively overruled the panel in EC–Biotech.Footnote 55 It considered that a balance had to be struck between ascertaining the common intentions of all WTO members and the need for systemic unity of international law in deciding which international legal instrument to have recourse to for interpreting WTO law.Footnote 56 This obiter dictum suggests that there is no strict rule of parity of membership before a non-WTO international legal instrument can be used for a contextual interpretation. Because of the near universal support the ICCPR and ICESCR enjoy, it is suggested they would be an appropriate instrument to take account of in interpreting the TBT Agreement.
According to Beqiraj's analysis, the Human Rights Committee considered that Article 27 of the ICCPR extended to economic activities, including ways of life associated with the use of land, even if modern technologies are used.Footnote 57 It also affirmed that states are under a duty to ensure that the cumulative effects of measures are not such as to erode the right in Article 27.Footnote 58 The Sami population in Sweden is a minority with a seal hunting tradition. Sweden therefore has a human rights obligation vis-à-vis its minority to ensure their culture and culture-related economic activities are protected. Given that Sweden is so obliged and that all other EU Member States are bound by the ICCPR, it is not so far-fetched to argue that it was legitimate in the sense of Article 2.1 for the EU to take account of the human rights obligations of its Member States in drawing up its regulatory policy on seals. Pursuant to the requirement of even-handedness in the application of the measure, the same exceptions would have to be extended to similarly situated products of indigenous communities outside of the EU.
In addition to Article 2.1, international human rights documents are also arguably relevant to interpreting the content of public morals since they express normative commitments of the international community to concepts such as dignity and justice. As a result, they are also relevant to the question of whether or not the objectives of a technical regulation are legitimate. Subsections D, E, and Section 4 indirectly show why such documents would be relevant.
3.3 The relationship between Article 2.1 and 2.2
Articles 2.1 and 2.2 both aim at verifying whether or not a trade-affecting technical regulation is legitimate. In the way they have been interpreted, Article 2.1 contains an inquiry into the rational connection of the policy to a regulatory objective and the way it is applied but not an inquiry into less trade-restrictive alternatives, and Article 2.2 sets out a necessity test, including whether less trade-restrictive alternatives are available.Footnote 59
This sequence of analysis has been criticized by Mavroidis who argues that the panel should examine necessity under Article 2.2 first (similar to the paragraphs of GATT Article XX) and then whether the manner of application is non-discriminatory under Article 2.1 (similar to the chapeau analysis of Article XX).Footnote 60 Indeed, why should a panel even consider the even-handedness in application of a measure under Article 2.1 before it has satisfied itself that no less trade-restrictive but equally effective alternative is available, which might have to replace the contested measure?
Note that the panel considered the distinction between commercial and IC hunts to be legitimate in principle under Article 2.1 and yet found that this legitimate regulatory distinction reduced the Seals Regime's contribution to the protection of public morals concerns on seal welfare under Article 2.2. How the substantive aspects of a regulatory distinction can be legitimate under one rational connection test and yet count against the justifiability of the regulation under another rational connection test escapes me. This aspect will be returned to in Subsection 3.5.
3.4 Public morals on seal welfare as legitimate objectives under Article 2.2
The panel's approach towards public morals as LOs is problematic for conceptual and technical legal reasons. At the outset, it is useful to recall the panel's definition of public morals in US–Gambling as ‘standards of right or wrong conduct maintained by or on behalf of a community or a nation’, which can vary depending, inter alia, on prevailing values.Footnote 61 The Gambling panel considered that WTO members should be given some discretion to define and apply this concept in accordance with their own system and scales of values.Footnote 62 For most philosophers, bar moral relativists, ‘morality’ expresses the notion of a regulative idea, of a norm with which human behavior must comply even if people actually have other preferences and opinions.Footnote 63
Recall how the panel established that public seal welfare concerns were about public morals. It asked itself: Do EU citizens say that killing seals is bad? Do EU law and the legislative process state that EU citizens consider the killing of seals to be bad? Both elements (what people say and what the law and the process of its making say they say) are empirical facts. EU public opinion is an empirical fact of what people think is good or bad.
The conceptual problem is that empirical facts alone cannot ground moral obligations because the obligations are supposed to regulate the facts. We would end in a tautology where the facts auto-justify themselves. To clarify: the mere fact that EU citizens think something does not make it right for them to think this. And the mere fact that the EU legislator confirms that EU citizens really do think this does not make it right for them to think this either. The panel's approach, by giving so much weight to the evidence on public opinion adduced by the EU, emptied the terms ‘morals’ and ‘right or wrong conduct’ of prescriptive normative content. It treated the terms as if they meant ‘in accordance with expressed moral feelings of the public that the technical regulation also describes as moral feelings'.
Let us now look at how the panel analyzed whether public concerns on seal welfare were a LO. To aid the reader in understanding where the analysis is going, it is useful to point out that ‘legitimate’ must express a normative judgment of the type ‘it is right or just to do Y’.Footnote 64 What is legitimate cannot be grounded entirely in an empirical fact. For instance, what is legitimate cannot be equated with what the positive law requires (or some other fact like public opinion) because we could continue asking why it is legitimate of the law to demand a certain course of action. Obviously, the reply that it is legitimate for the law to demand so because it is the law according to the criterion the community applies for identifying what is law is no satisfactory answer to this question. To make the claim for legitimate authority successfully, law must give us a reason for action that lies at least partly outside of law in domains like justice and morality. I am thus committed to a non-positivistic theory of law the upshot of which is that legal terms such as ‘legitimate’ directly implicating the justification of law's authority should be interpreted in such a way as to give meaning to the non-legal, moral dimensions that are necessary to justify that very authority.
What I want to show is that the panel's treatment of the term ‘legitimate objectives’ together with its treatment of the term ‘public morals’ fails to give any weight to the necessary normative dimension of the term ‘legitimate’ because both remain limited to empirical facts in the panel's approach. To bring out this conclusion, the discussion of the two aspects of the case (LO and content of public morals) is interwoven.
With this as background, here is how the panel proceeded. It essentially asked itself: Does WTO law make reference to public morals? Answer: ‘yes’. Is a technical regulation with public morals as its objective legitimate? Answer: ‘yes’. Next, concerning the content of public morals: Do EU citizens express the opinion that killing seals is bad? Does EU law state that EU citizens think that killing seals is bad? Answer: ‘yes’ to both. Is the Seals Regime therefore about public morals? Answer: ‘yes’. Do other national and international legislative instruments address animal welfare concerns, and is the EU's seal welfare concern therefore legitimate because it is about public morals? Answer: ‘yes’. The panel thus attached decisive importance to two empirical facts: public opinion and the positive law (EU, TBT, GATT, GATS, other national and international legal instruments) for qualifying seal welfare as being about legitimate public morals objectives.
I submit that by doing so, the panel never properly asked itself whether or not protection of the welfare of seals and ethical consumer behavior was legitimate. Let me explain why the panel's approach is problematic conceptually.
If the answer to the question ‘is seal welfare legitimate?’ is ‘yes’, because EU law confirms that EU citizens think it is about good morals and EU citizens actually say they think so,Footnote 65 this just begs the further question why it is legitimate for the law to say and EU citizens to think so. If the answer to that question is ‘it is legitimate because it is also recognized as a concern in other laws’,Footnote 66 this again just begs the question in respect of these other laws. In other words, if we let empirical facts define legitimacy, we quickly end up begging the question.
One might be tempted to reply that animal welfare is legitimate because it is about public morals, which is in turn about standards of right or wrong and that the panel had said so.Footnote 67 And if something is right, surely it must also be legitimate. Recall at this point how the panel determined that seal welfare was about public morals. It asked itself: do EU citizens think about seal welfare as good morals and does EU law confirm that EU citizens think this way? But this again is just an empirical fact. In other words, we are back at square 1 and end up going in circles. So the panel commits a logical and a conceptual fallacy.
It is useful in this context to examine how the Appellate Body has defined legitimacy to get a sense of the origin of the problem in the case law. In US–Tuna II, it quoted the Shorter Oxford English Dictionary definition of ‘lawful; justifiable; proper’.Footnote 68 It then went on to state that ‘this suggests that a “legitimate objective” is an aim or target that is lawful, justifiable or proper’ [my emphasis].Footnote 69 Note the Appellate Body's interpretative slippage. It treats the terms ‘lawful; justifiable; proper’ as alternatives by inserting the word ‘or’ even though those terms are joined by semicolons, whose grammatical function is to link elements that are interdependent in the sense that they depend on each other for complete meaning.Footnote 70 As the Shorter Oxford English Dictionary evidences, the ordinary meaning of the term ‘legitimate’ goes beyond lawfulness and law as an empirical fact and points towards dimensions such as justice and rightfulness.
Let me point to an interim conclusion. Only if terms such as ‘legitimate objectives’ connote with these dimensions of justice and rightfulness can we get a meaningful argumentative scheme about legitimacy that increases in depth. If we understand legitimacy this way, we might reply to the question begging (for instance in relation to the IC exception) in the following way: ‘the law is legitimate because it protects positive human rights and it is legitimate to protect positive human rights because they enable positive and negative freedom and it is legitimate to protect freedom because being free is a crucial aspect of what it means to be an autonomous human being’. Contrariwise, as I have shown, if we limit the meaning of legitimacy to empirical facts, our argumentation quickly ends up being tautological.
Let me drive these points home with a hypothetical example closer to the WTO context. Assume that citizens in a WTO member A think that it is right to discriminate against Muslims. Assume WTO member A bans all halal labels on meat with the justification that its citizens think it is right to discriminate against Muslims. Assume that most other WTO members have similar legislation. The halal way of slaughter is likely reflected in the characteristic of the product in that the meat will have fewer residues of blood. As a result, it could be argued that the ban on the label relates to product characteristics or that it addresses a related PPM and that the TBT Agreement applies. Following the panel's approach to LOs in the Seals case, a WTO panel would have to find that the policy pursues LOs. Surely, this result would be absurd in an international treaty aimed in part at dismantling discriminatory trade barriers.
3.5 The panel's appreciation of a policy with multiple, conflicting concerns
Through its combination of a concern with seal welfare, addressing ethical consumer concerns, and protecting Inuit traditions and marine resources, the Seals Regime pursues multiple, conflicting concerns.Footnote 71 This may often be the regulatory reality, because trade-offs have to be made and influence the legitimacy of regulatory objectives being pursued.Footnote 72 The reason we need law is precisely to deal with the problem of conflicting concerns in a legitimate manner.Footnote 73 Against this backdrop, any requirement to pursue animal welfare without exceptions, as Canada would have it, is problematic and inconsistent with prior TBT case law which has recognized the possibility of conflicting multiple purpose technical regulations implicitly in US–Clove.Footnote 74
Note that, for the purpose of the Article 2.1 analysis, the panel accepted that the protection of Inuit traditions was legitimate and found fault only with the manner of application of the exception. In other words, the panel acknowledged that the IC exception pursued a different aim than the ban and that it was a legitimate regulatory distinction. For the Article 2.2 analysis, the panel declined that the IC and MRM exceptions established additional objectives to seal welfare and public morals and considered they were merely an accommodation of interests.Footnote 75
I submit that the panel's approach under Article 2.2 ignored the character of the IC and MRM exceptions as exceptions, that is, as instruments that are legally necessary to advance a special purpose unrelated to the principal animal welfare purpose because the two distinct purposes cannot be expressed otherwise than through a prohibition‒exception mechanism. It is unconvincing to characterize an exception that trumps the objective of halting all trade in seal and seal products as not establishing another, independent objective. A mere accommodation of interests looks different to me. Accommodating the interest of Inuit might, for instance, comprise longer transition periods for phasing in the ban towards IC hunts, possibly coupled with some assistance to develop other economic activities. Obviously, the IC and MRM are very different from phase-in periods. Note that the EU in its Appellee Submission criticizes the inconsistency of the panel's finding under Article 2.2 that the IC and MRM exceptions are not really independent of the main animal welfare public morals objective and then finding that they bear no rational relationship to the objective. Footnote 76 The EU's critique is instructive. Interest accommodation in the way I just characterized it would still acknowledge that the aim ultimately is to terminate seal hunting. A policy with a trumping concern does not, so the question is why the exceptions should be measured against goals they do not ultimately recognize under Article 2.2.
More troubling, as a result of the panel's approach, the exceptions diminished the ability of the EU to justify its policy under Article 2.2 because the panel considered that they reduced the contribution to fulfilling the principal seal welfare objective. This is problematic for several reasons: The panel's approach to Article 2.2 encourages WTO members to pursue single-purpose policies because policies without trumping exceptions for conflicting purposes or accommodation of other interests can be more easily justified as necessary.Footnote 77 However, we have reasons to want governments to pursue conflicting objectives and make hard trade-offs because, as stated above, this is what law is for.
Note in this connection that the panel also did not positively take account of the trade-permissiveness of the Seals Regime's IC and MRM exceptions in its assessment of the trade-restrictiveness.Footnote 78 It did not even ask itself to what extent the trade-permissiveness of the exceptions outweighed the reduction to the fulfillment of the objective of seal welfare in its necessity analysis. However, even allowing for the netting of trade-permissiveness and reduction to fulfillment of objectives remains unsatisfactory because an exception for an ostensibly valid regulatory reason might be only insignificantly trade-permissive yet reduce the contribution to the fulfillment of the primary objective a great deal. In other words, it is just not the case that trade-permissiveness of exceptions and reduction to fulfillment of the primary objective will always cancel each other out. But even if they do not, it would be troubling if such policies should fare less well under the necessity test in Article 2.2 than policies were such an evening out effect is present or policies that only pursue one objective.
By finding the contribution to fulfilling LOs reduced, just because of the multi-purpose nature of the Seals Regime, the panel would also in effect treat exceptions in multi-purpose policies with an illegitimate and those with a legitimate secondary purpose the same. The approach of the panel is all the more unconvincing as the TBT Agreement contains an open-ended list of objectives, which should make it easier to justify the necessity of technical regulations with multiple purposes under the TBT Agreement.
It might be argued that multi-purpose policies would not be disfavored compared to single-purpose policies under the necessity analysis if the contribution to the fulfillment of the primary objectives only gets reduced to the extent that exceptions have been shown to create an unnecessary cost on the primary objective under the Article 2.1 analysis. However, this is not what the panel said it did under the Article 2.2 analysis.
For the comparable situation under GATT Article XX, Bartels has suggested that the paragraphs of that Article should focus on the justification of the trade-restrictiveness of the measure whereas the chapeau should focus on the justification of discriminatory aspects of the measure.Footnote 79 In respect of the sequencing of these two tests, he argues that it is possible to vet the justification of discriminatory effects before vetting the trade-restrictive effects.Footnote 80 In the TBT context, this would amount to an analysis of Article 2.1 prior to an Article 2.2 analysis for multipurpose policies. A complainant may not, however, have invoked Article 2.1. This approach gives rise to the sequencing problem detected by Mavroidis and discussed in Section 3.3 that the manner of application of a regulatory measure is assessed first even though the measure would have to be replaced by a more trade-permissive but equally effective policy. Additionally, the justifiability of a regulatory distinction with differential impacts may also still have a role to play in the inquiry of whether or not a less trade-restrictive means is reasonably available. If the less trade-restrictive alternative would endanger the other policy goal, it would not be a reasonably available alternative. Consequently, a neat delineation of Article 2.2 as being about trade-restriction and 2.1 being about the justifiability of regulatory distinctions with differential impacts seems impossible.
Howse and Levy have also forcefully argued that a panel second guesses the right of a WTO member to set its level of protection if it considers the contribution towards fulfillment of LO reduced because a policy contains exceptions.Footnote 81 They suggest that the one question under Article 2.2 is whether there is a less trade-restrictive alternative to achieve the level of protection.Footnote 82 The right to set the level of protection autonomously has been affirmed by the Appellate Body repeatedly so that the panel should have taken care not to interfere with it.Footnote 83
It might be argued that Article 2.2 excludes the possibility of justifying multi-purpose policies because it speaks of technical regulations not being more trade-restrictive than necessary ‘to fulfill a legitimate objective’ [my emphasis]. The word ‘a’ could imply that technical regulations may pursue only one legitimate objective. Note that the panel never even addressed this issue. I consider that the word ‘a’ should not have this limiting effect on the number of LOs a technical regulation may pursue because the preamble of the TBT Agreement recognizes that a WTO member may take regulatory measures ‘at the level it considers appropriate’. This implies the freedom to conduct a comprehensive regulatory cost–benefit analysis, and therefore necessarily also the possibility to opt for exceptions with a lower or no level of protection should another important regulatory objective so warrant it. Additionally, the word ‘fulfill’ does not require that a technical regulation completely attains a legitimate objective and more partial contributions should therefore be allowed, especially if the WTO member has policy reasons for them. Lastly, the TBT Agreement intends to further the objectives of the GATT. A possible telos of the GATT is considered to lie in preventing the imposition of unjustified externalities on foreign trade interests but not in the pursuit of uniform conditions of competition and regulatory uniformity across WTO members.Footnote 84 The preamble and the word ‘fulfill’ as context and the telos in the interpretation of Article 2.2 militate against interpreting the word ‘a’ as restricting LOs to only one.
There are further legal arguments in favor of my position. Article 2.2 requires that technical regulations not be ‘more trade-restrictive than necessary to fulfill a legitimate objective, taking account of the risk non-fulfillment would create’ [my emphasis]. The Appellate Body has confirmed that this requires an assessment of the nature of the risks at issue and the gravity of the consequences of non-fulfillment.Footnote 85 Concerning non-fulfillment, the Seals panel limited itself to evaluating seal welfare and ethical risks.Footnote 86 However, nothing in the text requires the panel to limit itself only to the primary objective here. In fact, the panel could have characterized the nature of the risk to animal welfare EU public concerns relative to the minority rights-related concerns of the IC exception and it could have accordingly assessed that the gravity of non-fulfillment of the seal welfare EU public concern through the IC exception was not severe and justified the reduction in the contribution to seal welfare under the earlier part of the necessity test. Note that room would still be left to fine-tune the exact balancing of the conflicting concerns under the test of Article 2.1 pertaining to the even-handedness of the application of the regulatory distinction.
Arguably, the panel should also have drawn in its interpretation of necessity on the emergence of a general principle of international law concerning animal welfare and on human rights law obligations in respect of indigenous communities.Footnote 87 Had it done so, it could have realized that the international legal context of the terms ‘necessary for the fulfillment of a legitimate objective taking account of the risks not fulfillment would create’, constituted by human rights norms and emergent animal welfare protection, pulls in opposing directions, making trade-offs between the two necessary.Footnote 88 As shown, Article 2.2 does allow for trade-offs to be made without ‘punishing’ such policies relative to the treatment of single-purpose policies.
So how should the analysis of necessity be conducted further for a multi-purpose technical regulation? Clearly, it still is necessary to assess whether there was a necessary reason for killing nuisance seals inhumanely under the IC and MRM exception, for instance, whether protecting traditional Inuit ways of life necessarily required the inhumane nature of the killing.Footnote 89 If there was no such necessary reason and if an alternative could allow for more market access for well-regulated commercially hunted seals without engendering poorer seal welfare outcomes, the alternative is less trade-restrictive but equally respectful of the trade-off the policy-maker desires.Footnote 90 The ultimate burden for a multipurpose policy with conflicting objectives is to show that the policy allows poor seal welfare outcomes only to the extent that the conflict with the other policy purpose is unavoidable. Put differently, the EU should have shown that poor seal welfare outcomes are necessary to attain the other purposes of its seals policy (IC and MRM protection).Footnote 91 It is exactly this kind of inquiry that the Appellate Body has carried out when it completed the analysis of the chapeau of GATT Article XX.Footnote 92 Its inquiry on this aspect is nothing other than a regulatory necessity test between two conflicting regulatory objectives that produce a differential impact on trade.Footnote 93
The panel's problematic approach to multipurpose policies with conflicting objectives also becomes apparent in the uncritical importance it attached to EU public opinion on seal welfare. It noted that EU public opinion did not distinguish between the origin of the hunt in its rejection of killing seals inhumanely and therefore determined that the IC and MRM exceptions reduced the measure's contribution to addressing public morals concerns.Footnote 94 By attaching so much weight to public opinion, the panel ignored the statement of the panel in US–Gambling that relevant public morals standards of right or wrong conduct can be maintained ‘by or on behalf of a community or nation’ [my emphasis].Footnote 95 As the EU points out in its submission on appeal, the reference to standards of public morals being maintained on behalf of a community suggests that the government can enact standards of right or wrong conduct without support from public opinion.Footnote 96
The fact that the panel considered the concerns protected by the IC and MRM exception not to fall under EU public morals because they were not reflected in EU public opinion is indicative of its misunderstanding of the role public opinion should play in public morals policies. Regulatory intervention going against public opinion can be justified for instance if public opinion is unjust or unconstitutional. Perhaps EU public opinion failed to attach sufficient importance to protecting the rights of minorities, traditions, and marine resources forming the basis of the livelihood of fishermen in the eyes of the EU so that it decided to intervene. It should certainly not be per se impossible to justify departures from public opinion in overall public opinion-related policies as long as these departures are rationally connected to another legitimate purpose. Anything less would seriously undermine the right of WTO members to regulate.
As regards technical regulations pursuing multiple, conflicting objectives related to public morals, it is arguably inherent in the concept of public morals itself that trade-offs or accommodations need to be made. In the following section, I will develop an interpretation of the term ‘public morals’ that brings it close to deontological standards of justice. If the requirement of a public morals policy is to be reciprocally justifiable to legal subjects with different interests and views of the good, a WTO member would sometimes even have to make trade-offs in order to succeed in demonstrating that its technical regulation relates to public morals.
The pursuit of one demand of justice completely at the expense of a conflicting demand of justice would violate the condition of reciprocal justifiability because it would entail that the interests of people protected by the conflicting demand of justice are entirely set aside, which no-one would reasonably consent to. The following example should make the argument accessible: an unlimited right to freedom of expression would not be reciprocally justifiable because it would mean that people could not protect themselves against speech that violates their dignity or their privacy. Because the EU's ban on seals and seal products concerned civil, political, economic, social, and cultural rights of Inuit minorities, the EU arguably had to give sufficient protection to their rights in order to succeed with its claim that the Seals Regime's objective was to protect public morals. Whether or not the seal welfare and consumer concerns can also be seen as related to public morals will be examined in the next section.
4. Ethical-expressive policies under WTO law
Commentators have wondered whether WTO law should allow regulation based on moral feelings and which agreement would be applicable to such a measure. An amicus curiae brief to the Appellate Body argues that only GATT Article XX (a) but not the TBT Agreement applies to regulatory measures motivated by non-instrumental moral concerns.Footnote 97 Moral philosophy knows of three principal types of moral standards: instrumental or utilitarian standards, deontological standards, and finally ethical standards. The amici point to the word ‘technical’ in the term ‘technical regulation’ and to the omission of the term ‘public morals’ in Article 2.2 to make an argument that the TBT Agreement excludes moral matters from its scope.Footnote 98 They also argue that the necessity test inherently requires an instrumentalist or consequentialist inquiry into effectiveness that is at odds with measures whose central purpose is to express moral opprobrium.Footnote 99
It is unconvincing to argue that the absence of the term ‘public morals’ in the TBT Agreement excludes only one specific type of moral standard, namely ethical-expressive ones. But even the complete exclusion of any moral policies from the TBT Agreement is impossible to sustain. Article 2.2 makes reference to the protection of human health or safety and national security policies that are of a moral character in that they embody the moral obligation of the state to protect the right to life and physical integrity of its citizens. Another problem is that the amici place strong reliance on the word ‘technical’ in order to interpret the definition of the term ‘technical regulation’. The effect would be that even measures that regulate physical product characteristics on moral grounds would have to be considered as outside of the definition of a technical regulation (e.g. a ban on human embryonic stem cells in a drug). It is not clear that Article 31.1 of the Vienna Convention on the Law of Treaties would condone such heavy reliance on words in the context in the interpretation of treaty terms, all the more so when the terms are meant to be a definition of those words.
Perisin wonders whether WTO law should allow its members to regulate based on irrational feeling where the policy is not scientifically supported or where it singles out one species, even if others suffer too. In discussing the public morals justification, Perisin has tried to establish a difference between moral views that inherently cannot be rationally proven (her example is religion) and views that rest on misconceptions about facts and scientific evidence.Footnote 100 For the former, she seems to concede the possibility that a public morals justification could be available but not for the latter.Footnote 101
An obvious difficulty with Perisin's proposal is that it would open the door to abuse and all kinds of morally objectionable policies. For instance, countries could declare that the reason to prohibit halal labels is not rooted in any facts about Muslims but simply in their belief in the inferiority of Muslims, and on this ground advance a public morals justification for the ban. Why a view should automatically be immune from factual contestation as soon as it claims to be based in belief is unclear to me.
Perisin also mischaracterizes the EU's policy. Howse and Langille argue that the EU policy was about the ethical-expressive preferences of EU consumers and not only about some instrumentally valued aim of animal welfare.Footnote 102 Utilitarian political philosophers have defended animal welfare policies as an instrument to attain a telos, namely to minimize pain and maximize pleasure of sentient beings.Footnote 103 Singer argues that we have a moral duty to become vegetarian because modern farming techniques inevitably cause animal suffering.Footnote 104 The Seals Regime went beyond attaining only this utilitarian or instrumental telos of minimizing suffering. Howse and Langille submit that the EU seals policy was partly about the expression of intrinsic ethical beliefs of what constitutes cruelty, which cannot be substantiated by scientific policy analysis.Footnote 105
I suggest they are correct. Utilitarian consequentialism does not require all future or avoidable consequences but only reasonably foreseeable ones to be anticipated.Footnote 106 Utilitarian philosopher Singer, a key advocate of animal rights, does not argue that we may not consume meat from animals raised and killed humanely and he also does not postulate a moral duty to abstain from non-meat purchases in supermarkets and restaurants, even though we thereby provide income to firms complicit in immoral animal suffering.Footnote 107 The EU claimed that any trade in seal products would sustain the market for commercially hunted seal products indirectly by creating more market demand and higher prices.Footnote 108 This shows that the EU's policy went beyond consequentialist utilitarianism and is more properly described as motivated by virtue ethics.
Virtue ethics is about installing in individuals the traits required in order to be virtuous. Unlike utilitarianism and deontology, the other two strands of moral philosophy, its concern goes beyond action to character building. For a person to be virtuous, the virtues must apply all the way down, that is, the person must make these virtues her own and hold these all the time.Footnote 109 So a person subscribing to animal welfare ethics must live virtues of compassion and beneficence in her direct dealings with animals, but also in situations whenever her behavior remotely affects animals or implicates her moral feelings (such as market purchases of humanely hunted seals).
Once we understand the aim of the Seals Regime as one of consumer virtue, we can see that its validity does not hinge on scientific evidence on whether humane hunting is possible as long as inhumane hunting is also practiced and supported by any commercialization of seal products. Positions in virtue ethics derive validity from how the virtues are necessary to attain human well-being and for it to flourish, and an appreciation on the basis of practical wisdom of their situational appropriateness.Footnote 110 It is not the consequences of an action that matter for its rightfulness, it is the fit between what we ought to do in the abstract and our beliefs and dispositions in specific instances. Consumer ethics, like the ones at issue in the EU Seals Regime, therefore can have a rational basis.
Given considerable, reasonable pluralism in society about which virtues to follow, some political philosophers consider collective regulation binding as to the observance of ethical standards to be illegitimate.Footnote 111 In contrast, deontological moral standards, set as a result of reflective processes in which persons treat each other as ends in themselves and not means, can be subscribed to by all reasonable people regardless of their different views about what it means to lead a virtuous life.Footnote 112 Deontological moral standards are public:Footnote 113 They are shared, reciprocal, and generalizable agreements about how to order our public life together so that inherent interests of every human can be equally protected (e.g. in physical safety, education, income).
This understanding of deontological moral standards comes close to the WTO panel's definition of the term ‘public’ in US–Gambling as ‘of or pertaining to the people as a whole; belonging to, affecting, or concerning the community or nation’ [my emphasis].Footnote 114 According to the panel, the measure for which a public morals justification is claimed ‘must be aimed at protecting the interests of the people within a community or a nation as a whole’ [my emphasis].Footnote 115 This suggests that the relevant standards must be generalizable. Because deontological moral standards are public, they fit under the WTO term of public morals whereas ethical-expressive standards are moral but not public in this sense.
I would argue that the interpretation of the terms ‘legitimate objectives’ and ‘public morals’ necessarily connects WTO law to a knowledge system beyond law, namely to moral philosophy, because it is this system of knowledge that expounds the demands of legitimacy and morality. The notion of ‘legitimate objectives’ in the TBT Agreement and GATT Article XX (a) is just like other terms in WTO law such as ‘sufficient scientific evidence’ in the Agreement on Sanitary and Phytosanitary Measures, which connects law with the exact sciences.Footnote 116
Reliance on moral philosophy to interpret these WTO terms would also address the scope of ‘public morals’: whether the relevant normative standards must be universally shared or may be particularistic, culture-relative standards defined by each WTO member, and whether only interests within the jurisdiction of the regulating WTO member or also those located outside of its jurisdiction may be protected. Scholars in favor of evidentiary unilateralism argue that WTO members should be free to define the content of public morals themselves as long as they can provide relevant evidence of widespread public support.Footnote 117 Other scholars argue that the content of public morals requires actual universal agreement today or in the past arguing for an evolutionary or originalist interpretation.Footnote 118 The scholars differ in what they accept as evidence of actual endorsement, ranging from opinion polls, to legislative enactment, international law obligations, ius cogens, the type of national law and its hierarchy within the domestic legal system, national ordre public, and religious texts.Footnote 119 This difference is not material for my further argument because all share the view that the support must be an empirical convergence of opinion.
The central difficulty with their approach is that they disregard the ordinary meaning of the term ‘moral’ (or ‘morals’ in the plural) as a being a normative judgment about right or good conduct.Footnote 120 Because the concept of morality expresses a regulative ideal, not an empirical fact, convergence of opinion is not decisive. Normatively speaking, actual consensus given under appropriate circumstances is only indicative of moral standards. The notion of ‘appropriate circumstances’ points to free consent without coercion from force or material need, etc., and is already a moralized notion of consent and not just a fact of convergence of opinion.Footnote 121
Obviously, actual international consensus is still relevant for interpreting the term ‘public morals’ as a relevant rule of international law binding in the relations between the parties in the sense of Article 31.3(c) of the Vienna Convention on the Law of Treaties.Footnote 122 However, context is only one of three factors in interpretation, the other two being textual meaning and telos.
The TBT Agreement intends to further the objectives of the GATT.Footnote 123 If its telos is not the harmonization of regulatory conditions of competition across WTO members, it might be objected that my endorsement of moral unity imposes unwarranted regulatory unity. I think this does not follow because my position does not imply that facts affecting feasibility have to be disregarded completely, and that incremental steps towards the ideal are not allowed. Feasibility refers to the relatively immutable facts inherently necessary for the realization of the moral requirement.Footnote 124 Since these facts can differ between WTO members, regulatory responses can also legitimately differ.
It might be objected that the US–Gambling panel confirmed that the content of public morals is unilaterally to be determined.Footnote 125 I would counter that my interpretation is more consistent with its statement that public morals denotes ‘standards of right or wrong conduct maintained by or on behalf of a community or a nation’.Footnote 126 The panel might have simply been saying that actual commitment to the moral requirement is a precondition for being able to invoke the requirement as a justification. The scholars who interpret this statement as allowing for unilateral relativism commit the interpretative mistake of ignoring the terms ‘standards of right or wrong conduct’ by stressing only the last element. They also ignore that the panel had clearly suggested that a public morals policy must be in the interest of the members of a nation or community.Footnote 127 This statement suggests some check on the substantive fairness of a policy.
It is true that the Gambling panel gave some scope to members to ‘define and apply for themselves the concepts of “public morals” and “public order” in their respective territories, according to their own systems and scales of values’.Footnote 128 It also recognized that the content of public morals could vary in time and space, depending on prevailing social, cultural, ethical, and religious values, amongst other factors.Footnote 129 But this can be interpreted as a concession to feasibility, which is different from complete relativism regarding the content of morality.
Moral philosophy indicates also the permissibility of protecting interest outside of the state's territory.Footnote 130 When an action is morally required between strangers, such as not harming a stranger's life or health, the fact that the person is located in another state should be immaterial.Footnote 131 In contrast, where moral obligations only arise by virtue of special relationships within the state, extraterritorial action would not be warranted.Footnote 132
Does my analysis imply that an ethical-expressive regulation can never be ‘public’ and justifiable as a public morals policy under WTO law? With Hüller, I would suggest that ethical-expressive matters can legitimately become matters for public regulation in two ways.Footnote 133 First, because everyone reasonably agrees on the content of the ethical-expressive standard and no deontological moral standard (such as a human right) is infringed.Footnote 134 In this instance, there is no need for permitting pluralism because ethical judgments are not plural. Second, because the matter inherently requires a decision in favor of one or the other ethical view. Hüller gives the examples of whether or not to include religious education in the curriculum of primary school education or which city to choose as the capital.Footnote 135 Either religious education is in or it is out of the curriculum (even if non-believers can opt-out) and the capital can be located in only one place. These decisions therefore inherently require a collective disposition. I suggest that a panel tests the ‘public’ dimension of ethical-expressive matters in this way.
It might be objected that collectively binding regulation is per se unnecessary if there is collective agreement on the ethical standard of right conduct. I submit that collectively binding regulation can nevertheless be necessary to ensure that foreign traders who do not share in the ethical judgment conform to it, to enhance legal certainty by stipulating clearly the requirements or to provide non-apparent information relevant for the ethical-expressive judgment to consumers.
Was there European consensus on consumer ethics? Only Belgium and the Netherlands had in place bans on seal products prior to the EU Seals Regime, suggesting there possibly was European disagreement about the demands in consumer virtue ethics (not animal welfare). Consumers committed to the standard of virtue ethics can also follow it without collectively binging regulation. All they require is information about whether certain products are free of seal products.
For the foregoing reasons, I thus beg to differ with Howse and Langille who maintain that the ethical-expressive element of the EU Seals Regime should be considered a WTO-legal public morals measure and with Levy and Regan who consider that WTO members have a right to regulate in accordance with collective preferences as long as these are not protectionist.Footnote 136 If there is no normative reason for regulating the content of an ethical-expressive matter in a collectively binding manner, the fact that a majority supports the policy is not enough to confer legitimacy or a public morals dimension on a policy with prescriptive content.
Note that the reasons I offered for why collectively binding dispositions on ethical-expressive preferences are legitimate takes account of Howse's and Langille's policy concern with permitting pluralism of ethical views because it also permits intra-societal pluralism without requiring panels or the Appellate Body to examine the substance or validity of ethical-expressive beliefs.Footnote 137
My conclusion notwithstanding, the factual evidence before the panel about inherent seal suffering regardless of the method used to kill them suggests that the prescriptive aspect of the regime (the ban on any seal products regardless of the hunting method) could nevertheless be justified: As an animal welfare policy designed to protect life and health of animals under GATT Article XX (b) that adopts a very conservative level of protection necessary because of the inherent difficulty to hunt seals humanely.Footnote 138 Health of animals, the panel in US–Tuna has recognized, includes also animal welfare aspects.Footnote 139
5. Conclusion
Overall, the panel's findings in the Seals case reveal misunderstandings of the role, purpose, and the interaction of WTO provisions and of the aims and policy decisions being made in regulatory processes. This leads the panel into problematic findings on Article 2.1 and 2.2 and into giving too little leeway to defend standard regulatory cost–benefit analysis under the necessity test of Article 2.2. The panel also equates normatively loaded concepts like public morals or legitimacy with empirical facts. I have shown that such type of reasoning becomes tautological. It would have been desirable had the panel shown more zeal in understanding regulatory trade-offs and terms whose content points beyond the positive law to the law's claim to moral correctness. The excursus into normative political theory explains why the EU exhibited excessive zeal in regulating consumer virtues. It also explains why, absent the availability of another public policy justification, such types of policies are best regulated on the basis of positive product labels enabling consumers with high standards to exercise their ethical-expressive preference for products with the positive attributes rather than through a prescriptive public morals policy. In this connection, I have also defended the view that the concept of public morals should be interpreted to refer to deontological moral standards of right and wrong – such as are generally considered to be embodied in human rights.
Finally, I should point out what I have not argued in this article. I have not criticized the panel for having examined the evidence on EU public opinion and EU legislation. I have criticized it because it attached decisive weight to EU public opinion and the fact that EU legislative documents referred to public morals in this context. I have shown firstly that the weight accorded to the evidence is inconsistent with the proper meaning of the term public morals as deontological standards of right or wrong conduct. In EC–Hormones, the Appellate Body suggested that assessing the consistency with the covered agreements objectively required applying customary rules of interpretation and Article 11 of the DSU.Footnote 140 As far as the customary rules of interpretation are concerned, Section 4 has offered an interpretation that gives meaning to the terms ‘public’ and ‘morals’. Secondly, Section 3.4 has demonstrated that a legal argumentation about terms implicating the justification of law's authority that gives weight to moral dimensions of justice and rightfulness is better in the sense of being more likely to be true than one that only looks to empirical facts. If it is more likely to be true, it is also more objective. This article, however, implies no specific position under Article 11 of the DSU on what non-decisive weight panels should attribute to public opinion and legislation nor of on the meta-ethical question of whether moral feelings (as physical states of mind) have any role to play in grounding moral facts.
