1 Steele, J., Risks and Legal Theory (2004), 4, 6–7, 18–21, 29Google Scholar; Giddens, A., ‘Risk and Responsibility’, (1999) 62 Modern Law Review 1, at 3CrossRefGoogle Scholar. Other authors employ the term ‘risk’ differently. Beck's term ‘risk society’ denotes a society whose dominant feature has become the distribution of pervasive risks that can no longer be completely controlled. U. Beck, Risk Society – Towards a New Modernity (1992). Cultural theorists employ the term ‘risk’ to denote world views of actors concerning the resilience or sensitivity of the environment. M. Douglas and A. Wildavsky, Risk and Culture (1982).

2 Giddens stresses this element of enabling risky choices to obtain benefits. See Giddens, supra note 1, at 3–4.

3 C. Alexandru Parvu, ‘Deliberative Administration and Risk Regulation’, available at http://absp.spri.ucl.ac.be/dgt-actionpublique.htm, at 4; similarly, Wynne, B., ‘Uncertainty and Environmental Learning: Reconceiving Science and Policy in the Preventative Paradigm’, (1992) 2 Global Environmental Change 111CrossRefGoogle Scholar (distinguishing between risk, with known probability; uncertainty, due to lack of a probability determination, indeterminacy, when causation mechanisms cannot be predicted; and uncertainty). Two examples may be useful to illustrate the difference between situations with a lack of probability and those with ambiguity. A lack of probability exists where it is known that a substance causes cancer but no dose-response curve can be established. Ambiguity exists where it is known that a substance is harmful but the harmful effects can range from morbidity to mortality at the same exposure level.

4 Giddens, supra note 1, at 1.

5 N. Luhmann, ‘Die Beschreibung der Zukunft’, in N. Luhmann, Beobachtungen der Moderne (1992), at 129, 140–1, 146; N. Luhmann, ‘Risiko und Gefahr’, in N. Luhmann, Soziologische Aufklärung 5 Konstruktivistische Perspektiven (2005), at 126, 128, 130, 152.

6 Giddens, supra note 1, at 4.

7 Luhmann, ‘Risiko und Gefahr’, supra note 5, at 155.

8 There are no clear protocols for how to combine expertise from different scientific specializations within one discipline, let alone across disciplines. Determining an individual's risk of death from cancer from exposure to a substance may hinge on combining standard toxicology testing with molecular and cell biology and oncology.

9 This function encompasses the control and review of decisions through administrative law and judicial review of administrative decisions, but also through tort law.

10 Suffice it to say that there are many difficulties with the accurate valuation of costs and benefits, notably whether to use ‘objective’ evaluations or evaluations based on stated or revealed citizen preferences and difficulties with how to compare costs and benefits of two different activities. The rationality of decisions based on cost–benefit analysis is thus not clear-cut.

11 If two different hazards can be evaluated against a single benchmark (e.g. health expenditure), it becomes possible to compare their risks.

12 N. Luhmann, ‘Risiko und Gefahr’, supra note 5, at 142, 152.

13 Ibid., at 138–9, 148.

14 The Japan – Apples report already contained some suggestions to this effect. See A. Herwig, ‘The Precautionary Principle in Support of Practical Reason: An Argument against Formalistic Interpretations of the Precautionary Principle’, in C. Joerges and E. U. Petersmann (eds.), Multilevel Trade Governance, Social Regulation and the Constitutionalization of International Trade Law (2006), 301, at 314–17. However, the facts of Japan – Apples were different, as direct scientific evidence showed that the less corroborated indirect and circumstantial evidence adduced by Japan was baseless. Panel Report, Japan – Measures Affecting the Importation of Apples, adopted 10 December 2003, WT/DS245/R, paras. 8.114–19, 8.125–7, 8.132–5, 8.116; Appellate Report, Japan – Measures Affecting the Importation of Apples, adopted 10 December 2003, AB-2003–4, WT/DS245/AB/R. For another discussion of the report, see Goh, G., ‘Tipping the Apple Cart: The Limits of Science and Law in the SPS Agreement after Japan–Apples’, (2006) 40 Journal of World Trade 655Google Scholar.

15 Risk regulation can be analytically distinguished as the phases of risk assessment, risk evaluation, and risk mitigation. Risk assessment involves technical or scientific expert knowledge; risk evaluation – judging the acceptability of risk – draws on normative or ethical value considerations; while mitigation concerns the means taken to reduce risk to acceptable levels.

16 A. Arcuri, ‘The Post-discriminatory Era of the WTO: Toward World-Wide Harmonization of Risk Law?’, working paper, 2005, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=974981.

17 SPS Agreement, Arts. 2(2) and 5(1) Appellate Report, European Communities – Measures Concerning Meat and Meat Products (Hormones), adopted 13 February 1998, AB-1997-4, WT/DS26/AB/R, WT/DS48/AB/R, paras. 129, 250 (hereinafter Appellate Report, EC – Hormones); Appellate Report Australia – Measures Affecting Importation of Salmon, adopted 6 November 1998, AB-1998-5, WT/DS18/AB/R, para. 138.

18 SPS Agreement, Art. 5(7).

19 Appellate Report, EC – Hormones, supra note 17, at paras. 104, 172.

20 SPS Agreement, Art. 5(5).

21 SPS Agreement, Art. 5(6).

22 Victor, D., ‘The Sanitary and Phytosanitary Agreement of the World Trade Organization: An Assessment after Five Years’, (2000) 32 New York University Journal of International Law and Policy 865, at 872, 879–80, 913, 926Google Scholar; J. Söderbom, ‘Balancing National Sovereignty against Disguised Protectionism’, World International Community Experts Report, 2004, available at http://world-ice.com/Articles/SPS.pdf, at 9.

23 L. Gruszczynski, ‘The Role of Science in Risk Regulation under the SPS Agreement’, EUI Working Paper Law No. 2006/03, 2006, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=891114, at 25–6 (with further references).

24 Ibid.

25 Howse, R., ‘Democracy, Science, and Free Trade: Risk Regulation on Trial at the World Trade Organization’, (2000) 98 Michigan Law Review 2329, at 2330, 2334–8, 2341–4CrossRefGoogle Scholar.

26 Ibid.

27 C. Joerges, ‘Juridification Patterns for Social Regulation and the WTO: A Theoretical Framework’, Transstate Working Paper 17, 2005, available at http://www.sfb597.uni-bremen.de/homepages/joerges/arbeitspapierBeschreibung.php?ID=18&SPRACHE=de&USER=joerges, 30; C. Joerges and C. Godt, ‘Free Trade: The Erosion of National and the Birth of Transnational Governance’, in M. Zürn and S. Leibfried (eds.), Transformation of the State (2005), 93, at 111. Gruszczynski also notes but does not conclusively assert that the SPS Agreement fosters the external accountability of states for transboundary effects. Gruszczynski, supra note 23, at 29.

28 Joerges, supra note 27, at 16, 18–19, 21.

29 T. Makatsch, Gesundheitsschutz im Recht der Welthandelsorganisation (WTO): Die WTO und das SPS-Übereinkommen im Lichte von Wissenschaftlichkeit, Verrechtlichung und Harmonisierung (2004), passim.

30 See also Cheyne, I., ‘Risk and Precaution in World Trade Organization Law’, (2006) 40 Journal of World Trade 837, at 841Google Scholar.

31 Goh, supra note 14, at 675.

32 Joerges, supra note 27, at 18–19, 35.

33 J. Peel, ‘Risk Regulation under the WTO/SPS Agreement: Science as an International Normative Yardstick’, New York University School of Law Jean Monnet Working Paper, New York, 02/04, 2004, available at http://www.jeanmonnetprogram.org/papers/04/040201.pdf, 95–7.

34 Peel, supra note 33, at 95–6.

35 Ibid.

36 M. Echols, Food Safety and the WTO: The Interplay of Culture, Science and Technology (2001), at 3 ff. and 148 ff.

37 For an excellent overview of the various sources of uncertainty in risk assessment, see E. K. Silbergeld, ‘Risk Assessment and Risk Management: An Uneasy Divorce’, in D. G. Mayo and R. D. Hollander (eds.), Acceptable Evidence: Science and Values in Risk Management (1991), 99, at 104–7.

38 Ibid., at 99, 104, 110–11.

39 D. Mayo, ‘Sociological versus Metascientific Views of Risk Assessment’, in Mayo and Hollander, supra note 37, at 252–3, 257, 267–8. For instance, standards of good science can tell us something about whether an extrapolation from animal bioassays to humans is empirically reasonable even if science cannot prove that animal bioassays always correctly predict risk in humans. In that sense, uncertainty remains.

40 Ibid., at 253, 256, 259–61, 265–6. Mayo shows how collapsing science into values reaffirms positivist, neutral science as the unattainable ideal and makes scientific criticisms of scientifically doubtful evaluations of scientific evidence impossible.

41 Arcuri, supra note 16, at 22. As is well known, the Appellate Body in EC – Hormones rejected the panel's view that risk assessment must be kept free of risk management considerations. Appellate Body Report, EC – Hormones, supra note 17, para. 181. It remains unclear whether the Appellate Body thereby included a broad range of policy considerations or only a more restricted, largely technical notion of risk management as the process whereby risks are controlled or minimized.

42 See also Arcuri, supra note 16, at 15–16, 26–7.

43 Panel Report, European Communities – Measures Affecting the Approval and Marketing of Biotech Products, adopted 21 November 20006, WT/DS291/R, WT/DS292/R, WT/DS293/R, paras. 7.432 (hereinafter Panel Report EC – Biotech).

44 SPS Agreement, Annex A(1).

45 Panel Report EC – Biotech, supra note 43 at paras. 7.1379, 7.1382.

46 Ibid., at para. 7.1335.

47 Ibid., at para. 7.1381.

48 Ibid., at para. 7.1393.

49 Ibid., at para. 7.1359.

50 Ibid., at paras. 7.1371, 7.1378.

51 Ibid., at paras. 7.1373, 7.1375, 7.1377.

52 Ibid., at para. 7.1374.

53 Ibid., at paras. 7.1686, 7.1689.

54 Ibid., at paras. 7.1466–570.

55 Ibid., at paras. 7.1553–60, 7.1564–5, 7.1567. See also paras. 7.1524–7, 7.1529.

56 Ibid., at paras. 7.3054, 7.3082, 7.3103, 7.3124, 7.3134, 7.3154, 7.3174, 7.3192, 7.3208.

57 Ibid., at paras. 7.3060–7.3062, 7.3067–7.3069, 7.3085–7.3086, 7.3106, 7.3127, 7.3137, 7.3157, 7.3177, 7.3195, 7.3211.

58 Ibid., at paras. 7.3262, 7.3275, 7.3288, 7.3302, 7.3316, 7.3329, 7.3343, 7.3358, 7.3371.

59 Ibid., at paras. 7.3234–7.3246.

60 Ibid., at paras. 7.3234, 7.3237.

61 Ibid., at paras. 7.3235–7.3237.

62 Ibid., at para. 7.3244.

63 Appellate Report EC – Hormones, supra note 17, at para. 124.

64 Panel Report EC – Biotech, supra note 43, at paras. 7.3260, 7.3273, 7.3286, 7.3300, 7.3314, 7.3327, 7.3341, 7.3356, 7.3369.

65 Panel Report EC – Biotech, supra note 43, at para. 7.3240.

66 The panel's letter to the parties is discussed in Prévost, D., ‘Opening Pandora's Box: The Panel's Findings in the EC – Biotech Dispute’, (2007) 34 Legal Issues of Economic Integration 67, at 85 ffGoogle Scholar.

67 Panel Report EC – Biotech, supra note 43, at para. 7.3098. Lacewings are predatory insects that do not feed on maize.

68 Ibid., at para. 7.3098.

69 Ibid., at paras. 7.3099, 7.3147–7.3148. For one study a further reason was that it did not use the maize variety at issue in the safeguard to feed the prey.

70 Vienna Convention on the Law of Treaties, Art. 31(3)(c), available at http://untreaty.un.org/ilc/texts/instruments/english/conventions/1_1_1969.pdf.

71 Panel Report EC – Biotech, supra note 43, at para. 7.68.

72 Ibid., at para. 7.92.

73 A. Herwig and T. Hüller, ‘Zur normativen Legitimität der Welthandelsordnung’, in M. Hilf and T. Niebsch (eds.), Perspektiven des internationalen Wirtschaftsrechts (2008), 117, at 154 (lack of clear textual support in Vienna Convention for panel's interpretation).

74 Vienna Convention on the Law of Treaties, Preamble.

75 Marceau, G., ‘A Call for Coherence in International Law – Praises for the Prohibition against “Clinical Isolation” in WTO Dispute Settlement’, (1999) 33 Journal of World Trade 87, at 110Google Scholar (against interpreting WTO law without consideration of international law).

76 J. Pauwelyn, ‘A Typology of Multilateral Treaty Obligations: Are WTO Obligations Bilateral or Collective in Nature?’, (2003) 14 EJIL 907, at 925–41. If WTO obligations are inter partes and can be modified through a simple bilateral treaty between two parties, it only seems necessary to ascertain how the WTO members parties to another treaty understood terms in the WTO agreements. Pauwelyn himself would not go as far in J. Pauwelyn, ‘The Role of Public International Law in the WTO: How Far Can We Go?’, (2001) 95 AJIL 535, at 573–6 (only non-WTO treaties establishing the common intention of all WTO members could be drawn upon, including those accepted by acquiescence rather than express consent).

77 For instance, in fragile ecosystems such as the rainforest, the introduction of alien species can have more severe consequences than in more robust ecosystems. Because the threshold for harm (invasiveness) differs, it would be rational to accept evidence of lesser scientific corroboration as a sufficient ground for regulation in a regional treaty.

78 C.-D. Ehlermann and L. Ehring, ‘Decision Making in the World Trade Organization: Is the Consensus Practice of the World Trade Organization Adequate for Making, Revising and Implementing Rules on International Trade?’ (2005) 8 Journal of International Economic Law 51.

79 L. Busch et al., ‘Amicus Curiae Brief Submitted to the Dispute Settlement Panel in the Case of EC: Measures Affecting the Approval and Marketing of Biotech Products’, available at http://www.lancs.ac.uk/fss/iepp/wtoamicus/amicus.brief.wto.pdf, at 5.

80 Ibid., at 5.

81 Ibid., at 5; Millstone, E., Brunner, E., and Mayer, S., ‘Beyond “Substantial Equivalence”’, (1999) 401 Nature 525CrossRefGoogle Scholar; Kuiper, ‘Profiling Techniques to Identify Differences between Foods Derived from Biotechnology and their Counterparts’, Joint FAO/WHO Expert Consultation on Foods Derived from Biotechnology, 29 May–2 June 2000, 6.

82 Busch et al., supra note 79, at 5; National Research Council, Genetically Modified Pest-Protected Plants: Science and Regulation (2000), 56; GM Science Review Panel, ‘GM Science Review First Report’, available at www.gmsciencedebate.org.uk, at 80, 87.

83 A. O. Sykes, ‘Domestic Regulation, Sovereignty, and Scientific Evidence Requirements: A Pessimistic View’, repr. in P. C. Mavroidis and A. O. Sykes (eds.), The WTO and International Trade Law/Dispute Settlement (2005), 178, at 189.

84 Peel, supra note 33, at 95–7, 99.

85 Winickoff, D. et al. , ‘Adjudicating the GM Food Wars: Science, Risk and Democracy in World Trade Law’, (2005) 30 Yale Journal of International Law 81, at 85, 93 ffGoogle Scholar. and especially 113; Cheyne, supra note 30, at 840, 842. That laypersons not only judge risks by reference to the potential of a hazard but use certain heuristics has been demonstrated in literature on risk perception. To them, mortality has to be evaluated differently depending on the dread factor of a risk or the familiarity with the underlying technology. For an overview of recent studies in risk perception see P. Slovic, The Perception of Risk (2000), and for an early study see Slovic, P., ‘Perception of Risk’, (1987) 236 Science 280CrossRefGoogle ScholarPubMed. Although these heuristics may sometimes lead laypersons to underestimate certain risks and it is not claimed that they must be accepted at face value in decisions about how to regulate hazards, it cannot be said that laypersons' qualitative and contextual understanding of hazards is necessarily less rational than the abstract understanding of scientists focusing only on the hazard itself.

86 Alexandru Parvu, supra note 3, at 2.

87 Ibid.

88 The following paragraph is based on Herwig, supra note 14, at 306.

89 Similarly, Perez, O., ‘Anomalies at the Precautionary Kingdom: Reflections on the GMO Panel's Decision’, (2007) 6 World Trade Review 265, at 279 ffCrossRefGoogle Scholar.

90 Peel, supra note 33, at 95–7, 99.

91 Jasanoff, S., The Fifth Branch: Science Advisers as Policymakers (1990), 42, 77–9Google Scholar (arguing that filling gaps in the knowledge base, knowledge synthesis, and prediction of uncertain events are key features of regulatory science distinguishing it from laboratory science and that the same evidentiary standards should not apply to both).

92 Reasonableness tests are advocated by Cheyne, supra note 30, at 842; O. Perez, Ecological Sensitivity and Global Legal Pluralism: Rethinking the Trade and Environment Conflict (2004), 145 f., 156; Sykes, supra note 83, at 190; and Walker, V. R., ‘Keeping the WTO from Becoming the “World Trans-science Organization”: Scientific Uncertainty, Science Policy and Fact-Finding in the Growth Hormones Dispute’, (1998) 31 Cornell International Law Journal 251, 280 ffGoogle Scholar.

93 K. Lehrer, A Theory of Knowledge (2000), at 97–121.

94 Ibid., at 101, 105.

95 Sykes, supra note 83, at 189. Gruszczynski, supra note 23, at 17.

96 SPS Agreement, Annex A(4).

97 Appellate Report Japan – Measures Affecting Agricultural Products, adopted 19 March 1999, WT/DS76/AB/R, para. 84.

98 Ibid., at paras. 73–74.

99 SPS Agreement, Art. 2(1) and Preamble.

100 SPS Agreement, Art. 5(5).

101 Herwig, supra note 14, at 320–1.

102 Discussed in M. Pollack and G. Shaffer, ‘The Challenge of Reconciling Regulatory Differences: Food Safety and GMOs in the Transatlantic Relationship’, in M. Pollack and G. Shaffer (eds.), Transatlantic Governance in the Global Economy (2001), 153, at 168.