2. Brief summary of the factual and legal issues in dispute

The factual and legal basis for the EC–Biotech case is relatively straightforward. The three complainants – the United States,Footnote ⁸ Canada,Footnote ⁹ and ArgentinaFootnote ¹⁰ – all concentrated on two distinct matters. One of those was ‘the operation and application by the European Communities of its regime for the approval of biotech products’, which can be identified as food or food ingredients containing, consisting of, or produced from genetically modified organisms or GMOs.Footnote ¹¹ At the time the complaint was brought against the EC its approval regime consisted of Directive 2001/18/EC (replacing Directive 90/220/EEC) governing the ‘deliberate release into the environment of genetically modified organisms’ or GMOs (the ‘Deliberate Release Directive’)Footnote ¹² and Regulation 258/97/EC regulating ‘novel foods and novel food ingredients’ (the ‘Novel Foods Regulation’).Footnote ¹³

Aside from seeking to protect human health and the environment, the EC's regime for approval of biotech products was an elaborate pre-marketing approval system. The legal instruments that underpinned the approval system contained detailed administrative procedures for corporate entities and others seeking to obtain approval to place a biotech product on the European market and for which an application for approval first had to undergo evaluation.Footnote ¹⁴

The other matter concerned ‘certain measures adopted and maintained by EC Member States prohibiting or restricting the marketing of biotech products’.Footnote ¹⁵ Under the Deliberate Release Directive and the Novel Foods Regulation certain EC Member States – Austria, France, Germany, Greece, Italy, and Luxembourg – continued to maintain safeguard measures in respect of biotech products or GMOs that had obtained approval for EC-wide marketing under the EC regime because they considered them to be ‘too risky’.Footnote ¹⁶ Under EC legislation individual EC Member States could provisionally restrict or prohibit the use and/or sale of an approved biotech product in their own territory but they could only do so if they had ‘detailed grounds for considering, based on new or additional information or scientific knowledge, that the particular product pose[d] a risk to human health or the environment’.Footnote ¹⁷ A further requirement was that the relevant EC Member State, which intended to take such a safeguard measure, was under a duty to inform other EC Members and the Commission. This led to a decision being taken at Community level on that Member State's safeguard measure within a prescribed period of time.Footnote ¹⁸

The complainants' in EC–Biotech argued that the EC regime for approval of biotech products was effectively a moratorium on the review of application for the commercialization and marketing of GMOs. Specifically they alleged that the EC had suspended consideration of applications under relevant EC processes for the approval of products such as canola/grape seed, maize, soybean, cotton, chicory, sugar and fodder beet, potato, tomato, and lettuce. They also claimed that certain EC Member States had introduced and/or were maintaining measures that prohibited the importation, marketing, or sale of biotech products or GMOs.

3. Social science expertise, ‘post-normal’ science and the risk discourse

When trade negotiators set out to draft the SPS Agreement during the Uruguay Round (1986–1994) the emphasis on risk assessment reflected the prevailing view in the international trade community that a robust and transparent means of evaluating ‘food-borne’ risks and ‘pest’ or ‘disease-borne’ risks was necessary in order to counter forms of neo-protectionism that WTO Members might pursue through stringent regulatory regimes for the importation of agricultural and processed agricultural products into their territories. Hence, SPS measures must not be maintained without sufficient scientific evidenceFootnote ¹⁹ and must be based upon a risk assessment.Footnote ²⁰ Where relevant scientific information is deemed insufficient, and on the basis of any pertinent information, a provisional SPS measure may be maintained.Footnote ²¹ WTO Members also have to ensure that their management of SPS risks does not lead to discrimination or a disguised restriction on trade between situations or products, irrespective of their origin, and are required to work together in seeking to achieve consistency in the level of risk that is acceptable in different situations.Footnote ²²

While the conditions for the use of science in establishing and applying health and food safety regulations remain unchanged, the first decade of the SPS Agreement has been marked by a period of considerable uncertainty in the world of science and technology. In particular during the late 1990s there was a growing recognition of a need for a new ‘Social Contract for Science’ in order to deal with some of the pressing scientific and social issues raised by new technologies such as biotechnology or nanotechnology.Footnote ²³ The international community faced the challenge of how to focus on linkages between social, political, and economic systems on the one hand and biological, physical, and chemical systems on the other.Footnote ²⁴

Traditionally, applied scientific research took place under conditions that have more or less been described by Thomas KuhnFootnote ²⁵ as the conduct of ‘puzzle-solving normal science within a paradigm that ensures the steady accumulation of relatively stable and certain facts, employing recognized methods’.Footnote ²⁶ An alternative paradigm of post-normal science, offered by Funtowicz and Ravetz, is science for policy, which embraces a situation where scientific uncertainty and value issues are commonplace, where objectivity is not possible, where prediction and control are limited, and where society and politics have a key role to play.Footnote ²⁷

As a rule, post-normal science can be located in relation to other traditional strategies that can best be described by placing the criterion of ‘systems uncertainty’ along a horizontal axis, while ‘decision stakes’ is placed along a vertical axis. When both systems uncertainty and decision stakes are low, a situation of applied science pertains and results in a situation that is usually referred to as normal or ‘safe’ or ‘sound’ science. When either systems uncertainty or decision stakes are medium, the routine application of scientific knowledge is judged to be inadequate and there is a need to resort to the specific skills and judgements of professionals in the sense of professional consultancy. When both systems uncertainty and decision stakes are high, a situation of post-normal science prevails.Footnote ²⁸

A situation of post-normal science may pertain irrespective of the nature of the subject matter, the degree of scientific uncertainty or the level of ignorance and without all the factors of the problem having been observed. It applies as much to the depletion and/or replenishment of fish stocks or global warming brought about by climate change as it does to developments in the field of gene technology. Post-normal science usually presents itself in situations where the sheer complexity of the object of scientific research is coupled with issues of uncertainty or ignorance but governments and society need to deal with a situation immediately rather than waiting for a future event. It is in this context that post-normal science has been called upon as a problem-solving strategy.Footnote ²⁹

Risk has come to play an increasingly important role in decision-making in the field of environmental law and public health but it is something of a newcomer with respect to food safety. Not surprisingly, the main emphasis has been on the development of (quantitative) risk assessment techniques, while the role of risk assessment in standard setting, previously discernible in evidence before domestic and regional courts, has moved centre stage in international adjudicative fora.Footnote ³⁰ Even so, scientific risk assessment often cannot deal adequately with issues of uncertainty and variability, both of which may prevail with respect to certain categories of risks and certain types of products or processes. Put simply: ‘[T]here is no scientific definition of terms such as “acceptable”, “reasonable”, and “sufficient”’Footnote ³¹ and attempts to deal with these terms in legal practice may get lost in translation.

If one takes the post-normal science approach to the issues of assessing and managing risk, then ‘inevitably various sorts of uncertainty and value-commitments enter into any decision on risks, and because of that the scientific side of the work must be complemented by other considerations, deriving from its policy aspects.’Footnote ³² Not only must society cope with changes that often occur with increasing intensity, rapidity, and complexityFootnote ³³ but there is also a need to deal with the failure of public trust in scientific information.Footnote ³⁴

Lack of public trust in scientific information can result from the (in)ability of governments to manage risk, especially where there are difficulties in assessment of that risk, as occurred in the case of some high-profile environmental disasters such as Chernobyl and Seveso.Footnote ³⁵ It may also arise where there is an unresolved debate about instances of hazard involving a new disease with a real but unquantifiable incidence rate that creates a high level of uncertainty, such as occurred in the Bovine Spongiform Encephalopathy or BSE crisis in Europe,Footnote ³⁶ or threatens to occur if there is a significant increase in the incidences of Avian or bird flu in humans. Significantly, the sheer novelty of the technology coupled with high uncertainty and high decision stakes may lead to a complete erosion of public trust in the relevant scientific information, as is the case with respect to GM foods and food crops.

Aside from this, the objective scientific or ‘sound science’ approach, which underpins the form of risk assessment, prescribed by the SPS Agreement in Article 5.1Footnote ³⁷ and Annex A.4Footnote ³⁸ raises some serious questions, which are of relevance in this case and potentially others like it in future. Who are the scientific experts? What is the quality of the information available to such experts? What are the uncertainties and sources of ignorance that we know of with respect to GM products? How precautionary can one be with respect to certain products and should the element of precaution be included in the methodological choices of scientific investigation?

Perhaps the most important question is who should judge such matters? To which social science in general, and post-normal science in particular, would answer: those most affected by the design of policies, the quality of information, and the quality of the methodology used; in every instance it is the ordinary citizen or the consumer who should have a right to participate in these processes. Thus, post-normal science is identified with three methodological elements: the analysis of uncertainty and its management, the integration of different sources of knowledge across the lay-expert divide, and the participation of non-scientific experts in forms of extended peer review.

4. Differing approaches towards risk analysis: the regulatory divide

There are different regulatory approaches to risk in WTO Member states, largely based upon the public's perception of risk and differing social and cultural preferences. The regulatory divide is most apparent in the process of risk analysis that exists in the United States as opposed to that of the European Communities.

From the 1970s, onwards US government agencies like the Food and Drug Administration (FDA), the Environment Protection Agency (EPA), and the US Department of Agriculture (USDA) moved away from the precautionary approach in risk regulation towards adoption of a sound science approach. For the United States, as one of the complainants in EC–Biotech, a risk assessment is a factually grounded, objective, and value-free analytic exercise that precisely identifies possible harms to human health and the environment. The means for achieving this is a formal, expert-based assessment of the potential harm that GMOs may cause, i.e. a straightforward hazard identification process. A further feature of the US system is that decisions of governmental agencies are subject to judicial reviewFootnote ³⁹ and this leads to an ‘adversarial’ system, which stands for openness and transparency in the public scrutiny of governmental activity.

The regulatory approach towards risk regulation in the EC – at least in the field of food safety – is in the first instance not so different to that of the United States. This is clear from the establishment of the European Food Safety Authority (EFSA), the legislative basis of which anticipates three distinct processes of scientific risk assessment, political risk management, and risk communication.Footnote ⁴⁰ The EC's view of risk assessment is that the process is not the subject of a single methodology nor is it a purely scientific exercise. Indeed, where it has not been possible to make a quantitative assessment of the risk with reasonable scientific confidence the precautionary approach may be adopted,Footnote ⁴¹ which is an approach that is informed by the most appropriate science.Footnote ⁴² In strict procedural terms, risk assessment consists of ‘hazard identification, hazard characterization, appraisal of exposure and risk characterization’.Footnote ⁴³

Again for a long time in the US it was accepted that risk management was ‘purely a problem of the application of the conclusions of the relevant quantitative sciences’, although there was a growing appreciation that risk management involved governance issues as well.Footnote ⁴⁴ However, changes came about from the late 1970s, as a result of judicial review in US courts of agency action pertaining to risk assessment which was followed by legislative action. The latter led to a ‘clear conceptual distinction’ being made in the United States between ‘a “scientific process” of risk assessment, in which the scientific information about risk was collected and evaluated, and a “political” process of risk management, in which a policy decision was made about what regulatory action should be taken in relation to that risk’.Footnote ⁴⁵ Overall, the US pursues an approach to risk management that involves a balancing of environmental/quarantine risks against the costs and benefits of the policy pursued.Footnote ⁴⁶

In the case of the EC, when it comes to risk management, matters are less clear. The EC approach reduces risk management to balancing a policy of precaution against the costs and benefits of the policy pursued. It is also seen as assigning ‘subjective valuations of risk as an implicit rather than [an] explicit part of risk management’.Footnote ⁴⁷ Additionally, the whole process of scientific advice in policy making in the EC is essentially a political one. From one point of view it is considered to offer few guidelines for the process of evaluation and its consensual character has been described as ‘one in which members of the “club” meet to negotiate behind close doors’.Footnote ⁴⁸ From another view, however, the Court of First Instance of the European Communities in Pfizer has upheld the role of the EC's scientific committees in assisting political bodies in the process of scientific policy making. In the Court's view, such scientific advice is based on the principles of excellence, transparency, and independence and serves as ‘an important procedural guarantee whose purpose is to ensure the scientific objectivity of the measures adopted and preclude any arbitrary measures’.Footnote ⁴⁹

5. The five person amicus curiae brief before the EC–Biotech Panel

The amicus brief was submitted by five social scientists to the PanelFootnote ⁵⁰ in EC–Biotech.Footnote ⁵¹ Ultimately, it was not considered by the Panel,Footnote ⁵² nor taken up in the record. Nevertheless, it offers some important lessons for the future. It is distinguishable from other amicus briefs brought in the case in that it specifically seeks to provide expertise from a social science perspective. In so doing, it raises a number of key issues concerning the way in which the SPS Agreement is drafted and how it should be interpreted.

Of particular note is the way in which the amicus brief tries to convey the authors' views on how the concept of risk and risk assessment should be applied in the field of biotechnology. For example, it recommends full consideration of the range of relevant scholarship, including the social sciences, in interpreting terms such as ‘risk’, ‘risk assessment’, ‘rational’, ‘objective’, and the contestable issue of ‘sufficient scientific evidence’.Footnote ⁵³

It also recognizes that risk assessments of GMOs, which are conducted in specific national or institutional settings, may be limited, partial, and constrained by the decision-making cultures within which those assessments are produced.Footnote ⁵⁴ The amicus brief therefore advocates using extensive forms of public deliberation in order to fill some of the gaps in knowledge and values in the field. Absent widespread public inputs, there is an insufficient basis for conducting a risk analysis and for resolving such threshold questions as the proper difference between GM and non-GM products.Footnote ⁵⁵

The amicus brief recognizes that risk assessment is not a singular concept but varies with the context in which it is found. Thus, where there is a situation of low certainty about facts and low consensus about methods, as in the case of GMOs, the scientific basis for the regulation of the risk and the very notion of risk analysis remains a fluid one. However, the situation changes dependent upon the national decision-making context, which the authors illustrate by reference to US regulatory practice, and is not helped by the fact that international guidelines for analysing the risk of GM foods are still emerging and not yet universally accepted.Footnote ⁵⁶

As a result of the current status of risk assessment techniques with respect to GMOs and the role of public confidence in regulating new food technologies the European moratorium is not one of ‘undue delay’ under Article 8Footnote ⁵⁷ and Annex C.1(a)Footnote ⁵⁸ nor is the amount of time, which the European Commission considers necessary in order to collect additional information about GMOs, so as to conduct a better risk assessment, ‘unreasonable’.Footnote ⁵⁹ Tellingly also for our consideration of the potential competition between WTO Members' regulatory regimes is the fact that the social science experts' amicus considers that the panel's role in reviewing the parties' arguments should be largely a procedural and not a substantive one. In fact, this addresses the applicable standard of review in SPS cases especially since the Panel in EC–Biotech appears to have acted more as an adjudicatory body reviewing the substantive merits of each party's risk assessment than an (administrative) tribunal reviewing the adequacy of an executive decision-making process.Footnote ⁶⁰

A considerable portion of the amicus brief is devoted to considering the conceptual foundations of risk-based decision making. In support of the Appellate Body's view in EC–Hormones Footnote ⁶¹ the authors attack such issues as the fallacy of creating a strict separation between risk assessment (based on quantitative analysis of risks) and risk management (incorporating judgements of value), where the former is a technical phase and the latter a political one.Footnote ⁶² Instead, the authors of the amicus brief call for a risk analysis to be conducted within a political and cultural context that ‘influence both the initial identification of hazard … and subsequent attempts to assess the magnitude, seriousness and distribution of harms’.Footnote ⁶³ When assessing the risk of GMOs in the context of the SPS Agreement the amicus brief regards risk as embedded in the social system, i.e. within ‘human societies as they actually exist’. Footnote ⁶⁴ Whereas, the Appellate Body in the earlier case of EC–Hormones anticipated the resulting difficulties and the political resistance towards a process of risk assessment and risk management based strictly on scientific evidence, and in complete isolation from other deep-rooted societal considerations, the Panel in EC–Biotech is clearly of a different persuasion.

The arguments for a broader-based risk assessment are complemented in the amicus brief by a call for greater public deliberation in the process of risk assessment, as part of a wider notion of public deliberation in scientific policy making, which is one of the characteristics of post-normal science. In the view of the amicus brief, this includes ‘[D]efining what exactly is “at risk” … [as] properly a matter of democratic value-commitment and of whatever process is appropriate to meet these needs.’Footnote ⁶⁵ Moreover, there is recognition that ‘public attitudes and perceptions … [are] heightened in the GMO case where scientific knowledge is neither uniform nor complete, and because food has a special cultural status in human society’.Footnote ⁶⁶

Public participation and stakeholder deliberation (in short public deliberation) are central to successful risk-based decision making and the SPS case law should be read in support of this approach – particularly where there is a situation of low certainty and low consensus as exists in the case of GMOs. The amicus brief even claims that public consultations on GMOs – such as the farm-scale evaluations of GM crops in the UKFootnote ⁶⁷ – have been essential in dealing with the uncertainties that surround GM crop developments and any attempt to remove planned public consultations as a component of risk assessment ‘would be to undermine important democratic elements in the regulatory processes of Member States’.Footnote ⁶⁸

Given that, democracies can, and do, differ in their assessment and management of risk and the fact that the WTO is a supranational institution, which is trying to resolve a regulatory conflict, it is important to define the role of the judge/arbiter or decision maker in scrutinizing the scientific bases of food regulations. Once again, the amicus brief argues that any assessment of facts and law under the SPS Agreement by panelists or the Appellate Body should at best be procedural and not substantive. On a chastening note, they state: ‘science-based disciplines should not function as draconian enforcement of substantive standards that have been considered and rejected by the citizens of WTO Members. Rather, the standard of review in science-related rules at the WTO should facilitate an international discourse of rational decision making in the regulatory sphere.’Footnote ⁶⁹

The amicus brief, applying a post-normal science analysis of the situation, considers GM engineering and the products thereof to constitute low systems certainty with respect to the scientific knowledge and to entail low consensus among the public, in which case the requirement of a rational relationship between the risk and measure should be interpreted more broadly.Footnote ⁷⁰ Additionally, risk assessment in the case of GMOs should be seen as an evolving technical practice and should be locally based since it is ‘unrealistic to envisage a global risk assessment that could assess the environmental risks posed by particular GM varieties in all, or even many, environments’.Footnote ⁷¹

To what extent is this view sustainable given the situation in EC–Biotech where the Panel failed to acknowledge the arguments put forward by the EC, in support of Austria's invocation of a safeguard measure for Bt maize varieties, as a precautionary measure on the basis of Article 5.7 SPS? The situation is as follows. As is well established, Article 5.1 SPS when read in conjunction with Article 2.2 SPS requires that there must be a rational relationship between a risk assessment and the measure or in the words of Appellate Body in EC–Hormones the results of the risk assessment must ‘sufficiently warrant’ or ‘reasonably support’ the SPS measure at issue Footnote ⁷²

However, the Panel in EC–Biotech considered that in light of Article 5.1 SPS it was specifically called upon to determine whether each individual safeguard measure was based on a risk assessment which was appropriate to the circumstances existing at the time the Panel was established, i.e. as of 29 August 2003. Whereas in EC–Hormones the Appellate Body went to some lengths to stress that a risk assessment could encompass both mainstream scientific opinion as well as divergent views without detracting from the rational relationship test, ‘especially where the risk involved [was] life-threatening in character and [was] perceived to constitute a clear and imminent threat to public health and safety’,Footnote ⁷³ the Panel in EC–Biotech dismissed the Appellate Body's pronouncement in EC–Hormones as only relating to a hypothetical situation.

It went on to distinguish the situation of GMOs with that of beef hormones on two separate grounds. First, the Panel claimed that it was unaware of any divergent scientific evidence relating to Austria's risk assessment in support of its safeguard measure. Second, based on the evidence provided to it by the Panel's scientific experts, it had formed the view that potential risks arising from the deliberate release of various Bt maize varieties and other biotech products in dispute could not be considered as risks that were ‘life-threatening in character’ or that ‘constitute[d] a clear and imminent threat to public health and safety’.Footnote ⁷⁴

While it was made clear to the Panel that the Austrian safeguard measure in respect of Bt maize varieties was based on the availability of ‘new scientific information’, which justified Austria's differing risk assessment, the Panel remained intransient. It was adamant that where there is disagreement ‘with some or all of the conclusions’ contained in the main EC risk assessment then it would be up to individual Member States, like Austria, to explain ‘how and why they assess the risks differently, and to provide their revised or supplemental assessment of the risks’, i.e. to muster evidence that could pass a new risk assessment.

However, when it came to individual EC Member States’ invocation of safeguard measures with respect to GM crops, the Panel disagreed with the EC that the insufficiency of scientific evidence under Article 5.7 SPS must be assessed in relation to the appropriate level of protection of the importing Member. Instead, sufficiency of scientific evidence to support a risk assessment must be consistent with the requirements of Article 5.1’Footnote ⁷⁵ and as defined in Annex A.4.Footnote ⁷⁶ The Panel then goes further in finding that the individual safeguard measures, such as the Austrian one, could not be justified as precautionary under Article 5.7 SPS because the risk assessment conducted by the EC's Scientific Committee on Plants had concluded that the information provided by Austria ‘did not constitute new scientific information which would change the original risk assessment’ that had been carried out in the context of the original EC regime for approval of biotech products.Footnote ⁷⁷

In other words, since the EC was able to maintain its original risk assessment in the light of the ‘new scientific information’ provided by Austria this confirmed its original risk assessment and meant that individual EC Member States were no longer in a position to claim that there was insufficient scientific evidence.Footnote ⁷⁸ The Panel's reasoning is irrational and stems from its suggestion, implicit in its reasoning, that once one WTO Member (the EC) has been able to make a risk assessment based upon scientific evidence, another Member (Austria) can no longer claim that there is insufficient scientific evidence and is thereby precluded from relying upon Article 5.7 SPS. This does little to facilitate an international discourse of rational decision making in the SPS regulatory sphere as the social science experts propose in their amicus. Even the Panel appears to have been sensitive about its findings on this point (leaked at the interim report stage) when it issued an addendum to its Report (Annex K), no doubt in order correct any misunderstandings that a broader public, anxious about a novel technology and the state of the relevant science, might have.Footnote ⁷⁹

A final note of caution in the social science experts' amicus relates to the appropriate selection and use of (scientific) experts. Essentially, what is proposed is ‘that appointed experts not be afforded greater authority in matters of science policy than national authorities’.Footnote ⁸⁰ The concern here addresses the Panel's discretion, as the ‘trier of the facts’, which should extend only to taking the views of experts into account and not to allowing scientific truth to privilege the proper application of WTO measures. In EC–Biotech, the Panel sought expert advice and invited suggestions for experts from the parties before making its final selection on scientific expertise.Footnote ⁸¹ While the choice of experts, the conduct of the proceedings involving the hearing of scientific evidence, and records of those proceedings in no way suggest that these experts acted in excess of their expertise or authority in the relevant scientific field, in future WTO panels might do well to heed the caution in the amicus brief and seek to broaden the basis upon which they call upon expertise.

6. Some conclusions concerning the potential use of social science expertise

The EC–Biotech dispute highlights the evolving relationship between science and governance in the field of agrobiotechnology and multilateral trade. It is clear that there is a growing tension between WTO Members as to the proper role of risk assessment and risk management in SPS matters. However, it would appear from the findings of the Panel in the GMO dispute that thus far social science expertise in general and post-normal science in particular has made little or no impression at all upon the ongoing debate.

Even so, WTO Members continue to have legitimate concerns over key provisions in the SPS Agreement concerning the ‘scientific justification’ for any measure they may adopt and they have latent anxieties concerning the provision of an adequate ‘risk assessment’. The sound science approach, which underpins the SPS Agreement, places a high burden of justification on Members wishing to regulate health and safety measures. The burden is even greater where there is low certainty as to the science as in the case of GM products and there is low consensus among a Member's citizens as to the viability of such products. Meanwhile the SPS Agreement permits complainants to challenge those measures on the basis of sufficient scientific evidence and reasoning, whereas it would seem that the possibility of maintaining a measure of a provisional nature under Article 5.7 SPS might have its limitations.

One of the key points about the amicus brief is the authors’ claim that social science research offers different sources of knowledge, which can compare risk-based decision making across different national political systems and can demonstrate how variations in the framing of issues and the conduct of science policy ‘can lead to systematic transnational variations in the assessment of health, safety and environmental risks’.Footnote ⁸² Arguably, the SPS Agreement, given its open-ended understanding of science and notion of risk assessment is open to a diversity of regulatory approaches, including the precautionary approach, and multiple science-based arguments.Footnote ⁸³

However, WTO decision makers, as in the case of the Panel in EC–Biotech, are minded to apply an objective assessment of the facts in SPS cases. While there is nothing inherently wrong in this approach, Panels have tended to favour particular notions of science and risk assessment over others – the main one being the sound science approach. If nothing else, the amicus merits consideration for its rigorous exposure of the notion of risk assessment, in particular its critique of the sound science approach to risk assessment, which is conducted as ‘an objective, science-based, value-free analytic exercise requiring precise identification of possible harms to human health, agricultural production, and the environment’Footnote ⁸⁴ rather than a broader approach (possibly a post-normal one), which considers scientific evidence alongside other relevant social, cultural, political and regulatory aspects relating to risk.

While in theory such an approach could be achieved by means of a process of public deliberation, which draws upon a broader range of individuals or groups in WTO risk decision-making processes, the reality is that even in domestic systems like the US, and to some extent the EC, there is frequent lack of understanding as to what constitutes an effective public participation mechanism, in which constituencies should participate, and of how the process of public deliberation should function in the policy-making process. Transposed to the supranational level this suggestion presupposes that participatory reforms exist at the WTO to facilitate public deliberation but this is far from reality.

Even if social science expertise is not acceptable to a WTO panel, elsewhere it has been proposed that when seeking scientific advice, WTO panels ‘should step away from their preference of appointing individual scientists, to whom they then submit a number of questions, in favour of appointing an advisory experts group or community’.Footnote ⁸⁵ What is sought is a broadening of the scope of the scientific expertise upon which a panel may draw by allowing the expert scientific advisory group to establish its own mandate and to discuss and evaluate scientific matters, including social science arguments, before submitting its conclusions to the panel. The benefits that can be derived from this approach are two-fold. First, instead of WTO decisionmakers seeking to establish common ground between divergent scientific opinions, the scientific experts themselves would be in a position to do this. Second, an expert scientific advisory group would be better placed to deal with the formal requirements for assessing and depicting issues such as uncertainty and variability in different risk assessments and in demonstrating how risk varies between citizens in different societies.

Were evidence to be elicited from social scientists by panel-appointed experts, the panel hearing a particular complaint would at all times maintain control over the selection and appointment of persons most qualified to advise it. Again, from a procedural point of view it would be up to the panel in question to determine the weight to be attributed to any relevant evidence drawn from social science expertise that might assist it in reaching its findings with respect to a risk assessment or indeed any other scientific matter. From an evidential point of view, such experts, once appointed, could be subject to examination and cross-examination by parties in the dispute in much the same way as currently occurs with scientific experts but in a manner which would be more transparent than the medium of an amicus brief, however well-informed its authors might be.Footnote ⁸⁶

What is becoming apparent in the use of scientific expertise in WTO decision – and this was already recognized by the Appellate Body in EC–Hormones – is that notions of ‘risk’, sufficiency of ‘scientific evidence’, and situations of ‘scientific uncertainty’ are increasingly untenable in isolation and can no longer be separated from their social dimension. The social science experts’ amicus may have made no impact on the course, or the outcome, of the EC–Biotech dispute but it does contribute significantly to making the trade and scientific communities more aware of the need to engage in a holistic approach towards the introduction of novel technologies like GMOs and their application in our global society.