‘When I use a word’ Humpty Dumpty said in a rather scornful tone, ‘it means just what I choose it to mean – neither more nor less.’ ‘The question is’ said Alice, ‘whether you can make words mean different things.’ ‘The question is’ said Humpty Dumpty, ‘which is to be the master – that’s all.’Footnote 1
Introduction
Debates about constitutionalism have long been a part of international economic law. In particular, arguments about whether and to what extent the World Trade Organization (WTO) should be considered ‘constitutional’, and the implications of constitutional theory for the organization’s form and substance, have elicited thousands of pages of scholarship. These debates and the political and academic positions they represent have influenced perceptions of the WTO and its boundaries, and have affected the further development of the organization itself. For scholars of international economic law, then, constitutionalism and its related debates are clearly a ‘big deal’.
What is less clear, however, is what exactly the constitutional debate is about. A number of authors have observed that constitutionalism is an amorphous or difficult to define concept.Footnote 2 Definitions and uses of ‘constitutionalism’ seem to be nearly as numerous as the number of commentators who write about it. In different hands it can be an institutional concept, a functional concept, a theoretical concept, a political concept, and perhaps even a theological concept. It can apply to any organization at any level, or only to certain long-standing and well-recognized sovereign democratic states. It can be a fact or a process, a bright line or a sliding scale. And it can be used to imply a need for centralization or for decentralization, for coherence or for specialization, for form or for substance, for continuity or for change.
In other words, while everyone seems convinced of the importance of constitutionalism, no one seems quite able to define it in a broadly satisfactory way. Instead, it is something more along the lines of a floating signifier, a term like ‘justice’ or ‘democracy’ that stands at the end of a long chain of equivalences, and can be used to explain or justify a number of contradictory claims and practices, depending on the context and the intent of the speaker. It is on this point that this paper seeks to build.
This paper will argue that when it comes to ‘constitutionalism’, we would benefit from approaching the term a little less like Alice, and a little more like Humpty Dumpty. This is not to imply that constitutionalism has no meaning at all – far from it. It is a term that evokes a common set of associations (for example, something organized more like a state and less like a contract), a common set of functional questions (related to things like the division of powers and the protection of rights), and a common set of fears (of things like the concentration of power, and democratic deficits). Indeed, these common associations, questions, and fears are what give the word its power and make it an important site of contestation. But these shared meanings also mask the fact that the precise content of ‘constitutionalism’ remains essentially disputed.
In brief, this paper will approach ‘constitutionalism’ not as something that can be measured or assessed in any objective sense, but rather as a site of discursive contest. It is a term better viewed as a debate in which participants intervene on behalf of particular understandings of how the system does or should operate. Attempts to come up with objective normative or analytical definitions of constitutionalism will always be implicated in these discursive contests. The real question is not whether an organization is constitutional or what this does entail; it is ‘which is to be the master – that’s all’.
This approach adds to the literature on WTO constitutionalism a focus on what is created or produced by constitutional debates. The discursive contest to define constitutionalism is important because it generates knowledge about how the system works, or should work, and what the possibilities for change and action may be. Calling (or refusing to call) the WTO ‘constitutional’ or arguing about its ‘constitution’ in a formal or substantive sense is not a neutral endeavour. It actively constructs our understandings about how the institution functions and how it should be organized, who its subjects are and how they behave, and what the aims of law, economy, and society should be. These constitutional contests frame the fields of action of the WTO and its Member States, producing a locus of power and a subject in a mutually constitutive relationship with one another. Viewing constitutionalism as a discursive contest, as this paper proposes, allows us to see that there is more at stake in these debates than the boundaries of doctrinal categories: constitutional debates take place as part of a process of rethinking the identity of the organization, its members, and the international context in which they exist, and as such produce new understandings and avenues for the exercise of power.
This paper will proceed in three parts. The second Part will flesh out and expand on the idea of constitutionalism as discursive contest introduced here. In particular, it will draw on Foucault’s later work on the relationship between government and truth, and the idea of the essentially contested concept as outlined by William Connolly.
The third Part will approach the question of constitutional contestation more directly by defining a typology of three sets of constitutional contests:
1. contests over constitutional status – whether and what it means that an organization or system is ‘constitutional’ or not;
2. contests over constitutional form – what a ‘constitutional’ organization or system does or should look like from an institutional perspective; and
3. contests over constitutional substance – which normative values a ‘constitutional’ organization or system does or should have, and how these values are hierarchically arranged.
In exploring each of these three discursive contests, it will draw examples from the debates over constitutionalism that have taken place in the context of the WTO, paying particular attention to the ways in which these debates both construct and are constructed by the normative commitments of their participants.
The final Part will conclude by offering some summary thoughts on the benefits of understanding constitutionalism as a discursive contest, and how these contests have shaped and are continuing to shape the theory and practice of WTO law.
Constitutionalism as discursive contest
As noted above, one of the most striking features of academic discussions of international economic constitutionalism is the lack of a common definition of the term. Most analysts agree that a ‘constitution’ or ‘constitutionalism’ reflects ‘the organization of a community’ or, in Deborah Cass’s words, ‘a set of social practices, defined as law, and associated with Western industrialized democracies, which structure the division of public power within a given community’.Footnote 3 However, beyond these abstract descriptions, different scholars use the words ‘constitution’, ‘constitutionalism’, and ‘constitutionalizing’ to refer to numerous – and sometimes conflicting – situations.Footnote 4 ‘Constitutionalism’ describes issues and processes varying from ‘regime formation’, to ‘institutional architecture’, to ‘democratic governance’, to ‘the hierarchical ordering of norms’, to ‘common legal principles’, to ‘the protection of individual rights’, to the ‘metaphysical creation of a polity’, and many others. These definitions are applied, inter alia, descriptively to gauge the constitutional status of a particular institution or order, normatively to develop programmatic ideals, analytically to trace the functionality and effectiveness of regimes, and theoretically to examine the relationship between government and society and its implications for politics and identity.
Perhaps the only sure thing that can be said about constitutionalism, then, is that it is not a singular or concrete idea. Instead, it is an ‘essentially contested concept’. As William Connolly defines it:
When [a concept] is appraisive in that the state of affairs it describes is a valued achievement, when the practice described is internally complex in that its characterization involves reference to several dimensions, and when the agreed and contested rules of application are relatively open, enabling parties to interpret even those shared rules differently as new and unforeseen situations arise, then the concept in question is an ‘essentially contested concept’. Such concepts ‘essentially involve endless disputes about their proper uses on the part of their users’.Footnote 5
Constitutionalism, like ‘democracy’ or ‘justice’, is seen as valuable or significant. Criteria for assessing whether and how it has been achieved are weighed differently by opposing parties. And discussions about the proper use of the term turn on fundamental issues that can be debated, but are unlikely to be resolved.Footnote 6 Thus, for some, the central feature of a constitutional order is its hierarchical ranking of normative values; for others this is a marginal factor, and the central question is the existence of a common polity, or an institutional architecture featuring the division of powers and balancing. For some, constitutionalism is a descriptive term that is appropriate only for bodies or organizations that function very much like democratic states, for others it is an analytic framework for examining the functional characteristics of any formally organized group. The problem multiplies when we recognize that words like ‘values’ and ‘polity’ also require further explanation, and could themselves prompt discursive splits.
As an essentially contested concept, ‘constitutionalism’ expresses a normative standard, but does so in such a way as to leave open the precise contours of its nature, application, and core elements. Indeed, this paper argues that the term’s simultaneous meaningfulness and indeterminacy is central to its appeal and its discursive power. In that sense, it will define constitutionalism not as a fixed point or condition, but rather as a site of discursive contest. Constitutionalism is a contested space on which players attempt to inscribe their own vision of how an order or system does or should look. It is not something that can be discovered, but rather something that is constructed; not something that is static, but rather something that is constantly shifting. Constitutionalism is not something ‘out there’ that can be measured and defined, but rather something that is produced by legal acts, by social psychology, by material power – and by intellectual debate. At the same time, constitutionalism produces particular effects: ideas about the nature, structure, and scope of a particular regime shape both its own field of action, and those of the other systems, subjects, and constitutional orders with which it interacts.
As an essentially contested concept, constitutionalism invites dispute over meanings and the chains of equivalence to which they should attach. As they progress, these disputes produce particular types of knowledge about ontology, causation, and power. As ‘knowledge’, these discourses define certain ‘truths’ about the behaviour and activities of subjects, and the modes of action that are necessary to order these behaviours and actions.Footnote 7 As a result, they reinforce and justify particular sites and exercises of power, both through mechanisms of efficient institutional ordering as well as through the construction of self-disciplining subjectivities.
In his later work on governmentality, Foucault describes these ‘truths’ in terms of the rationalities that underlie particular forms of governmental power. In Nikolas Rose and Peter Miller’s terms, these rationalities are ‘the changing discursive fields within which the exercise of power is conceptualized, the moral justifications for particular ways of exercising power by diverse authorities, notions of the appropriate forms, objects, and limits of politics, and conceptions of the proper distribution of such tasks among spiritual, military and familiar sectors’.Footnote 8 These rationalities are intimately related to the idea of ‘truth’. What weFootnote 9 ‘know’ to be ‘true’ about society, the state, religion, science, human behaviour, and so on produces particular ideas about how we should govern ourselves and others.Footnote 10 These ‘truths’ in turn structure behaviour as they are operationalized via various technologies of government, and internalized in the form of subjectivity.Footnote 11
In this context, discourses about the (actual or desirable) scope, content, and form of the ‘constitution’ of an institution or order structure its material, social, and political capacity for action, both by shifting the imagined field of possibility as well as by prompting the internalization of certain ‘truths’ by those subjected to that order (the WTO, states, companies, scholars, individuals, and so on). This process of internalization leads to the construction of self-limiting subjects who freely act in accordance with these truths, and, in turn, actively (re)produce the system of power and knowledge. The more hegemonic a particular constitutional discourse becomes, the easier it will be to justify action on the basis of that understanding, and the more likely the subjects of the order will be to discipline themselves along the lines it suggests. As particular institutional truths regarding constitutionalism come to dominate the discussion, therefore, they structure the realm of possibility for future action.Footnote 12 As a result, discursive contests over constitutionalism are not merely semantic – they have real effects on perceived, and therefore actual, distributions of power.Footnote 13
To sum up: in international economic law, a ‘constitutional order’ is not a thing, but a discursive contest. There is no ‘constitutionalism’ that one can point to in a truly descriptive sense, the existence of which can be proven or disproven by an assertion of facts. Instead, there are only debates about constitutionalism. Particular positions within these debates can be more or less convincing, but not objectively true or false. And as these debates progress, they produce particular types of knowledge about the world that in turn structure our perceptions about the way things work, and our perceived possibilities for action. Debates about constitutionalism are far from the only discursive contests that produce truth, or shape the governmental practice of the WTO – indeed, they may ultimately be rather minor in comparison with the influence of political struggles, material interests, and normative debates regarding other aspects of international economic governance. However, constitutional contests are a representative example in terms of identifying the way that particular discourses exert influence with respect to shifting hegemonic conceptions of ‘truth’.
In keeping with this definition of constitutionalism as discursive contest, this paper will avoid making any attempt to describe whether a particular system is or is not constitutional, what the form of such an order does or should look like, or what the content of such an order is or should be. Instead, it will focus on the discursive practices of participants in constitutional debates – representatives of states, lawyers, academics, and institutions – and attempt to map out the playing field on which these constitutional contestations take place. In doing so, it hopes to convince the reader that understanding constitutionalism as a discursive contest provides a useful lens through which to assess constitutional debates, as it highlights the constructive or creative nature of constitutional talk, and challenges the neutrality of constitutional language.
Constitutional Contests: Constitutionalism and the WTO
In order to explore in more detail the way that constitutionalism operates as a set of discursive contests, this Part will define a three-part typology:
1. contests over constitutional status – whether and what it means that an organization or system is ‘constitutional’ or not;
2. contests over constitutional form – what a ‘constitutional’ organization or system does or should look like from an institutional perspective; and
3. contests over constitutional substance – which normative values a ‘constitutional’ organization or system does or should have, and how these values are hierarchically arranged.
This is certainly not the only possible way to understand constitutional contests. However, this typology presents three dimensions along which there has been a great deal of discussion in the WTO context, and which are therefore helpful for understanding and illustrating the way in which constitutional contests have played out with respect to that organization.
For each of these three dimensions, this Part will first examine briefly the field of constitutional contest, suggesting some of the stakes of the debate and the type of knowledge created by particular outcomes. Then it will give some examples from the WTO context, highlighting the ways in which these constitutional contests have been historically and politically productive of knowledge about the WTO and international society, and have structured the organization’s field of action in accordance with these discourses. In doing so, it aims to make concrete the relationship between power and knowledge and the implications of discursive contests over constitutionalism in the field of WTO law.
Contesting constitutional status
A first set of discursive contests centres on whether a particular organization or regime is or should become ‘constitutional’ or not. Because it is a normatively valuable concept that refers to a shared set of associations, calling an order ‘constitutional’ has certain political, psychological, and material implications. Constitutional orders are seen as weightier – as having ‘gravitas’, as Thomas Franck puts it.Footnote 14 As such, they are perceived as freer to adapt to new circumstances and transcend the boundaries of their founding texts in response to emerging circumstances. Constitutional orders are seen as able to make judgments, interpret rules, and weigh and balance the needs of their constituents against claims of the greater good. They imply some claim to legal authority or sovereignty. Constitutionalism may also entail a certain symbolic and political coherence, and an identifiable polity.Footnote 15 To say an order is ‘constitutional’ or ‘constitutionalizing’, then, is to paint it as strong, as independent, as integrated into a single (hierarchical) order, or as a sophisticated political community – or at least as moving in that direction.Footnote 16 In a more negative reading, ‘constitutional’ or ‘constitutionalizing’ might imply domination, lack of restraint, becoming hegemonic, and so on. Refusing to use constitutional language, or to understand a particular system as constitutional or constitutionalizing has equally important implications. In the presence of a debate over constitutionalism, declining to use constitutional language or to label a certain system ‘constitutional’ or ‘constitutionalizing’ is to paint it as potentially weak, as contractual, as sui generis, as dependent on the good will of its constituents, as static; or in more positive terms, as appropriately limited, primarily political, cautious, conservative, and so on. These debates over constitutional status are grounded in particular understandings (knowledge) about the way that the WTO does and should function in international society, and they help to produce an understanding of the system as having or not having certain attributes or relationships. This ‘truth’ about the WTO, in turn, affects perceptions of the particular avenues for and limits to the WTO’s field of action, and its potential for governing.
It is perhaps useful to understand the debate over the constitutional status of the WTO as involving three interrelated sub-contests:
a. the development of a self-consciously constitutional discourse;
b. the relationship between the individual constitutional order of the WTO and the broader system of international (constitutional?) law; and
c. the definition of a sphere of competence for the WTO constitutional order vis-à-vis other (constitutional?) orders in the international system.
The rest of this section will discuss each of these questions in turn.
The WTO as constitutional.
With regard to the first question, the WTO has been the subject of a long-running debate over the extent to which the institution should be thought of as self-consciously ‘constitutional’. Particularly in the early days of the WTO, the ‘c-word’ was a dangerous thing. It was this subject that landed former WTO Director-General Renato Ruggiero in hot water in 1998, when he was famously attacked for using the term ‘constitution’ in reference to the WTO.Footnote 17 When anti-globalization activists learned of his statements, they protested that using the term ‘constitution’ in this context implied a concentration of global power and the formation of an economic ‘police state’.Footnote 18 Needless to say, no Director-General has since made the same mistake. Indeed, to the best of the author’s knowledge, neither the Secretariat nor the WTO Dispute Settlement Body has made any further explicit reference to constitutions or constitutionalism.
Shortly thereafter, a serious debate erupted on the topic of constitutionalization in the WTO.Footnote 19 Scholars like Ernst-Ulrich Petersmann maintained that the WTO represents universal rules recognizing the human right to trade, and that its rules-based adjudicatory system should undergo increasing constitutionalization in order to permit the recognition of individual rights claims. For Petersmann and others, this would involve a consequent and much-needed diminishing of the influence of sovereign states and power politics.Footnote 20 On the opposite side of the argument, Petersmann’s position was criticized by scholars such as Philip Alston, Robert Howse and Kalypso Nicolaïdis, who argued that he had failed to appreciate the damage that could be done by ‘constitutionalization’.Footnote 21 Rather than leading to more rule-based decision-making, they argued, constitutionalizing the WTO in the manner Petersmann suggested could lead to the domination of trade interests, stagnation due to an inability to correct mistakes, increased democratic deficits, and/or the depoliticization of important political choices.Footnote 22
More recently, the debate has become less polarized and more academic. A good example of the new style can be found in the 2006 discussion in the European Journal of International Law between Jeffrey Dunoff and Joel Trachtman. In two interesting papers, Dunoff and Trachtman take opposite positions with respect to whether the WTO is constitutional or not, each maintaining that his own position is supported by objective criteria, and that the contrary position might be explained as rhetorical strategy. On the anti-constitutional side, Dunoff argues that despite the many scholarly assertions of WTO constitutionalism, ‘neither the WTO texts nor practice support this understanding’.Footnote 23 As evidence, he cites the lack of disputes regarding constitutional issues (such as separation of powers), the failure of the dispute settlement body to embrace a fundamental freedom to trade, and the paucity of evidence that the WTO thinks of itself in constitutional terms. WTO constitutionalism, he argues, is not a descriptive fact, but a rhetorical tool: ‘the invocation of constitutional discourse at the WTO – and elsewhere in international law – may be a rhetorical strategy designed to invest international law with the power and authority that domestic constitutional structures and norms possess’.Footnote 24 Trachtman, on the other hand, argues that ‘[t]here is no doubt that the WTO has a constitution in [a] technical sense’.Footnote 25 It has a set of rules, integrates various social values, reflects a group of people, limits the sphere of government authority, and promotes social solidarity.Footnote 26 The failure of some scholars to recognize the constitutional character of the WTO, he explains, might relate to the discursive power of constitutional language. Resistance to constitutionalism stems from ‘fears that something essential (but unidentifiable) will be lost by ascribing constitutional significance to some features of the WTO’.Footnote 27
In these debates, the contest over whether the WTO is or should be a constitutional order or not is the product of particular contextual understandings of the way that power operates on the global level. Director General Ruggiero and the anti-globalization protestors had different ideas about the WTO’s ontological status and causal mechanisms, and their different beliefs led them to take opposite positions with respect to the use of constitutional discourse. Petersmann and his critics took different positions with respect to whether constitutionalizing the WTO would lead to more or less democracy, and whose interests it would privilege. Dunoff and Trachtman both explicitly recognized the power of constitutional language, purposefully asserting or refusing constitutional claims in support of their own beliefs about the international system. Their debates, moreover, have helped to produce and reproduce knowledge about the WTO. As certain understandings gain influence, they alter the way that the organization and its constituents perceive its opportunities for action. The WTO (and Director-General Ruggiero) certainly altered its behaviour in response to the protests in Seattle, the protestors’ refusal to see the WTO as part of or as an international constitutional order, and their assertions about the way that global power functions. Likewise, debates like the ‘constitutionalization’ conflict in the early 2000s and the discussion between Dunoff and Trachtman have come about as a result of the increasing frequency with which constitutional vocabulary is used in WTO scholarship, and have triggered further reflection into the system’s legitimacy, representativeness, transparency, strength, and place within international society.Footnote 28 Taking a ‘constitutional’ position helps to construct the WTO as necessary, as integrated, and as a governing body that has both expanded power and expanded responsibility.Footnote 29 Likewise, denying that the WTO is or should be a ‘constitutional’ order emphasizes that the members are or ought to be the controlling power in the WTO, that the organization remains essentially a set of inter-state contracts, that it is a singular or sui generis entity, and that it has not or should not overstep its textual bounds.
Unification versus fragmentation.
Linked with this debate over whether the WTO is constitutional or not is the question of the organization’s place in the wider system of international law. This question has to do with the coherence of the international legal system, and with jurisdictional divisions among the various international legal regimes.Footnote 30 It generally reflects the issue of fragmentation versus unification that has bedevilled international law for the past decades.Footnote 31 In this area, constitutional claims can lead to two opposite outcomes. On the one hand, they may lead to legal unity. If we understand the international legal order (or some part thereof) as ‘constitutional’ or ‘constitutionalizing’, then individual international organizations should integrate these broader principles of international law into their mandates, and understand themselves as part of a wider web of legal actors. On the other hand, constitutional claims may also lead to fragmentation. If we understand an individual international organization as ‘constitutional’, then, as Jan Klabbers has argued, we run the risk that ‘constitutionalism will … only result in deeper fragmentation, as the various competing regimes and organizations will be locked firmly in constitutional place – and in battle with each other’.Footnote 32
With respect to the WTO, these questions translate into three general positions: (1) that the WTO is one part of a broader constitutional system, (2) that the organization is a constitutional regime unto itself, or (3) that neither the international system nor the WTO are constitutional regimes. Each of these positions has clear normative implications with respect to both the WTO in general and individual disputes between member states. The third position is dealt with in the prior section. The first two will be discussed below.
On the ‘unification’ or ‘international constitutionalism’ side of the debate stand scholars, lawyers, and analysts who see the WTO as part of the broader system of international law. WTO Director-General Pascal Lamy might be seen as one advocate of the ‘unification’ position. As he argued (although not in constitutional terms), ‘it is clear that WTO law is largely a circumstantial application of international law in general … [and] has also acted as a vehicle in the evolution of international law towards its contemporary form, and indeed is a driving force in the progressive transformation of international society into an international community’.Footnote 33 In support, Lamy cites evidence such as the WTO’s contribution to international peace and security. Proponents of international trade law have always linked it to the goals of avoiding armed conflict and promoting the peaceful settlement of disputes. Liberalizing international trade, the argument goes, helps to guarantee international peace both because it binds nations together economically, making conflict costlier, and because legal guarantees of free trade prevent and peacefully resolve trade disputes, which could otherwise spiral out of control into war. Additionally, Lamy notes that the WTO has embraced and elaborated other international ‘constitutional’ principles. The sovereign equality of states, for example, is a foundational value at the WTO, which often invokes its commitment to the formal equality of states regardless of size or power. The WTO dispute settlement body also makes frequent reference to the Vienna Convention, underlining its commitment to and embeddedness in, international law.Footnote 34
Joel Trachtman also comes down on the ‘unification’ side of the discussion, arguing that ‘we must also recognize that the WTO constitution is itself but a part of a broader structure for the global system’.Footnote 35 As such, ‘it is necessary to examine the WTO constitution in the context of the general public international law system, and in relation to other components of that system. Indeed, the general public international law system, including its subsystems, must be evaluated in constitutional terms’.Footnote 36
On the ‘separate constitutional regime’ side, are scholars like Deborah Cass, who argue that ‘the constitutionalization of trade law by norm-generation is qualitatively and quantitatively different from any constitutionalization of international law’ because of the different ‘constitutional content of international trade, its structure, and the legal community which is forming around those elements’.Footnote 37 The fact that the WTO cites the Vienna Convention’s rules of interpretation ‘cannot be the key indicator or engine of constitutionalization’.Footnote 38 For that reason, ‘international law remains … a more diffuse system of law, which does not closely resemble the more densely structured, local form of legal community which is known as international trade law’.Footnote 39
Whether and how the WTO ‘constitution’ is understood to relate to the wider international legal order is, like the contest over whether the WTO is or has a ‘constitution’ or not, a product of context and the normative commitments of the speaker. Interventions into the debate are made with a mind to capture the discursive space, and will, if successful, produce certain effects. If knowledge about the position of the WTO in relation to the broader system of international law shifts in favour of coherence, it will follow that the dispute settlement system, trade policy review mechanism, and other bits and pieces of the WTO should take this into account, incorporating developing international legal standards into their mandates. If, on the other hand, it becomes widely accepted that the WTO ‘is’ a separate constitutional order, this logically implies that the behaviour of the organization and its Member States should reflect this ‘truth’ instead. The precise types of actions that may be taken on the basis of this new knowledge will depend upon further discursive debates regarding the meaning of these truths in particular new contexts.
Spheres of jurisdiction.
A similar process can be seen with regard to the relationship between the WTO and other international organizations. In this context, there has been a great deal of concern about how the WTO should deal with multilateral environmental agreements, human rights treaties, international labour standards, regional trade mechanisms, and other sets of international norms and legal standards. This question is, in a sense, jurisdictional: what types of issues should be addressed by the WTO, and which should be addressed in some other international forum? What should be done in the case of a conflict of laws?
Using formal last-in-time or lex specialis rules to address these problems has been unsatisfactory to many, who argue that what is needed is the development of a broader ‘interfunctional constitution’.Footnote 40 And indeed, the existence of these tie-breaking standards has not prevented jurisdictional conflicts from arising. In Chile–Swordfish, for example, Chile and the EU filed contradicting claims in the WTO and UNCLOS – a situation that would have led to quite the international pickle if the parties had not eventually withdrawn the dispute.Footnote 41
In the 2006 EC–Biotech case, too, the EU and other nations were in conflict, inter alia, over the definition of the WTO’s sphere of competence within the broader system of international law. In particular, the question was whether the Convention on Biological Diversity and Biosafety Protocol could be used to interpret the SPS Agreement. The EU argued that:
The issues faced by the Panel have to be taken in their broader context. That context includes other relevant international instruments, which reflect the view of the international community as to the appropriate way to proceed on decision-making in relation to GMOs and GM products … . [A] failure by the Panel to have regard to this broader context will risk undermining the legitimacy of the WTO system. The Panel should therefore not accede to the Complainants’ arguments that this case may be decided in ‘clinical isolation’ from the rules of public international law more generally.Footnote 42
The EU’s intervention relies on an understanding of the WTO as part of a broader international order within which the scope of its jurisdiction should be limited. And within that broader ‘constitutional’ order, the EU would have the issue of GMOs fall under the purview of the Biosafety Protocol. The Panel ultimately rejected this approach and found against the EU on fairly narrow grounds. The dispute, however, had interesting implications for both the WTO and the EU. As Nico Krisch wrote, ‘the legitimacy of both institutions is relatively fragile, and they depend on cooperative relations to avoid serious challenges’.Footnote 43 As such, both have subsequently been cautious in making sweeping pronouncements on the GMO issue, and have cooperated, to a certain extent, within the bounds of their own ‘constitutional’ orders.
Because of its jurisdictional reach, the WTO will inevitably come into conflict with competing international regimes. And the boundaries between their fields of authority are far from self-evident. Where they are drawn in each case reflects particular understandings of the WTO’s constitutional mandate and its institutional telos. As Director-General Lamy argued, when the WTO ‘recognizes its limited competence and the specialization of other international organizations’ it ‘participates in the construction of international coherence and reinforces the international legal order’.Footnote 44 Determining where one regime ends and the next begins is a statement about the ‘constitutional’ structure of the WTO and the international system more broadly. For some scholars, like Sungjoon Cho, the solution lies in developing ‘a synergistic, nonentropic linkage within the constitutional structure of the global trading system’.Footnote 45 For others, the solutions to these border disputes should be sui generis, or pluralist with respect to state preferences.
Constitutional contests over the breadth and limits of the WTO system have an impact not only on how often these conflicts occur, but also on how they will be resolved when they do happen. These questions of course overlap with those discussed in the previous section, regarding the place of the WTO in the overall system of international law. The system, or lack thereof, for dealing with these encounters, will be developed on the basis of an ongoing, multi-level, mutually constitutive interplay among the WTO, states, commentators, and other international regimes, and will involve a clashing of discourses over the nature and limits of the WTO’s authority. And these ‘constitutional’ debates will in turn affect perceptions of how authority should be divided between the WTO and other international organizations, to what degree the WTO should take into account general principles of international law developed in ‘other’ regimes, and what should happen in the case of jurisdictional conflict.
These contests over the constitutional status of the WTO – both in terms of its existence as a self-consciously ‘constitutional’ or ‘constitutionalizing’ regime and its place within the broader international legal order and vis-à-vis other international bodies – are more productively seen not as exercises in positivist fact-finding, but rather as struggles for discursive influence. As one position or another becomes dominant, it shapes the accepted ‘truth’ about the way the organization works, what it is for, who its constituents are, and how it should progress. And as these discourses slowly become truth, they in turn produce the WTO as a locus of power, expanding and contracting its field of action.
Contesting constitutional form
A second area of constitutional contestation relates to the form of the ‘constitutional’ order. Disputed points include how a particular constitutional system should be set up, who it should see as its constituents, how those constituents should participate in the system, how it should manage the division of powers, how and when it should expand to address new problems or limit its own behaviour, and how it can maintain the proper balance among its various organs and between institutional authority and the will of its subjects. Different configurations of these elements will produce different allocations of power and different opportunities for intervention. Who is authorized to do what, when, and by whom, has implications for how the system will function, and what types of action will be deemed possible and effective.
In the WTO context, debates over constitutional form go back many decades. John Jackson, one of the most prominent advocates of international trade constitutionalism, has been writing about the issue in the GATT context at least since the 1960s.Footnote 46 Jackson’s advocacy and constitutional vision for the trading system were influential in the evolution of the WTO. To cite perhaps the most significant example, in 1990 Jackson published a widely read book called Restructuring the GATT System.Footnote 47 In that book, he proposed a constitutional status and structure as the best way to address the historical ‘birth defects’ of the GATT, such as the ‘provisional’ nature of trade obligations,Footnote 48 Contracting Parties’ ability to veto dispute settlement reports, and the difficulties resulting from the multiplicity of GATT-related texts. Such a constitutional movement, he argued, would help the international trading system to move forward along a civilizing trajectory: ‘To a large degree the history of civilization may be described as a gradual evolution from a power oriented approach, in the state of nature, towards a rule oriented approach’.Footnote 49 This forward movement, he argued, was essential both to move the trading system toward its rule-based telos, and to prevent it from collapsing into the ever-present danger of resurgent national protectionism. As Jackson wrote, a strong constitutional structure is necessary to allow for progress and adaptation, which are essential to the continued operation of the trade system.
It was partly as a result of Jackson’s work that the Uruguay Round eventually produced a draft Agreement Establishing the Multilateral Trade Organization, which culminated in the formation of the World Trade Organization.Footnote 50 The new WTO, along the lines Jackson had advocated, was organized around a more ‘constitutional’ structure, which included a Ministerial Conference, a General Council, a number of other councils and committees, an expanded set of commitments, and a highly developed dispute settlement system.
Scholars continue to debate the institutional architecture of the WTO from a constitutional perspective. One such ongoing contest over constitutional form is the question of whether and to what extent the WTO should permit interventions from non-state actors in the dispute settlement system. Deborah Cass has argued that the receivability of amicus briefs is a constitutional question for the Appellate Body.Footnote 51 The amicus question arose partially in response to popular protests like the ones that took place at the Seattle Ministerial Conference in 1999, and concerns about the ‘democratic deficit’ in the WTO and its consequent lack of sensitivity to non-trade policy preferences. These demands for a democratic shift in the WTO ‘constitution’ led the dispute settlement body to permit the acceptance of amicus briefs from non-state actors in Shrimp/Turtle and subsequent decisions.Footnote 52 Cass argues that this decision to expand its fact-finding powers demonstrates that the WTO’s Appellate Body ‘is the dynamic force behind constitution-building by virtue of its capacity to generate constitutional norms and structures during dispute resolution’.Footnote 53 The Appellate Body’s decision in the Shrimp/Turtle case to accept amicus briefs from non-state actors, Cass argues, ‘lend[s] credence to the constitutional claim and ha[s] significance from a democratic and constitutional design perspective’.Footnote 54 Because it increases the level of non-state participation in the trade system, she argues, it enhances the perceived legitimacy, fairness and public acceptance of the dispute settlement system’s decision-making power.Footnote 55
A third example of a contest over constitutional form is the evolving debate over the judicialization of the WTO dispute settlement system. Both before and following the movement from GATT to WTO, there was significant disagreement regarding the extent to which the dispute settlement bodies should be thought of as ‘adjudicatory bodies’ as opposed to ‘courts’; that is, over the extent to which Panels should be considered to have a legalistic, as opposed to a political-diplomatic, function.Footnote 56 With the entry into force of the WTO Agreements, the dispute settlement system made a significant leap in the ‘court’ direction, as the ability for losing parties to veto the adoption of reports was eliminated, and a new Appellate Body was constituted. The institution of the Appellate Body meant that there would be significantly more coherence in the system as a whole. However, in its early years the Appellate Body adopted a very narrow text-based approach, and there was (and is) still no formal doctrine of precedent at the WTO.Footnote 57 As its confidence and experience grew, however, the Appellate Body began to make bolder decisions, applying more abstract principles and demanding more coherence from Panels. This led to renewed conflicts over constitutional form: was the Appellate Body overstepping its bounds, upsetting the original contract between Member States and the WTO?
In the mid-2000s, the issues of precedent and textualism developed from a marginal skirmish into a substantial problem. In several cases regarding a rather obscure technical rule on the calculation of dumping margins (involving a methodology known as ‘zeroing’), Panels and the Appellate Body clashed over the WTO-legality of the practice, which was not explicitly prohibited in the WTO treaties, but which was, in the Appellate Body’s opinion, contrary to the spirit of WTO law.Footnote 58 In a surprising move, a subsequent Panel refused to take up the Appellate Body’s analysis of the issue, declining to find zeroing WTO-illegal on the grounds that this was not explicitly prohibited by the text of the treaties.Footnote 59 As the disputes went on, the Appellate Body overturned the Panel’s decision,Footnote 60 but a subsequent Panel also refused to abide by the Appellate Body’s interpretation of the treaties.Footnote 61 Scholars intervened on both sides of the issue, arguing either that the Appellate Body was engaging in extra-treaty rule-making, or that it was properly applying the spirit of the Agreements. In the end, it seems that the Appellate Body has – in fact, if not in theory or text – carved out a doctrine of precedent on the basis of its continual reversals of these contrary zeroing Panel reports, and also solidified its commitment to building the law of the WTO outside the explicit bounds of the treaties. Sungjoon Cho has referred to the zeroing debates as precipitating a ‘constitutional revolution’ in WTO practice.Footnote 62 As he argues:
This critical choice flows from the AB’s [Appellate Body’s] awareness of the immediate and powerful normative consequences that would affect the future of the WTO. In other words, the AB was well aware that the AB’s adjudication would ‘(re) constitute’ the WTO, at least as far as antidumping is concerned. Here, the AB departed from a conventional role as a triadic settler, or arbiter, of disputes and instead assumes the innovative role of a ‘constitutional court’.Footnote 63
Each of these interventions contested the constitutional form of the WTO as a way of correcting what they saw as defects in the international trading system. Jackson wanted to move toward a rule-based order to counteract historical and threatened abuses of power politics by states, and, indeed, it was just such threats (in particular the US’s assertion that it would use its Section 301 provisions to unilaterally impose trade disciplines outside the WTO) that led the Member States to agree to a drastic shift in constitutional form. Cass’s focus on the democratizing moves of the dispute settlement system highlighted the ways in which ‘constitutionalization’ has helped to alleviate the democratic deficit, and how, vice versa, discourse about the democratic deficit led to shifts in constitutional form. Now, scholarly debates over the issue of precedent and the judicialization of the dispute settlement body that have focused on concern over the WTO / Member State power balance and concerns that increased judicial ‘law-making’ would harm the WTO’s legitimacy are helping to construct new narratives about how the Appellate Body does and should behave. As each of these discourses produces knowledge about the world – including truths about the threat of power politics, the democratic deficit, and the fait accompli of Appellate Body review – the WTO’s ‘constitutional’ form has realigned toward a new institutional configuration. These constitutional disputes about the way in which the international trade system should work become knowledge about the way the WTO system does work and the threats that it faces, and eventually lead to shifts in its architecture to reflect these new realities.
Contesting constitutional substance
A third set of discursive contests has to do with the content of the constitutional order. To define a set of values as ‘constitutional’ is to say that they are of systemic importance. ‘Constitutional’ values are not just ‘good things’ that an order may desire, but crucial things that keep the order from collapsing. For example, in democratic systems it is often argued that constitutional values (such as the rule of law, or fundamental rights) protect subjects against abuses of power, and protect governments from short-sighted popular movements. Calling a particular value ‘constitutional’ is to assign that value weight and power within the systemic order. Indeed, some have argued that calling a value constitutional is so consequential that to constitutionalize is to depoliticize an issue – to remove it from the political arena of opinion and choice and transfer it to the realm of fact and necessity.Footnote 64 Labelling a particular value ‘constitutional’ is not just a description: it is a normative signal that serves to legitimate that value and grant it recognition as politically indispensable. Commitments that are regarded as ‘constitutional’ will be able to insert themselves more globally into debates, will be more convincing when used as arguments, and will be able to justify practical action in line with their description of the world. Conversely, to refuse to name a particular value ‘constitutional’ is to assert its contingency, its lack of significance, its peripheral nature. Commitments that are not ‘constitutional’, while potentially important or desirable, will be considered secondary to ‘constitutional’ values, will be less convincing when used as arguments, and will be viewed as ‘debatable’ by reasonable persons.
In the WTO context, contests over constitutional substance have centred on questions of the autonomy of international economic law and whether and to what extent the WTO should consider the ‘non-trade’ aspects of trade law. There has been significant disagreement over whether the WTO should incorporate norms on human rights, environment, labour, and other topics into its understanding of international trade law. This issue has become increasingly important as the WTO has moved from a system of rules prohibiting certain trade measures to a system of rules requiring affirmative ‘behind the border’ action by governments. The result of this shift has been more WTO monitoring and disciplining of domestic rules that may not be primarily trade-focused. As these trade regulations intrude further into domestic policy spheres, scholars, states and the dispute settlement bodies have struggled with when and how to incorporate and accommodate non-trade policy goals in WTO law.
Theorists of international trade have devoted many articles to debating the substance of the WTO constitution. In particular, these conflicts have focused on what are sometimes referred to as ‘trade and …’ issues. An interesting example from this category is the linkage between trade and human rights. On one side of this debate sit scholars like Ernst-Ulrich Petersmann, who has championed the ‘constitutional primacy of the inalienable core of human rights’ – in particular ‘economic freedoms’ such as a ‘right to trade’ – which he argues should be applied at a constitutional level in the WTO context.Footnote 65 Incorporating human rights norms into the trade regime, he and others argue, would increase the legitimacy of the organization and promote coherence in international law. In opposition to this position stand trade purists, who reject any insertion of non-trade concerns into international economic law (though there are fewer of these to be found each passing year), as well as more subtle writers like Philip Alston and Robert Howse, who warn of the dangers of bringing human rights into trade because of the detrimental effects such marriages might have for the interpretation and adjudication of human rights principles.Footnote 66 As Philip Alston wrote:
[T]his process of human rights-based (or more accurately human rights justified) ‘constitutionalization’ of the WTO is a highly contentious one. While it is true that some human rights, and many labour rights, proponents would like to see a significant role for the Organization in these respects … they certainly do not see it as an Organization which is designed, structured, or suitable to operate in the way that one with major human rights responsibilities would. The Agreement Establishing the WTO is not a constitutional instrument in the sense of constituting a political or social community, and its mandate and objectives are narrowly focused around the goal of ‘expanding the production of and trade in goods and services.’Footnote 67
Similar debates have taken place over whether and how trade law should be linked with environmental norms,Footnote 68 labour standards,Footnote 69 and other issues.
WTO Member States have not been blind to these debates. At the Doha Ministerial in 2001, for example, the WTO set out an agenda for the Doha Development Round that included room for discussion of a range of ‘trade and …’ issues including the environment, investment, and competition.Footnote 70 These topics have remained highly contentious, however. Developing countries, in particular, have been reluctant to see environmental, human rights, and labour brought more directly into the WTO context. Many of these states fear that such social issues will be used as a basis for protectionism and denial of market access.Footnote 71 Other states argue that the failure to incorporate these issues in the WTO context may lead the dispute settlement body to unacceptably limit domestic policy space.
The dispute settlement system, in the meantime, has been grappling with these issues in its own somewhat chaotic way. One example is the incorporation of ‘outside’ international law into the WTO dispute settlement system. Though the Appellate Body famously stated in US–Gasoline that ‘the GATT is not to be read in clinical isolation from public international law’,Footnote 72 it is not clear which international law, or how it should be incorporated into the WTO system. In the EC–Beef Hormones case, for example, the Appellate Body concluded that whatever status the precautionary principle had ‘under international environmental law’, it was not recognized by the WTO.Footnote 73 In the EC–Biotech case mentioned above, the Panel had to determine whether the 1992 Convention on Biological Diversity and its Biosafety Protocol should be taken into account when interpreting the SPS Agreement. It declined to do so on the grounds that because the US was not a part of the Biosafety Protocol, it could not be incorporated into the decision.Footnote 74 In the Shrimp/Turtle case, on the other hand, the Appellate Body referred to an international environmental agreement in order to assist its interpretation of GATT Article XX.
Another contest regarding the substance of the WTO constitution is the debate over what factors should be assessed as part of the ‘like products’ analysis. In brief, the issue is whether dispute settlement bodies should examine a measure’s regulatory context when determining whether products are ‘like’ for the purposes of most favoured nation and national treatment provisions. As the years have progressed, the ‘like products’ analysis has shifted back and forth from a focus on whether products are competitive as the most important criteria for determining whether they should be treated alike, to a view that takes into account the aims or purposes of a challenged regulation.Footnote 75 Many scholars have criticized the traditional test’s focus on competitiveness. They argue that the competitiveness test is ‘too intrusive’,Footnote 76 and that it affects the ‘very symbolism of political identity’ because it ‘establishes a normative hierarchy, whereby the default norm is liberalized trade, and, for competing norms to prevail, they have to be justified’.Footnote 77 In the GATT context, opponents also argue that the competitiveness test unnecessarily limits ‘legitimate’ domestic policy considerations to those listed in GATT Article XX.Footnote 78 Instead, they argue that the likeness test should be more deferential to Member State regulatory autonomy by including a consideration of public policy objectives.Footnote 79 This would shift the default value from ‘competition’ to ‘autonomy of political and moral identity which requires justification only if purposefully abused’.Footnote 80
These debates, over ‘trade and …’ topics, the inclusion of non-trade international law in dispute settlement proceedings, and the proper evaluation of ‘like products’, exemplify the way in which states, scholars, and the WTO dispute settlement bodies have struggled with the issue of constitutional substance. As they have developed, these discursive themes have produced new forms of knowledge about what the substance of the WTO constitution is and what it should be. As the assertion that the WTO ‘cannot avoid’ environmental, human rights and other issues has become a ‘truth’, it has led to shifts in the way the WTO’s substantive constitution is understood. This has led to the inclusion of environmental and other ‘non-trade’ issues for discussion in the Doha Round as well as the recognition of certain non-trade values by the dispute settlement body, and the incorporation of ‘constitutional’ understandings of the WTO’s mandate into dispute settlement.Footnote 81 Whether the ‘solution’ to this issue is to expand the WTO’s substantive constitution or to shrink it, the ‘truth’ that intersections between trade and other issue areas are inevitable produces particular understandings of the organization’s field of action, and the choices that are available to it.
Conclusion
As the previous Parts have demonstrated, debates about constitutionalism have formed an important subject of academic and political debate at the WTO. Constitutional claims are rhetorically powerful, and carry with them important sets of associations and normative implications. At the same time, however, the term is indeterminate and open enough that it can support a huge range of different interpretations, definitions, and practical consequences. Because of this simultaneous power and openness, the discursive outcome of debates about the constitutional status, form, and substance of the WTO can be very influential. As certain forms of knowledge about the organization become accepted as truth, they change the way that we think about the WTO and its ability to act. And these beliefs, in turn, lead to shifts in the way that the organization behaves.
It has not been the intent of this paper to argue that any individual attempt to define the analytical or normative content of constitutionalism is ‘correct’ or ‘incorrect’. Rather, it has sought to show that contests over the meaning of constitutionalism are not neutral, but rather carry with them a whole set of assumptions about how power functions, what the proper objects of government should be, and how the subjects of that government should and do conduct themselves.
Thinking about constitutionalism as a discursive contest, rather than as an empirical fact, allows us to trace the ways in which these discussions attempt to influence the lines of power and knowledge in international economic law. As such, it helps us to identify the function of ‘constitutional talk’, and the importance of discourse in shaping the international world. It also allows us to understand how the proliferation of definitions attached to constitutionalism can be of great strategic benefit, and how the failure to agree on a single meaning opens up further possibilities for intervention and adaptation. It permits us, in other words, to get a glimpse of all the things that Alice still has to learn.
Acknowledgements
I would like to thank Wouter Werner, Gareth Davies, Juan Amaya-Castro, Laurens Ankersmit and Nikolas Rajkovic, as well as the two anonymous reviewers for Global Constitutionalism for their helpful comments. An early version of this article was presented at the COST Action PhD Training School ‘The Vices and Virtues of International Constitutionalism’ at the European University Institute in Florence in October 2011. My thanks are due also to all of the participants in the stimulating discussions that took place both during and following this event.
