1. Introduction
The Dispute Settlement System (DSS) of the World Trade Organisation (WTO) is an interesting mechanism to settle international trade controversies by means of argumentative procedures. At least part of its interest may be accounted for by the fact that it combines, in an original way, procedural rules belonging to the ‘furniture’ of other international legal organizations and well-established dispute settlement (DS) methods (i.e. consultation, inquiry, mediation, arbitration, and the like). But, why is the DSS argumentative in nature and articulated through such sophisticated procedures?
This kind of question is not new in the literature dealing with institutional design problems.Footnote 1 However, most of the answers have addressed the matter from purely legal, political, or economic perspectives, and have concentrated on specific substantive issues (e.g. the efficiency of the DSS, its capacity to deliver substantially fair outcomes, or the effects of DS rulings on low-income countries). This paper aims to combine studies in the fields of politics, law, and economics through philosophical analysis of legal texts to look for a systemic answer to the above question in the inherent qualities of the procedures through which the DSS is articulated.
In order to make sense of the DSS's distinctive features, we propose resorting to Jürgen Habermas's discourse theory, as a hermeneutic device to disentangle the different kinds of orientations DS procedures may have (compromise, consensus, and mutual understanding). This approach leaves aside the features of the contingent outcomes to which DS procedures may lead, and concentrates on how differently oriented procedures shape the interaction between the parties. Such a procedural approach has the advantage of being suitable for application across different disputes independently of their substance, which will almost always not be comparable across cases.Footnote 2
The paper unfolds as follows. In Section 2, we shall try to explain what makes the WTO an interesting framework in which to study DS mechanisms. Section 3 will present and distinguish the classes of disputes to which the DSS applies. Section 4 will briefly introduce the reader to the structure of the DSS.
Section 5 will be devoted to explaining why Habermas's discourse theory is suitable for capturing the nature of disputes within the WTO and the distinctive features of the procedures to settle them. Against this theoretical backdrop, we shall introduce and explain the different orientations argumentative procedures may have, according to whether they put the parties in a position to reach compromise, consensus, or mutual understanding (Section 6). This analysis will allow us to show, in Section 7, how in the DSS a commitment to put the parties in a position to reach mutual understanding seems to have priority over alternative procedural orientations. Such a commitment serves the general aims of the WTO presented in Section 2, and contributes to the settlement of the classes of disputes identified in Section 3. In Section 8, we shall address some problems that may occur once DS procedures are translated from texts into practice. On consideration of the proposals for reforming the DSS to address such problems, we shall suggest that negotiators should take into account the impact that any proposal would have on the orientation of DS procedures. Section 9 will draw the relevant conclusions.
We would like to emphasize, in sum, that our analysis does not aim to offer recommendations on the ideal structuring of the WTO DSS in order to deliver ‘better’ outcomes (however, one may want to qualify them). We intend, rather, to investigate how it has been devised, with a view to showing the purposeful relations between its procedural features, the WTO's distinctive general aims, and the various types of disputes between Members. This said, as we shall show, our procedural analysis has significant bearings on the reform prospects of the DSS and is thus not free of normative implications.
Before proceeding, and in order to restrict our area of concern, some qualifications are in order regarding (i) our reasons for adopting a procedural and textual approach to the analysis of the DSS, (ii) the extent of our interest in Habermas's discourse theory, and (iii) the scope of our considerations on the WTO.
On (i), a procedural approach, which is by its very nature suitable for application across diverse contexts, seems particularly apt to make sense of a dispute settlement mechanism whose main function is to provide a coherent system of rules and procedures to regulate the interaction between the parties engaged in disputes varying across contexts and over time. Such a procedural approach requires that specific attention be paid to the inherent qualities of the procedures, rather than to their application to, or to the outcomes of, the cases brought before WTO adjudicating bodies.Footnote 3 Although the analysis of specific cases could reveal important pieces of information on the practical functioning and effectiveness of the DSS, it would be too exposed to contingencies to account for the reasons why the DSS is structured the way it is. A better strategy seems to resort to textual analysis, for it is in legal texts that procedures are spelled out in their purest and intelligible form, guarded against the possible distortions to which their contingent application may lead. Therefore, our analysis shall draw mainly on the Understanding on Rules and Procedures Governing the Settlement of Disputes (hereafter DSU), the Rules of Conduct for the Understanding on Rules and Procedures Governing the Settlement of Disputes, and the Working Procedures for Appellate Review (hereafter AB Rules).Footnote 4 References to cases will nonetheless be made at strategic points by way of examples to illustrate specific aspects.
As far as point (ii) is concerned, the idea of making sense of the distinctive features of the DSS through a Habermas-inspired discourse-theoretical lens looks promising as the WTO agreements may be seen as a sort of ‘speech’: a formalized context of interaction governed by mutually agreed norms. As discourse is triggered, in Habermas's view, as a normative tool to renew the background agreement during a linguistic interaction, its function comes very close to that granted by the DSS within the WTO (see Section 5).
Despite the obvious differences between Habermas's discourse (a linguistic coordination setting) and the WTO, we think it nonetheless promising to employ a Habermasian (although not, strictly speaking, Habermas's) conceptual framework because, as we shall endeavour to show, similarities outweigh differences by and large. To wit, the considerations we shall offer are meant to ‘plunder’ Habermas's work in order to make sense of the DSS's distinctive features.
Finally, on point (iii), we should like to emphasize that this is not a paper on the WTO as a whole, nor one meant to assess the (either actual or intended) fairness of its procedures (both ordinary and those meant for settling disputes).Footnote 5 In particular, we shall not discuss those procedures regulating the negotiation of new agreements or those devoted to ‘housekeeping’ operations, or to setting the WTO's agenda. As stated, the aim of this paper is rather to disentangle the procedural qualities and orientations of the mechanism to settle trade disputes on issues covered by the WTO agreements.Footnote 6
2. The WTO as a laboratory to study the settlement of international trade controversies
There is not much disagreement in the economic literature on the beneficial implications of free trade for the welfare of trading partners. Nonetheless, countries face a Prisoner's Dilemma in deciding to liberalize exchanges. Even though a cooperative solution of (at least) partial liberalization is Pareto optimal, it would be unlikely to result out of individual uncoordinated decisions.
A cooperative trade agreement is a means of overcoming this problem. Parties agree to give market access to their partners in exchange for (a) mutual concessions and (b) a commitment to abide by certain rules. Mutual concessions ensure mainly that all parties gain from the agreement, prevent strategic liberalization, and strengthen the domestic political constituencies in favour of trade. The commitment to abide by rules provides national governments with international constraints on the actionable set of policies to which they can resort. Furthermore, governments value reciprocal concessions because, by negotiating foreign liberalization, they can claim credit for the gains accruing to domestic exporters (see Ethier, Reference Ethier2004).
Even so, parties may still have incentives to deviate unilaterally from what was agreed. This reveals a problem of enforcement of international trade agreements (see Bagwell and Staiger, Reference Bagwell and Staiger2002). In theory, if trade is mutually beneficial and international exchanges are repeated over time, the prospective threat of retaliation by the affected foreign partners can be sufficient to render the agreement self-enforcing. In practice, however, tacit and self-enforcing trade agreements are hardly conceivable. Contract incompleteness and the presence of transaction costs require the establishment of institutions serving to clarify the scope of the agreements and to allow parties to deal with contingencies by means of temporary ad hoc deviations from some of their commitments.Footnote 7 Institutions are thus necessary to support efficient market allocations, to promote the peaceful cooperation of states, to balance the different goals built into the agreements, and to boost compliance. Should this latter fail, institutions are also required to facilitate the peaceful settlement of trade disputes (on this see Petersmann, Reference Petersmann, Ortino and Petersmann2004: 5).
A dispute settlement mechanism can contribute to compliance in two ways: either boosting the enforceability or strengthening the enforcement capacity of the parties. The latter refers to the actual capacity of the complaining party to retaliate credibly against a damaging infringement of a relevant agreement. On the other hand, enforceability can be defined, following Keck and Schropp (Reference Keck and Schropp2008), as the ability to identify a violation of the terms of the contract and is a function of the observability of contract infringements, of the independent verifiability of the legal basis of a claim, and of the quantifiability of damages incurred. Therefore, a DS procedure is part and parcel of any successful trade agreement: by facilitating their enforcement, it also makes agreements more likely, comprehensive, and long lasting.
2.1 The WTO DSS and its distinctive features
The WTO is not just an example of institutions born to serve multilateral agreements, but a case in point. To start, the WTO is a distinctive and comprehensive community of parties entertaining stable, repeated, and secure relationships. Cooperation between Members is not just a means of achieving some external goals – as is the case, say, in environmental accords (e.g. Kyoto Protocol) – but a goal in itself. Moreover, this community is grounded on politically, mutually recognized agreements, which provide for an overall balance of rights and obligations for its Members, and whose respect is ensured by a sophisticated DSS. Based on the DSU, encompassed in the package of agreements establishing the WTO, the DSS is one of the few mandatory DS systems applying to multilateral agreements.
WTO DS procedures are geared to prevent a dispute between some Members and its ungoverned settlement from corrupting the cooperative overall spirit of the agreements. For this reason, the DSS combines well-established procedural rules and methods of DS in an original ‘vertical’ DS system to reach a certain, prompt, and acceptable settlement of controversies.
More precisely, the DSS serves to: (i) provide ‘security and predictability to the multilateral trading system’ (Article 3.2 of the DSU); (ii) ‘preserve the rights and obligations of Members under the covered agreements’ (Article 3.2 of the DSU), so that a ‘proper balance between the rights and obligations of Members’ is maintained (Article 3.3 of the DSU); (iii) ‘clarify the existing provisions of those agreements in accordance with customary rules of interpretation of public international law’ (Article 3.2 of the DSU); (iv) guarantee rule-oriented rather than power-oriented DS procedures; (v) ensure a positive and prompt settlement of disputes; and (vi) foster the adoption of mutually acceptable solutions consistent with the covered agreements (Article 3.7 of the DSU).
As we shall see in Section 8, the presence of the DSS does not improve the enforcement capacity of the Members, as defined above, which remains inevitably dependent on their relative power and on the willingness of their counterparts to comply. Nonetheless, as Keck and Schropp (Reference Keck and Schropp2008: 804) maintain, it can ‘contribute to successful enforcement by enhancing enforceability’, insofar as it establishes mutually acceptable and binding procedures to (i) weigh claims and rebuttals, (ii) obtain and process information, and (iii) interpret the covered agreements. This characterization strengthens the importance of distinguishing and analyzing the procedural components of the DSS (see Sacerdoti, Reference Sacerdoti, Sacerdoti, Yanovich and Bohanes2006).
3. A typology of the disputes within the WTO
In what follows, we shall illustrate four classes of disputes that may occur between WTO Members. This presentation will be essential in Section 7 where we discuss the orientations of DS procedures. It goes without saying that actual disputes rarely fit into a single type entirely, but mixed cases are most likely to be found.
The first class has regard to disputes arising because of the existence of gaps in the text of the WTO agreements, which, like any other complex multilateral treaty, have an inherent ‘open-texture’ (Hart, Reference Hart1994: 135). Agreements are naturally incomplete in that they cannot possibly foresee all states of the world to which their provisions may contingently apply. Furthermore, Members may well fail to reach an agreement on certain issues and, therefore, deliberately leave the text silent. Thus, a gap remains to be filled.Footnote 8
The second class includes those disputes arising from ambiguities in some provisions of the WTO agreements. Often procedural and substantive terms and rules carry an intrinsic vagueness.Footnote 9 Furthermore, the WTO agreements pursue multiple goals, some of which are subject to other international agreements, and whose conceptualization has been internationally understood in different ways over time.Footnote 10 When ambiguity-related problems occur, Members may take measures that end up impairing others' rights and benefits.Footnote 11
The third class of disputes refers to disagreements among Members regarding the conformity of their national measures with WTO rules.Footnote 12 An example is when a trade restrictive measure is imposed by a Member in the conviction of their being permitted to act in such a way by the numerous exceptions and safeguard clauses contained in the WTO agreements. This case differs from those above as the parties do not face a textual disagreement due to gaps or vague terms, but disagree on whether a measure is actually justified by a right under the agreements and fulfils the established requirements.Footnote 13
The fourth class of disputes encompasses those controversies in which parties act in an instrumental way, that is either knowingly defending a patently WTO-inconsistent measure, or strategically complaining against one that is in fact acceptable, or intentionally misinterpreting the text of the covered agreements. These disputes may present themselves in the guise of those belonging to the classes presented above, but – unlike these latter – they would violate the requirement that parties enter the DSS in good faith. We shall not engage with this class of disputes as they are revealing of an improper use of the DSS.
It could be argued that, far from being specific to the WTO, most of the disputes arising between parties bound by an international agreement would fall into one of these classes. However, the disputes occurring within the WTO are characterized further by the following features. They concern the application and interpretation of provisions that are the subjects of previously established agreements between the Members. And, they regard domestic national measures which affect other Members' benefits, not international actions which infringe directly on the sovereignty of other Members, as happens in many international disputes.
The typology of disputes will be revisited at several stages throughout the paper and will play a crucial role in revealing the overlaps between the areas of concern of Habermas's theory and the WTO DSS, with a view to disentangling the components of the latter through the lenses provided by the former (Sections 5.2 and 7). To offer a basis for our proposed discourse-theoretical analysis, a preliminary, concise description of the DSS's basic components is in order. The following section aims to fulfil this function. The reader who is familiar with the provisions in the DSU could proceed to Section 5.
4. The WTO Dispute Settlement System: an overview
4.1 WTO consultations between Members
WTO DS procedures must be entered by WTO Members when one of them challenges the validity of a measure undertaken by another Member, and bilateral consultations between them have either failed or not occurred. In order to set the DS procedure in motion, the request to enter WTO consultations must be formally made in writing and indicate the challenged measure and the legal basis of the complaint (Article 4.5 of the DSU). The request must be addressed to the Dispute Settlement Body (DSB): the organ administering the DSS in which all WTO Members are represented. Being the sole politically representative body in the DSS, the DSB retains the exclusive right to make decisions which are binding on its Members.Footnote 14
The Dispute Settlement Understanding defines clear time limits for consultations, whose inception and outcomes must be notified to the DSB and other WTO bodies (Articles 4.3, 4.7, and 4.8 of the DSU). Consultations are confidential and occur without prejudice to the rights of the parties in any further proceedings (Article 4.6 of the DSU). Other Members may join consultations as third parties if they have ‘substantial trade interests’ and provided that directly involved parties agree. If the latter do not agree, new consultations can be established a latere.
Should WTO consultations not settle the dispute, four scenarios are possible: (i) the complainant drops the case; (ii) the parties seek to resolve their differences through good offices, conciliation, and mediation; (iii) the parties agree to resort to the flexible arbitration procedure provided by the Article 25 of the DSU;Footnote 15 (iv) the complaining party requests the DSB to establish a panel to adjudicate the case (Article 4.7 of the DSU).
In what follows, we shall concentrate on scenario (iv), in the event of which the DSS is fully triggered.
4.2 Panels and the adjudication of disputes
A panel is an ad hoc (non-standing) WTO adjudicating body consisting of three to five members chosen from an indicative list of suitably qualified persons (Article 8.1 of the DSU). A panel's official aim is to assist the DSB by drafting and submitting a report giving an objective assessment of facts and law on the disputed matter. The DSB retains the sole authority to make the final decision on the dispute by adopting or not adopting the panel's report.
Panels are required to consult regularly with the parties, to give them ‘adequate opportunity to develop a mutually satisfactory solution’, and ‘to make such other findings as will assist the DSB in making the recommendations and in giving the rulings provided for in the agreements’ (Article 11 of the DSU).
If a measure is contended by several parties, a single panel may be established (Article 9.1 of the DSU) provided that it takes the rights of all parties into consideration. In addition, when a WTO Member has a ‘substantial interest’ in an existing dispute, it can ask at any time to participate in the panel's works as a third party (Article 10 of the DSU).Footnote 16
Panels establish their own working procedures and a timetable for the various written submissions and the meetings with the parties. Most of the time, first written submissions are not simultaneous. The complainant's first submissions contain factual information regarding the disputed measure and the exposition of the legal theory underpinning the assessment of its inconsistency with WTO provisions. The respondent usually files its submissions later (either as a rebuttal or as an affirmative defence) and exposes its factual view of the disputed measure. Second written submissions are, instead, usually simultaneous (Article 12.6 of the DSU). They take the form of rebuttal submissions to both the first submissions and the oral statements at the first meeting. Additional submissions are possible in the form of answers to questions presented by the panel and by the other party. All submissions are made available to all parties (Article 10 of the DSU Appendix 3).
At the second meeting, the parties provide rebuttal arguments and the panel may review any issue at will. This is the last opportunity for the parties to make their case, as much as for the panel to ask questions and request additional information.Footnote 17 Besides parties' submissions, a panel can use (seek and receive) information provided by a third party and solicit ‘information and technical advice from any individual or body it deems appropriate’ (Article 13.1 of the DSU). It can also ask for an advisory opinion from an expert review group on certain aspects of the matter (Article 13.2 of the DSU). However, panels cannot use their investigating power to ‘make the case for the complaining party’.
After the written submissions and the oral hearings, the panel releases a draft report to the parties for comment, setting out the facts and the parties' arguments. An interim report follows. It contains a descriptive section and the panel's findings and conclusions (Article 15.1 of the DSU). The panel submits the interim report to the parties for comment. The parties may submit written requests to review specific parts, and require an additional meeting to review the issues identified in the requests (Article 15.2 of the DSU and Art. 12 of the DSU Appendix 3). If no review is required, the interim report becomes final. If the parties have not reached a mutually satisfactory solution, the final report would be detailed and explain the basic rationale behind any findings and recommendations (Article 12.7 of the DSU).
To be binding, the final report must be adopted by the DSB. Should any Member (including the parties in dispute) have objections, they must give their reasons in writing before the DSB meets and discusses the relevant report. The report is ‘automatically’ adopted by the DSB at a meeting occurring within 60 days of the circulation of the report, unless either there is consensus on blocking its adoption or if a party notifies its intention to appeal (Article 16 of the DSU).Footnote 18 Note that the DSB works under the principle of ‘reverse (or negative) consensus’: the DSB cannot refuse the adoption of a panel's report unless there is a unanimous consensus to the contrary.
4.3 Appellate review under the DSS
To counterbalance the automaticity of these procedures, the DSU provides for an appeal on points of law (Article 17.6 of the DSU). A permanent Appellate Body (AB) is established to perform the function of appellate review. In each appellate case, three of the seven members of the AB (a Division) are selected on the basis of a non-disclosed rotation.
Only the original parties in a dispute can appeal to the AB (Article 17.4 of the DSU). A party wishing to appeal has to notify the DSB in writing (Article 16.4 of the DSU) and file a Notice with the Secretariat (AB Rule 20.1) before a meeting of the DSB is held to adopt the panel report. This Notice must contain certain information including a brief (but precise) statement of the nature of the appeal (AB Rule 20.2).
The AB follows its own Working Procedures. To ensure ‘collegiality’, all AB members receive the documents on every case (Article 17.3 of the DSU) and can express comments and exchange views before the Division submits the report to the DSB (AB Rules 4.2 and 4.3).
The Appellant must file its written submission with the Secretariat and serve a copy to the other parties of the dispute and to third parties. The Appellant's submission must contain all information necessary to determine the nature and scope of the appeal (AB Rule 21.2). Similarly, Appellees wishing to participate have to file a written Appellee's submission containing detailed information (AB Rule 22).
According to AB Rules, an oral hearing is to be held in each appeal and all participants are given the chance of an opening statement (AB Rule 27). Members of the Division may address questions and request memoranda from all participants, to whom questions and responses are circulated.
Participants may withdraw from an appeal, in which case the appeal is dismissed (AB Rule 30.1). If they find a mutually agreed solution, this must be notified to the DSB and the AB, which would drop the case (AB Rule 30.2). Otherwise, the AB has to draw up a report with its findings and submit it to the DSB. The report is accepted unless the DSB decides by consensus not to adopt it (Article 17.4 of the DSU).
4.4 The implementation of recommendations and rulings
Clearly, the implementation of recommendations and rulings contained in the adopted reports is crucial for the enforcement of the WTO agreements. The DSU contains provisions dealing with the implementation and with the lack thereof. We shall describe them briefly since the main text focuses principally on the adjudicating phase.
Usually, the panel's and AB's reports contain the recommendation that the Member concerned brings its behaviour into conformity with the agreements. With the adoption of a report, the DSB is in charge of the surveillance over the compliance with recommendations and rulings (Article 21 of the DSU).
If immediate and full implementation is not possible, and the Member cannot bring a measure into conformity with the covered agreements, it is given a ‘reasonable period of time’ to comply (Article 21.3 of the DSU). If disputes arise on whether the measures taken to comply with the rulings are sufficient and/or consistent with the agreements, they must be resolved through panel adjudication (Article 21.5 of the DSU).
According to the Article 22 of the DSU, should a Member refuse or fail to comply with recommendations and rulings, two temporary remedies are available: the defaulting Member can offer compensation to the other party or, failing a ‘mutually acceptable compensation’, the aggrieved party can request authorization from the DSB to suspend the concessions made to the Member in violation (retaliation).Footnote 19
5. A Habermasian discourse-theoretical interpretation of the WTO DSS
As emerges from the characterization in the previous section, the establishment of argumentative procedures inspired by some commitment to the principles of procedural equality and fair hearing seems to be at the heart of the WTO DSS. However, its sophisticated procedures are too nuanced to be accounted for in virtue of the sole commitment to the principle of due process. The remainder of the paper tries to make sense of the DSS's distinctive features, in a way consistent with the context sketched in Sections 2 and 3 above. To this end, we propose to disentangle the procedural components of the DSS with the aid of a philosophical analysis inspired by Jürgen Habermas's theory of discourse.
5.1 Habermas's discourse theory: basic ideas
Discourse theory is a part of the theory of communicative action, through which Habermas has studied the linguistic interactions between partners in a social context. In particular, Habermas's aim is to study the ability of linguistic exchanges to ‘coordinate action in a consensual or cooperative way as opposed to a forced or manipulated one’ (Warnke, Reference Warnke and White1995: 120).
When using language communicatively, in Habermas's view, social agents aim to reach a common interpretation of a situation and of their respective claims with an ultimate view to harmonizing the pursuit of their individual and/or shared goals. In short, communicative action is characterized by a use of language aimed to reach a mutual understanding between the agents. Every time a communicative statement is made, the speakers commit themselves to support their claims with reasons – especially if they are to respond to a critique. Correspondingly, the hearers sustain their positions by reasons. On this ground, communication is successful when one agent understands another and the implications of her claims (see Habermas Reference Habermas1984 and 1998 – for a more detailed presentation of the idea of understanding see Section 6.3 below). For such an understanding to be possible, communicative action presupposes a shared basis of norms and well-known facts (Habermas, 1995). In other words, every communication between two or more agents always presupposes a consensual context of interaction: (i) shared rules, following which communication can be carried out, and (ii) background knowledge, in light of which what is communicated may be understood (Habermas, 1996: 35).
Communicative action, thus characterized, is contrasted with strategic action. When an agent uses language strategically, her primary aim is to reach success: to influence her opponents and exploit their desires and fears to achieve whatever goal she has. This does not mean that agents engaged in communicative action are not motivated to pursue their goals. They are so motivated, but – unlike strategic actors – they are ready to accept constraints on their conduct in a linguistic interaction. In particular, to use Habermas's words, ‘they pursue their individual goals under the condition that they can harmonize their plans of action on the basis of the common situation definitions’ (Habermas, Reference Habermas1984: 286).
When disagreement arises and the consensual context of interaction is questioned, communicative action is disrupted and may be restored only through discourse. Whilst the theory of communicative action had a distinctively explicative intent (how do people use language to coordinate their actions?), discourse theory aims to outline a normative ethics of communication under idealized conditions (how should people interact to coordinate their actions through language?). Within a discourse setting, certain conditions must hold, namely that all those who can make a relevant contribution be allowed an equal chance to stand up and make their claims, have the freedom to say what they are willing to say, and the duty to listen to the others. Equality here translates into two aspects: on the one hand, it implies the negative requirement that participants be given the chance and room to speak. On the other hand, agents are positively required to listen, respond, and justify their positions to each other. On this basis, a condition of fair play is also added to the invitation (addressed to the agents) to renounce behaving strategically and trying to persuade their partners by all means, including deception. Accordingly, all sources of coercion built into the procedures of discourse must be removed (Habermas, 2005: 89).
In sum, rules immunize discourse against coercion and inequality. Agents should convince their partners in communication by rational persuasion alone, by what Habermas has called ‘the unforced force of the better argument’. Such conditions should also represent a sort of warranty that there will be an opportunity for all relevant information and arguments to be presented and to have a chance to exercise some influence on the final outcome (Habermas, Reference Habermas1984).
Habermas is fully aware that conditions of full inclusion, equality and non-coercion can hardly be realized in practice (Habermas, 1990: 92). However, they still play an important regulative and critical role: outcomes (whatever their substance) may be considered acceptable if scrutiny of the process by which they were attained does not reveal obvious exclusions, coercion, and suppression of arguments (Habermas, 2003: 108, 2005: 91).
5.2 Discourse theory and the WTO
If, building on the characterization above, one considers the WTO agreements as rules of communication (as they serve a function of action-coordination), the DSS may be assimilated (mutatis mutandis) to a sort of discourse, which enters the picture when disputes arise in order to restore harmony and compliance with the agreed norms on mutually acceptable grounds. As explained, the primary function of discourse is to protect the interaction of social agents from instrumental and coercive uses of unilateral power, through a system of rules applicable across a range of diverse cases. This is certainly in line with the commitment to transform a power-oriented DS mechanism (as under the GATT) into a rule-oriented system within the WTO.Footnote 20
More precisely, within the DSS the conditions of equality and inclusion are realized by the requirements that all Members (despite their unequal relative power): (i) are given an equal opportunity to resort to the DSS, (ii) share equal opportunities to make their voices heard as parties (or third parties) to a dispute and qua members of the Dispute Settlement Body, and (iii) are bound by the same agreements and procedures.
Moreover, non-coercion is ensured at different stages. As outlined in Section 4, with a view to assisting the parties to find a mutually acceptable solution, the adjudicating bodies should consult with the parties at all stages (Article 11 of the DSU, AB Rule 27.1), and these latter are encouraged to reach an agreement at any time (Articles 3.7 and 12 of the DSU). The requirement that both cases made to the panel and appeals filed with the Appellate Body must be made in writing (Article 12 of the DSU and AB Rule 21.2) helps to remove, or limit, possible sources of distortion and manipulation in communication. The concern against coercion is particularly vivid with respect to developing countries, which often lack the competences and resources necessary to participate actively and effectively in the DS process. To contain the problems developing countries face in making their cases, the DSU provides for some forms of special and differential (favourable) treatment. For example, during WTO consultations, developed countries ‘should give special attention’ to the problems of developing countries (Article 4.10 of the DSU) and these latter can request at the panel level longer time-limits to prepare and present their arguments (Article 12.10 of the DSU). The same rationale seems to lie beneath the provision that panels' draft and interim reports be sent to the parties for comments, before being submitted to the DSB (Article 15 of the DSU). By the same token, and to protect the interests of all Members, third parties with a substantial interest in the disputed matter can receive and provide detailed information at both the panel and the appellate review stages (Articles 10 and 17 of the DSU and AB Rule 24).
It should be noted that over the years, Habermas has refined his theory, introducing a series of differentiations between types of discourse for which it is impossible to account in this paper. However, for our purposes, a relevant distinction holds between justificatory discourse and practical discourse.Footnote 21 The latter is meant for dispute settlement and relaxes the conditions of unanimity (as a requirement for decision making) and universalization that applies to the former.Footnote 22 Unlike justificatory discourse, practical discourse is not expected to build up the basic rules of a society. It is rather meant to deal with situations of conflict within already structured contexts. This account suggests that the type of discourse relevant for the DSS is practical. It is meant to address disputes that may arise in the interpretation and application of agreed norms.
Moreover, in a later work, Habermas operates a distinction between discourses of justification and of application (Habermas, 1995). Whilst the former are general in kind and meant to formulate and clarify the rules of interaction, the latter are primarily meant to establish the appropriateness of a norm to a specific situation, or of a given behaviour to a norm. Furthermore, the latter may also serve a ‘gap-filling’ function facing unforeseeable consequences and situations (Habermas, 1996: 217).Footnote 23
This distinction presents some significant areas of overlap with the characterization of the classes of disputes offered in Section 3. ‘Conformity-related’ disputes seem to belong to the realm of discourses of application, as they derive from a disagreement on the specific features of a certain measure vis-à-vis a certain WTO agreement. In ‘ambiguity-related’ disputes, the problem lies instead in the lack of clarity of the text of WTO agreements and of the basic concepts employed. Although in this case parties are not called upon to establish a new rule through the DSS, the intention is to clarify the existing provisions in order to make up for some ambiguities in their formulation. This function of the DSS thus comes quite close to those belonging to the realm of discourses of justification. A mixed case is represented by issues of ‘gap-filling’. In the WTO, gaps may be due either to the impossibility of making provisions for all possible future circumstances, or to the lack of agreement between the Members (see Section 3). In the former case, the DSS's function resembles that of a discourse of application. In the latter, the DSS takes on a more ‘creative’ role, thus getting closer to discourses of justification.Footnote 24
One last type of discourse we must mention is legal discourse, seen by Habermas as a complement to (not a sub-type of) moral discourse (see Habermas, 1996: 230–237). Legal discourses concern the authority, legitimacy, and rational justification of the rightness of legal norms and of the legal system. In brief, through the theorization of legal discourse, Habermas has concerned himself with the following question: what confers validity to a law? Our aim in this article is not to assess the authority, legitimacy, or rightness of the WTO DS mechanism but, as explained, to make sense of its distinctive features.Footnote 25 Despite this difference between our aim and concern and Habermas's, we thought it fruitful to apply Habermasian categories to the DSS from a fresh, but cogent perspective. Even though the letter of Habermas's legal studies would drive us in a direction not relevant for our present concern, we shall take their spirit and in particular endorse the idea that the application of discourse theory to legal processes may help to critically assess and make sense of them.Footnote 26 We hope that the argument outlined in this section was successful to account for the plausibility and fertility of our decision.
6. Discourse theory and the orientation of interactions
So much for the general coherence of a Habermas-inspired reading of the DSS. It is now time to introduce those specific aspects of Habermas's theory that may help us to understand how the different components of the DSS are consistent with, and meant to secure, the WTO's general aims facing the different classes of disputes between its Members.
In particular, we would like to draw attention to the three different types of orientations which, according to Habermas, a linguistic interaction may have: compromise (strategic action), consensus, and mutual understanding (communicative action). In what follows we shall explain each of these concepts.
6.1 Compromise
In Habermas's view, compromise stems out of a fortunate coincidence of different interests, as the result of a bargaining process in which each party makes use of its relative power to pursue its goals, to manipulate the use of information, and to exploit the other parties' desires and fears to influence their conduct to its advantage (see Habermas, 1996: 166). In such a bargaining scenario, the balance of different forces is expected to lead to a compromise which is deemed mutually advantageous to the parties on the basis of their different and particular reasons (Habermas, 1996: 140).
As is typical of bargaining processes, interactions oriented to compromise do not touch upon the merits of the disputed question. Accordingly, the disagreement is likely to outlive the compromise between the parties, who are not therefore expected to go through any change either in their attitudes towards, or appreciations of, the disputed matter. They are likely to accept a compromise simply because it maximizes their interests, and may equally likely be ready to abandon it when it no longer serves their purposes.
According to Habermas, compromise belongs to the realm of strategic action (see Section 5.1) and cannot, consequently, be an appropriate orientation for the interaction within discourse. Specifically, an interaction oriented to compromise, thus characterized, cannot immunize parties against coercive and manipulative strategies of persuasion. Therefore, within strategic action the provision that the ‘unforced force of the better argument’ be the only source of agreement is undermined.
6.2 Consensus
The dialogical nature of Habermas's model makes it impossible to attain a permanent definitive closure to discourse: new issues may always arise in the interaction between the different parties. Nevertheless, under ideal conditions, ‘Habermas's concept of discourse seems to imply, in principle, the possibility of full consensus toward something like the correct answer’ (Rehg and Bohman, Reference Rehg, Bohman, Schomberg and Baynes2002: 33).
For Habermas, consensus means a status of full substantive agreement between the different agents, who come to agree unanimously on something for the same reasons. It should be noted that it is not only confined to an ideal dimension (Habermas, 1996: 166). Consensus is conceived as an always potentially attainable result, whose possible realization should be what makes discussion keep going. From this perspective, agents engaged in communicative action should always strive to reach a full consensus qua the fullest degree of harmonization of their (either individual or shared) goals (see Section 5.1).
6.3 Mutual understanding
As anticipated in Section 5.1, within Habermas's theory the notion of understanding plays a paramount role. It is not merely thought of as a possible end for a discursive interaction; it is, in fact, conceived of as ‘the proper – and not only a derivative – mechanism for coordinating action’ (Berger, Reference Berger, Honneth and Joas1991: 166).
Habermas explains that the ‘term “reaching understanding” means, at the minimum, that at least two speaking and acting subjects understand a linguistic expression in the same way’ (Habermas, Reference Habermas1984: 307). This may apply, in a basic sense, both (i) to the context in which the communication takes place and (ii) to the claims put forward by the parties.
Reaching understanding does not have to mean that agents need to find a substantive consensus, especially when what is at stake is a dispute featuring two opposed claims. For communication to be successful, they need primarily to understand the claims they put forward against a shared background. In this sense, Habermas sees understanding as the appropriate orientation of any discursive interaction qua the essential pre-condition for any prospect of coordination, reconciliation, and harmonization between different goals pursued by social agents (Habermas, 1996: 18).
7. Disentangling the DSS components: procedural orientations in context
Bearing in mind that our aim is neither exegetical nor interpretative of Habermas's own view, his characterization of the three different kinds of orientation for linguistic interaction will be employed with one significant amendment: we shall regard them as applicable to procedures rather than to agents participating in discourse. The DSS is structured in such a way as to transform strategic interaction (where power considerations and coercion apply) into a ‘communicative’ one, conducive to shared agreement and to the maintenance of an acceptable balance of rights and obligations. If this is accurate, and in line with our procedural approach, what is relevant for our purposes is not so much the actual orientation of the parties as that of the DS procedures, and their capacity to shape the interaction between the parties.
In what follows, we shall make use of the three orientations presented above (Section 6) in turn as instruments for a philosophical analysis meant to make sense of the distinctive features of the DSS, its different components and their purposeful connection with the aims of the WTO and the disputes between its Members.
7.1 Compromise
As seen in Section 4.1, prior to starting the procedure for the establishment of a panel, the complaining party must ask the challenged Member to enter into WTO consultations. The aim of this phase is to find a ‘satisfactory adjustment to the matter’ (Article 4.5 of the DSU). It seems plausible to argue that such a characterization draws consultations rather close to what is generally known as negotiation, central to which is the idea that all parties should find some satisfaction in the outcome. Since compromise is usually seen as the appropriate orientation for negotiation, one might think that this also applies to WTO consultations.
However, the above-mentioned Article 4 of the DSU reads: ‘If a request for consultations is made pursuant to a covered agreement, the Member to which the request is made shall … enter into consultations in good faith … with a view to reaching a mutually satisfactory solution’ (Article 4.3 of the DSU, emphasis added). This seems to reveal at least two points of departure from an orientation to strategic compromise as defined in Habermas's discourse theory (see Section 6.1). More precisely, parties do not seem to be allowed to act as purely strategic actors (thus employing all means to persuade their opponents and win the game) as (i) they must be in good faith and (ii) they must renounce imposing, with the use of threats and exploitation of their relative power, solutions that may not be regarded as mutually satisfactory by all concerned parties.Footnote 27 In addition, WTO consultations cannot be carried out in secret but the request for their establishment must be notified to the DSB, and the parties involved in them are required by Article 3.5 of the DSU to find solutions to the dispute consistent with the agreements and that do not ‘impair benefits accruing to’ any WTO Member.
It seems, in other words, that constraints applying to discourse are at work even at this preliminary stage. But why is this the case? In line with our initial declaration of intents, our answer looks at the relations between the orientation of DS procedures and the WTO's general aims. As explained, WTO aims to keep peaceful and stable trade relations between Members and a certain politically agreed balance of rights and obligations. Following the characterization offered in Section 6.1, procedures oriented to strategic compromise cannot immunize the parties' interaction from (i) the effects of their uneven bargaining power and (ii) manipulation and distortion of information. These would jeopardize the pursuit of peace and stability in trade relations by risking exacerbating, rather than smoothing, the tensions between the parties. Moreover, procedures oriented to strategic compromise would expose the fate of multilateral agreements to the contingent outcomes of power-based bilateral negotiations. This would be in striking contrast with the requirement that any agreed solution to a dispute be consistent with the overall balance of rights and obligations of all Members.
Notably, bargaining (oriented to strategic compromise) may take place before entering this even preliminary stage of the DSS, during bilateral consultations. Accordingly, strategic compromise is left outside the DSS procedures and thus seems to play only a residual role in the settlement of disputes.Footnote 28
This hypothesis regarding the residual role of compromise seems to be reinforced by consideration of such remedial procedures as those leading to compensation (Article 22 of the DSU). As anticipated in Section 4.4, parties are allowed to try to reach a compromise on a satisfactory compensation, yet this is considered a sub-optimal and temporary alternative to the preferred solution: the ‘full implementation of a recommendation to bring a measure into conformity with the covered agreements’ (Article 22.1 of the DSU). Given the general commitment to the overall stability of the trading system, parties are not free to strike any compromised agreement on compensation.Footnote 29 Article 22.3 of the DSU outlines precise procedural conditions aimed to limit the systemic effects of this remedy and its duration.Footnote 30
In sum, strategic compromise is residual as it may occur either before WTO consultations or, in a debatable constrained form, at the level of remedial procedures leading to compensation. Therefore, the sort of compromise allowed within the DSS is not open to the ‘brute’ power relations between the parties. It is procedurally regulated and its object must be consistent with the general balance of rights and obligations of all Members.
7.2 Consensus
In light of the characterization in Section 6.2, a procedural orientation to consensus would reveal a concern with unanimity in dispute settlement. Although consensus used to occupy a more prominent position under the GATT, it still seems to play some role in the WTO. Unanimity by consensus is paramount in the adoption of the panel's and AB's reports: although no explicit approval should be given, (reverse) consensus is demanded to reject them (Articles 16.4 and 17.14 of the DSU – see Section 4 above). Moreover, positive consensus is required in a few either rare or minor cases.Footnote 31
This seems to accord consensus an interstitial but still significant place. It is interstitial as, in line with the general provisions of the DSS, replies to disputes must be prompt so as to limit the potentially destabilizing effects of disputes on the general stability of the system. Unanimous positive consensus at all stages would be too demanding an orientation for procedures that are devised to attain a certain degree of efficiency. Moreover, consensus would imply veto power for losing countries, which risks rendering the DSS ineffective and unpredictable. The role of (negative) consensus is still significant though, as it is respondent to the importance of the idea that a dispute is not merely a matter for the directly concerned parties, but rather affects the whole WTO community.
This latter point may appear overestimated as parties retain the right to drop a case if they reach a mutually accepted solution before the DSB adopts the reports of the adjudicating bodies. This provision is in fact consistent with the aims of the WTO in two main regards. First, if parties find in good faith a mutually acceptable agreement between themselves, it must mean that their relationship is not compromised and their dispute no longer represents a threat to the general trading system.Footnote 32 Second, as explained above, not literally any mutually agreed solution would do. The general balance of rights and obligations must be preserved, so that no Member will be deprived of the benefits acquired under the relevant agreements (see Articles 3.4, 3.5 and 3.6 of the DSU). To ensure this, all agreed solutions must be notified to the DSB whose members are entitled to raise any point related to them.
It should be noted that the idea of consensus at work in the DSS is not as demanding as that favoured by Habermas (hereafter Habermasian consensus). Article IX of the WTO Agreement provides that a WTO body ‘shall be deemed to have decided by consensus on a matter for its consideration, if no Member, present at the meeting when the decision is taken, formally objects to the proposed decision’. Here it is verified that consensus should be unanimous and on matters of substance, but no mention is made of the fact that it should be reached by agents motivated by reasons they find convincing in the very same way, as was the case on Habermas's account (see Section 6.2).
In sum, consensus is present at all stages in its reverse form (and in a very few circumstances in its positive form) to ensure the consistency of dispute settlement with the covered agreements. However, it has an interstitial role with a view to ensuring that the DSS's reply to disputes is prompt and certain.
7.3 Mutual understanding
Looking carefully at the DSS, the relevance of understanding emerges quite clearly as a priority orientation for argumentative procedures. As explained in Section 4.1, from the very beginning of the process, when requests for WTO consultations are submitted, every effort seems to be made to ensure that all relevant pieces of information are disclosed both to the parties in a dispute and to the other Members.
As Palmeter and Mavroidis put it, WTO consultations are meant to ‘enable parties to gather relevant and correct information – both to assist them in reaching a mutually agreed solution or, failing that, to assist them in presenting accurate information to the panel’ (Palmeter and Mavroidis, Reference Palmeter and Mavroidis2004: 87). This reading is reinforced in the AB report on the Mexico–Corn Syrup case:
Through consultations parties exchange information, assess the strengths and weaknesses of their respective cases, narrow the scope of differences between them, and, in many cases, reach a mutually agreed solution in accordance with the explicit preference expressed in Article 3.7 of the DSU. Moreover, even where no such agreed solution is reached, consultations provide the parties an opportunity to define and delimit the scope of the dispute between them. Clearly, consultations afford many benefits to complaining and responding parties, as well as to third parties and to the dispute settlement system as a whole.Footnote 33
The provision that proceedings of consultations be confidential might be regarded as an occasion for the parties to behave strategically behind closed doors in the absence of an adjudicating body acting independently to seek and receive information on the matter. However, as explained in Sections 4.1 and 7.1, WTO consultations are limited by a requirement that the parties be in good faith and try to reach a mutually satisfactory solution. This seems crucial to remove occasions for strategic behaviour involving dissimulating actions and responsibilities.
A distinct element of the DSS consists in the role of such independent adjudicating bodies as the panels and the AB. An orientation to understanding may be found in the procedures through which these bodies work.
The attempt to put the parties in a position to reach a mutual understanding by means of argumentative procedures is clear in the case of the panel (Article 12 of the DSU). Following the illustration in Section 4.2, its main function is indeed to clarify the factual evidence and the legal nature of the dispute. This objective is pursued, in the first place, through procedures allowing continuous argumentative interactions (Article 11 of the DSU) between the panel and the parties by means of a series of subsequent submissions and hearings. Moreover, understanding is sought through the submission of both draft and interim reports to the parties for comment. In the second place, panels are entitled to seek information to clarify facts by addressing questions to the parties and by contacting external advisory experts. In the third place, third parties have the right to be heard by and make submissions to the panel.Footnote 34 Finally, the provision that the panel does not directly apply previous findings to adjudicate new disputes is intended to encourage the parties to assess and engage with each single dispute. These provisions, taken together, enhance the DSS procedural commitment to put the parties in a position to reach an understanding of their respective claims on the disputed matter.
The pursuit of mutual understanding through such procedures is meant to facilitate the appraisal of the conformity of a given measure with WTO provisions, and to allow the interpretation of the contended legal basis of the claims. This provides a solid ground on which ‘conformity-related’ disputes may be addressed (see disputes belonging to the third class in Section 3) in line with the WTO's general aim to maintain the overall balance of rights and obligations, by helping to clarify how a given measure may affect it.
The panel's procedures, however, may not be enough to secure full understanding. Parties may therefore resort to an appellate review aimed to clarify ‘issues of law covered in the panel report and legal interpretations developed by the panel’ (Article 17.6 of the DSU). Following the characterization in Section 4.3, AB procedures provide for a further occasion for the participants to interact and clarify their views through detailed submissions.Footnote 35 Note also that the AB is a standing body following a principle of collegiality (Article 17 of the DSU and AB Rule 4). This ensures that its legal interpretations are coherent and consistent across rulings, thus providing a common basis of knowledge guiding the ways in which Members can interpret WTO provisions with a view to reducing possible sources of misunderstanding.Footnote 36 Accordingly, AB procedures may be seen in line with the WTO's general aim to secure the stability of the overall trading system.
The orientation to understanding make AB procedures apt to address what we have called ‘ambiguity-related’ disputes (see disputes belonging to the second class in Section 3), helping to dissolve the vagueness inherent to the formulation of certain WTO provisions. In addition, the AB has come to exercise an albeit controversial, ‘gap-filling’ function as far as its rulings contribute to specifying the scope of the agreements on issues on which the text is silent (see disputes belonging to the first class in Section 3).
This characterization is strikingly in line with Habermas's concern with mutual understanding as the appropriate orientation for discourse as a means of restoring agreement and secure cooperation. However, it should be remembered, to avoid over-enthusiastic simplifications, that Habermas's discourse is ideally time-limit-free a form of interaction. According to Habermas, communication should continue until all participants have reached a full understanding of the others' claims and of the situation that surrounds them. Although very precise time constraints are rather established by the DSU and allocated to different phases of the process, it should be noted that the parties are allowed at any stage to request a time extension, should they deem it necessary to reach a fuller understanding of the case and, ultimately, a mutually acceptable agreement.Footnote 37
In sum, if in the DSS the orientation to compromise plays a residual role and the orientation to consensus an interstitial one, mutual understanding is given pride of place as procedures oriented to it are particularly apt to address the kinds of disputes arising between Members in a way consistent with the WTO's general aims.
8. Shortcomings of the DSS and procedural implications for its reform
As declared in the Introduction, our primary aim has been to disentangle the purposeful connections between the procedural features of the DSS, the classes of disputes and the WTO's general aims. Accordingly, we have concentrated on the textual analysis of the DSU and related documents. This strategy is consistent with our twofold concern with (i) the inherent qualities of DS procedures, and (ii) issues of enforceability, rather than enforcement capacity (see Section 2). However, textual analysis alone may conceal some significant shortcomings that have occurred in the translation of procedures into practice. We should like to draw attention to four points mainly.
First, despite the procedural guarantees of formal equality present in the text of the DSU (see especially Section 5.2), some Members may de facto end up being excluded from active participation in the DSS, either because of its costs or as a result of unequal power relations. This is often the case with developing countries that lack legal and technical human resources and the financial means to support the costs of the proceedings, and that may be afraid of challenging within the WTO a more powerful state.
Second, despite their orientation, material procedures may prove incapable of actually leading to mutual understanding. The lack of transparency (often complained of by certain Members and NGOs) and the absence of effective control on the completeness of the information provided by the parties represent a serious obstacle to the actual attainment of a full understanding between them.
Third, the ultimate implementation of WTO rulings depends on the losing party's willingness to comply, which depends in part on the winning party's enforcement capacity. This emerges clearly on consideration of the observed time delays in compliance when the losing party is economically powerful and has a strong interest in the matter. The enforcement capacity of the winning party is certainly not enhanced by the absence of both provisional and retrospective remedies, which factually allows for a sanction-free period of non-compliance.Footnote 38
Fourth, the ‘gap-filling’ function often played by the AB in adjudicating disputes is controversial and has been highly debated. As explained in Section 4.3, the AB is required to clarify the existing provisions without altering the rights and obligations of the WTO Members, and its rulings are exclusively binding on the parties in the dispute and do not set a precedent as in common law.Footnote 39 In fact, the AB has never declined to adjudicate a dispute, even when the text did not provide sufficient legal grounds to reach a decision and on which, therefore, a political body would have been normally competent. In so doing, it ended up making rulings enlarging the scope of the WTO agreements. Moreover, the AB has frequently considered its previous rulings as a form of non-binding precedent.Footnote 40 This shows how striking a balance between the law-making function of the political bodies and the law-interpreting functions of the adjudicating ones is particularly delicate.
These and other problems affecting the practical working of the DSS have driven Members, in the 2001 Ministerial Conference launching the Doha round, to agree to negotiate clarifications and improvements on the DSU (see Ortino and Petersmann, Reference Petersmann, Ortino and Petersmann2004; Zimmermann, Reference Zimmermann2006).Footnote 41 It is worth noticing that any amendment to the text of the DSU would affect the orientation of the DS procedures.
As an example, consider the several proposals regarding the introduction of a remand procedure at the appellate stage. When a panel adopts a strict ‘judicial economy’ principle and does not decide on issues of law that turn out to be important in the appellate review, the AB can either force the participants to start a new panel proceeding to address such issues, or it can take a decision by itself (de novo), but it cannot remand the issue to the original panel. If the AB decides de novo, as often occurs, the parties are deprived of the opportunity to participate in the analysis and discussion that would occur during the panel's proceeding. The introduction of a remand procedure would enhance the orientation of DS procedures to understanding by giving the parties the opportunity to resume discussion on controversial facts and interpretations.
On a different note, some have lamented a couple of shortcomings at the interim review stage when panels circulate draft and interim reports (see Section 4.2): (i) descriptive draft reports are often very long but little informative; and (ii) third parties lack the opportunity to receive and comment on interim reports. Sub (i), if expunging draft reports may increase the speed of the procedure, such a decision would certainly have a negative impact on the DSS's orientation to mutual understanding, depriving the parties of an occasion to enhance their appreciation of the matter. Sub (ii), on the contrary, enhancing third parties' participation in the interim review stage may increase transparency (thus contributing to a better understanding of the matter), but lengthen the process by forcing the panel to consider and discuss third parties' opinions while issuing the final report.
To conclude, if our analysis is correct in emphasizing the key role of an orientation of DSS procedures to mutual understanding, any change undermining such an orientation may risk having a negative impact on the capacity of the DSS to serve the WTO's general aims in light of the disputes arising between its Members. Therefore, we claim that this kind of procedural consideration should inform the DSU reform proposals as much as substantive concerns.
9. Conclusion
The characterization of the DSS offered in this paper has shown that the development of a structured but flexible DS mechanism is not a mere idiosyncrasy of the WTO. The formulation of a multi-layered, argumentative mechanism, including a complex system of procedures is, rather, the result of a purposeful adaptation of the GATT provisions to the WTO's general aims and to the nature of the disputes arising between its Members.Footnote 42
Our Habermas-inspired procedural reading was intended to propose a method to evaluate and make sense of the DSS's distinctive features, going beyond the ‘mere’ reference to the quality of the outcomes DSS procedures produce. In particular, we have used Habermas's account of diverse orientations for interaction (compromise, consensus, and mutual understanding) and applied it to the diverse procedural components of the DSS.
The analysis we have carried out has revealed that the crucial procedural components of the DSS have been devised with the primary orientation of putting the parties in a position to reach a mutual understanding, as a basis to finding a mutually satisfactory settlement of disputes. Following our proposed typology of disputes, the DSS comes to play three main functions: (i) to fill gaps in the agreements, (ii) to clarify the ambiguities of rules, and (iii) to evaluate the conformity of measures with WTO provisions. Vis-à-vis such disputes, the orientation to mutual understanding has proved to be important to secure the system's stability and predictability, as well as to keep smooth relationships between Members. In contrast, an orientation to either strategic compromise or consensus would have been respectively hardly compatible with the need for stability of multilateral agreements and ineffective in ensuring the prompt settlement of disputes.
This does not mean or imply that the very ideas of compromise or consensus are absent from the DSS. In fact, the system allows for these types of orientations to enter the scene at different stages, albeit not in a distinctively Habermasian form. In particular, some residual room for compromise and some interstitial, though crucial space for consensus may be found. The former is located in bilateral (pre-WTO) consultations and, in a constrained form, in the negotiations leading to the establishment of compensation. The latter is present at all stages in its reverse form but is attributed an interstitial role so as to ensure that the DSS's reply to disputes be prompt and certain.
In sum, to our initial question concerning the reasons why the DSS is argumentative in nature and articulated through sophisticated procedures, our proposed answer is as follows: to be conducive to mutual understanding which is in turn conducive to the general aims of the WTO, vis-à-vis the different kinds of disputes that may arise between its Members. From this perspective, argumentative procedures, allowing the participation and hearing of all relevant parties, are essential. Similarly, DS procedures are quite sophisticated as the orientation to mutual understanding requires a complex, multi-layered system ensuring spaces for confrontation and discussion (involving the parties and other interested WTO Members), allowing disclosure and consideration of all relevant information, and providing a final phase of adjudication.
We hope that our discourse theory-inspired reading has succeeded in making sense of the DSS's distinctive features and, in so doing, proved to be helpful from a twofold angle. At an applied level, in improving the understanding of the DSS, it may inform diplomats who are engaged in Doha negotiations on improvements and clarifications of the DSU. At a theoretical level, it represents a distinctively procedural-oriented contribution to the recent interdisciplinary literature addressing institutional design problems.
