Risk Regulation and Administrative Constitutionalism 9781472560117, 9781841130330

Over the last decade the regulatory evaluation of environmental and public health risks has been one of the most legally

188 65 2MB

English Pages [302] Year 2007

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Risk Regulation and Administrative Constitutionalism
 9781472560117, 9781841130330

Citation preview

(A) Fisher Prelims

10/5/07

10:15

Page v

For Roderick

(A) Fisher Prelims

10/5/07

10:15

Page vi

(A) Fisher Prelims

10/5/07

10:15

Page vii

ACKNOWLEDGEMENTS This book has had a very long gestation period. Its origins lie in my taking two undergraduate courses in the third year of my Arts/Law degree in 1990 at the University of New South Wales—a course on administrative law as part of my Law Degree and a course on the sociology of risk in the School of Science and Technology Studies. The former was a straightforward doctrinal course on Australian administrative law and the latter a very theoretical and interdisciplinary course concerned with technological risk decision-making. Despite these differences, I not only found myself having to think about the role and nature of public administration in both, but scholars and decision-makers seemed to be struggling with the exact same issues. Yet with that said, there was very little dialogue between these two disciplines. To me this seemed a pity because the scholars in each discipline seemed to be holding different pieces of a larger jigsaw puzzle. Over the past decade and a half I have been dedicated myself to putting together that jigsaw puzzle. While I have been publishing on these issues for several years, this book is my first attempt to provide a more comprehensive overview of these issues. I say first, because the purpose of this book is to reframe debate in this area rather than be a last word on the issue. Indeed, as will become clear this book raises more questions than it answers, but the questions it raises are important ones and cannot be ignored. Since 1990 there are many people who have helped, inspired and deliberated with me and whom I would like to thank. I would like to thank Mark Aronson, Ronnie Harding and Gavan McDonell who were at the University of New South Wales while I was there. While working in very different disciplines, they each motivated me to pursue these issues and this book lies very much at the intersection of their interests. I would also like to thank Paul Craig, the supervisor of my DPhil while I was a graduate student at St John’s College, Oxford. Some of that doctoral work can be seen in Chapter Three of this book but Paul’s contribution to this book extends beyond that chapter. In particular, Paul taught me the importance of writing about what I wanted to write about rather than what I thought others wanted me to write about. Much of the work in this book has been given as seminar papers in a variety of different academic forums in Europe, the United States, and Australia. Several of the chapters in this book have also been published elsewhere in earlier, and shorter, versions. The list of those that have discussed these issues with me and given feedback on those papers and publications is a very long one. I cannot even attempt to name them all here but in particular I would like to thank Nick Barber, Peter Cane, Michelle Everson, David Faulkner, Adam Finkel, Mark Freedland,

(A) Fisher Prelims

10/5/07

viii

10:15

Page viii

Acknowledgements

Dan Kelemen, Sheila Jasanoff, Judith Jones, Christian Joerges, Pasky Pascual, Mahla Pearlman, Paul Stein, Andrew Stirling, Rene von Schomberg, Paddy van Zwanderberg, Ellen Vos, Wendy Wagner, Vern Walker, Nick Wikeley, David Wirth, Brian Wynne, and Lucia Zedner for taking the time to discuss these issues with me. While they may not agree with everything in this book I do believe that their critical and honest comments have made it a better piece of work. I would also like to thank Richard Hart who has been the most patient, encouraging and understanding of editors. In researching this book I also received support from a number of other quarters. I thank everyone at Corpus and the Law Faculty for making Oxford such a wonderful and vibrant place to work. I am also grateful to the Oxford University Faculty of Law for providing financial support for travel and research assistance. Jaswinder Kaur, Eloise Scotford, Sameer Singh and Sandeep Sreekumar were all brilliant and highly professional research assistants. Research for Chapter Three was part of my doctoral studies for which I had a Commonwealth Scholarship for. Research for Chapter Four was done while a Visiting Fellow at the Law Program at the Research School of Social Sciences at the Australian National University and the Faculty of Law, University of New South Wales in 2005. In conclusion I would like to thank Roderick’s and my family. Jill, Jo, Andrew, Victoria and Ian have always been encouraging and supportive of my work. My two wonderful sons Corin and Arthur, in their own boisterous and lovely way, remind me daily that the concept of reasonable action is contextual. My final and most important thank you is to Roderick Bagshaw who has been an endless source of loving support, academic rigour, intellectual inspiration, and humour. Oxford 1 January 2007

(A) Fisher Prelims

10/5/07

10:15

Page xiii

TABLE OF ABBREVIATIONS AAT AB ACGIH ACT AFMA AG ALARP APA BSE CFI CJD CMO Codex Communication CPSC CVO CVL DC DH EC ECJ EFTA EIA EIS EPA ERDC ESD EU FIS GATT GMO ICC IGAE LEC MAFF NHSTA

Administrative Appeals Tribunal Appellate Body American Conference of Governmental Industrial Hygienists Australian Capital Territory Australian Fisheries Management Authority Advocate General as low as reasonably practicable Administrative Procedure Act 1946 bovine spongiform encephalopathy Court of First Instance Creutzfeldt-Jakob disease Chief Medical Officer Codex Alimentarius Commission Communication from the Commission on the Precautionary Principle Consumer Product Safety Commission Chief Veterinary Officer Central Veterinary Laboratory deliberative-constitutive Department of Health European Communities European Court of Justice European Free Trade Association environmental impact assessment environmental impact statement Environmental Protection Agency Environmental Resources and Development Court ecologically sustainable development European Union fauna impact statement General Agreement on Tariffs and Trade genetically modified organism Interstate Commerce Commission Intergovernmental Agreement on the Environment Land and Environment Court Ministry of Fisheries and Food National Highways and Safety Transport Administration

(A) Fisher Prelims

10/5/07

xiv NIOSH NRC NSW OIE OMB OSHA OSH Act PEC PEL ppm Qld RI SA SEAC SIS SPS Agreement SVS Tas TEC TBT Agreement TSE UK US Vic v-CJD WA WTO

10:15

Page xiv

Table of Abbreviations National Institute of Occupational Safety and Health National Research Council New South Wales Office international des épizooties Office of Management and Budget Occupational Safety and Health Administration Occupational Safety and Health Act 1970 Planning and Environment Court permissible exposure limit parts per million Queensland rational-instrumental South Australia Spongiform Encephalopathy Advisory Committee species impact statement Sanitary and Phytosanitary State Veterinary Service Tasmania Treaty establishing the European Community Technical Barriers to Trade Agreement transmissible spongiform encephalopathy United Kingdom United States Victoria variant Creutzfeldt-Jakob disease Western Australia World Trade Organisation

(Aa) Fisher TABLES

4/7/07

11:46

Page xv

TABLE OF CASES Australia Acquaro v Great Lakes Council [2005] NSWLEC 582 ........................................... 144 Ajka Pty Limited v Australian Fisheries Management Authority [2001] AATA 258 .......................................................................................................... 144 Aldekerk Pty Ltd v City of Port Adelaide Enfield and Environment Protection Authority [2000] SAERDC 47 ................................................................... 132, 153 Alliance to Save Hinchinbrook v Environmental Protection Agency [2006] QSC 84 ........................................................................................... 128, 142 Alumino (Aust) Pty Ltd v Minister Administering the Environmental Planning and Assessment Act 1979 [1996] NSWLEC 102 ................................................ 145 Australian Conservation Foundation v Commonwealth (1980) 146 CLR 493 ..... 129 Australian Pork Ltd v Director of Animal and Plant Quarantine [2005] FCA 671 ............................................................................................................ 130 Bell v Minister for Urban Affairs and Planning (1993) 95 LGERA 86 ........... 130, 138 Bennett Taylor Pty Ltd v North Sydney Municipal Council [1988] NSWLEC 77 ...................................................................................................... 138 Bentley v BGP Properties Pty Ltd [2006] NSWLEC 34 .................................. 132, 159 BGP Properties Pty Limited v Lake Macquarie City Council [2004] NSWLEC 399 .................................................................................................... 132 Blank v Australian Fisheries Management Authority [2000] AATA 1027 ..............151 Booth v Bosworth [2001] FCA 1453 ...................................................................... 132 Bridgetown/Greenbushes Friends of the Forest Inc v Executive Director of Conservation and Land Management (1997) 18 WAR 102 ............... 133, 142, 145 BT Goldsmith Planning Services v Blacktown City Council [2005] NSWLEC 210 .................................................................................................... 144 Carstens v Pittwater Council (1999) 111 LGERA 1 ............................................... 142 Commerical Crash Repairs Pty Ltd v City of Adelaide [2000] SAERDC 83 ....................................................................................................... 153 Conservation Council of South Australia v Development Assessment Commission & Tuna Boat Owners Association (No 2) [1999] SAERDC 86 ............................................................................ 142, 147–9, 153, 154 CSR Limited v Caboolture Shire Council [2001] QPE 013 .................................... 149 David Kettle Consulting v Gosford City Council [2005] NSWLEC 519 ......... 153, 155 De Brett Investments Pty Ltd v Australian Fisheries Management Authority [2004] AATA 704 .................................................................. 151–3, 155 Director of Animal and Plant Quarantine v Australian Pork Limited [2005] FCAFC 206 ............................................................................................ 130

(Aa) Fisher TABLES

xvi

4/7/07

11:46

Page xvi

Table of Cases

Dixon v Australian Fisheries Management Authority [2000] AATA 442 ........................................................................................... 132, 145, 151 Dow Chemicals (Australia) Ltd v Chemicals Notification and Assessment [1999] AATA 1023 ..................................................................................... 132, 146 Dubler v Ku-ring-gai Muncipal Council (NSWLEC, unreported, 21 December 2001) ........................................................................................... 153 Elliott v Brisbane City Council [2002] QPEC 013 ................................................. 153 Enfield City Corporation v Development Assessment Commission (2000) 199 CLR 135 ...................................................................................................... 129 Fast Buck$ v Byron Shire Council (1999) 103 LGERA 94 ...................................... 137 Friends of Hinchinbrook Society Inc v Minister for the Environment (1997) 142 ALR 632 .............................................................................................. 133, 146 Gales Holdings Pty Limited v Tweed Shire Council [2006] NSWLEC 85 ....... 144, 155 Golden v Coffs Harbour City Council (1991) 72 LGRA 104 .................................. 129 Gray v Minister for Planning [2006] NSWLEC 720 ........................................ 158–60 Green v Australian Fisheries Management Authority [2004] AATA 426 ............... 145 Greenpeace Australia Ltd v Redbank Power Company Pty Ltd (1994) 86 LGERA 143 .................................................................................................... 143–4 Greentree v Colac Otway Shire Council [2005] VCAT 815 .................................... 144 Grishin v Conservator of Flora and Fauna [1998] ACTAAT 250 ............... 132–3, 146 Guthega Development Pty Ltd v Minister Administering the National Parks and Wildlife Act 1974 (1986) 61 LGRA 401 ............................................... 129, 140 Hale v Parramatta City Council (1982) 47 LGRA 269 .......................................... 139 Hasan v Moreland City Council [2005] VCAT 1931 ............................................. 153 Histpark Pty Ltd v Maroochy Shire Council [2001] QPEC 59 ....................... 133, 150 Humane Society International v Minister for the Environment and Heritage [2006] AATA 298 .............................................................................................. 144 Jones v Pristine Waters Council (NSWLEC, unreported, 24 May 2002) ............... 144 Justice v Australian Fisheries Management Authority [2002] AATA 49 ................ 145 Kroger v Southern Rural Water [2001] VCAT 1334 ....................................... 144, 146 Lainson v Sutherland Shire Council [1998] NSWLEC 87 ..................................... 138 Leatch v National Parks and Wildlife Service (1993) 81 LGERA 270 ...................................................................................... 132, 140, 143, 146, 155 Liverpool City Council v Roads and Traffic Authority (1991) 74 LGERA 265 ....... 129 Maxnox Pty Ltd v Hurstville City Council [2006] NSWLEC 146 ......................... 137 McDonald v Director General of Social Security (1984) 6 ALD 6 ...........................139 Mees v Kemp [2004] FCA 366 ............................................................................... 142 Miltonbrook Pty Ltd v Kiama Municipal Council [1998] NSWLEC 281 .............. 145 Minister for Aboriginal Affairs v Peko-Wallsend Ltd (1986) 162 CLR 24 ...... 128, 141 Minister for Immigration and Ethnic Affairs v Pochi (1980) 4 ALD 139 ............... 138 Minister for the Environment and Heritage v Queensland Conservation Council Inc [2004] FCAFC 190 ......................................................................... 129 Mohr v Great Barrier Reef Marine Park Authority [1998] AATA 805 ................... 150 Mol Pty Ltd v City of Mitcham [2002] SAEDRC 55 .............................................. 153

(Aa) Fisher TABLES

4/7/07

11:46

Page xvii

Table of Cases

xvii

Murrumbidgee Groundwater Preservation Association v Minister for Natural Resources [2004] NSWLEC 122 .................................................................... 144–5 Nicholls v Director General of National Parks and Wildlife Service (1994) 84 LGERA 397 .............................................................................. 45, 142, 146, 152 North Queensland Conservation Council v Great Barrier Reef Marine Park Authority [2000] AATA 925 ........................................... 132–3, 139, 144, 146, 155 Northcompass Inc v Hornsby Shire Council [1996] NSWLEC 213 ....................... 145 P & E Turner v Launceston City Council [1998] TASRMPAT 27 .......................... 146 Port Vincent Progress Association v DAC & Colimion P/L [1999] SAERDC 7 ...... 155 Price v Water Administration Ministerial Corporation of New South Wales [2002] NSWLEC 147 ........................................................................................ 132 Prineas v Forestry Commission of New South Wales (1983) 49 LGRA 402 .... 129, 159 Providence Projects Pty Ltd v Gosford City Council [2006] NSWLEC 5 ............... 145 Rashleigh v Environment Protection Authority [2005] ACTSC 18 ........................ 142 Re Costello and Secretary, Department of Transport (1979) 2 ALD 934 ................ 126 Re Minister for Immigration and Multicultural and Indigenous Affairs; Ex parte Applicant S 20/2002 (2003) 198 ALR 59 ........................................................... 129 Re National Parks and Nature Conservation Authority: Ex Parte McGregor [2001] WASCA 368 ................................................................................... 132, 142 Re Pochi and Minister for Immigration and Ethnic Affairs (1979) 2 ALD 33 ........ 138 Rosemount Estates Pty Ltd v Minister for Urban Affairs and Planning (1996) 90 LGERA 1 ....................................................................................................... 130 Rowe v Linder [2006] SASC 176 ............................................................................ 153 Schaffer Corporation Ltd v Hawkesbury City Council (1992) 77 LGRA 21 ...................................................................................... 129–30, 138, 140 Shannon v Dalby Town Council [2004] QPEC 062 ................................. 149–50, 154 Simpson v Ballina Shire Council [1994] NSWLEC 43 ................................... 140, 142 St Ives Development Pty Ltd v Mandurah [2003] WATPAT 5 ............................... 154 Telstra Corporation Ltd v Caloundra City Council [2004] QPEC 85 .................... 153 Telstra Corporation Ltd v Hornsby Shire Council [2006] NSWLEC 133 ....................................................................................................... 132, 155–60 Terminals Pty Ltd v Greater Geelong City Council [2005] VCAT 1988 ................. 153 Timbarra Protection Coalition Inc v Ross Mining NL (1999) 46 NSWLR 55 ........ 129 Tuna Boat Owners Association of South Australia Inc v Development Assessment Commission [2000] SASC 238 ............................................ 142, 148–9 ULV Pty Ltd v Scott (1980) 69 LGRA 212 .............................................................. 137 Yamauchi v Jondaryan Shire Council [1998] QPELR 452 ..................................... 155 Zhang v Canterbury City Council [1999] NSWLEC 209 ...................................... 139

Germany Brunner v European Union Treaty [1994] 1 CMLR 57 .................................. 170, 217

(Aa) Fisher TABLES

4/7/07

11:46

xviii

Page xviii

Table of Cases

United Kingdom Bolton Municipal Council v Malrod Insulations Ltd [1993] ICR 358 ..................... 68 Briscoe v Shattock [1999] Env H 108 ....................................................................... 68 R (on the application of Amvac Chemical UK Ltd) v Secretary of State for Environment Food and Rural Affairs [2001] EWHC Admin 1011 ................... 229 R (On the Application of Thomas Bates & Son Ltd) v Secretary of State for Transport Local Government and the Regions [2005] 2 P & CR 11 ................... 214 R v Board of Trustees for the Science Museum [1993] ICR 876 ................................ 68 R (on the application of Davies) v Carmarthenshire County Council [2004] EWHC 2847 ...................................................................................................... 214

United States Abbott Laboratories v Gardner 387 US 136 (1967) ................................................. 97 AFL-CIO v Marshall 617 F 2d 636 (DC Cir 1979) ............................................ 104–5 AFL-CIO v OSHA 965 F 2d 962 (11th Cir 1992) ............................................ 118–19 Alabama Power Co v OSHA 89 F 3d 740 (11th Cir 1996) ....................................... 19 American Airlines Inc v Civil Aeronautics Board 359 F 2d 624 (DC Cir 1966) ....... 97 American Dental Association v Martin 984 F 2d 823 (7th Cir 1993) ............ 117, 120 American Federation of Labour v Brennan 530 F 2d 109 (3rd Cir 1975) ...... 109, 110 American Iron & Steel Institute v EPA 115 F 3d 979 (DC Cir 1997) .............. 120, 122 American Iron & Steel Institute v OSHA 577 F 2d 825 (3rd Cir 1978) .................................................................................................. 109, 110, 112 American Lung Association v EPA 134 F 3d 388 (DC Cir 1998) ........................... 119 American Petroleum Institute v OSHA 581 F 2d 493 (5th Cir 1978) .............. 110–12 American Ship Building Co v National Labour Relations Board 380 US 300 (1965) .................................................................................................................. 97 American Trucking Associations v EPA 175 F 3d 1027 (DC Cir 1999) .................. 113 American Waterworks Association v EPA 39 ERC 1897 (DC Cir 1994) ................ 120 Amoco Oil Co v EPA 501 F 2d 722 (DC Cir 1974) ................................................. 101 Appalachian Power Co v EPA 135 F 3d 791 (DC Cir 1998) .................................. 120 Aqua Slide N’ Drive Corp v CPSC 569 F 2d 831 (5th Cir 1978) ............................ 108 Arkon, Canton & Youngsville Railway Company v US 261 US 184 (1923) .................................................................................................................. 94 ASARCO v OSHA 647 F 2d 1 (9th Cir 1981) ........................................................ 116 ASARCO v OSHA 746 F 2d 483 (9th Cir 1984) .................................................... 117 Asbestos Information Association of North America v OSHA 727 F 2d 415 (5th Cir 1984) ................................................................................................... 117 Associated Industries of New York Inc v US Dept of Labour 487 F 2d 342 (5th Cir 1973) .................................................................................... 109, 111, 112 Board of Railway Commissioners of the State of North Dakota v Great Northern Railway Company 281 US 416 (1931) ................................................ 95

(Aa) Fisher TABLES

4/7/07

11:46

Page xix

Table of Cases

xix

Builders and Construction Trades Department v Brock 838 F 2d 1258 (DC Cir 1988).................................................................................................... 117 Bunker Hill Co v EPA 572 F 2d 1286 (9th Cir 1977) ............................................. 101 Central Arizona Water Conservation District v EPA 990 F 2d 1531 (9th Cir 1993) ................................................................................................... 119 Centre For Auto Safety v Federal Highway Administration 956 F 2d 309 (DC Cir 1992) ................................................................................................... 122 Centre For Auto Safety v Peck 751 F 2d 1336 (DC Cir 1985) ................................. 123 Certified Colour Manufacturers Association v Matthews 543 F 2d 284 DC Cir 1976) ..................................................................................................... 101 Chemical Manufacturers Association v EPA 870 F 2d 177 (5th Cir 1989) ............ 119 Chicago Burlington & Quincy Railway Company v Babcock 204 US 585 (1907) .................................................................................................................. 94 Citizens to Preserve Overton Park Inc v Volpe 401 US 402 (1971) ......................... 108 Colour Pigments Manufacturers Association v OSHA 16 F 3d 1157 (11th Cir 1994) ................................................................................................. 119 Competitive Enterprise Institute v NHTSA 45 F 3d 481 (DC Cir 1995) ............... 251 Competitive Enterprise Institute v NHTSA 901 F 2d 107 (DC Cir 1990) ...... 117, 251 Competitive Enterprise Institute v NHTSA 956 F 2d 321 (DC Cir 1992) ...... 120, 251 Consolidated Edison Co v National Labour Relations Board 305 US 197 (1938) .................................................................................................................. 95 Corrosion Proof Fittings v EPA 947 F 2d 1201 (5th Cir 1991) ......................... 119–21 Daubert v Merrell Dow Pharmaceuticals 509 US 579 (1993) ............................... 156 Dithiocarbamate Task Force v EPA 98 F 3d 1394 (DC Cir 1996) .......................... 119 Dry Colour Manufacturers Association Inc v Department of Labour 486 F 2d 98 (3rd Cir 1973) ................................................................................ 110–11 East Tennessee Virginia & Georgia Railway Company v ICC 181 US 1 (1901) ....... 93 Edison Electric Institute v EPA 2 F 3d 438 (DC Cir 1993) ..................................... 119 Environmental Defense Fund v EPA 510 F 2d 1292 (DC Cir 1975) ...................... 101 Environmental Defense Fund v Ruckelshaus 439 F 2d 584 (DC Cir 1971) .......................................................................................................... 102, 103 Ethyl Corp v EPA 541 F 2d 1 (DC Cir 1976) ....................... 16, 89, 101, 103, 105, 124 Federal Communications Commission v RCA Communications 346 US 86 (1952) .................................................................................................................. 97 Federal Power Commission v Hope Natural Gas Co 320 US 591 (1944) ................. 95 Federal Power Commission v Natural Gas Pipeline Co 315 US 575 (1942) ............. 95 Federal Radio Commission v Nelson Brothers Bond & Mortgage Co 289 US 266 (1933) .................................................................................................................. 95 Fleming v Florida Citrus Exchange 358 US 153 (1958) ........................................... 97 Florida Peach Growers Association v US Department of Labour 489 F 2d 120 (5th Cir 1974) ........................................................................................... 110–111 Flue-Cured Tobacco Co-op v EPA 4 F Supp 2d 435 (MD NC, 1998) ............... 16, 117 Forging Industry Association v Secretary Of Labour 773 F 2d 1436 (4th Cir 1985) ................................................................................................... 119

(Aa) Fisher TABLES

xx

4/7/07

11:46

Page xx

Table of Cases

Greater Boston Television Corp v FCC 444 F 2d 841 (DC Cir 1970) ............. 102, 103 Gulf South Insulation v CPSC 701 F 2d 1137 (5th Cir 1983) ........................ 117, 119 ICC v Brimson 154 US 447 (1894) .......................................................................... 93 ICC v Louisville & Nashville Railway Company 227 US 88 (1913) ......................... 94 ICC v Union Pacific Railway Company 222 US 541 (1912) .................................... 94 In Re Permian Basin Area Rate Cases 390 US 747 (1968) ..................................... 109 Industrial Union Dept AFL-CIO v American Petroleum Institute 448 US 607 (1980) Benzene ...................................................................... 19, 113–16, 119, 248 Industrial Union Dept, AFL-CIO v Hodgson 499 F 2d 467 (DC Cir 1974) ..... 108–12 International Association of Machinists, Toolmakers & Diemakers Lodge No 35 v National Labour Relations Board 311 US 72 (1940) .............................. 95 International Harvester v Ruckelshaus 478 F 2d 615 (DC Cir 1973) ......... 102–4, 120 International Union, UAW v Pendergrass 878 F 2d 389 (DC Cir 1989) Formeldehyde................................................................................................ 116–7 Kelley v Selin 42 F 3d 1501 (6th Cir 1995) ............................................................ 120 Leather Industries of America v EPA 40 F 3d 392 (DC Cir 1994) ...................... 118–9 Love v Thomas 858 F 2d 1347 (9th Cir 1988) ........................................................ 119 Mobil Oil Corp v FPC 483 F 2d 1238 (DC Cir 1973) ............................................ 111 Motor Vehicles Manufacturers Association v State Farm Mutual Automobile Insurance Company 463 US 29 (1983) .............................................................. 117 National Asphalt Pavement Association v Train 539 F 2d 775 (DC Cir 1976) ...... 101 National Grain and Feed Association v OSHA 866 F 2d 717 (5th Cir 1989) .................................................................................................. 111, 117, 119 National Lime Association v EPA 627 F 2d 416 (DC Cir 1980) ............................. 101 Natural Resources Defence Council v Administrator of EPA 902 F 2d 962 (DC Cir 1990) ................................................................................................... 122 Natural Resources Defense Council v EPA 822 F 2d 104 (DC Cir 1987) ............... 120 Natural Resources Defense Council v EPA 824 F 2d 1211 (DC Cir 1987) ............. 119 Natural Resources Defense Council v Nuclear Regulatory Commission 547 F 2d 633 (DC Cir 1976) ............................................................................ 105–6, 252 Natural Resources Defense Council v Thomas 805 F 2d 410 (DC Cir 1986) ......... 119 New York v EPA 133 F 3d 987 (7th Cir 1998) ........................................................ 119 National Labour Relations Board v Wyman-Gordon 394 US 759 (1969) ............... 97 Nor Am Agricultural Products Inc v Hardin 435 F 2d 1151 (7th Cir 1970) ........... 101 Portland Cement Association v Ruckelshaus 486 F 2d 375 (DC Cir 1973) ...................................................................................................... 102–3, 116 Public Health Research Group v Tyson 796 F 2d 1479 (DC Cir 1986) ................... 113 Reserve Mining Co v EPA 514 F 2d 492 (8th Cir 1975) ......................................... 101 Society of Plastics Industry Inc v OSHA 509 F 2d 1301 (2nd Cir 1975) .......... 109–13 Sunshine Anthracite Coal Co v Adkins 310 US 381 (1940) ...................................... 95 Synthetic Organic Chemical Manufacturers Association v Brennan 503 F 2d 1155 (3rd Cir 1974) ............................................................................ 108–11 Synthetic Organic Chemical Manufacturers Association v Brennan 506 F 2d 385 (3rd Cir 1974) ............................................................................... 109–12

(Aa) Fisher TABLES

4/7/07

11:46

Page xxi

Table of Cases

xxi

Tagg Brothers & Moorhead v US 280 US 420 (1930) .............................................. 95 Texas Independent Ginners Association v Marshall 630 F 2d 398 (5th Cir 1980) ............................................................................................. 111–12 Thompson v Clark 741 F 2d 401 (DC Cir 1984) ................................................... 120 United Steelworkers of America v Auchter 763 F 2d 728 (3rd Cir 1985) ................ 112 United Steelworkers of America v Marshall 647 F 2d 1189 (DC Cir 1980) .... 117, 119 Universal Camera Corp v National Labour Relations Board 340 US 474 (1951) .............................................................................................................. 96–7 US v Morgan 313 US 409 (1940) ............................................................................. 95 Vermont Yankee Nuclear Power Corp v Natural Resources Defense Council 435 US 519 (1978) ............................................................................... 99, 107, 121 Walter Holm & Co v Hardin 449 F 2d 1009 (DC Cir 1971) .................................. 102 Welford v Ruckelshaus 439 F 2d 598 (DC Cir 1971) .............................................. 102 Weyerhaeuser Co v Costle 590 F 2d 1011 (DC Cir 1978) ...................................... 101

European Free Trade Authority EFTA Court, Report for the Hearing in Case E-3/00 prepared by Carl Baudenbacher, Judge-Rapporteur (E-3/00/53, 2000) ....................................... 230 Case E-3/00 EFTA Surveillance Authority v Norway [2001] 2 CMLR 47 ......................................................................................................... 229–30, 238 Case E-4/04 Pedicel AS v Sosial- OG Helsedirektoratet (Directorate for Health and Social Affairs) [2005] 2 CMLR 7 ................................................... 231

European Union Case 9/56 Meroni v ECSC High Authority [1957-8] ECR 133 ...................... 172, 212 Case 8/74 Procureur du Roi v Dassonville [1974] ECR 837 .................................. 222 Case 120/79 Rewe-Zentral AG v Bundesmonopolverwaltung für Branntwein [1979] ECR 649 ................................................................................................. 213 Case 53/80 Koninklijke Kaasfabriek Eyssen BV [1981] ECR 409 .............. 216, 221–2 Case 174/82 Sandoz BV [1983] ECR 2445 ........................................ 221–2, 230, 237 Case 227/82 Bennekom [1983] ECR 3883 ........................................................ 221–2 Case 176/84 Commission v Greece [1987] ECR 1193 ................................ 216, 221–2 Case 178/84 Commission v Germany [1987] ECR 1227 ........................... 216, 222–3 Case 247/84 Motte [1985] ECR 3887 ........................................................ 216, 221–2 Case 304/84 Muller [1985] ECR 1511 ........................................................... 216, 222 Case 54/85 Mirepoix [1986] ECR 1067 ..................................................... 216, 221–3 Case 68/86 United Kingdom v Council [1988] ECR 855 ................................... 218–9 Case C-331/88 R v Minister of Agriculture, Fisheries and Food and Secretary of State for Health, ex parte FEDESA [1990] ECR I-4023 ..................... 214, 218–9 Case C-2/90 Commission v Belgium [1992] ECR I-4431 ..................................... 216

(Aa) Fisher TABLES

xxii

4/7/07

11:46

Page xxii

Table of Cases

Case C-52/92 Commission v Portugal [1993] I-2961 ............................................ 220 Case C-284/95 Safety Hi-Tech [1998] ECR I-4301 .............................................. 219 Case C-149/96 Portugal v Council [1999] ECR I-8395 ......................................... 213 Case C-180/96 United Kingdom v Commission [1998] ECR I-2265 .............. 219–20 Case C-67/97 Anklagemyndigheden v Ditlev Bluhme [1998] ECR I -8033 .......... 216 Case C-293/97 R v Secretary of State for the Environment, ex parte Standley [1999] ECR I-2603 .................................................................................... 212, 220 Case C-418-419/97 Arco Chemie Nederland Ltd v Minister Van Volkshuis vesting, Ruimtelijke Ordening En Milieubeheer [2000] ECR I-4475 ................. 220 Case C-318/98 Fornasar [2000] ECR I-4785 ................................................ 212, 220 Case C-376/98 Germany v Parliament and Council [2000] ECR I -8149 ............. 224 Case C-473/98 Kemikalienspektionen v Toolex Alpha AB [2000] ECR I-5681 ............................................................................................ 215–6, 222 Case C-6/99 Association Greenpeace France v Ministere de l’Agriculture et de la Peche [2000] ECR I-1651 ................................................................... 212, 215, 220 Case C-3/00 Denmark v Commission [2003] ECR I-2643 .................... 212, 231, 236 Case C-9/00 Palin Granit OY and Vehmassalon Kansanterveystyön Kuntayhtymän Hallitus v Lounais-Suomen Ympäristökeskus [2002] ECR I-3533 .................... 212 Case C-24/00 Commission v France [2004] ECR I-1277 .......................... 231, 237–8 Case C-378/00 Commission v Council [2003] ECR I-937 .................................... 231 Case C-192/01 Commission v Denmark [2003] ECR I-9693 ........ 230–1, 237–8, 240 Case C-236/01 Monsanto Agricoltura Italia SpA v Presidenza del Consiglio dei Ministri [2003] ECR I-8105 ..................................................... 212, 230–1, 236 Case C-41/02 Commission v Netherlands [2004] ECR I-11375 ...................................................................... 213, 230–1, 237–9, 241 Case C-95/01 Criminal Proceedings Against Greenham and Abel [2004] ECR 1-1333......................................................................... 213, 215, 230–1, 237–8 Case C-127/02 Landelijke Vereniging tot Behoud van de Waddenzee v Staatssecretaris van Landbouw, Natuurbeheer en Visserij [2004] ECR I-9405 ........................................................................................ 212, 230, 235 Case C-286/02 Bellio F.lli Srl v Prefettura di Treviso [2004] ECR I-3465 ............. 230 Case C-434/02 Arnold André GmbH & Co KG v Landrat SWA Kreises Herford [2004] ECR I-4023 ........................................................... 215, 230–1, 234 Case C-98/03 Commission v Germany [2006] ECR I-53 ...................................... 236 Case C-132/03 Minstero della Salute v CODACONS [2005] ECR I-4167 .... 231, 236 Case C-244/03 France v Parliament [2005] ECR I-4021 ...................................... 231 Case C-453/03 ABNA Ltd v Secretary of State for Health [2005] ECR I-10423 .................................................................................. 231, 234–5, 239 Case C-6/04 Commission v United Kingdom [2005] ECR I-9017 ........................ 236 Case C-154/04 R (on the application of: Alliance for Natural Health) v Secretary of State for Health [2005] ECR I-6451 ...................................................... 215, 239

(Aa) Fisher TABLES

4/7/07

11:46

Page xxiii

Table of Cases

xxiii

Court of First Instance Case T-105/96 Pharos v Commission [1998] ECR II-285 ..................................... 219 Case T-125 & 152/96 Boehringer v Council and Commission [1999] ECR II-3427 ................................................................................................ 218–19 Case T-199/96 Laboratories Pharmaceutiques Bergaderm SA v Commission [1998] ECR II-2805 .................................................................................... 219–20 Case T-13/99 Pfizer Animal Health SA v Council [2002] ECR II-3305 .................................................. 16, 157, 203, 215–7, 231–2, 233, 236 Case T-70/99 Alpharma v Council [2002] ECR II-3495 ............................... 232, 236 Case T-74/00 Artegodan GmbH v Commission [2002] ECR II-4965 ........................................................................................ 209, 231, 234, 240 Case T-392/02 Solvay Pharmaceuticals BV v Council [2003] ECR II-4555 .............................................................................................................. 234 Case T-366/03 Land Oberosterreich v Commission [2005] ECR II-4005 ...... 212, 236

WTO Cases Appellate Body Reports Appellate Body Report, European Communities - Measures Concerning Meat and Meat Products (Hormones), WT/DS26/AB, 16 January 1998 ............................................ 41, 170, 174, 176, 178, 183–8, 194, 201, 213, 225 Appellate Body Report, Australia - Measures Affecting Importation of Salmon, WT/DS18/AB/R, 20 October 1998 .................. 178, 184, 192–3, 205, 213 Appellate Body Report, Japan - Measures Affecting Agricultural Products, WT/DS76/AB/R, 22 February 1999 ...................................... 174–5, 178, 184, 190, 196–7, 198, 201, 202 Appellate Body Report, United States - Definitive Safeguard Measures on Import of Wheat Gluten from the European Communities, WT/DS166/ AB/R, 19 January 2001 ...................................................................................... 170 Appellate Body Report, Japan - Measures Affecting the Importation of Apples, WT/DS245/AB/R, 26 November 2003 .............................. 184, 190, 197, 198, 201 Panel Reports Panel Report, Australia - Measures Affecting Importation of Salmon, WT/DS18/R, 12 June 1998 ................................. 190, 192, 194, 195, 197, 199, 205 Panel Report, European Communities Measures Concerning Meat and Meat Products (Hormones), WT/DS26/R/USA, 18 August 1997 ................ 16, 174, 183, 185–89, 192–4, 195, 199–201, 203, 213–4, 225–6 Panel Report, Japan - Measures Affecting Agricultural Products, WT/DS76/R, 27 October 1998 ..................................................................... 184, 190, 193–5, 202 Panel Report, Japan - Measures Affecting the Importation of Apples, WT/DS245/R, 15 July 2003 ................................................... 184, 190, 196–8, 201

(Aa) Fisher TABLES

xxiv

4/7/07

11:46

Page xxiv

Table of Cases

Panel Report, European Communities - Measures Affecting the Approval and Marketing of Biotech Products, WT/DS291-3/R, 29 September 2006+ .............................................................. 175, 198, 200–2, 213

(Aa) Fisher TABLES

4/7/07

11:46

Page xxv

TABLE OF LEGISLATION Australia Commonwealth Administrative Appeals Tribunal Act 1975 (Cth) ............................................ 134–6 Administrative Appeals Tribunal Act 1989 (Australian Capital Territory) ..... 134–5 Administrative Decisions (Judicial Review) Act 1977 (Cth) ............................... 134 Environment Protection (Impact of Proposals) Act 1974 (Cth) (now repealed) .................................................................................................. 129 Environmental Protection and Biodiversity Act 1999 (Cth) ............................... 131 Fisheries Management Act 1994 (Cth) ..................................................... 130–1, 145 Great Barrier Reef Marine Park Act 1975 (Cth) ................................................... 131 Ombudsman Act 1976 (Cth) ................................................................................ 134 Gene Technology Act 2000 (Cth) ......................................................................... 128 Australian Capital Territory Administrative Appeals Tribunal Act 1989 (ACT) .............................................. 134 Planning and Land Act 2002 (ACT) ..................................................................... 131 Waste Minimisation Act 2001 (ACT) ................................................................... 131 New South Wales Administrative Decisions Tribunal Act 1997 (NSW) .......................................... 134 Contaminated Land Management Act 1997 (NSW) ........................................... 131 Environmental Planning and Assessment Act 1979 (NSW) ........................ 129, 158 Land and Environment Court Act 1979 (NSW) ...................................... 134–6, 138 National Parks and Wildlife Act 1974 (NSW) ...................................................... 140 Protection of the Environment Administration Act 1991 (NSW) ...................... 130 Threatened Species Conservation Act 1995 (NSW) .................................... 128, 130 Queensland Integrated Planning Act 1997 (Qld) ............................................. 130–1, 134–6, 149 Local Government (Planning and Environment) Act 1990 (Qld) ...................... 134 South Australia Development Act 1993 (SA) ................................................................................. 148 Environment, Resources and Development Court Act 1993 (SA) .......... 134–6, 148

(Aa) Fisher TABLES

4/7/07

11:46

xxvi

Page xxvi

Table of Legislation

Tasmania Resource Management and Planning Appeal Tribunal Act 1993 (Tas) ........... 135–6 Victoria Environmental Protection Act 1970 (Vic) ........................................................... 131 Victorian Civil and Administrative Tribunal Act 1998 (Vic) .............................. 134 Western Australia Environment Protection Act 1993 (WA) ............................................................. 130 Environmental Protection Act 1986 (WA) .......................................................... 131 State Administrative Tribunal Act 2004 (WA) ..................................................... 134 Town Planning and Development Act 1928 (WA) .............................................. 135

United Kingdom Primary Legislation Agriculture Act 1984 ............................................................................................... 72 Alkali Act 1863 ........................................................................................................ 67 Alkali Act 1874 ........................................................................................................ 67 Animal Health Act 1981 .................................................................................... 72, 86 Environment Act 1995 ............................................................................................ 53 Food Act 1984 .................................................................................................... 72, 86 Food and Drugs Act 1860 ....................................................................................... 66 Food Standard Act 1999 .................................................................................... 53, 86 Pharmacy Act 1862 ................................................................................................. 66 Rivers Pollution Prevention Act 1876 .................................................................... 66 Sale of Food and Drugs Act 1875 ............................................................................ 66 Slaughterhouse Act 1974 ........................................................................................ 72 Statutory Instruments Act 1946 ............................................................................. 73 Secondary Legislation Bovine Spongiform Encephalopathy Order 1988 SI 1988/1039 ........................... 79 Bovine Spongiform Encephalopathy (Amendment) Order 1988 SI 1988/1345 ....................................................................................................... 79 Bovine Spongiform Encephalopathy Compensation Order 1988 SI 1988/1346 ....................................................................................................... 79 Bovine Spongiform Encephalopathy (No 2) Order SI 1988/2299 ........................ 82 Bovine Offal (Prohibition) Regulations 1989 SI 1989/2061 .................................. 84

United States Administrative Procedure Act 1946 .................................................................. 53, 95

(Aa) Fisher TABLES

4/7/07

11:46

Page xxvii

Table of Legislation

xxvii

Clean Air Act ....................................................................................................... 98–9 Consumer Product Safety Act ................................................................................ 98 Data Quality Act ................................................................................................... 116 Federal Insecticide, Fungicide, and Rodentcide Act .............................................. 98 Federal Water Pollution Control Act .................................................................. 98–9 Food, Drug and Cosmetic Act ................................................................................ 98 Interstate Commerce Commission Act .................................................................. 93 Occupational Safety and Health Act ............................... 57, 89–90, 98–100, 107–15 Resource, Conservation and Recovery Act ............................................................. 98 Safe Drinking Water Act .................................................................................. 98, 100 Toxic Substances Control Act ......................................................................... 98–100

European Union Law Decisions Council Decision 468/99/EC of 28 June 1999 laying down the procedures for the exercise of implementing powers conferred on the Commission [1999] OJ L184/23 .............................................................................................. 21 Directives Directive 67/548 on the classification, labelling and packaging of dangerous substances [1967] OJ 196/1 .............................................................................. 210 Directive 70/524/EEC concerning additives in feeding stuffs [1998] OJ L351/4 ............................................................................................... 215–6, 232 Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora OJ [1992] L 206/7 .................................. 235 Directive 2002/46 on the approximation of the laws of the Member States relating to food supplements [2002] OJ L183/51 ............................................ 238 Regulations Council Regulation No 2821/98 amending, as regards withdrawal of the authorisation of certain antibiotics, Directive 70/224/EEC concerning additives in feeding stuffs [1998] OJ L351/4 .................................................... 232 Regulation 178/2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety [2002] OJ L31/1. .................... 210, 212, 229 Regulation No 2371/2002 on the conservation and sustainable exploitation of fisheries resources under the Common Fisheries Policy [2002] OJ L358/59 ........................................................................................................ 212 International Law Biodiversity Convention ................................................................................. 40, 147 Cartagena Protocol on Biosafety ............................................................................ 40

(Aa) Fisher TABLES

xxviii

4/7/07

11:46

Page xxviii

Table of Legislation

General Agreement on Tariffs and Trade ...................................... 167–9, 175–6, 179 Rio Declaration on Environment and Development ............................................ 40 Vienna Convention on the Law of Treaties ........................................... 169, 185, 189 World Trade Organisation Dispute Settlement Understanding .......161, 183–4, 192 World Trade Organisation Sanitary and Phytosanitary Agreement ......................................................... 5, 18, 173–206, 213, 246, 248, 251 World Trade Organisation Technical Barriers To Trade Agreement ........... 175, 198

(B) Fisher Ch1

4/7/07

16:58

Page 3

1 Risk Evaluation through the Lens of Administrative Constitutionalism This book is a study of legal debates in a number of different jurisdictions concerning how administrative decision-makers should appraise environmental and public health risks and/or set standards in relation to such risks. My aim in this book is a radical one—to recast the way in which such debates are characterised. Typically such conflicts have been understood as requiring a state, loosely defined, to make a choice between primarily basing its decision on science or on the basis of values. As such, technological risk decision-making is understood as either predominantly a matter for experts or for democracy. In either case, the role of law is not a significant one. Rather it is characterised as crudely instrumental, irrelevant, or obstructive. My argument is that this is an incorrect depiction of risk regulation and the legal disputes it gives rise to. Legal disputes over standard setting and risk appraisal are essentially disputes about what is, and should be, the role and nature of ‘good’ public administration in risk regulation. That role and nature is influenced by closely related understandings about the nature of technological risks, science, democracy, and the potential of law to constitute public administration, limit it, and hold it to account. As this is the case these disputes are essentially disputes over administrative constitutionalism and a study of these disputes highlights the symbiotic relationship between law, public administration, and technological risk. Moreover, this relationship can be seen across a range of contexts including administrative law and trade law and in numerous different national, supranational, and transnational jurisdictions. In putting forward this argument I appreciate that focusing on public administration and administrative law is unfashionable. In recent years both have been incorporated into the far more magnetic concept of governance, and the legitimacy of public administration is decidedly questionable. But that is exactly my point—it is the questionable legitimacy of public administration and the legal disputes that this gives rise to which are fundamental but ignored aspects of risk regulation. By viewing standard setting and risk appraisal through the lens of administrative constitutionalism the nature and challenges of this area of regulation can be understood more precisely. In particular, rather than understanding standard setting and risk appraisal as requiring a choice, or compromise, between science and democracy it ought to be understood as requiring a choice between different paradigms of

(B) Fisher Ch1

4/7/07

4

16:58

Page 4

Risk Evaluation through the Lens of Administrative Constitutionalism

administrative constitutionalism—what I call the deliberative-constitutive and rational-instrumental paradigms—which are an amalgamation of particular and very different understandings of law, public administration, and technological risk. These paradigms, like all paradigms, are both prescriptive and descriptive—they dictate what ought to be as well as purporting to be descriptions about what is. By identifying these paradigms a better understanding can be gained of technological risk decision-making and in particular its legal dimension. While the primary aim of this book is to recast how scholars and policy-makers think about debates over standard setting I am also concerned to highlight two points which are made obvious by such a re-characterisation. First, nearly everyone looking at these paradigms will find one more normatively attractive than the other, but in any legal dispute or any risk regulation regime both paradigms will nearly always coexist. As such, administrative decision-makers are invariably operating in contexts where they are subject to different legal and regulatory views about what their role and nature is, and should be. This, I argue, is an inescapable and probably desirable state of affairs.1 Second, in an era of globalisation there is a particular need to appreciate the central role that administrative constitutionalism plays in standard setting/risk appraisal. Globalisation brings with it the increased overlap of legal cultures which creates new arenas for debating administrative constitutionalism, introduces different issues into those debates, and results in administrative decision-makers being regulated by ideas of administrative constitutionalism generated from different legal cultures. The present tendency of characterising risk regulation in terms of a science/democracy dichotomy is dangerous in that it not only results in legal actors, scholars, and policy-makers understanding globalisation as requiring a choice to be made between science and democracy but also leads to an ignoring of the hard questions concerning how ideas of administrative constitutionalism generated in different legal cultures can and should interact. I should stress at the outset that recasting this area of risk regulation in terms of administrative constitutionalism does not immediately result in any solutions to the difficult problems of technological risk regulation. Indeed, in some cases it makes the intractability of many of those problems even more obvious. Administrative decision-makers are invariably acting in circumstances where competing paradigms of administrative constitutionalism are in operation and globalisation brings with it some difficult questions concerning legal culture. What I hope, however, is that by altering the lens through which scholars and policy-makers view this area of the law this book will facilitate more constructive debate on these matters. This bulk of this book is a series of studies of legal disputes over standard setting in various jurisdictions. Chapter two examines how what was understood to be the role and nature of the Southwood Working Party that gave advice to government departments in relation to bovine spongiform encephalopathy (BSE) was shaped 1

Chapter 7, section II.

(B) Fisher Ch1

4/7/07

16:58

Page 5

Risk Evaluation through the Lens of Administrative Constitutionalism

5

by understandings of administrative constitutionalism. Chapter three studies two interrelated debates concerning the scope of judicial review of risk regulation rulemaking in the United States, and chapter four is an analysis of how the precautionary principle has been interpreted by Australian courts and tribunals. Chapter five is a study of dispute settlement proceedings concerning the World Trade Organisation (WTO) Sanitary and Phytosanitary Agreement (SPS Agreement), and chapter six considers how the European Court of Justice and Court of First Instance have reviewed the legal validity of Community and Member State technological risk evaluation in relation to precautionary reasoning and the application of the precautionary principle. My aim is not to provide an exhaustive comparative account of risk regulation but rather to present a series of ‘snapshots’ which illustrate how disputes over standard setting and risk appraisal in risk regulation are disputes over administrative constitutionalism. The examples are drawn from a range of legal cultures and the legal disputes examined vary in nature. Some are ‘legal’ in that they are disputes over what should be the nature of legislation or a particular legal doctrine. Others, however, are concerned with the ways in which decision-makers are constituted and held to account by means which are less conventionally legal in nature. Likewise, some of these disputes concern specific controversies while other debates are ongoing and concern how a particular principle should be interpreted or what role judges should play in reviewing decisions. But, before turning to those case studies, I need to put forward my main argument—that legal disputes over technological risk evaluation are in essence disputes over administrative constitutionalism. That I do in this chapter. The first section is an overview of technological risk evaluation and how it has been characterised as involving a choice between a democratic or scientific approach to risk evaluation. This characterisation is problematic, particularly in the role it ascribes to law. In the second section I put forward a different approach to conceptualising technological risk evaluation. The starting point for this approach is the institutional context for the vast majority of technological risk evaluation—public administration. While the role of public administration is necessary it is also contentious and this has resulted in technological risk evaluation giving rise to disputes over administrative constitutionalism. In the third section I describe and compare the two paradigms of administrative constitutionalism that I see debates as having polarised around—the rational instrumental and the deliberative constitutive. In the fourth section I discuss how administrative constitutionalism is a form of thick legal culture. In the final section I give a brief example of thinking about risk evaluation in terms of administrative constitutionalism. I do this by examining the precautionary principle and show not only how it can be interpreted in both rational-instrumental and deliberative-constitutive terms but also how those definitions have been complicated by the legal concept of ‘burden of proof’. I should make two clarifications before starting in relation to the book’s audience. First, this book is primarily addressed to two distinct groups operating in two very different disciplinary realms. The first group to whom the book is addressed

(B) Fisher Ch1

4/7/07

6

16:58

Page 6

Risk Evaluation through the Lens of Administrative Constitutionalism

comprises those legal scholars and legal practitioners working in this regulatory area. For that group, the message of this book is that they need to question and deconstruct the concepts of science, public administration and risk that they are deploying in their scholarship and practice.2 To put it simply—technological risk regulation is far more complex and contingent than most of legal scholarship presently recognises. Yet at the same time this book also has an audience of those working in the technological risk field more generally, whether in the policymaking or scholarship context. Most of that group already has a sound appreciation of the epistemological and normative challenges involved in technological risk regulation but what they do not have is a subtle appreciation of the normative malleability of law. This book is a demand for them to take a more critical and nuanced approach to the legal aspects of their subject. I recognise that the starting points for each of these groups is very different and in places this book maybe more accessible to one group than the other. Likewise, each group may be disappointed because of my failure to engage in detail with specific academic debates. It should be remembered, however, that this book is a first step in reorientation and not an attempt to be the last word on these issues. In so being, I hope it will go a long way to contributing to a dialogue between these two different groups. The second clarification is that this book is concerned only with the regulation of technological risks that are threats to the environment and/or human health. As such, this book does not address the many other arenas in which ‘risk’ is now being regulated, such as in the fields of criminal justice, counter-terrorism, and financial regulation. Nor is this book a contribution to the meta discourses about risk governance which are starting to emerge. These other issues are interesting and important, and I have addressed them briefly elsewhere,3 but the argument in this book is solely about technological risk evaluation and not about these other issues. With that said, as discussed in the conclusion, there is an argument that the general lens of administrative constitutionalism may be applicable in other areas.

I The Science/Democracy Dichotomy in Regulating Technological Risk In the last two decades the collective public regulation of environmental and public health risks that arise from human industrial activities has been a constant source of socio-political and legal conflict both within and across jurisdictions. 2 I made a similar point in E Fisher, ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds), Environmental Law for Sustainability (Oxford, Hart Publishing, 2006) at 116–25. 3 E Fisher, ‘The Rise of the Risk Commonwealth and the Challenge for Administrative Law’ (2003) Public Law 455. Nor is this book concerned with the management of technological risk outside the regulatory context.

(B) Fisher Ch1

4/7/07

16:58

Page 7

The Science/Democracy Dichotomy in Regulating Technological Risk

7

These risks I call technological risks. Some of the debates over these risks have concerned whether the state should regulate but many of these conflicts have arisen in circumstances where the powers to regulate exist and the issue is how the state should regulate. This book focuses on these latter debates and in these cases the conflict is normally concerning risk evaluation. That is the process by which ex ante standards are set and individual activities or products are appraised for their potential harm to the environment and public health, the latter amounting to a form of risk appraisal. Among other things, risk evaluation is a process of deciding whether a risk is acceptable or not—focusing as it does on whether an activity or product is ‘unsafe’ and thus requiring regulatory action to be taken in relation to it.4 As is clear from the concept of acceptability such decisions are not merely factual but also require value decisions to be made about what is, and is not, a desirable state of affairs.

A Technological Risks The reason why risk evaluation has been the focus of risk controversies has much to do with three particular features of technological risks. First, such risks are scientifically uncertain. In some cases, particularly when dealing with new technologies and activities, scientific uncertainty is due to the fact that risks relate to future outcomes of action which is inherently unpredictable. In most cases, however, scientific uncertainty takes a more complex form. This is because collective knowledge about the risks is poor and such risks are occurring in open-ended systems in which the vagaries of both the natural environment and human behaviour must be taken into account. In relation to many technological risks there are no firm probabilities, and a better understanding of a threat cannot often be gained through simply doing more research. This is because the existence of scientific uncertainty reflects a whole series of methodological, epistemological and even ontological problems inherent in determining whether a threat exists and, if it does, what its nature is.5 There are simple experimental and sampling uncertainties in the collection of data, particularly in the ecological context. There are technological limitations placed on the process of monitoring. The assessment of risk invariably relies on modelling tools and it is difficult to assess whether a model is a constructive simplification or a misunderstanding of the reality it is attempting to represent. Likewise, there are methodological problems in assessing risks, particularly in relation to long-term health risks such as cancer. Thus, for example, there is disagreement over whether animal studies or epidemiology is the most 4 M Douglas, Risk Acceptability According to the Social Sciences (London, Routledge & Kegan Paul, 1985); and S Rayner and R Cantor, ‘How Fair is Safe Enough?: The Cultural Approach to Societal Technology Choice’ (1987) 7 Risk Analysis 39. 5 B Wynne, ‘Uncertainty and Environmental Learning’ (1992) 2 Global Environmental Change 111; M Smithson, ‘Ignorance And Science’ (1993) Knowledge: Creation, Diffusion and Utilization 133; and E Fisher, ‘Drowning by Numbers: Standard Setting in Risk Regulation and the Pursuit of Accountable Public Administration’ (2000) 20 Oxford Journal of Legal Studies 109 at 115–16.

(B) Fisher Ch1

4/7/07

8

16:58

Page 8

Risk Evaluation through the Lens of Administrative Constitutionalism

reliable discipline on which to assess the risks of cancer in humans.6 There are also epistemological uncertainties that are the types of uncertainties which arise because ‘we don’t know what we don’t know’.7 This is particularly the case in relation to something such as BSE where its nature and cause is only just beginning to be understood.8 Scientific uncertainty is also caused by the indeterminacy9 of the natural environment, it being a holistic system made up of complex and littleunderstood interrelationships. Causal links are not always linear and activities may not cause adverse effects until far into the future. Thus, for example, air pollution can cause acid rain which leads to a series of knock-on effects for the environment and public health.10 Likewise, the adverse environmental and health effects of nuclear waste may only become serious decades later.11 In part, however, such uncertainty is due to the second feature of technological risks, which is that they are behaviourally uncertain due to the fact that the nature and existence of a risk will often depend on human behaviour. Thus, for example, a cause of any particular adverse effect may be a combination of mutually interacting variables, such as was the case in relation to Three Mile Island and Chernobyl, where accidents occurred due to numerous inter-linked factors including management, operational error, and the design of the technology.12 Assessing the exposure to a risk can also be difficult due to the vagaries of human behaviour. Likewise, the polycentricity of technological risk problems may create uncertainty.13 In all cases the problem is that the social phenomena are not easily subject to predictive analysis and as Rayner and Malone note, we ‘have inaccurate and conflicting theories about how and why people make choices, for themselves and in societies’,14 something which is highlighted by the fact that many of the risks are the unintended and unexpected consequence of human activity.15 The third feature of technological risk problems is that whether a risk is acceptable or not is dependent upon cultural context. The polycentric nature of 6 National Research Council, Science and Judgement in Risk Assessment (Washington DC, National Academy Press, 1994) at 58–60. 7 S Jasanoff and B Wynne, ‘Science and Decision Making’ in S Rayner and E Malone (eds), Human Choice and Climate Change—Volume One (Pacific Northwest National Laboratory, Battelle Press, 1998). 8 See ch 2, this volume, and the Report of the Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Two: Science (London, HMSO, 2000). 9 Wynne, above n 5. 10 S Boehmer-Christiansen and J Skea, Acid Politics: Environmental and Energy Policies in Britain and Germany (London, The Belhaven Press, 1991). 11 K Shrader-Frechette, Burying Uncertainty: Risk and the Case Against the Geological Disposal of Nuclear Waste (Berkeley, University of California Press, 1993). 12 President’s Commission on the Accident at Three Mile Island, Final Report (Washington DC, Government Printing Office, 1980). 13 L Fuller, ‘The Forms and Limits of Adjudication’ (1978) 92 Harvard Law Review 353 at 395–7. 14 S Rayner and E Malone (eds), Human Choice and Climate Change—Volume Four (Pacific Northwest National Laboratory, Battelle Press, 1998) at 120. 15 K Erikson, A New Species of Trouble: The Human Experience of Modern Disasters (New York, WW Norton & Co, 1994); and U Beck, Risk Society: Towards A New Modernity (London, Sage Publications, 1992).

(B) Fisher Ch1

4/7/07

16:58

Page 9

The Science/Democracy Dichotomy in Regulating Technological Risk

9

technological risks highlights the ‘social vulnerability’16 of individuals in a society, and, as such, what makes a risk acceptable to an individual is not just a question of personal preference, magnitude and likelihood (the last two which might be impossible to assess) but also is dependent upon the acceptability of the context to the individual. A simple example of this is where an individual may bear the risk of an activity where they do not receive the benefit of it. Individual risk acceptability may depend upon matters such as institutional trust, control, risk distribution, and how voluntary any risk-taking activity is, as well as the acceptability of broader social arrangements.17 To put it crudely, and as one anonymous commentator has quipped, ‘never say a risk is negligible unless the listener shares your philosophy of life’. What this means is that issues of risk acceptability are deeply interwoven with broader societal arrangements, and technological risk disputes raise questions concerning power and social control.18 As Rayner and Cantor have noted, the question is not ‘how safe is safe enough’ but ‘how fair is safe enough’.19 These different features of technological risks have led to them becoming a matter for state action. Individuals are ‘trapped in a web of contingencies and uncertainties’20 which renders them unable to make effective risk management decisions or, at least, creates the perception that that is the case. Likewise, because the acceptability of risk raises issues about the acceptability of the way in which societies are organised and governed it produces a politics of the public sphere.21 As a matter of state action, however, the regulation of technological risks is not straightforward. In many cases the ‘facts’, which are often stressed as being important for the operation of public reason,22 are either not available or open to varying interpretations. This leads to a situation where different actors may have

16 W Freudenburg, ‘Risk and Recreancy: Weber, the Division of Labour, and the Rationality of Risk Perceptions’ (1993) 71 Social Forces 909 at 915. 17 For different analyses of this issue see P Slovic, The Perception of Risk (London, Earthscan, 2000); J Bradbury et al, ‘Trust and Public Participation in Risk Policy Issues’ in G Cvetkovich and R Lofstedt (eds), Social Trust and the Management of Risk (London, Earthscan, 1999); Douglas, above n 4; and C Jaeger et al, Risk, Uncertainty, and Rational Action (London, Earthscan, 2001). 18 For different views on this point see W Freudenburg and S Pastor, ‘Public Responses to Technological Risks: Towards a Sociological Perspective’ (1992) 33 Sociological Quarterly 389 at 391; G Cvetkovich and R Lofstedt (eds), Social Trust and the Management of Risk (London, Earthscan, 1999); L Clarke and J Short, ‘Social Organization and Risk: Some Current Controversies’ (1993) 19 Annual Review of Sociology 375; R Kuehn, ‘The Environmental Justice Implications of Quantitative Risk Assessment’ (1996) University of Illinois Law Review 103; and M Douglas and A Wildasky, Risk and Culture (Berkeley, University of California Press, 1982). 19 Rayner and Cantor, above n 4. 20 Jaeger et al, above n 17 at 250. 21 A range of different and not always reconcilable reasons are given for the state bearing the ultimate responsibility for risk evaluation. See U Beck, ‘Risk Society and the Provident State’ in S Lash and B Szerszynski (eds), Risk, Environment and Modernity: Towards a New Ecology (London, Sage Publications, 1996) (risk as politics); D Moss, When All Else Fails: Government as the Ultimate Risk Manager (Cambridge, Harvard University Press, 2002) (the state as efficient risk manager); and C Hilson, Regulating Pollution: A UK and EC Perspective (Oxford, Hart Publishing, 2000) (market failure and ethical reasons). 22 C Sunstein, Free Markets and Social Justice (New York, Oxford University Press, 1997).

(B) Fisher Ch1

4/7/07

10

16:58

Page 10

Risk Evaluation through the Lens of Administrative Constitutionalism

differing views over whether a risk exists or not.23 Furthermore, as the US National Research Council (NRC) has noted, regulating environmental risk is not so much about ‘estimating and reducing risks but about developing a shared vision of desired conditions of an ecosystem’.24 Most significantly, however, if technological risk issues need to be collectively managed, and such risks raise issues concerning the acceptability of governance arrangements then how such technological risks are managed will affect whether particular actors will view such risks as acceptable. This is because standard setting and risk appraisal will ‘frame’25 the issue being regulated but in so doing will tend to favour one understanding of risk acceptability over another. This is particularly the case in circumstances of high scientific uncertainty where there will be differing and often incommensurable views over the nature of the risk itself and, as such, risk identification and risk acceptability collapse into each other. Thus, for example, for some commentators genetically modified organisms (GMOs) are simply a natural extension of crossbreeding techniques while for others they represent a novel technology and need to be assessed as such.26 In other words, a state cannot provide a ‘commonly accepted or legitimate meta-rationality’ in regards to its description of a risk and thus it must privilege one perspective.27 All of this leads to the situation that the process of evaluating technological risks will shape, and be shaped by, individual risk acceptability, and for many actors the process of regulating a risk is part of the technological risk problem.28 The challenges that this creates for decision-makers may lie on a continuum. At one end, it may mean that technological risk evaluation recognises, but marginalises, particular understandings of risk acceptability. At the other end, however, it may mean that the way in which a public decision-maker frames a technological risk problem fails to recognise that the problem has other dimensions that need to be addressed. In all cases, however, the problem is that there is an ‘institutional neglect of issues of public meaning’ as well as the ‘presumptive imposition’ of other understandings of risk problems onto the public.29 Such neglect and imposition often leads to a growing mistrust of the institutions that are engaged in risk 23 ESRC Global Environmental Change Programme, The Politics of GM Food: Risk, Science and Public Trust (Special Briefing No 5) (1999). 24 National Research Council, Understanding Risk: Informing Decisions in a Democratic Society (Washington DC, National Academy Press, 1996) at 18. 25 S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005) at 23–6; and S Jasanoff, ‘Technological Risk and Cultures of Rationality’ in National Research Council (ed), Incorporating Science, Economics and Sociology in Developing Sanitary and Phytosanitary Standards in International Trade: Proceedings of a Conference (Washington DC, National Academies Press, 2000). 26 R Grove White et al, Uncertain World: Genetically Modified Organisms, Food and Public Attitudes in Britain (Lancaster, Lancaster University, 1997). 27 Jaeger et al, above n 17 at 251. 28 Douglas and Wildasky, above n 18; and B Williams and A Matheny, Democracy, Dialogue and Environmental Disputes: The Contested Languages of Social Regulation (New Haven, Yale University Press, 1995). 29 B Wynne, ‘Seasick on the Third Wave? Subverting the Hegemony of Propositionalism: Response to Collins & Evans (2002)’ (2003) 33 Social Studies of Science 401 at 402.

(B) Fisher Ch1

4/7/07

16:58

Page 11

The Science/Democracy Dichotomy in Regulating Technological Risk

11

evaluation30 as well as an increasing scepticism that such institutions are legitimate.31 As such, the challenging of technological risk evaluation decisions is often a challenge to the validity of the institutions. Likewise, arguments for reform in this area are invariably attempts to overhaul how technological risk issues are governed. For this reason, a number of scholars have argued that technological risks have become the focal point for arguing for more general reforms in how the state governs in a liberal democracy as well as one of the catalysts for the rise of the new social movements.32

B The Science/Democracy Dichotomy The discussion so far has merely sketched a range of issues which have been the subject of intense scholarly debate. What such a sketch highlights is that technological risk disputes are not just discrete technical or ethical disagreements. They concern how collective institutions should identify, understand and take action in relation to such risks. Indeed, questions about ‘good’ risk evaluation, risk acceptability, and legitimate institutions collapse into each other. In such circumstances it comes as no surprise that technological risk evaluation has been so controversial and commentators have been primarily concerned with reforming technological risk decision-making to ensure ‘better’ risk evaluation—if an important aspect of a technological risk controversy is how risk is characterised then a necessary solution will be overhauling those characterisation processes. Indeed, for the last three decades it has been risk evaluation, not risk itself, which has been the focus of much of the literature.33 By and large, the institutional focus for overhauling risk evaluation processes is the state, invariably loosely defined. This is even the case in an era of governance and globalisation where such a formal institution has apparently been deconstructed and standard setting and risk evaluation have in some cases been delegated out of the public sector, and even out of the jurisdiction.34 The state and its conceptual baggage, however much maligned, are still the starting points for conceptualising this area of decision-making. In part this has to do with the fact that it is only sovereign states that have the innate power to regulate risk, and in 30 M Edelstein, ‘When the Honeymoon is Over: Environmental Stigma and Distrust in the Siting of a Hazardous Waste Disposal Facility in Niagara Falls, New York’ (1993) 5 Research Issues in Social Problems and Public Policy 75; and ESRC Global Environmental Change Programme, above n 23. 31 Douglas and Wildasky, above n 18. 32 U Beck, ‘The Reinvention of Politics: Towards a Theory of Reflexive Modernization’ in U Beck et al (eds), Reflexive Modernization: Politics, Tradition and Aesthetics in the Modern Social Order (Cambridge, Polity Press, 1994); and F Ewald, ‘Risk in Contemporary Society’ (2000) 6 Connecticut Insurance Law Journal 365. On new social movements see J Cohen and A Arato, Civil Society and Political Theory (Cambridge, MIT Press, 1992); and N Luhmann, Risk: A Sociological Theory (New York, Aldine de Gruyter, 1993) at ch 7. 33 Fisher, above n 2. 34 O Perez, Ecological Sensitivity and Global Legal Pluralism (Oxford, Hart Publishing, 2004); and H Schepel, The Constitution of Private Governance (Oxford, Hart Publishing, 2005).

(B) Fisher Ch1

4/7/07

12

16:58

Page 12

Risk Evaluation through the Lens of Administrative Constitutionalism

part it has to do with the fact that ‘the state’ as an abstract concept has such a powerful hold over academic imagination it is difficult to conceptualise collective decision-making without reference to it. In arguing for an overhaul of the state’s approach to technological risk evaluation, commentators tend to align themselves into two groups—those that argue science and expertise are the primary basis for making decisions about technological risk, and those that argue democracy and ethical values are the primary basis. I say primary, because both groups recognise a role for the other, but often only in marginal terms. For the first group, the solution to technological problems is to promote a scientific or expert approach to decision-making.35 This is because, while risk evaluation does involve value choices and is affected by scientific uncertainty, risk is a scientific concept that can be properly understood only in scientific terms. As this is the case, a state needs to ensure that decision-making is as objective as possible and decision-makers are experts in their field. Scientific uncertainty is recognised, but it is something that is manageable. Values are also relevant but only need to be integrated into decision-making after a proper assessment of the risk has been carried out. Science, for this group, needs to be brought into politics, into regulation, and into the courtroom so that technological risk can be properly understood. At the same time science needs to be kept separate from politics, policy, and law so as to ensure its integrity. In particular there is a danger that irrational fears and arbitrary values will contaminate proper scientific analysis. For another group, this is the wrong approach to evaluating risk. For this group, risk evaluation is inherently value laden.36 Science is a limited tool for making decisions due to: its epistemological limits, the fact that its ‘objectivity’ can act as a cloak for ideology; and because decision-making about risk is about how communities collectively wish to live their lives. For this group, the state needs to take a more democratic approach to the handling of risk which emphasises the importance of ethical values, liberal autonomy, and trust building for determining whether any particular risk is acceptable. The state needs to promote public participation in technological risk decision-making and to ensure an explicit discussion of the value conflicts raised by particular technological risks. The role of law is thus to largely enable greater participation, particularly by enabling greater access to regulatory processes and to the courts. The dichotomy in operation here is essentially a dichotomy between expertise and democracy and it has been characterised in other ways—as one between science and values, as one between science and politics, and/or as one between technocracy and participatory democracy. From a theoretical point of view these 35 Examples include National Research Council, Risk Assessment in the Federal Government: Managing the Process (Washington DC, National Academy Press, 1983); J Graham, ‘Making Sense of Risk’ (2000) 20 Risk Analysis 302; and C Sunstein, Risk and Reason: Safety, Law and the Environment (Cambridge, Cambridge University Press, 2002). 36 Examples include D Fiorino, ‘Environmental Risk and Democratic Process: A Critical Review’ (1989) 14 Columbia Law Review 501; and R Kennedy, ‘Risk, Democracy and the Environment’ (2000) 20 Risk Analysis 306.

(B) Fisher Ch1

4/7/07

16:58

Page 13

The Science/Democracy Dichotomy in Regulating Technological Risk

13

do not amount to the same distinction, a fact that hints at the distinction’s malleable nature. What is being understood in each case is that the evaluation of technological risk is either principally a matter of fact carried out in the realm of science and expertise or principally a matter of values carried out in the realm of democracy and socio-political debate. For the former group, the purpose of risk evaluation is to accurately identify technological risks, while for the latter it is to develop a sophisticated model of risk acceptability that reflects the sociopolitical complexities of a pluralistic society. It is not that each group does not recognise a role for the other but they do so only in subsidiary terms. To do otherwise, would be to threaten the legitimacy of the risk evaluation process. The fundamental division between these two groups is perhaps well summed up in a comment by Collins and Evans when they noted: Technical decision-making in the public domain is where the pigeons of much recent social science are coming home to roost. The problem can be stated quite simply: Should the political legitimacy of technical decisions in the public domain be maximized by referring them to the widest democratic processes, or should such decisions be based on the best expert advice?37

Admittedly, there have been proposals to base risk evaluation more squarely on both, particularly in recent policy documents.38 Thus, for example, the NRC talked of the need for an ‘analytic-deliberative process’ in which science and deliberation occur in a reiterative process.39 Likewise, much of the academic field of science and technology studies has in the last three decades been concerned with examining the co-production of science and culture,40 and Collins and Evans were putting the question above so as to challenge it and replace it with a normative theory of expertise.41 The reality is, however, that the distinction between science and democracy has remained not only as the starting point for analysis but it has also been assumed to be immutable—science and democracy have been presumed to be two incommensurable spheres of human activity.42

37 H Collins and R Evans, ‘The Third Wave of Science Studies: Studies of Expertise and Experience’ (2002) 32 Social Studies of Science 235 at 235–6. 38 Presidential/Congressional Commission on Risk Assessment and Risk Management, Framework for Environmental Health Risk Management—Volume One (Washington DC, 1997); and Royal Commission on Environmental Pollution, Setting Environmental Standards, 21st Report (London, HMSO, 1998). 39 National Research Council, Understanding Risk: Informing Decisions in a Democratic Society, above n 24. 40 S Jasanoff, Science at the Bar: Law, Science and Technology in America (Cambridge, Harvard University Press, 1995); B Latour, Politics of Nature: How To Bring the Sciences Into Democracy (Cambridge, Harvard University Press, 2004); and H Nowotny et al, Re-thinking Science: Knowledge and the Public in an Age of Uncertainty (Cambridge, Polity Press, 2001). 41 Collins and Evans, above n 37. 42 See Royal Society, Risk, Analysis, Perception and Management (London, Royal Society, 1992) where there was a failure to reach consensus on the definition of risk.

(B) Fisher Ch1

4/7/07

14

16:58

Page 14

Risk Evaluation through the Lens of Administrative Constitutionalism

C The Role of Law None of the above discussion accounts for the role of law. The important role of law is acknowledged in the social science and policy literature but is subject to little examination.43 In many cases, legal obligations are treated as ‘truths’ that cannot be questioned and are not open to interpretation.44 Likewise, there is often an embracement of legal terms and concepts with little thought of whether such terms are appropriate for the context. A good example here is that the precautionary principle, a principle concerned with scientific uncertainty in technological risk decision-making,45 is often described as requiring a ‘shifting the burden of proof’.46 The problem with this characterisation, however, is that a ‘burden of proof’ is an adjudicatory concept regulating the treatment of evidence, while the principle primarily regulates administrative discretion.47 This will be discussed in more detail below.48 The expertise/democracy dichotomy has been readily embraced by legal scholars, particularly in the trade law context,49 as a framework for critiquing regulatory regimes. It is, however, rarely accompanied by a detailed discussion of how that dichotomy relates to the law.50 Rather, law has tended to be treated in three different ways. First, in most circumstances it has been construed in instrumental terms as a tool for enhancing either the scientific or democratic aspects of risk evaluation. It does the former by requiring decision-makers to use science and scientific tools51 and it does the latter by creating opportunities for public participation.52 In both cases, however, law is not regarded as having any internal logic or injecting any fundamental issues into the debate. The law is treated as being 43 M Douglas, Risk and Blame: Essays in Cultural Theory (London, Routledge, 1992) at 24; and Luhmann, above n 32 at ch 3. 44 See the discussion of Commission of the European Communities, Communication from the Commission on the Precautionary Principle COM (2000) 1 final in ch 6, this volume, and Fisher, above n 2. 45 See chs 4 and 6. 46 J Morris (ed), Rethinking Risk and the Precautionary Principle (Oxford, Butterworth-Heinemann, 2000). See ch 4. 47 For a discussion of these problems see E Fisher, ‘Is the Precautionary Principle Justiciable?’ (2001) 13 Journal of Environmental Law 317; and J Jones and S Bronitt, ‘The Burden and Standard of Proof in Environmental Regulation: The Precautionary Principle in an Australian Administrative Context’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). 48 See section V.C below. 49 See ch 5. 50 D Williams, ‘Environmental Law and Democratic Legitimacy’ (1994) 4 Duke Environmental Law and Policy Forum 1; J Fraiberg and M Trebilcock, ‘Risk Regulation: Technocratic and Democratic Tools for Regulatory Reform’ (1998) 43 McGill Law Journal 835; Fiorino, above n 36; and C Sunstein, ‘Book Review: The Perception of Risk’ (2002) 115 Harvard Law Review 1119. 51 See the risk assessment reforms proposed by the Republicans in the mid- to late 1990s discussed in J Applegate, ‘A Beginning Not an End In Itself: The Role of Risk Assessment in Environmental Decision Making’ (1995) 63 University of Cincinnati Law Review 1643. 52 See the interest representation model in R Stewart, ‘The Reformation of American Administrative Law’ (1975) 88 Harvard Law Review 1661.

(B) Fisher Ch1

4/7/07

16:58

Page 15

The Science/Democracy Dichotomy in Regulating Technological Risk

15

content free: ‘law can’t think’, as one commentator has noted.53 As such, suggested reforms to the law in this area are often procedural in nature54 or relate to the creating of democratic or scientific institutions.55 In all cases, law is really just ‘carving out a quasi autonomous place’ for either science or democracy that preserves either democratic or scientific authority.56 Second, for some, law is seen as largely irrelevant because there has been an abandonment of traditional law for more fashionable concepts of regulatory governance. This development has taken many simultaneous forms including the abandonment of command and control regulation;57 the creation of reflexive law;58 and the negating of the public/private divide.59 In all cases, however, there appears to be not much of a role for law, which is seemingly replaced by other forms of decision-making, in particular negotiation,60 and/or other forms of accountability mechanisms such as audit.61 Third, when there is recognition that legal reasoning has a significant role to play, this only seems to result in recognising intellectual divisions between law and science or law and democracy.62 Thus for example, scholars exploring the science/law relationship note the differing burdens of proof 63 or the misuse of science by lawyers.64 As Mashaw has commented the ‘law-science literature is voluminous. It is also heterogeneous but it is seldom congratulatory or optimistic concerning the law-science interface’ which it characterises as ‘a vast intellectual

53 P Howard, The Death of Commonsense: How Law is Suffocating America (New York, Random House, 1994) at 186, as quoted in D Wirth and E Silbergeld, ‘Risky Reform’ (1995) 95 Columbia Law Review 1857 at 1889. 54 J Bohanes, ‘Risk Regulation in WTO Law: A Procedure-Based Approach to the Precautionary Principle’ (2002) 40 Columbia Journal of Transnational Law 323. 55 S Breyer, Breaking the Vicious Circle: Towards Effective Risk Regulation (Cambridge, Harvard University Press, 1993). 56 S Jasanoff, ‘Law’s Knowledge: Science for Justice in Legal Settings’ (2005) 95 American Journal of Public Health S49 at S52. 57 I Ayres and J Braithwaite, Responsive Regulation—Transcending the Deregulation Debate (New York, Oxford University Press, 1992). 58 See E Orts, ‘Reflexive Environmental Law’ (1995) 89 Northwestern University Law Review 1227; and J Holder, Environmental Assessment: The Regulation of Decision-Making (Oxford, Oxford University Press, 2005). 59 J Freeman, ‘The Private Role in Public Governance’ (2000) 75 New York University Law Review 543. 60 J Freeman, ‘Collaborative Governance in the Administrative State’ (1997) 45 UCLA Law Review 1. 61 M Power, The Audit Society: Rituals of Verification (Oxford, Oxford University Press, 1997). 62 This has been a point of particular interest in the exploration of the relationship between law and theories of the state. See P Cane, ‘Theory and Values in Public Law’ in P Craig and R Rawlings (eds), Law and Administration in Europe: Essays in Honour of Carol Harlow (Oxford, Oxford University Press, 2003). 63 C Cranor, Regulating Toxic Substances: A Philosophy of Science and the Law (Oxford University Press, 1993); and P Pascual, ‘Wresting Environmental Decisions from an Uncertain World’ (2005) 35 Environmental Law Reporter 10539. 64 W Wagner, ‘The Science Charade in Toxic Risk Regulation’ (1995) 95 Columbia Law Review 1613’; and T McGarity, ‘Our Science is Sound Science and Their Science is Junk Science: Science-Based Strategies For Avoiding Accountability and Responsibility for Risk-Producing Products and Activities’ (2004) 52 University of Kansas Law Review 897.

(B) Fisher Ch1

4/7/07

16

16:58

Page 16

Risk Evaluation through the Lens of Administrative Constitutionalism

divide’.65 Law in these circumstances is largely obstructive and preventing the proper operation of scientific or democratic processes.

D Problems with the Science/Democracy Dichotomy There are three main problems with understanding risk evaluation in terms of the science/democracy dichotomy and understanding the role of law in instrumental, obsolete, and/or obstructive terms. All are to do with the fact that this characterisation of technological risk evaluation misconstrues what actually occurs in practice. The first problem is that this characterisation of technological risk decisionmaking seems to suggest that the disputes over technological risk decision-making are fundamentally a conflict between facts on one side and values on the other. This, however, is not the case. Technological risk disputes invariably involve conflicts over what is the available science and how it should be interpreted. Likewise, as socio-political conflicts, they also involve disagreement over the relevant normative values that should operate and how such risks should be managed by governing institutions. This can be seen in legal disputes over risk evaluation. The different parties are often disputing what are the ‘facts’ and how those facts should be assessed.66 Likewise, each party has a different understanding of what issues are reasonable and rational to take into account in risk evaluation.67 It is not a case that one group is arguing for a decision to be based on knowledge and another group is arguing that it should not, but that there is a disagreement between both groups about what counts as knowledge. The second problem with this characterisation is that science and democracy are treated as different realms of public life but there is no explanation of how they interrelate and what is meant be these terms. Discussion seems to presume that science and democracy are fixed arenas of activity and that the decision to be made about technological risk is to allot decision-making to one realm or the other. This is because in alluding to ‘the state’ in grandiose and vague terms, there is little discussion of institutional context. Yet clearly the validity of a scientific or democratic approach will depend on that context and without knowing that context it is very difficult to determine what the problems with a certain approach to technological risk are. Thus for example, Collins and Evans note that while making decisions on a solely scientific basis carries with it a ‘problem of legitimacy’, recognising the need for democratic input carries with it the ‘problem of extension’ due to the difficulties in determining who should participate in decision-making.68 Yet the 65 J Mashaw, ‘Law and Engineering: In Search of the Law-Science Problem’ (2003) 66 Law and Contemporary Problems 135. 66 Flue-Cured Tobacco Co-op v EPA 4 F Supp 2d 435 (MD NC, 1998); and Panel Report, European Communities Measures Concerning Meat and Meat Products (Hormones), WT/DS26/R/USA, 18 August 1997. 67 Case T-13/99 Pfizer Animal Health SA v Council [2002] ECR II-3305; and Ethyl Corp v EPA 541 F 2d 1 (DC Cir 1976). 68 Collins and Evans, above n 37 at 236.

(B) Fisher Ch1

4/7/07

16:58

Page 17

The Science/Democracy Dichotomy in Regulating Technological Risk

17

magnitude and nature of each of these problems will vary with context. If a democratic legislature is making a decision then the problem of legitimacy is more serious than if the decision is being made by a public research laboratory. Likewise, the problem of extension is less of a problem in parliamentary politics than it would be in the courtroom. The seriousness of each depends on what is expected of the institutions making technological risk decisions and referring in loose terms to ‘the state’ gives very little elucidation on that context. Moreover, problems of lack of contextualisation are not assisted by the fact that much of the non-legal literature in this area largely characterises the law–science interface as occurring mainly in the courtroom in the context of rules of expert evidence. There seems to be little appreciation that the regulatory administrative context is a very different legal realm, shaped by its own separate logic and normative concerns. The third problem is that the present depiction of risk evaluation controversies tends to sideline the role of legal discourse in technological risk decision-making. Yet, as will be shown in this book, the legal dimension is a significant one and the body of law in this area is complex, dynamic and constantly shaping understandings about what is, and what is not, a legitimate way to make decisions about technological risks. This is best illustrated by examining how the role of law is not instrumental, irrelevant, or obstructive. It is not instrumental because although legal frameworks create the conditions for technological risk decision-making they also do more. Legal disputes are the sites for debating the legitimacy of particular approaches to technological risk decision-making and that debate is in legal terms. Moreover, that legal discourse is not just concerned with technological risk per se but with how technological risk decision-making relates to broader legal concepts and ideas. Likewise, such a discourse will shape technological risk decision-making. Thus, for example, as will be shown in these chapters, many decision-making techniques and policy statements are the direct product of legal debates.69 Likewise, law will authoritatively define what is understood by the terms ‘science’, ‘democracy’, ‘values’, and ‘expertise’. Law is not just a tool for ensuring more democracy or more science but is an influential discourse that actually constructs these concepts. This relates to the fact that law is also not irrelevant, because legal frameworks and discourse will always be creating the conditions for technological risk decision-making. This is even in the era of regulatory governance where law seems to have a marginal role. Law will be allowing the involvement of other regulatory actors and regulating the basis for their involvement.70 Much of this has to do with the fact that as a ‘state’ activity the exercise of power needs to be authorised, no matter which institution is involved.71 69

See chs 2 and 6 in particular. L McDonald, ‘The Rule of Law in the “New Regulatory State”’ (2004) 33 Common Law World Review 197; E Fisher, ‘Unpacking the Toolbox: Or Why the Public/Private Divide is Important in EC Environmental Law’ in M Freedland and J-B Auby (eds), The Public Law/Private Law Divide: Une entente assez cordiale? (Oxford, Hart Publishing, 2006); and M Egan, Constructing A European Market: Standards, Regulation and Governance (Oxford, Oxford University Press, 2001). 71 C Sampford, ‘Law, Institutions and the Public/ Private Divide’ (1991) 20 Federal Law Review 185. 70

(B) Fisher Ch1

4/7/07

18

16:58

Page 18

Risk Evaluation through the Lens of Administrative Constitutionalism

Finally, the notion that law is obstructive is only valid if law is understood in purely instrumentalist terms but, as noted above, law is creating the conditions for action. With that said, particular laws may be construed as obstructive by an actor because they do not accord with that actor’s understanding of technological risk decision-making. Indeed, as will be shown time and time again in this book, ‘law’ does not speak with a uniform voice. Public decision-makers are often caught in a legal web of competing rules and principles, and the law is in a constant state of flux—a point which once again points to the non-instrumentalist nature of law.

II Technological Risk, Public Administration, and Administrative Constitutionalism I appreciate that the discussion above may be seen by some as the construction of a straw man. I also recognise that there is much literature out there which takes a more sophisticated view than that described above.72 But, in presenting earlier versions of my arguments, I was constantly faced with the problem that audiences would reinterpret what I had to say in terms of the science/democracy dichotomy.73 What this highlights is that while the dichotomy is problematic and simplistic it has a powerful hold on how risk evaluation is conceptualised. Moreover, the same audiences tended to understand the challenges for legal scholars as taking the form of needing to ‘master’ public participation or science. There was little attention paid to how legal processes contributed to how risk was assessed beyond law being thought of in an instrumental sense. If my argument is that much of the present debate over technological risk regulation suffers from the flaw of incorrect characertisation of the problem, that raises the question about how scholars and policy-makers should think about technological risk decision-making. My starting point for this exercise is one of the important features of technological risk seen above—that technological risk controversies are primarily disputes over how risk is evaluated. If one returns to this point what becomes clear is that the institutional context for risk evaluation will have an important role to play in determining the acceptability of risk evaluation processes. That institutional context is neither science nor democracy, however— it is public administration. 72

The most obvious example being the work of Sheila Jasanoff. For earlier published examples see E Fisher and R Harding, ‘The Precautionary Principle and Administrative Constitutionalism: The Development of Frameworks for Applying the Precautionary Principle’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006); E Fisher, ‘Precaution, Law and Principles of Good Administration’ (2005) 52 Water Science and Technology 19; and E Fisher, ‘Beyond the Science/Democracy Dichotomy: The World Trade Organisation Sanitary and Phytosanitary Agreement and Administrative Constitutionalism’ in C Joerges and E-U Petersmann (eds), Transnational Trade Governance and Social Regulation: Tensions and Interdependencies (Oxford, Hart Publishing, 2006). 73

(B) Fisher Ch1

4/7/07

16:58

Page 19

Technological Risk, Public Administration, Administrative Constitutionalism 19

A The Necessary Role of Public Administration Look at nearly any risk evaluation process that is part of a regulatory regime in most Western jurisdictions and one will see that in the vast majority of cases the specific business of risk evaluation has been delegated to an unelected body by a primary lawmaker.74 Such administrative bodies may take many forms and may include both private and transnational bodies.75 Indeed, as will become clear, the flexibility of the administrative form is an inherent feature of disputes over technological risk evaluation. It is useful to briefly consider why technological risk evaluation has been delegated to administrative decision-makers. For ease of analysis, I limit myself to thinking about public administration in a tripartite separation of powers setting, although as will become clear, my argument is not limited to that realm. There are four main reasons which can be identified, all of which are concerned with the resource-intensive nature of such standard setting. First, and as already noted, the central question in standard setting, what is an acceptable risk, has an important socio-political aspect that needs to be addressed in the standard setting process. Legislation can provide some prior guidance on how this can be done but reaching a specific balance in regard to any particular standard requires more detailed consideration than can be achieved by legislative consensus. This is both due to the need to consider particular factual circumstances and because specific issues raise distinct socio-political concerns. Thus, for example, US legislation states that occupational safety and health standards should be ‘reasonably necessary or appropriate to provide safe or healthful employment and places of employment’.76 How this prescription should be applied to the wearing of clothing to protect against electrocution is very different from how it should be applied to protecting against the long-term health risks of being exposed to benzene.77 Each risk is of a different nature, occurs in a different employment context, raises distinct economic issues, and has its own politics. Second, standard setting requires the collection and organisation of information about a technological risk, a process for which the legislature has neither the time nor the resources.78 That information may relate to the nature of impacts, the causes of impacts, and the consequences of any regulatory action taken. What is held to be ‘relevant’ and ‘necessary’ information will depend upon what the problem is perceived to be and what are understood to be reliable sources of knowledge. Such information may include scientific information, economic data, 74

This will become obvious from the subsequent chapters in this book. See M Aronson, ‘A Public Lawyer’s Response to Privatisation and Outsourcing’ in M Taggart (ed), The Province of Administrative Law (Oxford, Hart Publishing, 1997); Freeman, above n 59; and Schepel, above n 34. 76 29 USC § 652(8). 77 Compare Alabama Power Co. v OSHA 89 F 3d 740 (11th Cir 1996) and Industrial Union Dept AFL-CIO v American Petroleum Institute 448 US 607 (1980). 78 This is a traditional reason for delegation. See C Sunstein and E Ullmann-Margalit, ‘SecondOrder Decisions’ (1999) 110 Ethics 5 at 17. 75

(B) Fisher Ch1

4/7/07

20

16:58

Page 20

Risk Evaluation through the Lens of Administrative Constitutionalism

lay observations, and judgements based on experience. Such information must also be organised and interpreted. There are many means of doing this, including a variety of risk assessment and cost/benefit analysis methodologies. Risk assessment can take many forms although it is primarily concerned with identifying and assessing the nature, level, and potential of a particular harm and characterising it in such a way as to be of relevance to the decision-making process.79 The use of these different methodologies leads on to the third necessary feature of any institutional response to technological risk problems—expertise. The need for specialised input, as with the need for information collection, is one of the classic justifications for the growth of public administration and in regards to risk regulation is particularly important for interpreting information. Expertise refers to a wide range of disciplines including the sciences, social sciences, professional vocations, and those who have experience. Experts are not isolated and will be part of their own disciplinary and professional communities. Expertise should not be understood in monolithic terms and while the phrase is often closely aligned with scientific rationality it does not necessarily need to refer to those with scientific training. Moreover, understandings of expertise are shaped by the context in which such expertise operates.80 The most important feature of expertise is that it is relative. An expert is thus someone who can lay claim to specific skills, experience, or knowledge which others do not have. Moreover, the expertise required in relation to the regulatory process is quite different from expertise required in other contexts in that greater emphasis is given to knowledge synthesis and prediction than to knowledge production.81 The final feature of risk standard setting is that it requires some form of communication between those involved in decision-making. At its simplest this is required because no single individual can possess all the information and expertise required for a particular risk problem. Communication and dialogue can also be a means of identifying different characterisations of technological risk and different views over their acceptability. Thus, depending on the risk problem, communication may involve the collection of information, reference to expertise, normative debate, or collaborative problem-solving. It also may take many different forms and embody many types of public participation.82 Thus, for example, it may be discussion among a small elite or it may be participation on a mass scale. It may be a form of information dissemination, interest group bargaining, or a 79 For an example of variations see Interdepartmental Liaison Group on Risk Assessment, Use of Risk Assessment Within Government Departments (London, Health and Safety Executive, 1996); and L Rhomberg, A Survey of Methods for Chemical Risk Assessment Among Federal Regulatory Agencies (Washington DC, National Commission for Risk Assessment and Management, 1997). 80 S Jasanoff, ‘Breaking the Waves in Science Studies: Comment on HM Collins and Robert Evans, “The Third Wave of Science Studies”’ (2003) 33 Social Studies of Science 389 at 393–396. 81 S Jasanoff, The Fifth Branch: Science Advisers as Policy Makers (Cambridge, Harvard University Press, 1990) at 77. 82 Compare S Funtowicz and J Ravetz, ‘Three Types of Risk Assessment and the Emergence of Post Normal Science’ in S Krimsky and D Golding (eds), Social Theories of Risk (Westport, Praeger, 1992) l; and J Rossi, ‘Participation Run Amok: The Costs of Mass Participation for Deliberative Agency Decision-Making’ (1997) 92 North Western University Law Review 173.

(B) Fisher Ch1

4/7/07

16:58

Page 21

Technological Risk, Public Administration, Administrative Constitutionalism 21 deliberative discussion in which the outcome is greater than the sum of individual preferences. In a traditional national context, only public administration can provide the flexible institutional space that can encompass these four features of risk evaluation and integrate them. The legislature, while the most democratic, does not have the expertise, the institutional capacity, nor the time to deal with specific risks.83 Likewise, the courts do not have the expertise, or the institutional means to provide for information collection or deliberation. Nor can they provide ongoing oversight of particular problems but only post hoc remedies that are at the instigation of the parties.84 As such, risk regulation represents the classic type of social problem that led to the rise of the administrative state in the first half of the twentieth century.85 The recent developments in both governance and globalisation have done little to halt this trend but rather have increased the role of administrative power.86 Moreover, due to the need to integrate these different factors into standard setting and risk appraisal the scope of administrative power is often considerable, as well as being highly discretionary in nature. By pointing out that the framework for technological risk decision-making is public administration I am not arguing that science and democracy are not important. Nearly all administrative regimes engaged in technological risk decision-making rely on both expert and public consultation, nearly all require reference to facts and values, and nearly all draw on both science and democracy. The nature of science, participation, values, and expertise will be defined very differently in different regimes but all contribute to the risk evaluation process. This can be seen from the various case studies in this book as well as by looking at nearly any risk regulation regime in operation. Nor is my argument that there is not a distinction between science and democracy or that that distinction is not important.87 Rather, I am arguing that scholars and policy-makers are looking at regulatory technological risk evaluation through the wrong prism. By focusing on the science/democracy dichotomy they are overlooking the fact that our understanding of these concepts is actually processed through another institutional form which will shape our understanding of risk evaluation problems. Moreover, those scholars who tend to think about the science–law interface in technological risk decision-making solely in a courtroom context are also deploying an incorrect lens. Technological risk decision-making is not occurring in the bipolar context of 83 R Posner, ‘The Rise and Fall of Administrative Law’ (1997) 72 Chicago Kent Law Review 953. This is even the case in those jurisdictions where it is the primary lawmaker that formally makes the ultimate decision (I am thinking here of Comitology procedure in the EU). See Council Decision 468/99/EC of 28 June 1999 laying down the procedures for the exercise of implementing powers conferred on the Commission [1999] OJ L184/23. See ch 6. 84 C Sunstein, The Partial Constitution (Cambridge, Harvard University Press, 1993) at 320 and 334–8. 85 F Frankfurter, The Public and its Government (New Haven, Yale University Press, 1930). 86 Schepel, above n 34. 87 Indeed, the distinction, as well as the distinction’s negation, is arguably a necessary feature of modernity. See B Latour, We Have Never Been Modern (Cambridge, Harvard University Press, 1993).

(B) Fisher Ch1

4/7/07

22

16:58

Page 22

Risk Evaluation through the Lens of Administrative Constitutionalism

the adversarial trial, where concepts of evidence and expertise are regulated by rules of evidence. Rather, it is occurring in the polycentric, malleable and openended institutional context of public administration.

B The Contentious Role of Public Administration, Law, and Administrative Constitutionalism As already noted, technological risk decision-making is conflict bound. Much of this has to do with the uncertainties over technological risks, and these disputes are in part disputes over how communities are governed. If risk evaluation is primarily carried out by public administration then such disputes will also be over the nature of public administration. Yet here is an added, and important, complication. Even ignoring issues of technological risk, the role of public administration is an uneasy and contentious one that gives rise to legal debates over how it should be constituted and limited. The simple reason for this that while public administration is a well-entrenched and necessary feature of democratic life it has long been perceived to be both undemocratic and unconstitutional and its role has defied principled explanation—it is the ‘awkward family heirloom’ of the contemporary constitutional state.88 Even at the dawn of the twenty-first century, the legal scholar Bruce Ackerman states that what is desperately needed is a constitutional design that accepts the need for supplementary bureaucratic lawmaking in the ongoing regulatory enterprise but that self-consciously confronts the serious legitimation problems involved.89

As opposed to being accommodated into any constitutional design public administration is invariably viewed as an intruder - ‘the great cuckoo in the nest, elbowing out the historic actors in the drama of government’.90 As Cook notes, the position of public administration begs the question of ‘how can a long-range, stable, even permanent exercise of governmental authority be reconciled with a regime of popular sovereignty?’91 The result of this situation is that there are ongoing attempts to explain, justify, and legitimise administrative power which have resulted in a range of theories which often prescribe quite different roles to administrative bodies. These include attempts to democratise public administration,92 control it,93 88 C Farina, ‘The Consent of the Governed: Against Simple Rules for a Complex World’ (1997) 72 Chicago Kent Law Review 987 at 987. 89 B Ackerman, ‘The New Separation of Powers’ (2000) 113 Harvard Law Review 633 at 696. 90 MJC Vile, Constitutionalism and the Separation of Powers, 2nd edn (Indianapolis, Liberty Fund, 1998) at 409. 91 B Cook, Bureaucracy and Self Government: Reconsidering the Role of Public Administration in American Government (Baltimore, Johns Hopkins University Press, 1996) at 3. 92 M Dorf and C Sabel, ‘A Constitution of Democratic Experimentalism’ (1998) 98 Columbia Law Review 267. 93 T Lowi, The End of Liberalism: The Second Republic of the United States, 2nd edn (New York, WW Norton & Co, 1979).

(B) Fisher Ch1

4/7/07

16:58

Page 23

Technological Risk, Public Administration, Administrative Constitutionalism 23 and/or to replace it with decentralised governance networks.94 A constant feature of the administrative state in nearly every jurisdiction has been a continuous reworking of its nature and role.95 Indeed, the focus on the science/democracy dichotomy can be seen as part of this state of denial and reworking. The problem is that, as an exercise in reworking, it bears little relationship to the actuality of public administration and provides few tools to make sense of how public administration is constituted, limited, and held to account. It is here that law comes in, and in these circumstances law has primarily two purposes. First, law plays the primary (albeit not the only) role in constituting, limiting, and holding public administration to account so as to ensure that it has at least some tenuous legitimacy. It does so by being the main instrument by which power is granted to public administration by the legislature and by providing the mechanisms by which such power can be policed, such as by judicial review. The processes of constitution, limitation, and accountability are ongoing ones and, due to the normative nature of law, legal frameworks, rules, and doctrines will reflect different understandings about what is and should be the role and nature of public administration.96 This is the case whether it be decisions over how a legal term should be interpreted, what legal concepts are relevant, or how a framework should be designed. As Mashaw notes in relation to administrative procedure, it contributes to the ‘construction of an operationally effective and symbolically important normative regime’.97 As such, he goes on to note, it shapes ‘administrative decision-making in accordance with our fundamental (but perhaps malleable) images of the legitimacy of state action’.98 Thus, and to paraphrase, behind any body of administrative law lies a theory of the administrative state.99 This relates to the second role of law, which is that it provides both arenas and discourses for disputing the role and nature of public administration. Law as such is not just instrumental—it is not content free.100 Legislative reform debates, judicial review cases, or other forms of calling to account are sites for determining and shaping what is, and should be, the role and nature of public administration. Likewise, the law itself is the discourse through which this is done. Law is not just another site for carrying on a scientific or political debate. Legal disputes over technological risk decision-making are carried on in legal terms, and law has its own internal logic and philosophy which will influence these debates. Legal imperatives will shape understandings of the nature and role of public administration 94

C Scott, ‘Accountability in the Regulatory State’ (2000) 27 Journal of Law and Society 38. P Light, The Tides of Reform: Making Government Work 1945–1995 (New Haven, Yale University Press, 1997); and H Parris, Constitutional Bureaucracy: The Development of the British Central Administration since the Eighteenth Century (London, George Allen & Unwin Limited, 1969). 96 J Mashaw, Greed, Chaos and Governance: Using Public Choice to Improve Public Law (New Haven, Yale University Press, 1997). 97 Ibid at 108. 98 Ibid. 99 P Craig, Public Law and Democracy in the United Kingdom and the United States of America (Oxford, Clarendon Press, 1990). 100 Cane, above n 62. 95

(B) Fisher Ch1

4/7/07

24

16:58

Page 24

Risk Evaluation through the Lens of Administrative Constitutionalism

and the nature of the problems that public administration is dealing with. At the same time understandings of public administration, and the problems they deal with, will shape the law. The role that law plays in relation to technological risk is essentially a form of administrative constitutionalism. Most of administrative law is essentially concerned with administrative constitutionalism but, as will be seen, debates over administrative constitutionalism arise in other contexts. The term ‘administrative constitutionalism’ captures the normative nature of law in this context and its constantly shifting nature due to the operation of different understandings of principled and constitutional government.101 Indeed, we might chose to think of administrative constitutionalism as an ‘essentially contested concept’ akin to concepts such as democracy and the rule of law.102 As this is the case it comes as no surprise that administrative law scholarship is replete with examples of different normative models of public administration which are said to underpin administrative law.103 The use of the term ‘constitutionalism’ in the administrative context may look odd to those who tend to link constitutionalism with democracy, particularly when standard setting can never be democratic. It reflects, however, the more traditional connotations of constitutionalism, which is that constitutionalism is concerned with the constituting and limiting of government so as to ensure its principled operation where there are divergences of opinion over what this means and entails.104 The term makes clear that while meta-constitutional principles such as the rule of law and the separation of powers are relevant, they are wholly inadequate by themselves to address, in full, the issue of how administrative power should be constituted and limited. Instead, the process of ‘constituting’ and ‘limiting’ public administration is invariably distinct from the more overarching processes of constitutionalism, and any particular constitutional structure can accommodate different models of good administration—a fact evidenced by the clear distinction between constitutional and administrative law in most jurisdictions. At the same time, however, and as will be seen in the chapters of this book, broader constitutional structures and cultures will influence understandings of administrative constitutionalism. One of the most significant features of the concept of administrative constitutionalism is that it highlights how the process of calling public administration to account is often a process of challenging what is understood to be the nature and 101 Note that constitutionalism is a concept ‘cloudy’ in its analytic and descriptive concept. See T Grey, ‘Constitutionalism: An Analytical Framework’ in J Pennock and J Chapman (eds), Constitutionalism: Nomos XX (New York, New York University Press, 1979). 102 WB Gaillie, ‘Essentially Contested Concepts’ (1956) 56 Proceedings of the Aristotelian Society 167; and J Waldron, ‘Is the Rule of Law an Essentially Contested Concept (in Florida)?’ (2004) 21 Law and Philosophy 137. I am grateful to Sheila Jasanoff for drawing my attention to these concepts. 103 C Harlow and R Rawlings, Law and Administration, 2nd edn (London, Butterworths, 1997); G Frug, ‘The Ideology of Bureaucracy in American Law’ (1984) 97 Harvard Law Review 1276; M Shapiro, Who Guards the Guardians: Judicial Control of Administration (Athens, University of Georgia Press, 1988); and Stewart, above n 52. 104 C McIlwain, Constitutionalism: Ancient and Modern (Ithaca, Cornell University Press, 1947) at 3.

(B) Fisher Ch1

4/7/07

16:58

Page 25

Technological Risk, Public Administration, Administrative Constitutionalism 25 role of public law decision-makers. As Davies notes, accountability has four major elements: the setting of standards; the obtaining of an account; the judging of such an account; and finally a decision about the consequences that arise from such a judgment.105 Most administrative law literature focuses on the last three of these, but the first is in many ways the most significant because those standards will be grounded in a particular understanding about good public administration and the process of holding someone to account is an attempt to enforce those standards. Accountability mechanisms thus become vehicles for particular understandings of administrative legitimacy and administrative law accountability. Thus, rather than being something that stabilises understandings of public administration, the operation of accountability can actually be, within limited boundaries, a highly destabilising force within a governance structure.106 Each case, or legislative reform is another attempt to either entrench or to overthrow a particular understanding of good public administration. The result is that administrative decision-makers often find themselves operating in a context of multiple and conflicting accountabilities in which each accountability mechanism is based on different standards of what is understood to be good public administration.107 It is important to appreciate that the concept of administrative constitutionalism is a broad one. This is so in a number of different ways. First, law is not the only means by which public administration is constituted, limited, and held to account.108 Debates over administrative constitutionalism can take place in nonlegal fora and concern issues other than law.109 Likewise, the ‘law’ in operation does not always accord with what is understood as law in a traditional sense.110 Administrative lawyers have become accustomed to dealing with both these issues in identifying other forms of administrative accountability and control, as well as thinking about the different forms of legal pluralism that have developed.111 In this regard, the scholarly and policy focus on governance should not be seen as an abandonment of administrative constitutionalism.112 Non-elected secondary lawmakers are just being limited by different means. Nor are debates over administrative constitutionalism only operating in the context of administrative law and related areas of governance. As will be shown in Part Two of this book, these debates can occur in other contexts such as trade law. The common theme in all these different circumstances is that the concept of administrative constitutionalism reflects the 105 A Davies, Accountability: A Public Law Analysis of Government by Contract (Oxford, Oxford University Press, 2001) at 81. 106 E Fisher, ‘The European Union in the Age of Accountability’ (2004) 24 Oxford Journal of Legal Studies 495. 107 Fisher, above n 73. 108 R Thomas, ‘Deprofessionalisation and the Postmodern State of Administrative Law Pedagogy’ (1992) 42 Journal of Legal Education 75. See ch 2. 109 See ch 2. 110 See chs 4 and 5. 111 ND Lewis, Law and Governance (London, Cavendish Publishing, 2001); and H Arthurs, ‘Without the Law’: Administrative Justice and Legal Pluralism in Nineteenth Century England (Toronto, University of Toronto Press, 1985). 112 Fisher, above n 70; and Holder, above n 60.

(B) Fisher Ch1

4/7/07

26

16:58

Page 26

Risk Evaluation through the Lens of Administrative Constitutionalism

symbiotic relationship operating between administrative law, public administration, and the problems that public administration addresses. Indeed, debates over administrative constitutionalism can be seen right across the administrative state, but this book is only a study of these disputes in the technological risk regulation context. As such, what follows should not be taken as applying to other areas of administrative law, although there will be overlaps.

III Two Paradigms of Administrative Constitutionalism in the Risk Regulation Context What is clear from above is that the role of law in technological risk regulation is not merely instrumental, irrelevant, or obstructive. Legal debates as a form of administrative constitutionalism are an important part of technological risk decision-making. Disputes over how to govern technological risk will merge into disputes over the legal validity of public administration. It is not just that technological risk regulation gives rise to disputes over administrative constitutionalism, however. At the same time, this area of regulation has given rise to a disproportionately large number of these disputes in comparison with other areas of administrative activity.113 This is for two reasons. First, the copious number of disputes reflects the contentious nature of this area, which was discussed in the first section. The undemocratic and malleable nature of public administration increases the perception that decision-makers are challengeable and changeable. Legal challenges and legal reform are based on the presumption that they can act as vehicles for potentially improving how risk is evaluated. Second, technological risk decision-making requires the delegation of considerable discretionary power to public administration in circumstances where the exercise of that power is not easily policed due to the expert and uncertain nature of decision-making. Technological risk decision-making may give rise to many different types of disputes over administrative constitutionalism. In saying this I am highlighting the fact that these disputes are legal in nature but are informed by understandings of public administration and the problems that public administration addresses. These disputes relate to all aspects of standard setting and risk appraisal including: the right of a court to review decision-making; the way in which a risk is characterised; the right of individuals to participate and challenge decision-making; the information used in evaluation; the form of communication employed; the values taken into account; the procedure before the court; the duty on a decision-maker to take other factors to account, and so on. In particular, a recurring theme in these legal disputes is whether there has been a ‘reasonable’ and ‘non-arbitrary’ 113 Indeed, risk regulation is often the focus for studies to do with the legitimacy of the administrative state. See Cook, above n 91; and H Richardson, Democratic Autonomy: Public Reasoning About the Ends of Policy (New York, Oxford University Press, 2002).

(B) Fisher Ch1

4/7/07

16:58

Page 27

Administrative Constitutionalism in the Risk Regulation Context

27

exercise of discretion on the part of an administrative decision-maker in evaluating technological risk. This is because this question relates directly to the process of technological risk evaluation and different understandings of what is ‘reasonable’ and ‘non-arbitrary’ will directly bear on the issue of how decision-makers should frame and exercise their powers in relation to technological risk. A recurring theme of legal disputes in relation to technological risk regulation is thus the malleability of the concept of reasonableness, the malleability reflecting the complex nature of law, public administration, and technological risk. That malleability, as well as the more general flexibility in the way in which law is interpreted is not unbounded, however. Rather, legal debate has been dominated by two different paradigms of administrative constitutionalism—the rational-instrumental and the deliberative-constitutive paradigms. These are ‘models’ or ‘paradigms’ which are primarily normative, and thus prescriptive, but also have a powerful impact upon how we understand what is. As Mashaw notes, such paradigms are: descriptions of how our political world is organised and how it works. They give us mental images of what to look for in political life and what to expect from it . . . the influence of pictures and themes is not just on what we expect and what we see, but also on what we demand and what we affirm.114

As this is the case, these paradigms are descriptions and prescriptions not only about the nature of public administration but also about the nature of science, public participation, and technological risks. As shall be shown throughout these chapters, each paradigm has had a powerful impact on legal disputes about technological risk evaluation and on the nature of technological risk itself. As such, the paradigms highlight the mutually dependent relationship between our understandings of the role of law, public administration, and technological risk. Below I outline and compare the two paradigms in some detail. The description of each of these paradigms directly relates to technological risk evaluation and thus the paradigms are described in those terms. With that said, and recognising the regime-specific nature of public administration, these paradigms do reflect more general paradigms about the role and nature of the administrative state. In putting forward these two paradigms and analysing legal disputes in terms of them I am acutely aware that I may be accused of two offences. The first is that I have placed too much weight on the existence of these paradigms and, as such, I am expecting them ‘to do too much work’. My response to this is that I am not treating them as reified and fixed realities. The purpose of the level of each paradigm’s detail is not to prove its existence in the minds of particular legal actors but rather to illustrate the close relationship between specific understandings of law, public administration, science, democracy, and technological risk.115 The detail shows how particular definitions of these concepts are co-produced and symbiotic. The second 114

Mashaw, above n 96 at 1. In this sense it is useful to think of these models as educative tools. See M Hesse, Models and Analogies in Science (Notre Dame, University of Notre Dame Press, 1966). 115

(B) Fisher Ch1

4/7/07

28

16:58

Page 28

Risk Evaluation through the Lens of Administrative Constitutionalism

offence I may be accused of is oversimplifying the disputes over administrative constitutionalism in the technological risk evaluation context by arguing that there are only two paradigms. I concede that there are many different paradigms of administrative constitutionalism in operation in relation to technological risk regulation but these two paradigms, representing the polar and incommensurable opposite understandings of administrative constitutionalism, aid understanding. I also hope that by identifying them this will lead to others identifying other models in operation.

A The Rational-Instrumental Paradigm The rational-instrumental (RI) theory of administrative constitutionalism116 construes public administration to be an ‘instrument’ of the legislature—a ‘robot’ or ‘transmission belt’117 whose task is strictly to obey the pre-ordained democratic will as it is expressed in legislation. In so doing, it should act effectively and efficiently. Public administration may not be democratic, but it can be structured to ensure the efficient pursuit of goals generated by the democratic process. More general versions of the RI theory have had a high profile throughout the twentieth century for precisely this reason—it seems a common-sense way to constrain administrative power for democratic purposes and its most obvious expression can be seen in the Weberian model of bureaucracy.118 In terms of technological risk evaluation, public administration under the RI paradigm is understood to be entrusted with a series of discrete tasks. The first task is to identify and assess a specific risk as well as assess the possible consequences of possible regulatory actions to manage that risk. This should involve a collection of all the information available and an assessment of that information by experts. As there is concern to constrain discretion, such expertise should deploy analytical methodologies which can both guide discretion and allow it to be easily assessed. An expert is thus someone who has training in particular rational methodologies. The guiding quality criterion in assessing that information is objectivity, which in itself will act as a constraint on administrative discretion.119 The second task is to identify the relevant value preferences. This is ideally done through interest representation so as to enable the proper and accurate commu116 Elements of this model can be seen in the following models: Cook, above n 91 at 4–5 (instrumental model); M Weber, From Max Weber: Essays In Sociology (London, Routledge, 1991) at ch 7 (theory of bureaucracy); T McGarity, Reinventing Rationality: The Role of Regulatory Analysis in the Federal Bureaucracy (Cambridge, Cambridge University Press, 1991) at ch 1 (comprehensive analytical rationality); M Shapiro, The Supreme Court and Administrative Agencies (New York, Free Press, 1968) at 67–91 (synoptic model); Stewart, above n 52 (interest representation model); and C Sunstein, The Cost Benefit State: The Future of Regulatory Protection (Washington DC, American Bar Association, 2002). 117 Stewart, above n 52. 118 Weber, above n 116 at ch 7. 119 T Porter, Trust in Numbers: The Pursuit of Objectivity in Science and Public Life (Princeton, Princeton University Press, 1995).

(B) Fisher Ch1

4/7/07

16:58

Page 29

Administrative Constitutionalism in the Risk Regulation Context

29

nication of preferences in relation to a specific risk. Interest representation is a way of gaining an account of the ‘will of the people’ and the role of the standard setter is simply to be an umpire overseeing the process—the power of public administration being to regulate on the fairness rather than on the substance of the interest representation process.120 Interest representation, as a surrogate legislative process, is also a way of gaining a more accurate understanding of how a prescription should apply in a certain circumstance. If public administration is construed as a body entrusted with a series of discrete analytical tasks then, for it to be effective, technological risk must be largely understood as something that can be managed through analytical techniques. Ideally, technological risk is quantifiable and the impact of any scientific uncertainty that exists is reducible through rational methodologies. Likewise, the value disputes inherent in technological risk can be described in terms of a series of discrete conflicts between endogenous preferences that can be expressed through a process of interest representation. The solution to disputes between those holding difference preferences is thus the striking of a bargain, or a compromise, and the role of public administration is to ensure such a bargain is a fair one. This understanding of technological risk is a direct result of concerns over the democratic legitimacy of the governing institutions and the paramount preoccupation of the RI paradigm is with ensuring that public administration has a limited and controlled role to play in public life so that it will not usurp democratic processes. The means by which it is ensured that such usurpation does not occur has two aspects. First, under the RI paradigm there is a limited role for general administrative discretion. Decision-making must adhere as closely as possible to a legislative scheme and to the methodologies described above. In the constituting of public administration the tasks delegated to it should be made explicit and the boundaries of standard setting clearly defined. Any legal or institutional framework for decision-making should ideally be one that promotes rational and objective approaches to decisions. Discretionary judgement based on unverifiable grounds is kept to a minimum and is effectively ‘filtered out’ by requiring the decision-maker to justify their decision on the basis of a narrow reading of their legislative mandate, objective knowledge and the accurate depiction of public preference. Second, because that is the case any decision is theoretically easy to assess and verify.121 Accountability mechanisms such as regulatory impact statements and risk assessment can be used to ensure that decision-makers are not acting beyond their legal boundaries. Post-decision accountability is largely a process of ensuring that the decision-maker has kept within their legislative mandate, kept within their methodology, and acted pursuant to public preferences. As such, it is portrayed by RI adherents as a relatively uncontroversial and straightforward process concerned with policing preordained boundaries. It gives very little discretion to those 120 121

Stewart, above n 52 at 1684; and Frug, above n 103 at 1369. On verification and objectivity see Power, above n 61.

(B) Fisher Ch1

4/7/07

30

16:58

Page 30

Risk Evaluation through the Lens of Administrative Constitutionalism

doing the reviewing, an added benefit for those concerned with the hegemony of the administrative state.

B The Deliberative-Constitutive Paradigm In contrast to the RI paradigm, the deliberative-constitutive (DC) paradigm of administrative constitutionalism122 promotes a model of public administration that is designed to address the factual and normative complexities of technological risk evaluation by granting to public administration substantial and ongoing problem-solving discretion in relation to particular issues. This power is needed so that the processes of technological risk evaluation can adapt to the uncertainties and issues involved in relation to specific technological risks. Public administration to do its task effectively must be a semi-independent and permanent institution. In the words of Cook, it must be a ‘political institution’ that ‘helps to create, to express, and to realise . . . public purposes’.123 Under the DC paradigm, legislation is less a set of strict commands and more akin to a constitution that sets out a series of general principles and the ‘broad parameters’ for the exercise of discretion. Public administration is not an ‘agent’ of a primary lawmaker but rather a body constituted by the legislature which exercises its own free will. The need for constituting such a body is because standard setting cannot be a set of simple preordained methodological formulae and positivist scientific method ‘all cold and iced’ is inadequate.124 All this is to do with the nature of risk evaluation itself. What is ‘relevant’ information will depend on the particular problem and can include all types of knowledge and information including experience and the lay observations of the public.125 Information is also understood to be dynamic and its ongoing flow needs to be encouraged. Scientific uncertainties mean that there is a constant need to take into account new scientific developments. Preferences over risks are not fixed and will change as information and management regimes change. The actual problems involved in standard setting may also evolve as standard setting proceeds and new issues are identifiable. Following on from this, the expertise needed for such standard setting is more broadly based. Experts are not simply those trained in scientific methodology but can also include those with experience such as professionals.126 Indeed, under the DC paradigm considerable trust in placed in 122 The following are useful sources for this model: Cook, above n 91 (constitutive model); M Seidenfeld, ‘A Civic Republican Justification for the Bureaucratic State’ (1992) 105 Harvard Law Review 1512; National Research Council, above n 24; C Sunstein, ‘Beyond the Republican Revival’ (1988) 97 Yale Law Journal 1539; Shapiro, above n 103 (deliberative model); and Richardson, above n 113. 123 Cook, above n 91 at 16. 124 W Lippmann, Drift and Mastery (New York, Mitchell Kennerley, 1914) at 290–3. 125 A Irwin, Citizen Science (London, Routledge, 1995); and S Krimsky, ‘Epistemic Considerations on the Value of Folk-Wisdom in Science and Technology’ (1984) 3 Policy Studies Review 246. 126 On this broader form of expertise see Frankfurter, above n 85 at 81, and AN Whitehead, Science and the Modern World (Cambridge, Cambridge University Press, 1928) at 244–88.

(B) Fisher Ch1

4/7/07

16:58

Page 31

Administrative Constitutionalism in the Risk Regulation Context

31

professional judgement, judgement which should ‘be kept articulate with democracy’.127 As with determining what relevant information is needed for technological risk evaluation, what is understood to be relevant expertise will depend upon the nature of the regulatory problem to be addressed. As this is the case it is important for DC administration to be actively engaged in thinking about how risk acceptability is best assessed in particular circumstances. It cannot passively rely on methodologies or formulae. As such, the most important duty placed on DC public administration is to deliberate. Deliberation is the means by which the issues for standard setters can be defined, the relevance of information and expertise established, and risk ultimately evaluated. Deliberation can take many forms. As the NRC states, deliberation is ‘any informal process for communication, and for raising and collectively considering issues’ and can include both consensual and adversarial processes of communication.128 Deliberation can involve a wide array of actors and depending on the problem at hand may involve a small group or be on a mass scale. An important feature of deliberative process is that it is insulated from the mainstream political process, which is over-responsive to particular political interests.129 The focus of deliberation is not votes, power, or interest group bargaining, but public reason.130 Such deliberative dialogue is not just a clash of interests but is transformative in nature in that it is an ‘iterative process that moves towards closure’ in which different actors ideally learn from the process and reconsider their perspectives.131 Public administration must play an important role in shaping and directing the debate as well as making the final decision. Deliberation does not bring legitimate authority to standard setting because it is democratic, albeit that is a side benefit, but because it acts as a ‘superior’ means of collective problem-solving.132 The process of calling standard setters to account is highly significant under the DC paradigm because the legislative boundaries are often so broad. Accountability processes are concerned with ensuring that decision-makers engaged in an active deliberative process that was within their ‘constituted’ jurisdiction. The giving of substantive reasons is essential to any such accountability process and those reasons should not only set out the grounds for why one course was adopted but also why others were rejected. In regards to post hoc accountability, the DC paradigm requires those reviewing decisions to do more than simply police the boundaries of methodology; rather, they must seek to ensure that a proper deliberative process took place. In light of the very flexible and fluid nature of problem-solving, those doing the holding to account are also exercising their own discretion and that in itself raises the problem of ‘who guards the guardians’.133 As such, the DC 127

H Croly, Progressive Democracy (New York, Macmillan Co, 1915) at 373. National Research Council, above n 24 at 73. 129 Seidenfeld, above n 122 at 1515 and 1542. 130 Richardson, above n 113 at chs 10–12. 131 National Research Council, above n 24 at 74. 132 J Steele, ‘Participation and Deliberation in Environmental Law: A Problem Solving Approach’ (2001) 21 Oxford Journal of Legal Studies 415 at 432–3. 133 Shapiro, above n 103. 128

(B) Fisher Ch1

4/7/07

32

16:58

Page 32

Risk Evaluation through the Lens of Administrative Constitutionalism

paradigm also requires a more substantive role to be recognised for courts and others involved in overseeing decision-making for accountability reasons, although any power is likely to be checked through the separation of powers, or some other form of interaction between public institutions.134

C The Paradigms Compared Each of these paradigms represents a different amalgamation of understandings about law, public administration, and technological risk. Looking at the paradigms together makes clear that the divisions involved in technological risk regulation are far more complex then the science/democracy dichotomy suggests. In particular, they also show the way in which understandings of law, public administration, and technological risks are ‘co-produced’ and that there is a ‘mutual embedding of natural knowledge and the social order’.135 At the same time, however, these two paradigms are not negating the type of conflict that have led scholars and policy-makers to characterise this area of decision-making in divided terms. Rather, my argument is that these divisions must be seen as being between different institutional forms and that demands for ‘more democracy’ or ‘more science’ are actually demands for particular types of public administration. These paradigms present a far more sophisticated understanding of the divisions involved in technological risk regulation debate. Moreover, as will be shown in every chapter of this book, they provide an important tool for making sense of complex legal arguments. The differences between the two paradigms can be seen in Table One. As is obvious from the table these two paradigms represent a very stark intellectual division. As the DC and RI paradigms co-exist within any legal culture, this intellectual division results in a highly fractured legal discourse in which different actors are attempting to shape legal frameworks in accordance with incommensurable understandings of what is and what should be. Likewise, any particular legal framework will be the product of a history of the interaction between these two worldviews. The end result is thus rarely a legal framework built on one paradigm but rather one that represents an awkward compromise between the two. This can be clearly seen throughout the chapters in this book. As Mashaw notes in regard to a similar divide, These groups largely talk past each other. Their separate conversations have few points of tangency, much less common understandings and vocabularies.136

The RI paradigm is deeply attractive in that it addresses the problem of the legitimacy of public administration by ensuring that it is limited and controlled and 134

C Sunstein, Designing Democracy: What Constitutions Do (New York, Oxford University Press,

2001). 135 136

682.

Jasanoff, above n 80 at 392. J Mashaw, ‘Deconstructing Debate, Reconstructing Law’ (2002) 87 Cornell Law Review 682 at

(B) Fisher Ch1

4/7/07

16:58

Page 33

Administrative Constitutionalism in the Risk Regulation Context Rational-Instrumental

33

Deliberative-Constitutive Technological Risk

Nature of Technological Risk

Objective and quantifiable Scientific uncertainty as manageable

Complex socio-political disputes involving values and epistemological problems

Public Administration and Law Relationship with Principal/Agent Primary Law Maker Limits on Discretion Accountability

Legislation Analytical methodology Interest representation Policing the methodology of decision-making and ensuring that decision-makers have kept within legislative limits.

Constitutive ongoing authority granted by primary lawmaker Constitutive structure Deliberative process Requires those reviewing the decisions to engage in a substantive review of decision-making.

Risk Regulation Standard Setting Values

Information

Expertise

Communication

Embodied in legislation but limited role for preferences in administrative decision-making Scientific and heavy reliance on analytical tools such as risk assessment and cost/benefit analysis Rational and kept strictly within methodological boundaries which serves as the primary basis for administrative action Interest representation

Inherent in all aspects of decision-making A range of information the relevance of which will depend on the nature of the problem. Broadly defined and one input into decision-making process Deliberation

Science and Democracy Science

Democracy

Objective

Legislation Interest Representation

Generally reliable knowledge but subject to methodological and epistemological limits Legislation Deliberation

Table One: The Rational Instrumental and Deliberative-Constitutive Paradigms of Administrative Constitutionalism

(B) Fisher Ch1

4/7/07

34

16:58

Page 34

Risk Evaluation through the Lens of Administrative Constitutionalism

has no authority except that derived from other sources. In that sense it is the classic model for lawyers. Likewise, it offers a means of assessing the success of a risk regulation regime by seeing whether certain legislative commands were followed through. This is particularly so if risk is understood in quantitative terms. Yet at the same time, the RI model does not adequately address the complex nature of technological risk. Control is seemingly at the expense of missing the point. Indeed, RI public administration is often seen as the cause of many technological risk controversies.137 In contrast, the DC paradigm starts with the messiness of technological risk problems and acknowledges that they require a flexible institutional response. As such, public administration is recognised as a permanent feature of the democratic landscape and is given a wide discretion. While such an institutional structure may provide an appropriate framework for decisionmaking it also threatens to usurp the legislature. Thus for lawyers it is a more difficult model to conceptualise and accept even though, as shall be shown, numerous regimes have been built on it. Public administration has its own internal authority but it is difficult to provide a justification for it other than necessity. Likewise, the process of holding a decision-maker to account is not an easy one. If a decision is said to be the product of complex deliberation and the judgement of the public administration then it is difficult to know when a decision-maker got it right and when a decision-maker got it wrong. Despite these differences there are also a number of common features of the two paradigms. First, both paradigms involve both sides of the science/democracy dichotomy seen above. Both recognise and require a role for science, politics, values, expertise, and participation. These concepts are defined in very different ways, however, and what is made clear by these paradigms is that disputes over technological risk evaluation are not between science advocates and participation advocates but between those who view both in different terms. Second, and following on from this, both paradigms are concerned with the democratic nature of public administration. In part this is due the fact that administrative constitutionalism, by seeking to ensure constitutional government, is seeking to ensure democratic government. In part, however, it is due to the fact that under each paradigm there is an attempt to democratise public administration. For the RI paradigm it is through interest representation138 and for the DC paradigm it is through deliberation.139 In both cases, these attempts largely fail in that such processes can never ensure that ‘political power is ultimately in the hands of the whole adult population’.140 Moreover, if such attempts were successful then arguably the institutional context would no longer be administrative. Third, both models require trust to be 137

Beck, above n 15; and Wynne, above n 5. Stewart, above n 52. 139 Dorf and Sabel, above n 92. 140 See D Robertson, The Penguin Dictionary of Politics, 2nd edn (London, Penguin, 1993) at 129. Also see Dahl’s four criteria for democracy in R Dahl, ‘Democracy’ in N Smelser and P Baltes (eds), International Encyclopaedia of the Social and Behavioural Sciences (Oxford, Elsevier Science Ltd, 2001). For a discussion of this point in the environmental context see Steele, above n 132. 138

(B) Fisher Ch1

4/7/07

16:58

Page 35

Administrative Constitutionalism as a Form of Legal Culture

35

placed in some aspect of social interaction.141 Thus under the RI paradigm that trust is in objectivity and for the DC paradigm it is in the human capacity for civic virtue and public reason. Moreover, the worthiness of both of these things to be trusted can be questioned. Relying on scientific method may shield ideological debate and also be a ‘shallow ritual in verification’.142 Likewise, the concept of civic virtue can overburden the citizen and place far too much faith in citizens to be able to rise above self-interest and for deliberation not to oust particular perspectives.143 As such, one model cannot be described as based on distrust and another on trust. The issue is what exactly is being trusted.

IV Administrative Constitutionalism as a Form of Legal Culture The discussion so far has focused on the concept of administrative constitutionalism and on the detail of the DC and RI paradigms. What it has not done is to explain in detail what administrative constitutionalism is. The answer to that question is that administrative constitutionalism is a form of legal culture. But that is an answer which raises many of its own questions, mainly because the concept of legal culture is controversial and, for many, its utility is questionable.144 Much of this criticism derives from those that seek to use the concept as an explanatory tool to predict developments.145 Understanding administrative constitutionalism as a form of legal culture is not an exercise in determinism, however. Rather, it highlights the interpretative power of such a concept.146 Legal culture, in effect, denotes legal norms, rules, and institutions and the interaction between them.147 It can, as Nelken notes, refer to everything from basic facts about a legal system to ‘more nebulous aspects of ideas, values, aspirations and mentalities’.148 Administrative constitutionalism is largely concerned with the latter and how those ‘ideas, aspirations and mentalities’ shape understandings of the administrative world. Thinking about administrative constitutionalism as legal culture highlights three important aspects of it—its ‘thick’ 141 Trust being an important aspect of technological risk decision-making. See Cvetkovich and Lofstedt (eds), above n 18. 142 Power, above n 61. 143 J Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (Cambridge, Polity Press, 1996) at 487; and C Sunstein, Infotopia: How Many Minds Produce Knowledge (Oxford, Oxford University Press, 2006). 144 D Nelken, ‘Towards a Sociology of Legal Adaptation’ in D. Nelken and J. Feest (eds), Adapting Legal Cultures (Oxford, Hart Publishing, 2001). 145 A point well made by J Webber, ‘Culture, Legal Culture, and Legal Reasoning: A Comment on Nelken’ (2004) 29 Australian Journal of Legal Philosophy 27 at 27. 146 Ibid. 147 See generally D Nelken, ‘Disclosing/Invoking Legal Culture’ (1995) 4 Social and Legal Studies 437; and P Legrand, Fragments On Law As Culture (Deventer, WEJ Tjenkk Willink, 1999). 148 D Nelken, ‘Using the Concept of Legal Culture’ (2004) 29 Australian Journal of Legal Philosophy 1 at 1.

(B) Fisher Ch1

4/7/07

36

16:58

Page 36

Risk Evaluation through the Lens of Administrative Constitutionalism

nature, its close symbiotic relationship with the rest of social life, and the inherent challenges involved in the process of globalisation. All of these are points which I have made already, but are useful to reiterate from a legal culture perspective. First, and as already noted, one of the most serious problems with the science/democracy dichotomy has been the way in which law has been primarily characterised in instrumental terms. There has been very little unpacking of legal concepts or mapping of legal ideas and legal inconsistencies. Law has been treated as a system of rules which are not only often seen as instrumental, irrelevant, or obstructive but also non-negotiable—they are fixed points on the landscape. Legal culture offers up a far ‘thicker’ description of the role and nature of law149: one in which legal reasoning is taken seriously and in which there is an appreciation of the importance of both history and conflict. Legal culture is not a static ideal but constantly changing due to changes in arguments and ideas. Its thickness is in part due to it operating at the point of contact between ‘sociological description and normative assessment’150 My constant reference to the thickness of legal culture is a thus a reference to its substantive, complex, and dynamic nature. Second, identifying administrative constitutionalism as legal culture recognises the symbiotic and malleable relationship between law and the rest of society. Law should not, in the words of Geertz, be understood as a ‘separate and “selfcontained” legal system struggling to defend its legal integrity in the face of the conceptual and moral sloppiness of ordinary life’.151 The line between law and non-law, between legal culture and other culture, is not easy to draw. While this may be disquieting for some it reflects the fact that legal discourse is embedded within society. For most administrative lawyers this blurring of the distinction between law and the ‘other’ is nothing new but rather a constant feature of public law.152 The practical implication of it is that the means by which public administration is constituted, limited, and held to account is not only legal, and public lawyers must look beyond the law to understand the constitutionality and accountability of administrative institutions.153 Tackling legal pluralism is the raison d’être of administrative law scholars154 and when I use the term ‘law’ in this book I am using it in its loosest sense. Yet the concept of administrative constitutionalism does carry this process of blurring law with ‘the other’ one step further in that it is based on the recognition that there is a constantly ongoing coproduction between law, technological risk, and public administration. It identifies the fact that there is not only a symbiotic relationship between law and administration but also between law, public administration, and the way in which the problems that administration addresses are characterised. 149

C Geertz, Local Knowledge (New York, Fontana Press, 1993). Webber, above n 145 at 36. 151 Geertz, above n 149 at 214. 152 Thomas, above n 108; and C Harlow, ‘Changing the Mindset: The Place of Theory in English Administrative Law’ (1994) 14 Oxford Journal of Legal Studies 419. 153 C Scott, ‘Analysing Regulatory Space: Fragmented Resources and Institutional Design’ (2001) Public Law 329. 154 Arthurs, above n 111. 150

(B) Fisher Ch1

4/7/07

16:58

Page 37

Administrative Constitutionalism as a Form of Legal Culture

37

Finally, understanding administrative constitutionalism as legal culture makes clear why there are divergences in discourses over how risk should be regulated in different jurisdictions and why there is no simple way of interrelating those discourses.155 It is not just a case of different jurisdictions being driven by different policies or starting assumptions156 but fundamental differences in culture. This creates a dilemma. As will be shown in the latter half of this book, there has been a shift to more ‘global’ approaches to risk evaluation both due to economic interdependencies and the creation of new supranational and international political communities. The profound influence of administrative constitutionalism on risk regulation regimes means that the globalisation of risk regulation involves some difficult questions about legal culture. Globalisation is not just a functional exercise enabling more efficient economies of scale. Nor is it just about power relationships: whether the local should trump the global or vice versa.157 Rather, what it involves is a whole series of issues concerned with the integration of different ‘thick’ discourses. Debates over administrative constitutionalism in national settings are dynamic, and inherent in them are competing arguments about the role and nature of public administration. Globalisation is thus not the bringing together of national ideas about regulation but rather the integration of legal discourses concerning administrative constitutionalism. It is also made more complicated by the creation of new supranational and international legal cultures. The end result is not so much a process of harmonisation but the creation of an interconnected web of different discourses spanning over a multitude of legal cultures. Much of the above is nothing new to legal scholars, and in particular administrative law scholars but the identification of the significance of legal culture is perhaps surprising to many non-lawyers working in the discourse who have tended to treat law in a ‘plug and play’ manner. For those who perhaps think I am overemphasising the significance of legal culture, what the case studies in this book constantly highlight is its thickness. In particular, in chapters three, four, and five, I show how administrative constitutionalism debates over risk evaluation are interlaced with debates over the role and nature of legal adjudication. The common law concept of legal adjudication158 is a somewhat amorphous one159 but broadly speaking it refers to the mode of decision-making utilised by courts in 155 See a similar argument about framing in D Winickoff et al, ‘Adjudicating the GM Food Wars: Science, Risk and Democracy in World Trade Law’ (2005) 30 Yale Journal of International Law 81. 156 For examples of work that takes this approach see J Hammitt et al, ‘Precautionary Regulation in Europe and the United States: A Quantitative Comparison’ (2005) 25 Risk Analysis 1215, and E Millstone et al, Science in Trade Dispute Related to Potential Risks: Comparative Case Studies, EUR 21301 EN (Seville, European Commission, Directorate General: Joint Research Centre, European Science and Technology Observatory, Institute for Prospective Technological Studies, 2004). 157 See S Jasanoff and M Long Martello (eds), Earthly Politics: Local and Global in Environmental Governance (Cambridge, MIT Press, 2004) for a good discussion of this point. 158 For the purposes of this book I am not directly concerned with civil law notions of inquisitorial adjudication. 159 Fuller, above n 13; D Galligan, Due Process and Fair Procedures (Oxford, Clarendon Press, 1996) at 241–52; Australian Law Reform Commission, Review of the Adversarial System of Litigation: Federal Tribunal Proceedings, Issues Paper No 24 (Canberra, 1998) at ch 2; and A Chayes, ‘The Role of the Judge in Public Law Litigation’ (1976) 89 Harvard Law Review 1281 at 1282.

(B) Fisher Ch1

4/7/07

38

16:58

Page 38

Risk Evaluation through the Lens of Administrative Constitutionalism

which there are two parties who are in an adversarial relationship and a judge who is ruling between them on the basis of their arguments and claims of proof. In talking about adjudicative method I am thus not concerned with technical rules of evidence but rather with basic tenets of adjudicative/adversarial procedure— those concerned with how facts should be proved, the role of the judge, and what rights parties should have to argue their case. Indeed, for the purposes of this book adjudication and adversarialism are taken to be roughly synonomous concepts. The relevance of adjudication to the subject-matter in this book may not seem immediately obvious. Technological risk evaluation is an exercise in ex ante rulemaking and thus bears very little in common with post hoc bipolar adjudicative processes concerned with determining rights and responsibilities.160 But the legal concept of adjudication has an important role to play in debates over administrative constitutionalism for two reasons. First, in most common law jurisdictions, adjudicative procedure was the historical blueprint on which administrative decision-making was based.161 It was quickly realised however that adjudicative procedure was not appropriate for this new realm of government activity, and the history of administrative law in all these jurisdictions has been a history of breaking free from the shackles of that procedure.162 At the same time, however, adjudicative procedure has remained a reference point for determining what is a fair procedure, particularly because it is a practical manifestation of the rule of law.163 While the generalist concept of administrative adjudication cannot be directly mapped onto the DC and RI paradigms (which are risk evaluation-specific) what can be seen is that administrative adjudication is based on a prescription for instrumental public administration and, as such, promotes a model of public administration that is aligned to the RI paradigm. The role of a decision-maker is constrained to a few predefined questions, fact-finding is regulated by preordained rules of proof, and there is no deliberation between the parties. In other words, adjudicative procedure is often the promotion of the RI paradigm and what we will see in chapter three is that in the US debates over how courts should judicially review risk evaluation decisions were complicated because of this. The second reason why adjudicative procedure is relevant can be seen in chapters four and five. In these two chapters what can be seen is that it is the concept of adjudicative procedure that pertains to how the reviewing tribunal goes about its task which is influential. Thus in chapter four it is tribunals’ shifting perceptions 160

Sunstein, above n 84 at ch 11. G Hewart, The New Despotism (London, Ernest Benn Ltd, 1929); and J Landis, The Administrative Process (New Haven, Yale University Press, 1938). 162 AV Dicey, ‘The Development of Administrative Law in England’ (1915) 31 Law Quarterly Review 148; Attorney General’s Committee on Administrative Procedure, Final Report of the Attorney General’s Committee on Administrative Procedure (Washington DC, Government Printing Office, 1941); Kerr Committee, Commonwealth Administrative Review: Committee Report, Parliamentary Paper No 144 (Canberra, Commonwealth Government Publishing Office, 1971); and Arthurs, above n 111. 163 Hewart, above n 161, and H Arthurs, ‘Rethinking Administrative Law: A Slightly Dicey Business’ (1979) 17 Osgoode Hall Law Journal 1. 161

(B) Fisher Ch1

4/7/07

16:58

Page 39

The Precautionary Principle and Administrative Constitutionalism

39

about merits review and adjudication which results in RI and DC understandings of the precautionary principle. Likewise, in Chapter Five it is Dispute Settlement Panels’ assumptions about the role and nature of WTO dispute settlement which shape understandings about what is reasonable action on the part of national administrations. In both cases it is the ambiguous nature of the tribunal’s task which causes adjudication to play such an important role, and both cases are important reminders that the role of law is never simply instrumental. Indeed, what all of the above makes clear is that legal culture is not just an irritant in the process of comparative analysis. Administrative constitutionalism is a form of legal culture and in so being emphasises the fact that the symbiosis between law, public administration, and technological risk is culturally dependent. In highlighting this I could be accused of being contradictory. I have stressed the importance of institutional and cultural context as a major reason for the irrelevance of the science/democracy dichotomy and in characterising administrative constitutionalism as a form of legal culture. At the same time, however, I have described two abstract paradigms of administrative constitutionalism which seem to owe very little to a specific context. If context and culture are important then of what value are abstract paradigms? Such a criticism misses the point, however. The RI and DC paradigms are not representations of realities but rather tools for understanding the interrelationship between understandings of technological risk, law, and public administration. They represent competing and irreconcilable intellectual impulses in debates over public administration in Western democracies committed to liberal constitutionalism. The importance of these paradigms is that they move debate beyond the stalemate of the science/democracy dichotomy. They highlight how debates are driven forward by debates over the legitimacy of public administration and how the role of law and legal discourse in those debates is not instrumental.

V An Example: The Precautionary Principle and Administrative Constitutionalism Assertion by itself is never enough and to establish the utility of the lens of administrative constitutionalism I need to do two things. First, I need to show how administrative constitutionalism operates as a lens through which to understand technological risk evaluation. Second, I need to show how such a lens clarifies and brings into perspective debates over risk evaluation. This I do in the next five chapters. In each chapter I examine a particular controversy or legal development in terms of administrative constitutionalism and show not only how what is under analysis makes sense in terms of administrative constitutionalism but also how understanding it in those terms highlights issues and problems which have tended to be ignored but are crucial in developing regulatory systems of risk evaluation.

(B) Fisher Ch1

4/7/07

40

16:58

Page 40

Risk Evaluation through the Lens of Administrative Constitutionalism

It is useful to give a brief example of thinking in terms of administrative constitutionalism before diving headlong into those more substantive chapters, however. This I do in this section by considering one of the most controversial principles of contemporary risk regulation—the precautionary principle. I have a twofold purpose in examining this principle at this stage. The first reason is that such an exercise quickly makes clear the utility of the concept of administrative constitutionalism because understanding the precautionary principle in terms of administrative constitutionalism brings into focus the nature of the principle and why it has become such an issue for debate. The second reason is that the precautionary principle is the focus for discussion in both chapter four and chapter six and having a general understanding of the principle before examining it in the context of two fundamentally different legal cultures is important.

A The Precautionary Principle Broadly speaking, the precautionary principle is a principle concerned with risk evaluation which requires decision-makers to pay particular attention to the problems created by scientific uncertainty. The principle, in its present form, had its origins in West German environmental policy of the 1970s but from the late 1980s onwards was transplanted into a range of national, supranational, and international regimes.164 Within those regimes the principle has been interpreted and characterised in many different ways.165 In nearly all cases, the precautionary principle has proved controversial and, although well established in many jurisdictions,166 it has also been subject to strong criticism.167 While there are many different characterisations of the precautionary principle in different legal cultures the most common formulation of it is as follows: where there is a threat to human health or environmental protection a lack of full scientific certainty should not be used as a reason to postpone measures that would prevent or minimise such a threat.168

164 N de Sadeleer, Environmental Principles: From Political Slogans to Legal Rules (Oxford, Oxford University Press, 2002) at ch 3. 165 E Fisher, ‘Precaution, Precaution Everywhere: Developing a “Common Understanding” of the Precautionary Principle in the European Community’ (2002) 9 Maastricht Journal of European and Comparative Law 7. 166 E Fisher et al, ‘Implementing the Precautionary Principle: Perspectives and Prospects’ in E Fisher, et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006) at 5–6. 167 C Sunstein, Laws of Fear: Beyond the Precautionary Principle (Cambridge, Cambridge University Press, 2005); and G Majone, ‘What Price Safety? The Precautionary Principle and its Policy Implications’ (2002) 40 Journal of Common Market Studies 89. 168 A version of this definition can be found in Principle 15 of the Rio Declaration on Environment and Development; the Preamble of the Convention on Biological Diversity; and Art 10(2) of the Cartagena Protocol on Biosafety. On other versions see R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999) at Annex One.

(B) Fisher Ch1

4/7/07

16:58

Page 41

The Precautionary Principle and Administrative Constitutionalism

41

As a principle it requires decision-makers to focus directly on the complex and dynamic nature of scientific uncertainty, something which has been relatively ignored by lawyers and policy-makers alike.169 Furthermore, the precautionary principle is a legal principle and, as such, it does not direct a particular outcome to occur. Rather, like all legal principles, it ‘states a reason that argues in one direction but does not necessitate a particular decision’.170 In other words, as a principle, it regulates the reasons for a decision and the process by which a decision is made.171 Like all principles, the principle can only be substantively defined in the context that it is operating in. As already noted, the principle is primarily a public law principle. In so being, it can apply to both the internal and external exercise of state authority.172 From the internal perspective, the precautionary principle is a principle that governs what is understood as legitimate regulatory action in circumstances of scientific uncertainty. This close interrelationship between the precautionary principle and administrative constitutionalism means that how the principle is defined, applied, and decision-making pursuant to it subjected to review, will depend on ideas of administrative constitutionalism.173 In terms of the external exercise of state authority, the principle also raises issues of administrative constitutionalism, albeit in a more indirect way. The precautionary principle may be relevant in this setting to how a state exercises its power in an international, transnational, or supranational context. In these circumstances the principle acts as a reason to require sovereign states to take action174 or acts as a reason for a state to derogate from their international obligations, usually in the context of trade regimes.175 In each case the precautionary principle needs to be interpreted in light of the framework of legal obligations, the purpose of such regimes, and the legal culture.176 At the same time, concepts of administrative constitutionalism will also have as role to play for two reasons. First, the same institutions are nearly always being used for the internal and external exercise of 169 Wynne, above n 5, and S Dovers and J Handmer, ‘Ignorance, Sustainability and the Precautionary Principle’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). 170 R Dworkin, Taking Rights Seriously (London, Duckworth, 1977) at 26. For a lengthier discussion on the nature of the precautionary principle as a legal principle see Fisher, above n 165 at 15–18. 171 E Fisher and R Harding, ‘The Precautionary Principle: Towards a Deliberative, Transdisciplinary, Problem-Solving Process’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999), and J Peel, The Precautionary Principle in Practice: Environmental Decision-making and Scientific Uncertainty (Sydney, Federation Press, 2005) at 156–9. 172 Fisher, above n 165 at 20–24; and O Godard, ‘The Precautionary Principle and Catastrophism on Tenterhooks: Lessons from Constitutional Reform in France’ in E Fisher et al (eds), Applying the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). 173 Fisher and Harding, above n 73. 174 J Cameron and J Abouchar, ‘The Precautionary Principle: A Fundamental Principle for the Protection of the Global Environment’ (1991) 14 Boston College International and Comparative Law Review 1. 175 Bohanes, above n 54; and Appellate Body Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), WT/DS26/AB, 16 January 1998. 176 See chs 5 and 6.

(B) Fisher Ch1

4/7/07

42

16:58

Page 42

Risk Evaluation through the Lens of Administrative Constitutionalism

state authority. This is particularly the case in an era in which non-discriminatory risk regulation measures are subject to trade obligations due to the fact that national risk regulation standards often conflict with such trade obligations.177 Second, and adding more complexity to the matter, what is a ‘reasonable’ exercise of external state authority will involve a determination of what is an appropriate model of administrative constitutionalism to be operating in that context.178 This is because in deciding what is a ‘reasonable derogation’ reference must not only be had to the purposes of trade regimes, but also to common understandings about how risk regulation standards are set.179 The relationship between administrative constitutionalism and trade regimes is not of direct concern to me here but will be examined in considerable detail in chapters five and six.

B The Precautionary Principle and the DC and RI Paradigms As a principle that raises issues of administrative constitutionalism, the precautionary principle is directly concerned with the role that information, and the scientific uncertainty embedded in that information, has for establishing the reasonableness of administrative action. In particular, the principle is capable of being interpreted in both RI and DC terms, and both DC and RI interpretations of the precautionary principle can be found throughout law and policy.180 Indeed, much of the debate over the legitimacy of the precautionary principle has been over the paradigm of administrative constitutionalism that the principle is seen to promote. I would go so far as to argue that the major reason why the principle has become so controversial, and in particular attracted the interest of public law scholars, is that the principle has become the contemporary reference point for debates over administrative constitutionalism.181 Thus, for example, some of the most strident criticism against the principle has been from those that perceive it as a mandate for DC public administration and thus a threat to the RI paradigm.182 Likewise, those who promote it are often arguing strongly against the logic of the RI paradigm in the risk evaluation context.183 Debates over the precautionary principle are thus, in essence, debates over the legitimacy of the administrative 177 For an example of a discussion of this overlap see N McNelis, ‘The Role of the Judge in the EU and WTO: Lessons from the BSE and Hormones Cases’ (2001) 4 Journal of International Economic Law 189. 178 Fisher, above n 73. 179 M Maduro, We, The Court: The European Court of Justice and the European Economic Constitution (Oxford, Hart Publishing, 1998). 180 Fisher and Harding, above n 73; and Fisher, above n 73. 181 J Graham and S Hsia, ‘Europe’s Precautionary Principle: Promise and Pitfalls’ (2002) 5 Journal of Risk Research 371; and A Klinke and O Renn, ‘A New Approach to Risk Evaluation and Management: Risk-Based, Precaution Based, and Discourse Based Strategies’ (2002) 22 Risk Analysis 1071. 182 Sunstein, above n 167; and G Marchant and K Mossman, Arbitrary and Capricious: The Precautionary Principle in the European Union Courts (Washington DC, The AEI Press, 2004). 183 A Stirling et al, ‘A Framework for the Precautionary Governance of Food Safety: Integrating Science and Participation in the Social Appraisal of Risk’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006).

(B) Fisher Ch1

4/7/07

16:58

Page 43

The Precautionary Principle and Administrative Constitutionalism

43

state. This can best be seen by comparing DC and RI understandings of the principle. From a DC perspective, the precautionary principle can be understood as a natural outgrowth of the DC paradigm. If public administration is constituted to address complex and uncertain technological problems in an ongoing and flexible fashion then the need for decision-makers not to be constrained by a requirement of ‘full scientific certainty’ before they can take action is obvious.184 In such circumstances the principle is not a ‘prescribed formula’ but a flexible principle that ensures that administrative decision-makers are not ignoring problems of scientific uncertainty.185 Both ‘information’ and ‘scientific uncertainty’ are interpreted broadly and it is a principle that applies to all aspects of the decision-making process. In other words, the principle is a confirmation of the DC ideal that public administration should exercise its power flexibly and conscientiously. Or to put it more simply—the principle empowers decision-makers to act in accordance with DC principles. Information about environmental risks is only one constituent element in that decision-making process, and the principle requires the decision-maker to look at the role of information in that context. That context is a dynamic and polycentric one in which a decision-maker needs to be engaging in an ongoing process of analysis and deliberation. Such flexibility does not mean decision-making is unaccountable. Accountability under the DC paradigm is established by scrutinising the decision-making process in context to show that a decision-maker has seriously addressed the complexities of the problem.186 At the same time, however, the precautionary principle can be understood as an RI principle that operates in certain limited circumstances. Such an interpretation of the principle arises because of the way that the principle is phrased as a negative (scientific uncertainty should not be used as a reason) which operates if a certain threshold is met (where there are threats). From this perspective, the principle is allowing decision-makers to deviate from the normal definition of reasonable action—that decision-makers should act on the basis of proof—but only in limited circumstances. On this basis, the principle is interpreted far more narrowly. The principle is primarily understood as applying to the decision to act after there has been the assessment of information if that assessment has identified a threat.187 In such circumstances, the principle gives a margin of discretion to decision-makers but it is not an enabling principle in the way it is under the DC paradigm. In particular, the principle is only applying to how information is used in risk evaluation and, in so doing, is not altering the fact that the starting point for judging the reasonableness of decision-making is proof. Those that promote this RI ideal of the precautionary principle are often also conceptualising the administrative process in interest representation terms in which there is very little 184 Fisher and Harding, above n 73 at 119–24; and A Deville and R Harding, Applying the Precautionary Principle (Sydney, Federation Press, 1997). 185 Fisher and Harding, above n 73 at 123. 186 Stirling et al, above n 183; and Deville and Harding, above n 184. 187 Fisher and Harding, above n 73 at 124–6.

(B) Fisher Ch1

4/7/07

44

16:58

Page 44

Risk Evaluation through the Lens of Administrative Constitutionalism

room for deliberation. In particular, public participation is understood to be a source of irrational fear.188 What can thus be seen is that these different interpretations are embodied in different models of legitimate administrative governance which encompass different understandings about the role and nature of information, communication, expertise, and the nature of legislative prescriptions.189 Thus from a DC perspective, the precautionary principle is operating in a context where expertise and information are broadly defined and communication is deliberative. In contrast, from an RI perspective, expertise and information are strictly defined in methodological terms and communication is understood as a form of interest representation. Different definitions of the precautionary principle thus come accompanied with different administrative baggage.

C The Precautionary Principle and the Burden of Proof That is not the only baggage that definitions of the precautionary principle come accompanied with, however. As already stressed, administrative constitutionalism is not just instrumental, but a form of thick legal culture which is also shaped by legal concepts and must be made to make some sense in legal terms. The importance of this can be seen in the legal illogicality of a common characterisation of the precautionary principle—that it requires a shifting of the onus or burden of proof in circumstances of scientific uncertainty.190 This characterisation seems to suggest that in decision-making the burden would be normally on an objector to an activity to prove that it presents risks, but in cases where the objector establishes a ‘threat’ then the application of the precautionary principle requires the burden to be placed on those wishing for the activity to occur. Such a definition has been used by lawyers and non-lawyers alike and can be found in both policy and commentary. Such a description of the principle is superficially appealing if a burden of proof is understood in a general sense as a form of reasoning strategy, the aim of which is to ‘structure vast areas of indeterminacy’.191 Yet, from a legal perspective this conceptualisation makes no sense.192 A preliminary problem is that such descriptions are rarely elaborated upon and there has been little attempt to map the precautionary principle onto conventional legal understandings about such burdens.193 Thus while there have been some discussions about standards of proof,194 there has been virtually no examination of 188

Sunstein, above n 167. Fisher and Harding, above n 73 at 126–9, and see Table One in this chapter. 190 For a good reflective discussion of this type of formulation see Jones and Bronitt, above n 47. 191 R Gaskins, Burdens of Proof in Modern Discourse (New Haven, Yale University Press, 1992) at 20. 192 Fisher, above n 47 at 332. 193 C Tapper, Cross and Tapper on Evidence, 10th edn, (London, Lexis-Nexis, 2004) at ch 3; and Jones and Bronitt, above n 47. 194 Jones and Bronitt, above n 47 at 142–5; and P Harremoës et al, ‘Twelve Late Lessons’ in P Harremoës et al (eds), The Precautionary Principle in the Twentieth Century: Late Lessons from Early Warnings (London, Earthscan, 2002) at 208. 189

(B) Fisher Ch1

4/7/07

16:58

Page 45

The Precautionary Principle and Administrative Constitutionalism

45

any other aspects of legal proof. Thus, for example, it is not obvious whether such a burden is an evidential burden or a persuasive burden. Cross and Tapper on Evidence describes the distinction between the two in the following terms: The evidential burden has been defined as the obligation to show, if called upon to do so, that there is sufficient evidence to raise an issue as to the existence or non-existence of a fact in issue. The persuasive, ‘legal’, or ‘probative’ burden has been defined as the obligation of a party to meet the requirement of a rule of law that a fact in issue must be proved or disproved.195

The requirement to prove a ‘threat’ would seem an evidential one but the requirement then placed on those wishing for the development to go ahead would seem a persuasive one, particularly as there is no way to prove the safety of an activity.196 But if this is the case, then the shifting of the burden is largely illusory. Also missing from these discussions of the precautionary principle as a shifting of the burden of proof is much discussion about what actually has to be ‘proved’. Burdens of proof normally apply to the establishment of historical facts but in cases where the precautionary principle is being applied the relevant ‘facts’ are predictions about future events and about compliance with regulatory decisionmaking. It is not clear that legal burdens of proof are particularly useful in this context, as the discharge of either burden is probably philosophically impossible. Most significantly, however, the concept of a burden of proof is an adjudicative one and operates in a bipolar adversarial context. The role of such burdens is to allocate both the responsibilities for the production of evidence and the risk of errors if a trial does not produce truth.197 Moreover, within the adjudicative context, ‘it is for the parties to determine whom and what [judges and juries] they see or hear, but not how they evaluate and reason from evidence’.198 As Ganz notes: The judicial process is an adversary process. The judge is an impartial arbiter between two parties. He does not step into the arena. It is for the parties to make out their case. The judge is concerned with the issues between them not with finding the best solution to a particular problem.199

The problem, however, is that this is not the context in which the precautionary principle is applying. Rather, it is applying in the administrative sphere where decision-making is polycentric and the administration has an active role in factfinding. As Chayes has noted, the phrase in administrative decision making should be ‘fact evaluation’ rather than ‘fact finding’.200 Moreover, while the concept of the burden of proof may be appropriate for creating ‘formal accuracy’ in bipolar

195

Tapper, above n 193 at 137. Nicholls v Director General of National Parks and Wildlife Service (1994) 84 LGERA 397 at 419. 197 A Stein, Foundations of Evidence Law (Oxford, Oxford University Press, 2005). 198 W Twining, Rethinking Evidence: Exploratory Essays (Evanston, Northwestern University Press, 1994) at 194–5. 199 G Ganz, Administrative Procedures (London, Sweet and Maxwell, 1974) at 19. 200 Chayes, above n 159 at 1297. 196

(B) Fisher Ch1

4/7/07

46

16:58

Page 46

Risk Evaluation through the Lens of Administrative Constitutionalism

litigation201 concerned with historical facts, it is nonsensical in the administrative context where fact-finding tends to be more predictive. All of the above provides a good example of where policy-makers and commentators have taken a legal concept without thinking about the context it is operating in. Law is not just an instrument; it is a discourse in which legal concepts are closely interrelated with other concepts. At the same time, however, this does not mean that the precautionary principle can never operate as a shifting of the burden of proof but that if it does, it raises some fundamental legal questions. This will be seen in chapter four. Moreover, while such a characterisation may be grounded in concepts of adjudication, it does also have implications for administrative constitutionalism. This is because, as seen in the last section, the reliance on concepts from the adjudicative/adversarial sphere often reflects RI understandings about the role of public administration. The same can be seen here. Under this formulation of the principle, a decision-maker can normally act when a certain burden of evidentiary proof is met. Thus, for example, a risk is proved on the balance of probabilities or the evidence meets a certain statutory threshold. The principle adjusts this burden of proof but only when a threshold threat has been established. That threshold usually amounts to an evidentiary requirement itself. Thus, when the principle is operating, the onus is on a different regulatory presumption—regulation should occur except if it can be proved that it should not. In both cases, the decisionmaker has little active discretion of their own but must simply assess information in accordance with the evidentiary rules.

VI Conclusion At the outset of this chapter I stressed that this book is radical exercise in recharacterisation. In this chapter, I have provided a new lens through which to view technological risk evaluation. It is, I argue, a far more powerful lens than those that presently exist. In particular, it is a lens which gets rid of the ‘blindspots’202 which are created by the lens which has dominated this area of scholarship—the science/democracy dichotomy. The blindspot that has been created has obscured the role of both public administration and law in debates over risk evaluation. In the conclusion, I will consider some of the implications of removing this blindspot but before that I will examine five different examples of risk evaluation issues through the lens of administrative constitutionalism. There are two points worth making about my focus in the next five chapters. First, my analyses in most of these chapters concern how administrative constitu201

R Posner, The Problems of Jurisprudence (Cambridge, Harvard University Press, 1990) at 216. I have explored the concept of ‘blindspots’ and scholarship elsewhere. See E Fisher and P Schmidt, ‘Seeing the Blindspots in Administrative Law: Theory, Practice, and Rulemaking Settlements in the United States’ (2001) 30 Common Law World Review 272. 202

(B) Fisher Ch1

4/7/07

16:58

Page 47

Conclusion

47

tionalism has shaped understandings about the role and nature of science and expertise rather than on how they have shaped the role and nature of communication and participation. In part, this is due to the fact that science and expertise have been the primary focus of attention in debates over technological risk regulation—a situation which is a product of the fact that under the DC and RI paradigms science and expertise play a very different role in being sources of authority. In part, however, I am focusing on science and expertise because they have remained ‘black boxes’ to legal scholars while concepts of public participation have been subject to greater analysis.203 Despite my focus on understandings of expertise and science it is important to remember that these understandings do not exist in isolation. They are always part of broader theories of administrative constitutionalism and as such are integrated with particular understandings of the role and nature of public participation. The second point worth making is that in these chapters I focus on controversies and legal developments with which most working in the risk regulation field would be acquainted. This is on purpose: what I want to do is replace many of the existing prisms which have been used to view these controversies and developments—those narratives have primarily been based on the science/democracy dichotomy—with what I argue is the more powerful lens of administrative constitutionalism. In doing this I wish to show that, while thinking in terms of administrative constitutionalism may be an act of radical re-characterisation, it is one that is likely to yield a far more fruitful line of inquiry for policy-makers, legal scholars and legal practitioners.

203 Rossi, above n 82, and C Sunstein, ‘Factions, Self Interest and the APA: Four Lessons Since 1946’ (1986) 72 Virginia Law Review 271.

(B) Fisher Ch1

4/7/07

16:58

Page 48

(C) Fisher Intro to part 1

31/5/07

09:57

Page 51

Introduction to Administrative Constitutionalism in National Legal Cultures At the end of the last chapter I stated that to prove the utility of the concept of administrative constitutionalism I need to do two things First, I need to show how administrative constitutionalism operates as a lens through which to view debates and legal developments in relation to risk evaluation. Second, I need to show how the lens clarifies and brings into perspective these debates and legal developments In particular, I need to show how the lens of administrative constitutionalism helps scholars, lawyers, and policy-makers make sense of the complex detail of these debates and how it gets rid of the ‘blindspots’ which have resulted in fundamental issues in these debates being ignored. In this first part, I examine three different controversies and legal developments in three different national jurisdictions—the UK, the US, and Australia. In chapter two, I examine what was understood to be the role and nature of the Southwood Working Party in the bovine spongiform encephalopathy (BSE) crisis in the UK. Understandings of that role and nature were influenced by the rationalinstrumental (RI) and deliberative-constitutive (DC) paradigms of administrative constitutionalism. In chapter three, I examine debates in the 1970s about how US Federal courts should judicially review Federal risk regulation rule-making and, in particular, those debates concerning ‘hard look review’ and the operation of the ‘substantial evidence’ judicial review standard in relation to Occupational Safety and Health Administration (OSHA) rule-making. I argue that that these debates should be understood as debates between grounding judicial review, and thus public administration, in either the RI or DC paradigm. In chapter four, I provide an account of the interpretation of the precautionary principle by courts and tribunals in Australia. Most of this process of interpretation has occurred in the context of merits review by specialist environmental courts, and I show not only how interpretation of the principle has been in DC and RI terms but that these different interpretations of the principle have been co-produced alongside different understandings of merits review. The subject matter and scope of analysis is very different in each of these chapters In the UK chapter, the study is of decision-making around a particular event in central government decision-making. In the US chapter, the analysis is of a series of judicial decisions and the legal framework in which those decisions were

(C) Fisher Intro to part 1

52

31/5/07

09:57

Page 52

Introduction to Part One

operating. In Australia, the study is of a series of merits review decisions, concerning natural resources and planning made by Federal, State, and local administrative decision-makers What this makes clear is that this book is not an exercise in straightforward comparative analysis1 These ‘snapshots’ through the administrative constitutionalism lens are not about finding ideal solutions to the problems of risk regulation, comparatively evaluating national responses to similar problems, or trying to capture the quintessential nature of a particular legal culture. Rather, I am showing simultaneously the commonalities and differences between these ‘snapshots’: commonalities because these ‘snapshots’ are ‘snapshots’ of administrative constitutionalism, and differences due to the fact that administrative constitutionalism is a form of legal culture and thus is manifesting itself in different ways In particular, placing these studies side by side draws attention to two interrelated matters First, it highlights how thinking in terms of the science/democracy dichotomy not only has resulted in an overlooking of the administrative nature of technological risk evaluation but also of the administrative, legal, and institutional detail of standard setting. Indeed, one of the reasons why these case studies cannot stand as representative examples of a particular culture is because so much of the debate and detail is to do with particular institutions, particular laws, and particular administrative arrangements. Second, this side-by-side comparison highlights the fact that the differences between legal cultures and administrative constitutionalism are multidimensional and such differences cannot be summed up in simplistic terms such as one culture being more legalistic than another.2 It is useful to elaborate on these differences in a little more detail by highlighting some of differences in administrative constitutionalism in each of these case studies. Again, it is important to remember that these case studies are not necessarily representative of the legal culture they operate in, but a rough comparison is useful to make my point—that the differences are multidimensional. Three differences can be noted—differences in the nature of administrative constitutionalism, in the role of law, and in the interrelationship between legal concepts and regulatory regimes for technological risk evaluation.

I Nature of Administrative Constitutionalism The first difference between the UK, the US, and Australia is that, despite having common legal foundations, each of these jurisdictions has a very distinct legal culture, and thus different forms of administrative constitutionalism. In the UK, 1 Of course, comparative legal analysis is never straightforward: see P Legrand and R Munday, Comparative Legal Studies: Traditions and Transitions (Cambridge, Cambridge University Press, 2003). 2 For an interesting multidimensional examination of different ‘civic epistemologies’ see S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005) at 258–69.

(C) Fisher Intro to part 1

31/5/07

09:57

Page 53

Nature of Administrative Constitutionalism

53

administrative constitutionalism is a blurring of law, policy, political theory, ideology, and convention.3 Moreover, due to the dominance of the Westminster model, administrative and constitutional law have tended to be collapsed into each other, and thus are not distinct.4 All this can be disconcerting and historically has resulted in some commentators being critical of the substantive context of UK public law5 and suggesting it is a ‘deserted territory’6 in which no meaningful debate has occurred. But, as Harlow and Rawlings have noted: [t]he history of political ideas in Britain is peculiarly rich. The tradition of public law less so. It is none the less a pluralist tradition . . .7

In other words, UK administrative constitutionalism has been pluralistic but more ‘political’ than legal in nature, a fact evidenced by the dominance of the concept of the ‘political constitution’.8 Likewise, in relation to risk regulation, developments have been incremental and less marked by the creation of rule-bound legal regimes.9 The implications of all of this for thinking about UK administrative constitutionalism are twofold. First, administrative constitutionalism cannot be easily delineated from more general debates about constitutionalism.10 Second, the ‘thickness’ of UK administrative constitutionalism is less a legal ‘thickness’ but more the product of a language of normative discussion, the set of historical reference points, the range of solutions proposed in the past, the institutional norms taken for granted, given a particular context of repeated social interaction.11

Indeed, UK administrative law is profoundly pluralistic in its legal forms.12 In contrast, in the US, Federal administrative constitutionalism is clearly distinct from general debates over constitutionalism, and is very legal in nature. The creation of the Federal administrative state was through an explicit process of debate, particularly around the Administrative Procedure Act 1946 (APA)13 and the ‘rights revolution’ of the late 1960s and early 1970s.14 Most risk regulation administrative agencies were the product of this later era and were created in a 3

C Harlow and R Rawlings, Law and Administration, 2nd edn (London, Butterworths, 1997). M Freedland, ‘The Evolving Approach to the Public/Private Distinction in English Law’ in M Freedland and J-B Auby (eds), The Public Law/Private Law Divide: Une entente assez cordiale (Oxford, Hart Publishing, 2006) at 95. 5 KC Davis, ‘English Administrative Law—An American View’ (1962) Public Law 139. 6 B Schwartz and W Wade, Legal Control of Government (Oxford, Clarendon Press, 1972) at 4. 7 Harlow and Rawlings, above n 3 at 90. 8 A Tomkins, Public Law (Oxford, Clarendon Press, 2003). 9 Although arguably this is changing with the creation of the Environment Act 1995 and the Food Standards Act 1999. 10 See the discussion in Tomkins, above n 8. 11 J Webber, ‘Culture, Legal Culture, and Legal Reasoning: A Comment on Nelken’ (2004) 29 Australian Journal of Legal Philosophy 27 at 32. 12 H Arthurs, ‘Rethinking Administrative Law: A Slightly Dicey Business’ (1979) 17 Osgoode Hall Law Journal 1; and Harlow and Rawlings, above n 3. 13 M Shapiro, ‘A Golden Anniversary? The Administrative Procedure Act of 1946’ (1996) 3 Regulation 40. 14 C Sunstein, After the Rights Revolution (Cambridge, Harvard University Press, 1990). 4

(C) Fisher Intro to part 1

54

31/5/07

09:57

Page 54

Introduction to Part One

short space of time. Indeed, concepts of administrative constitutionalism are easy to identify in the US because of the way in which discourse has focused so obviously on the legitimacy of the administrative state.15 Moreover, Federal administrative constitutionalism is carried on in legal arenas and legal terms, and the courts, particularly the District of Columbia Circuit of the Federal Court of Appeals, have dominated that discourse. This means that debates over administrative constitutionalism are easily accessible and it is no accident that my inquiries into the concept of administrative constitutionalism began in that jurisdiction.16 In contrast to these two jurisdictions, Australian administrative constitutionalism is different again. Although influenced by US and UK administrative law doctrines,17 it has developed on its own terms. There is, particularly at the Federal level, a distinct body of administrative law which is striking in terms of its doctrinally rule-bound nature.18 Moreover, Australian administrative constitutionalism is unique because of reforms in the 1970s which led to the creation, as part of an ‘administrative law package’, of the Administrative Appeals Tribunal which limits and holds administration to account on the basis of merits review.19 Administrative constitutionalism discourses have thus been shaped by two very different types of review—judicial and merits review. From a risk regulation perspective, matters are more complicated by the fact that much of the administration of environmental issues has occurred at the State and local level due to the fact that the States primarily have constitutional competence in this area. At that level there has been the creation of specialist environmental courts carrying out merits review.20 What this means is that debates over administrative constitutionalism have occurred in a series of decentralised forums created through a process of ‘state sponsored legal pluralism’21 which carry out a peculiarly ‘Australian’ type of review.22 Despite being so decentralised, legal developments in Australian tend to 15 For examples see M Shapiro, Who Guards the Guardians: Judicial Control of Administration (Athens, University of Georgia Press, 1988); and B Cook, Bureaucracy and Self Government: Reconsidering the Role of Public Administration in American Government (Baltimore, Johns Hopkins University Press, 1996). 16 E Fisher, ‘The Risks of Quantifying Justice: The Use of the Substantial Evidence Test in Judicial Review of OSHA Rule-Making’ in R Baldwin (ed), Law and Uncertainty, Risks and Legal Processes (The Hague, Kluwer Law International, 1997); and E Fisher, Risk, Expertise and Judicial Review: Scope of Review and Decision-making under Scientific Uncertainty, Thesis in Law Submitted for a DPhil Degree (Oxford, University of Oxford, 1998). 17 P Cane, ‘The Making of Australian Administrative Law’ in P Cane (ed), Centenary Essays for the High Court of Australia (Sydney, LexisNexis Butterworths, 2004) at 319. 18 M Aronson et al, Judicial Review of Administrative Action (Sydney, Law Book Co., 2004) at 12; and Cane, above n 17 at 330. 19 Cane, above n 17 at 323; and G Taylor, ‘The New Administrative Law’ (1977) 51 Australian Law Journal 804. 20 See the discussion of them in M Grant, Environmental Court Project. Final Report. (London Department of Transport, Environment and the Regions, 2000). 21 H Arthurs, ‘Without the Law’: Administrative Justice and Legal Pluralism in Nineteenth Century England (Toronto, University of Toronto Press, 1985). 22 Cane, above n 17 at 323; and P Stein, ‘Specialist Environmental Courts: the Land and Environment Court of New South Wales, Australia’ (2002) 4 Environmental Law Review 5.

(C) Fisher Intro to part 1

31/5/07

09:57

Page 55

The Role of Law

55

be understood in overarching terms, particularly in relation to the precautionary principle, which is understood as a meta-principle which has a freestanding status.23 The end result is, as will be seen in chapter four, Australia has one of the most legally and normatively complicated regimes of administrative law in the common law world.

II The Role of Law The discussion above highlights the very different forms of administrative constitutionalism that are operating in these different jurisdictions Implicit in those different forms is a series of different roles for law, and it is worth highlighting these roles separately. It is worth remembering in relation to this point that, while administrative constitutionalism is a form of legal culture, that does not mean it consists of a series of legal rules and principles strictly understood. Indeed, the distinct forms of administrative constitutionalism in these three jurisdictions emphasise the centrality of ‘legal pluralism’ in administrative law,24 as well as the embedded nature of legal culture. As already noted above, in the UK, administrative constitutionalism has traditionally not consisted of ‘law’ strictly understood. Historically, the courts have played a marginal role in holding decision-making to account,25 and the legal frameworks for executive action in the UK left much to the imagination in that they delegated broad powers to administration. The constituting, limiting, and holding of decision-makers to account has been through organisational structures, political processes, policies, managerial technique and through the promotion of an ethos. Moreover, the lack of ‘hard law’ is particularly the case in the risk regulation field, where regulatory strategy has historically been based on a more flexible model in which there is a significant role for negotiation between the regulated and the regulator.26 The setting and enforcement of standards has often been one and the same process, and the legal frameworks relatively loose.27 There is also little in the way of judicial review litigation in this area, and thus the courts have played no role in promoting particular understandings of administrative constitutionalism. As shall be seen in the next chapter, the implication of this is that in studying UK administrative constitutionalism in the risk regulation field there is very little attention paid to law, as narrowly understood, because it simply does not 23 B Preston, ‘Ecologically Sustainable Development in the Courts in Australia and Asia’, Paper for seminar held at Buddle Findlay, Lawyers (Wellington New Zealand, 28 August 2006). 24 Arthurs, above n 21. 25 Although arguably this is changing: see P Craig, Administrative Law, 5th edn (London, Thompson/Sweet & Maxwell, 2003). 26 K Hawkins, Environment and Enforcement (Oxford, Oxford University Press, 1984). 27 Royal Commission on Environmental Pollution, Best Practicable Environmental Option, Twelfth Report, Cmnd 310 (London, HMSO, 1988).

(C) Fisher Intro to part 1

56

31/5/07

09:57

Page 56

Introduction to Part One

play a significant role in constituting, limiting, and holding public administration to account. The situation is very different in the US, where administrative constitutionalism has been a discourse carried out in the form of legislation and case-law. This is not to say that the constituting and limiting of public administration has not occurred by other means,28 but that the debates have primarily being carried out in judicial and legislative terms. This is particularly in relation to the former because of the way in which ‘adversarial legalism’ is an inherent feature of US legal culture.29 Nearly every risk regulation rule passed will be subject to judicial review, and thus the courts have become a major forum for debating the legitimacy of administrative action. A study of US administrative constitutionalism in relation to risk regulation is thus largely a study of case-law. In the Australian context, the role of law is a far more complex one. On the one hand, the legal frameworks for risk regulation decision-making are well delineated and established. Whether it be Federal regulators regulating natural resources or local planners assessing the environmental risks of a project, the main means of constituting and limiting such bodies is through legislation. Likewise, Australian judicial review doctrine is a complex body of case-law which clearly reflects ideas of administrative constitutionalism. On the other hand, the process of limiting and holding to account in the risk regulation field is mainly done by a series of quasi-judicial bodies that are carrying out a form of review that is simultaneously both administrative and judicial in nature—merits review. The legal status of such review is open to question30 and that means that the role that law, narrowly understood, plays in administrative constitutionalism is ambiguous. It is a significant one, however, and, despite the legally pluralistic nature of merits review, it has been its legal technicalities which have been influential in debates over administrative constitutionalism. In comparing the role of law in isolation in each of these case studies, once again it can be seen that the concept of ‘comparing’ each of these jurisdictions against one another is almost impossible. The ways in which public administration is being constituted, limited, and held to account in each jurisdiction are so different that each legal culture needs a different scope of analysis. The lens of administrative constitutionalism must be adjusted each time and the focus and expectations must be different.

28 The most obvious being the requirement of regulatory impact statements by the President’s Office of Management and Budget: see T McGarity, Reinventing Rationality: The Role of Regulatory Analysis in the Federal Bureaucracy (Cambridge, Cambridge University Press, 1991). 29 R Kagan, Adversarial Legalism: The American Way of Law (Cambridge, Harvard University Press, 2003). 30 L Pearson, ‘Merits Review Tribunals’ in R Creyke and J McMillan (eds), Administrative Law: The Essentials (Canberra, Australian Institute of Administrative Law Inc., 2002); and P Cane, ‘Merits Review and Judicial Review—The AAT as a Trojan Horse’ (2000) 28 Federal Law Review 213.

(C) Fisher Intro to part 1

31/5/07

09:57

Page 57

Regulatory Regimes for Technological Risk Evaluation

57

III The Relationship Between Legal Concepts and the Regulatory Regimes for Technological Risk Evaluation The final difference to note between these three case studies is the difference in the interrelationships between legal concepts and the regulatory regimes for technological risk evaluation. As noted in the first chapter, an important aspect of administrative constitutionalism is that it highlights the fact that the role of law is not just instrumental. Legal discourse has its own internal logic and this will often shape understandings about what is technological risk evaluation and how it should be carried out. Thus, for example, the influence of common law concepts of adjudication can be seen in a number of these chapters and, in particular, chapters three and four.31 The thickness of administrative constitutionalism as a form of legal culture can be seen in many different ways in these chapters, however. In the UK context there is very little formal law to focus on but what can be seen is how understandings and prescriptions of how public administration should be constituted, limited, and held to account to ensure its legitimacy impacted upon how the task of dealing with BSE was understood. Thus, historical debates over the ‘generalist civil service’ shaped the expert capacity of administrative decisionmaking, and standards of accountability influenced what was understood to be an acceptable basis for action in relation to BSE. Indeed, the UK context is a good starting point for understanding the relationship between administrative constitutionalism and the characterisation of technological risk because the debate is not cluttered by legal technicalities The situation is different in the US, where although discourses of administrative constitutionalism are more explicit they are also complicated by legal concepts such as adjudication and historically evolved concepts of scope of review. Thus, for example, I will show how confusion over the legal operation of the ‘substantial evidence’ standard of judicial review in the Occupational Safety and Health Act shaped debates over administrative constitutionalism in the risk regulation field. That standard had a conventional legal definition which made no legal sense when placed into the rule-making context of OSHA rule-making. Much of the case-law of the 1970s was concerned with deciphering this legislative anomaly, a process which ultimately impacted upon what was understood to be the role and nature of public administration. These debates are as much debates over legal detail as they are normative debates of administrative constitutionalism. In Australia, it has been the ambiguous nature of merits review which has had such a profound effect on how the precautionary principle has been interpreted in terms of administrative constitutionalism. Reviewing courts and tribunals have understood their legal roles in different ways and this has resulted in them promoting different models of administrative constitutionalism. Some of them have 31

See section IV of ch 1.

(C) Fisher Intro to part 1

58

31/5/07

09:57

Page 58

Introduction to Part One

seen their role as akin to judicial review, while others have seen it akin to a form of merits decision-making in an adjudicative context. These divergent legal conceptualisations have resulted in different understandings of what it is that risk regulators do. Again, what can be seen is that debates about administrative constitutionalism are embedded in legal discourses which are often quite complex in nature. An implication of this is that in all three jurisdictions, specific debates about administrative constitutionalism and technological risk evaluation must always be seen in the context of broader debates over the legitimacy of the administrative state. Indeed, as already noted in chapter one, the RI and DC paradigms which are specific to risk regulation reflect more generalist paradigms about the role and nature of public administration A study of technological risk evaluation through the lens of administrative constitutionalism thus makes it clear that, as an area of administrative activity, it cannot be studied in isolation

IV Conclusion This introduction to the study of administrative constitutionalism in national legal cultures is the second step in my process of radical reorientation. In chapter one, I argued the need for technological risk evaluation to be understood in terms of administrative constitutionalism. In this second step, I have explored the first implication of what that means—that administrative constitutionalism as a form of legal culture is complex. That complexity is rarely identified when technological risk evaluation is understood in terms of the science/democracy dichotomy, and my next step is to engage in that complexity.

(D) Fisher Ch2

31/5/07

09:57

Page 59

2 BSE, Expertise, and Administrative Constitutionalism: Examining the Role of the Southwood Working Party In late 1986 a new disease, bovine spongiform encephalopathy (BSE), was identified by the Ministry of Agriculture, Fisheries and Food (MAFF) in cattle in the UK. Little was known about the disease and whether it was a risk to human health. In April 1988 MAFF and the Department of Health (DH) set up a small expert committee to advise widely on BSE. That committee was the Southwood Working Party, and it concluded in its report in February 1989 that: From the present evidence, it is likely that cattle will prove to be a ‘dead-end host’ for the disease agent and most unlikely that BSE will have any implications for human health. Nevertheless, if our assessments of these likelihoods are incorrect, the implications would be extremely serious.1

Despite the caveat, and the discussion about the uncertainties in the report, this conclusion was treated by government decision-makers as authority for the proposition that BSE did not present a human health risk. This was the case until 20 March 1996 when another expert committee, the Spongiform Encephalopathy Advisory Committee (SEAC), advised that the ‘most likely explanation’ for the appearance of a new and fatal disease in humans, variant Creutzfeldt-Jakob disease (v-CJD), was exposure to BSE through the eating of beef and the use of beef products.2 At that time there were 10 confirmed deaths from v-CJD,3 a number which by March 2007 had risen to 112.4 The establishment, operating, and reporting of the Southwood Working Party was not the only event in what has become known as the ‘BSE crisis’,5 but it is its most defining and puzzling. It is a defining event because the government treated 1 R Southwood et al, Report of the Working Party on Bovine Spongiform Encephalopathy (London, Department of Health, Ministry of Agriculture, Fisheries and Food, 1989) at para 9.2. 2 Hansard, HC Deb, vol 275, col 375 (20 March 1996). 3 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Six: Human Health (London, HMSO, 2000) at para 7.40. 4 See the monthly v-CJD statistics at www.dh.gov.uk. 5 I appreciate that the word ‘crisis’ is a value-laden one and use it here simply as shorthand to refer to the events in relation to BSE in the UK between 1986 and 1996.

(D) Fisher Ch2

31/5/07

60

09:57

Page 60

BSE, Expertise, and Administrative Constitutionalism

the Working Party Report as the authoritative expert basis for administrative action, and the only significant regulatory actions they took in relation to the human health risks from BSE were justified in terms of that report. It is puzzling because the Working Party never expected their report to carry such authority and sought to make it plain that they were doing their best on very limited data, that much further research was necessary, that their assessment of risk might be wrong and that, were it wrong, the consequences would be extremely serious.6

In this chapter I make sense of this puzzling state of affairs by viewing the Southwood Working Party through the lens of administrative constitutionalism. My focus is upon what was understood as legitimate administrative action in relation to the human health risks from BSE, and what the Working Party was understood to contribute to that legitimacy. Put simply, that Working Party was understood in two very different ways due to the simultaneous operation of the deliberative-constitutive (DC) and rational-instrumental (RI) paradigms of administrative constitutionalism. On the one hand, UK public administration has been traditionally DC in nature, and expertise was broadly defined and understood as one input into the decision-making process. It was in these terms that the Southwood Working Party understood themselves, as did others working in MAFF and DH at the time. On the other hand, by the mid-1980s the UK administrative state was being remodelled on the basis of the RI paradigm of administrative constitutionalism and, as such, expertise was increasingly being understood in more RI terms. Due to this, the Working Party’s report was treated as the principal source of authority for administrative action. In other words my argument in this chapter is that the role given to the Southwood Working Party was a bifurcated one due to the nature of administrative constitutionalism operating within UK legal culture at that time. To understand this, the focus must not only be on the Southwood Working Party, but also on how this state of affairs came to be. As such, a large segment of this chapter is about the history of UK public administration. This chapter is structured as follows. In the first section, I illustrate how the human health risks from BSE are classic examples of technological risks and, as such, give rise to issues of administrative constitutionalism. In the second section, I show how the DC paradigm of administrative constitutionalism traditionally dominated UK public administration and technological risk evaluation, but by the mid-1980s administrative and regulatory reform was increasingly on the basis of the RI paradigm. In the third section, I examine the institutional context that the BSE crisis occurred in to show how both the RI and DC paradigms shaped that context. In the fourth section, I examine the establishment, operation and reporting of the Southwood Working Party to show how both paradigms of administrative constitutionalism influenced what was understood to be the role and nature of that Working Party. In the final section, I 6 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9 (London, HMSO, 2000) at para 1.1.

(D) Fisher Ch2

31/5/07

09:57

Page 61

Thinking of BSE in Terms of Administrative Constitutionalism

61

briefly consider events in this area after the Southwood Working Party to show how administrative decision-making since 1989 has been firmly based on the RI paradigm. Four points should be made before starting. First, as noted in chapter one, in describing action as consistent with either the RI or the DC paradigm I am not saying that particular actors identified those paradigms explicitly but rather that their actions can be seen as consistent with the ideals encapsulated in those paradigms. The second point is that this chapter is not a comprehensive account of the BSE crisis and for that readers will need to look elsewhere.7 The third point is that the analysis in this chapter relies heavily on the account of events in the Inquiry into BSE and v-CJD in the UK (‘the BSE Inquiry’)8 and the evidence given to that inquiry. This is not simply for the sake of convenience. Witnesses giving evidence in that quasi-judicial forum were likely to present their actions in terms which corresponded with what they understood to be legitimate administrative action.9 For that reason the Inquiry report, and the evidence accompanying it, provide a very accurate picture of the understandings of administrative constitutionalism that were operating at that time.10 Third, it is important to keep in mind the cultural ‘thickness’ of UK administrative constitutionalism and, in particular, the fact that the ways in which UK public administration has been constituted, limited, and held to account have not had a formal legal status. Rather, UK administrative constitutionalism has taken the form of convention, policy, institutional structure, and organisational culture.11

I Thinking of BSE in Terms of Administrative Constitutionalism To start with it is important to understand why, and how, the evaluation of the potential human health risks from BSE gave rise to issues of administrative constitutionalism. This is not mere pedantry on my part. An understanding of the intractable nature of the BSE problem makes clear that the evaluation of the risks arising from it was always going to be controversial. BSE, a disease in cattle, was first identified in 1986 and Creutzfeldt-Jakob Disease (CJD), a disease in humans, was first described in 1920.12 A variation of 7 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report (London, HMSO, 2000); P van Zwanenberg and E Millstone, BSE: Risk, Science and Governance (Oxford, Oxford University Press, 2005); and R Packer, The Politics of BSE (Basingstoke, Palgrave MacMillan, 2006). 8 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, above n 7. 9 S Hinchliffe, ‘“Indeterminacy In-Decisions”—Science, Policy and Politics in the BSE Crisis’ (2001) 26 Transactions of the Institute of British Geographers 182 at 187. 10 All that material is available at www.bseinquiry.gov.uk. 11 See the Introduction to Part One. 12 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Two: Science (London, HMSO, 2000) at para 2.21.

(D) Fisher Ch2

31/5/07

62

09:57

Page 62

BSE, Expertise, and Administrative Constitutionalism

CJD, v-CJD, was also identified in the early 1990s, and the issue at the heart of the BSE crisis, as it emerged in the mid 1990s, was whether the cases of v-CJD had been caused by exposure to BSE. The Southwood Working Party predated the identification of v-CJD, however, and in the 1980s when it reported, there was no link made between BSE and particular diseases in humans. With that said, it is useful to have an overview of the nature of both BSE and v-CJD to have a feeling for just how complex both these diseases were. Not only were the risks of each shrouded in scientific uncertainty, but their evaluation was also made problematic by behavioural uncertainty and because the acceptability of such risks was embedded deep in other socio-political disputes. The types of scientific uncertainties in relation to BSE and v-CJD were, and still are, many. BSE, CJD, and v-CJD are all fatal and progressive neurodegenerative diseases known as transmissible spongiform encephalopathies (TSEs) that cause the appearance of microscopic holes in the brain.13 TSEs have been identified in a number of other species including mink, deer, and sheep, but not all TSEs have the ability to be transmitted from species to species.14 While TSEs have long been recognised15 there was in 1986 very little solid scientific understanding about their cause, how they were transmitted, and their infectivity. In 1967 the prion hypothesis had been put forward that suggested that the causative agent of these diseases was a self-replicating protein, and this theory, while controversial and still subject to criticism, has been the basis for understanding TSEs ever since.16 Moreover, as TSEs have long incubation periods the only means of confirming them was by post-mortem.17 Different TSEs were found to have different modes of transmission including maternally, laterally, and iatrogenically.18 The implications of this for evaluating the risks from BSE was that uncertainties existed concerning how the disease was transmitted among cattle19 and whether the disease was transmissible to humans. The situation was made even more complicated by uncertainties in relation to CJD and v-CJD. There was some limited understanding about CJD, and in 1985 it had come to light that CJD had been transmitted to children treated with human growth hormone taken from human cadavers.20 There had also been some studies in the 1970s into whether there was a link between CJD and scrapie in sheep after it had been found that CJD could be transmitted to primates in experimental conditions.21 The charting of the emergence of a new form of CJD, v-CJD, in the early 1990s was also difficult. Variant CJD had much in common 13 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Two: Science (London, HMSO, 2000) at paras 2.4–2.34. 14 Ibid at paras 2.112–2.138. 15 The first case of scrapie was recorded in 1732: ibid at para 2.10. 16 Ibid at paras 2.66–2.70. 17 Ibid at para 1.1. 18 Ibid at para 2.101. 19 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years (London, HMSO, 2000) at chs 1–3. 20 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Eight: Variant Creutzfeldt-Jakob Disease (London, HMSO, 2000) at paras 2.3–2.6. 21 Ibid at paras 2.3–2.6.

(D) Fisher Ch2

31/5/07

09:57

Page 63

Thinking of BSE in Terms of Administrative Constitutionalism

63

with CJD but also was distinct in that it tended to be in younger people, was longer in duration, and had more of a psychiatric presentation, with the victim experiencing anxiety and depression.22 As well as these scientific uncertainties, there also existed a number of behavioural uncertainties in relation to the human health risks from BSE. The evaluation of the nature and extent of such risks depended crucially on human behaviour but such behaviour was difficult to assess for a number of reasons. Thus, for example, there was a lack of knowledge over slaughterhouse practices, which resulted in uncertainties in relation to the potential for diseases to be transmissible by those practices.23 Beef and beef by-products were also used in a wide range of products including animal feed, a variety of foods for human consumption, medical products, and cosmetic products. Yet while many of these uses were known, there was very little official centralised information about them, which made tracking the use of beef products in the food chain very difficult.24 The potential for BSE to result in large-scale and unexpected adverse effects was thus quite high, and in light of the incubation period of both BSE and v-CJD, could be over a considerable time period. As well, BSE presented a series of occupational health and safety risks for those working in the slaughter and waste industries, and a general pollution risk in relation to beef and beef product waste.25 Likewise, if humans were also infected that created a series of new risks such as those from blood transfusions and other medical procedures.26 Mapping the risks from BSE was thus very much about mapping a polycentric web in which countless actors were entangled. A proper evaluation of the risks required evaluation of all their behaviour. Finally, debates over the acceptability and nature of the risks arising from BSE were highly culturally contextual and embedded in a number of deeper sociopolitical disputes. Thus, for example, from an ethical perspective, there were some in the community who found the feeding of cattle to cattle (the cause of transmission of BSE among cattle) as barbaric.27 From a consumer perspective, the BSE controversy was only one in a series of food safety crises that had led to a lack of trust in government pronouncements over the safety of products.28 Likewise, the transmission of CJD through human growth hormone treatment had raised its own ethical questions. From an economic perspective, BSE presented a number of threats to the viability of the farming industry, particularly in regard to 22

Ibid at para 5.103. Packer, above n 7 at ch 3. 24 This can be seen in relation to government inquiries into the ingredients and manufacturing processes for agricultural feed. See especially Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years, above n 19 at ch 3. 25 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Six: Human Health, above n 3 at chs 8 and 10, respectively. 26 See the history of the transmission of CJD in Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Eight: Variant Creutzfeldt-Jakob Disease, above n 20 at chs 2 and 4. 27 A Irwin, Citizen Science (London, Routledge, 1995) at 22. 28 Agriculture Committee, Salmonella in Eggs, HC 108-I (London, HMSO, 1989); and R Lacey, Unfit for Human Consumption: Food in Crisis (London, Grafton, 1992). 23

(D) Fisher Ch2

31/5/07

64

09:57

Page 64

BSE, Expertise, and Administrative Constitutionalism

international trade.29 In such circumstances, how the government framed and conceptualised the risks would also impact upon how acceptable those risks were to members of the public. Two things are clear from the above. First, the problems that the Southwood Working Party were addressing were scientifically uncertain, behaviourally uncertain, and socio-politically embedded. They did not have the overall perspective that can be seen above but rather an incomplete understanding of the issues. This means that their task was a highly complex one. Second, the analysis above makes clear that BSE was a classic technological risk problem. As seen in chapter one, the practical consequence of a risk being a technological risk is not only that it is a candidate for collective regulation by the state, but that such regulation is the province of public administration, public administration being the only institution of the state which can integrate normative prescriptions, information, expertise, and communication in the risk evaluation process. Indeed, during the BSE crisis all regulatory action was taken by administrative bodies including the passing of 30 pieces of delegated legislation before 1996.30 The important role of public administration in the BSE crisis has been widely recognised.31 Commentators have noted that the BSE crisis drew attention to the ‘folkways’ of UK public administration,32 the ‘ethos of the government machine’,33 and the systems of administrative decision-making.34 Returning to my analysis in chapter one, there were three implications of the evaluation of the human health risks from BSE being a matter for public administration. First, if risk evaluation was the responsibility of public administration it could not solely be a scientific or democratic exercise. In other words, both the human health risks from BSE and their evaluation transcended the science/democracy dichotomy.35 Second, if the evaluation of the health risks from BSE was an administrative exercise then what was understood to be an acceptable and legitimate exercise in risk evaluation would be determined by understandings of good public administration and thus by administrative constitutionalism. This was noted in other terms by the BSE Inquiry when it stated:

29 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Ten: Economic Impact and International Trade (London, HMSO, 2000) at ch 4. 30 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Sixteen: Reference Material (London, HMSO, 2000) at ch 1. 31 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration (London, HMSO, 2000). 32 B De Marchi and J Ravetz, ‘Risk Management and Governance: A Post-Normal Science Approach’ (1999) 31 Futures 743 at 750. 33 Packer, above n 7 at 247. See also G Little, ‘BSE and the Regulation of Risk’ (2001) Modern Law Review 730 at 731. 34 R Gerodimos, ‘The UK BSE Crisis as a Failure of Government’ (2004) 82 Public Administration 911. 35 B Latour, Politics of Nature: How To Bring the Sciences Into Democracy (Cambridge, Harvard University Press, 2004) at 111–14; and A Irwin, ‘The Politics of Talk: Coming to Terms with the “New” Scientific Governance’ (2006) 36 Social Studies of Science 299 at 301.

(D) Fisher Ch2

31/5/07

09:57

Page 65

Thinking of BSE in Terms of Administrative Constitutionalism

65

The conventions under which the executive operated between 1986 and 1996 helped to determine the nature and quality of the response to the emergence and subsequent course of BSE.36

As such, to understand administrative action in relation to the BSE crisis there is a need to understand administrative constitutionalism as it operated in UK legal culture. Third, it must be appreciated that there is no single paradigm of administrative constitutionalism that operates in relation to technological risk evaluation. Rather, I identified two different extremes of administrative constitutionalism, the RI and the DC, both of which conceptualise technological risk and the role of public administration in very different ways. As all of this is the case, to make sense of the BSE crisis and the role of the Southwood Working Party in that crisis the starting point must be UK administrative constitutionalism and how it impacted upon those making decisions in relation to BSE. In particular, in thinking about the Southwood Working Party, the focus needs to be on how understandings of administrative constitutionalism shaped understandings about the role and nature of expertise. This is because, on any account, the role of that body in the decision-making process was in providing expertise in the risk evaluation process. The point is that expertise is defined differently under the RI and DC paradigms, due to broader understandings of good public administration being different in relation to each. From an RI perspective, expertise is highly rationalistic and kept strictly within methodological boundaries. The role of such expertise is to provide the facts which are the primary basis for administrative action. This is consistent with the more general RI understanding of public administration being a rational and constrained instrument of the legislature which acts only on the basis of evidence and legislative command. In contrast, from a DC perspective, expertise is far more broadly based and someone may be an expert due to experience and wisdom. The role of such expertise is to be one input into the administrative problem-solving process. Such an understanding of expertise is integral to the DC understanding of public administration as a constitutive institution that addresses complex problems through ongoing deliberation. While not the focus in this chapter, each concept of expertise will also be accompanied by understandings about the role and nature of communication and participation.37 Overall, what is clear is that the BSE crisis cannot be studied in a vacuum and the Southwood Working Party is not some atypical artefact—a scientific body floating in a sea of politics. Rather, to make sense of both the BSE crisis and the Working Party, an understanding is needed of the institutional context to see what understandings of administrative constitutionalism were in operation. To do that, one needs to understand the history of administrative constitutionalism in the UK as well as understand how theories of administrative constitutionalism influenced 36 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at para 1.1. 37 See sections III.A and III.B in ch 1.

(D) Fisher Ch2

31/5/07

66

09:57

Page 66

BSE, Expertise, and Administrative Constitutionalism

the institutional context in which the Southwood Working Party was operating. Both these issues are considered in the next two sections.

II Technological Risk Regulation and Administrative Constitutionalism in the UK: A Brief History Expertise has always been a central issue in debates over UK administrative constitutionalism.38 Much of this has to do with the fact that the development of technological risk regulation and administrative constitutionalism occurred simultaneously and the need for expertise in technological risk evaluation highlighted the problematic nature of these new non-elected institutions. There was no tradition of central state intervention in the UK before the mid-nineteenth century and the first major examples of it occurred in the technological risk regulation field due to the adverse side effects of industrialisation and urbanisation.39 New administrative bodies were created, such as the Factories Inspectorate, the Public Health Board, the Railways Inspectorate, and the Alkali Inspectorate. In the latter part of the century this was added to by more comprehensive regimes for generic pollution40 and food safety.41 The key question in the development of these bodies was what their role and nature should be, and in particular what should be the role and nature of expertise within such bodies. Architects of these new regulatory schemes, such as the Benthamite Edwin Chadwick, largely justified and conceptualised public administration in RI terms.42 For this group, public administration was an expert body acting on the basis of science in relation to specific issues.43 Expertise and science were synonymous, technological risks were discrete isolated problems, and the basis for regulatory action was extensive statistical analysis.44 Most of the early 38 R Macleod (ed), Government and Expertise: Specialists, Administrators, and Professionals 1860–1919 (Cambridge, Cambridge University Press, 1988); J Roberts, ‘The Professional Expert and Administrative Control’ (1929) 7 Public Administration 247; and CP Snow, Science And Government (London, Oxford University Press, 1961). 39 For general discussions on these developments see O MacDonagh, ‘The Nineteenth Century Revolution in Government: A Reappraisal’ (1958) 1 Historical Journal 52; and D. Roberts, Victorian Origins of the British Welfare State (New Haven, Yale University Press, 1960). 40 Rivers Pollution Prevention Act 1876. 41 Food and Drugs Act 1860; Pharmacy Act 1862; and Sale of Food and Drugs Act 1875. 42 R Lewis, Edwin Chadwick and the Public Health Movement 1832–1854 (London, Longmans, Green & Co, 1952); C Hamlin, Public Health and Social Justice in the Age of Chadwick, Britain 1800–1854 (Cambridge, Cambridge University Press, 1998); and S Finer, The Life and Times of Sir Edwin Chadwick (London, Methuen & Co Ltd, 1952). 43 D Eastwood, ‘Amplifying the Province of the Legislature: The Flow of Information and the English State in the Early Nineteenth Century’ (1989) 62 Historical Journal 275; G Mooney, ‘Professionalisation in Public Health and the Measurement of Scientific Progress in Nineteenth Century England and Wales’ (1997) 10 Social History of Medicine 53, and Macleod (ed), above n 38. 44 E Chadwick, Report on the Sanitary Condition of the Labouring Population of Great Britain (Edinburgh, Edinburgh University Press, 1964) at 382 and 392. Also see parliamentary debates

(D) Fisher Ch2

31/5/07

09:57

Page 67

Technological Risk Regulation & Administrative Constitutionalism in the UK 67 administrative experiments were on the basis of the RI paradigm, and legislation stated precise rules for an inspectorate or local body to enforce with very little discretion given to them in that process of enforcement.45 Juxtaposed against the promotion of public administration grounded on the RI paradigm were those who argued that the role of public administration, and in particular the role and nature of expertise, should be understood in DC terms. This view was taken for a number of reasons. For some, particularly those working inside the new RI regimes,46 such a position was taken because they appreciated that technological risks were more uncertain and socio-politically embedded than suggested by those who promoted the RI paradigm.47 Others promoted the DC paradigm because for them DC ideals embodied the traditions of local government, and thus were an important bulwark against what they saw as the RI tendencies of centralisation which resulted in ‘universal obedience to the pedantic schemes of a few closet theorists’.48 In both cases, the argument was not that expertise was not required in the process of risk evaluation but that the term should also refer to wisdom and experience, and expertise should be recognised as being only one input into decision-making. This DC concept of expertise gained legislative expression in the 1874 amendments to the Alkali Acts that introduced the concept of ‘best practicable means’.49 This was a far more flexible and ambiguous concept for inspectors to enforce rather than the quantitative limits of the earlier Acts. This type of thinking underpinned risk regulation, and in particular enforcement,50 throughout the twentieth century and was further strengthened in the 1970s in the Robens Report on Health and Safety at Work and by the Royal Commission on Environmental Pollution in its report on air pollution control.51 Where the DC paradigm can be seen to be promoted most aggressively, however, was in relation to the concept of the generalist civil service. This was an ideal concerning public health legislation during this period: Hansard, HC Deb, vol 91 (3rd Ser), cols 618–23 (30 March 1847); and Hansard, HL Deb, vol 96 (3rd Ser), cols 394–402 (10 February 1847). 45 Section 4 of the Alkali Act 1863. 46 See especially Chief Inspector of the Alkali Inspectorate, Second Report of the Alkali Inspectorate (London, HMSO, 1866) at 8; Chief Inspector of the Alkali Inspectorate, Third Report of the Alkali Inspectorate (London, HMSO, 1867) at 48–53; Chief Inspector of the Alkali Inspectorate, Fourth Report of the Alkali Inspectorate (London, HMSO, 1868) at 5. 47 This was one of the reasons for the failure of the Public Health Board in 1854. See Hamlin, above n 42 at 332; and J Brand, Doctors and the State: The British Medical Profession and Government Action in Public Health, 1870–1912 (Baltimore, Johns Hopkins Press, 1965) at 7. 48 J Toulmin Smith, Local Government and Centralisation (London, John Chapman, 1851) at 56; and J Toulmin Smith, Government By Commissions: Illegal and Pernicious (London, Sweet, 1849) at 292. 49 Alkali Act 1874, s 5. 50 K Hawkins, Environment and Enforcement (Oxford, Oxford University Press, 1984). 51 Committee on Safety and Health at Work, Health and Safety at Work (London, HMSO, 1972) at 35, 69–70; and Royal Commission on Environmental Pollution, Air Pollution Control: An Integrated Approach, Fifth Report, Cmnd 6371 (London, HMSO, 1976) at 4. This was reiterated in Royal Commission on Environmental Pollution, Best Practicable Environmental Option, Twelfth Report, Cmnd 310 (London, HMSO, 1988).

(D) Fisher Ch2

31/5/07

68

09:57

Page 68

BSE, Expertise, and Administrative Constitutionalism

that had its roots in the nineteenth century as well and became the organising ethos of UK public administration throughout the twentieth century.52 The Northcote-Trevelyan report of 1853 has invariably been seen as the founding document of that ethos. In that report, the generalist civil servant was promoted over the specialist.53 Training in specific subjects, it was thought, would result in a narrow mindset and the favouring of ‘mechanical’ over ‘intellectual’ behaviour.54 Departmental promotion would: cramp the energies of the whole body, to encourage the growth of narrow views and department prejudices, to limit the acquisition of experience and to repress and almost extinguish the spirit of emulation and competition . . .55

In particular, the report emphasised that the authority of the civil service did not derive from rational knowledge but rather from its ability to reason and deliberate. This DC thinking about public administration, and the role and nature of expertise within it, was ultimately reflected in the way in which the British administrative state grew throughout much of the twentieth century. Public administration was constituted as a permanent deliberative institution and its purpose was the addressing of complex problems. The promotion of this understanding of administrative constitutionalism can be seen in three ways. First, administrative departments were constituted, limited, and held to account in DC terms. Administrative departments were constituted (usually through the use of prerogative powers) on a broad basis and were given wide-ranging legislative powers. Attempts to rationalise or systematise this state of affairs were seen as getting in the way of the complex business of administrative governance.56 Judicial review, often viewed as too deferential,57 operated on the basis that administration needed discretionary judgment to deal with complex problems, that decision-makers needed to actively and conscientiously exercise their discretion, and that statutory mandates should be interpreted very broadly.58 Thus, for example, phrases such as ‘risk’, ‘dangerousness’, and ‘safety’ were not interpreted by courts as factual concepts.59

52 See generally P Hennessy, Whitehall (London, Pimlico, 2001) at 31–51; H Parris, Constitutional Bureaucracy: The Development of the British Central Administration Since the Eighteenth Century (London, George Allen & Unwin Limited, 1969) at chs 3 and 4; and Ministry of Reconstruction, Report of the Machinery of Government Committee, Cd 9230 (London, HMSO, 1918). 53 Fulton Committee, The Civil Service—Volume One, Cmnd 3638 (London, HMSO, 1968) at Appendix B. 54 Ibid at 115. 55 Ibid at 110. 56 Committee on Ministers’ Powers, Report of Committee of Ministers’ Powers, Cmd 4060 (London, HMSO, 1932); L. Scott, ‘Evolution of Public Law’ (1932) 14 (3rd Ser) Journal of Comparative Legislation 163 at 170; and Statutory Instruments Act 1946. 57 P Devlin, ‘The Common Law, Public Policy and the Executive’ (1956) Current Legal Problems 1 58 P Craig, Administrative Law, 5th edn (London, Thompson Sweet & Maxwell, 2003), and E Fisher, ‘Is the Precautionary Principle Justiciable?’ (2001) 13 Journal of Environmental Law 317. 59 R v Board of Trustees for the Science Museum [1993] ICR 876 at 882; Briscoe v Shattock [1999] Env H 108 at 124; and Bolton Municipal Council v Malrod Insulations Ltd [1993] ICR 358 at 369.

(D) Fisher Ch2

31/5/07

09:57

Page 69

Technological Risk Regulation & Administrative Constitutionalism in the UK 69 The second and more significant way in which the DC paradigm permeated British public administration was through the general culture of the permanent civil service and its guiding principles of impartiality, integrity, and objectivity in its non-scientific sense.60 While the central government departments were there to implement government policy, they were not simply the instruments of the governing party. There was not a change of personnel with a change of government, and the civil service was to serve the public interest not just the party interest.61 Problems were understood to be multidimensional and complex.62 The dominance of the generalist administrator was recognition of the complexity of governing, and civil servants were expected to be imaginative, responsive to change, and humane.63 The third way in which the promotion of the DC paradigm can be seen is in the role of experts in public administration, particularly in relation to technological risk regulation and, as in the nineteenth century, expertise remained a pressure point in debates over legitimate public administration.64 While there were some examples of specialist administrations being set up to address specific risks,65 in the main technological risk evaluation was entrusted to large generalist departments that carried out risk evaluation alongside many other tasks.66 This is not to say there was not a role for experts but that their role was as one input into the decision-making process—a fact reflected by their placement on separate civil service grades which did not enable them to progress fully up the administrative hierarchy.67 The relevant Minister and generalist civil servants retained ultimate responsibility for making decisions. Few government science facilities were set up and when research was required, it tended to be carried out through co-ordination by independent research councils68—an arrangement which attempted to harness the rich complexity of scientific activity while ensuring that government science was not the product of ‘tactical or administrative convenience’.69 Scientific advice was also provided by the creation of ad hoc expert committees set up to consider

60 For an interesting discussion of this see D Woodhouse, In Pursuit of Good Administration: Ministers, Civil Servants and Judges (Oxford, Clarendon Press, 1997) at ch 2. 61 Ibid at 32–4. 62 Ministry of Reconstruction above n 52. 63 W Armstrong, The Role and Character of the Civil Service (London, Oxford University Press, 1970) at 1. 64 A Zimmern, ‘Democracy and the Expert’ (1930) 1 Political Quarterly 7; and Fulton Committee, above n 53. 65 Royal Commission on Environmental Pollution, above n 51. 66 Although note that enforcement was often in the hands of local government. 67 P Gummett, Scientists in Whitehall (Manchester, Manchester University Press, 1980) at 70; and R Macleod, ‘Introduction’ in R Macleod (ed), Government and Expertise: Specialists, Administrators and Professionals 1860–1919 (Cambridge University Press, Cambridge, 1988) at 16–17. 68 See Ministry of Reconstruction, above n 52; F Dainton, Doubts and Certainties: A Personal Memoir of the Twentieth Century (Sheffield, Sheffield Academic Press, 2001) at ch 8; and Gummett, above n 67 at 20–29. 69 F Dainton, The Future of the Research Council System (The BSE Inquiry Report, Evidence M19 tab 2B, 1971) at 10–12.

(D) Fisher Ch2

31/5/07

70

09:57

Page 70

BSE, Expertise, and Administrative Constitutionalism

specific issues.70 That expertise was defined in broad, deliberative terms and the expertise of the official scientific advisor lay not in a ‘monopoly of scientific wisdom’ but rather in practical experience in the way in which government worked.71 Knowledge and expertise were required to be drawn from the wider community, and while distinguishable from mainstream decision-making, needed to be properly integrated into it.72 All of this reflected the fact that the basis for administrative action was deliberative reason not scientific analysis—science was just one input into the process of deliberative reasoning and it needed to be ensured that experts were, in the terms of the clichéd expression of the time, ‘on tap not on top’. The dominance of the DC paradigm did not go unchallenged, however. In regard to general pubic administration, there was, from the late 1970s onwards, a promotion of ideas of good administration clearly based on the RI paradigm. In part these developments were a direct product of problems encountered in technological risk evaluation, such as the controversy over salmonella in eggs in the late 1980s.73 There was a perception that standard setting in the risk regulation area was secretive, inconsistent, and that it was difficult to know whether standards were due to careful deliberation or a product of ‘fudge and smudge, a quagmire of intellectual fuzziness . . . and administrative laxity’.74 In other words, the deliberative capacities of the permanent civil service were starting to be doubted. The promotion of the RI paradigm was also the product of two other closely related forces. First, there was a loss in faith in the concept of the generalist civil service. The civil service was increasingly viewed as elitist and its claims to professionalism seen as a means of protecting self-interest. Moreover, many understood it as an unaccountable ‘state within a state’.75 From the 1968 Fulton Report onwards there was an ongoing challenge to the validity of the notion of the generalist civil service and an argument for public administration to be more specialist in nature.76 In 1985 the Armstrong Memorandum appeared to rewrite the role of the civil service in RI terms with its statement that the civil service should serve the government of the day.77 Second, the promotion of the RI paradigm was also due to a new ideology which could be traced back to the late 1960s. The Fulton Report was based on a more 70 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at ch 4. 71 S Zuckerman, Advice and Responsibility: Romanes Lecture for 1975 (Oxford, Clarendon Press, 1975) at 33. 72 Ministry of Reconstruction, above n 52. 73 See Agriculture Committee, above n 28 . 74 Rose, as quoted in A Jordan, ‘The Impact on UK Environmental Administration’ in P Lowe and S Ward (eds), British Environmental Policy and Europe (London, Routledge, 1998) at 183. 75 B Castle, ‘Mandarin Power’ in P Barberis (ed), The Whitehall Reader (Buckingham, Open University Press, 1996) at 61. See also Woodhouse, above n 60 at 37–8. These developments were linked to more general attacks on the concept of professionalism. See E Freidson, Professionalism Reborn: Theory, Prophecy, and Policy (Cambridge, Polity Press, 1994). 76 Fulton Committee, above n 53. 77 Cabinet Office, ‘The Armstrong Memorandum’ in P Barberis (ed), The Whitehall Reader (Buckingham, Open University Press, 1996) at 114.

(D) Fisher Ch2

31/5/07

09:57

Page 71

The Administrative Constitutionalism Context of the BSE Crisis

71

instrumental vision of administration,78 and in 1972 the Rothschild Report on the Organisation and Management of Government R&D recommended that research commissioned by government should not be on the historical ‘arm’s length’ terms that it had been, but on a ‘customer–contractor’ basis.79 The role of research was to serve particular purposes rather than being a more public-spirited enterprise. RI reforms increased with the coming into power of the Conservative government in 1979.80 Much of their reform was about remodelling the civil service to be more like private sector institutions,81 with generalist administrators being replaced with trained managers.82 The logic behind this was that such reforms would constrain unelected administration and result in an elected government gaining greater control over it. The actual reforms put in place by the Conservative government were numerous and included downsizing, budget management, privatisation, deregulation, and institutional reforms.83 Likewise, there was constant pressure to reduce the scientific research budget by scaling down research activities and encouraging industry to do the research.84

III The Administrative Constitutionalism Context of the BSE Crisis What is clear from the last section is that by the mid-1980s understandings of administrative constitutionalism were in a state of flux. This can be seen even more clearly when one looks at the specific administrative/regulatory context in which the BSE crisis arose. The two main administrative institutions involved in making decisions about BSE were MAFF and DH, with the former taking the lead role. Generally speaking, both departments were organised along DC lines in which generalist administrators dominated, as did the ethos of the permanent generalist civil service.85 Technological risk evaluation was only one of a number of responsibilities of these large sprawling departments, which had a three-pronged 78

Fulton Committee, above n 53. Lord Rothschild, The Organisation and Management of Government R&D (BSE Inquiry Report, Evidence M19 tab 2A, 1972). 80 Although note the historical antecedents of this in the Fulton Report. See Fulton Committee, above n 53. 81 See generally D Osborne and T Gaebler, Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector (New York, Plume, 1992); and C Foster and C Plowden, The State Under Stress (Buckingham, Open University Press, 1996). 82 Foster and Plowden, above n 81 at 175–8. 83 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at chs 6 and 7; Efficiency Unit, Improving Management in Government: The Next Steps (London, HMSO, 1988); Hennessy, above n 52 at ch 14; and Foster and Plowden, above n 81 at ch 8. 84 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Two: Science, above n 12 at paras 6.47–6.50. 85 For a description see Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at ch 2. 79

(D) Fisher Ch2

31/5/07

72

09:57

Page 72

BSE, Expertise, and Administrative Constitutionalism

role: providing advice, formulating policy, and implementing that policy.86 Such departments were also responsible for an array of different matters, and MAFF, for example, was concerned both with the protection of health and the promotion of interests of the agricultural sector. This was a combination that many have seen as leading to a direct conflict of interest,87 but not problematic from a DC perspective because of the ability of deliberation to diffuse the power of self-interest.88 The administrative powers given to these departments by statute were in very broad DC terms. Ministers and their departments were entrusted with wide-ranging powers to deal with complex problems as they arose. To this end, most legislation granted significant powers to pass delegated legislation to address issues. In relation to BSE, MAFF and DH had relevant powers under the Animal Health Act 1981, the Slaughterhouse Act 1974, the Food Act 1984, and the Agriculture Act 1984.89 Thus, for example, under section 1 of the Animal Health Act 1981: The Ministers may make such orders as they think fit— (a) generally for the better execution of this Act, or for the purpose of in any manner preventing the spreading of disease; and (b) in particular for the several purposes set out in this Act, and for prescribing and regulating the payment and recovery of expenses in respect of animals.

This is a good example of legislation being based on a DC understanding of administrative constitutionalism. The Act entrusted the Minister with responsibilities to address particular problems and did not limit such power unnecessarily in recognition of the complexity of such problems. Moreover, nowhere did the Act regulate the process of technological risk evaluation or require a factual threshold. Likewise, section 13(1) of the Food Act 1984 stated: The Ministers may make such regulations as appear to them expedient for securing the observance of sanitary and cleanly conditions and practices in connection with— (a) the sale of food for human consumption, or (b) the importation, preparation, transport, storage, packaging, wrapping, exposure for sale, service or delivery of food intended for sale or sold for human consumption, or otherwise for the protection of the public health in connection with those matters.

Subsection (2) listed illustrative examples of the type of regulations which may be passed under this provision but which did not limit the power of the Minister. The Act also allowed delegated legislation to amend it (what is known as a Henry VIII clause) so long as there was a consultation process.90 86 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at chapter 2 and annex one. 87 van Zwanenberg and Millstone, above n 7 at 40–65. 88 See section III.B in ch 1, this volume; and C Sunstein, ‘Interest Groups in American Public Law’ (1985) 38 Stanford Law Review 29. 89 For a lengthy discussion of the legal framework, including other relevant pieces of legislation, see Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fourteen: Responsibilities for Human and Animal Health (London, HMSO, 2000). 90 Section 118.

(D) Fisher Ch2

31/5/07

09:57

Page 73

The Administrative Constitutionalism Context of the BSE Crisis

73

A common perception about the BSE crisis was that decision-makers were operating in an accountability vacuum.91 From a strictly legal perspective, this is perhaps true—judicial review was not only deferential but there were few cases in which decisions had been challenged. The constraints of the Statutory Instruments Act 1946, the statute that regulated the passing of delegated legislation, were also minimal.92 Likewise, established mechanisms such as ministerial accountability and accountability to Parliamentary Select Committees, were perceived to be ineffectual.93 Indeed, in the early years of the BSE crisis there was little discussion about BSE in Parliament.94 Despite this, the threat of such mechanisms did influence decision-makers’ actions.95 Moreover, decision-makers were also being held to account in numerous other ways. As such, they were operating in a context in which they were subject to multiple accountabilities, albeit very few of them having a legal status. Departments were financially accountable to a number of different bodies,96 and operating under internal management mechanisms, individual appraisal systems, budgeting systems, and the Deregulation Initiative. Nearly all these forms of accountability required decision-makers to be more RI in nature, allowing them to act only on the basis of evidence. The Deregulation Initiative was particularly important in this regard as it required departments to justify their regulatory action by proving the factual basis for such regulation on the basis of cost/benefit analysis or risk assessment.97 One official involved in the crisis described the Initiative not as stopping regulation but required that it was ‘tightly drawn’.98 This tight drawing was largely on an RI basis—decision-makers proving that they were only carrying out the regulatory tasks assigned to them. Where administrative constitutionalism can be seen to have its greatest influence, however, was in relation to the way in which MAFF and DH were constituted. As already noted, the institutional culture in central government departments at this time was primarily DC in nature. Decisions and delegated legislation were the product of a ‘long and tortuous process’ in which the emphasis was on deliberation and the putting across of a number of different points of view.99 As one MAFF official put it, policy decisions required:

91 van Zwanenberg and Millstone, above n 7 at Conclusions; Little, above n 33; and P James, Food Standards Agency: An Interim Proposal (1997). 92 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at paras 3.48–3.50. 93 This was particularly seen in the ‘Arms to Iraq’ controversy involving events that occurred at roughly the same time: A Tomkins, The Constitution After Scott: Government Unwrapped (Oxford, Oxford University Press, 1998). 94 J MacGregor, Statement of Evidence to the BSE Inquiry (The BSE Inquiry/Statement 302, 1998) at 4. 95 Treasury Solicitor, The Judge Over Your Shoulder: A Guide to Judicial Review for UK Government Administrators, 3rd edn (London, Treasury Solicitor’s Department, 2000). 96 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at paras 6.5–6.9. 97 Ibid at ch 7. 98 Ibid at para 7.13. 99 Ibid at 10. Also see Woodhouse, above n 60 at 34 for a lengthier analysis of this.

(D) Fisher Ch2

31/5/07

74

09:57

Page 74

BSE, Expertise, and Administrative Constitutionalism a build-up of available information, discussion amongst officials either orally or through correspondence, perhaps preliminary discussion with Ministers, [written] submissions to Ministers with argument and options, discussion with Ministers, further discussion between Ministers, decisions and an action timetable.100

Most of that process occurred within the executive, with the involvement of the Minster only at a later stage.101 The aim was for a frank and wide-ranging consideration of issues with input from numerous different sources, albeit that conversation only involving a small group.102 At the same time, by the late 1980s a more RI understanding of the role of the civil service was becoming apparent and civil servants began to see that the giving of advice was less in line with the tradition of the permanent civil service and more providing the information that would support an already decided-upon policy position or course of action. Thus a MAFF official stated that advice would depend on the Minister’s thinking, and that advice would not be given ‘if there was very little chance of it being accepted’.103 This development was noted by others and became more apparent throughout the 1990s. As Woodhouse described it, what was seen was the rise of a ‘rational choice or managerial model of good administration’.104 It was primarily an RI development—the civil service was being transformed from a permanent DC institution to an RI agent of the ruling party. This set of changes can be seen in relation to what was understood to be the role and nature of expertise and information. Both were understood to have important roles to play in administration, particularly risk evaluation, but the role and nature was an increasingly bifurcated one. From one perspective, the role and nature of expertise was understood in DC terms as one input into decision-making rather than its source of authority.105 Within both departments it was seen as important to have generalist and specialist administrators working together in a ‘cooperative’ problem-solving process in which the role of each would depend on the nature of the problem.106 Those who headed the specialist branches also saw themselves in DC terms. In MAFF, the Chief Veterinary Officer (CVO) saw his role as requiring him to engage in open, frank discussion with generalist civil servants in which there was ‘reliance’ on the judgement of each.107 In DH, the Chief 100 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at para 3.4. 101 Ibid at para 3.5. This can be seen in relation to all BSE decision-making. 102 Ibid at para 3.6. See also the deliberation in relation to the slaughter and compensation policy. See Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years, above n 19 at ch 5. 103 Cruickshank, quoted in Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at para 3.7. 104 Woodhouse, above n 60 at ch 3. 105 A DH internal management review, quoted in Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at para 4.2. 106 Ibid at paras 4.31–4.39 and 4.45–4.54; and D Andrews, Statement of Evidence to the BSE Inquiry (The BSE Inquiry Evidence/Statement 281, 1998) at para 19. 107 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at para 4.16.

(D) Fisher Ch2

31/5/07

09:57

Page 75

The Administrative Constitutionalism Context of the BSE Crisis

75

Medical Officer (CMO) was a figure of professionalism, acting in the public interest, and unswayable by private interest.108 Expertise was thus broad based and not simply confined to the methodologies of a discipline. Experts were involved in policy-making, which was understood as a multidisciplinary exercise.109 Moreover, during the early years of the BSE crisis, quantitative methodologies were not seen as necessary in the risk evaluation process.110 While such methods were seen as relevant to areas where knowledge allowed their use, the limits of such tools were recognised in areas where there were considerable scientific uncertainties.111 Rather, the concepts of reasonableness and proportionality were understood as broad guiding principles, as was general precautionary thinking.112 Likewise, the concept of ‘best practicable means’, developed in the nineteenth century, had continued to be the basis for action and also given rise to associated concepts such as ‘as low as reasonably practicable’ (ALARP).113 All this flowed from an understanding that technological risk problems were complex and thus would not yield easily to strict methodological approaches. At the same time, however, the role and nature of expertise was being transformed from being understood in DC terms to being grounded in an RI understanding of public administration. This can be seen in each of the three main ways in which expertise played a role in the decision-making process in relation to BSE. The first of these, ad hoc scientific committees, will be discussed in the next section. The second form in which expertise was utilised was in the commissioning of research, and this was done on the customer/contractor basis required by the Rothschild Report, with the government department acting as an ‘intelligent customer’.114 Here the concern was less with expert advice and more with specialist information and the obtaining of information was increasingly viewed in instrumental terms, both due to the customer/contractor concept and also due to limits in budgets for research which required departments to justify expenditures in this area.115 Thus, for example, in 1985 an independent body advising the UK government concluded that research into animal diseases was disproportionate and it should be reduced.116 This advice rested on the assumption that the role and nature of public administration could be pre-defined, an assumption that the CVO saw as problematic due to the way in which MAFF needed to be able to respond quickly to the emergence of new diseases.117 Third, there was very limited ability for in-house research by specialist administrators. Thus, for example, in MAFF the State Veterinary Service (SVS), headed 108

Ibid at paras 4.24–4.25 Ibid at paras 4.50 and 4.55. 110 Ibid at ch 5. 111 Ibid at paras 5.22 and 5.38. 112 Ibid at paras 5.22 and 5.28. 113 Ibid at para 5.9. 114 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Two: Science, above n 12 at paras 6.11–6.13. 115 Ibid at paras 6.42–6.50. 116 Ibid at para 6.49. 117 Ibid at para 6.49. 109

(D) Fisher Ch2

31/5/07

76

09:57

Page 76

BSE, Expertise, and Administrative Constitutionalism

by the CVO, included a number of different groups of specialists including the Central Veterinary Laboratory (CVL). Those within the CVL saw it as a scientific unit with no independent power of its own, and as such, in RI terms. As an administrative body, it was required to carry out the research requested of it, and the reasonableness of its action was understood in terms of the quality of its research.118 The problems with this understanding were identified by Richard Packer, the permanent secretary within MAFF for part of the BSE crisis. He noted, in discussing the CVL, that ‘vets, like other professionals, do not think like civil servants and are reluctant to involve others in decisions they regard as falling within their remit’.119 Such a comment harks back to the Northcote-Trevelyan report.

IV The Southwood Working Party What can be seen from the last section is that the decision-making context in which MAFF and DH were operating was a profoundly complex one and while it was recognised that action should be ‘appropriate, proportionate, legally feasible and, in practice, achievable’,120 this could mean two different things. This state of affairs was due to the understandings of administrative constitutionalism operating at the time and what this meant was that the concept of expertise was increasingly bifurcated. Expertise was simultaneously being understood as a deliberative input into decision-making and as the analytical foundations for decisionmaking. The ambiguous nature of expertise can be seen in how the Southwood Working Party came to be set up, how it operated, and how its report was received. The first case of BSE was identified by the CVL in 1986, although it took a number of months for it to be confirmed authoritatively as a TSE.121 The CVL’s characterisation of BSE was influenced by its RI understanding of its own role.122 For those in the CVL, a lack of definitive evidence meant that they could not rule on the nature of the new disease, and the CVL only started to disclose information about BSE to others within government once they felt the evidence justified it.123 Moreover, while the possibility of health risks was identified early on, these were not seen as matters for the CVL so it did not pursue them.124 MAFF Ministers were informed about BSE in June 1987 in a memo sent to them by the CVO in which he stated: 118 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Two: Science, above n 12 at para 6.116. 119 Packer, above n 7 at 35. 120 The Head of MAFF’s Food Safety Directorate, as quoted in Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at para 3.1. 121 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years, above n 19 at ch 1. 122 As quoted in ibid at para 1.37. 123 Ibid at paras 2.99 and 4.9. 124 Ibid at para 5.6.

(D) Fisher Ch2

31/5/07

09:57

Page 77

The Southwood Working Party

77

There is no evidence that the bovine disorder is transmissible to humans. In the absence of such evidence, and in the absence also of the epidemiological knowledge or of a definitive test to establish the disorder’s presence in a live host, it does not seem appropriate to impose restrictions on affected farms or on the sale of produce from cattle in affected herds.125

Senior civil servants on receiving the memo did not interpret the statements of there being ‘no evidence’ of health risks meaning that there was no risk or that they should only act on the basis of scientific advice and/or evidence.126 The discourse among the senior civil servants after this memorandum was very much DC in nature. First, civil servants focused upon the possibility that there might have been health risks arising from BSE, and they deliberated about the nature of the uncertainties involved and how one would go about establishing whether there was a health risk.127 They planned a programme of research,128 but also saw their role as needing to take precautionary action even if such a risk could not be strictly proved.129 An options paper was circulated among MAFF officials in early 1988 which reflected wide-ranging deliberations.130 There was very little in that paper that focused on issues of evidence, even though BSE remained a little understood disease. In February 1988 a submission was made to the MAFF Minister, John MacGregor, about what action should be taken in relation to BSE. In that submission, senior officials recognised the considerable scientific uncertainties involved, others issues such as welfare and trade, and MAFF’s responsibility in relation to human health.131 Before that submission was given to the Minister it was circulated among MAFF officials with a covering noted that stated that: We do not know where this disease came from, we do not know how it is spread and we do not know whether it can be passed to humans. The last point seems to me to be the most worrying aspect of the problem. There is no evidence that people can be infected but we cannot say that there is no risk.132

Even though this was the case, the submission recommended that precautionary action of a slaughter and compensation scheme should be taken.133 This scheme was controversial, however, because it was in conflict with the policy that was then in operation to recover costs from industry.134 Moreover, if compensation was to 125

Ibid at para 2.54. See also the statement at para 2.88. Ibid at paras 2.57 and 5.20. 127 Ibid at paras 2.85–2.89, 5.20, 5.30, 5.37, and 5.44. 128 Ibid at paras 2.57 and 5.20. 129 Andrews, above n 106 at 34; Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years, above n 19 at ch 5 and Annex 2. This stance was also taken among those in the CVL: see BSE Meeting, 9 am on 8 January in CVO Room (BSE Inquiry Yearbooks, YB 88/-1/04/1.1-1.4, 4 January 1988). 130 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years, above n 19 at para 5.40. 131 Ibid at paras 5.43–5.55. 132 Ibid at para 5.44. 133 Ibid at para 5.49. 134 Ibid at paras 5.49–5.55; and Packer, above n 7 at 37–8. 126

(D) Fisher Ch2

31/5/07

78

09:57

Page 78

BSE, Expertise, and Administrative Constitutionalism

be paid, MAFF would need to show that such expenditure would yield benefits— a difficult thing to do in circumstances of scientific uncertainty but logical in a context in which the Treasury was attempting to control the administrative state.135 The submission was put to the relevant Minister, John MacGregor, with accompanying advice from the Permanent Secretary, which stated: I do not see how you could defend taking no action now unless you had the support of the Chief Medical Officer. But, on the face of it, it seems unlikely that he would feel able to endorse a wholly reassuring statement of the likely risks of transmission of this disease to man until we have much more information available.136

However, MacGregor made it clear that action would not be taken without seeking the advice of the CMO.137 This stance was due to the fact that he understood that reasonable administrative action would need to be based on expert advice and/or evidence.138 His reasons for taking this stance was that it was consistent with the RI paradigm which was being promoted by the Treasury’s budgetary policies, the Deregulation Initiative, and by the perception that evidence was needed to withstand any judicial review challenges.139 Moreover, he saw a direct analogy between BSE and rhizomania in sugar beet and was concerned that paying compensation in relation to the former would require compensation in relation to the latter.140 This was a simplistic understanding of both problems, and ignored the serious health implications of BSE.141 The CMO, on being informed, recognised the problems of assessing the potential health risks, the need to investigate the issue further, and the need to seriously consider regulatory action. He recommended the setting-up of an expert committee to give advice on the matter.142 In evidence to the BSE Inquiry, the CMO suggested that this recommendation was for a number of reasons, including scientific uncertainty; that the CMO giving such advice ‘off the cuff’ was inappropriate; and that expert advice would be needed to justify a slaughter and compensation scheme to the Treasury.143 In other words, a committee was needed for both DC and RI reasons.144 The Committee that was set up was the Southwood Working Party. 135 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years, above n 19 at paras 5.51 and 5.52. 136 Ibid at para 5.53. For an account see Packer, above n 7 at 37 137 Packer, above n 7 at 37. 138 MacGregor, above n 94 at 6 and 19; and van Zwanenberg and Millstone, above n 7 at 88. 139 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Six: Human Health, above n 3 at para 3.159; and Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years, above n 19 at paras 5.58–5.63. 140 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Three: The Early Years, above n 19 at para 5.58 141 Ibid at para 5.179. 142 Ibid at para 5.67. 143 Ibid at paras 5.65, 5.157, and 5.208. 144 See Lord Phillips’ questioning of Sir Derek Andrews in relation to issue of whether different reasons for setting up expert committees may exist: The BSE Inquiry, Oral Transcript, 10 November 1998 at 10.

(D) Fisher Ch2

31/5/07

09:57

Page 79

The Southwood Working Party

79

The group met for the first time on 20 June 1988 and it reported in February 1989. Before this first meeting, MAFF had banned the use of ruminants in ruminant feed and made BSE notifiable.145 Straight after their first meeting the Working Party made interim recommendations that the already suggested compulsory slaughter and compensation scheme for animals should be introduced and that a research committee should be set up. These recommendations were both accepted by the government.146 In their final report, they recommended that the regulatory authorities should address the possible health risks from medicines and occupational exposure as well as recommending a ban on the use of thymus and offal in baby food.147 My concern is not with what the Working Party recommended, however, but what was understood to be their role and nature in the risk evaluation process. As can be seen above, the Working Party was operating in a context in which both the DC and RI paradigms were in operation, and thus it should come as no surprise that what was understood as the role and nature of such an expert body was bifurcated. As seen in the last section, the Working Party was created mainly because it was perceived that regulatory action could not be taken without evidence and/or expert advice. This RI understanding of reasonable administrative action was a departure from traditional DC ideas, but the actual establishment and operation of the Working Party was along DC lines. This can be seen in what was understood to be the expertise of the Working Party, how it operated, and what it perceived its role to be in risk evaluation. In regard to the first issue, the expertise of the Working Party was initially understood in very broad terms. As Jasanoff has put it, the committee was staffed by the ‘great and the good’, whose authority derived not from their specific disciplinary expertise but their ability to exercise ‘caution, empiricism, and constraint’.148 The Working Party consisted of four eminent research scientists who did not have expertise in TSEs but rather more general expertise in the area.149 This was seen as desirable because it was felt that those who had direct expertise in TSEs would be ‘too close to the front line to take the slightly broader view needed’.150 In asking Sir Richard Southwood to chair the Working Party, the CMO said that Southwood was appropriate because ‘it is an ecological food chain problem [Southwood’s area of expertise] and because I know you are an independent chap’.151 In oral evidence to the BSE Inquiry, Southwood said he did not see the Working Party as ‘experts in the narrow sense of carrying out research into BSE’, and it was for this reason that the committee was called a Working 145

Bovine Spongiform Encephalopathy Order 1988, SI 1988/1039 Bovine Spongiform Encephalopathy (Amendment) Order 1988, SI 1988/1345, and Bovine Spongiform Encephalopathy Compensation Order 1988, SI 1988/1346. 147 Southwood et al, above n 1. 148 S Jasanoff, ‘Civilization and Madness: The Great BSE Scare of 1996’ (1997) 6 Public Understanding of Science 221 at 227–8. 149 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9 above n 6 at paras 1.9–1.15. 150 Ibid at para 1.12. 151 R. Southwood, Statement to BSE Inquiry (The BSE Inquiry/Statement No 1, 1998) at para 2. 146

(D) Fisher Ch2

31/5/07

80

09:57

Page 80

BSE, Expertise, and Administrative Constitutionalism

Party.152 This is not to say that the Working Party ignored those with expertise in TSEs, but rather they consulted them, particularly those who had long experience in the area.153 The broadness of the Working Party’s expertise was not only in terms of scientific expertise, however, but also in the fact that most of those on the Working Party had experience on other government committees and thus understood their task in broad public interest terms.154 In particular, they saw that their deliberations, while taking account of the policy context, should not be influenced by specific interests.155 This is consistent with the DC ideals of the civil service and an appreciation that technological risks are not easily manageable. The second way in which the expertise of the Working Party accorded with the DC paradigm is in the deliberative way that the Working Party conducted itself. The Committee saw itself as a ‘scoping committee’ that needed to see what ‘pointers they could discover’ about BSE.156 As this was the case, the fact that they did not have a grasp on all issues was not necessarily problematic. They saw themselves as the first, not the last, word on the problem. Because of their need to ‘scope’, those on the committee felt that very broad terms of reference were important because ‘one could not have been restricted and operated properly’.157 The Working Party needed to take an expansive perspective on BSE and all the problems it might create. Indeed, the Working Party’s inquiries and deliberations covered a range of issues, including: the nature of the disease itself; problems inherent in ruminant feed; health risks to animals; and human health risks arising from medicines, food, and occupational exposure. In so doing, those on the Working Party were acutely aware that they were operating in a context where there were many scientific and behavioural uncertainties.158 Thus, for example, Southwood, in discussing these issues with MacGregor, the MAFF Minister, stressed that the Working Party was operating in ‘uncharted waters’.159 The Working Party consulted formally and informally with an array of people albeit within a reasonably small circle.160 Deliberations were supported by advice and information resources from DH and MAFF.161 The Working Party did not carry out quantitative risk assessments but did operate on the ALARP principle, although not in a systema152

BSE Inquiry, Oral Transcript, 11 March 1998 at 6. Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9, above n 6 at para 1.12. 154 Southwood, above n 151 at para 17; and Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9, above n 6 at paras 1.9 and 1.21. 155 Southwood, above n 151 at 21. 156 BSE Inquiry, Oral Transcript, 11 March 1998 at 12. 157 BSE Inquiry, Oral Transcript, 11 March 1998 at 14. See also Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9, above n 6 at para 1.8. 158 Ibid at para 6.7; Southwood, above n 151 at paras 30–31; BSE Inquiry, Oral Transcript, 11 March 1998 at 13. 159 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9 above n 6 at para 6.6. 160 Ibid at paras 3.5 and 5.13 161 Ibid at para 1.27. 153

(D) Fisher Ch2

31/5/07

09:57

Page 81

The Southwood Working Party

81

tised way162 The focus was far more on deliberating the scientific issues in light of general principles of proportionality and reasonableness. The third way in which members of the Working Party saw themselves in DC terms is that they saw their report not as an authoritative basis for action but as one input into the deliberative process. Indeed, the bulk of the report was less than 20 pages long (35 pages with appendices) and, while a good overview of the state of knowledge about BSE at that time, it contained little in-depth analytical analysis of data. Most of its analysis was in qualitative terms, and the report was not carefully footnoted. Moreover, as already noted in the introduction, the conclusions in the report about human health contained an important caveat about uncertainties. Throughout the report they stressed the problems of scientific uncertainty,163 problems created by the long incubation of the disease,164 and the need to do more research.165 The scientific and evidentiary issues were discussed in terms of theories, assumptions, and hypotheses.166 The report reads as a judgement on the available information rather than an exercise in proving risks. Indeed, in evidence to the Inquiry, those on the Working Party stressed that their conclusions were based on judgement and that ‘good and wise men and women may reach different sorts of conclusions’.167 Moreover, they hoped that due to the progress in research their report would quickly be out of date.168 As Sir Anthony Epstein, one of the members of the Working Party, stated: we were simply asked to alert them to the implications, and that is what we did. They had their army of civil servants and Government scientists. It was for them to take over at that stage, and decide what policy should be followed by government. It was not the remit of the Working Party to enter into that at all.169

Moreover, some of their conclusions about the nature of BSE were consistent with what MAFF officials had already concluded and were, in the words of the BSE Inquiry, conclusions that would have been ‘shared by any scientists, or indeed intelligent layman’ informed of the issues.170 The problem, however, was that the Southwood Working Party’s report was not treated as one input into an ongoing process of risk evaluation. Rather, it was treated by those in MAFF and DH as if it contained definitive conclusions based on an evaluation of adequate data by expert scientists in relation to the extent both of the risk and of the precautionary measures necessary to counter that risk.171 162

Ibid at para 10.38–10.39. Southwood et al, above n 1 at 3, 5, 7, 11, 13, 14, and 21. 164 Ibid at i and 7. 165 Ibid at i and 13. 166 Ibid at 5, 9, 11, 12, 13, and 14. 167 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9, above n 6 at para 11.3. 168 Ibid at para 11.4. 169 Ibid at para 11.12. 170 Ibid at para 10.9. 171 Ibid at para 1.1. 163

(D) Fisher Ch2

31/5/07

82

09:57

Page 82

BSE, Expertise, and Administrative Constitutionalism

This interpretation of the report, while entirely at odds with a DC understanding of the role and nature of expertise, was very consistent with the concept of expertise embodied in the RI paradigm of administrative constitutionalism. As public administration should only act where rigorous analysis had established a factual basis, then the Working Party’s report took on a new significance. First, the report was interpreted as an authoritative risk evaluation that had concluded that the human health risks from BSE were remote. Until the statement of 20 March 1996, the UK government consistently stated that beef was safe to eat and preventative measures reflected this. Such statements repeatedly focused on the scientific evidence. ‘There is no scientific justification to avoid eating British beef ’172 and ‘there is no scientific evidence of a casual link between BSE in cattle and CJD in humans’173 were common phrases.174 Emphasis was placed on measures being ‘based on the best scientific expertise available’175 and that the most ‘eminent and distinguished scientists’176 had found no risk. The assumption was that the conclusions of the Working Party could not be revisited unless there was new evidence or an improved risk evaluation.177 As such, expert committees also relied heavily on the report until the evidence became such that they felt a new risk evaluation could be undertaken.178 Second, the advice of the Working Party was treated as the primary basis for carrying out administrative and regulatory action. As one MAFF Minister noted, it was a ‘bible’,179 and others saw it as ‘the foundation of government policy’.180 The practical consequence of this was that all regulatory action needed to be justified in terms of the Working Party’s report and that any action that went beyond that advice needed to be strictly justified.181 The main regulatory initiatives that the UK government took in relation to human health—the slaughter and compensation policy and the specified offal ban—were both a direct product of the advice of the Southwood Working Party, as were other more minor regulatory actions.182 Moreover, administrators, even specialist administrators, did not see it as their task to question the report. In a statement to the BSE Inquiry, the CVO at the time of the Southwood Report stated that:

172 MAFF minute sent to health and local authorities on 15 May 1990: Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Six: Human Health, above n 3 at para 4.545. 173 Dr Calman, Chief Medical Officer (11 March 1993): ibid at para 5.331. 174 Also see ibid at paras 4.476 and 4.541–4.577. 175 MAFF press release (185/90): ibid at para 4.561. 176 Meat and Livestock Commission press release (17 May 1990): ibid at para 4.594. 177 For example, ibid at paras 3.179–3.181, 3.243, and 3.266. 178 Packer, above n 7 at 46. 179 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9, above n 6 at para 11.7. 180 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Eleven: Scientists After Southwood (London, HMSO, 2000) at para 1.1. 181 Ibid at para 1.5. 182 For example, the ban on milk from cattle suspected of being infected: see Southwood et al, above n 1 at 10.4; and Bovine Spongiform Encephalopathy (No 2) Order 1988, SI 1988/2299.

(D) Fisher Ch2

31/5/07

09:57

Page 83

The Southwood Working Party

83

the Southwood Working Party had considered the then available scientific evidence (from work on scrapie), had conducted a risk assessment and had concluded that any risk of transmission of BSE to humans was remote. I had no reason to question that conclusion and nor did I have responsibility to do so. Similarly, I believe that my colleagues and Ministers did not question the conclusion that the risk was remote.183 [emphasis supplied]

The CVO saw his task as acting on the basis of evidence. The CVO understood his responsibility as a conduit for information and expertise rather than as a deliberating official. Moreover, he understood the role and nature of the Working Party in rationalistic terms as providing the relevant methodological analysis of evidence. Thus he described their role as carrying out a ‘risk assessment’ even though the Working Party never used that language themselves or purported to carry out any formal risk assessment. This stance on the role of expertise is very different from that seen in the last section, where expert input was seen as the starting point for deliberation rather than the closure of it. In relation to the Working Party’s report there were no significant attempts by MAFF or DH to review it formally even though it was obvious among MAFF and DH officials that there were problems with the advice.184 Thus, for example, it became increasingly clear that the Working Party had not taken into account slaughterhouse practices, an issue which was important in thinking about how any regulations were to apply.185 More significantly, however, one of the report’s recommendations was quickly found to be not well thought-through. In the body of the report, the Working Party had made a recommendation that ruminant offal and thymus should not be used in the manufacture of baby food.186 This recommendation had seemingly been based on the assumption that these parts of cattle may be infective in subclinical animals and, if so, babies would be particularly vulnerable.187 There were a number of problems with this recommendation, however. First, thymus was not used in baby food and it was not clear what was meant by ‘ruminant offal’.188 It quickly became apparent that the recommendation had not been based on a thorough analysis of the issue. MAFF’s and DH’s response to this was to ask Sir Richard Southwood for clarification rather than using the recommendation as a starting point for deliberation.189 Second, a number of officials saw that there was a disjunction between the suggested action to be taken to protect babies and the lack of similar action in relation to children and adults. While babies were clearly more 183 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Six: Human Health, above n 3 at para 3.275. Note that this was a different CVO from the CVO who described his role in DC terms above. 184 Ibid at paras 3.91 and 3.103. 185 Packer, above n 7 at 42. 186 Southwood et al, above n 1 at para 5.3.5 187 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9, above n 6 at para 10.46. 188 Inquiry into BSE and vCJD in the United Kingdom, The BSE Inquiry Report, Volume Six: Human Health, above n 3 at paras 3.28–3.42. 189 Ibid at para 3.51.

(D) Fisher Ch2

31/5/07

84

09:57

Page 84

BSE, Expertise, and Administrative Constitutionalism

susceptible to TSEs, it did not follow that if there was a risk to babies there was no risk for adults.190 Third, the Southwood Working Party had recommended a slaughter and compensation policy for those animals obviously suffering from BSE, a policy based on the assumption that such animals should not be fed to humans. Considering, however, the long incubation period of the disease and the fact that it could be infective in sub-clinical animals, the taking of no precautions in relation to cattle not diagnosed with BSE but which could be infected with it seemed odd.191 These were all issues deliberated and discussed among MAFF and DH officials, many of whom were experts in the area, but despite this, the expert advice of the Southwood Working Party was still seen as the only legitimate basis for discussing the possibilities of regulatory action.192 Ultimately the recommendation on baby food transformed into a wider ban on certain specified offal being consumed by humans.193 Such a ban turned out to be an eminently sensible one and entirely consistent with the application of the traditional DC principle of ALARP.194 The ban, however, was not justified in DC terms. Rather, the ban was seen by most in MAFF and DH as an application of the Working Party’s expert advice, the ban’s wider nature being explained as due to the technicalities of the legal system which meant it was the only way to implement that advice.195 Whatever deliberations that had occurred in MAFF and DH were not seen as the legitimate basis for administrative action, and many did not believe the ban was necessary because there was no basis in the scientific advice for it.196

V After Southwood The focus in this chapter has been on the Southwood Working Party and the bifurcated understanding of the nature and its role. It is useful before concluding to note the developments after the Southwood Working Party. Those developments have largely been RI in nature. This is evident in a number of ways. First, the RI reforms discussed above continued. The Deregulation Initiative quickened pace in the early 1990s, and decision-makers increasingly had to justify regulatory action in terms of risk assessment and regulatory impact assessment.197 Likewise during 190 Inquiry into BSE and vCJD in the United Kingdom, The BSE Inquiry Report, Volume Six: Human Health, above n 3 at paras 3.111 and 3.171. 191 Ibid at para 3.104. 192 Ibid at ch 3. 193 Bovine Offal (Prohibition) Regulations 1989, SI 1989/2061. 194 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Six: Human Health, above n 3 at para 3.276. 195 Ibid at paras 3.32 and 3.278. 196 On the widespread nature of this perception see Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume One: Findings and Conclusions, (London, HMSO, 2000) at paras 232–4. 197 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Fifteen: Government and Administration, above n 31 at ch 7 and Annex 3.

(D) Fisher Ch2

31/5/07

09:57

Page 85

After Southwood

85

this time, quantitative forms of risk assessment were progressively more promoted.198 Among senior civil servants the RI reforms resulted in them understanding their role in more instrumental terms.199 The introduction of reforms such as the creation of Next Step Agencies also consolidated an instrumental understanding of public administration. Thus, for example, the RI nature of the CVL was bolstered in 1990 when it was turned into a Next Step Agency, and its relationship with MAFF came to be governed by a series of formal research contracts.200 Most significantly however, regulatory action became primarily guided by expert committees. One of the Southwood Working Party’s early conclusions had been that a consultative committee on research should be set up and this was constituted in February 1989 (known as the Tyrell Committee). It made a report in mid-1989 that recommended a series of research projects, many of which the government then undertook.201 Much of that research was co-ordinated by MAFF in instrumental terms.202 In April 1990 SEAC was set up with broader terms of reference and was relied on heavily for both scientific and policy advice,203 and met 28 times between May 1990 and March 1996.204 As Richard Packer has noted about SEAC: Ministers quickly recognized that they could not act against SEAC advice. To be more precise, they recognised that they could not take fewer precautions than SEAC had recommended. Taking more precautions than SEAC recommended needed careful justification but was sometimes desirable and possible.

As such, the sprawling civil service and its Ministers were deferring to a small expert group on the basis that that committee’s advice was the only legitimate basis for administrative action. Indeed, it was this committee that in March 1996 advised that the ‘most likely explanation’ for the appearance of v-CJD was due to exposure to BSE from eating beef and beef products before the introduction of a Specified Offal Ban in 1989.205 This statement prompted dramatic and farreaching regulatory action, including a ban on the use of cattle over 30 months old, which led to an effective cull of the national herd.206 Moreover, the RI paradigm remained the template for public administration after the fall of the Conservative government in 1997. The incoming Labour 198

Ibid at Annex 2. See statements at ibid at 10–11, and Andrews above n 106 at para 19. 200 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Two: Science, above n 12 at para 6.109. 201 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Eleven: Scientists After Southwood, above n 180 at chs 2 and 3. 202 Packer, above n 7 at 47. 203 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Eleven: Scientists After Southwood, above n 180 at ch 4. 204 Ibid at para 4.92. 205 Ibid at paras 4.627–4.632. 206 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume One: Findings and Conclusions, above n 196 at para 103. 199

(D) Fisher Ch2

31/5/07

86

09:57

Page 86

BSE, Expertise, and Administrative Constitutionalism

government had as one of its commitments the overhauling of the food safety system.207 In 2000 they created a new Food Standards Agency, an independent agency which would provide advice about standard setting to central government. The role of the Agency was understood in the following terms: The Agency’s assessment of food standards and safety will be unbiased and based on the best available scientific advice, provided by experts invited in their own right to give independent advice.208

This is a very RI understanding of the role of administration in the process of risk evaluation. The Agency’s structure was to be based on ‘openness, transparency and best science’209 and was to ensure that the Agency was ‘driven rationally’.210 To that end, the Food Standards Act 1999 placed the following obligations on those evaluating risks: 23(2) The Agency, in considering whether or not to exercise any power, or the manner in which to exercise any power, shall take into account (among other things)— (a) the nature and magnitude of any risks to public health, or other risks, which are relevant to the decision (including any uncertainty as to the adequacy or reliability of the available information); (b) the likely costs and benefits of the exercise or non-exercise of the power or its exercise in any manner which the Agency is considering; and (c) any relevant advice or information given to it by an advisory committee (whether or not given at the Agency’s request).

This is a very long way from the provisions of the Animal Health Act 1981 and Food Act 1984 seen above. Not only are discretionary administrative powers subject to more regulation, but they are conceptualised in RI terms. The task of the Agency is primarily to engage in an analytical process of risk assessment. As such, technological risks are largely thought to be manageable through such techniques. These RI developments have not only occurred in relation to food safety issues but right across the risk regulation sphere.211 Moreover, there has also been a general overhaul of the expert committee system. Guidelines for expert committees were introduced in 2001,212 and general guidance on the role of such committees is now regularly updated.213 The focus in those documents is very much on understanding public administration, and the role of experts within it, in RI terms. It is 207

James, above n 91. UK Government, The Food Standards Agency: A Force for Change, Cm 3830 (London, HMSO, 1998) at ch 2. 209 Second Reading Speech, House of Commons, Hansard, col 786, 21 June 1999. 210 Ibid at col 793. 211 E Fisher, ‘Drowning by Numbers: Standard Setting in Risk Regulation and the Pursuit of Accountable Public Administration’ (2000) 20 Oxford Journal of Legal Studies 109. 212 Office of Science and Technology, Code of Practice for Scientific Advisory Committees (London, HMSO, 2001). 213 Office of Science and Technology, Guidelines 2000: Scientific Advice and Policy Making (London, HMSO, 2000); and HM Government, Guidelines on Scientific Analysis in Policy Making (London, HMSO, 2005). 208

(D) Fisher Ch2

31/5/07

09:57

Page 87

Conclusions

87

expected that some form of systematic risk assessment214 will be used and the role of such committees is to provide analysis so that public administration can be evidence-based.215 Within UK administrative constitutionalism, DC conceptions of the administrative state and technological risk evaluation seem an increasingly fading memory.216 This state of affairs is somewhat ironical. While one should not over-emphasise the ability of DC public administration to tackle problems as complex as BSE, the promotion of the RI paradigm clearly led to a particular characterisation of the human health risks from BSE which encouraged regulatory inaction and discouraged more wide-ranging deliberation. Yet recent reforms have not only continued to frame problems in this narrow RI manner but also have limited the role for more DC understandings of expertise and participation.

VI Conclusions In pondering the disjunction between the Southwood Working Party’s advice and the way in which such advice was treated, the BSE Inquiry wondered [h]ow was it that this message [the conditional and uncertain nature of their advice] seems to have been lost on so many?217

Commentators’ answers to that question have nearly always been in terms of the science/democracy dichotomy in that the Working Party’s conclusions were the product of the wrong relationship between science and politics.218 The role of the Working Party has been described as ‘a cautionary tale of mixing the conditionalities of science with the glib certainties of politics’219 and as a ‘legitimatory resource—a means of taking the politics out of policy-making’.220 On this account, the problems with how the Southwood Working Party’s advice was interpreted and utilised were due to an illegitimate relationship between science and politics,221 a state of affairs many see as due to the ‘awfulness’ of British institutions.222 For these commentators, the solutions to the problems encountered in the BSE crisis are said to lie in greater transparency, regulation, and the promotion 214

Office of Science and Technology, above n 212 at 10. HM Government, above n 213. 216 Although see Health and Safety Executive, Reducing Risks, Protecting People (London, Health and Safety Executive, 1999). 217 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9 above n 6 at para 1.4. 218 van Zwanenberg and Millstone, above n 7; and M Jacob and T Hellström, ‘Policy Understanding of Science, Public Trust, and the BSE–CJD Crisis’ (2000) 78 Journal of Hazardous Materials 303. 219 Obituary, ‘Sir Richard Southwood’ (2005) The Telegraph, 3 November. 220 van Zwanenberg and Millstone, above n 7 at 96. 221 Ibid at ch 10. 222 Jasanoff, above n 148, quoting that argument at 226, although this is an argument she firmly disagrees with. See also D Marquand, Decline of the Public (Cambridge, Polity Press, 2004) at 12–17. 215

(D) Fisher Ch2

31/5/07

88

09:57

Page 88

BSE, Expertise, and Administrative Constitutionalism

of communication between science and politics,223 such reforms being the means to right the wrongs of the British administrative state and to place the science/democracy relationship on a proper footing. My analysis in this chapter has highlighted the superficiality in these approaches in two ways. First, the story of the BSE crisis and the role that the Southwood Working Party played in it was not so much a story about the interface between science and politics but rather a story about how the nature and role of expertise was shaped by understandings of administrative constitutionalism. The Southwood Working Party were experts whose role and nature was a product of debates over administrative constitutionalism, rather than a set of values internal to science. To argue that the problems encountered in the BSE crisis were due to an incorrect relationship between science and politics completely ignores that administrative context. Second, what such a line of analysis also ignores is that all those exercising their power in relation to the health risks from BSE, whether that power was being exercised on the basis of the DC or the RI paradigm, believed that they were making decisions that would contribute to good public administration. The Southwood Working Party was operating in a context where their role and nature was understood in bifurcated terms. The traditional DC ethos of the civil service was in many ways well suited to addressing a technological risk problem such as BSE and, as we saw, early deliberations captured the complexity of the issues. The RI paradigm seemed to provide more objective constraints on public administration, a desirable state of affairs in a liberal democracy. Likewise, both paradigms were problematic. The DC paradigm entrusted considerable power into the hands of officials, while the RI entrusted considerable power into scientific analysis. As this is the case, it is not clear that there is any obvious proper footing on which the evaluation of the health risks from BSE should have been based. Moreover, whatever the case, general arguments about accountability and transparency do little to address these deeper issues of administrative constitutionalism. What this means is that those thinking about the BSE crisis and the implications of it for UK government must move beyond the science/democracy dichotomy and engage with issues of administrative constitutionalism. In so doing there are unlikely to be any simple answers. However, a more subtle understanding of decision-making processes is likely to be gained.

223 van Zwanenberg and Millstone, above n 7 at ch 10; Little, above n 33; and Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report, Volume One: Findings and Conclusions, above n 196 at para 1290.

(E) Fisher Ch3

31/5/07

09:57

Page 89

3 Hard Looks and Substantial Evidence: Scope of Review of US Risk Regulation Rule-making in the 1970s In 1976 Judge Leventhal of the District of Columbia Circuit of the United States (US) Federal Court of Appeals asked What does, and should, a reviewing court do when it considers a challenge to technical administrative decision-making?1

Leventhal was asking the question in the context of judicially reviewing a standard based on the evaluation of a technological risk, and he was not simply being rhetorical. In the early 1970s the US Federal government had created a number of administrative agencies entrusted with the task of evaluating and regulating risk. The decisions of those agencies had proved controversial and were invariably subject to judicial review actions. The central issue in those actions was nearly always how a court should review the factual and discretionary aspects of risk evaluation—what is often described as a question of scope of review. Throughout the 1970s the Federal judiciary carried on a robust, and at times heated, discourse over what that scope of review should be. That discourse was a discourse of administrative constitutionalism in which understandings about scope of review were grounded in the deliberative-constitutive (DC) and rational-instrumental (RI) paradigms. Like all such discourses, it was shaped by history, and the judiciary struggled to reconcile these new regimes with established legal understandings of administrative constitutionalism and scope of review. In this chapter I provide an account of two of these struggles over scope of review. The first struggle relates to the concept of ‘hard look review’. In a number of cases, judges characterised their scope of review as ensuring that public administration took a ‘hard look’ at the issues before them. Such a hard look was understood in both DC and RI terms, which resulted in two very different types of scope of review. The second struggle concerned the attempts by the Federal courts to make sense of what they saw as the legislative mismatch between the rule-making processes and scope of review provisions in the Occupational Safety and Health 1

Ethyl Corp v EPA 541 F 2d 1 (DC Cir 1976) at 6.

(E) Fisher Ch3

31/5/07

90

09:57

Page 90

Hard Looks and Substantial Evidence

Act 1970 (the OSH Act). In both cases, these struggles highlight not only how understandings of public administration, law, and technological risk are coproduced in the US, but also the particularly important role that law plays in that process of co-production due to adversarial legalism.2 A consequence of this has been legal concepts have often dominated debates about administrative constitutionalism in the US. This chapter is structured as follows. In the first section, I give an overview of the concept of scope of review and how it has developed in US Federal administrative law. In the second section, I examine the legislative and institutional frameworks of the risk regulation agencies created in the 1970s and show how they embodied both RI and DC understandings of administrative constitutionalism. In the third section, I examine the debates over hard look review between Chief Judge Bazelon and Judge Leventhal in the District of Columbia Circuit of the Federal Court of Appeals. In the fourth section, I analyse the judicial response to one particular regulatory scheme—that of the OSH Act. In both cases these debates resulted in judicial review doctrine being dominated by RI understandings of administrative constitutionalism, and the consequences of this for judicial review during the 1980s are explored in the fifth section. In the final section, I evaluate three consequences of this state of affairs: the rise of analytical opportunism on the part of litigants; ossification of rule-making; and a growing disillusionment in administrative process. Two points should be made clear before starting. First, this chapter does not provide an exhaustive account of how understandings of administrative constitutionalism shaped technological risk regulation in the US during this time. Such an account is beyond the scope of this chapter and would need to take into account not only other aspects of judicial review doctrine but also other policy debates and developments.3 Second, as stated in the last chapter in describing doctrines or legislation as being consistent with either the RI or DC paradigm, I am not saying that either of these paradigms was explicitly in operation. Rather, those doctrines and that legislation can be understood to rest on ideals encapsulated in those paradigms.

I Scope of Review and Administrative Constitutionalism in Historical Perspective The focus in this chapter is upon the co-production of law, public administration, and understandings of technological risk in the context of US judicial review. This 2 R Kagan, Adversarial Legalism: The American Way of Law (Cambridge, Harvard University Press, 2003). 3 For a discussion of those see E Fisher, Risk, Expertise and Judicial Review: Scope of Review and Decision-Making Under Scientific Uncertainty, Thesis in Law Submitted for a DPhil Degree, (University of Oxford, 1998).

(E) Fisher Ch3

31/5/07

09:57

Page 91

Scope of Review and Administrative Constitutionalism

91

focus is no accident and reflects the fact that ‘accountability through litigation’4 has been the leitmotif of US administrative law, particularly during the period on which this chapter focuses. Judicial review is a legal phenomenon embedded in legal processes, and an important starting point is understanding something about its nature. On the face of it, judicial review of administrative action should be a relatively straightforward enterprise—the role of the courts is to review administrative decision-making so as to ensure that administrative decision-makers stay within their powers. In nearly every legal culture, however, the law and theory of judicial review is profoundly complex and fraught. While judicial review of administrative action may be ‘psychologically necessary’5 in a democracy, the form that that review should take is open to question. Much of this has to do with the fact that the legal boundaries of administrative power are rarely bright lines, and this is particularly the case when public administration is delegated considerable discretion. As seen in chapter one,6 this is a particular problem with technological risk evaluation and one of the reasons why it has proved so controversial. My concern is not with all aspects of US Federal judicial review doctrine, however, but only those aspects of it which proved to be contentious in the judicial review of risk regulation decisions in the 1970s. Judicial review has numerous different aspects governed by both legislative frameworks and case-law, which has resulted in courts taking different approaches depending on whether what is being reviewed is a question of law, procedure, discretion, or fact. The importance of these distinctions relates to questions of competence and, as Davis and Pierce note: Courts usually substitute judgement on the kind of questions of law that are within their special competence, but on other questions they limit themselves to deciding reasonableness; they do not clarify the meaning of reasonableness but retain full discretion in each case to stretch it in either direction.7

In thinking about judicial review of risk evaluation decisions my concern is with these ‘other questions’, because it is these which directly relate to how risk evaluation should be carried out. These ‘other questions’ are often described as issues concerned with scope of review8 and include grounds of review such as the ‘arbitrary and capricious’ standard and the ‘substantial evidence’ standard. This area of review has proved not easily subject to analysis. For some commentators, scope of review holds ‘no more substance than a seedless grape’.9 For others, judicial intervention on this ground is largely pragmatic,10 and for others it is 4

Kagan, above n 2 at 215. L Jaffe, Judicial Control of Administrative Action (Boston, Little, Brown, & Co, 1965) at 320. 6 See section II.B of ch 1. 7 KC Davis and R Pierce, Administrative Law Treatise, 3rd edn (San Diego, KC Davis Publishing Co, 1994) at §29.1. 8 Note that some commentators draw a distinction between scope, standard, and intensity of review but, generally speaking, these terms have been used interchangeably by courts and academics alike. 9 E Gelhorn and G Robinson, ‘Perspectives on Administrative Law’ (1975) 75 Columbia Law Review 771 at 780–81. 10 Davis and Pierce, above n 7 at §29.1. 5

(E) Fisher Ch3

31/5/07

92

09:57

Page 92

Hard Looks and Substantial Evidence

ideological.11 Edley has noted, the whole issue of scope of review is ‘just one big problem’ and to describe it as a law or doctrine presumes a clarity that does not exist.12 Whatever the case, scope of review doctrine is constantly evolving.13 As DeLong has commented, it is like a philosophy exam where the question remains the same but the ideal answer keeps changing.14 Although not easily subject to analysis it is clear that embodied in scope of review doctrines are understandings of constitutionalism and administrative constitutionalism. The former plays a role because what courts think are appropriate answers to these questions will depend in part upon how courts understand their own constitutional and institutional competence.15 Courts only have the power to review the legality not the merits of the decisions, and the growth of public administration is seen as due to the institutional limits of courts and adjudicative procedure.16 Scope of review also raises issues of administrative constitutionalism, because what courts understand to be the role and nature of the body they are reviewing will directly affect how they carry out review. If the courts understand their role as ensuring that decisions are legally ‘reasonable’ then what is understood as reasonable will depend upon prescriptions about good public administration. The starting point is what the court understands as the appropriate paradigm of administrative constitutionalism that should be in operation. Specific doctrines of scope of review will be grounded in, and promote, different understandings of administrative constitutionalism. It is in this way that law, public administration, and understandings of technological risk are co-produced in the context of judicial review. Thus, for example, if a court understands the role of a decision-maker in RI terms then what is reasonable action on the part of that decision-maker will be identifying, assessing, and applying information to the legislative mandate. Interest representation will be the form of communication, and technological risks will be understood as manageable. Scope of review will be concerned with ensuring that a decision-maker has managed a problem through rigorous processes of analytical assessment that ensure the decision-maker does not stray outside legislative boundaries and is not swayed improperly by interest representation. In contrast, if a court understands the role of an administrative body in more DC terms then reasonable administrative action by such a body will be in terms of how it exercised all aspects of its discretion in relation to a particular problem. A court’s 11 R Revesz, ‘Environmental Regulation, Ideology and the D.C. Circuit’ (1997) 83 Virginia Law Review 1717; and R Pierce, ‘Legislative Reform of Judicial Review of Agency Actions’ (1995) 44 Duke Law Journal 1110. 12 C Edley, Administrative Law: Rethinking Judicial Control of Bureaucracy (New Haven, Yale University Press, 1990) at 96. 13 R Levin, ‘Scope of Review Legislation: The Lessons of 1995’ (1996) 31 Wake Forest Law Review 647 at 652. 14 J DeLong, ‘Informal Rulemaking and the Integration of Law and Policy’ (1979) 65 Virginia Law Review 257 at 285. 15 Edley, above n 12. 16 F Frankfurter, ‘The Task of Administrative Law’ (1927) 75 University of Pennsylvania Law Review 614.

(E) Fisher Ch3

31/5/07

09:57

Page 93

Scope of Review and Administrative Constitutionalism

93

scope of review will be less concerned with ensuring a decision is kept within pre-ordained boundaries and more with a decision-maker showing that it has thoroughly addressed the problem at hand in a process of deliberation and analysis. In both cases, the reasonableness of the decision needs to be established in relation to the role of information, expertise, and communication. Likewise, under either paradigm, a court may be more intensive or deferential in its review—the key difference is not the degree of the court’s scrutiny but the template of ‘reasonableness’ that it is using to assess the legality of the decision. Of course, scope of review is not simply a crude vehicle for shaping public administration. It is also a legal concept embedded in a legal discourse. This can be best seen from its history in the US, a history which is also important because it was the starting point for judges reviewing risk regulation decisions in the 1970s. One of the first major independent Federal executive institutions to be created was the Interstate Commerce Commission (ICC)17 and it was entrusted with setting railroad rates as well as laying down rates in any given case by ad hoc adjudication.18 The concept of adjudicative procedure was discussed in chapter one;19 and in the early twentieth century it was the template for executive decision-making. On the one hand, this reflected a lack of other institutional alternatives at that time but it also was as an expression of RI understandings of public administration because such procedures, particularly in relation to fact-finding, were seen to be an effective means of constraining and regimenting administrative discretion.20 The ICC’s task was akin to risk regulation in that it was a collective enterprise needing expertise and information but it was one that combined judicial, legislative and adjudicative power and there was considerable debate over what its nature and role should be.21 For some, the ICC was an instrument of specialised factfinding and for others it was a ‘counsellor between the railroads and the public’.22 Whatever the case, many of the decisions were challenged before the US Supreme Court, challenges which directly raised questions of scope of review. In early cases, because of the adjudicative nature of the ICC’s decision-making, the court saw its task as one of ensuring that the ICC had limited itself to the finding of the facts in accordance with an adjudicative procedure.23 This was an understanding of scope of review based on the RI paradigm. By 1910, however, this was giving way to a more DC approach to judicial review in which the courts were 17

Interstate Commerce Commission Act 1887. T McGraw, Prophets of Regulation (Cambridge, The Belknap Press, 1984) at ch 2. 19 See section IV of ch 1. 20 R Pound, ‘The Growth of Administrative Justice’ (1924) 2 Wisconsin Law Review 321; and H Bevis, ‘Administrative Commissions and the Administration of Justice’ (1928) 2 University of Cincinnati Law Review 1. 21 B Cook, Bureaucracy and Self Government: Reconsidering the Role of Public Administration in American Government (Baltimore, Johns Hopkins University Press, 1996) at 79–86; and S Skowronek, Building A New American State: The Expansion of National Administrative Capacities 1877–1920 (New York, Cambridge University Press, 1982) at 138–50. 22 Railroad executive Albert Fink, as quoted in Cook, above n 21 at 83. 23 ICC v Brimson 154 US 447 (1894); and East Tennessee Virginia & Georgia Railway Company v ICC 181 US 1 (1901) at 12. 18

(E) Fisher Ch3

31/5/07

94

09:57

Page 94

Hard Looks and Substantial Evidence

recognising the complexity of the rate-making process and the need for deliberation and judgement. Not only did they see the irrelevance and limits of legal rules of evidence in this field,24 but they also recognised the need for ‘an intuition of experience that outruns analysis’.25 As such, they used the ‘substantial evidence’ test from appellate review of a jury’s decision as the ground of judicial review.26 The analogy between an administrative agency and a jury rested not so much on their role as fact-finders but rather on their role as deliberative bodies, and the courts were quick to note that the fact/law distinction was not a clear one.27 In other words, the courts were borrowing from those aspects of adjudicative procedure which seemed more appropriate to apply to DC public administration. They downplayed those aspects of adjudicative procedure which constrained factfinding and emphasised those aspects which rested on a more deliberative understanding of decision-making. Thus within legal discourse they were using legal concepts to adapt scope of review so that it was more in line with what they saw as the role of the ICC, and such a process of adaptation was brought about with a mixture of prescription and legal doctrine. This shift towards a DC vision of the emerging administrative state was also apparent in wider academic commentary. The Progressive literature of this era depicted public administration as ‘official custodians of a certain part of the accepted social program’ that was concerned with the enhancement of human life.28 Likewise, such administrative bodies needed to be ‘dignified by an element of independent authority’ so as to be institutions that would work towards ‘social enlightenment and up-building’.29 In particular, considerable discretion was needed so as to address the problems that were arising out of an increasingly complex market economy.30 The Progressive Era paved the way for the New Deal during which several independent regulatory commissions were created and given wide-ranging and open-ended powers.31 The legitimate authority of these bodies rested in their permanence, the professional judgement and expertise of their staff, and their dedication to a particular regulatory problem.32 Throughout the 1930s the courts also characterised public administration in DC terms. and it was understood that judicial review should promote in administration ‘that rare combination of intelligence and rectitude without which any

24

ICC v Louisville & Nashville Railway Company 227 US 88 (1913) at 94. Chicago Burlington & Quincy Railway Company v Babcock 204 US 585 (1907) at 598. 26 See especially ICC v Union Pacific Railway Company 222 US 541 (1912) at 547, and Louisville, above n 24 at 97. 27 Union Pacific, above n 26 at 547; Louisville, above n 24 at 94; and Arkon, Canton & Youngsville Railway Company v US 261 US 184 (1923). 28 H Croly, Progressive Democracy (New York, Macmillan Co, 1915) at 360. 29 Ibid at 361 and 354. 30 Bevis, above n 20 at 15; and J Wigmore, ‘Administrative Boards and Commissions: Are the JuryTrial Rules of Evidence in Force for Their Inquiries? ’ (1922) 17 Illinois Law Review 263 at 263. 31 See generally Cook, above n 21 at ch 5. 32 See especially J Landis, The Administrative Process (New Haven, Yale University Press, 1938), and F Frankfurter, The Public And Its Government (New Haven, Yale University Press, 1930). 25

(E) Fisher Ch3

31/5/07

09:57

Page 95

Scope of Review and Administrative Constitutionalism

95

program of administration may prove to be a curse’.33 To this end, the US Supreme Court developed a ‘zone of reasonableness’ test in reviewing complex expert decisions such as rating decisions, which recognised that there were a number of different ways in which such a task could be carried out.34 Likewise, the Supreme Court continued to define ‘substantial evidence’ in non-adjudicatory terms as such ‘relevant evidence as a reasonable mind might accept as adequate to support a conclusion’.35 This definition also stressed the test in non-quantitative terms, as the courts were quick to note that an administrator was not just a ‘bookkeeper posting items in a ledger’.36 Related to this was the explicit recognition by the courts that adjudicative procedures and evidentiary burdens ‘did not comport with the administrative process’,37 as the type of fact-finding an administrative body engaged in was flexible and embedded in experience and policy.38 The central issue in establishing the reasonableness of an administrative decision was the quality of argument not the quality of evidence.39 In 1946 the Administrative Procedure Act (APA) was passed which codified rule-making processes and the grounds of judicial review and thus provided an important starting point for understanding the US administrative process, and thus for administrative constitutionalism. Much of the Act was based on the Attorney General’s Committee on Administrative Procedure Report of 194140 and it was a report which promoted a DC understanding of public administration. In particular, the Committee clearly recognised the limits of adjudicative procedure in rule-making and recommended the development of a new form of rule-making that it called informal rule-making.41 The legal framework for informal rulemaking, or notice and comment rule-making as it became known, was a major aspect of the APA and was an explicit embracing of the DC paradigm. Rulemaking of this type was a three-step process: publication of a proposed rule; a period for notice and comment; and the publication of a final rule.42 Such a process was seen as advantageous because it promoted the ‘full utilisation of 33 Chief Justice Hughes, as quoted by J Davison, ‘Administration And Judicial Self-Limitation’ (1936) 4 George Washington Law Review 291 at 324. See also Federal Radio Commission v Nelson Brothers Bond & Mortgage Co 289 US 266 (1933) at 276, and Board of Railway Commissioners of the State of North Dakota v Great Northern Railway Co 281 US 416 (1931) at 422. 34 See especially Federal Power Commission v Hope Natural Gas Co 320 US 591 (1944) at 602, and Federal Power Commission v Natural Gas Pipeline Co 315 US 575 (1942) at 585 35 Consolidated Edison Co v National Labour Relations Board 305 US 197 (1938) at 229. 36 US v Morgan 313 US 409 (1940) at 415. 37 Sunshine Anthracite Coal Co v Adkins 310 US 381 (1940) at 398. 38 Tagg Brothers & Moorhead v US 280 US 420 (1930) at 442; J Frank, If Men Were Angels: Some Aspects of Government in a Democracy (New York, Harper & Bros, 1942) at 141; and International Association of Machinists, Toolmakers & Diemakers Lodge No 35 v National Labour Relations Board 311 US 72 (1940) at 79. 39 KC Davis, ‘An Approach to Problems of Evidence in the Administrative Process’ (1942) 55 Harvard Law Review 364 at 418. 40 Attorney General’s Committee on Administrative Procedure, Final Report of the Attorney General’s Committee on Administrative Procedure (Washington, Government Printing Office, 1941). 41 Ibid at 61. 42 5 USC §553.

(E) Fisher Ch3

31/5/07

96

09:57

Page 96

Hard Looks and Substantial Evidence

concentrated experience’, allowed significant consultation, and the substantive exercise of discretion in light of that deliberation.43 The Committee also recognised that different types of decision-making processes with their different fact-finding processes should result in different scopes of review.44 This is a good example of the co-production of law and public administration. Consequently, the APA codified the grounds of judicial review by linking scope of review to particular types of decision-making processes.45 The ‘arbitrary and capricious’ standard was to apply to informal rule-making and the ‘substantial evidence on the record’ test to formal rule-making where a formal record was created by adjudicative procedure. The practical consequence of it was that the ‘substantial evidence’ test became strongly associated with adjudicative procedure, and while it continued to be interpreted in DC terms, it was understood to be based on a more constrained RI understanding of public administration.46 In contrast, informal rule-making was the embodiment of the DC paradigm and it was that which came to dominate the US administrative process.47 What is also interesting to note about the report is that the Attorney General’s Committee explicitly flagged up the problems that risk regulation rule-making created for scope of review. The Federal government at this time had limited risk regulation responsibilities in regards to food and drugs, and this regulatory area had proved contentious.48 The Committee outlined at length the problem of scope of review. In considering now whether judicial review of a detailed kind is desirable, attention should be paid to the nature and complexity of the questions of fact involved. To take a comparatively simple example, suppose the problem to be that of prescribing regulations for a particular type of poisonous spray residue to be permitted upon raw apples shipped in interstate commerce. The following questions would seem to have a bearing upon the final result: (a) the quantity of the particular poison, consumed within, say, a year, that will have a definitely harmful effect upon ordinary individuals; (b) the proportion of individuals that would be similarly affected by smaller quantities, and what quantities; (c) the quantity of unpeeled apples, and hence of poison upon apples consumed by individuals in, say, a year; (d) the quantity of the same poison consumed by individuals upon other products in the same time; (e) the physical practicability and; ( f ) the cost of reducing the amount of spray residue to various quantities and of eliminating it entirely before the apples are shipped; (g) the probable distribution between growers and consumers of the added cost incident in the removal of spray residue, in light of (h) the effect of higher prices upon consumption and (i ) the counter effect of knowledge by consumers that apples carry poison.49 43

Attorney General’s Committee on Administrative Procedure above n 40 at 19, 46, 61 and 103. Ibid at 92. 45 5 USC §706. 46 Universal Camera Corp v National Labour Relations Board 340 US 474 (1951). 47 KC Davis, Discretionary Justice: A Preliminary Enquiry (Urbana, University of Illinois Press, 1965). 48 R Fuchs, ‘The Formulation and Review of Regulations under the Food Drug and Cosmetic Act’ (1939) 6 Law and Contemporary Problems 43. 49 Attorney General’s Committee on Administrative Procedure, above n 40 at 118. 44

(E) Fisher Ch3

31/5/07

09:57

Page 97

Scope of Review and Administrative Constitutionalism

97

Every one of these questions is diffcult to answer, let alone to judicially review the answer administration may come to. Not surprisingly, the report concluded that in such cases it was not ‘desirable’ for the courts to impose any heavy procedural or analytical demands on a decision-maker.50 They did so on the basis of the DC paradigm—that administrative decision-making in such areas was highly complex and decision-makers need to exercise their discretion flexibly. The 1950s saw a continuation of DC thinking in judicial review doctrine that developed out of the schemata set down by the APA.51 The existence of a legal framework for informal rule-making was taken as recognition of the fact that the ‘legislative type’ rules passed pursuant to such a framework did not need the establishment of particular facts but rather the creative exercise of discretion.52 By the late 1950s, however, this faith in, and emphasis on, flexible discretion and professional judgement was giving way to distrust. The Hoover Commission report,53 the Landis report54 and the Ash Council55 all pointed to the problems of agency inertia and the capture of regulatory programmes by business interests.56 Bureaucracy was viewed as being ‘tainted with an ineradiciable lust for power’57 and did not, as perhaps once thought, ‘combine the celerity of Mercury, the wisdom of Minerva, and the purity of Diana’.58 As such, from a legislative and legal perspective, there was a greater emphasis on controlling administrative bodies and constraining their discretion.59 This shift could also be seen in the case-law.60 Despite this distrust, however, there was an expectation that government would take on new regulatory agendas and the creation of new administrative programmes were a means of putting a ‘democratic wish’ into action.61

50

Ibid at 120. Federal Communications Commission v RCA Communications 346 US 86 (1952); Universal Camera, above n 46 at 488–9; and Fleming v Florida Citrus Exchange 358 US 153 (1958) at 163. 52 See especially KC Davis, Administrative Law Treatise (St. Paul, West Publishing, 1958) at §15.03. For case examples see American Airlines Inc v Civil Aeronautics Board 359 F 2d 624 (DC Cir 1966) at 633, and National Labour Relations Board v Wyman-Gordon 394 US 759 (1969) at 765. 53 Hoover Commission, Report on Organisation of the Executive Branch of Government (New York, McGraw-Hill Book Co Inc, 1949) at 4–7. 54 J Landis, Report on Regulatory Agencies to the President Elect (Washington DC, 1960). 55 Task Force on Legal Services and Procedure, Task Force Report on Legal Services and Procedure (Washington DC, Commission on Organization of the Executive Branch of the Government, 1955) at 4. 56 M Bernstein, Regulating Business by Independent Commission (Princeton, Princeton University Press, 1955). 57 N Long, ‘Bureaucracy and Constitutionalism’ (1952) 46 American Political Science Review 808. 58 H Friendly, The Federal Administrative Agencies: The Need for Better Definition of Standards (Cambridge, Harvard University Press, 1962) at 2. 59 See especially Davis, Discretionary Justice: A Preliminary Enquiry, above n 47. 60 American Ship Building Co v National Labour Relations Board 380 US 300 (1965) and Abbott Laboratories v Gardner 387 US 136 (1967). 61 J Morone, The Democratic Wish: Popular Participation and the Limits of American Government (Revised ed, New Haven, Yale University Press, 1998); and C Reich, The Greening of America (New York, The Penguin Press, 1970). 51

(E) Fisher Ch3

31/5/07

98

09:57

Page 98

Hard Looks and Substantial Evidence

II Administrative Constitutionalism and Risk Regulation Regimes in the Early 1970s Part of that democratic wish was the creation of new regulatory regimes to protect the public from a wide range of environmental and health risks. These regimes were inevitably administrative in nature62 and thus immediately raised issues of administrative constitutionalism. The cornerstones of these new regimes were administrative bodies that were entrusted with a wide range of rule-making powers so as to set ex ante standards. These bodies included the Environmental Protection Agency (EPA), the Occupational Safety and Health Administration (OSHA), the Consumer Product Safety Commission (CPSC), the National Highways and Safety Transport Administration (NHSTA), and the Nuclear Regulatory Commission. Each agency had a unique structure, a particular agenda, and was created in a different manner. In most cases, the structure and positioning of these bodies represented a step back from the independent commissions of the New Deal in that there was a greater focus on legislative and central executive control.63 Each of these bodies was entrusted with a variety of programmes and research powers.64 They were also given a diverse range of legislative mandates. On the one hand, these agencies were entrusted with DC regulatory powers which were wide ranging in recognition of the complexity and seriousness of the problems being dealt with.65 These mandates invariably granted agencies openended power to deal with problems as they arose while often providing a checklist of issues that a decision-maker should have regard to.66 On the other hand, Congress wished to avoid granting agencies large amounts of unfettered discretion, and, as such, delegated to these new agencies specific RI tasks in which discretion was limited to a set of clearly defined goals, although the goals were often unreasonably ambitious.67 Such standards often gave deadlines for action68 and/or were very specific about subject matter.69

62

See section II.A of ch 1, this volume. B Ackerman and W Hassler, Clean Coal/Dirty Air: Or How the Clean Air Act Became a Multimillion Dollar Bail Out for High Sulfur Coal Producers and What Should be Done About It (New Haven, Yale University Press, 1981) at 2 and 9. 64 The EPA’s research powers under the Clean Air Act, 42 USC §7403; the Federal Water Pollution Control Act, 33 USC §1254–§1264; the Safe Drinking Water Act, 42 USC §300j-1; the Federal Insecticide, Fungicide, and Rodentcide Act, 7 USC §136r; and the Toxic Substances Control Act, 15 USC §2609(a). 65 Federal Water Pollution Control Act, 33 USC §1317; Resource, Conservation and Recovery Act, 42 USC §6922(a); and Consumer Product Safety Act, 15 USC §2056(a). 66 Occupational Safety and Health Act, 29 USC §655(b)(5) and Clean Air Act, 42 USC §7409(b)(1). 67 Food, Drug and Cosmetic Act, 21 USC §348(c)(3)(A) (known as ‘the Delaney Clause’); Safe Drinking Water Act, 42 USC §300g-1(b)(4); and Toxic Substances Control Act, 15 USC §2605(a). 68 The original s 111 of the Clean Air Act required the EPA to propose national ambient air quality standards 30 days after the enactment of the legislation. 69 Polychlorinated biphenyls in the Toxic Substances Control Act 1976, 15 USC §2605. 63

(E) Fisher Ch3

31/5/07

09:57

Page 99

Administrative Constitutionalism and Risk Regulation in the Early 1970s

99

The ambiguity over the role and nature of these bodies was also heightened by the processes for rule-making laid down in this legislation. The main means of decision-making for these new administrative bodies was rule-making, and each of these statutes set out the processes that agencies needed to comply with in making rules. However, while informal rule-making formed the basis for rule-making in risk regulation in nearly all cases, the process was amended by legislation,70 by the agencies themselves,71 and by the courts.72 These adjusted processes were generically described as forms of hybrid rule-making.73 In each case, the amendments were due to a feeling that informal rule-making was inadequate but the reasons given for the inadequacy varied. On the one hand, some amendments aimed to bolster the DC aspects of the process by increasing deliberation through longer periods for comment,74 and informal, public hearings.75 There was a perception that rule-making in relation to risk evaluation needed to be unshackled from restrictive fact-finding procedures.76 As Skelly Wright noted, extrajudicially informal rule-making ‘mandate[d] a dialogue and not a trial’, and accuracy was less important than requiring good faith consideration of all the issues at hand.77 On the other hand, however, the amendments to rule-making procedure were attempting to ground the authority of decision-makers in rational analysis, and thus in terms of the RI paradigm. In particular, there was an emphasis on making rule-making a scientific process in which information was collected and documents produced that set out the factual basis and the methodology used.78 Adjudicative procedures such as cross-examination were also re-introduced.79 The final confusing aspect of these new regimes, and the aspect that proved particularly problematic for the courts, was the legislated grounds of review. Many of these new statutes set out the grounds and opportunities of review but in a way that significantly deviated from how grounds of review had been previously 70

Clean Air Act, 42 USC §7607(d)(3)(A)–(B) See generally W Pederson, ‘Formal Records and Informal Rulemaking’ (1975) 85 Yale Law Journal 38. 72 The court’s role in procedural reform was strongly rejected by the Supreme Court in Vermont Yankee Nuclear Power Corp v NRDC 435 US 519 (1978). 73 S Williams, ‘“Hybrid Rulemaking” under the APA: A Legal and Empirical Analysis’ (1976) 42 University of Chicago Law Review 401. 74 Occupational Safety and Health Act, 29 USC §655(a)(4); and Federal Water Pollution Control Act, 33 USC §1317(a)(2). In most cases, rule-making procedures were lengthened from 30 days to 60 days. 75 Occupational Safety and Health Act, 29 USC §655(a)(3); Federal Water Pollution Control Act, 33 USC §1317(a)(2); and Clean Air Act, 42 USC §7607(d)(5). 76 B Boyer, ‘Alternatives to Administrative Trial Type Hearings for Resolving Complex Scientific, Economic and Social Issues’ (1972) 71 Michigan Law Review 111 at 112; R Cramton, ‘A Comment on Trial Type Hearings in Nuclear Power Plant Siting’ (1972) 58 Virginia Law Review 585; and R Hamilton, ‘Procedure for the Adoption of Rules of General Applicability: the Need for Procedural Innovation in Administrative Rulemaking’ (1972) 60 California Law Review 1276 at 1290. 77 J Wright, ‘The Courts and the Rulemaking Process: The Limits of Judicial Review’ (1974) 59 Cornell Law Review 375 at 380 and 392. 78 Clean Air Act, 42 USC §7607(d)(3)(A)–(B). 79 Toxic Substances Control Act 15 USC §2605(c)(3)(A)(ii) and Federal Water Pollution Control Act, 33 USC §1317(a)(2) 71

(E) Fisher Ch3

31/5/07

100

09:57

Page 100

Hard Looks and Substantial Evidence

understood. Thus, for example, a number of the new statutes required that the standard of review was the substantial evidence test,80 even though hybrid rulemaking processes were adjusted from informal rule-making. As such, these new schemes were at odds with the APA schemata for judicial review. The problems this created will be seen in Section IV. It was the courts engaged in judicial review which ultimately had to make sense of these frameworks. This was particularly because there was an increasing anxiety that decision-making under these regimes should be accountable and that accountability was best ensured through judicial review.81 As one judge noted extrajudicially: It is as if sensitivity to the fact that it has failed to define grants of authority with precision has prompted Congress to rely on the courts to keep the delagatees within proper bounds in their execution of the trusts reposed in them.82

In this regard, judges described courts and agencies as ‘collaborative instrumentalities’ that worked towards a common goal of environmental protection and good administration.83 These regimes also appeared at a time where there was an exponential growth in administrative law litigation, and many rules under these new regimes were subject to judicial review challenges by actors right across the political spectrum.84 The consequence of this adversarial legalism was that the courts became the arenas for debate about the role and nature of these administrative agencies, and law became the discourse. Judicial review actions thus became forums of debates over administrative constitutionalism in which law, public administration, and understandings of technological risk were coproduced. That process of co-production, however, was not occurring on the basis of a clean slate. Not only were the courts struggling with interpreting the concept of reasonable action in light of these new legislative frameworks but they were also doing so in light of long-established understandings about administrative process and judicial review. It is not possible in the space here to review all the case-law from this period. What it is useful to do, however, is to look at two examples of these types of debates that developed over scope of review in this area. The first example given is that of the debate carried out concerning the nature of ‘hard look’ review. The second example is that of a series of courts struggling to make sense of one legislative regime—the OSH Act. 80 29 USC §655(f) (OSHA); 15 USC §2058 (CPSC); and 15 USC §2618 (EPA, under the Toxic Substances Control Act); and 42 USC §300g-1 (EPA, under the Safe Drinking Water Act) 81 A Marcus, Promise and Performance: Choosing and Implementing an Environmental Policy (Westport, Greenwood Press, 1980) at 44. 82 C McGowan, ‘Congress, Court and the Control of Delegated Power’ (1977) 77 Columbia Law Review 1119 at 1124. 83 H Leventhal, ‘Environmental Decision Making and the Role of the Courts’ (1974) 122 University of Pennsylvania Law Review 509 at 512. 84 Kagan, above at n 2 at 48–50; and R Stewart, ‘The Discontents of Legalism: Interest Group Relations in Administrative Regulation’ (1985) Wisconsin Law Review 655.

(E) Fisher Ch3

31/5/07

09:57

Page 101

Hard Look Review

101

III Hard Look Review Many of the judicial review challenges to risk regulation decisions were brought in the District of Columbia Circuit of the Federal Court of Appeals and, consequently, this court became an important forum for debating the role and nature of judicial review, and in particular, the issue of scope of review. Judicial perspectives on these issues were shaped by both the DC and RI paradigms. In regard to the former, there were those that saw technological risk evaluation as complex, scientifically uncertain, and subject to socio-political conflict.85 As such, it was perceived that risk regulators needed considerable discretion so as to exercise professional judgement and engage in discourse.86 To that end, judges interpreted statutory mandates purposively,87 and review focused on reasons and explanations rather than administrators proving findings.88 For others, however, risk evaluation was envisaged in RI terms. Government was not to be ‘by whim’,89 and informal rulemaking was not a ‘freewheeling’, ‘brainstorming’ and unaccountable process.90 Standard setting was understood as a primarily analytical enterprise, and decisionmakers needed to ensure a ‘rational basis’ for decision-making through the application of scientific principles.91 These two distinct approaches manifested themselves in many ways, the most explicit being the extremes of the debate over hard look review between Chief Judge Bazelon and Judge Leventhal. The concept of hard look review was developed in the late 1960s as an interpretation of the nature of scope of review in relation to the arbitrary and capricious standard. Its central idea was that the courts should ensure that the agency had taken a ‘hard look’ at the issue before them. The hard look concept was a way of dealing with the institutional competence problem created by a generalist court reviewing a specialist administrator.92 The question, of course, was what should be the nature of that ‘hard look’ and what can be seen is that the concept was interpreted in both RI and DC terms. Judge Leventhal took the first steps towards concretising the concept of hard look review and did so in RI terms. Leventhal, before being on the bench, was a 85 Reserve Mining Co v EPA 514 F 2d 492 (8th Cir 1975) at 519; Environmental Defense Fund v EPA 510 F 2d 1292 (DC Cir 1975) at 1299; and Environmental Defense Fund v EPA 465 F 2d 528 (DC Cir 1972) at 535–6. 86 National Asphalt Pavement Association v Train 539 F 2d 775 (DC Cir 1976) at 783; Amoco Oil Co v EPA 501 F 2d 722 (DC Cir 1974) at 737; and Weyerhaeuser Co v Costle 590 F 2d 1011 (DC Cir 1978) at 1029. 87 Amoco Oil, above n 86 at 741; and Reserve Mining, above n 85 at 529. 88 Ethyl, above n 1; and Amoco Oil, above n 86. 89 Nor Am Agricultural Products Inc v Hardin 435 F 2d 1151 (7th Cir 1970) at 1166. 90 W Rodgers, ‘A Hard Look at Vermont Yankee: Environmental Law under Close Scrutiny’ (1979) 67 Georgetown Law Journal 699 at 717. 91 Bunker Hill Co v EPA 572 F 2d 1286 (9th Cir 1977) at 1299; Certified Colour Manufacturers Association v Matthews 543 F 2d 284 (DC Cir 1976) at 294; and National Lime Association v EPA 627 F 2d 416 (DC Cir 1980) at 452–3. 92 Leventhal, above n 83.

(E) Fisher Ch3

31/5/07

102

09:57

Page 102

Hard Looks and Substantial Evidence

member of the Hoover Commission93 and, in line with that commission, he was concerned to ensure that administrative expertise was not an excuse for unarticulated irrational reasoning94 and that informal rule-making was not a ‘seed bed for the weed of industry domination’.95 As communication was interest representation, then the courts needed to ensure that agency capture did not result in decision-makers deviating from their legislative mandates. In Greater Boston Television Corporation v Federal Communications Commission96 Leventhal first set out in detail his understanding of ‘hard look review’. He stated: Expertise is strengthened in its proper role as the servant of government when it is denied the ‘opportunity to become a monster which rules with no practical limitations on its discretion’.97

Judge Leventhal described the task of the agency as being threefold—keeping within legislative boundaries, finding facts, and ‘selecting’ policies.98 As such, Leventhal was clearly understanding administrative agencies in RI terms—risk regulators were instrumental appendages of the legislatures. In establishing that it had taken a hard look, an agency would need to show that it had identified the crucial facts, taken into account the different expert opinions, and not been inappropriately swayed by interest representation.99 This was best done by the agency showing that it had stayed within the boundaries of rational knowledge. Leventhal felt the hard look doctrine had particular applicability to risk regulation because the consequences of regulation were so serious and widespread and thus any decision required more scrutiny.100 He wanted to ensure an agency had taken a ‘hard look at its hard problems’101 by ensuring that there was a factual basis to the decision.102 Thus he noted in Portland Cement Association v Ruckelshaus (the Portland Cement decision)103 that: It is not consonant with the purpose of a rule-making proceeding to promulgate rules on the basis of inadequate data, or on data that, to critical degree is known only to the agency.104

This factual basis acted as a replacement for a record developed by adjudicative procedure. Moreover, courts, Leventhal argued, had an excellent grasp of what it meant to ‘prove something’.105 In this case he found that the EPA had the burden 93

Hoover Commission, above n 53. Environmental Defense Fund v Ruckelshaus 439 F 2d 584 (DC Cir 1971) at 597–8. 95 Walter Holm & Co v Hardin 449 F 2d 1009 (DC Cir 1971) at 1016 per Judge Leventhal. 96 Greater Boston Television Corp v FCC 444 F 2d 841 (DC Cir 1970). 97 Ibid at 850. 98 Ibid at 851. 99 Ibid at 851. 100 Welford v Ruckelshaus 439 F 2d 598 (DC Cir 1971) at 601; and International Harvester v Ruckelshaus 478 F 2d 615 (DC Cir 1973). 101 Environmental Defense Fund, above n 94 at 541. 102 Leventhal above n 83 at 511. 103 Portland Cement Association v Ruckelshaus 486 F 2d 375 (DC Cir 1973). 104 Ibid at 393. 105 Leventhal, above n 83 at 538. 94

(E) Fisher Ch3

31/5/07

09:57

Page 103

Hard Look Review

103

of proving that its analysis was correct.106 To do so it had to refute or take into account other material scientific viewpoints.107 Any material factual submission would have to be dealt with before such a burden would be discharged.108 Likewise, Leventhal in carrying out hard look review wanted to ensure that any decision was methodologically rigorous.109 As he noted in International Harvester Co v Ruckelshaus (the International Harvester decision)110 ‘the underlying issue was the reasonableness and reliability of the Administrator’s methodology’.111 This meant that procedures should be dedicated to strengthening analysis.112 The agency would need to set out in detail its methodology and if it deviated from petitioner’s views or departed from past practices explain why it did. Leventhal did acknowledge the problems of scientific uncertainty. He insisted, however, that an important distinction should be made between ‘prophecy’ and ‘prediction’.113 Only the latter was valid for an agency, and in predicting the agency would need to rely heavily on methodological analysis.114 Leventhal remanded decisions on a number of grounds. In International Harvester he did so because the standards had been based on unexplained assumptions and variations from the methodology of other government agencies, and because the statistical reliability of predictions was too weak.115 In Portland Cement he did so because the manufacturers had not been allowed to comment because the details of methodology had not been disclosed.116 In Environmental Defense Fund v EPA117 he remanded the decision because the EPA had ordered a total suspension of the use of aldrin and dieldrin where the facts suggested less drastic action would be desirable. Leventhal’s ‘rule of administrative law’118 was a rule of the RI paradigm. A risk regulator’s legitimacy lay in analytical methodology and the legislative will. As such, the court’s task in judicial review was probing the quality of the analytical reasoning. The practical implications of this were that the administrative record needed to set out in detail the methodology of decision-making and its associated reasoning. As Leventhal’s primary concern was with control and the limiting of discretion, he had no time for approaches to judicial review that characterised an administrative decision as the product of an ‘intuition of experience which outruns analysis’.119 106

International Harvester, above n 100 at 648. Portland Cement, above n 103 at 393. 108 Ethyl, above n 1 at 69. 109 Portland Cement, above n 103 at 392. 110 International Harvester, above n 100. 111 Ibid at 643. 112 Ibid at 632. 113 Ibid at 642; and Portland Cement, above n 103 at 391. 114 International Harvester, above n 100 at 642. 115 Ibid at 648. 116 Portland Cement, above n 103 at 402. 117 Environmental Defense Fund, above n 85 at 537 and 541. See also Environmental Defense Fund, above n 94 at 1298. 118 Leventhal, above n 83 at 514. 119 Greater Boston, above n 96 at 852, quoting Justice Holmes. 107

(E) Fisher Ch3

31/5/07

104

09:57

Page 104

Hard Looks and Substantial Evidence

This, however, was not the only way in which hard look review was interpreted. Chief Judge Bazelon also saw the role of judicial review as one concerned to ensure that administrative agencies took a hard look, but of a very different sort from that of Leventhal—one grounded in the DC paradigm of administrative constitutionalism. For Bazelon, the starting point was the inherent complexity of decisionmaking in regard to technological risk that required decision-makers to focus on the broader problems inherent in risk regulation.120 He sharply distinguished the role of risk regulators from the role of scientists and, in particular, pointed to the fact that administrative agencies had been set up because of the problems of proving harm in other forums such as the common law courts.121 Technological risk problems were inherently complex and required consideration of numerous different factors. Likewise, Bazelon was acutely aware that scientific uncertainty required many decisions to be made on ‘judgement calls’,122 and regulators needed to act in ‘spite of uncertainty’ as opposed to scientists ‘who sort to conquer it’.123 Bazelon was thus defining regulatory expertise as separate from scientific expertise and thus in a very different way from Leventhal. The nature of these problems also required a wider group of actors to participate in that process.124 He stressed that while in many cases better evidence may be obtainable, the important issue was whether the facts available allowed both the agency and the wider public to engage in a proper debate.125 For him: Complex questions should be resolved in the crucible of debate through the clash of informed but opposing scientific and technological viewpoints.126

There needed to be full disclosure, including of the agency’s ignorance as opposed to uncertainty.127 This was not to say that the task of the agency was merely to broker interest group trade-offs, but rather that answers could be found through leading a process of substantive dialogue.128 Administration would need to be flexible, and trial and error had a role to play.129 Standard setting was thus a deliberative problem-solving exercise. As such, judicial review should aim to ensure that different groups had a meaningful opportunity to participate and, more importantly, that the final agency outcome had been a product of a deliberative process.130 Thus he noted:

120 D Bazelon, ‘Coping with Technology through the Legal Process’ (1977) Cornell Law Review 817 at 827. 121 D Bazelon, ‘Science and Uncertainty: A Jurist’s View’ (1981) 5 Harvard Environmental Law Review 209 at 213. 122 AFL-CIO v Marshall 617 F 2d 636 (DC Cir 1979) at 651. 123 Bazelon, above n 121 at 213. 124 Bazelon, above n 120 at 818 and 831. 125 AFL-CIO, above n 122 at 658. 126 International Harvester, above n 100 at 651. 127 Bazelon, above n 121 at 212 and 214. 128 Bazelon, above n 120 at 831. 129 Environmental Defense Fund, above n 85 at 598. 130 AFL-CIO, above n 122 at 652.

(E) Fisher Ch3

31/5/07

09:57

Page 105

Hard Look Review

105

The court’s role is . . . to monitor the agency’s decision-making process—to stand outside both the expert and political debate and to assure that all the issues are thoroughly ventilated.131

To that end he focused upon administrative procedures which would ensure that this would occur132 rather than on matters of analytical methodology. It is on this point that he most obviously differed from Leventhal.133 He also noted the important role that a system of checks and balances played.134 Bazelon’s decision in Natural Resources Defense Council v Nuclear Regulatory Commission (the Table S-3 decision)135 is a prime example of his approach and of judicial review grounded in a DC paradigm. That case involved the licensing of a nuclear power plant and a rule-making by the Nuclear Regulatory Commission. The purpose of the rule-making was to consider the environmental effects of the nuclear fuel cycle. The Nuclear Regulatory Commission had concluded that nuclear waste presented only ‘insignificant’ environmental effects and it developed a series of formulae and numerical tables for cost/benefit analysis which rested on this assumption.136 The evidence on which it based this conclusion was a 20-page statement of an expert from the Atomic Energy Commission which was submitted in oral hearings. The expert concluded that technology would be found to solve the problem of nuclear waste disposal and that the ‘bugaboo’ of waste disposal should not hold up development of the nuclear industry even though no reliable techniques for disposal had yet been established.137 Bazelon characterised the role of the courts as being one to ensure that ‘genuine opportunities to participate in a meaningful way were provided’ and that there was a ‘thorough ventilation of issues’ through dialogue.138 He stated that: A reviewing court must assure itself not only that a diversity of informed opinion was heard, but that it was genuinely considered. ‘(T)he dialogue that the APA’s rulemaking section contemplates cannot be a sham.’.139

Procedure should focus not only on the facts but on a whole range of issues.140 Likewise, the respective role of policy and fact would depend on the issue at hand.141 In the case before him, he concluded such dialogue had not taken place 131

Bazelon, above n 121 at 211. AFL-CIO, above n 122 at 650. 133 Ethyl, above n 1 at 67; and Bazelon, above n 120 at 822. Also see T McGarity, ‘Substantive and Procedural Discretion in Administrative Resolution of Science Policy Questions: Regulating Carcinogens in EPA and OSHA’ (1979) 67 Georgetown Law Journal 729 at 797, and S Jasanoff, Science at the Bar: Law, Science and Technology in America (Cambridge, Harvard University Press, 1995) at 76–7. 134 Bazelon, above n 121 at 281. 135 Natural Resources Defense Council v Nuclear Regulatory Commission 547 F 2d 633 (DC Cir 1976). 136 Ibid at 637–8. 137 Ibid at 648–9. 138 Ibid at 644. 139 Ibid at 646, quoting from an article by Judge Skelly Wright. 140 Ibid at 645. 141 Ibid at 656. 132

(E) Fisher Ch3

31/5/07

106

09:57

Page 106

Hard Looks and Substantial Evidence

and that because the issue of nuclear waste would be such a long-term problem this was clearly not appropriate.142 The conclusions of the Nuclear Regulatory Commission had been based on ‘conclusory’ statements, and this was not acceptable where expert administration was leading the community in highly technical and weighty decisions with long-term consequences.143 Moreover, the expert who had given the advice had acknowledged at a later date that there were serious longterm problems with nuclear waste disposal.144 Bazelon stressed that there were many different ‘procedural devices’ which the Nuclear Regulatory Commission could have used, including formal conferences between intervenors and staff, document discovery, interrogatories, technical advisory committees comprised of outside experts with differing perspectives, limited cross examination, funding independent research by intervenors, detailed annotation of technical reports, surveys of existing literature, memoranda explaining methodology.145

These procedures were directed at the more technical aspects of decision-making but Bazelon also stressed in his analysis of the record (much of this in footnotes) the failure of the Nuclear Regulatory Commission to deal with the complex nature of the problems involved.146 He concluded his judgment by stating: It has become a commonplace among proponents of nuclear power to lament public ignorance. The public—the ‘guinea pigs’ who will bear the consequences of either resolution of the nuclear controversy—is apprehensive. But public concern will not be quieted by proceedings like the present.147

In other words, Bazelon was requiring decision-makers to engage in wide-ranging processes of analysis and deliberation. It is tempting to see the differences between Leventhal and Bazelon as a difference between a scientific and a democratic approach to risk regulation, or a substantive and a procedural approach to hard look review. Both those conceptualisations are wrong, however. The first of these, besides overlooking the administrative context of these debates, ignores the fact that both judges saw a role for science and participation, but characterised them differently. For Leventhal, science was defined in analytical terms, and participation could never amount to more than crude interest representation. For Bazelon, regulatory science was a complex and uncertain exercise, and deliberation a substantive problem-solving exercise. Likewise, both types of hard look review had substantive and procedural aspects, albeit Bazelon’s giving greater emphasis to the latter.148 In either case, 142

Natural Resources Defense Council v Nuclear Regulatory Commission at 652. Ibid at 651. 144 Ibid at 650. 145 Ibid at 653. 146 Ibid at 650–51. Also see Bazelon, ‘Coping With Technology Through the Legal Process’, above n 120 at 827 and 831, discussing why technical procedures are not enough. 147 Table S-3, above n 135 at 655. 148 Ibid at 660 per Judge Tamm. 143

(E) Fisher Ch3

31/5/07

09:57

Page 107

Substantial Evidence and the Occupational Safety and Health Act

107

what can be seen is that administrative constitutionalism is a thick legal discourse in which understandings of public administration were being co-produced with legal concepts. Indeed, it was the perception that Bazelon’s view of hard look review promoted adjudicative procedure which ultimately led to it being rejected by the US Supreme Court in Vermont Yankee Nuclear Power Corp v Natural Resources Defense Council (the Vermont Yankee decision),149 where it was held that courts should not, save in limited circumstances,150 impose extra procedural requirements, particularly of an adjudicative nature, on agencies above and beyond statutory requirements.151 The logic behind this decision was the de-shackling of administrative procedure from adjudicative method152 and to stop judicially imposed procedural requirements being a form of ‘Monday morning quarterbacking’153 in that petitioners could always argue that more procedural requirements should have been required. In rejecting the DC approach of Bazelon, however, this case was an implicit confirmation of Leventhal’s approach,154 and thus of the RI approach to judicial review, although that rejection was based less on an explicit debate about the role and nature of public administration in risk evaluation and more upon general questions concerning the role of adjudicative procedure in the administrative process. This reflects the fact that debates over administrative constitutionalism and risk regulation rarely take place in splendid isolation—they are embedded in more general debates over legitimate administrative governance.

IV Substantial Evidence and the Occupational Safety and Health Act The centrality of issues of administrative constitutionalism cannot be seen only in regard to general philosophies of judicial review as developed by the District of Columbia Circuit, but also in regard to different Federal Appeals courts struggling to make sense of the legislative schemes for judicial review that were included in the new risk regulation regimes. A prime example of this was the judicial review of OSHA standard setting under the OSH Act. Under that Act, OSHA (acting on behalf of the Secretary of Labour) was given a general power to set occupational health and safety standards. A standard was defined as: 149

Vermont Yankee, above n 72. Where the court was required to afford constitutional due process or where there had been a ‘totally unjustified’ departure from normal agency procedure: ibid at 542. 151 Ibid at 543. 152 A Scalia, ‘Vermont Yankee: The APA, the DC Circuit, and the Supreme Court’ (1978) Supreme Court Review 345 at 378; and M Shapiro, Who Guards the Guardians: Judicial Control of Administration (Athens, University of Georgia Press, 1988) at 164. 153 Vermont Yankee, above n 72 at 547. 154 Scalia, above n 152 at 354. 150

(E) Fisher Ch3

31/5/07

108

09:57

Page 108

Hard Looks and Substantial Evidence

a standard which requires conditions, or the adoption or use of one or more practices, means, methods, operations or processes, reasonably necessary or appropriate to provide safe or healthful employment and places of employment.155

This was a relatively open-ended definition. There were a number of different types of standards that OSHA could adopt including those in relation to toxic materials which were regulated by §655(b)(5). That section stated: The Secretary, in promulgating standards dealing with toxic materials or harmful physical agents under this subsection, shall set the standard which most adequately assures, to the extent feasible, on the basis of the best available evidence, that no employee will suffer material impairment of health and functional capacity even if such employee has regular exposure to the hazard dealt with by such a standard for the period of his working life. Development of standards under this subsection shall be based on research, demonstrations, experiments and such information as may be appropriate. In addition to the attainment of the highest degree of health and safety protection for the employee, other considerations shall be the latest available scientific data in the field, the feasibility of standards and experience gained under this and other health and safety laws. Wherever practicable, the standard promulgated shall be expressed in terms of objective criteria and of the performance desired.

The rule-making procedure under the OSH Act was hybrid. Thus while essentially informal it allowed for public hearings and consultation with advisory committees.156 The standard of judicial review, however, was the ‘substantial evidence’ test.157 As seen above, the substantial evidence test under the APA normally applied to formal rule-making where there was an adjudicative hearing in which the presiding officer was unbiased and the rules of evidence were applicable.158 The ‘substantial evidence’ in question was that on a formal adjudicative record and, as such, the ground of review was traditionally understood as the ‘handmaiden of formal procedures’.159 It was difficult to apply to informal rule-making because there was no formal hearing and because no physical administrative record was produced.160 Most significantly, the test evoked an understanding of public administration in RI terms, while informal rule-making was understood in largely DC terms. DC thinking heavily influenced early judicial review of OSHA rule-making.161 Standard setting was understood as complex and needing the substantive exercise 155

29 USC §652(8). 29 USC §655(b)(2)-(4). 157 29 USC §655(f). 158 Aqua Slide N’ Drive Corp v CPSC 569 F 2d 831 (5th Cir 1978) at 837. 159 Note, ‘Judicial Review under the Occupational Safety and Health Act: The Substantial Evidence Test as Applied to Informal Rulemaking’ (1974) Duke Law Journal 459 at 462. 160 Citizens to Preserve Overton Park Inc v Volpe 401 US 402 (1971); and C Auberach, ‘Informal Rulemaking: A Proposed Relationship Between Administrative Procedures and Judicial Review’ (1977) 72 North Western University Law Review 15 at 16 and 23. 161 Industrial Union Dept, AFL-CIO v Hodgson 499 F 2d 467 (DC Cir 1974); and Synthetic Organic Chemical Manufacturers Association v Brennan 503 F 2d 1155 (3rd Cir 1974) (the Ethyleneimine decision). 156

(E) Fisher Ch3

31/5/07

09:57

Page 109

Substantial Evidence and the Occupational Safety and Health Act

109

of discretion on the part of the agency.162 The courts recognised that the task of OSHA lay on the ‘frontiers of scientific knowledge’ and therefore ‘explicit factual findings were not possible’.163 Administrators were expected to rely heavily on policy, judgement, and deliberation in making decisions.164 Thus, for example, extrapolation from animals to humans was seen as a policy issue which was not required to be grounded in fact.165 Moreover, the public interest of worker protection would be of overriding importance in making a decision.166 The courts stressed that the inclusion of advisory committees and the public participation requirements were also important in an area with so much scientific uncertainty.167 The court shaped its review accordingly and it stressed that the informal rulemaking provisions of the OSH Act were an important signifier that Congress did not require methodological rigour.168 It also interpreted the statute in a broad, purposive manner, and saw that constraint was to be provided through a deliberative reasoning process. Thus in Society of Plastics Industry Incorporated v OSHA169 (the Society of Plastics Industry decision) the court upheld a standard relating to vinyl chloride even though there was no proof to justify a standard set at such a low level. In actual fact, there was no evidence as to its safety at that level, but in light of the possible consequences of exposure, OSHA had set the lowest standard feasible.170 Judge Clark, in upholding the standard, noted that the aim of the OSH Act was the protection of human lives and that this should be the driving rationale of any rule-making process. Moreover—he commented, and though the factual finger points, it does not conclude. Under the command of OSHA, it remains the duty of the Secretary to act to protect the working man, and to act even in circumstances where existing methodology or research is deficient.171

Thus, in cases where there could not be accurate calculation, the courts construed the boundaries of the discretion in accordance with the public interest that agencies such as OSHA were seeking to protect.172 Thus, for example, the statutory hurdle of ‘feasibility’ as stated in §655(b)(5), was also interpreted broadly.173 162 Ethyleneimine, above n 161 at 1159–60; Society of Plastics Industry Inc v OSHA 509 F 2d 1301 (2nd Cir 1975) at 1308; and Hodgson, above n 161 at 474. 163 Hodgson, above n 161 at 474. 164 American Iron & Steel Institute v OSHA 577 F 2d 825 (3rd Cir 1978) at 831; and Ethyleneimine, above n 161 at 1159. 165 Ethyleneimine, above n 161 at 1159–60; and Society of Plastics Industry Inc, above n 162 at 1308. 166 Hodgson, above n 161 at 475. 167 Synthetic Organic Chemical Manufacturers Association v Brennan 506 F 2d 385 (3rd Cir 1974) (the MOCA decision) at 389. 168 Hodgson, above n 161 at 475; Associated Industries of New York Inc v US Dept Of Labour 487 F 2d 342 (5th Cir 1973) at 344–5; and American Federation of Labour v Brennan 530 F 2d 109 (3rd Cir 1975) at 115–16. 169 Society of Plastics Industry Inc, above n 162. 170 Ibid at 1308. 171 Ibid at 1308. 172 Hodgson, above n 161 at fn 18, referring to the rate-making case of In Re Permian Basin Area Rate Cases 390 US 747(1968). 173 This requirement arose out of the wording of 29 USC §655(b)(5) of the OSH Act.

(E) Fisher Ch3

31/5/07

110

09:57

Page 110

Hard Looks and Substantial Evidence

This is not to say that OSHA could act on no evidence, but the evidentiary burden was not always a large one due to the courts understanding that evidentiary requirement in a wider context.174 In Synthetic Organic Chemical Manufacturers Association v Brennan (the Ethyleneimine decision)175 the Third Circuit held that two animal studies amounted to substantial evidence.176 Moreover, in Industrial Union Department, AFL-CIO v Hodgson (the Hodgson decision)177 the District of Columbia Circuit held that while the expert opinions of the National Institute of Occupational Safety and Health (NIOSH) were relevant they did not need to be directly followed.178 OSHA should exercise its discretion in a responsible and flexible fashion and need not strictly comply with advice. In particular, it was for OSHA to decide what was practical.179 Rather than being preoccupied with questions of evidence, the courts were particularly concerned to ensure that deliberation had taken place.180 In the Ethyleneimine case, therefore, the decision was partly remanded because of the failure of OSHA to give notice for specific provisions.181 Likewise, in Dry Colour Manufacturers Association Inc v Department of Labour the court remanded the decision because the reasons were inadequate.182 Moreover, in another case, Synthetic Organic Chemical Manufacturers Association v Brennan,183 the case was partly remanded because the notice and comment requirements had not been properly complied with. In all these cases the problem was that OSHA had not ensured there had been a proper ventilation and consideration of the issues. At the same time, courts were also attempting to make sense of the legislative scheme for standard of review. The primary response of the courts to the mismatch of standard of review and rule-making procedure was one of frustration, and it was described as an ‘illogic of legislative compromise’.184 Judge McGowan in Hodgson was highly critical of this combination, describing it as adding to the ‘angularity of the relationship’ between the courts and OSHA and imposing ‘additional burdens upon the court’.185 Many judges viewed the substantial evidence test as nearly impossible to apply.186 The major conceptual barrier for the courts 174 American Federation of Labour, above n 168 at 135; American Iron & Steel Institute, above n 164 at 835; and Society of Plastics Industry Inc, above n 162 at 1309. 175 Ethyleneimine, above n 161. 176 Ibid at 1157. 177 Hodgson, above n 161. 178 Ibid at 477. 179 MOCA, above n 167 at 391. 180 Ethyleneimine, above n 161 at 1159–60; Hodgson, above n 161 at 474; and American Iron & Steel Institute, above n 164. 181 Ethyleneimine, above n 161 at 1160. 182 Dry Colour Manufacturers Association Inc v Department of Labour 486 F 2d 98 (3rd Cir 1973) at 104–5. 183 MOCA, above n 167. 184 Hodgson, above n 161 at 469. Also see Ethyleneimine, above n 161 at 1157. 185 Hodgson, above n 161 at 469. 186 Society of Plastics Industry Inc, above n 162 at 1304; and Ethyleneimine, above n 161 at 1157. For other discussions of the problems see Associated Industries of New York, above n 168 at 347–50; Florida Peach Growers Association v US Department of Labour 489 F 2d 120 (5th Cir 1974) at 127–9; American Petroleum Institute v OSHA 581 F 2d 493 (5th Cir 1978) at 497.

(E) Fisher Ch3

31/5/07

09:57

Page 111

Substantial Evidence and the Occupational Safety and Health Act

111

was that the record produced under the hybrid provisions of the OSH Act did not yield easily to a substantial evidence analysis. Moreover, the test bore little relationship to what was understood as the nature of the rule-making process in relation to risk evaluation. Judge McGowan in the Hodgson case described the record in that case as consisting in the main: of a melange of written statements, letters, reports, and similar materials received outside the bounds of the oral hearing and untested by anything approaching the adversary process.187

Likewise in Society of Plastics the review of the 4,000-page record was described as a prodigious task . . . aggravated by duplications of testimony, irrelevant exhibits and letters, almost illegible reproductions of documents and a generally blunderbuss approach in petitioner’s briefs.188

The labyrinthine nature of the rule-making record was not merely due to it being an informal process but also due to the fact that in areas of scientific controversy any record on which a decision was based would be an intricate web of conflicting scientific information, deliberation, and policy.189 There were two solutions developed to address this mismatch, both of which pointed in the direction of the RI paradigm because both emphasised the factual basis of decision-making. In both cases it was legal concepts that were affecting what was understood to be the role and nature of public administration. First, it was argued that the substantial evidence test required more explicit agency articulation and stringent review than that under the arbitrary and capricious test.190 On this analysis, the distinction between the substantial evidence test and the arbitrary and capricious test was merely ‘semantic’ and about intensity.191 The practical result of this interpretation of the test, however, was to require decision-makers to justify their decisions in factual terms.192 A second interpretation was that the substantial evidence test required that the facts on which the agency was basing its decision needed to be proved with a greater degree of certainty than in normal informal rule-making.193 The substantial evidence test applied to the judicial review of the facts, and the arbitrary and capricious standard would apply to the 187

Hodgson, above n 161 at 474. Society of Plastics Industry Inc, above n 162 at 1304. See Florida Peach, above n 186 at 129. 189 S Jasanoff, ‘The Problem of Rationality in American Health and Safety Regulation’ in Smith and Wynne (eds) Expert Evidence: Interpreting Science in the Law (London, Routledge, 1989) at 156. 190 National Grain and Feed Association v OSHA 866 F 2d 717 (5th Cir 1989) at 728; and Associated Industries of New York, above n 168 at 348. 191 Associated Industries of New York, above n 168 at 349–50; American Petroleum Institute, above n 186 at 497; and Texas Independent Ginners Association v Marshall 630 F 2d 398 (5th Cir 1980) at 404. 192 Dry Colour Manufacturers Association, above n 182 at 104–105. 193 Mobil Oil Corp v FPC 483 F 2d 1238 (DC Cir 1973) at 1258–9. Some doubted whether this was the case. See A Scalia and F. Goodman, ‘Procedural Aspects of the Consumer Product Safety Act’ (1973) 20 University of California Law Review 899 at 934; American Iron & Steel Institute, above n 164 at 831; Ethyleneimine, above n 161 at 1159. 188

(E) Fisher Ch3

31/5/07

112

09:57

Page 112

Hard Looks and Substantial Evidence

rest of the decision.194 Yet even under this non-adjudication interpretation of the test there were still serious problems with carrying out judicial review in this manner. Judge Staley in the MOCA case195 commented that applying the substantial evidence test to informal rule-making was a ‘good illustration of the difficulty in attempting to measure a legislative policy decision against a factual yardstick’. Moreover, the courts appreciated that OSHA’s decision-making was ‘essentially a prediction based upon pure legislative judgement’, and that rule-making not only required ‘factual findings subject to evidentiary development’ but also ‘legislative like policy judgements’.196 An attempt to make a distinction between fact and policy was not only artificial but in some cases impossible.197 The perception was that to carry out judicial review, however, such a distinction needed to be made.198 Increasingly, then, the DC approach of the courts was replaced with a concern that decision-makers establish that there was some factual basis to their decision. In Texas Independent Ginners Association v Marshall199 the Fifth Circuit once again struck down a decision on the basis that it lacked substantial evidence. This time it was in relation to a cotton dust standard for the cotton ginning industry. The court described its task as involving the scrutiny of a number of matters. These included: (a) whether enough facts are available and have been investigated to render rational the making of a policy judgement; (b) whether the factual premises underlying that policy judgement are ‘supported by substantial evidence’ although there may be other conflicting evidence in the record considered as a whole; and (c) whether the policy judgement is reasonably related to those substantially supported factual premises so that ‘the Secretary carried out his essentially legislative task in a manner reasonable under the state of the record’.200

In the language of Leventhal, OSHA now had a heavy informational burden to shift before it could regulate.201 The task of OSHA was being conceptualised in RI terms as one of an expert fact-finder carrying out analysis so as to apply facts to a specific legislative mandate. The emphasis was on knowledge collection and analysis rather than deliberation. In American Petroleum Institute v OSHA (the American Petroleum decision)202 the Fifth Circuit struck down OSHA’s standard in relation to benzene stating that:

194 American Iron & Steel Institute, above n 164 at 831; Hodgson, above n 161 at 473; and Associated Industries of New York, above n 168 at 347. Note this was bolstered up with an interpretation of ‘determination’ in 29 USC §655(f). 195 MOCA, above n 167 at 1158. 196 American Petroleum Institute, above n 186 at 497. 197 Society of Plastics Industry Inc, above n 162 at 1304. 198 Hodgson, above n 161 at 474. 199 Texas Independent Ginners Association, above n 191. 200 Ibid at 405. 201 United Steelworkers of America v Auchter 763 F 2d 728 (3rd Cir 1985) at 763. 202 American Petroleum Institute, above n 186.

(E) Fisher Ch3

31/5/07

09:57

Page 113

Substantial Evidence and the Occupational Safety and Health Act

113

OSHA’s failure to provide an estimate of expected benefits for reducing the permissible exposure limit, supported by substantial evidence, makes it impossible to assess the reasonableness of the relationship between expected costs and benefits. This failure means the required support is lacking to show reasonable necessity for the standard promulgated.203

OSHA decided to regulate at the lowest extent feasible, and to this end it set the standard at one part per million (ppm) even though they had no evidence (of either safety or harm) at any level below 10 ppm. It did so in accordance with its generic carcinogenic policy,204 and a similar approach in relation to its vinyl chloride standard had been upheld by the Second Circuit in Society of Plastics.205 American Petroleum was appealed to the US Supreme Court where it became known as the Benzene decision.206 The Supreme Court affirmed the Fifth Circuit’s decision by a five-to-four majority, including a plurality opinion written by Justice Stevens. This judgment entrenched an RI understanding of risk regulation not only in relation to OSHA but also more generally. Indeed, Justice Rehnquist in his concurring judgment attempted to revive the non-delegation doctrine, a highly RI concept, as it ensured that ‘important choices of social policy’ were made by Congress.207 While the plurality decision was not explicit in its reliance on the non-delegation doctrine it was clearly influenced by it, and the judgment was an attempt to ‘define more precisely the specific contours of the OSH Act’.208 In particular, the plurality opinion can be understood as ensuring that the OSH Act did not give ‘unbridled’ discretion.209 This was in two ways. First, in an act of highly creative statutory interpretation, the majority required that OSHA establish a threshold of ‘significant risk’ before regulating.210 The plurality argued that the test was implicit in the words ‘reasonably necessary’ as set out in §652(8) and that it was required because otherwise §652(8) would be meaningless. As Justice Stevens stated: In the absence of a clear mandate in the Act, it is unreasonable to assume that Congress intended to give the Secretary the unprecedented power over American industry that would result from the Government’s view of s 3(8) [§652(8)] and s 6(b)(5) [§655(b)(5)], coupled with OSHA’s cancer policy.211

Underlying such a statement was, of course, a highly RI understanding of public administration. Justice Stevens also attempted to define the concept of ‘significant 203

Ibid at 505. OSHA, ‘Identification, Classification, and Regulation of Potential Occupational Carcinogens’ 45 Federal Register 5002 (22 January 1980). 205 Society of Plastics Industry Inc, above n 162. 206 Industrial Union Dept AFL-CIO v American Petroleum Institute 448 US 607 (1980). 207 Ibid at 685–6. As in the 1930s those promoting the non-delegation doctrine were ultimately unsuccessful. See R Stewart, ‘Beyond Delegation Doctrine’ (1987) 36 American University Law Review 323; and American Trucking Associations v EPA 175 F 3d 1027. 208 Public Health Research Group v Tyson 796 F 2d 1479 (DC Cir 1986) at 1484. 209 Benzene, above n 206 at 614 per Justice Stevens, and at 662 and 663 per Chief Justice Burger. 210 Ibid at 639–40 per Justice Stevens, at 662 per Chief Justice Burger, and at 665 per Justice Powell. 211 Ibid at 645. 204

(E) Fisher Ch3

31/5/07

114

09:57

Page 114

Hard Looks and Substantial Evidence

risk’ more accurately, and while he stated that it should not be a ‘mathematical straitjacket’,212 he suggested that if the odds are one in a thousand that regular inhalation of gasoline vapors that are 2% benzene will be fatal, a reasonable person might well consider the risk significant and take appropriate steps to decrease or eliminate it.213

Defining a ‘significant risk’ in quantifiable terms was attractive because it made the legislative boundaries more precise.214 The authority of OSHA was thus being derived from science and, in particular, the identification of specific type of risks. Second, the requirement of this new factual threshold was also accompanied by a confirmation that the substantial evidence test was to be understood in RI terms. Justice Stevens, while recognising the problem of scientific uncertainty, stated: Thus so long as they are supported by a body of reputable scientific thought, the Agency is free to use conservative assumptions in interpreting data with respect to carcinogens, risking error on the side of over protection rather than under protection.215 [emphasis supplied]

Substantial evidence was defined in relation to scientific knowledge, and the quality of the substantial evidence was to be judged on the quality of the science. The starting point for analysis was scientific evidence, and reliance on policy was restricted to marginal cases.216 In this case, Justice Stevens argued that OSHA had not even attempted to ‘carry its burden of proof’.217 The prime example he gave of such failure was the fact that OSHA rejected industry testimony that a dose response curve could be constructed which suggested benzene exposure presented only a de minimis risk at low levels of exposure. In doing so, OSHA did not scrutinise this evidence carefully218 and thus its decision was not underpinned by substantial evidence. He disallowed OSHA’s attempt to set a limit for the exposure to benzene in the workplace at 1 ppm, because there was only evidence of a ‘significant risk’ at 10 ppm. There was no discussion of the problems of scientific uncertainty, or the consequences of such a risk, or the policy of worker protection. Under the majority’s interpretation of the substantial evidence test, ‘evidence on the record’ was now equated solely with scientific evidence and not any information on the record whatever its form it might take.219 The concept of substantial evidence was thus de-shackled from adjudicative procedure, but in so being was interpreted in profoundly RI terms. 212

Benzene, above n 206 at 655. Ibid at 655. 214 Ibid at 646 per Justice Stevens, and at 663 per Chief Justice Burger. 215 Ibid at 656. 216 On this see E Schwartzbauer and S Shindell, ‘Cancer and the Adjudicative Process: The Interface of Environmental Protection and Toxic Tort Law’ (1988) 14 American Journal of Law and Medicine 1 at 22. 217 Benzene, above n 206 at 653, such language being reminiscent of Judge Leventhal. 218 Ibid at 654. 219 Only Justice Powell (dissenting on this point) took the more traditional ‘evidence on the record’ approach. 213

(E) Fisher Ch3

31/5/07

09:57

Page 115

Substantial Evidence and the Occupational Safety and Health Act

115

In his dissenting opinion, Justice Marshall described the plurality’s opinion as ‘extraordinarily arrogant and extraordinarily unfair’.220 It was arrogant because, he argued, they were effectively engaging in de novo review. It was unfair because: its characterization of the Secretary’s report bears practically no resemblance to what the Secretary actually did in this case.221

In contrast to the majority’s characterisation, Justice Marshall reasoned from a DC starting point. From the outset he described the problem that OSHA was attempting to handle.222 Moreover, he summarised both the factual and policy basis of the standard.223 He constantly stressed that consideration of both aspects was a vitally important aspect of the exercise of responsible administrative discretion in this area.224 While the majority seemingly feared that any approach not grounded in the RI paradigm would lead to unchecked discretion, Justice Marshall’s decision is a classic example of a searching review under the DC paradigm. One of Justice Marshall’s primary concerns was that OSHA had not ‘blindly’ relied on its generic carcinogen policy.225 Rather, he required deliberation and proof that OSHA had conscientiously considered the issues. He argued that this had taken place because, although there was no disagreement that the policy should apply, OSHA still gathered over 50 volumes of evidence and conducted 17 days of hearings. The witnesses included were not only scientists but political economists and members of the workforce. It was not so much the quantity of the evidence collected which concerned him, but what OSHA considered. Thus he examined in detail OSHA’s technical and policy reasons, and what they had taken into account. This included the type of uncertainties involved, including the fact that leukaemia had a long latency period and that, although the evidence was poor, it did suggest that the risk would be reduced by a lowering in exposure levels. Moreover, from a feasibility perspective, in many cases industries were complying with the 1 ppm level already, and/or replacing benzene with other substances. In light of this searching analysis, Justice Marshall upheld OSHA’s standard. Marshall’s dissent fell on deaf ears, however. This was even though the reasoning in it was similar to that in the early case-law. Rather, scope of review of OSHA decision-making was now firmly on an RI basis—a state of affairs that had as much to do with judges attempting to make sense of illogical legislative frameworks than with an explicit decision to embrace the RI paradigm.

220 221 222 223 224 225

Benzene, above n 206 at 695. Ibid at 695. Ibid at 691. Ibid at 689. Ibid at 706. Ibid at 695.

(E) Fisher Ch3

31/5/07

116

09:57

Page 116

Hard Looks and Substantial Evidence

V Scope of Review under the RI Paradigm By the mid-1980s, the dominance of the RI paradigm was not simply in regard to judicial review but also in other areas of administrative law and policy.226 The Presidential Office of Management and Budget (OMB), under the conservative Reagan administration, introduced requirements for regulatory impact statements.227 There was also the development of new analytical methodologies such as comparative risk assessment,228 and in the mid-1990s there were attempts to legislate risk assessment methodology, which resulted in the Data Quality Act.229 But with that said, these scope of review decisions made an important contribution to received understandings about what was ‘reasonable’ administrative action in relation to risk evaluation. Benzene led directly to the National Research Council (NRC) report Risk Assessment in the Federal Government: Managing the Process,230 known as the Red Book. That report became the primer for all risk regulatory agencies and divided standard setting up into a scientific process of risk assessment and a political process of risk management. As well, Benzene was interpreted as requiring risk regulators to establish at least a 1 in 1000 risk before regulating.231 Most importantly, the Benzene decision, along with the other doctrinal developments in judicial review discussed above, saw a dramatic shift in how decision-makers justified their decisions. Standard setting was increasingly being understood as primarily an analytical process in which there was little role for professional judgement, deliberation, or administrative policy choice. The administrative lawyer Kenneth Culp Davis writing in 1980 noted Leventhal’s comment in Portland Cement 232 about decision making requiring a factual basis and then went on to note: The statement probably was not the law before it was written. It may not yet be the general law, but it maybe in the process of coming so.233 226 E Fisher, ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds), Environmental Law for Sustainability (Oxford, Hart Publishing, 2006); and Fisher, above n 3. 227 Order 12,291 replaced by Order 12,866. See generally T McGarity, Reinventing Rationality: The Role of Regulatory Analysis in the Federal Bureaucracy (Cambridge, Cambridge University Press, 1991). 228 See particularly United States Environmental Protection Agency, Unfinished Business: A Comparative Assessment of Environmental Problems (Washington DC, 1987), and Science Advisory Board, Reducing Risk: Setting Priorities and Strategies for Environmental Protection (Washington DC, US Environmental Protection Agency, 1990). 229 C Sunstein, Free Markets and Social Justice (New York, Oxford University Press, 1997) at ch 14; and M Lacko, ‘The Data Quality Act: Prologue to a Farce or a Tragedy?’ (2004) 53 Emory Law Journal 305. 230 National Research Council, Risk Assessment in the Federal Government: Managing the Process (Washington DC, National Academy Press, 1983). 231 International Union, UAW v Pendergrass 878 F 2d 389 (DC Cir 1989) (the Formaldehyde decision) at 392; ASARCO v OSHA 647 F 2d 1 (9th Cir 1981); and National Research Council, Science and Judgement in Risk Assessment (Washington DC, National Academy Press, 1994) at 36. 232 Portland Cement, above n 103 at 393. 233 KC Davis, ‘Facts In Law-Making’ (1980) 80 Columbia Law Review 931 at 931.

(E) Fisher Ch3

31/5/07

09:57

Page 117

Scope of Review under the RI Paradigm

117

What can be seen from the 1980s onwards is a consolidation of scope of review doctrine in this area on the basis of the RI paradigm. This shift did not necessarily result in more or fewer decisions being struck down but rather was a change in the way in which risk regulation was characterised and thus the way in which administrative decision-makers understood their role.234 Leventhal’s ‘hard look’ doctrine was given explicit Supreme Court approval in Motor Vehicles Manufacturers Association v State Farm Mutual Automobile Insurance Company.235 Risk regulators were construed in RI terms and thus the task of the courts was to police the methodology and analysis involved in decisionmaking. Likewise, as this body of judicial review developed, the courts also became more concerned with reviewing the validity of particular assessments whether they be risk assessments,236 cost/benefit analyses,237 or comparative risk assessments.238 In all cases, the courts found that there should be little room for policy or professional judgement. Thus, for example, OSHA was now required to ‘demonstrate substantial evidence for all matters of determinable fact’ and articulate its reasons for choosing among scientific predictions.239 The courts would, among other things: inquire into methodology in great detail;240 look into the quality of the studies underpinning risk assessments;241 assess the experimental conditions of studies;242 and compare studies of OSHA with those of petitioners.243 In some cases they would also stress the importance of peer review,244 but in general this was a second-place concern. While scientific uncertainty was still accepted as a problem it merely made quantification more difficult.245 Needless to say, administrative records needed to be carefully ordered.246

234 For discussions of the statistics in relation to the outcome of cases, see generally D Elliott et al, ‘Science, Agencies and the Courts: Is Three a Crowd? ’ (2001) 31 Environmental Law Reporter 10125, and P. Wald, ‘Regulation At Risk: Are Courts Part of The Solution or Most of the Problem?’ (1994) 67 Southern California Law Review 621. 235 Motor Vehicles Manufacturers Association v State Farm Mutual Automobile Insurance Company 463 US 29 (1983). 236 Flue-Cured Tobacco Co-op v EPA 4 F Supp 2d 435 (MD NC, 1998) 237 American Dental Association v Martin 984 F 2d 823 (7th Cir 1993) 238 Competitive Enterprise Institute v NHTSA 901 F 2d 107 (DC Cir 1990). 239 Formaldehyde, above n 231 at 492. 240 Asbestos Information Association of North America v OSHA 727 F 2d 415 (5th Cir 1984) at 421. 241 Ibid at 421. See also Gulf South Insulation v CPSC 701 F 2d 1137 (5th Cir 1983) at 1146. 242 Formaldehyde, above n 231 at 393; and ASARCO v OSHA 746 F2d 483 (9th Cir 1984) at 493. 243 National Grain and Feed Association, above n 190 at 740. 244 Asbestos Information Association of North America, above n 240 at 421. 245 Builders and Construction Trades Department v Brock 838 F 2d 1258 (DC Cir 1988) at 1264; Asbestos Information Association of North America, above n 240 at 425; Formaldehyde, above n 231 at 392; and United Steelworkers Of America v Marshall 647 F 2d 1189 (DC Cir 1980) at 1206–7; and National Grain and Feed Association, above n 190 at 729. 246 American Dental, above n 237 at 833, arguing the record in that case was a ‘hodge podge of findings’.

(E) Fisher Ch3

31/5/07

118

09:57

Page 118

Hard Looks and Substantial Evidence

An example of all of this can be seen in AFL-CIO v OSHA (the Air Contaminants decision),247 where the Eleventh Circuit vacated a generic air contaminants standard for 428 toxic substances on the ground that OSHA had not based the standard on substantial evidence. The standards updated recommendations of the American Conference of Governmental Industrial Hygienists (ACGIH) which were new or more protective than standards already in place. The ACGIH has been a historical and well-respected source for such standards, and this was an attempt by OSHA to update regulation in this area in light of new evidence. If rule-making had proceeded on a case-by-case basis many workers would not have been protected. Although the standard prescribed different ‘permissible exposure levels’ (PELs) for each substance, the method of reduction was in many cases the same. Moreover, OSHA was passing the standard knowing that it would need to be updated again in several years’ time to take into account new research.248 The ACGIH, however, did not base their PELs on risk assessments.249 Judge Fay, for the court, stated that while these standards could theoretically be set through a generic rule-making, substantial evidence was required for each separate substance.250 As such, OSHA would need to establish in relation to each substance that there was substantial evidence of a significant risk, that the proposed standard would reduce that risk, and that the standard was both technologically and economically feasible. This large methodological burden offset any advantages OSHA would gain through a generic standard. On the first ground, the court found that OSHA had not made a finding of significant risk. OSHA had grouped the substances into 18 different categories depending on their health effects and then discussed a number of studies. It broadly concluded that the standard would result in saving 683 lives and avoiding 55,000 occupational illnesses annually.251 Moreover, the court found that OSHA had not established that the standards would reduce a risk. Rather, OSHA, taking into account scientific uncertainty, had factored in a number of ‘safety factors’ to each of the PELs. The safety factor differed with each substance depending on the quality of the data. This use of differentiated safety factors had not only been condoned by other courts but actively promoted as being a more accurate way of accounting for scientific uncertainty.252 The court here, however, rejected the use of safety factors when the evidence was incomplete or inadequate. The court stated: The lesson of Benzene is clearly that OSHA may use assumptions, but only to the extent that those assumptions have some basis in reputable scientific evidence. If the agency is 247AFL-CIO v OSHA 965 F 2d 962 (11th Cir 1992). For critical discussions of this case see A Scordo, ‘A No-Frills Approach to Rulemaking Leaves Workers at Risk: OSHA’s Failed Attempt to Enact “Generic” Air Contaminants Standard’ (1994) 16 Journal of Products and Toxics Liability 167, and K Strader, ‘OSHA’s Air Contaminants Standard Revision Succumbs to Substantial Evidence Test’ (1993) 62 University of Cincinnati Law Review 351. 248 Air Contaminants, above n 247 at 977. 249 62 Federal Register 1498 (19 January 1997). 250 Air Contaminants, above n 247 at 972. 251 Ibid at 975. 252 Leather Industries of America v EPA 40 F 3d 392 (DC Cir 1994) at 400.

(E) Fisher Ch3

31/5/07

09:57

Page 119

Scope of Review under the RI Paradigm

119

concerned that the standard should be more stringent than even a conservative interpretation of the existing evidence supports, monitoring and medical testing may be done to accumulate the additional evidence needed to support that more protective limit. Benzene does not provide support for setting standards below the level substantiated by the evidence.253

This statement is a strong restatement of the Benzene decision and thus of the RI paradigm. Not only did the decision have to be primarily grounded in scientific knowledge but the assumptions taken into account also needed to be as well. Broad policies of worker protection had no role. ‘Substantiated by the evidence’ had thus become a stringent test in which proof was required before any regulatory action could be taken. The courts did recognise a ‘zone of reasonableness’, just as the courts had done in the New Deal, but rather than a zone of professional discretion it was a zone of estimation. ‘Leeway’ was given but in the carrying out of a scientific task.254 The nature and role of public administration was being transformed by the changing definitions of scope of review. The emphasis on substantial scientific evidence was not just in regard to OSHA, however, and the standard-setting practices of all agencies were also subject to such analysis. Arguments of petitioners included: the data underlying the risk assessment was incomplete or flawed;255 the agency did not make its findings in relation to specific industries and specific geographical areas;256 the interpretation of the data was incorrect;257 the methodology used inappropriate;258 the mismanagement scenario was unrealistic;259 the statistical analysis poor;260 and the agency misunderstood the nature of the scientific problem.261 Moreover, the courts became increasingly concerned to ensure that the analysis of agencies in risk evaluation encompassed other physical risks. Thus, in Corrosion Proof Fittings v EPA262 the court was not only concerned with the methodological rigour of the cost/benefit analysis but also with what other risks the agency had considered and, in particular, what risks any specific standard might create. Thus it found that ‘the EPA’s explicit failure to consider the toxicity of likely substitutes

253

Air Contaminants, above n 247 at 979. United Steelworkers, above 245 at 1253; National Grain and Feed Association, above n 190 at 738; Forging Industry Association v Secretary of Labour 773 F 2d 1436 (4th Cir 1985) at 1443; and Air Contaminants, above n 247 at 977. 255 Gulf South Insulation above n 241 at 1145; and Natural Resources Defense Council v EPA 824 F2d 1211 (DC Cir 1987) at 1216. 256 Colour Pigments Manufacturers Association v OSHA 16 F 3d 1157 (11th Cir 1994) at 1162; and Love v Thomas 858 F 2d 1347 (9th Cir 1988) at 1360. 257 American Lung Association v EPA 134 F 3d 388 (DC Cir 1998) at 392; and Central Arizona Water Conservation District v EPA 990 F 2d 1531 (9th Cir 1993) at 1542. 258 Natural Resources Defense Council v Thomas 805 F 2d 410 (DC Cir 1986) at 425. 259 Dithiocarbamate Task Force v EPA 98 F 3d 1394 (DC Cir 1996) at 1400; and Edison Electric Institute v EPA 2 F 3d 438 (DC Cir 1993) at 446–7. 260 Chemical Manufacturers Association v EPA 870 F 2d 177 (5th Cir 1989) at 184. 261 New York v EPA 133 F 3d 987 (7th Cir 1998) at 989; and Leather Industries of America, above n 252 at 405. 262 Corrosion Proof Fittings v EPA 947 F 2d 1201 (5th Cir 1991). 254

(E) Fisher Ch3

31/5/07

120

09:57

Page 120

Hard Looks and Substantial Evidence

thus deprives its order of a reasonable basis’.263 While the court argued that this did not mean that the agency had to search out and analyse all substitutes, it did mean that any ‘credible’ evidence about the risks of substitutes brought to its attention needed to be analysed and a rational weighing-up of the factors engaged in. Likewise, in Competitive Enterprise Institute v NHTSA264 the court remanded a proposed fuel economy standard because the petitioners had introduced evidence which suggested that if the fuel economy standard was lowered (which was what NHTSA intended to do) manufacturers would be required to produce smaller cars, which would increase safety risks. The majority found that the agency had not properly analysed the trade-off. The logic behind this approach to the review of the substantive exercise of discretion was highly RI. The task of the agency was understood as one of reducing risk in numerical terms. All risks were treated the same and, as noted in Corrosion Proof, a ‘death is a death’.265 There was little appreciation of the complexity of technological risks. The focus on analysis did not mean that there was no role for communication and public participation, but the courts understood the role of public participation in RI terms—that is, as a form of interest representation. As such, they needed to ensure that such representation played a role in information collection and improving methodology, but did not result in agency capture.266 Indeed, informal rule-making, once the cornerstone of the DC paradigm, was recast as a series of one-way communication processes rather than a process in which complex questions were ‘resolved in the crucible of debate’.267 This meant that courts placed less emphasis on deliberation and far more on methodology.268 The failure to respond to comments was only important insofar as those comments may have given rise to relevant information that the agency should have factored into its analysis.269 Moreover, courts were concerned to ensure that rule-making should not be railroaded by ‘amorphous public extravaganzas’270 and that rules should not be the product of ‘public hysteria’.271

263

Corrosion Proof Fittings v EPA 947 F 2d 1201 (5th Cir 1991) at 1221. Competitive Enterprise Institute v NHTSA 956 F 2d 321 (DC Cir 1992). 265 Corrosion Proof Fittings, above n 262 at 1221. 266 Natural Resources Defense Council v EPA 822 F2d 104 (DC Cir 1987) at 104; Thompson v Clark 741 F 2d 401 (DC Cir 1984) at 409; and American Waterworks Association v EPA 39 ERC 1897 (DC Cir 1994) at 1903. 267 International Harvester, above n 100 at 651. 268 In Corrosion Proof Fittings, above n 262 at 1211 there is a downplaying of deliberation in favour of extensive methodological analysis. 269 American Iron & Steel Institute v EPA 115 F 3d 979 (DC Cir 1997) at 1005; and Appalachian Power Co v EPA 135 F 3d 791 (DC Cir 1998) at 815. 270 Kelley v Selin 42 F 3d 1501 (6th Cir 1995) at 1514. 271 American Dental, above n 237 at 831. 264

(E) Fisher Ch3

31/5/07

09:57

Page 121

Reflections

121

VI Reflections The analysis above only provides a glimpse of how scope of review case-law has developed in this area, but what can be seen from that glimpse is that the courts engaged in judicial review now primarily understand risk evaluation in RI terms. Analysis such as that of Bazelon’s is a dim and distant memory. Moreover, as the RI paradigm has dominated elsewhere in US risk regulation, what can now be seen in the US is that risk evaluation is now primarily modelled on the RI paradigm. Cass Sunstein has gone so far as to describe this development as a ‘constitutional moment’ in which there has been the rise of the ‘cost/benefit state’.272 In such an administrative state, risk evaluation requires a highly quantitative approach to risk control, based on careful specification of both costs and benefits, and on close attention to relevant trade offs, including the risks sometimes introduced by regulation.273

The developments in scope of review have not been the only catalyst for this development,274 and nor is it the case that there have not been attempts to promote the DC paradigm,275 but the RI paradigm now dominates understandings about what should be the role and nature of risk evaluation, and scope of review doctrine has been a factor in that domination. Moreover, this state of affairs is not the product of crude political discourse occurring in the courts, but rather due to the legal thickness of administrative constitutionalism debates. While the purpose of this chapter is not to critique the RI paradigm’s dominance,276 it is useful to reflect on the consequences of these developments. In particular, the US experience highlights some of the problems with the RI paradigm which were identified in chapter one.277 With that said, the problems with the DC paradigm must also be kept in mind. Here I highlight three particular consequences—analytical opportunism, ossification, and disillusionment. The first of these is that ironically the RI paradigm has given rise to the serious problem of ‘Monday morning quarterbacking’ that the Supreme Court in Vermont Yankee was so keen to prevent. This problem is better described as one of analytical opportunism in that nearly any flaw in methodology can now be used as a basis for arguing that the decision was arbitrary and capricious or not based on substantial 272 Sunstein, above n 229 at ch 14; and C Sunstein, The Cost Benefit State: The Future of Regulatory Protection (Washington DC, American Bar Association, 2002). 273 C Sunstein, ‘Regulating Risks After ATA’ (2001) Supreme Court Review 1 at 1. 274 Fisher, above n 3; and Fisher, above n 226. 275 National Research Council, Understanding Risk: Informing Decisions in a Democratic Society (Washington DC, National Academy Press, 1996); and J Freeman and L Langbein, ‘Regulatory Negotiation and the Legitimacy Benefit’ (2000) 9 New York University Environmental Law Journal 60. 276 That has been done eloquently elsewhere. See T McGarity, ‘A Cost Benefit State’ (1998) 50 Administrative Law Review 7, and W Wagner, ‘The Science Charade in Toxic Risk Regulation’ (1995) 95 Columbia Law Review 1613. 277 See section III.C of ch 1.

(E) Fisher Ch3

31/5/07

122

09:57

Page 122

Hard Looks and Substantial Evidence

evidence. The opportunism arises due to the fact that such flaws will always exist in technological risk evaluation. There are two reasons for this. First, in this area, the amount of data that can be collected is almost infinite. This is because in every single case the impact of a chemical could possibly be different because of the interrelationship between countless different factors. Thus, for example, the effect of chemical exposure can vary depending on gender, smoking habits, exposure to other chemicals, and age. Likewise, if there is an expectation that an agency considers comparative risks there are no obvious limitations on how far the analysis can stretch. Second, because technological risk is so complex and uncertain then policy and judgement will always have a role to play. The problem is that under the RI paradigm data gaps are perceived as a threat to accountable and legitimate decision-making rather than as a reality of regulatory decision-making that requires a very different response to analysis.278 The impact of scientific uncertainty is only at the margins.279 The second consequence to note is that of ossification or ‘paralysis by analysis’.280 Judicial review has not been the only catalyst for this but it has clearly been influential, and the rule-making process has slowed down considerably so that analysis can take place.281 Rule-making periods have got longer as agencies have attempted to carry out more exhaustive and extensive analyses. Likewise, even after years of analysis, petitioners from both industry and public interest groups will seek judicial review.282 Thus, for example, OSHA’s methylene chloride rule published in 1997 was over 100 pages long,283 had taken over 10 years to develop, and had been based on a 48,000-page record. It was, however, immediately subject to a to judicial review challenge.284 A final consequence to note is that of disillusionment, and there has been an increasing perception that neither rule-making processes nor judicial review are contributing to good administration. As Elliott has noted in regards to the former, it is a ‘highly stylised process for displaying in a formal way the essence of something which in real life takes place in other venues’.285 The collection of more information has not necessarily resulted in better regulation,286 although it has 278

American Iron & Steel Institute, above n 269 at 992. Centre for Auto Safety v Federal Highway Administration 956 F 2d 309 (DC Cir 1992) at 316; Leather Industries of America, above n 252 at 396. 280 T McGarity, ‘Some Thoughts on “DeOssifying” the Rulemaking Process’ (1992) 41 Duke Law Journal 1385 and R. Pierce, ‘Seven Ways to Deossify Agency Rulemaking’ (1995) 47 Administrative Law Review 59. 281 Carnegie Commission on Science Technology and Government, Risk And The Environment: Improving Regulatory Decision Making (New York, 1993) 282 Natural Resources Defence Council v Administrator of EPA 902 F 2d 962 (DC Cir 1990) and Natural Resources Defense Council, above n 255. 283 62 Federal Register 1494 (10 January 1997). 284 Although it later settled. On the relevance of settlement see E Fisher and P Schmidt, ‘Seeing the Blindspots in Administrative Law: Theory, Practice, and Rulemaking Settlements in the United States’ (2001) 30 Common Law World Review 272. 285 D Elliott, ‘Reinventing Rulemaking’ (1992) 41 Duke Law Journal 1490 at 1492. 286 Carnegie Commission on Science Technology and Government, above n 281 at 108. 279

(E) Fisher Ch3

31/5/07

09:57

Page 123

Reflections

123

resulted in a more expensive administrative process.287 Indeed, Elliott notes that if an administrator is ‘genuinely’ interested in obtaining input from interested parties, informal rule-making is not the means to do it. Likewise, Mashaw comments that: the machinery of Federal rulemaking is widely viewed as so creaky and accident-prone that administrators will resort to almost any other technique to attempt to get their jobs done.288

This has seemingly resulted in attempts to de-legalise administrative process289 and a growing tendency for issues to be deliberated in informal forums such as in judicial review settlement.290 Of course, the ‘de-legalised’ nature of these forums is often a chimera, and processes such as settlement are equally dominated by legal concepts and ideas.291 Alongside this there has been a growing disillusionment with the promises of judicial review and, after three decades of judicial review of risk regulation, nearly all commentators regard the area as deeply problematic.292 Again this is due to the perception that courts are not concentrating on what is important because the legal framework bears very little relationship to what is actually going on and, moreover, such a framework is subject to abuse. Thus, for example, in Centre for Auto Safety v Peck293 Judge Scalia, then on the District of Columbia Circuit, spent nearly 30 pages analysing the methodology of a cost/benefit analysis in relation to a rule concerned with reducing the primary impact test speed for bumper bars from 5 miles per hour to 2.5 miles per hour. He affirmed the rule because the analysis was so ‘painstaking’.294 In actual fact, the rule had been passed in accordance with a White House policy, and as Judge Skelly Wright noted in dissent: [The] NHTSA proceeded to perform a cost benefit analysis that appears, given the contortions that the agency went through to reach its final conclusions, to have been solely a formalistic exercise aimed at justifying a preordained result.295

This highlights one of the most troubling aspects of the RI paradigm—both courts and agencies can expend many resources on analysis but yet the real reason why a decision is being made is not made explicit. Moreover, because of the uncertainties involved, methodology can be manipulated to support any outcome. 287 R Pierce, ‘Judicial Review of Agency Actions in a Period of Diminishing Resources’ (1997) 49 Administrative Law Review 61 at 71. 288 J Mashaw, Greed, Chaos and Governance: Using Public Choice to Improve Public Law (New Haven, Yale University Press, 1997) at 159. 289 K Werhan, ‘Delegalising Administrative Law’ (1996) University of Illinois Law Review 423. 290 Fisher and Schmidt, above n 284; and J Rossi, ‘Bargaining in the Shadow of Administrative Procedure: The Public Interest in Rulemaking Settlement’ (2001) 51 Duke Law Journal 1015. 291 P Schmidt, Lawyers and Regulation: The Politics of the Administrative Process (Cambridge, Cambridge University Press, 2005) at chs 3 and 4. 292 Levin, above n 13 at 665; and P Wald, ‘Judicial Review in Midpassage: The Uneasy Partnership Between Courts and Agencies Plays On’ (1997) 32 Tulsa Law Journal 221 at 231. 293 Centre For Auto Safety v Peck 751 F 2d 1336 (DC Cir 1985). 294 Ibid at 1369. 295 Ibid at 1371.

(E) Fisher Ch3

31/5/07

124

09:57

Page 124

Hard Looks and Substantial Evidence

In making these comments, I am not attempting to engage in a definitive critique of the RI paradigm, but what they do highlight is that conceptualising the role of risk regulators as rational instruments of the legislator does not magically result in effective and efficient risk regulation regimes.296 Of course, my whole point in this chapter is that that process of conceptualisation was so embedded in legal discourse and studded with legal concepts that it was not, in relation to scope of review at least, a conceptualisation that was the product of conscious choice. Both these issues are important to keep in mind when we come to the analysis in chapter five where we consider how US RI concepts of risk evaluation have attempted to be transplanted into the World Trade Organisation (WTO) context. That process of transplantation has rarely been accompanied with any discussion of how these concepts are embedded within US debates over administrative constitutionalism, and thus US legal culture.297

VII Conclusion At the start of this chapter, I presented the reader with Judge Leventhal’s question: What does, and should, a reviewing court do when it considers a challenge to technical administrative decision-making?298

In this chapter, I have shown that any answer to this question is concerned with administrative constitutionalism and any answer depended upon how understandings of law, public administration, and technological risk were co-produced. This was seen to have two closely interrelated aspects. First, a court’s scope of review was shaped by understandings about the legitimate role of public administration. As Edley noted in regard to the hard look review debate: Bazelon and Leventhal were talking about two sides of the same coin, that coin being judicial activism motivated by a concern for sound governance.299

Second, this process of activism was not a product of crude political debate but of law. The ultimate domination of the RI paradigm of scope of review doctrine was the product of a legal discourse over administrative constitutionalism in which courts were struggling to make sense of novel legislative regimes in light of established legal concepts. As such, the analysis in this chapter has not only highlighted how law and public administration were co-produced, but the role that a thick legal culture played in that process of co-production. 296 The converse is also true: understanding public administrators as constituted to be deliberative problem solvers does not result in effectiveness and efficiency either. 297 An exception is D Winickoff et al, ‘Adjudicating the GM Food Wars: Science, Risk and Democracy in World Trade Law’ (2005) 30 Yale Journal of International Law 81. 298 Ethyl, above n 1 at 6. 299 Edley, above 12 at 227.

(F) Fisher Ch4

31/5/07

09:58

Page 125

4 The Precautionary Principle and Merits Review in Australia As noted in chapter one, over the last decade the precautionary principle has become a touchstone for debates over administrative constitutionalism.1 This is not surprising due to the fact that the principle directly raises the issue of how information, and the scientific uncertainty embedded in that information, contributes to the establishing of the reasonableness of administrative action. As also shown, the principle is capable of being interpreted in both deliberativeconstitutive (DC) and rational-instrumental (RI) terms.2 In DC terms, the principle empowers flexible, deliberative and problem-solving public administration,3 while in RI terms the principle is a limited exception to the analytical modus operandi of RI public administration and operates in constrained circumstances when certain conditions precedent are met. As administrative constitutionalism is part of the thickness of legal culture then whether either or both of these interpretations operates within a jurisdiction will be a product of that jurisdiction’s legal culture.4 In this chapter I provide a case study of how the precautionary principle has been shaped by understandings of administrative constitutionalism existing within Australian legal culture. In particular, I show how a unique feature of that legal culture—merits review—has had a profound impact upon how the principle has been interpreted in the Australian context. My choice of Australia is no accident. Over the last decade, Australian courts and tribunals have discussed, interpreted, and applied the precautionary principle in the context of reviewing administrative decision-making, and nowhere else in the common law world can such a large body of case-law be seen. Uniquely, most of these cases have been in 1

See section V of ch 1. E Fisher and R Harding, ‘The Precautionary Principle and Administrative Constitutionalism: The Development of Frameworks for Applying the Precautionary Principle’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). 3 E Fisher and R Harding, ‘The Precautionary Principle: Towards a Deliberative, Transdisciplinary, Problem-Solving Process’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999) 4 E Fisher, ‘Precaution, Precaution Everywhere: Developing a “Common Understanding” of the Precautionary Principle in the European Community’ (2002) 9 Maastricht Journal of European and Comparative Law 7. 2

(F) Fisher Ch4

31/5/07

126

09:58

Page 126

The Precautionary Principle and Merits Review in Australia

the context of merits review, a process by which a court or tribunal ‘stands in the shoes’ of the primary decision-maker.5 Merits review is not an easy legal concept to pin down and it can be understood either as a process akin to judicial review or as a decision-making process that is a substitute for the original decision.6 Each of these different understandings leads to merits review being conducted in divergent ways, which also leads to the precautionary principle being defined and applied in different ways as well. In other words, understandings of merits review and the precautionary principle have been co-produced. In this chapter I examine this process of co-production, and the chapter is structured as follows. First, I briefly sketch Australian environmental law and show how public administration in this area has historically been DC in nature. In the second section I briefly examine how the precautionary principle was introduced into Australian legal culture, how it came to be the subject of public law litigation, and how in such a context it has been interpreted in both RI and DC terms. This state of affairs is odd when one considers the DC nature of Australian environmental law but is due to the ambiguous nature of merits review. In the third section I examine the role and nature of the courts and tribunals in carrying out merits review and show that it can be understood either as a form of external review or as a form of internal review. If understood in the former terms, merits review promotes the DC paradigm due to the fact that the starting point for fashioning such a review is mainstream administrative law doctrine, which tends to be DC in nature. In contrast, if merits review is understood as a form of internal review, it tends to promote the RI paradigm of administrative constitutionalism, due to the fact that such a review is based on an adjudicative model of decision-making rather than a DC one. The practical consequences of this for interpreting the precautionary principle are considered in Sections IV and V, where I show that DC and RI interpretations of the precautionary principle are co-produced with external and internal understandings of merits review respectively. In the final section I show how a recent decision, while representing an attempt to find a universal definition of the principle, not only highlights the inherent ambiguity and tensions in merits review but also raises some important questions about administrative constitutionalism. As always, a number of caveats should be made before starting. First, as stated in the previous two chapters, the DC and RI paradigms are being used as shorthand to describe the different normative assumptions underlying different legal and policy positions. The paradigms have never been referred to explicitly by actors but are representations of the normative presumptions that those actors hold. Second, this chapter does not contain a basic discussion of the precautionary principle, its relevance to administrative constitutionalism, and how it can be interpreted in DC and RI terms. That discussion can be found in chapter one.7 5

Re Costello and Secretary, Department of Transport (1979) 2 ALD 934 at 943. P Cane, ‘Merits Review and Judicial Review—The AAT as a Trojan Horse’ (2000) 28 Federal Law Review 213 at 225. 7 See section V of ch 1. 6

(F) Fisher Ch4

31/5/07

09:58

Page 127

Australian Environmental Law

127

Third, the focus in this chapter, due to Australian socio-political and legal culture, is upon the evaluation of environmental and public health risks as part of the regulation of urban development, natural resource management, and primary industry. While these types of risks are different from those seen in other chapters, they are still very much technological risks in that they are scientifically and behaviourally uncertain as well as socio-politically embedded. Moreover, as shall be seen, while risk assessment methodologies have not played a role, environmental impact assessment (EIA) has. Finally, this chapter is very much a sketch and does not attempt an exhaustive examination either of all Australian case-law concerning the precautionary principle or of the legal contexts it arises in. This is because my purpose is not to provide a definitive account of the principle in Australian law but rather to illustrate the utility of the lens of administrative constitutionalism.

I Australian Environmental Law and the DC Paradigm of Administrative Constitutionalism Australian environmental law is a complex web of regimes stretched over Federal, State, and local government. The multi-level nature of these regimes reflects the fact that all levels of government have responsibilities in relation to environmental issues,8 and the complex nature of these regimes is due to the pivotal role that environmental politics has played in Australia.9 Environmental law also includes planning law, regulated at the State level, which has been the main constraint on developments that impact upon the environment whether those impacts are from certain types of industrial activities or the expansion of urban areas. The main decision-makers in all these regimes have been administrative whether in the form of generalist government departments, specialist administrative bodies, or the planning officers of local councils.10 Moreover, as shall be shown below, because governments in Australia have set up specialist courts and tribunals, many of the disputes over environment and planning have ended up being litigated.11 As such, these courts and tribunals have been important sites for debating administrative constitutionalism. 8 B Galligan and C Fletcher, New Federalism: Intergovernmental Relations and Environmental Policy (Canberra, Resource Assessment Commission, 1993). 9 P Toyne, The Reluctant Nation: Environment, Law and Politics in Australia (Sydney, ABC Books, 1994); and D Mercer, A Question of Balance: Natural Resources Conflict Issues in Australia, 3rd edn (Sydney, Federation Press, 2000). 10 Note that not all local planning decisions are delegated to planning officers, but some are made by democratically elected councils. However, in both cases, guidance for principles of good decisionmaking is administrative in nature. See Land and Environment Court Working Party, Report of the Land and Environment Court Working Party (Sydney, Land and Environment Court, 2001) at 21, and P Ryan, ‘Court of Hope and False Expectations: Land and Environment Court 21 Years On’ (2002) 14 Journal of Environmental Law 301 at 311. 11 T Bonyhady, Places Worth Keeping: Conservationists, Politics and the Law (St Leonards, Allen & Unwin, 1993).

(F) Fisher Ch4

31/5/07

128

09:58

Page 128

The Precautionary Principle and Merits Review in Australia

The assessment of the environmental impacts of activities has been a major theme in Australian environmental law.12 As such, EIA and a variety of other assessment tools have had a central role in decision-making. At the same time, however, the problems created by scientific uncertainty for that process of assessment have long been obvious. Decision-makers have been required to assess consequences in the context of open-ended ecological systems where uncertainties arise due to synergies, feedback, threshold effects, and ignorance.13 It has also been obvious to those involved in such assessments that knowledge about these systems is limited, consequences tend to be irreversible, and the type of scientific uncertainties involved are more numerous than in the assessment of health risks from a particular chemical.14 Likewise, there has been an acute awareness of how environmental problems are embedded, and shaped by, deeper socio-political conflicts.15 In such circumstances, it comes as no surprise that administrative decisionmaking in this area has historically been understood in DC terms, as the problems that decision-makers have had to face are not easily manageable through RI administration. The DC paradigm can be seen in legislative frameworks which often list the factors to take into account and grant wide discretion to decisionmakers in how they exercise their power.16 Likewise, considerable legislative emphasis has been placed on the creation of deliberative committees and forums.17 The DC paradigm can also be seen significantly in judicial review doctrine which emphasises the importance of administrative decision-makers exercising flexible, wide-ranging discretion where the primary decision-maker, so long as he considers all relevant factors, has the right to give such factors what weight he thinks appropriate.18 Likewise, the courts have stressed that decision-makers 12

B Preston, Environmental Litigation (Sydney, Law Book Company Ltd, 1989) at 1. K Walker, ‘Introduction’ in K Walker (ed), Australian Environmental Policy: Ten Case Studies (Sydney, UNSW Press, 1992) at 7–8; and S Dovers and J Handmer, ‘Ignorance, Sustainability and the Precautionary Principle’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). 14 A Underwood, ‘Precautionary Principles Require Changes in Thinking About and Planning Environmental Sampling’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999); and R Carpenter, ‘Limitations in Measuring Ecological Sustainability’ in T Trzyna (ed), A Sustainable World: Defining and Measuring Sustainable Development (Sacramento, IUCN/World Conservation Union, 1995). 15 K Walker, ‘Statist Developmentalism in Australia’ in K Walker and K Crowley (eds), Australian Environmental Policy 2: Studies in Decline and Devolution (Sydney, UNSW Press, 1999); and Bonyhady, above n 11. 16 Environmental Planning and Assessment Act 1979 (NSW), s 79C; Fisheries Management Act 1994 (Cth), s 3; and Development Act 1993 (SA), s 33. See also DE Fisher, Environmental Law: Text, Cases and Materials (Sydney, Law Book Company, 1993) at 393–409 for more examples. 17 Threatened Species Conservation Act 1995 (NSW). s 129; and Gene Technology Act 2000 (Cth), Part 8. 18 Minister for Aboriginal Affairs v Peko-Wallsend Ltd (1986) 162 CLR 24 at 41; Alliance to Save Hinchinbrook v Environmental Protection Agency [2006] QSC 84 at para 24; and J Whitehouse, ‘Will the Precautionary Principle Affect Environmental Decision-Making and Impact Assessment?’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999) at 68–70. 13

(F) Fisher Ch4

31/5/07

09:58

Page 129

Australian Environmental Law

129

should engage in conscientious reasoning rather than simple fact-finding.19 Moreover, due to the constitutional and institutional limitations of the courts, they do not review questions of fact or entertain evidence on such issues except in limited circumstances, such as where it either goes to the question of jurisdiction or obviously affects the rationality of the decision-making.20 This has give rise to a doctrinally complex body of case-law where a guiding theme is that the decisionmaker should be engaged in a proper consideration of issues.21 Where DC thinking can particularly be seen is in the legal doctrine that has developed in relation to the role of environmental impact statements (EISs), which are part of EIA. Australia was one of the earliest jurisdictions to pass legislation requiring EIA as part of environmental decision-making22 but, unlike in the US context, such analytical tools have not been understood as means of establishing the factual basis for action. Rather, EIA enables decision-makers to deliberate about the most significant issues.23 As one Australian lawyer notes, EIA was one of the tools developed to ensure that ‘the deliberative obligation’ is discharged.24 What this means is that there has been less emphasis on EIA as an accurate analytical method and more on its role in aiding an understanding of a problem. Thus, for example, in Schaffer v Hawkesbury City Council, Pearlman J summed up the principles in relation to a court reviewing an EIS. She noted: 1. An environmental impact statement must be sufficiently specific to direct a reasonably intelligent and informed mind to the possible environmental consequences of the proposed development (per Cripps J in Prineas (at 417). 2. The purpose of an environmental impact statement is to alert the decision-maker and the public to the inherent problems of the proposed development, to encourage public participation, and to ensure that the decision maker takes a hard look at what is proposed (per Cripps J in Prineas (at 417) and per Cripps J in Liverpool City Council (at 278)). 3. The environmental impact statement is not required to be perfect. It need not cover every topic or explore every avenue (per Cripps J in Prineas (at 417) and per Hutley JA on appeal (at 163). 4. The environmental impact statement must not be superficial, subjective or noninformative (per Cripps J in Prineas (at 417)). 19 Minister for the Environment and Heritage v Queensland Conservation Council Inc [2004] FCAFC 190 at paras 51–62. 20 Enfield City Corporation v Development Assessment Commission (2000) 199 CLR 135; Timbarra Protection Coalition Inc v Ross Mining NL (1999) 46 NSWLR 55; and Re Minister for Immigration and Multicultural and Indigenous Affairs; Ex parte Applicant S 20/2002 (2003) 198 ALR 59. 21 M Aronson et al, Judicial Review of Administrative Action (Sydney, Law Book Co, 2004) at chs 4 and 5. 22 Environment Protection (Impact of Proposals) Act 1974 (Cth) (now repealed), and Environmental Planning and Assessment Act 1979 (NSW). 23 Australian Conservation Foundation v Commonwealth (1980) 146 CLR 493 at 525; Prineas v Forestry Commission of New South Wales (1983) 49 LGRA 402; Guthega Development Pty Ltd v Minister Administering the National Parks and Wildlife Act 1974 (1986) 61 LGRA 401; Liverpool City Council v Roads and Traffic Authority (1991) 74 LGERA 265; and Golden v Coffs Harbour City Council (1991) 72 LGRA 104. 24 Fisher, above n 16 at 409. See also B Preston, ‘Adequacy of Environmental Impact Statements in New South Wales’ (1986) 3 Environmental and Planning Law Journal 194 at 203.

(F) Fisher Ch4

31/5/07

130

09:58

Page 130

The Precautionary Principle and Merits Review in Australia

5. It should be comprehensive in its treatment of subject matter and objective in its approach (per Cripps J in Prineas (at 417). 6. Changes to the proposed development may be made between the exhibition of the environmental impact statement and the decision of the decision-maker but not so as to result in a completely different proposal (per Stein J in Golden (at 108)).25

This is clearly a DC statement of the role of information has been adopted in other cases.26 It reflects the fact that while information is understood to be important in decision-making its importance is in the context of ensuring deliberative problem-solving. As such, there has been very little litigation concerning the accuracy of EIA or any other analytical tool.27

II The Precautionary Principle in Australian Environmental Law In light of the above, it comes as no surprise that when the precautionary principle was introduced into Australian environmental law in the early 1990s it was interpreted in DC terms.28 The introduction of the principle occurred during an ‘exceptional period of environmental institutional innovation’.29 The Federal government introduced a number of comprehensive policy and institutional experiments and State governments engaged in a process of overhauling their environmental and planning regimes.30 Most of this reform either retained, or strengthened, the DC nature of public administration by entrenching the importance of public participation, principles, and the need to take a wide range of factors into account in decision-making.31 The precautionary principle was included as a guiding principle in two of the Federal government’s initiatives—the National Strategy on Ecologically 25

Schaffer Corporation Ltd v Hawkesbury City Council (1992) 77 LGRA 21 at 31. Bell v Minister for Urban Affairs and Planning (1993) 95 LGERA 86; and Rosemount Estates Pty Ltd v Minister for Urban Affairs and Planning (1996) 90 LGERA 1 at 33. 27 Although see Australian Pork Ltd v Director of Animal and Plant Quarantine [2005] FCA 671. That case was overturned in Director of Animal and Plant Quarantine v Australian Pork Limited [2005] FCAFC 206. 28 For a history of this process see R Harding and E Fisher, ‘The Precautionary Principle in Australia’ in T O’Riordan and J Cameron (eds), Interpreting the Precautionary Principle (London, Earthscan, 1994). 29 P Christoff, ‘Out of Chaos, a Shining Star?: Towards a Typology of Green States’ in J Barry and R Eckersley (eds), The State and the Global Ecological Crisis (Cambridge, MIT Press, 2005) at 32. 30 Integrated Planning Act 1997 (Qld); Environment Protection Act 1993 (WA); and Threatened Species Conservation Act 1995 (NSW). 31 Fisheries Management Act 1994 (Cth); Integrated Planning Act 1997 (Qld); and Protection of the Environment Administration Act 1991 (NSW). On the importance of principles see R Harding, ‘Sustainability: Principles to Practice—Outcomes’, Fenner Conference on the Environment 1994 (Canberra, 13–16 November 1994), and B Preston, ‘Judicial Implementation of the Principles of Ecologically Sustainable Development in Australia and Asia’, Law Society of New South Wales Regional Presidents Meeting (Sydney, 21 July 2006). 26

(F) Fisher Ch4

31/5/07

09:58

Page 131

The Precautionary Principle in Australian Environmental Law

131

Sustainable Development (ESD)32 and the non-binding Intergovernmental Agreement on the Environment (IGAE). The latter was signed by the Federal government, all State governments, and the Local Government Association of Australia. In that Agreement the precautionary principle was formulated in the following terms: Where there are threats of serious or irreversible environmental damage, lack of full scientific certainty should not be used as a reason for postponing measures to prevent environmental degradation. In the application of the precautionary principle, public and private decisions should be guided by: i) careful evaluation to avoid, wherever practicable, serious or irreversible damage to the environment; and ii) an assessment of the risk-weighted consequences of various options.33

The practical consequence of the inclusion of the principle in these two documents was that it came to be a relevant principle for administrative decisionmakers working in all fields of environment, planning and natural resource management.34 The legal relevance of the principle was strengthened by the fact that it was then included in numerous pieces of legislation, both as part of a composite definition of ESD,35 and on its own.36 Thus overnight the principle became an integral part of Australian environmental decision-making and applied to all aspects of planning, environmental, and natural resource decision-making. In so being, it was largely understood in DC terms, and many decision-makers did not understand how it was different from their existing practices.37 Moreover, because the principle, like the other principles of ESD, was understood to be a general guiding principle it emphasised that decision-making was not rulebound but discretionary and flexible.38 With all that said, some argued that the principle required an even more deliberative and flexible approach in pursuit of

32

This policy ultimately was abandoned. See Christoff, above n 29 at 33. Section 3.5.1 of the IGAE. 34 Harding and Fisher, ‘The Precautionary Principle in Australia’, above n 28 . 35 There are many examples, including s 10 of the Contaminated Land Management Act 1997 (NSW); s 6 of the Waste Minimisation Act 2001 (ACT); and s 3 of the Fisheries Management Act 1994 (Cth). 36 Again there are many examples, including s 391 of the Environmental Protection and Biodiversity Act 1999 (Cth); s 1C of the Environmental Protection Act 1970 (Vic); s 39Z of the Great Barrier Reef Marine Park Act 1975 (Cth); s 1.2.3 of the Integrated Planning Act 1997 (Qld); s 74 of the Planning and Land Act 2002 (ACT); and s 4A of the Environmental Protection Act 1986 (WA). 37 B Robinson, ‘Is “Cleaner Production” the Answer to the Precautionary Principle?’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999); and P Waterman, ‘Possible Implications of the Precautionary Principle for Australian Local Government’, in The Precautionary Principle (Institute of Environmental Studies, University of New South Wales, Sydney, 1993) at 97. 38 A Deville and R Harding, Applying the Precautionary Principle (Sydney, Federation Press, 1997); and J Peel, The Precautionary Principle in Practice: Environmental Decision-Making and Scientific Uncertainty (Sydney, Federation Press, 2005) at ch 1. See the survey in Harding and Fisher, above n 28. 33

(F) Fisher Ch4

31/5/07

132

09:58

Page 132

The Precautionary Principle and Merits Review in Australia

environmental protection than already existed.39 Others also defined the principle in RI terms as part of the promotion of the RI paradigm.40 Debate about the principle has not been primarily carried out in public policy arenas, however. Rather, debate has mainly occurred in the context of public law litigation in numerous different Federal and State courts and tribunals which vary in terms of their powers and expertise. There are now over 100 decisions of courts and tribunals where the precautionary principle has been the subject of sustained legal analysis.41 Most, but not all,42 of these decisions are in the context of merits and judicial review, with most of the cases falling into the former category. In both judicial and merits review, the principle is used by litigants to challenge the validity of the primary decision43 or by the primary decision-maker to justify their decision.44 In all these cases the principle relates to the application of the exercise of administrative discretion in the authorising of an activity or the granting of a licence or some form of permission. In particular, the principle regulates what is understood to be a suitable basis for action. The validity of the principle has been widely recognised, although the reasons vary and include that the principle is: included in the governing legislation,45 is a relevant consideration as a question of law,46 and that it is common sense to consider it.47 Despite the considerable variation in regulatory and legal context, the principle is treated as having a freestanding status.48 A key issue in all these cases has been what the principle actually means and what its implications are for a primary decision-maker and a reviewing court.49 These are questions of administrative constitutionalism and thus, while courts and tribunals have generally agreed that the principle requires a decision-maker to take ‘a cautious approach’,50 what that means in practice is dependent on the under39 V Brown, ‘Top Down, Ground Up or Inside Out? Community Practice and the Precautionary Principle’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999); and W Gullett, ‘The Precautionary Principle in Australia: Policy, Law and Potential Precautionary EIAs’ (2000) 11 Risk: Health, Safety And Environment 93. 40 D Peterson, ‘Precaution: Principles and Practice in Australian Environmental and Natural Resources Management’ (2006) 50 Australian Journal of Agricultural and Resource Economics 469. 41 For studies of this case-law see P Stein, ‘Are Decision-Makers Too Cautious with the Precautionary Principle?’ (2000) 17 Environmental and Planning Law Journal 3; E Fisher, ‘Is the Precautionary Principle Justiciable?’ (2001) 13 Journal of Environmental Law 317; and Peel, above n 38 42 For other areas see Booth v Bosworth [2001] FCA 1453 (imposition of an injunction), and Bentley v BGP Properties Pty Ltd [2006] NSWLEC 34 (sentencing in relation to environmental offences). 43 Telstra Corporation Ltd v Hornsby Shire Council [2006] NSWLEC 133; and Re National Parks and Nature Conservation Authority: Ex Parte McGregor [2001] WASCA 368. 44 Dixon v Australian Fisheries Management Authority [2000] AATA 442; and Price v Water Administration Ministerial Corporation of New South Wales [2002] NSWLEC 147. 45 Dixon, above n 44. 46 BGP Properties Pty Limited v Lake Macquarie City Council [2004] NSWLEC 399; and North Queensland Conservation Council v Great Barrier Reef Marine Park Authority [2000] AATA 925. 47 Dow Chemicals (Australia) Ltd v Chemicals Notification and Assessment [1999] AATA 1023. 48 Telstra Corporation Ltd, above n 43 at para 127. 49 Fisher, above n 41. 50 First stated in Leatch v National Parks and Wildlife Service (1993) 81 LGERA 270 at 282. Reference to the duty to be cautious can be found in the majority of cases, but for examples see: Aldekerk Pty Ltd v City of Port Adelaide Enfield and Environment Protection Authority [2000] SAERDC 47 at 23; Grishin

(F) Fisher Ch4

31/5/07

09:58

Page 133

Generalist Tribunals, Environmental Courts, and Merits Review

133

standings of administrative constitutionalism embedded in legal culture. In light of the above, it is not surprising that much of the case-law concerning the precautionary principle has interpreted the principle in DC terms as an enabling principle which requires the flexible exercise of public administration.51 What is more surprising, however, is that there is also a significant body of case-law which has interpreted the principle in RI terms as requiring a shifting of the evidential burden of proof if certain conditions precedent are met.52 While such an interpretation does accord with how some lawyers and legal scholars have conceptualised the principle53 it seems oddly discordant with both understandings of EIA and administrative decision-making in Australia. The RI understanding of the principle has not been a product of explicit understandings of public administration, however. Rather, it has been a product of how courts and tribunals have understood their own role when carrying out merits review. To understand this, some understanding is needed of these courts and tribunals and of merits review.

III Generalist Tribunals, Environmental Courts, and Merits Review As already noted, the cases concerning the precautionary principle have mainly taken place in the context of merits review carried out by generalist administrative tribunals and specialist environmental courts and it is this context which has been fundamental in shaping understandings of the precautionary principle. Australia has led the world in the creation of such reviewing bodies,54 which are forms of ‘state sponsored legal pluralism’ created by the Australian Federal and State governments due to the perception that traditional courts were inadequate in v Conservator of Flora and Fauna [1998] ACTAAT 250 at 14; Histpark Pty Ltd v Maroochy Shire Council [2001] QPEC 59 at 19; North Queensland Conservation Council, above n 46 at 168; Friends of Hinchinbrook Society Inc v Minister for the Environment (1997) 142 ALR 632 at 677; and Bridgetown/Greenbushes Friends of the Forest Inc v Executive Director of Conservation and Land Management (1997) 18 WAR 102 at 118. 51 See section IV. 52 See section V. For a discussion of the relationship between the RI paradigm and understanding the principle as requiring a shifting of the burden of proof see section V.C of ch 1. 53 See the discussion in Whitehouse, above n 18 at 65–8, and J Jones and S Bronitt, ‘The Burden and Standard of Proof in Environmental Regulation: The Precautionary Principle in an Australian Administrative Context’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). 54 P Cane, ‘The Making of Australian Administrative Law’ in P Cane (ed), Centenary Essays for the High Court of Australia (Sydney, LexisNexis Butterworths, 2004) at 331; and M Grant, Environmental Court Project. Final Report. (London Department of Transport, Environment and the Regions, 2000) at 4, 168, and 255. Note that merits review was not new to Australian law but these new schemes were on a far larger scale: J Fitzgerald, ‘Aspects of the System of Fact-Finding and Rules of Evidence’ in K Cole (ed), Administrative Law and Public Administration: Form vs Substance (Canberra, Australian Institute of Administrative Law, 1996); and Land and Environment Court Working Party, above n 10 at 43

(F) Fisher Ch4

31/5/07

134

09:58

Page 134

The Precautionary Principle and Merits Review in Australia

resolving public law disputes or ensuring accountability.55 As such, these courts and tribunals have an important role in limiting and holding public administration to account and thus as forums for debating administrative constitutionalism. The first major initiative in this exercise of legal pluralism in Australia was the creation of the generalist Federal Administrative Appeals Tribunal (AAT) in 197656 as part of a comprehensive administrative law package that also included a legislative codification of the grounds of review, and the creation of an ombudsman.57 Subsequently, a number of generalist State administrative tribunals were also set up to carry out merits review.58 Accompanying these developments in general administrative law were similar reforms in the environmental area. From the late 1970s onwards a number of specialist environmental and planning courts were created at the State level to hear administrative appeals. The first and most high profile of these was the Land and Environment Court (LEC) which was created in 1979 in New South Wales (NSW).59 As with the AAT, this court was created as part of a comprehensive package of environmental reforms and it was described by the Minister introducing the legislation into Parliament as: [a] somewhat innovative experiment in dispute resolution mechanisms. It attempts to combine judicial and administrative dispute resolving techniques.60

The NSW LEC was not only resolving disputes, however; it was also understood to be an enforcer of principles of good administration in the environmental context.61 In the 1990s most other States also set up such courts:62 the Environment Resources and Development Court (ERDC) in South Australia (SA) in 1994;63 the Planning and Environment Court (PEC) in Queensland (Qld) in 1990;64 the

55 H Arthurs, ‘Without the Law’: Administrative Justice and Legal Pluralism in Nineteenth Century England (Toronto, University of Toronto Press, 1985) at 192. See also Land and Environment Court Working Party, above n 10 at 44; and G Brennan, ‘Twentieth Anniversary of the AAT: Opening Address’ in J McMillan (ed), The AAT Twenty Years Forward (Canberra, Australian Institute of Administrative Law, 1998) at 11. 56 Administrative Appeals Tribunal Act 1975 (Cth). 57 Administrative Decisions (Judicial Review) Act 1977 (Cth); and Ombudsman Act 1976 (Cth). For background see Kerr Committee, Commonwealth Administrative Review: Committee Report (Canberra, Parliamentary Paper No 144, Commonwealth Government Publishing Office, 1971). 58 Administrative Decisions Tribunal Act 1997 (NSW); Administrative Appeals Tribunal Act 1989 (ACT); Victorian Civil and Administrative Tribunal Act 1998 (Vic); and State Administrative Tribunal Act 2004 (WA). 59 Land and Environment Court Act 1979 (NSW). Ryan above n 10. 60 Minister Paul Landa, as quoted in S Smith, A Review of the Land and Environment Court, Briefing Paper 13/2001 (Sydney, Parliament of New South Wales, 2001). 61 Land and Environment Court Working Party, above n 10 at 44. 62 For a general overview see Grant, above n 54 at chs 5–8. A strong impetus for this development in the 1990s was B Hayes and C Trenorden, Combined Jurisdiction for Development Appeals in States and Territories (Canberra, AGPS, 1990). 63 Environment Resources and Development Court Act 1993 (SA) 64 Local Government (Planning and Environment) Act 1990 (Qld), although note that its role and nature were substantially transformed by the Integrated Planning Act 1997 (Qld).

(F) Fisher Ch4

31/5/07

09:58

Page 135

Generalist Tribunals, Environmental Courts, and Merits Review

135

Resource Management and Planning Appeal Tribunal in Tasmania in 1993;65 and the Town Planning Appeal Tribunal in Western Australia (WA) in 2003.66 These courts and tribunals follow no common template, occupy different places in the judicial hierarchy67 and have different jurisdictions and powers.68 Most courts and tribunals are staffed with judges and non-legally trained experts but in different arrangements.69 Most of the legislative frameworks encourage mediation, although this ranges from promoting traditional forms of adversarial settlement70 to creating more innovative forms of community dispute resolution.71 In most cases these courts and tribunals do not have to comply with the rules of evidence and have powers to regulate their own procedures ‘as they see fit’.72 As part of this, such courts are directed to conduct their proceedings with as ‘little formality as possible’, with the expectation that they will dispose of the expensive and tangential trappings of the adversarial process.73 Yet, these requirements are not uniform and some statutes do regulate some aspects of court and tribunal procedure.74 There is also variation between courts in how adversarial proceedings are operating in practice.75 The nature of these courts and tribunals is further complicated by the enigma that is merits review. Merits review means that these courts or tribunals were entrusted with the power to remake the decision under review as if they were the primary decision-maker. From the start this is a curious concept because the reality is that a court or a tribunal is an entirely different institutional form from public administration. It is if, to use a very crude and extreme analogy, a lion was 65

Resource Management and Planning Appeal Tribunal Act 1993 (Tas). Town Planning and Development Act 1928 (WA), s 36. 67 The NSW LEC is a superior court of record. See s 5(i) of the Land and Environment Court Act 1979 (NSW). The Qld PEC and SA ERDC are equivalent to the district court in the judicial hierarchy. See s 4.1.1(2) of the Integrated Planning Act 1997 (Qld), and s 5 of the Environment Resources and Development Court Act 1993 (SA). 68 The NSW LEC has a number of different classes of jurisdiction, not all being merits reviews: see ss 16–21B of the Land and Environment Court Act 1979 (NSW). The Qld PEC has a relatively narrow merits review jurisdiction: see s 4.1.2 of the Integrated Planning Act 1997 (Qld). The SA ERDC has a more ad hoc jurisdiction: see s 7 of the Environment Resources and Development Court Act 1993 (SA). 69 Land and Environment Court 1979 (NSW), ss 6 and 12; Resource Management and Planning Appeal Tribunal Act 1993 (Tas), s 10(1); and Town Planning and Development Act 1928 (WA), s 38. 70 Integrated Planning Act 1997 (Qld), s 4.1.48; and Environment Resources and Development Court Act 1993 (SA), s 28B. 71 Land and Environment Court Act 1979 (NSW), Part 5A. 72 Environment Resources and Development Court Act 1993 (SA), s 21; Land and Environment Court Act 1979 (NSW), s 38; Administrative Appeals Tribunal Act 1989 (ACT), s 32; Town Planning and Development Act 1928 (WA), s 51; Resource Management and Planning Appeal Tribunal Act 1993 (Tas), ss 16 and 19; and Administrative Appeals Tribunal Act 1975 (Cth), s 33. 73 M Allars, ‘Neutrality, the Judicial Paradigm and Tribunal Procedure’ (1991) 13 Sydney Law Review 377 at 380. 74 The prime example being the Qld PEC, which is required to comply with the rules of court as well as a series of other evidentiary rules: see Integrated Planning Act 1997 (Qld), ss 4.1.7, 4.1.9, and 4.1.50. A number of these statutes also require courts to comply with the principles of natural justice: see Town Planning and Development Act 1928 (WA), s 51(1)(a) and (2); and Resource Management and Planning Appeal Tribunal Act 1993 (Tas), s 16(1)(d). 75 Grant, above n 54 at 288–9. 66

(F) Fisher Ch4

31/5/07

136

09:58

Page 136

The Precautionary Principle and Merits Review in Australia

standing in the shoes of an elephant. Both are entirely different beasts, which have their own strengths and weaknesses, their own innate competences, their own practices, and their own traditions and cultures. The problem is further complicated by the fact that there is no fixed understanding of the concept of merits review. Thus, for example, the legislative frameworks for these courts and tribunals describe merits review in different ways. The AAT ‘may exercise all the powers and discretions that are conferred by any relevant enactment on the person who made the decision’.76 The NSW LEC ‘for the purposes of hearing and disposing of an appeal [has] all the functions and discretions which the person or body whose decision is the subject of the appeal had in respect of the matter the subject of the appeal’.77 The court shall also ‘have regard to this or any other relevant Act, any instrument made under any such Act, the circumstances of the case and the public interest’.78 The SA ERDC ‘must act according to equity, good conscience and the substantial merits of the case and without regard to legal technicalities and forms’.79 An appeal to the Qld PEC ‘is by way of hearing anew’,80 and the Tasmanian Resource Management and Planning Appeal Tribunal should ‘hear matters afresh’.81 Likewise, each institution has developed its own body of caselaw concerning what is, and is not, appropriate review for the purposes of merits review.82 All of the above reflects the fact that merits review, and the institutions that carry it out, occupy an awkward space in Australian administrative law.83 Not only have they been simultaneously and specifically developed as forums for both dispute resolution and holding decision-makers to account, they also sit uneasily between the administration and the judiciary.84 This state of affairs has resulted in merits review being understood in two distinct ways—as a form of external review and as a form of internal review.85 As the former, merits review is a variation on judicial review in that while a court or tribunal can remake a decision they are still acting on the basis that they have a different institutional and constitutional competence from the primary decision-maker.86 As the latter, merits review is a form of dispute resolution in which a court or tribunal is remaking a decision on the basis that they have the same institutional and constitutional competence as the

76

Administrative Appeals Tribunal Act 1975 (Cth), s 43(1). Land and Environment Court Act 1979 (NSW), s 38(2). Also see s 39(2). 78 Ibid, s 39(4). 79 Environment Resources and Development Court Act 1993 (SA), s 21(1)(c). 80 Integrated Planning Act 1997 (Qld), s 4.1.52(1). 81 Resource Management and Planning Tribunal Act 1993 (Tas), s 16(1)(e). 82 See, for example, the discussion of AAT merits review case-law in R Creyke and J McMillan, Control of Government Action: Text, Cases and Commentary (Sydney, LexisNexis Butterworths, 2005) at ch 3. 83 G Fleming, ‘Administrative Review and the “Normative” Goal—Is Anybody Out There?’ (2000) 28 Federal Law Review 61. 84 This is particularly problematic at the Federal level. See Cane, above n 6 at 214–15. 85 Ibid at 225. 86 See ibid for an example of this thinking. 77

(F) Fisher Ch4

31/5/07

09:58

Page 137

Generalist Tribunals, Environmental Courts, and Merits Review

137

decision-maker, or at least they are an effective substitute for that competence.87 The latter has tended to be understood as review carried out within a department,88 but it has also dominated where merits review by a court or tribunal is understood as a superior form of dispute resolution.89 Moreover, while these two understandings may look complementary—merits review could potentially be simultaneously a form of decision-making and a forum for ensuring the validity of good decision-making—in actual fact they lead to legally constructing merits review in two different ways. If understood as a form of external review, merits review is understood as being akin to judicial review in that it is a forum in which the principles of good public administration are articulated, debated, and laid down.90 This is not to say that the two are not recognised as distinct practices91 but rather that merits review is understood as similar to judicial review in that both forms of review serve the same purpose.92 A court or tribunal is thus judging public administration against a pre-ordained standard of reasonableness, and merits review is an activist variation on a check on the power of public administration.93 As such, merits review raises the same type of institutional and constitutional competence issues that were discussed in chapter three.94 In particular, in engaging in merits review as a form of external review a court or tribunal does not see the scrutiny of evidence as a priority. Moreover, as in judicial review, the evidence before the court tends to be accepted without lengthy cross-examination. As merits review is about promotion of good administration through the articulation of norms, then decisions, while not necessarily strictly legally binding, are ‘precedents’ that provide guidance to decisionmakers about what is reasonable administrative action and what they should and should not take into account.95 Likewise, merits review is understood less as a bipolar dispute and more about achieving the ‘best community outcome’.96 87 Administrative Review Council, Better Decisions: Review of the Commonwealth Merits Review Tribunals (Canberra, ARC 39, 1993). 88 Administrative Review Council, Environmental Decisions and the Administrative Appeals Tribunal (Canberra, Report No 36, AGPS, 1994). 89 Administrative Review Council, Better Decisions: Review of the Commonwealth Merits Review Tribunals above n 87. 90 Land and Environment Court Working Party, above n 10 at 37–8; P Stein, ‘Specialist Environmental Courts: the Land and Environment Court of New South Wales, Australia’ (2002) 4 Environmental Law Review 5 at 5; and N Bignold, ‘NSW Land and Environment Court—Its Contribution to Australia’s Development of Environmental Law’ (2001) 18 Environmental Planning and Law Journal 256. 91 ULV Pty Ltd v Scott (1980) 69 LGRA 212, and Fast Buck$ v Byron Shire Council (1999) 103 LGERA 94. 92 Cane, above n 6. 93 G Taylor, ‘The New Administrative Law’ (1977) 51 Australian Law Journal 804 at 805; Brennan, above n 55 at 11; P Stein, ‘A Cautious Application of the Precautionary Principle’ (2000) 2 Environmental Law Review 1; and Stein, above n 41 at 12. 94 See section I of ch 3, although note that the constitutional competence issues are complicated in the AAT because of the concept of judicial power in the Australian constitution. See Cane, above n 6 at 214–19. On the close relationship between merits and judicial review see N Pain, ‘Environmental Decision-Making Processes’ (1996) 79 Canberra Bulletin of Public Administration 73 at 76. 95 Ryan, above n 10 at 311. 96 Maxnox Pty Ltd v Hurstville City Council [2006] NSWLEC 146 at para 57.

(F) Fisher Ch4

31/5/07

138

09:58

Page 138

The Precautionary Principle and Merits Review in Australia

Understanding merits review as a form of external review is made plausible by a number of factors including that the same judges are often carrying out both sorts of review;97 merits review decisions are treated as precedents by public administration;98 questions of law are raised in merits review and are treated as such;99 and environmental law scholarship often fails to make any real distinction between the two forms of review.100 In contrast, if merits review is understood as a form of internal review then the task of a court or tribunal is to remake a decision. On this basis, merits review is a form of dispute resolution, and such a remaking will be legitimate only if the decision is remade in a more accountable, fairer, and superior form.101 For a court or tribunal these goals are inevitably achieved by modelling decision-making on adjudicative method. This is because the adjudicative method is understood as the best means to ensure that, in resolving a dispute, a court or tribunal is acting in an acceptable manner, particularly by protecting the neutrality of the judge as umpire.102 As Brennan J noted in an early AAT case: To depart from the rules of evidence is to put aside a system which is calculated to produce a body of proof which has rational probative force.103

Adjudicative method, from this perspective, is the ‘natural means of fact finding’,104 and the superiority of an administrative tribunal or court is in the superiority of their fact-finding procedures.105 As such, courts and tribunals engaged in merits review have borrowed procedures from the adjudicative/adversarial process, particularly basic principles concerning the treatment and assessment of evidence.106 Thus, from early on, the AAT relied on common law evidentiary principles in cases involving serious implications for individual rights to conclude that the basis for decisions ‘should be established, on the balance of probability, to its satisfaction by some rationally probative evidence’.107 The AAT also makes use of cross-examination, written statements, and other basic rules of adjudicative procedure.108 Similar practices can be seen in other courts, and thus, 97 In the NSW LEC, judges can conduct both merits review and judicial review proceedings. See Land and Environment Court Act 1979 (NSW), ss 17 and 20. 98 Ryan, above n 10 at 311. 99 Schaffer Corporation Ltd, above n 25, and Bell, above n 26. 100 Gullett, above n 39; and Stein, above n 41. 101 Administrative Review Council, above n 87. 102 Allars, above n 73 at 382. 103 Re Pochi and Minister for Immigration and Ethnic Affairs (1979) 2 ALD 33 at 41. 104 Brennan, above n 55 at 17. 105 M Jeffrey, ‘Appropriateness of Judicial and Non Judicial Determination of Environmental Issues’ (1988) 5 Environmental Planning and Law Journal 265; Land and Environment Court Working Party, above n 10 at 45; G Brennan, ‘The AAT: Early Issues’ (2001) 9 Australian Journal of Administrative Law 5 at 12; and Grant, above n 54 at 215. 106 Creyke and McMillan, above n 82 at 155–79 for an excellent discussion on the development of these principles in the AAT. See also Lainson v Sutherland Shire Council [1998] NSWLEC 87, and Bennett Taylor Pty Ltd v North Sydney Municipal Council [1988] NSWLEC 77. 107 Minister for Immigration and Ethnic Affairs v Pochi (1980) 4 ALD 139 at 156 per Deane J. 108 Fitzgerald, above n 54; J McMillan, ‘Merit Review and the AAT: A Concept Develops’ in J McMillan (ed), The AAT—Twenty Years Forward (Canberra, Australian Institute of Administrative

(F) Fisher Ch4

31/5/07

09:58

Page 139

Generalist Tribunals, Environmental Courts, and Merits Review

139

for example, a judge of the NSW LEC describes that court’s procedure as striking a ‘happy medium’ between old fashioned formality and extreme informality’109— both concepts being understood as variations on traditional adversarial procedure. With that said, the process of borrowing is by no means exhaustive, uniform, or non-reflexive. Neither all features of adjudicative procedure nor all rules of evidence are utilised by these courts and tribunals,110 and others have been adapted for specialist use.111 These two different understanding of merits review do reflect the fact that public law litigation has always been simultaneously a vehicle for dispute resolution and for checking the power of public administration. Yet in the merits review context they are two fundamentally different types of activities—either a more activist form of judicial review or dispute resolution in an adjudicative forum. Moreover, these two different types of activities have implications for which paradigm of administrative constitutionalism is promoted. If merits review is understood as external review then theoretically it can advance either the DC or RI paradigm of administrative constitutionalism. This is because, as seen in chapter three, standards of judicial review are capable of being based on either paradigm. As already seen, however, Australian judicial review doctrine in the environmental area promotes a DC understanding of public administration. As such, merits review when understood as a form of external review does the same, drawing as it does on established judicial review principles. In contrast, when merits review is understood as a form of internal review it promotes the RI paradigm of administrative constitutionalism. This is because, in being modelled on adjudicative method, such a form of merits review characterises the court or tribunal as having very little discretion itself but rather it is limited to assessing the information and arguments provided by the parties. Although arguably this RI understanding applies only to the review decision-making process and not to the original decisionmaking process as well, it does result in the guidance that is given to the original decision-maker, particularly on issues in relation to evidence, tending to be RI in nature. This will be seen below. In both cases, it is understandings about the role of courts and tribunals which are shaping understandings about administrative constitutionalism. Returning to the precautionary principle, what can be seen is that it is the divergence of understanding about the nature of merits review which has led to courts Law Inc, 1998) at 48, and the cases discussed there; and A O’Neill, ‘The Rules of Evidence and Administrative Law’ in K Cole (ed), Administrative Law and Public Administration: Form vs Substance (Canberra, Australian Institute of Administrative Law, 1996). 109 Stein, above n 90 at 8. 110 O’Neill, above n 108; and Fitzgerald, above n 54, concepts of the onus of proof being a case in point. See the different approaches in McDonald v Director General of Social Security (1984) 6 ALD 6 at 11; North Queensland Conservation Council, above n 46 at para 204; Hale v Parramatta City Council (1982) 47 LGRA 269; and Zhang v Canterbury City Council [1999] NSWLEC 209 at para 37. 111 Land and Environment Court Working Party, above n 10 at 52–61; and G Downes, ‘Expert Evidence: The Value of Single or Court-appointed Experts’, Australian Institute of Judicial Administration: Expert Evidence Seminar (Melbourne, 2005).

(F) Fisher Ch4

31/5/07

140

09:58

Page 140

The Precautionary Principle and Merits Review in Australia

and tribunals interpreting the principle in both DC and RI terms. In other words, what can be seen is the co-producing of understandings of merits review and understandings of the precautionary principle. The relationship between the external review understanding of merits review and a DC interpretation of the principle is considered in the next section, and the relationship between the internal understanding of merits review and the RI interpretation of the principle in the section after that.

IV DC Interpretations of the Precautionary Principle As already stated, Australian environmental administration has evolved on the basis of the DC paradigm of administrative constitutionalism. Thus it comes as no surprise that the earliest statements about the precautionary principle were in DC terms,112 and Leatch v National Parks and Wildlife Service (the Leatch decision),113 the first significant case on the principle, was a classic example of DC reasoning. Leatch concerned a challenge by a third party objector to a decision to grant a local council a licence to ‘take or kill’ fauna under relevant biodiversity protection legislation.114 The licence was needed by the council because a road it was building was likely to have an adverse impact upon a number of species. The third party challenger raised a number of arguments, including that the fauna impact statement (FIS) that had been produced as part of the decision-making process was invalid and that the decision-maker had failed to apply the precautionary principle. Stein J’s judgment was clearly grounded in an understanding of merits review as a form of external review. His starting point for judging the legitimacy of the decision was the legal framework and principles of judicial review, and he was not particularly concerned with the adjudicative handling of evidence before him.115 In assessing the validity of the FIS he drew on the large body of case-law concerning the role of environmental impact statements (EISs).116 He concluded that an FIS was not an end in itself but rather was designed to ‘assist the decision-maker in its task and to inform the public and enable its participation’117 and this was entirely consistent with that case-law.118 Moreover, such a characterisation of the role of an FIS characterised the whole decision-making process in deliberative terms. Within such a context, it comes as no surprise that the precautionary principle was also understood in DC terms. Yet that was not the only reason for such a 112

For a very early brief DC mention, see Simpson v Ballina Shire Council [1994] NSWLEC 43 . Leatch, above n 50. 114 Section 92B of the National Parks and Wildlife Act 1974 (NSW). 115 Leatch, above n 50 at 277–8 and 282. 116 Schaffer Corporation Ltd, above n 25 at 31. See also Prineas, above n 23, and Guthega Development Pty Ltd, above n 23. 117 Leatch, above n 50 at 278. 118 See n 23 above. 113

(F) Fisher Ch4

31/5/07

09:58

Page 141

DC Interpretations of the Precautionary Principle

141

characterisation. Stein J argued that the relevance of the principle was a question of law and, relying on judicial review authority,119 concluded that the principle was legally relevant even though it was not included in any relevant legal scheme but only in policy, international law, and legislation in other areas.120 As such, he embedded the principle in more general DC judicial review doctrine. It is in this context that he described the principle as a duty to be cautious: In my opinion the precautionary principle is a statement of common sense and has already been applied by decision-makers in appropriate circumstances prior to the principle being spelt out. It is directed towards the prevention of serious or irreversible harm to the environment in situations of scientific uncertainty. Its premise is that where uncertainty or ignorance exists concerning the nature or scope of environmental harm (whether this follows from policies, decisions or activities), decision makers should be cautious.121

For him, the principle was a broad, enabling DC principle, and throughout the rest of the decision, Stein J treated it as applying to all aspects of decision-making. That decision-making process required a ‘balancing of considerations’122 and a thorough examination of alternatives.123 It required a decision-maker to be wide ranging in his/her analysis and flexible in reaching a decision. The evidence of environmental conditions and the potential impact of the road were both important, but there was no expectation that either of these matters could be fully proved. Thus in the case before him, Stein J was not only concerned with exploring many of the different types of uncertainty involved in the decision but also with how such uncertainties impacted upon the values that the statutory scheme was trying to uphold.124 Likewise, the issue of scientific uncertainty was not an isolated evidentiary issue but rather forced decision-makers to reflect upon how to balance the need for a road with biodiversity conservation.125 Moreover, Stein J understood this as an ongoing issue that would need to be revisited.126 Stein J’s DC interpretation logically flows from his understanding of merits review as a form of external review, akin to judicial review. His starting points for understanding the principle were general judicial review principles and the NSW LEC’s case-law on EIA,127 both of which promoted the DC paradigm. It is thus not surprising that his description of the principle as requiring a ‘cautious approach’ has been relied on as precedent in both judicial review and merits review actions.

119 120 121 122 123 124 125 126 127

Minister for Aboriginal Affairs, above n 18. Leatch, above n 50 at 281–2. Ibid at 282. Ibid at 285. Ibid at 286. Ibid at 284. Ibid at 286. Ibid at 287. Above n 23.

(F) Fisher Ch4

31/5/07

142

09:58

Page 142

The Precautionary Principle and Merits Review in Australia

There have been fewer cases in the first category, which includes both judicial review of merits review decisions128 and judicial review of primary decisions129 In the former subcategory, there has been little discussion of the principle, although in most cases the way in which the principle was interpreted as part of merits review has been upheld.130 More discussion has occurred in the latter subcategory, where discussion of the principle has been clearly in DC terms. Thus, for example, the DC approach can clearly be seen in the approach taken by Wheeler J in the Western Australian Supreme Court case of Bridgetown/Greenbushes Friends of the Forest Inc v Executive Director of Conservation and Land Management.131 This was a judicial review challenge to forestry operations in which the precautionary principle was raised in argument. In considering the principle and Stein J’s judgment in Leatch, Wheeler J noted: a requirement that a decision maker ‘be cautious’ says something about the way in which the decision must be made. There must be some research, or reference to available research, some consideration of risks, and a more pessimistic rather than optimistic view of the risks should be taken. However, such a requirement does not in any particular case specify precisely how much research must be carried out, or when a risk should be considered to be so negligible that it may safely be disregarded. Still less, does such an approach dictate what courses of action must be taken after the possibilities have been cautiously weighed.132

As already noted, this DC understanding of the principle fits comfortably in with Australian judicial review doctrine as well as understandings of judicial competence in judicial review cases.133 The role of a reviewing court is not to dictate outcomes but rather to police the process of decision-making to ensure its validity.134 Where, however, far more actions have been brought is in the context of merits review, and here the DC interpretation of the principle has been promoted in those cases where merits review is understood as a form of external review. Thus in Nicholls v Director General of National Parks and Wildlife Service (the Nicholls decision)135 Talbot J was engaged in the review on the merits of a fauna licence decision, and characterised the nature of decision-making and FISs in the same DC terms as Stein J did in Leatch.136 Talbot J, while finding no duty on the decision-maker to apply the precautionary principle, found that a careful evaluation 128 Rashleigh v Environment Protection Authority [2005] ACTSC 18 at 94; Simpson, above n 112; and Carstens v Pittwater Council (1999) 111 LGERA 1. 129 Re National Parks and Nature Conservation Authority: Ex Parte McGregor, above n 43; and Friends of Hinchinbrook Society Inc, above n 50. 130 Tuna Boat Owners Association of South Australia Inc v Development Assessment Commission [2000] SASC 238, but see Rashleigh, above n 128 at 94. 131 Bridgetown/Greenbushes Friends of the Forest Inc, above n 50. 132 Ibid at 118. See also Mees v Kemp [2004] FCA 366 at 34; and Re National Parks and Nature Conservation Authority: Ex Parte McGregor, above n 43 at 157. 133 Fisher, above n 41. 134 J Jowell, ‘Of Vires or Vacuums: The Constitutional Context of Judicial Review’ in C Forsyth (ed), Judicial Review and the Constitution (Oxford, Hart Publishing, 2000) at 332–3. 135 Nicholls v Director General of National Parks and Wildlife Service (1994) 84 LGERA 397. 136 Ibid at 417–18.

(F) Fisher Ch4

31/5/07

09:58

Page 143

DC Interpretations of the Precautionary Principle

143

of considerations in decision-making was ‘axiomatic’ in the context of environmental assessment.137 As such, while there was no strict duty to apply the principle he concluded that licences should be subject to ongoing surveys because of scientific uncertainties over impacts.138 For Talbot J, public administration was thus understood in DC terms. His characterisation of the precautionary principle was more equivocal, however. On the one hand, he understood it in DC terms, hence his conclusions above. On the other hand, Talbot J described the principle in narrower RI terms as a principle applying to evidence within litigation. From this perspective he was more sceptical about the principle and suggested it ‘could have the potential to create interminable forensic argument’139 due to scientific uncertainty making the operation of such rules of evidence difficult. As a DC principle applying to administrative process, he saw it making sense. As an RI principle applying to evidence, he did not. Talbot J’s ambivalence towards the principle reflected both the ambiguity over the definition of the principle as well as the ambiguity over the nature of merits review. Whatever the case, he clearly opted for a DC approach. The broad DC approach to applying the principle can also be seen in Greenpeace Australia Ltd v Redbank Power Company Pty Ltd (the Greenpeace decision),140 which concerned a merits review of an EIS in relation to a power plant that used mining tailings as fuel. The public interest group Greenpeace argued that the precautionary principle should operate in relation to taking into account the carbon dioxide emissions from the plant. Pearlman CJ did not refer to any general judicial review principles but cited the IGAE definition of the principle and noted that there were many different uncertainties in relation to the development, including the disposal of the tailings which would occur if the power plant did not operate. She then stated: The important point about the application of the precautionary principle in this case is that ‘decision makers should be cautious’ (per Stein J in Leatch v National Parks & Wildlife Service & Anor (1993) 81 LGERA 270 at 282). The application of the precautionary principle dictates that a cautious approach should be adopted in evaluating the various relevant factors in determining whether or not to grant consent . . .141

She thus understood the principle as a broad DC principle promoting flexible and wide-ranging deliberation.142 In particular, it was a principle that was understood to apply to the process of decision-making. Moreover, in discussing the expert evidence she was less concerned with the experts establishing facts and more with how their opinions helped in a general understanding of the problem.143 Similar approaches can be seen in other merits review cases where the principle was understood to be regulating the whole process of decision-making. From this 137 138 139 140 141 142 143

Ibid at 419. Ibid at 421. Ibid at 419. Greenpeace Australia Ltd v Redbank Power Company Pty Ltd (1994) 86 LGERA 143. Ibid at 154. Ibid at 154–6. Ibid at 150–3.

(F) Fisher Ch4

31/5/07

144

09:58

Page 144

The Precautionary Principle and Merits Review in Australia

perspective the principle has been described as a ‘central element in decisionmaking’ and as a method of ensuring the proper pursuit of environmental protection goals.144 The principle is operating both as an enabling principle and as a background principle and requires wide-ranging considerations.145 As such, the principle not only applies to the assessment of evidence or to a final decision but has also been found to apply to management plans,146 to whether a Species Impact Statement (SIS) is needed,147 as well as to the general reasoning process of a decision-maker.148 A result of this is that the principle is one of a number of factors that a decision-maker must take into account but does not necessitate a particular outcome.149 Moreover, its relevance is in assessing the overall validity of the decision, legally and otherwise, and it does not just concern how evidence is assessed. In other words, a decision-maker in applying the principle is understood as a ‘discretion exerciser not a fact finder’.150 As such, in cases where the principle is being understood in DC terms, the court or tribunal is concerned to assess the evidence but in so doing it is not particularly concerned with matters of procedure and defines ‘evidence’ loosely to include evidence of facts and evidence of policies and practices.151 This is because merits review is primarily being understood as a form of external not internal review, and in most cases a court or tribunal is concerned with reasoning rather than proof.152 This is entirely consistent with how evidence is understood in the judicial review context, although it does amount to more intensive review than judicial review. Thus, for example, in Acquaro v Great Lakes Council the applicant’s application contained inaccurate drawings and an inadequate explanation of the development, which negated the quality of its evidence and reasoning.153 The issue in external review is the ‘proper consideration and due consideration’ of issues rather than the discharging of any evidentiary onuses.154 Thus, in some cases, courts have accepted evidence without there being any cross-examination of it,155 and the pur144 Murrumbidgee Groundwater Preservation Association v Minister for Natural Resources [2004] NSWLEC 122 at paras 178 and 186. 145 Jones v Pristine Waters Council (NSWLEC, unreported, 24 May 2002); and Murrumbidgee Groundwater Preservation Association, above n 144 at 186. 146 North Queensland Conservation Council, above n 46. 147 BT Goldsmith Planning Services v Blacktown City Council [2005] NSWLEC 210 at para 72; and Gales Holdings Pty Limited v Tweed Shire Council [2006] NSWLEC 85 at para 68. 148 Humane Society International v Minister for the Environment and Heritage [2006] AATA 298 at para 51; and Ajka Pty Limited v Australian Fisheries Management Authority [2001] AATA 258 at para 86. 149 North Queensland Conservation Council, above n 46 at 171; and Greentree v Colac Otway Shire Council [2005] VCAT 815 150 D Farrier, ‘Factoring Biodiversity Conservation into Decision-Making Processes: The Role of the Precautionary Principle’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999) at 109. 151 Gales Holdings Pty Limited, above n 147 at para 68. 152 Greenpeace Australia Ltd, above n 140. 153 Acquaro v Great Lakes Council [2005] NSWLEC 582 at para 74. 154 Kroger v Southern Rural Water [2001] VCAT 1334 at para 55. 155 Murrumbidgee Groundwater Preservation Association, above n 144 at para 100.

(F) Fisher Ch4

31/5/07

09:58

Page 145

DC Interpretations of the Precautionary Principle

145

pose of evidence is understood as an aid to the court in getting a better understanding of the issue before them.156 What this has meant is that when courts and tribunals understand their role as a form of external review they have resisted attempts to define the precautionary principle in RI terms as a shifting of the burden of proof. Thus, for example in Dixon v Australian Fisheries Management Authority (the Dixon decision)157 the AAT adjusted the interpretation of the principle given by an expert witness who understood it as a shifting of the burden of proof that only operated when a threshold was met.158 Rather, the tribunal concluded that the principle could operate any time but was required in circumstances where a threshold threat was established.159 In both circumstances, the principle was required to be weighed up against other objectives and not given undue weight,160 a conclusion which reflected the competing objectives included in the Australian Fisheries Management Authority’s (AFMA) framework legislation.161 This understanding of the principle in DC terms can also be seen in other AFMA cases162 and has led to the idea that AFMA can legitimately pursue the precautionary principle as one of its aims.163 In all these cases where the precautionary principle is understood in DC terms, it is operating as an enabling principle by providing the ‘philosophical authority’ for DC administration.164 Yet while a DC understanding of the principle has often been held to be relevant, it seems to have had very little impact upon decisionmaking beyond requiring a decision-maker to be cautious.165 This often seems a very low threshold to cross166 due to the principle not dictating a ‘specific course of action to the exclusion of others’.167 This state of affairs can be understood as due to two interrelated issues. First, it reflects the fact that judicial review has always played a limited role in controlling discretion and in reviewing the assessment of the facts.168 If merits review when understood as external review is 156 Miltonbrook Pty Ltd v Kiama Municipal Council [1998] NSWLEC 281 at para 281; and Murrumbidgee Groundwater Preservation Association, above n 144. There are exceptions where, although the precautionary principle has been interpreted in DC terms, there has been some attention paid to quality of the evidence and evidentiary burdens: see North Queensland Conservation Council, above n 46 at para 204, and Providence Projects Pty Ltd v Gosford City Council [2006] NSWLEC 5 at para 81. 157 Dixon, above n 44 158 Ibid at paras 103, 117 and 120. 159 Ibid at para 120. 160 Ibid at paras 174 and 182. 161 Fisheries Management Act 1994 (Cth), s 3. 162 Justice v Australian Fisheries Management Authority [2002] AATA 49 at para 37; and Green v Australian Fisheries Management Authority [2004] AATA 426 at para 35. 163 W Gullett et al, ‘Substantive Precautionary Decision-making: The Australian Fisheries Management Authority’s Lawful Pursuit of the Precautionary Principle’ (2001) 2 Australasian Journal of Natural Resources Law and Policy 95. 164 Stein, above n 93. 165 Gullett et al, above n 163; and E Fisher, ‘Changing the Mindset? Implementing the Precautionary Principle in Australia’ (1999) 7 Research in Social Problems and Public Policy 183. 166 Alumino (Aust) Pty Ltd v Minister Administering the Environmental Planning and Assessment Act 1979 [1996] NSWLEC 102 and Northcompass Inc v Hornsby Shire Council [1996] NSWLEC 213. 167 Bridgetown/Greenbushes Friends of the Forest Inc, above n 50 at 119. 168 Fisher, above n 41; and Jowell, above n 134.

(F) Fisher Ch4

31/5/07

146

09:58

Page 146

The Precautionary Principle and Merits Review in Australia

modelled on judicial review then it is not surprising that it takes a similar approach. Second, in some cases the principle has been understood to be an articulation of a philosophy already entrenched in Australian public law and public administration, and decision-makers are characterised as engaging in cautious consideration of issues already.169 Thus in a number of cases where the principle was not held to be legally relevant, the decision-maker was nonetheless found to be acting in a precautionary manner.170 The operation of the principle in DC terms was thus seemingly legitimating existing patterns of decision-making.171 This state of affairs reflects the fact that DC ideals have always underpinned Australian environmental administration, and interpreting the principle in DC terms has meant that the courts and tribunals have tended to defer to the judgment of the decision-maker so long as it has demonstrated that it has exercised its discretion cautiously.172

V RI Interpretations of the Precautionary Principle and Merits Review The precautionary principle has not only been understood in DC terms, however. It has also been interpreted in line with the RI paradigm of administrative constitutionalism. This has primarily occurred in cases where the principle has been understood as requiring a ‘shifting of the burden proof’ onto the proponent. As noted above, this characterisation has been popular among some scholars and some litigants,173 but makes little sense if merits review is understood as a form of external review. If, however, merits review is understood as a form of adjudicative/adversarial dispute resolution then such a conceptualisation is more plausible. As such, merits review is understood in RI terms and a consequence of this, as shall be seen below, is that substantive guidance for primary decision-makers is also understood in RI terms. This RI conceptualisation of the principle was raised in one of the earliest legal disputes in which the principle was considered, Queensland Nickel Management Pty Ltd v Great Barrier Reef Marine Park Authority (the Queensland Nickel Management decision),174 which concerned whether approval should be given for 169

Nicholls, above n 135 at 419. Ibid at 419; Friends of Hinchinbrook Society Inc, above n 50 at 678–9; P & E Turner v Launceston City Council [1998] TASRMPAT 27; and Kroger, above n 154. 171 Leatch, above n 50 at 282. 172 Grishin, above n 50 at 15; Dow Chemicals (Australia) Ltd, above n 47 at 13; and North Queensland Conservation Council, above n 46 at para 171. 173 See n 53 above, and the arguments put in Dixon, above n 44. 174 As this case settled, there was no decision made by the AAT. A lengthy analysis of the proceedings can be found at H Prokuda, ‘The Precautionary Principle—A Case Study’, The Precautionary Principle (Institute of Environmental Studies, University of New South Wales, Sydney, 1993). 170

(F) Fisher Ch4

31/5/07

09:58

Page 147

RI Interpretations of the Precautionary Principle and Merits Review

147

an ore off-loading facility in the Greater Barrier Reef Marine Park. That case settled, but only after 92 sitting days of hearings, 78 witnesses, and 400 exhibits, in which the AAT, which heard the case, substantively reviewed the evidence.175 While the principle was largely accepted as requiring a shifting of the burden of proof, it was also seen as problematic in that it raised questions about the relationship between legal and scientific proof, the different roles of the parties, and whether proof of environmental impact and mitigation measures needed to be established separately.176 Moreover, the case also caused the Administrative Review Council to argue that merits review should be reformed in relation to environmental decision-making so that it focused more on the procedural aspects of decision-making so as to ensure that there had been proper deliberation.177 With that said, the RI conceptualisation of the principle was also accepted by a number of different legal actors, including one of the AAT members who sat on the Queensland Nickel Management case.178 The real starting point for the RI approach can be seen in the SA ERDC decision of Conservation Council of South Australia v Tuna Boat Owners Association (No 2)179 (the Tuna Boat Owners decision). This merits review decision concerned the licensing of tuna farms off the coast of SA. The project was challenged for not being ecologically sustainable and part of this argument was that the precautionary principle had not been applied properly. Judge Trenorden characterised the principle as being concerned with the limitations in scientific method and as applying to the question of what evidence was needed to make the decision.180 In terms of the case before her, she understood the application of the principle in an evidentiary context. Thus she saw it as relevant to determining ‘who has the onus of satisfying us that the proposed development would be carried out in an ecologically sustainable way, and located, sited, designed, constructed and managed to be ecologically sustainable’.181 She was thus characterising the precautionary principle in terms of a bipolar dispute resolution in which she, standing in the shoes of the primary decision-maker, had to assess whether parties to the case had discharged their burdens of proof. This is a very RI characterisation of decisionmaking, suggesting as it does that both the judge and the original decision-maker are assessing evidence rather than exercising discretion. Moreover, this characterisation can be seen more obviously in the way in which Judge Trenorden described how the precautionary principle should be applied. She stated that a person proposing a development would normally need to prove 175 Administrative Review Council, Review of Commonwealth Environmental Impact Assessment Decisions: Discussion Paper (Canberra, AGPS, 1993) at 10–11. 176 Prokuda, above n 174 at 190–94. 177 Administrative Review Council, above n 88. 178 Prokuda, above n 174 at 193–4, and E Christie, ‘The Eternal Triangle: the Biodiversity Convention, Endangered Species Legislation and the Precautionary Principle’ (1993) 10 Environmental and Planning Law Journal 470. 179 Conservation Council of South Australia v Development Assessment Commission & Tuna Boat Owners Association (No 2) [1999] SAERDC 86. 180 Ibid at para 24. 181 Ibid at para 24.

(F) Fisher Ch4

31/5/07

148

09:58

Page 148

The Precautionary Principle and Merits Review in Australia

that their development was consistent with the Development Plan.182 In relation to the precautionary principle, the party challenging the development would need to ‘show that there is a prospect of serious or irreversible damage to the environment, should the proposed development proceed’.183 They would not need to ‘prove that the development will threaten serious or irreversible environmental damage’ because scientific uncertainty would make this highly difficult. But Judge Trenorden advised that a challenger should keep in mind the legislative provisions which allow the court to dismiss vexatious or delaying litigation.184 Again this is an adjudicative, and thus RI, characterisation of decision-making in which considerable emphasis is placed on litigants proving their case. Judge Trenorden then went on to note that if the challenger does discharge its burden of proof: the burden of proof switches to the proponent and it will be necessary for the proponent to show, in order to have his or her development classified as ecologically sustainable, the following: —the measures that the proponent will take (within the limits of practicability) to avoid serious or irreversible damage to the environment; and —that the risk-weighted consequences of the development assessed together do not suggest that serious or irreversible environmental damage would be sustained. ... The proponent would have to satisfy the burden of proof by evidence as to the likely consequences of the proposal, including scientific evidence (with its limitations), evidence as to the proposed management regime and measures, and evidence to assist the Court in the assessment of the risk-weighted consequences of the proposal.185

On this characertisation, the precautionary principle is an adjustment to the burdens of proof in merits review litigation. This is a very different characterisation from that seen in the last section, where the principle was understood as applying generally to the exercise of discretion. In this case the court refused the licence mainly on the basis that the legislative framework was not appropriate for putting in place the necessary adaptive management framework that was required to ensure ecologically sustainable development.186 Judge Trenorden’s RI characterisation of decision-making did not stop at the precautionary principle, however. She also characterised the role of those giving expert evidence in RI terms and criticised expert witnesses for misunderstanding their legislative role in providing advice.187 Such witnesses she characterised as only providing scientific information, and she noted by analogy: 182 Conservation Council of South Australia, above n 179 at para 24, although note the ambiguity on this point in Tuna Boat Owners Association of South Australia Inc, above n 130 at paras 27–30. 183 Ibid at para 24. 184 Ibid at para 24, referring to Environment, Resources and Development Court Act 1993 (SA), s 17(4). 185 Ibid at 24–25. 186 Ibid at para 41. 187 Development Act 1993 (SA), s 37.

(F) Fisher Ch4

31/5/07

09:58

Page 149

RI Interpretations of the Precautionary Principle and Merits Review

149

In the same vein, a scientist, whose task is to collect and analyse data, test it against a hypothesis and draw a conclusion, merely reports the results. It is not the task of the scientist to carry out a risk assessment in the context of the community’s expressed social and political goals.188

Again, this is a very RI understanding of the role of public administration and rests on the assumption that factual assessment and other aspects of decision-making can be easily distinguished. This is not an assumption seen where the principle has been interpreted in DC terms. Tuna Boat Owners was appealed to the Supreme Court of South Australia, where it was overturned because Doyle CJ found that the legislative framework could accommodate an adaptive management approach.189 At the same time, however, he also upheld the ERDC’s reasoning in regards to the precautionary principle, stressing that it was acceptable for an onus to be placed on the applicant to prove that the development was ecologically sustainable,190 although he also noted that the principle was not necessarily a proposition of law.191 This ‘burden of proof’ approach has not been limited to the SA ERDC and can also be seen in other jurisdictions, particularly those where there has been a heavy borrowing from adjudicative method in formulating merits review. Thus, for example, the Qld PEC is another court in which the precautionary principle has largely been characterised in terms of burdens of proof.192 This is not surprising in light of the fact that the court operates as a court and its governing statute regulates who must prove a case.193 Thus, for example, in considering the principle that court has noted: This statement of principle does not, as I read it, depart in any important way from the approach which this court has taken conventionally with such matters prior to the statements becoming part of the legislation. It has generally been held that the appellant faces a burden of establishing, on the balance of probabilities, that when a particular activity might pose a risk of environmental harm, those risks have been evaluated and measures intended to provide protection from environmental harm are feasible and likely to be put in place.194

Thus, for example, in Shannon v Dalby Town Council 195 the court was considering a third party challenge to planning permission being granted to an ethanol plant. The objector argued, among other things, that the council had not properly taken into account the precautionary principle.196 The court understood this as the objector arguing that the ‘expert evidence adduced for the applicant fell short of 188

Conservation Council of South Australia, above n 179 at para 45. Tuna Boat Owners Association of South Australia Inc, above n 130 at paras 55–60. 190 Ibid at paras 27–8. 191 Ibid at para 30. 192 The precautionary principle has had a significant role to play in Qld town planning law because of its inclusion in the Integrated Planning Act 1997, see s.1.2.3(2). 193 Ibid, s 4.1.50. 194 CSR Limited v Caboolture Shire Council [2001] QPE 013 at para 52. 195 Shannon v Dalby Town Council [2004] QPEC 062. 196 Ibid at para 20. 189

(F) Fisher Ch4

31/5/07

150

09:58

Page 150

The Precautionary Principle and Merits Review in Australia

establishing the plant could operate safely’.197 Thus the court held that the application of the principle would be: by considering the evidence which might attract it and determining whether, on balance and by reference to the requisite standard of proof there remains, nevertheless, a state of uncertainty about environmental impacts. Importantly, that does not involve placing an onus upon the applicant to extinguish, with absolute certainty, the risk of environmental harm.198

The court thus proceeded to assess each piece of expert evidence to see whether it addressed the concerns over environmental impacts adequately. In some cases, evidence was discounted as not believable199 but generally the evidence was accepted. Thus the court saw itself in largely passive terms. Thus, for example, in relation to the issue about the safety of emissions from the plant, the court concluded that: The breadth of the evidence about these matters and the plain tenor of it supports the applicant’s contention that there is no reasonable ground for concern as to the effect of emissions, and the application should be accepted subject to the conditions imposed by the independent concurrence agencies which, for the reasons given earlier, is an appropriate course. Those conclusions carry the necessary inference that, contrary to the appellant’s contentions, the scientific evidence about emissions and odour is more than sufficient to extinguish uncertainty to the requisite degree.200

The operation of the precautionary principle was thus largely understood as relating to the operation of evidence between the two parties. The court had no responsibility in relation to the principle otherwise. A similar approach can be seen in other Qld PEC cases.201 An RI interpretation of the precautionary principle can also be seen in a number of decisions in the AAT. Thus in Mohr v Great Barrier Reef Marine Park Authority 202 the tribunal noted that the Great Barrier Reef Marine Park Authority was bound by the precautionary principle and then quoted the IGAE version of the principle. The tribunal then stated: We observe that an assessment of risk weighted consequences [the requirement in the IGAE] is necessary to ensure procedural fairness in the decision making process where developmental interests (as reflected in this case by the mariculture proposal) and conservation interests compete. Given the fact that no party has provided any evidence on the risk weighted consequences of the mariculture proposal, the Tribunal has not considered the operation of the precautionary principle to its decision making process.203

197 198 199 200 201 202 203

Shannon v Dalby Town Council [2004] QPEC 062 at para 23. Ibid at para 28. Ibid at para 33. Ibid at para 41. Histpark Pty Ltd, above n 50 at para 66. Mohr v Great Barrier Reef Marine Park Authority [1998] AATA 805 . Ibid at para 124.

(F) Fisher Ch4

31/5/07

09:58

Page 151

RI Interpretations of the Precautionary Principle and Merits Review

151

Here the tribunal does not see itself as under a pro-active duty to apply the principle the way we saw the NSW LEC was in the decisions discussed in the last section. As the parties had brought no evidence to the table there was no need to consider it. The AAT, in the shoes of the primary decision-maker, is thus a passive assessor of information rather than a deliberative problem-solver. The RI approach to interpreting the precautionary principle can also be seen in relation to some of the more recent AAT AFMA case-law, in which the principle has been understood as requiring the consideration of ‘risk-weighted consequences’ where onuses of proof are in operation and a threshold threat needs to be established. As seen above, expert witnesses in cases concerning AFMA did characterise the principle as a ‘shifting of a burden of proof ’, and while such a characterisation was not fully accepted in Dixon204 it has been in some other cases.205 In De Brett v AFMA (the De Brett decision)206 the Federal AAT was reviewing an AFMA decision to attach a condition to fishing permits in a certain fishery so that each permit ‘prohibits the possession, carrying and landing of shark fins that are not attached to the trunk [of the shark]’.207 The reason for this condition was that evidence had come to light that the unregulated practice of shark finning208 was a threat to the shark levels as well as being a cruel and wasteful activity.209 The challengers to the decision argued that the decision-maker, among other things, had ‘misconstrued or misapplied’ the precautionary principle.210 The tribunal described its role as carrying out ‘de novo review on the merits’ and noted that, while there was technically ‘no general rule’ that one or other party bore a burden of proof,211 in practice each party had an ‘evidentiary burden’ in that each party would want to have evidence before the tribunal which would support their case.212 Moreover, the inquisitorial powers of the tribunal were limited by resources.213 As such, the tribunal tends to largely depend upon the material given to it by the parties, either at their own instigation or at the instigation of the Tribunal, and tested to the extent that it can be tested through the examination and cross examination of each witness and its own questioning of witnesses.214

This is a highly RI conceptualisation of decision-making. The decision-maker is an umpire assessing the information before them which is provided by the parties. Moreover, in discussing the principle the tribunal understood its importance in relation to scientific evidence and discussed the concept of scientific proof at some 204 205 206 207 208 209 210 211 212 213 214

Dixon, above n 44. Blank v Australian Fisheries Management Authority [2000] AATA 1027 at para 51. De Brett Investments Pty Ltd v Australian Fisheries Management Authority [2004] AATA 704. Ibid at para 1. Removing the fin of a shark at sea and then discarding the shark back into the sea. De Brett Investments Pty Ltd, above n 206 at paras 37–9. Ibid at para 115. Ibid at para 126. Ibid at para 127. Ibid at para 129. Ibid at para 53.

(F) Fisher Ch4

31/5/07

152

09:58

Page 152

The Precautionary Principle and Merits Review in Australia

length, particularly how it related to legal burdens of proof.215 In regards to the threshold ‘threat’, the tribunal concluded that it needed to be established on the ‘balance of probabilities’—a standard which could vary depending on the circumstance.216 Moreover, the precautionary principle was understood in evidentiary terms. It is useful to quote from the tribunal’s decision at length. At the practical level for us, the precautionary principle means that we must assess whether there is an indication that there will be some serious or irreversible environmental damage if a certain course is followed, including the course of taking no action at all. That means that we must assess the possible consequences and gravity of those courses’ being followed together with the risk of those consequences’ occurring. That assessment must be carried out having regard to all sources of evidence; it is not limited to scientific evidence. If the assessment leads to us to conclude on the balance of probabilities that there is a threat of serious or irreversible damage to the environment that is not a bare possibility, full scientific certainty in the sense we have explained that concept should not be used as a reason for postponing measures to prevent environmental degradation. Caution should be exercised. The outcome of our assessment in applying the precautionary principle must be weighed with the other objectives in s. 3(1) of the Act and a decision reached.217

This is a broader characterisation of the evidence than seen in other cases but it is still very much an RI conceptualisation of the principle in that the principle is adjusting the burden of proof in the assessment of the evidentiary aspects of the decision. In regard to the decision before it, the tribunal concluded that it could not assess whether there was a threat of serious or irreversibly environmental damage because of the lack of information before it.218 It ultimately upheld AFMA’s decision, on other grounds. Besides being a good example of understanding the precautionary principle in RI terms, there are two other interesting features of the De Brett decision. The first is that the tribunal is aware of the distinction between scientific evidence and the evidence before it but at the same time is attempting to collapse each into the other. Scientific concepts of proof are thus mapped onto adjudicative/adversarial notions of proof. The second feature is that the RI understanding of the precautionary principle is, to a certain extent, unworkable, a point made by Talbot J in Nicholls.219 The tribunal is faced with the problem that the complexities of both scientific and behavioural uncertainties mean that it is very difficult for any decision-maker to assess ‘risks’ in a formulaic way. Thus, in this case it had very little information about sharks and shark stocks, very little information about bycatch practices, and very little knowledge about the consequences of finning on sharks.220 215 216 217 218 219 220

Ibid at paras 148–52. Ibid at para 162. See also Jones and Bronitt, above n 53 at 142. De Brett Investments Pty Ltd, above n 206 at para 162. Ibid at para 170. Nicholls, above n 135 at 419. De Brett Investments Pty Ltd, above n 206 at paras 167–70.

(F) Fisher Ch4

31/5/07

09:58

Page 153

RI Interpretations of the Precautionary Principle and Merits Review

153

There are a number of different consequences of the RI interpretation of the principle. First, the application of the precautionary principle is being circumsrcibed quite narrowly. The principle is not an enabling principle as it was when understood in DC terms; rather, it guides only the assessment of evidence. Moreover, it must take a subsidary role to the objective evaluation of benefits and costs221 and to the statutory scheme.222 A second consequence of the RI approach to the precautionary principle is that the concept of ‘threat’ becomes an important threshold that those arguing for the application of precautionary principle must meet.223 The concept of threat has two dimensions. First, the risk must not be a small one,224 and, second, there must be some evidence about the risk. In regard to the latter point, in Mol Pty Ltd v City of Mitcham the SA ERDC stated that for the principle to apply ‘evidence should be presented to demonstrate that a natural resource is at some form of risk’.225 At the same time, scientific uncertainty about that risk must also be shown.226 In Aldekerk Pty Ltd v City of Port Adelaide Enfield the ERDC concluded that: the precautionary principle should only be applied, we think, in those cases where there is genuine uncertainty or ignorance on relevant scientific matters and there is a risk of substantial impairment to the environment. It cannot be used to prop up a decision which is unsupported by tenable evidence.227

The practical result of these cases is that they create an evidentiary burden for those challenging decisions. The nature of that evidentiary burden varies from case to case. Thus in some cases the burden is simply about environmental impact,228 while in others it includes other information about a project.229 Moreover, if there is sufficient evidence of a threat then the principle does not apply. Thus in Terminals Pty Ltd v Greater Geelong City Council it was concluded that due to a risk assessment being conducted which was ‘scientifically based’ the precautionary principle was inapplicable.230 A third consequence, as seen in De Brett 231 is that the issue of standard of proof becomes important. Thus in David Kettle Consulting v Gosford City Council the NSW LEC concluded that, while there was enough evidence that a certain volume of water could be sensibly extracted from an aquifer, the court was not satisfied on the balance of probabilities that such an activity should occur on an ‘absolutely unconstrained and open-ended basis’ and thus limited the licence period.232 221

Hasan v Moreland City Council [2005] VCAT 1931 at para 25. Rowe v Linder [2006] SASC 176 at para 77. 223 Telstra Corporation Ltd v Caloundra City Council [2004] QPEC 85 at para 59. 224 Dubler v Ku-ring-gai Muncipal Council (NSWLEC, unreported, 21 December 2001) at para 18. 225 Mol Pty Ltd v City of Mitcham [2002] SAEDRC 55 at para 99; and Commercial Crash Repairs Pty Ltd v City of Adelaide [2000] SAERDC 83 at para 43. 226 Elliott v Brisbane City Council [2002] QPEC 013 at para 39. 227 Aldekerk Pty Ltd, above n 50 at para 25. 228 Conservation Council of South Australia, above n 179. 229 De Brett Investments Pty Ltd, above n 206. 230 Terminals Pty Ltd v Greater Geelong City Council [2005] VCAT 1988 at para 140. 231 De Brett Investments Pty Ltd, above n 206. 232 David Kettle Consulting v Gosford City Council [2005] NSWLEC 519 at paras 54–5. 222

(F) Fisher Ch4

31/5/07

154

09:58

Page 154

The Precautionary Principle and Merits Review in Australia

It is useful to note that, as with the DC intepretation of the principle, there are problems with the RI interpretation as well. First, as has been highlighted in a number of cases,233 discharging such a burden of proof, even on the balance of probabilities, has a conceptual impossibility to it. This is particularly because the decision-maker is not only garnering information but also evaluating and critiquing information, and information gathered in circumstances of scientific uncertainty will always be open to critcism.234 Just as the DC interpretation of the principle in the hands of a reviewing court is perhaps too flexible to provide any real constraint on administrative action, the RI interpretation is too great a restraint. Not only can a decision-maker not act in the face of scientific uncertainty until certain conditions precedents are met, but when they are met, there also seems to be little discretion. Second, the RI interpretation of the principle is very vague on the information in relation to which the burden of proof operates. Clearly it is more than just the scientific information and would seem to include information about how a development operates.235 At the same time, however, the burden of proof characterisation is often justified by reference to science.236 Third, the application of the principle is caught on the horns of needing to prove that there is a threat but not proving it to such an extent that the principle does not apply. In light of the fluid and flexible nature of scientific information, these two requirements are almost paradoxical.237

VI The Search for a Uniform Interpretation What can be seen in the previous two sections is that understandings of the precautionary principle and merits review are being co-produced. In many of these cases, the starting point is how the courts and tribunals understand their task in merits review, which then influences how they interpret the precautionary principle. In some other cases, however, a certain interpretation of the precautionary principle influences how a court or tribunal understands its role. Thus, for example, the WA Town Planning Appeal Tribunal has commented in the context of a town planning merits review: As the Tribunal has pointed out in several decisions, there is no onus on the Appellant on an Appeal, aside from an evidentiary onus. However, the precautionary principle creates an onus on the Appellant to prevent or offer a solution to the risk of damage from impacts that are not known or understood.238

233 234 235 236 237 238

Prokuda above n 174 and Nicholls above n 135. Prokuda above n 174. Shannon, above n 195. Conservation Council of South Australia, above n 179. See a similar interpretation of the principle in ch 6, below. St Ives Development Pty Ltd v Mandurah [2003] WATPAT 5 at para 62.

(F) Fisher Ch4

31/5/07

09:58

Page 155

The Search for a Uniform Interpretation

155

Thus an RI interpretation of the principle leads the Town Planning Appeal Tribunal to an adjudicative/adversarial understanding of its role. In light of this, it is better to understand the developments discussed above as processes of coproduction in that different interpretations of the principle co-exist alongside different understandings of merits review. The consequences of these processes of co-production are that in Australia there are two divergent understandings of the precautionary principle in operation. As the AAT noted in De Brett about the precautionary principle: its interpretation is not quite as clear as was perhaps the case when Stein J decided Leatch v National Parks and Wildlife Service.239

Moreover, it is not simply the case that either interpretation is isolated to a single court or tribunal, and both interpretations can be seen right across the different courts and tribunals, and in any particular court or tribunal both interpretations can be seen in operation.240 Thus, while the RI interpretation is particularly prevalent in the SA ERDC and the Qld PEC, DC interpretations can also be found in each of these courts.241 Likewise, RI interpretations of the principle can be seen in NSW LEC decisions, even though a DC interpretation of the principle has dominated in that court.242 What these processes of co-production also highlight is that understandings of both merits review and the precautionary principle are ambiguous in the Australian context. As seen in other chapters of this book, that uncertainty is inherent in administrative constitutionalism and thus it is not surprising that there are two competing interpretations of the precautionary principle. Very recently there has been a significant attempt by the Chief Justice of the NSW LEC, Preston CJ, to provide an authoritative analysis and definition of the precautionary principle, in Telstra Corporation Ltd v Hornsby Shire Council (the Telstra decision).243 The case concerned an appeal on the merits by a mobile phone company in relation to the refusal by the local council to grant a development application to it for placing two mobile phone antennae on a building in an urban area.244 The conclusion was that the appeal should be allowed, an unsurprising result that is consistent with other authorities on the issue.245 What is interesting about the decision is Preston CJ’s discussion of the precautionary principle, and extrajudicially Preston CJ has described the case as ‘one of the most detailed and comprehensive elaborations of the precautionary principle in a 239

De Brett Investments Pty Ltd, above n 206 at para 158. In the NSW LEC compare David Kettle Consulting, above n 232 with Gales Holdings Pty Limited, above n 147. In the AAT compare North Queensland Conservation Council, above n 46, with De Brett Investments Pty Ltd, above n 206. 241 Port Vincent Progress Association v DAC & Colimion P/L [1999] SAERDC 7 at para 36; and Yamauchi v Jondaryan Shire Council [1998] QPELR 452. 242 David Kettle Consulting, above n 232. 243 Telstra Corporation Ltd, above n 43. 244 Ibid at para 7. 245 Ibid at paras 185–198. 240

(F) Fisher Ch4

31/5/07

156

09:58

Page 156

The Precautionary Principle and Merits Review in Australia

national court decision to date’.246 In it he attempted to reconcile not only some of the Australian authorities on the principle but overseas ones as well. His starting point for analysing the precautionary principle was a very RI one in that he characterised the principle as operating when two different thresholds were met. He noted: The application of the precautionary principle and the concomitant need to take precautionary measures is triggered by the satisfaction of two conditions precedent or thresholds: a threat of serious or irreversible environmental damage and scientific uncertainty as to the environmental damage. These conditions or thresholds are cumulative. Once both of these conditions or thresholds are satisfied, a precautionary measure may be taken to avert the anticipated threat of environmental damage, but it should be proportionate.247

He also stated that ‘the threat of environmental damage must be adequately sustained by scientific evidence’248 and, in doing so, referred to the leading US case concerning the treatment of expert evidence in the context of a trial—Daubert v Merrell Dow Pharmaceuticals.249 He also described the second threshold to be met in scientific terms.250 If these two conditions precedent are met, then Preston CJ described the principle as requiring the following: At this point, there is a shifting of an evidentiary burden of proof. A decision-maker must assume that the threat of serious or irreversible environmental damage is no longer uncertain but is a reality. The burden of showing that this threat does not in fact exist or is negligible effectively reverts to the proponent of the economic or other development plan, programme or project. The rationale for requiring this shift of the burden of proof is to ensure preventative anticipation; to act before scientific certainty of cause and effect is established. It may be too late, or too difficult and costly, to change a course of action once it is proven to be harmful. The preference is to prevent environmental damage, rather than remediate it. The benefit of the doubt is given to environmental protection when there is scientific uncertainty. To avoid environmental harm, it is better to err on the side of caution. The function of the precautionary principle is, therefore, to require the decisionmaker to assume that there is, or will be, a serious or irreversible threat of environmental damage and to take this into account, notwithstanding that there is a degree of scientific uncertainty about whether the threat really exists.251

This is clearly an RI understanding of the principle, in which the decision-maker is required to follow a certain course of action if both a threat and scientific uncertainty are established. The evidence of threat is also relatively narrow and does not include evidence about mitigation measures or how a development will operate in 246 B Preston, ‘Ecologically Sustainable Development in the Courts in Australia and Asia’, Paper for seminar held at Buddle Findlay, Lawyers (Wellington, New Zealand, 28 August 2006) at 18. 247 Telstra Corporation Ltd above n 43 at para 128. 248 Ibid at para 134. 249 Ibid at para 135; citing Daubert v Merrell Dow Pharmaceuticals 509 US 579 (1993). 250 Ibid at paras 140–49. 251 Ibid at paras 150–51.

(F) Fisher Ch4

31/5/07

09:58

Page 157

The Search for a Uniform Interpretation

157

practice. Moreover, the meeting of these thresholds is important for establishing the reasonableness of administrative action. Later in the judgment, and using European Union case-law as authority,252 Preston CJ stated: Rationality dictates that the precautionary principle and any preventive measures cannot be based on a purely hypothetical approach to risk, founded on mere conjecture which has not been scientifically verified.253

Moreover, when a threshold threat is established the decision-maker must presume that there is a threat of serious or irreversible environmental damage and can only not do so if the developer discharges their burden of proof. Preston CJ does not understand the whole of the decision-making process in these RI terms, however. In some places in his judgment he defines ‘threat’ very broadly254 and argues that ‘different judgments, values and cultural perceptions of risk, threat and required action play a role in the assessment process’.255 More significantly, he argues that the shifting of the evidentiary burden of proof is only operating in relation to the ‘question of environmental damage’. The assumption about damage that such a shifting creates is only one factor in the decision-making process. He states: This assumed factor [the threat of serious or irreversible environmental damage] must be taken into account in the calculus which decision-makers are instructed to apply under environmental legislation . . . There is nothing in the formulation of the precautionary principle which requires decision-makers to give the assumed factor (the serious or irreversible environmental damage) overriding weight compared to the other factors required to be considered, such as social and economic factors, when deciding how to proceed.256

This, in contrast with the approach he took in relation to environmental damage, is understanding the precautionary principle to apply to DC decision-making, and, while Preston CJ is requiring the evidence of environmental impact to be assessed on RI terms, he is also understanding the wider decision-making framework to be DC in nature.257 Preston CJ thus argues strongly for adaptive management258 and is critical of the use of cost/benefit analysis in the assessment of proportionality.259 In relation to the latter, he does promote as its replacement another analytical tool—multi-criteria analysis.260 Preston CJ’s judgment is clearly an attempt to make sense of the large body of case-law and commentary on the precautionary principle both in Australia and 252 In particular, Case T-13/99 Pfizer Animal Health SA v Council of the European Union [2002] ECR II–3305. 253 Telstra Corporation Ltd, above n 43 at para 159. 254 Ibid at paras 129–131. 255 Ibid at para 132. 256 Ibid at para 154. 257 Ibid at paras 154–5, 174, 179, and 183. 258 Ibid at paras 163–5. 259 Ibid at paras 174–6. 260 Ibid at para 176.

(F) Fisher Ch4

31/5/07

158

09:58

Page 158

The Precautionary Principle and Merits Review in Australia

overseas. In that regard, his judgment suffers from problems of failing to take comparative law methodology into account in that he is taking references to the principle out of their legal context, particularly the European authorities.261 At the same time, however, this judgment is an attempt at an explicit reconciliation of the DC and RI interpretations of the principle. But it is a reconciliation that can only occur by dividing the decision-making process into an RI process and a DC process. Thus, first, he requires a structured RI response to assessment of impact, which relies on tools such as a careful risk assessment.262 This is what the shifting of the burden of proof operates in relation to. Second, he is not limiting the application of the principle to an assessment of the evidence about environmental impact impact, however. Preston CJ is also maintaining that the principle applies to the overall decision-making process, which is DC in nature. Whatever conclusions are reached in relation to information about environmental impact, they do not dictate a final outcome, and decision-makers should be flexible, not limited by analytical tools, and they should promote adaptive management.263 The reasoning in Telstra was applied in a recent NSW LEC judicial review decision, Gray v Minister for Planning (the Gray decision).264 The case concerned a challenge by a third party to a decision by the Director General of the Department of Planning that an environmental assessment for a large coal mine was adequate even though it had not taken into account the environmental impact of the greenhouse gases which would be produced by end users of the coal. The case was in the context of a complex legislative scheme for environmental assessment of major infrastructure projects265 and was also somewhat moot because the information about such impacts had been later submitted. The judge, Pain J, found that while the environmental assessment complied with the environmental assessment requirements laid down by the Director General, the Director General had failed to take into account the principles of intergenerational equity and the precautionary principle. In coming to this conclusion she characterised EIA as playing an important role in implementing ESD principles.266 In doing so, she cited Preston CJ from an earlier decision where he noted: Prior environmental impact assessment and approval are important components in a precautionary approach. The precautionary principle is intended to promote actions that avoid serious or irreversible damage in advance of scientific certainty of such damage. Environmental impact assessment can help implement the precautionary principle in a number of ways including: (a) enabling an assessment of whether there are threats of damage to threatened species, populations or ecological communities;

261 262 263 264 265 266

Fisher, above n 4. Telstra Corporation Ltd, above n 43 at paras 161–2, 167, and 176. Ibid at paras 163–5. Gray v Minister for Planning [2006] NSWLEC 720. Environmental Planning and Assessment Act 1979 (NSW), Part 3A. Gray, above 264 at paras 116–17.

(F) Fisher Ch4

31/5/07

09:58

Page 159

The Search for a Uniform Interpretation

159

(b) enabling an evaluation of the conclusiveness or certainty of the scientific evidence in relation to the threatened species, populations or ecological communities or the effect of proposed development on them; (c) enabling informed decisions to be made to avoid or mitigate, wherever practicable, serious or irreversible damage to the threatened species, populations or ecological communities and their habitats; and (d) shifting the burden of proof (evidentiary presumption) to persons responsible for potentially harmful activity to demonstrate that their actions will not cause environmental harm.267

This decision, Bentley v BGP Properties Pty Ltd,268 was decided just before Telstra and is a far less careful analysis of the principle and, in particular, how it relates to the burden of proof.269 The statement above is, however, a very RI understanding of the role of EIA and, as such, a departure from the earlier DC understandings seen in the first section of this chapter. The role of EIA under this newer formulation is to enable an ‘evaluation of the conclusiveness or certainty of the scientific evidence’ rather than be a means of directing a ‘reasonably intelligent and informed mind to the possible environmental consequences’.270 Returning to Gray, Pain J elaborated on this RI understanding of EIA in determining a role for the precautionary principle in the assessment of information. She stated: inherent in the precautionary principle . . . is the need for careful evaluation to avoid serious or irreversible damage to the environment and an assessment of the risk weighted consequences for various options. The role of environmental assessment is to assist in providing information to the decision-maker to enable him or her to consider that scientific uncertainty in relation to the serious, irreversible environmental threat, in this case climate change/global warming, as identified above at par[agraph] 116 in the passage from Bentley . . .271

Yet this operation of the precautionary principle is not determinative. Pain J thus states: if the precautionary principle did apply so that there was a shifting of the evidentiary burden of proof to a proponent in relation to environmental damage this is but one of the factors a decision maker under the EP&A Act must consider and is not determinative of the outcome of that decision making process.272

In other words, the principle as requiring a form of RI environmental assessment acts as a ‘precautionary enabling device’273 for the DC consideration of the issues, 267

Bentley, above n 42 at para 68. Ibid. 269 It should also be noted that the case involved the issue of sentencing and was not a judicial review. 270 Prineas, above n 23 at 417. 271 Gray, above n 264 at para 131. 272 Ibid at para 139. 273 Ibid, quoting J Cameron, ‘The Precautionary Principle: Core Meaning, Constitutional Framework and Procedures for Implementation’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999) at 52–4. 268

(F) Fisher Ch4

31/5/07

160

09:58

Page 160

The Precautionary Principle and Merits Review in Australia

and without such an assessment a proper consideration of the DC principle could not take place.274 The final decision about planning permission had not been given, however, and was not the subject of litigation. But Pain J did note that, in making the final decision, the Minister for Planning would need to take into account all relevant considerations and that the principle was not determinative of outcome.275 What can be seen in these two cases, one a merits review and one a judicial review, is a new characterisation of the decision-making process so as to accommodate the RI and DC definitions of the precautionary principle. As admirable an exercise in consolidation as this is, there are serious doubts over whether such a division is workable in practice. One should be careful of accepting the neatness of this division for two reasons. First, it is not clear how it would operate in practice in relation to merits review. In Telstra276 Preston CJ’s analysis of the expert evidence was in very RI terms and beyond making general statements about the precautionary principle there is little in his decision which takes the form of external review. Second, it has been shown in other jurisdictions that the distinction between the scientific process of risk assessment and the political process of risk management is impossible to achieve in practice.277 Even from these two judgments, similar problems can be forecast here. Thus Preston CJ argues that socio-political issues need to be taken into account in thinking about threat but it is not clear how these issues are reconciled with the discharging of an evidentiary burden of proof. Likewise, while the overall decision-making is understood to be DC in nature, at the same time he requires decision-making to have a clear evidentary basis.

VII Conclusions The very recent nature of these decisions makes it too early to tell how the case-law will develop in this area. But what all these cases highlight is that the way in which the precautionary principle is interpreted is highly dependent on understandings of administrative constitutionalism embedded in legal culture. Moreover, the state of affairs I have described above is less a product of explicit debate about the appropriate paradigm of administrative constitutionalism and far more to do with the conceptual ambiguities implicit in merits review. The RI concept of the precautionary principle as involving a shifting of a burden of proof has been legitimate because it accords with an understanding of merits review as internal review modelled on adjudicative/adversarial procedure. DC concepts of the precaution274

Gray, above n 264 at para 138. Ibid at paras 133 and 139. 276 Telstra Corporation Ltd, above n 43 277 E Fisher, ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds) Environmental Law for Sustainability (Oxford, Hart Publishing, 2006). 275

(F) Fisher Ch4

31/5/07

09:58

Page 161

Conclusions

161

ary principle as enabling deliberative power have also been legitimate because they are consistent with traditional judicial review doctrine. From one perspective, this is an ironic state of affairs. Merits review was developed as a means of promoting better administration without the constraints of legal technicality. In relation to the precautionary principle, however, it has been the legal technicality which has had the most influence. The Australian experience with the precautionary principle has not had as high a profile as the issues explored in other case studies in this book. In many ways, however, it is the most interesting of these case studies, highlighting as it does how the thickness of legal culture shapes how a principle is interpreted. The major purpose of this chapter has been to show that understandings of the precautionary principle and administrative constitutionalism have been embedded in, and coproduced alongside, legal concepts, legal procedures, and legal doctrines. The principle is not a free-floating ‘duty to be cautious’ but rather a principle whose interpretation will be dependent on context. That context in Australia has been merits review. In showing this, I have highlighted three important points. First, the precautionary principle is a touchstone for debates about administrative constitutionalism and is capable of being interpreted in both DC and RI terms. This was discussed in the first chapter and will be returned to again in chapter six. Second, in thinking about the legal thickness of administrative constitutionalism, the focus should not only be on those principles which constitute and limit public administration but also on the rules and procedures which regulate those who hold public administration to account. This was hinted at in chapter three but is far more obvious here. Another example of this will be seen in the next chapter when I show how understandings about the role of the World Trade Organisation Dispute Settlement Panels have shaped debates over administrative constitutionalism in relation to the Sanitary and Phytosanitary Agreement. Finally, the operation of the precautionary principle in the Australian context has highlighted just how legally complex and ambiguous merits review is—a complexity and ambiguity that is only now, after nearly three decades, just beginning to be identified.278 Indeed, the analysis in this chapter raises as many questions for that form of legal pluralism as it does for thinking about the interface between technological risk and administrative constitutionalism. But again, there is nothing surprising in that as the raising of such questions is inherent in the symbiotic relationship between technological risk, public administration, and law.

278

Cane, above n 6; and Fleming, above n 83.

(F) Fisher Ch4

31/5/07

09:58

Page 162

(G) Fisher Intro to Part 2

31/5/07

09:58

Page 165

Introduction to Administrative Constitutionalism and Risk Regulation Beyond the State In the first part of this book my focus was upon how the interrelationship between administrative constitutionalism and technological risk regulation operates in a number of different national legal cultures. Chapters two, three, and four were each snapshots of how disputes or issues concerning technological risk regulation had been shaped by, and co-produced with, understandings of administrative constitutionalism. As a series of snapshots they have established at least two things. First, they have shown the utility of administrative constitutionalism as a lens through which to view disputes over technological risk regulation by showing how that lens provides a new perspective on three technological risk issues which were relatively well known in risk regulation scholarship. In each case, understandings of law, public administration, and technological risk have been shaped by theories of administrative constitutionalism. What that means is that such disputes are in essence debates over what is good public administration rather than debates over technical detail or whether scientific or democratic means are the better way to address technological risk issues. Second, and related to this, each of these case studies has highlighted the particular importance of legal culture. In each chapter, the way in which debates over administrative constitutionalism have manifested themselves have been very different, and so has been the role of law. Such a conclusion is particularly important for the growing body of scholarship on comparative risk regulation, which has failed to address this issue adequately.1 The reality is, however, that in the last two decades decision-making and debate about technological risk regulation have moved beyond the confines of the traditional constitutional state. Not only do supranational regimes such as the European Union (EU) have an ever-increasing role to play in technological risk regulation,2 but 1 J Wiener and M Rogers, ‘Comparing Precaution in the United States and Europe’ (2002) 5 Journal of Risk Research 317; J Hammitt et al, ‘Precautionary Regulation in Europe and the United States: A Quantitative Comparison’ (2005) 25 Risk Analysis 1215; and D Vogel, ‘Risk Regulation in Europe and the US’ (2004) 3 Yearbook of European Environmental Law 1. 2 E Vos, Institutional Frameworks of Community Health and Safety Legislation: Committees, Agencies and Private Bodies (Oxford, Hart Publishing, 1999).

(G) Fisher Intro to Part 2

166

31/5/07

09:58

Page 166

Introduction to Part Two

also the ability of states to regulate technological risks is increasingly being limited through trade law whether it be the trade rules of bodies like the EU or international frameworks such as that created by the World Trade Organisation(WTO). In light of these developments it could be argued that a focus on administrative constitutionalism is missing the point—in a globalising world the impact of understandings of good public administration is marginal. Moreover, in an era in which international organisations are having an increasingly important role to play in risk regulation then our focus should be on those new organisations rather than on outmoded national administrative structures. This is the era of governance and not government, and administrative constitutionalism seems to be in the realm of the latter. In this Part, I wish to show that the concept of administrative constitutionalism, far from being irrelevant, is of central importance, both in the interface between trade law and national technological risk regulation and in the creation of supranational regimes. I do so in two different ways. In chapter five I illustrate how the WTO Sanitary and Phytosanitary Agreement (SPS Agreement) is a document that regulates administrative constitutionalism and how different interpretations of ‘risk assessment’ in that document in the dispute settlement process have been shaped by the deliberative-constitutive (DC) and rational-instrumental (RI) paradigms of administrative constitutionalism. In chapter six, I examine how the precautionary principle legally operates within the EU context and show that there is a tension between needing to recognise that the precautionary principle is operating in many different contexts and the inevitable promotion of a uniform definition of the principle in an integrating polity. What can also be seen is that there has been a shift from the DC to the RI paradigm with the publication of the European Commission’s Communication on the Precautionary Principle.3 My concern in these two chapters is to establish the validity of administrative constitutionalism as a lens through which to view technological risk regulation disputes in the supranational and international trade law context. As such, each of these chapters is far more concerned with explaining the relevance of the lens than with exploring the implications of understanding disputes in these contexts in such terms. This is entirely on purpose and is for two different reasons. First, over the last decade debates about international trade law and supranational regimes have been dominated by discussions concerning market integration, democracy, and constitutionalism.4 Discussion about the role of administrative institutions is only now beginning to occur,5 but is hampered by a lack of empirical insight into these new institutions.6 Moreover, the challenge confronting researchers is not just one of 3 Commission of the European Communities, Communication from the Commission on the Precautionary Principle, COM (2000) 1 final. 4 C Joerges and E-U Petersmann (eds), Constitutionalism, Multilevel Trade Governance and Social Regulation (Oxford, Hart Publishing, 2006)l and C Joerges et al (eds), Mountain or MoleHill: A Critical Appraisal of the Commission White Paper on Governance (Florence, European University Institute, 2002). 5 B Kingsbury et al, ‘The Emergence of Global Administrative Law’ (2005) 68 Law and Contemporary Problems 63; and P Craig, EU Administrative Law (Oxford, Oxford University Press, 2006). 6 This is not to say that there has not been groundbreaking work. In the EU context see, in particular, Vos, above n 2.

(G) Fisher Intro to Part 2

31/5/07

09:58

Page 167

The WTO and EU: An Overview

167

unearthing institutional practices but also in deciding what is, and is not, relevant to analyse.7 This is made all the more difficult because what is in operation is not just new ‘public’ institutions but complex webs of interrelated public and private institutions that span across international, supranational, national, and sub-national legal cultures.8 In such circumstances, highlighting administrative constitutionalism is not so much about reorientation of an old debate but about providing a perspective on a new debate. Second, the consequence of highlighting administrative constitutionalism in these areas is akin to opening a Pandora’s box. While such a description is a hackneyed cliché it does reflect the fact that in both chapters five and six some very difficult questions are raised with no easy answers. In particular, what can be seen is that national, supranational, and transnational technological risk regulation regimes, as well as international trade law regimes, are a complex network of forums in which different but interlinked debates over administrative constitutionalism are occurring. As administrative constitutionalism is a form of legal culture then while these debates may be existing in networks they do not easily interact. Before launching into these two case studies it is useful to sketch the nature of these legal cultures in brief. I say legal cultures, rather than administrative constitutionalism, because as I have already highlighted, our understandings of these cultures is still embryonic. In particular, it is useful to highlight: the fundamental differences between each legal culture; the role of law in each legal culture; and the role of administrative institutions in each legal culture.

I The WTO and EU: An Overview In much contemporary discourse the WTO and EU systems are paired as examples of globalisation. Such a pairing suggests that there share much in common but the reality is that they are fundamentally different legal cultures. While both have their roots in post-World War II developments, each legal culture had a different starting point and has evolved in a fundamentally divergent way. The WTO trading system has evolved from the public international law framework of the General Agreement on Tariffs and Trade 1947.9 The system applied only to goods and among its many provisions were two basic principles—the principles of Most Favoured Nation and National Treatment—the combined effect amounting to a relatively ‘light touch’ approach to trade liberalisation as compared to the EU.10 The system was also accompanied with little formal 7

See many of the interesting essays in Joerges and Petersmann (eds), above n 4, exploring this issue. Ibid; B Morgan, ‘Turning Off the Tap: Urban Water Service Delivery and the Social Construction of Global Administrative Law’ (2006) 17 European Journal of International Law 215; and A-M Slaughter, A New World Order (Princeton, Princeton University Press, 2004). 9 For a history and overview see J Jackson, The Jurisprudence of GATT and the WTO (Cambridge, Cambridge University Press, 2000) at Part 2. 10 Ibid; and M Matsushita et al, The World Trade Organization: Law, Practice and Policy, 2nd edn (Oxford, Oxford University Press, 2006) at ch 1. 8

(G) Fisher Intro to Part 2

168

31/5/07

09:58

Page 168

Introduction to Part Two

institutional structure, and dispute settlement took a diplomatic rather than legal form. In particular, it worked on the basis of consensus.11 The WTO system itself only came into being in 1995 after several years of negotiation. It still had at its core the GATT but added a series of new Agreements which both strengthened the obligations under GATT and also broadened the scope of WTO jurisdiction.12 Among these Agreements was the SPS Agreement which is the subject of analysis in chapter five. Another important feature of the new WTO system was the creation of a new institutional structure including a General Council, Secretariat, Director General, and, most significantly, a system of dispute settlement. The nature and legitimacy of that system is increasingly open to debate but the reality is that it is still a system grounded in public international law and diplomacy.13 In contrast, the EU was created as a sui generis polity in 1958 pursuant to the Treaty of Rome, and right from the beginning involved aggressive negative and positive economic and regulatory integration aimed at ‘an ever closer union’.14 That process of integration has been achieved through a mixture of Treaty amendments and revision, the sophistication of EU law, dramatic deregulation and re-regulation, the creation of a network of national and community institutions, and limited political integration. What has resulted is a complex multi-level polity which, although defying easy explanation, is clearly involved in the governing of the EU in many different areas of economic, political, and social life. It is thus not surprising that focus in EU scholarship and practice is now upon EU constitutionalism.15 Central to the EU is a series of institutions—the Council, Commission, and Parliament—which operate on the principle of ‘institutional balance’ with each other and with Member States.16 Most significantly, the European Court of Justice (ECJ) has been an active and sophisticated engine of integration.17 To focus only on these institutions is to miss a major aspect of European integration, however. Not only does it ignore the significant role of Member States but also the range of committees that link Member States and Community institutions.18 Moreover, as 11 D Cass, The Constitutionalisation of the World Trade Organisation: Legitimacy, Democracy and Community in the International Trading System (Oxford, Oxford University Press, 2005) at 5–15. 12 Matsushita et al, above n 10 at 6–17. 13 Cass, above n 11; Joerges and Petersmann (eds), above n 4 at Part II; and J Pauwelyn, ‘The Transformation of World Trade’ (2005) 104 Michigan Law Review 1. 14 D Dinan, An Ever Closer Union: An Introduction to European Integration, 2nd edn (Basingstoke, Macmillan, 1999); and P Craig, ‘The Nature of the Community: Integration, Democracy and Legitimacy’ in P Craig and G de Búrca (eds) The Evolution of EU Law (Oxford, Oxford University Press, 1998). 15 J Weiler and M Wind (eds), European Constitutionalism Beyond the State (Cambridge, Cambridge University Press, 2003); and J Weiler, The Constitution of Europe: Do the New Clothes Have an Emperor? (Cambridge, Cambridge University Press, 1999). 16 K Lenaerts and A Verhoven, ‘Institutional Balance as a Guarantee for Democracy in EU Governance’ in C Joerges and R Dehousse (eds) Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002); and S Prechal, ‘Institutional Balance: A Fragile Principle With Uncertain Contents’ in T Heukels et al (eds), The European Union After Amsterdam (The Hague, Kluwer Law International, 1998). 17 M Maduro, We, The Court: The European Court of Justice and the European Economic Constitution (Oxford, Hart Publishing, 1998); and A Stone Sweet, The Judicial Construction of Europe (Oxford, Oxford University Press, 2004). 18 See Craig, above n 5 for a discussion of these frameworks.

(G) Fisher Intro to Part 2

31/5/07

09:58

Page 169

The Role of Law

169

will be seen in chapter six, those committees have a particularly important role to play in technological risk regulation.

II The Role of Law The differences between the WTO and the EU can be seen most starkly when one considers the role of law in each.19 It is not just the case that there is less law in the WTO than in the EU but that the nature of that law is fundamentally different in each jurisdiction. As noted above, part of the new WTO system was the creation of a dispute settlement system. The dispute settlement process under GATT began life as primarily a diplomatic process. It was called ‘conciliation’, those sitting on panels were unlikely to be lawyers, and any report of a panel had to be agreed to by all contracting parties.20 It was not until the Uruguay Round (which led to the creation of the WTO) and the setting-up of the Dispute Settlement Body on the basis on a negative consensus principle that the dispute settlement process evolved into something akin to a legal-type body of adjudication. The problem, however, is that this ‘thickening of legality’21 has not necessarily been accompanied by a clear vision of what is the nature and role of the WTO dispute settlement system. A key aspect of the dispute settlement system is its binding nature, but to state that what is produced by that system is law, as it is traditionally understood, would be to make too great a leap. The new system has two tiers—ad hoc Panels, which were created for specific disputes, and a permanent Appellate Body (AB)—both reporting to the Dispute Settlement Body. The ambiguity of their role can be seen in the fact that Panels and the AB, while almost akin to courts, do not produce ‘decisions’ but ‘reports’, which are then adopted by the Dispute Settlement Body. Panels are still ad hoc and modelled on a diplomatic model, while the AB takes more the guise of a judicial-type body.22 Both tend to adhere to basic public international law principles, and, in particular, those principles of literalism contained in the Vienna Convention on the Law of Treaties.23 There may be increasing consensus that dispute settlement, in producing interpretations of WTO law, is creating a ‘broad normative binding matrix’,24 but there is no agreement on what that means, and what the role of the Panels is in creating 19 A difference put nicely by S Dillion, ‘Observations on Trade Law and Globalisation’ (2005) 33 International Journal of Legal Information 103 at 109 20 D Palmeter and P Mavroidis, Dispute Settlement in the World Trade Organisation, 2nd edn (Cambridge, Cambridge University Press, 2004) at 7–9. 21 Lafer, as quoted in ibid at 303. 22 J Weiler, ‘The Rule of Lawyers and the Ethos of Diplomats: Reflections on the Internal and External Legitimacy of WTO Dispute Settlement’ (2002) 13 American Review of International Arbitration 177. 23 See chapter five, and D McRae, ‘What is the Future of WTO Dispute Settlement?’ (2004) 7 Journal of International Economic Law 3 at 5. 24 Weiler, above n 22 at 190.

(G) Fisher Intro to Part 2

170

31/5/07

09:58

Page 170

Introduction to Part Two

that matrix.25 To make matters more complicated, the frameworks that govern these institutions are not paradigms of logical drafting. Thus, for example, while the Dispute Settlement Understanding contains no explicit statement about the standard of review it does state that a Panel should ‘make an objective assessment of the matter before it, including an objective assessment of the facts’,26 and this has been taken by the AB as the starting point for understanding a Panel’s standard of review.27 The implications of this can be seen in chapter five. In contrast, in the EU the decisions of the ECJ amount to a complex jurisprudence which is breathtaking in its scope, nuance, and ambition. Not only has the Court developed a rich body of law in relation to the fundamental freedoms,28 but it has also developed a significant body of ‘general principles’.29 Moreover, over the last decade the Court of First Instance (CFI) has also had a significant role to play in developing a body of distinctly EU law. With that said, the development of this body of jurisprudence has not been straightforward. In particular, there have been two jurisprudential challenges. The first, which is yet to be resolved, is the intractable interrelationship of EU legal supremacy with the legal sovereignty of Member States.30 This question is made even more intractable due to the fact that the EU is not only a legal culture unto itself but also spans across the legal cultures of Member States. Views on supremacy and sovereignty differ dramatically depending on which legal culture is used as a starting point. This is particularly the case when a Member State’s constitution is in the mix.31 The second challenge faced by the ECJ is the development of a common body of jurisprudence in such a context, and what can be seen is that the Court has liberally borrowed and adapted concepts from a range of different legal cultures in quite innovative ways.32 This is particularly the case in relation to the development of general principles of Community law.33 With all that said, what can be seen in both the WTO and EU is a willingness of actors to bring their disputes to WTO dispute settlement or to the European 25 R Steinberg, ‘Judicial Lawmaking at the WTO: Discursive, Constitutional, and Political Constraints’ (2004) 98 American Journal of International Law 247. 26 Art 11, which describes the functions of panels. 27 Appellate Body Report, United States—Definitive Safeguard Measures on Import of Wheat Gluten from the European Communities, WT/DS166/AB/R, 19 January 2001, at para 175; and Appellate Body Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), WT/DS26/AB, 16 January 1998, at para 74. Also see M Oesch, Standards of Review in WTO Dispute Resolution (Oxford, Oxford University Press, 2003) at 83–8. 28 C Barnard, The Substantive Law of the EU: The Four Freedoms (Oxford, Oxford University Press, 2004). 29 T Tridimas, The General Principles of European Community Law, 2nd edn (Oxford, Oxford University Press, 2006). 30 Brunner v The European Union Treaty [1994] 1 CMLR 57; and N MacCormick, Questioning Sovereignty (Oxford, Oxford University Press, 1999). 31 See the interesting discussion on this point in relation to the inclusion of the precautionary principle in the French constitution in O Godard, ‘The Precautionary Principle and Catastrophism on Tenterhooks: Lessons From Constitutional Reform in France’ in E Fisher et al (eds), Applying the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006) at 80. 32 K Lenaerts, ‘Interlocking Legal Orders in the European Union and Comparative Law’ (2003) 52 International and Comparative Law Quarterly 873. 33 Tridimas, above n 29.

(G) Fisher Intro to Part 2

31/5/07

09:58

Page 171

Governance and Administrative Constitutionalism

171

courts. In relation to the WTO, this was particularly the case in the early years, and many have argued that it is proof that the system is effective at resolving disputes.34 In the EU, the growth of litigation arguably reflects the growth of adversarial legalism in the EU,35 and the growing desire among legal actors to hold Member States and Community institutions to account in their exercise of power pursuant to EU risk regulation regimes.36

III Governance and Administrative Constitutionalism, The final issue to note in relation to these two legal systems is the nature of governance and administrative constitutionalism in each. What can be seen is that in both systems there is a vibrant debate over governance, with a particular focus on constitutionalism and democratic legitimacy.37 For my purposes, my main focus is on these issues in the EU. As I shall show in chapter five, administrative constitutionalism does have a role to play in interpreting the SPS Agreement but one that flows more out of the nature of that Agreement than from the governance arrangements of the WTO. As already noted, the governance arrangements of the EU are recognised to be sui generis. They have also been largely administrative in nature, although it should be stressed that there is no conventional understanding of the separation of powers operating in the EU. The governance arrangements are administrative because in most cases the power of the Community institutions has not been directly derived from a democratic source but rather has been delegated to them from primary lawmakers. The nature of Community ‘administration’ is open to question, however. Thus there are those commentators who tend to understand all institutions at the Community level in administrative terms38 and those that focus mainly on the committees and agencies which have developed in the crevices between Community institutions and the Member States.39 What this distinction highlights is that within the Community there is a very rough distinction between a ‘meta’ form of delegation from the Member States to Community institutions40 and a more traditional 34 W Davey, ‘The WTO Dispute Settlement System: The First Ten Years’ (2005) 8 Journal of International Economic Law 17; and K Leitner and S Lester, ‘WTO Dispute Settlement 1995–2004: A Statistical Analysis’ (2005) 8 Journal of International Economic Law 231. 35 D Kelemen, ‘Suing for Europe: Adversarial Legalism and European Governance’ (2006) 39 Comparative Political Studies 101. 36 C Harlow, Accountability in the European Union (Oxford, Oxford University Press, 2002) at ch 6. 37 Cass, above n 11; and Weiler and Wind (eds), above n 15. 38 P Lindseth, ‘Democratic Legitimacy and the Administrative Character of Supranationalism: The Example of the European Community’ (1999) 99 Columbia Law Review 628. 39 E Chiti, ‘The Emergence of a Community Administration: The Case of European Agencies’ (2000) 37 Common Market Law Review 309; and European Institute of Public Administration, Governance by Committee: The Role of Committees in European Policy Making and Policy, Research Paper 00/GHA (Maastricht, European Institute of Public Administration, 2000). 40 Art 7 TEC.

(G) Fisher Intro to Part 2

172

31/5/07

09:58

Page 172

Introduction to Part Two

form of delegation from those institutions to committees and agencies.41 I say rough, because the reality is that there is no clear legal line that can be drawn between the two, and the matter is made even more complicated by the fact that these different institutions are closely interrelated with Member State institutions. The end result of all of this is that much of the recent debate about EC governance has been about its exact nature, and many different concepts have been coined to describe the EU governance system. Thus, for example, academics talk of ‘deliberative supranationalism’,42 ‘regulatory space’,43 and the EU ‘regulatory state’.44 What is interesting about much of this dialogue is that it translates the descriptive language of governance into a prescriptive theory of normative legitimacy.45 Moreover, the ambiguous nature of Community institutions is omnipresent. Thus, for example, even after a decade of debate, the role of comitology committees is not clear.46 What is clear is that there is an increasing focus on accountability within the EU.47 This has not only resulted in an increasing policy focus on the accountability of Community institutions48 but also in the rapid emergence of a body of EU administrative law.49 That body of case-law is an increasingly complex one in which the CFI has had a significant role to play. As I shall show in chapter six, the elaboration of the precautionary principle has been an important part of these developments.

IV Conclusion My sketches above are not meant to be exhaustive examinations of the WTO and EU legal systems. What they do make clear, however, is that these two legal cultures are fundamentally different not just in purpose but in the role of governance and administrative institutions within them. What I will show in the next two chapters, however, is that despite this being the case, administrative constitutionalism discourses are operating in both systems. 41 This form of delegation is more strictly policed. See Case 9/56 Meroni v ECSC High Authority [1957–8] ECR 133. 42 C Joerges and J Neyer, ‘From Intergovernmental Bargains to Deliberative Political Processes: The Constitutionalisation of Comitology’ (1997) 3 European Law Journal 273; and C Joerges, ‘Deliberative Supranationalism: Two Defences’ (2002) 8 European Law Journal 133. 43 C Scott, ‘Analysing Regulatory Space: Fragmented Resources and Institutional Design’ (2001) Public Law 329. 44 G Majone, Regulating Europe (London, Routledge, 1996). 45 E Fisher, ‘Unpacking the Toolbox: Or Why the Public/Private Divide is Important in EC Environmental Law’ in M Freedland and J-B Auby (eds) The Public Law/Private Law Divide: Une entente assez cordiale? (Oxford, Hart Publishing, 2006). 46 E Fisher, ‘The European Union in the Age of Accountability’ (2004) 24 Oxford Journal of Legal Studies 495 at 512. 47 Ibid; and Harlow, above n 36. 48 Commission of the European Communities, European Governance: A White Paper COM(2001) 428 final. 49 Craig, above n 5

(H) Fisher Ch5

31/5/07

09:58

Page 173

5 Risk Assessment, The World Trade Organisation Sanitary and Phytosanitary Agreement, and Administrative Constitutionalism* Over the last decade the science/democracy dichotomy has dominated discussion about the interface between national risk regulation standards and international trade law. Nowhere is this more the case than with debates about the relationship between national standards protecting the life and health of animals, plants, and humans and the World Trade Organisation’s (WTO) Sanitary and Phytosanitary Agreement (SPS Agreement). In terms of the science/democracy dichotomy, national risk regulation standards are the product of a sovereign nation’s democratic process and, as such, reflect the values that that nation’s citizenry espouses. In contrast, the SPS Agreement, with its obligation that a national SPS measure is to ‘be based on scientific principles’ and ‘not maintained without adequate scientific evidence’,1 is understood as requiring standards to be based on objective scientific evidence. In this chapter, I argue two things. First, I argue that this characterisation is wrong because the SPS Agreement is a document concerned with administrative constitutionalism and can be interpreted in terms of both the deliberativeconstitutive (DC) and rational-instrumental (RI) paradigms, and has been interpreted in both terms by Panels and the Appellate Body (AB) as part of the dispute settlement process. Second, I argue that the present characterisation of the SPS Agreement in terms of the science/democracy dichotomy is also dangerously unhelpful because it has resulted in disputes concerning the SPS Agreement being understood solely in scientific terms. This has resulted in the inadvertent promotion of the RI paradigm and the dispute settlement process becoming problematic due to incoherent decisions, dispute settlement becoming an exercise in * An earlier and shorter version of this chapter can be found in E Fisher, ‘Beyond the Science/Democracy Dichotomy: The World Trade Organisation Sanitary and Phytosanitary Agreement and Administrative Constitutionalism’ in C Joerges and E-U Petersmann (eds), Transnational Trade Governance and Social Regulation: Tensions and Interdependencies (Oxford, Hart Publishing, 2006). 1 Art 2.2.

(H) Fisher Ch5

31/5/07

174

09:58

Page 174

Risk Assessment, The WTO and Administrative Constitutionalism

proving the evidentiary basis of standards, the subsequent strains on the institutional framework, and the failure of commentators to see the close relationship between national risk regulation and transnational trade regulation. The structure of this chapter is as follows. In the first section, I give an overview of the SPS Agreement and how it has been understood in terms of the science/democracy dichotomy. In the second section, I show how the Agreement is, in actual fact, a document concerned with administrative constitutionalism, albeit one that raises issues that are distinct from those issues raised in national legal cultures. These differences include: that there are no theories of administrative constitutionalism operational in the WTO context; that such theories will be shaped by perceptions about the trade law purposes of the SPS Agreement; that dispute settlement is characterised in RI terms; and that there has been a failure to recognise fully the normative nature of dispute settlement. In the third section, I analyse the Panel’s and AB’s decisions in the first dispute concerning the SPS Agreement, EC—Hormones,2 to show how each of these rulings was shaped by the RI and DC paradigms, respectively. In the fourth section, I consider the four other disputes concerning the SPS Agreement and show that the development of the WTO jurisprudence in this area has inadvertently promoted the RI paradigm of administrative constitutionalism. I say inadvertently because it is not obvious that the promotion of the RI paradigm is a product of conscious choice. In the fifth section, I consider the problems with this present body of jurisprudence. In the final section, I reflect on this state of affairs and suggest two future lines of inquiry—one concerned with thinking more carefully about the WTO regime, and the other to do with exploring the interrelationship between the WTO regime and national risk regulation regimes. A number of clarifications should be made before proceeding. First, as always, the DC and RI paradigms are being used as shorthand to describe the different normative assumptions underlying different legal and policy positions. The paradigms have never been referred to explicitly by actors but are representations of the normative presumptions that those actors hold. Second, this chapter is not directly concerned with the legitimacy and constitutionalisation of the WTO itself.3 This chapter’s agenda is far less ambitious. It is simply to get the right questions being asked by scholars about the interface between trade and social regulation. Third, this chapter is not an exhaustive analysis of Panel and AB decisions concerning the SPS Agreement.4 Rather, I am just focusing on one legal issue it 2 Panel Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), WT/DS26/R/USA, 18 August 1997; and Appellate Body Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), WT/DS26/AB, 16 January 1998. 3 D Cass, The Constitutionalisation of the World Trade Organisation: Legitimacy, Democracy, and Community in the International Trading System (Oxford, Oxford University Press, 2005); and C Joerges and E-U Petersmann (eds), Constitutionalism, Multilevel Trade Governance and Social Regulation (Oxford, Oxford University Press, 2006). 4 For such examinations see J Paulwelyn, ‘The WTO Agreement on Sanitary and PhytoSanitary (SPS) Measures as Applied in the First Three Disputes: EC—Hormones, Australia—Salmon and Japan—Varietals’ (1999) 2 Journal of International Economic Law 641; and D. Victor, ‘The Sanitary and Phytosanitary Agreement of the World Trade Organisation: An Assessment After Five Years’ (2000) 32 New York University Journal of International Law and Politics 865.

(H) Fisher Ch5

31/5/07

09:58

Page 175

The WTO SPS Agreement

175

has raised to illustrate how the SPS Agreement is directly concerned with administrative constitutionalism. In doing this, I appreciate that my analysis may seem too narrowly circumscribed but it should be remembered that the purpose of this chapter is an exercise in reorientation rather than comprehensive analysis.5

I The WTO SPS Agreement The SPS Agreement was developed as part of the Uruguay Round and came into force in 1994 along with the associated but less substantively stringent Technical Barriers to Trade Agreement (TBT).6 The SPS Agreement applies to ‘measures’ that are applied to protect animal, plant, and human life and health from a variety of different risks ranging from the spread of pests to additives in food.7 As such, the Agreement regulates technological risk evaluation. The Agreement entrenches the right of Member States to act so as to protect human, animal and/or plant health,8 but regulates the process by which such standards can be set. By process, I mean a concept distinct from procedure,9 in that the Agreement is regulating both the way SPS measures are set as well as the basis on which they are set. The Agreement thus includes provisions relating to the basis for decisions,10 what can be taken into account,11 decision-making transparency,12 who should be consulted,13 and how trade-restrictive standards may be.14 While it does not require Member States to carry out all these processes themselves, a Member State needs to show that these different processes have occurred.15 In other words, the Agreement defines, in some detail, what is an ‘acceptable’ risk regulation standard and, as such, is directly concerned with the ‘reasonableness’ of standard setting. It thus is a considerable departure from the more modest trade liberalisation objectives of the General Agreement on Tariffs and Trade (GATT) and has not surprisingly been the subject of considerable academic comment.16 5 In particular, it should be noted that the emerging jurisprudence on Art 5.7 also deserves analysis in terms of administrative constitutionalism. 6 Note that an earlier version of the TBT Agreement was developed as part of the Tokyo Round. 7 Annex A.1. 8 A right also evidenced in Art XX(b) of the General Agreement on Tariffs and Trade (GATT). 9 Process should be distinguished from procedure. On the distinction see J Jowell, ‘Of Vires or Vacuums: The Constitutional Context of Judicial Review’ in C Forsyth (ed), Judicial Review and the Constitution (Oxford, Hart Publishing, 2000) at 332–3. 10 Arts 2.2, 3.2, and 5.1 11 Arts 5.2 and 5.3. 12 Art 7 and Annex B 13 Arts 3.4, 9, and 12. 14 Arts 5.4, 5.5, and 5.6. 15 Appellate Body Report, above n 2 at para 190; and Panel Report, European Communities— Measures Affecting the Approval and Marketing of Biotech Products, WT/DS291-3/R, 29 September 2006 at para 7.3024. 16 D Roberts, ‘Sanitary and Phytosanitary Risk Management in the Post-Uruguay Round Era: An Economic Perspective’ in National Research Council (ed), Incorporating Science, Economics, and Sociology in Developing Sanitary and Phytosanitary Standards in International Trade: Proceedings of a

(H) Fisher Ch5

31/5/07

176

09:58

Page 176

Risk Assessment, The WTO and Administrative Constitutionalism

The Agreement contains a number of ‘basic rights and obligations’.17 The two most relevant for my purpose are Articles 2.1 and 2.2: 1. Members have the right to take sanitary and phytosanitary measures necessary for the protection of human, animal or plant life or health, provided that such measures are not inconsistent with the provisions of this Agreement. 2. Members shall ensure that any sanitary or phytosanitary measure is applied only to the extent necessary to protect human, animal or plant life or health, is based on scientific principles and is not maintained without sufficient scientific evidence, except as provided for in paragraph 7 of Article 5.

The Agreement also strongly encourages Member States to base their national SPS measures on international standards. Article 3 states: 1. To harmonize sanitary and phytosanitary measures on as wide a basis as possible, Members shall base their sanitary or phytosanitary measures on international standards, guidelines or recommendations, where they exist, except as otherwise provided for in this Agreement, and in particular in paragraph 3. 2. Sanitary or phytosanitary measures which conform to international standards, guidelines or recommendations shall be deemed to be necessary to protect human, animal or plant life or health, and presumed to be consistent with the relevant provisions of this Agreement and of GATT 1994. 3. Members may introduce or maintain sanitary or phytosanitary measures which result in a higher level of sanitary or phytosanitary protection than would be achieved by measures based on the relevant international standards, guidelines or recommendations, if there is a scientific justification, or as a consequence of the level of sanitary or phytosanitary protection a Member determines to be appropriate in accordance with the relevant provisions of paragraphs 1 through 8 of Article 5.

It is not clear what is the nature of these obligations. Some have interpreted Article 3.1 in mandatory terms,18 while others have argued it is more voluntary.19 The most relevant international standard-setting body is the Codex Alimentarius Commission (Codex), an international organisation that sets food safety standards.20 Codex is essentially an administrative body that raises its own legitimacy concerns but it is not of direct concern in this chapter.21 Conference (Washington DC, National Academies Press, 2000); C Button, The Power to Protect: Trade, Health and Uncertainty in the WTO (Oxford, Hart Publishing, 2004); V Walker, ‘Keeping the WTO From Becoming a World Trans-scientific Organisation’ (1998) 31 Cornell International Law Journal 251; and G Sampson, ‘Risk and the WTO’ in D Robertson and A Kellow (eds), Globalisation and the Environment: Risk Assessment and the WTO (Cheltenham, Edward Elgar, 2001). 17 Art 2. 18 M Matsushita et al, The World Trade Organization: Law, Practice and Policy, 2nd edn (Oxford, Oxford University Press, 2006) at 503. 19 Appellate Body Report, above n 2 at paras 165–169. 20 J Braithwaite and P Drahos, Global Business Regulation (Cambridge, Cambridge University Press, 2000) at ch 16. 21 M Livermore, ‘Authority and Legitimacy in Global Governance: Deliberation, Institutional Differentiation, and the Codex Alimentarius’ (2006) 81 New York University Law Review 766; A Herwig, ‘Transnational Governance Regimes for Foods Derived from Bio-Technology and their Legitimacy’ in C Joerges et al (eds), Transnational Governance and Constitutionalism (Oxford, Hart Publishing, 2004); and T Huller and M Maier, ‘Fixing the Codex? Global Food Safety Governance

(H) Fisher Ch5

31/5/07

09:58

Page 177

The WTO SPS Agreement

177

The process for setting SPS measures is set out in Article 5 of the Agreement. Article 5.1 states that: Members shall ensure that their sanitary or phytosanitary measures are based on an assessment, as appropriate to the circumstances, of the risks to human, animal or plant life or health, taking into account risk assessment techniques developed by the relevant international organizations.

How this Article has been interpreted in dispute settlement is the focus of this chapter. It is has been universally recognised that there is a close relationship between Article 5.1 and Article 2.2, but the nature of that relationship is by no means clear.22 It is also important to remember that the concept of risk assessment can refer to many different types of processes.23 Moreover, as also seen in other chapters in this book, it is a term which is shaped by understandings of administrative constitutionalism.24 The SPS Agreement does define the concept of risk assessment further. In Annex A.4, risk assessment is defined as [t]he evaluation of the likelihood of entry, establishment or spread of a pest or disease within the territory of an importing Member according to the sanitary or phytosanitary measures which might be applied, and of the associated potential biological and economic consequences; or the evaluation of the potential for adverse effects on human or animal health arising from the presence of additives, contaminants, toxins or diseasecausing organisms in food, beverages or feedstuffs. [my emphasis]

This is both a broad and dual definition of risk assessment in that not only is it open to interpretation but it contains two distinct definitions of risk assessment. The first definition relates to risk assessment in a quarantine context and to decisions about whether certain products should be allowed to enter into a jurisdiction, and the second definition relates to the assessment of health risks from products. Each definition is requiring a different process so that the first requires an evaluation of likelihood and the second an evaluation of potential. As well as these two provisions, Articles 5.2 and 5.3 state what should be taken into account in carrying out a risk assessment: 2. In the assessment of risks, Members shall take into account available scientific evidence; relevant processes and production methods; relevant inspection, sampling and testing methods; prevalence of specific diseases or pests; existence of pest- or disease-free areas; relevant ecological and environmental conditions; and quarantine or other treatment. Under Review’ in C Joerges and E-U Petersmann (eds), Constitutionalism, Multi-Level Trade Governance and Social Regulation (Oxford, Hart Publishing, 2006). 22 See sections III and IV below. 23 E Fisher, ‘The Rise of the Risk Commonwealth and the Challenge for Administrative Law’ (2003) Public Law 455. 24 See, in particular, chs 2 and 3. See also E Fisher, ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds), Environmental Law for Sustainability (Oxford, Hart Publishing, 2006).

(H) Fisher Ch5

31/5/07

178

09:58

Page 178

Risk Assessment, The WTO and Administrative Constitutionalism

3. In assessing the risk to animal or plant life or health and determining the measure to be applied for achieving the appropriate level of sanitary or phytosanitary protection from such risk, Members shall take into account as relevant economic factors: the potential damage in terms of loss of production or sales in the event of the entry, establishment or spread of a pest or disease; the costs of control or eradication in the territory of the importing Member; and the relative cost-effectiveness of alternative approaches to limiting risks.

These two provisions are requiring, and allowing, many different things to be taken into account in risk assessment. Risk assessment thus requires broadranging analysis. The rest of Article 5 regulates other aspects of the process of setting measures, including that Member States should: take into account ‘the objective of minimizing negative trade effects’;25 that they ‘shall avoid arbitrary or unjustifiable distinctions in the levels it considers to be appropriate in different situations, if such distinctions result in discrimination or a disguised restriction on international trade’;26 and they should take the least trade-restrictive measures necessary.27 Article 5.7 also allows for temporary measures to be put in place in circumstances where there is ‘insufficient scientific evidence’. From the outset the obligations under the SPS Agreement were perceived to be lacking in clarity.28 In part, this imprecision can be understood as a product of political compromise and thoughtless drafting, particularly as so many of the problems in interpreting the Agreement lie in the unclear relationship between its different provisions.29 Thus, for example, the relationship between Articles 2.2 and 5.1 is not clear30 and nor is that between Articles 3.1 and 3.3.31 Thus, while the Agreement is generally understood as striking a balance between Member State autonomy and the regulation of standard setting in this area, the nature of this balance is not obvious.32 Much of this lack of clarity can be understood to derive from conflict over the purpose of the Agreement. Indeed, the Agreement can be seen to have two not always mutually consistent objectives. First, SPS measures can be seen as needing special regulation under the WTO regime because, historically, Member States had used SPS measures as covert vehicles for trade protectionism.33 From this per25

Art 5.4. Art 5.5. 27 Art 5.6. 28 T Christoforou, ‘Settlement of Science-based Trade Disputes in the WTO: A Critical Review of the Developing Case Law in the Face of Scientific Uncertainty’ (2000) 8 New York University Environmental Law Journal 622 at 625–6. 29 D Wirth, ‘The Role of Science in the Uruguay Round and NAFTA Trade Disciplines’ (1994) 27 Cornell International Law Journal 817. 30 Appellate Body Report, above n 2 at para 180; Appellate Body Report, Australia—Measures Affecting Importation of Salmon, WT/DS18/AB/R, 20 October 1998 at paras 130–38; and Appellate Body Report, Japan—Measures Affecting Agricultural Products, WT/DS76/AB/R, 22 February 1999 at paras 109–13. 31 Appellate Body Report, above n 2 at paras 158–71. 32 Button, above n 16 at 43. 33 W Maruyama, ‘A New Pillar of the WTO: Sound Science’ (1998) 32 International Lawyer 651; Roberts, above n 16 at 35; and T Weiler, ‘International Regulatory Reform Obligations’ (2000) 34 Journal of World Trade Law 71. 26

(H) Fisher Ch5

31/5/07

09:58

Page 179

The WTO SPS Agreement

179

spective, regulating SPS measures is an extension of the discrimination principles, and the Agreement is a fleshing out of the Article XX(b) exception included in the GATT.34 The second purpose of the Agreement can be seen as reducing regulatory heterogeneity which can act as a form of trade barrier, due to the fact that different regulatory standards in different states cause extra costs to importers.35 The abandonment of tariffs has made such non-tariff trade barriers more obvious, and SPS standards have had a particularly high political profile in this regard.36 From this perspective, the Agreement is a form of ‘policed decentralisation’ in that it addresses the regulatory heterogeneity problem by creating common rules for national SPS standard setting.37 On this interpretation, Article 3.1 takes the form of a compulsory obligation because Member States adopting international standards is the best way to reduce regulatory heterogeneity. Due to the ambiguous nature of the Agreement’s obligations and the unclear relationship between those obligations, the Agreement is open to interpretations informed by either purpose.38 As noted in the introduction to this chapter, most writing on the SPS Agreement has tended to characterise it in terms of the science/democracy dichotomy, where science is the logic of trade and democracy is the logic of national risk regulation regimes. Thus, while commentators have attributed to the Agreement divergent purposes, there has been a general consensus that the SPS Agreement is about promoting a ‘scientific’ approach to standard setting.39 The Agreement is described as: requiring reliance on ‘sound science’;40 creating an objective system;41 and giving to science the ‘key role’ in determining whether a standard is legitimate.42 This is not surprising because the Agreement makes clear that an SPS measure must be ‘based on scientific principles’,43 and if Member States wish to introduce or maintain measures which result in a ‘higher level’ of protection than international standards they can only do so if there is a ‘scientific justification’.44 A corollary of the SPS Agreement being understood as requiring a scientific approach to SPS standard setting is that it is also assumed that Member States would carry out standard setting by more democratic or political means if 34 Preamble SPS Agreement. See J Trachtman, ‘International Trade as a Vector in Domestic Regulatory Reform: Discrimination, Cost-Benefit Analysis, and Negotiations’ (2000) 24 Fordham International Law Journal 726 for a discussion on how the national treatment principle can be extended into a discussion of the rationality of national regulation. 35 A Sykes, ‘The (Limited) Role of Regulatory Harmonisation in International Goods and Services Markets’ (1999) 2 Journal of International Economic Law 49. 36 M Trebilcock and R Howse, ‘Trade Liberalization and Regulatory Diversity: Reconciling Competitive Markets with Competitive Politics’ (1998) 6 European Journal of Law & Economics 5 at 6. 37 Sykes, above n 35 at 61–5. 38 See section III below. 39 Button, above n 16 at 44. 40 Maruyama, above n 33. 41 K. Ambrose, ‘Science and the WTO’’ (2000) 31 Law and Policy in International Business 861. 42 O Perez, Ecological Sensitivity and Global Legal Pluralism (Oxford, Hart Publishing, 2004) at 117. 43 Art 2.2. 44 Art 3.2.

(H) Fisher Ch5

31/5/07

180

09:58

Page 180

Risk Assessment, The WTO and Administrative Constitutionalism

they were not bound by the Agreement.45 Moreover, whether science or democracy is a better basis for risk evaluation depends on a scholar’s starting assumptions.46 If national regulatory intervention is largely understood as another mode of trade protectionism then the SPS Agreement and the objectivity it ‘requires’ is desirable.47 In contrast, if one sees the objectivity of science as a mirage and/or SPS measures as part of the democratic responsibility of a state then the SPS Agreement is problematic.48 There are some scholars who take a more sophisticated view of the dichotomy in that they see science and democracy working in tandem. Even for them, however, the dichotomy is still an important one.49

II The SPS Agreement Through the Lens of Administrative Constitutionalism There are two significant problems with understanding the SPS Agreement in terms of the science/democracy dichotomy. The first is that such a characterisation is wrong and the second is that such a characterisation is unhelpful. In either case, the problems are to do with the fact that there has been a failure to understand that the SPS Agreement is a document directly concerned with administrative constitutionalism. At first sight, this relationship is not obvious due to the fact that the Agreement’s purpose is with preventing unacceptable barriers to trade rather than with the legitimacy of public administration. It is that focus on ‘unacceptability’ that provides the link between it and administrative constitutionalism, however. As already noted, what is an ‘unacceptable’ SPS measure is ambiguous, and the term could refer either to a disguised discriminatory measure or to any SPS measure which is inadvertently a trade barrier. The trade law purpose of the Agreement cannot be the only reference point for determining the acceptability of SPS measure, however. That is because, beyond stating that either trade protectionism or regulatory variation are undesirable, ideals of trade integration contain no substantive guidance about the nature of standard setting. Rather, in providing a framework for SPS standard setting there must be some starting assumptions 45 Remembering that in terms of the science/democracy dichotomy these two concepts are understood as interchangeable. See section I.B of ch 1 above; A Guzman, ‘Food Fears: Health and Safety at the WTO’ (2004) 45 Virginia Law Review 1; and J Bohanes, ‘Risk Regulation in WTO Law: A Procedure-Based Approach to the Precautionary Principle’ (2002) 40 Columbia Journal of Transnational Law 323. 46 See A Sykes, ‘Regulatory Protectionism and the Law of International Trade’ (1999) 66 University of Chicago Law Review 1, and Trebilcock and Howse, above n 36 for a discussion of these assumptions. 47 Sampson, above n 16. 48 V Walker, ‘The Myth of Science as a “Neutral Arbiter” for Triggering Precautions’ (2003) 26 Boston College International and Comparative Law Review 197. 49 R Howse, ‘Democracy, Science, and Free Trade: Risk Regulation on Trial at the WTO’ (2000) 98 Michigan Law Review 2329; and D Winickoff et al, ‘Adjudicating the GM Food Wars: Science, Risk and Democracy in World Trade Law’ (2005) 30 Yale Journal of International Law 81.

(H) Fisher Ch5

31/5/07

09:58

Page 181

The SPS Agreement Through the Lens of Administrative Constitutionalism 181 about what type of activity standard setting is, and those assumptions must inevitably derive from the actual practice of standard setting. As shown in the rest of this book, that practice has been overwhelmingly carried out by public administration. As this is the case, then the Agreement is, due to its subject matter, a document concerned with administrative constitutionalism, regulating as it does the reasonableness of the standard-setting process. It is not just the case that the SPS Agreement must inevitably be concerned with administrative constitutionalism, however—its framework also rests on an assumption that SPS standard setting is an activity of public administration. First, the Agreement regulates SPS standard setting which is carried on by a state. This is inevitable given its public international law nature,50 but in doing so the Agreement presupposes and entrenches the right of states to act in this area.51 Second, the Agreement is based on an implicit assumption that standard setting in this area is administrative in nature. There is no mention of democracy, legislatures, or democratic rights throughout the Agreement, and if any references are made to institutions they are to different levels of governments and regulatory bodies.52 More significantly, the Agreement in creating a framework for SPS standard setting regulates the four different aspects of standard setting discussed in chapter one—the application of a regulatory prescription to a specific issue, information, expertise, and communication.53 Thus there are provisions concerned with the role of normative prescriptions54 and ensuring that they are ‘appropriate’ in any particular circumstance.55 There are a considerable number of provisions regulating the type of information on which standards can be set.56 Implicit in these provisions is the utilisation of a range of different expertises57 including the expertise of international bodies58 and national regulatory bodies.59 The Agreement also includes a number of provisions regulating communication and transparency,60 particularly concerning consultations between Member States.61 Finally, the Agreement requires that Member States must base their SPS measures on a ‘risk assessment’,62 50

Although note Art 13. Art 2.1. 52 Arts 9.1 and 13. 53 See section II.A of ch 1. 54 Art 2.1 and the Preamble, although note that the concept of ‘level’ of protection (see Arts 3.3 and 5.6) suggests that the issue of normative prescription is linear when, as seen in the rest of this book, it is multidimensional. Also note that other provisions of the Agreement will limit the discretion of a Member State in relation to how a normative prescription applies. See Art 2.2 (extent necessary), Art 5.4 (need to take into account minimising negative trade effects in determining the ‘appropriate level of protection’), and Annex A.5. 55 Arts 5.4 and 5.6. 56 Art 3.1 (information for international bodies); Art 2.2 and Art 3.3 (scientific principles and scientific justification); Arts 5.1–5.3 and Annex A.4 (a range of information), 57 See, in particular, the provisions of Arts 5.2 and 5.3. 58 Art 3.1 and Annex A.3. 59 Art 9. 60 Art 7 and Annex B. 61 Arts 3.4, 9 and 12. 62 Art 5.1. 51

(H) Fisher Ch5

31/5/07

182

09:58

Page 182

Risk Assessment, The WTO and Administrative Constitutionalism

an activity which is not only inevitably administrative in nature but an institutional construct of administrative constitutionalism.63 The Agreement is thus not, as many commentators describe it, laying down requirements just in relation to the scientific aspects of SPS standard setting, but for the whole standard-setting process which is understood to be administrative. With that said, the Agreement does put significant weight on SPS measures being based on scientific principles, scientific justification, and scientific evidence,64 but that emphasis is in the context of the wider understanding of the standard-setting process seen above. Moreover, as stressed in chapter one, all risk evaluation has a scientific basis—the important question is how that basis is defined. A consequence of the SPS Agreement regulating the reasonableness of SPS standard setting, and such standard setting being administrative, is that the Agreement is a document of administrative constitutionalism which is concerned with constituting, limiting, and holding standard setters to account. For the Agreement to operate in practice there must be reference to understandings of administrative constitutionalism. Interpretations and applications of it must rest on a theory of good public administration. Likewise, the operation of the Agreement will also result in the promotion of particular paradigms of administrative constitutionalism. This means that the Agreement is a new influence shaping technological risk regulation, and dispute settlement concerning the Agreement is a new site for debating administrative constitutionalism. As a document of administrative constitutionalism, the SPS Agreement raises four issues which are very different from those seen in the national settings. First, within WTO legal culture there is no entrenched and long-running discourse over administrative constitutionalism as seen in national legal orders. There is the Codex Alimentarius, and associated international organisations, but these are only beginning to give rise to such a discourse.65 Moreover, as noted in the introduction to this Part, WTO legal culture is still emerging. What this means is that theories of administrative constitutionalism must be derived from other legal cultures, primarily those of national Members. The second point is that it is not only understandings of public administration which will influence understandings of administrative constitutionalism. Understandings about the trade liberalisation purposes of the Agreement also have a role to play. Indeed, the two possible purposes of the SPS Agreement seen above result in two different paradigms of administrative constitutionalism being promoted. If the purpose of the Agreement is understood to be to stop disguised protectionism then in understanding what is reasonable administrative action the focus is upon preventing invalid exercises of discretion rather than upon preventing any exercise of discretion. Indeed, this understanding of the Agreement often 63

Fisher, above n 24. Arts 2.2 and 3.3. 65 See references, above n 21. For the connection between administrative constitutionalism discourses concerning supranational organisations and administrative constitutionalism discourses concerning trade law see ch 6 below. 64

(H) Fisher Ch5

31/5/07

09:58

Page 183

The SPS Agreement Through the Lens of Administrative Constitutionalism 183 flows from an appreciation that standard setting is highly discretionary due to scientific and behavioural uncertainty as well as the socio-political nature of these disputes.66 As such, the DC paradigm of administrative constitutionalism tends to be promoted when the Agreement is understood in these terms, and the acceptability of decision-making is judged by the quality of deliberative problem solving.67 In contrast, if the Agreement is understood as an attempt to reduce regulatory heterogeneity between Member States then reasonable standard setting is that which is the most consistent with standard setting in other jurisdictions. Such consistency is best achieved through limiting administrative discretion as much as possible through clear rules and analytical methods. Moreover, such an understanding of standard setting makes sense only if risks are understood to be manageable. As such, this understanding of the SPS Agreement promotes an RI understanding of standard setting. The third issue that means that thinking about administrative constitutionalism and the SPS Agreement is very different from the type of considerations in national settings is that the WTO dispute settlement process, and in particular the role of Panels, promotes the RI paradigm. Under the Dispute Settlement Understanding (DSU), the function of a Panel is described in the following terms: The function of panels is to assist the DSB [Dispute Settlement Body] in discharging its responsibilities under this Understanding and the covered agreements. Accordingly, a panel should make an objective assessment of the matter before it, including an objective assessment of the facts of the case and the applicability of and conformity with the relevant covered agreements, and make such other findings as will assist the DSB in making the recommendations or in giving the rulings provided for in the covered agreements. Panels should consult regularly with the parties to the dispute and give them adequate opportunity to develop a mutually satisfactory solution.68

Article 11 has been understood by the AB as the Panel’s standard of review.69 This was most significantly articulated by the AB in EC—Hormones in which it stressed the importance of a standard of review reflecting ‘the balance established in the Agreement between the jurisdictional competences conceded by the Members for themselves’.70 Article 11, it stated, articulates with great succinctness but with sufficient clarity the appropriate standard of review in respect of both the ascertainment of such facts and the legal characterisation of such facts under the relevant agreements.71

66

Appellate Body Report, above n 2 at para 187. Howse above n 49. 68 Dispute Settlement Understanding, Art 11. 69 Appellate Body Report, above n 2 at para 116. For an account of this see D Palmeter and P Mavroidis, Dispute Settlement in the World Trade Organisation, 2nd edn (Cambridge, Cambridge University Press, 2004) at 152–5, and M Oesch, Standards of Review in WTO Dispute Resolution (Oxford, Oxford University Press, 2003). 70 Appellate Body Report, above n 2 at para 116. 71 Ibid at para 116. 67

(H) Fisher Ch5

31/5/07

184

09:58

Page 184

Risk Assessment, The WTO and Administrative Constitutionalism

Such review, it noted, is neither de novo nor total deference but requires an ‘objective assessment of the facts’.72 A consequence of this understanding of the function of Panels and their standard of review is to characterise the decisionmaking that they are reviewing as consisting of two discrete activities: accurate fact-finding and correct application of the law. There is very little room to understand decision-making as a discretionary practice, and this characterisation is reinforced by the fact that Panels can commission expert evidence to assist them in the fact-finding process.73 There is very little discussion of the role of discretion in the risk evaluation process, despite the fact that this has been the focus in national jurisdictions, as seen in chapters two to four. Panels have also tended to understand SPS standard setting as the practice that occurs before them in the process of dispute settlement.74 Thus the requirements of Member States in relation to standard setting are largely understood as a set of evidentiary requirements before a Panel. This is similar to what was seen in chapter four in relation to merits review in Australia.75 Thus, for example, considerable emphasis has been placed on who bears the burden of proof,76 and Panels refer to evidentiary concepts from an adjudicative/adversarial context.77 In some of the early SPS cases, the Panel also collapsed the requirements of showing that a Member State had complied with the SPS Agreement with the operation of the burden of proof, although the AB ultimately ruled this was an incorrect approach.78 The ultimate consequence is that the dispute settlement process at the Panel level promotes an RI approach to standard setting in that it has resulted in standard setting being understood as the correct assessment of the facts and the law before the Panel. The notion that the SPS Agreement does not regulate the procedure of standard setting but rather the substance of SPS standard setting reinforces this point79—what is relevant is the evidence before a Panel rather than showing that a Member State went through a particular procedure. The final point to note about the difference between administrative constitutionalism operating in a national legal system and within the WTO SPS context is that there are strong incentives not to recognise dispute settlement as an arena for 72

Appellate Body Report, above n 2 at para 117. DSU, Art 13, and SPS Agreement, Art 11.3, although note that the wording of these provisions is different. 74 Panel Report, Japan—Measures Affecting Agricultural Products, WT/DS76/R, 27 October 1998 at para 8.42; and Panel Report, Japan—Measures Affecting the Importation of Apples, WT/DS245/R, 15 July 2003 at para 8.105. 75 See section V of ch 4. 76 Appellate Body Report, above n 2 at paras 97–108; Appellate Body Report, Australia—Measures Affecting Importation of Salmon, above n 30 at paras 257–61; Appellate Body Report, Japan—Measures Affecting Agricultural Products, above n 30 at paras 118–131; and Panel Report, Japan—Measures Affecting the Importation of Apples, above n 74 at para 122. For commentary see Palmeter and Mavroidis, above n 69 at 143–50; M Grando, ‘Allocating the Burden of Proof in WTO Disputes: A Critical Analysis’ (2006) 9 Journal of International Economic Law 615; and J Paulwelyn, ‘Evidence, Proof and Persuasion in WTO Dispute Settlement’ (1998) 1 Journal of International Economic Law 227. 77 Panel Report, Japan—Measures Affecting the Importation of Apples, above n 74 at paras 8.94–8.95. 78 Appellate Body Report, above n 2 at paras 97–108. 79 Ibid at para 109. 73

(H) Fisher Ch5

31/5/07

09:58

Page 185

EC—Hormones and the Definition of Risk Assessment

185

debating the normative legitimacy of different SPS standard-setting regimes. WTO dispute settlement has been constructed on the basis that it is an objective and neutral form of dispute settlement80 in which neither the Panels nor the AB are lawmakers but are conduits for applying the law.81 Both, in line with the Vienna Convention, are theoretically meant to be only giving ‘ordinary meaning’ to Treaty terms82 and, as seen above, the task of Panels is to carry out an ‘objective assessment of the facts’.83 Indeed, the operation of the science/democracy dichotomy in this area reinforces this idea of neutrality in that the Agreement is simply understood as imposing scientific objectivity rather than regulating a Member State’s normative choices about the role and nature of SPS standard setting. The Agreement is thus using the ‘epistemic authority’ of science to find the appropriate balance between trade liberalisation and health protection.84 The problem, of course, is that the SPS Agreement does raise issues of administrative constitutionalism, and Panels and the AB are not just ruling on the trade liberalisation and health protection balance but also on the reasonableness of standard-setting processes. Questions of standard of review are not just about an objective assessment of the facts but also about developing a body of doctrine that must encompass issues of vertical, horizontal, and institutional competence which are embedded in, but transcend, public international law principles.85 What can thus ultimately be seen is that there is a failure properly to recognise the fundamental issues that are raised by the SPS Agreement. These issues are discussed in the next two sections.

III EC—Hormones and the Definition of Risk Assessment In discussion so far I have simply asserted that there is a relationship between the SPS Agreement and administrative constitutionalism. What I now wish to do is to show how issues of administrative constitutionalism have been raised in Panel and AB reports in relation to disputes concerning the SPS Agreement and, in particular, how Article 5.1 of the Agreement has been defined. An excellent example of the importance of paradigms of administrative constitutionalism in interpreting Article 5.1 can be seen by comparing the Panel and AB reports in 80 A von Bogdandy, ‘Law and Politics in the WTO: Strategies to Cope with a Deficent Relationship’ (2001) 5 Max Planck Yearbook of United Nations Law 609 at 617. 81 C-D Ehlermann, ‘Reflections on the Appellate Body of the WTO’ (2003) 6 Journal of International Economic Law 695 82 Vienna Convention on the Law of Treaties, Art 31(1). 83 Appellate Body Report, above n 2 at 117. 84 Winickoff et al, above n 49 at 84. See also Button, above n 16 at 44. 85 For a thoughtful discussion of these issues see Oesch, above n 69 at ch 2. See also C-D Ehlermann and N Lockhart, ‘Standard of Review in WTO Law’ (2004) 7 Journal of International Economic Law 491.

(H) Fisher Ch5

31/5/07

186

09:58

Page 186

Risk Assessment, The WTO and Administrative Constitutionalism

EC—Hormones.86 The former is a classic example of RI reasoning and the latter a classic example of DC reasoning. In the EC—Hormones dispute, the US challenged the EU’s ban on beef being sold from cattle that had been treated with certain growth hormones. Among other things, both the Panel and the AB found that the EU measure had not been consistent with Article 5.1 but their interpretation of that provision diverged. The Panel defined it in RI terms while the AB defined it in DC terms. In both cases, they understood the relevant definition of risk assessment to be that set out in the second half of Annex A.4. The Panel’s starting point for its analysis was that Article 5.1 must be understood in terms of the division between a scientific process of risk assessment and a political process of risk management.87 Its characterisation of the standard-setting process in this way seemed to have little to do with the text of the Agreement and to be more due to the way in which the arguments were put to it by the parties.88 In particular, this characterisation of standard setting came from the US regulatory context.89 The division rests upon a presumption that standard setting can be divided into a wholly scientific process of analysing the facts and a political process of applying those facts to the relevant normative prescription.90 That division makes sense only if risk assessment is understood to be an objective process of ‘scientific’ information assessment, which is indeed the way in which the Panel characterised it. It stated that Article 5.1 should be understood as a ‘specific application’ of the obligations set out in Article 2.2.91 Moreover, it stated that the assessment of risks required the ‘scientific examination of data and factual studies’92 and applied a relatively narrow definition of risk assessment techniques by relying on documents from international bodies even though the Codex Alimentarius had not at that time issued a formal statement on risk assessment.93 The RI nature of risk assessment was reinforced by the fact that the Panel also stated that a Member State should evaluate the ‘potential or probability’ of the adverse effects occurring.94 The term ‘probability’ is not included in the Agreement and it is a primarily quantitative notion.95 Furthermore, the Panel interpreted the phrase ‘based on’ in Article 5.1 to mean that the scientific conclu86 Panel Report, European Communities Measures Concerning Meat and Meat Products (Hormones) above n 2 and Appellate Body Report, European Communities—Measures Concerning Meat and Meat Products (Hormones) above n 2. 87 Panel Report, above n 2 at paras 8.91–8.97. 88 Ibid at para 8.95. 89 See ch 3 above, and National Research Council, Risk Assessment in the Federal Government: Managing the Process (Washington DC, National Academy Press, 1983). 90 Panel Report, above n 2 at paras 8.94, 8.96 and 8.97. 91 Ibid at para 8.93. 92 Ibid at para 8.94. 93 Ibid at para 8.103. The authority for relying on these documents is Art 3.1. 94 Ibid at para 8.98 95 Although one having a variety of definitions: see MG Morgan and M Henrion, Uncertainty: A Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis (Cambridge, Cambridge University Press, 1991) at 48–50.

(H) Fisher Ch5

31/5/07

09:58

Page 187

EC—Hormones and the Definition of Risk Assessment

187

sions of the risk assessment and the scientific conclusions reflected in the measure were in conformity with each other.96 For the Panel, the task of a Member State’s regulatory body in setting a standard was to apply the facts to a normative prescription by using an analytical methodology, and its decision is a perfect example of defining risk assessment in RI terms. Standard setting was largely characterised as a compartmentalised process in which the standard setter identified the facts and then applied those facts to a pre-ordained normative prescription. As this was the case, the process of assessing the EU’s compliance with the SPS Agreement was understood as requiring the scrutinising of the analytical methodology of the risk assessment and the methodological rigour of the scientific basis. The burden on the EU was also primarily understood as an evidential one,97 and the Panel relied heavily on appointed experts in its assessment of that evidence.98 Furthermore, there was little role for principles such as the precautionary principle or for discussion of the problems that scientific uncertainty created for standard setting.99 Risk assessment was understood to not be complicated by scientific uncertainty or socio-political complexity. The AB, in contrast, took a very different approach and one that was far more aligned with the DC paradigm in that the AB characterised the process of standard setting in far more complex terms. First, it noted that there was no authority for the distinction that the Panel made between risk assessment and risk management, and that the utilisation of such a distinction resulted in ‘a restrictive notion of risk assessment’.100 Moreover, while the AB stated that Article 5.1 should be interpreted in light of Article 2.2 it was less adamant that it was a specific application of it.101 Moreover, it stated that the Panel’s use of ‘probability as an alternative for potential’ created a ‘significant concern’ because it implied both a ‘higher degree of a threshold of potentiality or possibility’ and its use introduced ‘a quantitative dimension to the notion of risk’.102 It went on to state that the imposition of such a quantitative requirement had no basis in the Agreement, although it did note that there were conceptual problems concerned with ascertaining risk.103 Here we can see that the AB took an approach to standard setting that is not about characterising it as a series of discrete tasks which could be constrained by a strict methodology. Rather, it recognised that risk assessment required a broader definition. It stated that the Panel’s approach to risk assessment was too narrow because it purported to exclude from the scope of risk assessment

96

Panel Report, above n 2 at para 8.117. Ibid at para 8.100. 98 This is particularly in relation to the issue of whether a measure was based on a risk assessment: see ibid at para 8.133. 99 Ibid at para 8.157-8. 100 Appellate Body Report, above n 2 at para 1.81. 101 Ibid at para 180. 102 Ibid at para 184. 103 Ibid at para 186. 97

(H) Fisher Ch5

31/5/07

188

09:58

Page 188

Risk Assessment, The WTO and Administrative Constitutionalism

all matters not susceptible of quantitative analysis by the empirical or experimental laboratory methods commonly associated with the physical sciences.104

For the AB this would have been clearly inconsistent with Article 5.2 of the Agreement, and it stated that Article 5.1 was not only concerned with risk ascertainable in a science laboratory operating under strictly controlled conditions, but also risk in human societies as they actually exist, in other words, the actual potential for adverse effects on human health in the real world where people live and die.105

Moreover, the AB also noted that there was flexibility in the requirements of Article 5.1106 Ultimately, risk assessment could take into account a range of different forms of information and could be quantitative and qualitative in nature. This could also be evidenced by Articles 5.2 and 5.3. Implicit in the AB’s approach is an appreciation of the complexities in assessing risk and the problems of scientific uncertainty. The AB was not being ‘scientific’ or ‘anti scientific’ but rather was requiring the assessment of risk to be on a broader basis than understanding risk assessment as an analytical straightjacket would allow. In other words, the AB was characterising standard setting more as a reasoning process that encompassed a fact-finding process rather than solely as a factfinding process. This can also be seen in the AB’s discussion of the precautionary principle where it acknowledged the principle’s relevance right across the standardsetting process.107 The principle and the concept of precaution were understood to inform a Member State’s general approach to standard setting rather than simply applying in discrete circumstances. Furthermore, the AB ruled that the Panel’s interpretation of ‘based on’ was far too narrow. Rather, it argued that Article 5.1 required that a risk assessment ‘must sufficiently warrant—that is to say, reasonably support’108 an SPS measure but that there was no need for a risk assessment to come to a monolithic conclusion.109 Indeed, the AB recognised that there may be a range of different scientific conclusions and problems created by scientific uncertainty.110 The AB’s approach can be treated as being underpinned by a DC approach. Standard setting was understood to be a complex enterprise not easily kept within the boundaries of stringent risk assessment methodologies. Rather, the AB assessed the compatibility of the EU’s measure with the Agreement by determining whether the EU had carried out a coherent process of reasoning.111 Looking at these two reports, three things are very clear. First, the Panel and the AB are not simply engaging in an exercise of giving risk assessment its ‘ordinary

104

Appellate Body Report, above n 2 at para 187. Ibid at para 187. 106 Ibid at para 129. 107 Ibid at para 124. 108 Ibid at para 193. 109 Ibid at para 194. 110 Ibid at para 194. 111 Ibid at paras 205 and 207, although note here the limits of the Appellate Body’s review. For a discussion of this see above n 30 at para 261. 105

(H) Fisher Ch5

31/5/07

09:58

Page 189

EC—Hormones and the Definition of Risk Assessment

189

meaning’.112 This is because risk assessment has no ordinary meaning and takes its definition from the different institutional cultures in which it operates.113 Moreover, one cannot define risk assessment without reference to deeper principles of administrative constitutionalism.114 The differences here can be seen as stemming from a number of disparate sources. The Panel was clearly influenced by quantitative understandings of risk while the AB was not. The Panel and AB also understood the SPS Agreement as serving divergent purposes.115 The Panel largely characterised the Agreement as being a means of reducing regulatory heterogeneity and thus promoted the RI paradigm.116 In contrast, the AB understood the Agreement as being far more about ensuring that bogus SPS measures were invalidated, and thus promoted the DC paradigm.117 Furthermore, the Panel’s standard of review inevitably led it to understand its role and that of an SPS standard setter in RI terms, while the AB was not constrained by such understandings. Alternatively, it could be understood that the Panel and the AB were choosing between different national regulatory styles. Thus, for example, the Panel’s risk assessment/risk management distinction was a product of US regulatory politics118 and in embracing that distinction the Panel also embraced the RI paradigm that has dominated US risk regulation for the last 20 years.119 In contrast, the AB could be understood as reflecting the more deliberative nature of EU regulatory standard setting. Second, and following on from this, both the Panel and the AB are making a normative choice about what is reasonable standard setting. They are not neutral arbiters simply policing the science but rather laying down principles about how standard setting should occur. As is clear from above, in doing so they will be influenced by many different factors. Finally, both these interpretations are equally valid, and neither the Panel nor the AB is distorting the SPS Agreement. Rather, each is emphasising different aspects of it, as well as interpreting it in different ways. The SPS Agreement thus has the ability to be interpreted in both RI and DC terms.

112

Vienna Convention on the Law of Treaties, Art 31(1). Winickoff et al, above n 49. 114 Fisher, above n 24. 115 The Panel in EC—Hormones interpreted Art 3.1 as the general rule in departing from which a Member State bore a burden of proof (in line with the Agreement being concerned with reducing regulatory heterogeneity): see Panel Report, above n 2 at para 8.86. The Appellate Body, in contrast, did not adopt this approach: see Appellate Body Report, above n 2 at paras 102–4 and 177. 116 This can be seen in its ruling that Art 3.1 creates a general rule for which Art 3.3 is an exception: see Panel Report, above n 2 at para 8.86. 117 Appellate Body Report, above n 2 at paras 104 and 124. 118 Fisher, above n 24; and S Jasanoff, The Fifth Branch: Science Advisers as Policy Makers (Cambridge, Harvard University Press, 1990). 119 See ch 3 above. 113

(H) Fisher Ch5

31/5/07

190

09:58

Page 190

Risk Assessment, The WTO and Administrative Constitutionalism

IV Defining Risk Assessment in Disputes since EC—Hormones: The Inadvertent Pursuit of the RI Paradigm? In light of the divergence of opinion that could be seen between the Panel and the AB in EC—Hormones over the nature of risk assessment, one would expect that after these two reports there would have been an opening-up of a debate over administrative constitutionalism, or at least a realisation that risk assessment can mean many different things and that dispute settlement is not an objective exercise. Yet, while some commentators highlighted the normative issues raised by EC—Hormones,120 the dispute was generally understood as raising tricky questions about factual review and/or state sovereignty rather than about administrative constitutionalism.121 Moreover, in disputes since EC—Hormones, SPS standard setting has been increasingly understood as involving the question of whether a measure was based on ‘sufficient scientific evidence’.122 A consequence of this has been that the dispute settlement process has inadvertently promoted the RI paradigm, and this can be seen most obviously in Japan—Varietal Products II 123 and Japan—Apples.124 I say inadvertently because in these decisions the RI promotion has had more to do with a focus on science than a focus on understandings of good public administration. This is very different from what has occurred in national legal cultures, because the promotion of either the RI or DC paradigm is always closely connected to debates over the legitimacy and accountability of public administration. That link has not been explicitly made in these cases. The next dispute after EC—Hormones in which the interpretation of Article 5.1 arose was Australia—Salmon.125 This case concerned a challenge by Canada to an Australian ban on the importation of fresh, chilled, or frozen salmon which had not undergone a prescribed heat treatment process so as to remove the risk that such products would carry diseases into Australia.126 Among the many arguments 120

Walker, above n 16. J Pauwelyn, ‘Does the WTO Stand for “Deference to” or “Interference with” National Health Authorities when Applying the Agreement on Sanitary and Phytosanitary Measures (SPS Agreement)?’ in T Cottier and P Mavroidis (eds), The Role of the Judge in International Trade Regulation: Experience and Lessons for the WTO (Ann Arbor, University of Michigan Press, 2003), and C Correa, ‘Implementing National Health Policies in the Framework of WTO Agreements’ (2000) 34 Journal of World Trade 89. 122 Art 2.2. See Button, above n 16 at 44–55 for a discussion of this development. 123 Panel Report, Japan—Measures Affecting Agricultural Products, above n 74; and Appellate Body Report, Japan—Measures Affecting Agricultural Products, above n 30. 124 Panel Report, Japan—Measures Affecting the Importation of Apples above n 74; and Appellate Body Report, Japan—Measures Affecting the Importation of Apples, WT/DS245/AB/R, 26 November 2003. 125 Panel Report, Australia—Measures Affecting Importation of Salmon, WT/DS18/R, 12 June 1998; and Appellate Body Report, Australia—Measures Affecting Importation of Salmon, above n 30. 126 Panel Report, above n 125 at para 2.11. 121

(H) Fisher Ch5

31/5/07

09:58

Page 191

Defining Risk Assessment in Disputes since EC—Hormones

191

brought by Canada was that the ban was not consistent with Article 5.1 of the SPS Agreement. Australia had only done a risk assessment for ocean-caught Pacific salmon and thus was clearly in breach in relation to the ban of other salmon products. Most of the analysis about Article 5.1 thus concerned the ocean-caught Pacific salmon risk assessment. Although this dispute concerned the definition of Article 5.1, the context was a very different one from that of EC—Hormones. First, as the dispute concerned the entry of a pest or disease into a jurisdiction, Australia’s actions were governed by the first definition of risk assessment included in Annex A.4 rather than the second, which was the focus in EC—Hormones. A consequence of this was that the analytical requirements placed on Australia were different. Under that first definition, the reference was to ‘likelihood’, while under the second definition, the reference was to ‘possibility’. As well, international guidelines on import risk assessment had been developed by the ‘Office international des épizooties’ (OIE), which needed to be taken into account under Article 5.1 due to these guidelines being ‘risk assessment techniques developed by the relevant international organizations’.127 These guidelines referred to a risk assessment establishing probabilities.128 In both cases, there was thus an expectation that a risk assessment would produce a more analytically rigorous analysis of risks than was possible in EC— Hormones. This would suggest a more RI approach to understanding standard setting but the Panel mainly followed the DC guidance set out by the AB in EC—Hormones. Its focus was upon the reasoning of the Member State, and they characterised risk assessment not in terms of a set of analytical tools but rather as three distinct reasoning processes. It stated: (1) the definition of risk assessment implies that the Member imposing the sanitary measure first has to identify the disease(s) whose ‘entry, establishment or spread’ within its territory it wants to prevent as well as the ‘associated potential biological and economic consequences’; (2) the definition then requires an ‘evaluation of the likelihood’ of entry, establishment or spread of these diseases and of the associated potential biological and economic consequences’; and (3) the definition further requires that the evaluation of the likelihood of entry, establishment or spread of these diseases be conducted ‘according to the sanitary . . . measures which might be applied’.129 [their emphasis]

These three steps are based on the first definition of risk assessment in Annex A.4. The focus is upon establishing a logical line of reasoning that connects identifying a risk, an evaluation of the likelihood of that risk, and the ability of the SPS measure to reduce such a risk. Such reasoning needed to be in reasonable detail. Thus in relation to the first requirement the Panel stated that the risks identified must 127 128 129

Ibid at paras 2.11 and 8.49. Ibid at paras 8.78 and 8.86. Ibid at para 8.72.

(H) Fisher Ch5

31/5/07

192

09:58

Page 192

Risk Assessment, The WTO and Administrative Constitutionalism

be disease-specific, and a risk assessment cannot ‘simply address the overall risk related to the combination of all diseases of concern’130. Moreover, the Panel concluded that a Member State in ‘evaluating the likelihood’ would need to establish the probability of adverse consequences, which could be expressed in either qualitative or quantitative terms.131 It also started that for the second and third of the steps in a risk assessment, only ‘some’ evaluation of likelihood was required.132 In this case it concluded that the particular report that Australia was purporting to be a risk assessment was indeed one.133 The next question was whether the measures taken by Australia were ‘based on’ that risk assessment. The Panel’s starting point for understanding ‘based on’ was the AB’s statement in EC—Hormones in which it emphasised the requirements that ‘the results of the risk assessment must sufficiently warrant—that is to say, reasonably support—the SPS measure at stake’, and that there must be ‘a rational relationship between the measure and the risk assessment’.134 In this case it found that the measure was not based on the risk assessment because Australia had not established whether heat treatment would actually inactivate the pathogens. In particular, Australia had made ‘no substantive assessment of the risk or the risk reduction related to the heat treatment requirements in effect imposed by the measure at issue’.135 Again, the focus of the Panel was upon the reasoning of the Member State rather than directly on issues of proof. The Panel’s conclusions were appealed and the AB took a very different approach, one far more informed by the RI paradigm. The AB found that the Panel had erred in law by concluding that the SPS measure was the heat treatment requirement and not the import prohibition.136 As it had come to this conclusion it then decided that it needed to rule on whether the import prohibition was based on a risk assessment.137 In so doing the AB adopted the same three-pronged definition of risk assessment used by the Panel138 but interpreted it in a far more RI manner. In particular, it gave the concept of probability far greater emphasis.139 Moreover, it was not as deferential to the AB’s decision in EC—Hormones, perhaps because of the existence of the OIE guidelines and because they were concerned with the first definition of risk assessment as set out in Annex A.4.140 Indeed, as Perez notes, the AB in this case understood risk assessment as a probabilistic tool rather than as a more general reasoning process.141 130 131 132 133 134 135 136 137 138 139 140 141

Panel Report, above n 125 at para 8.74. Ibid at paras 8.77–8.80 Ibid at paras 8.80, 8.83, 8.89. and 8.91. Ibid at para 8.92. Ibid at para 8.94, quoting Appellate Body Report, above n 2 at para 198. Panel Report, above n 125 at para 8.98. Appellate Body Report, above n 30 at paras 103–4. Ibid at para 108. Note its reference to its limited role under Art 17. 6 of the DSU. Ibid at paras 121–2. Ibid at para 123. Ibid at para 123. Perez, above n 42 at 140.

(H) Fisher Ch5

31/5/07

09:58

Page 193

Defining Risk Assessment in Disputes since EC—Hormones

193

What this means is that the AB laid down far stricter analytical requirements for Member States.142 Thus while the AB in Australia—Salmon also concluded that probability did not need to be expressed in quantitative terms it did not agree with the Panel that only ‘some evaluation’ of likelihood was needed.143 Indeed, ultimately the AB concluded that in this case, Australia had not conducted a proper risk assessment because there was only ‘some evaluation’ of the probabilities rather than a comprehensive analysis.144 In particular, it noted that: the existence of unknown and uncertain elements does not justify a departure from the requirements of Articles 5.1, 5.2 and 5.3, read together with paragraph 4 of Annex A, for a risk assessment. We recall that Article 5.2 requires that ‘in the assessment of risk, Members shall take into account available scientific evidence’. We further recall that Article 2, entitled ‘Basic Rights and Obligations’, requires in paragraph 2 that ‘Members shall ensure that any sanitary . . . measure . . . is based on scientific principles and is not maintained without sufficient scientific evidence, except as provided for in paragraph 7 of Article 5’.145

This, again, is a very narrow RI reading of risk assessment in which the emphasis is upon risk assessment as a scientific tool, thus the reference to ‘available scientific evidence’ in Article 5.2 and to ‘sufficient scientific evidence’ in Article 2.2. Thus, unlike in EC—Hormones, the AB was not focusing upon the whole of the standard-setting process but only its scientific aspects. The practical consequence of this is that the RI paradigm is promoted in that standard setting is largely understood as the analytical application of the facts to the law without any regard for broader-ranging issues. Indeed, Australia—Salmon laid the foundations for understanding the SPS Agreement in solely scientific terms and thus in RI terms. This can be seen very clearly in the next two disputes. The next case in which Article 5.1 was considered was Japan—Agricultural Products II,146 in which the US challenged a Japanese regulation that required the testing of different agricultural products for codling moth before they would be allowed into Japan (the varietal testing requirement). This case once again concerned the first and stricter definition of risk assessment in Annex A.4, and the issue in this case was whether the need to test different varieties of agricultural products was consistent with the SPS Agreement. What is particularly interesting about the decision was that there was considerable evidence before the Panel concerning the scientific uncertainties over the risks and the seriousness of the risks, and that expert evidence rested upon opinion rather than fact.147

142

Ibid at 141. Appellate Body Report, above n 30 at para 124. 144 Ibid at paras 128, 132, and 135. 145 Ibid at para 130. 146 Panel Report, Japan—Measures Affecting Agricultural Products, above n 74, and Appellate Body Report, Japan—Measures Affecting Agricultural Products, above n 30. 147 Panel Report, Japan—Measures Affecting Agricultural Products, above n 30 at paras 8.30– 8.41. 143

(H) Fisher Ch5

31/5/07

194

09:58

Page 194

Risk Assessment, The WTO and Administrative Constitutionalism

As in all these disputes,148 the importance of the relationship between Article 2.2 and Article 5.1 was first stressed.149 In this case, however, the Panel concluded that if the measure was found not to be consistent with Article 2.2 then it would not be consistent with Article 5.1,150 which meant that the Panel concluded that they only needed to analyse the provision’s consistency with Article 2.2 and, in particular, the requirement that it could not be maintained ‘without sufficient scientific evidence’.151 This is a very narrow reading of the basis of decision-making which suggests that the only basis is science. It also represents a shift from understanding the SPS Agreement as being concerned with regulating the overall process of decision-making to conceiving it as just regulating the evidentiary basis, where that basis is understood solely in scientific terms. The consequence of this shift from an administrative constitutionalism perspective was that the RI paradigm was inadvertently promoted. Thus, for example, the Panel stated: In our view, for a phytosanitary measure to be ‘maintained without’ sufficient scientific evidence, there needs to be a lack of an objective or rational relationship between, on the one hand, the phytosanitary measure at stake (in casu, the varietal testing requirement) and, on the other hand, the scientific evidence submitted before the Panel in casu, in particular the six studies referred to by Japan).152

The reasonableness of the SPS measure was thus determined by the objective relationship between the evidence before the Panel and the measure. Moreover, this test would seem to be a more stringent evidentiary requirement than the Article 5.1 ‘based on’ requirement. The role of standard setter is thus largely one of producing proof. Interestingly, it was not the Panel itself that determined whether there was ‘sufficient scientific evidence’. Rather, the Panel referred to the experts advising the Panel.153 Moreover, while it was clear from the expert evidence that there was considerable scientific uncertainty over the issue, which made assessment and evaluation difficult,154 the Panel still concluded that Japan had not maintained its measure on the basis of sufficient scientific evidence. In so doing, the Panel understood its task as reviewing the evidence before it in accordance with the rules of burden of proof.155 Moreover, while it noted that Japan did have data which did ‘hint’ at ‘relevant varietal differences’, it also stated that: no evidence before this Panel makes the actual causal link between the differences in the test results and the presence of varietal differences.156 148 Panel Report, above n 2 at para 8.93; Appellate Body Report, above n 2 at para 180; and Panel Report, above n 125 at paras 8.52 and 8.56. 149 Panel Report, Japan—Measures Affecting Agricultural Products, above n 74 at para 8.15 150 Ibid at para 8.63. 151 Ibid at para 8.30. 152 Ibid at para 8.29. 153 Ibid at para 8.32 154 Ibid at paras 8.30–8.41. 155 Ibid at para 8.42. 156 Ibid at para 8.42.

(H) Fisher Ch5

31/5/07

09:58

Page 195

Defining Risk Assessment in Disputes since EC—Hormones

195

As such, it noted: On these grounds and after having carefully weighed the evidence and opinions of the experts advising the Panel submitted to us, we thus consider that the United States has raised a presumption that Japan’s varietal testing requirement is maintained without sufficient scientific evidence and that this presumption has not been sufficiently rebutted by Japan.157

This is very narrow RI understanding of standard setting and amounts to a test of requiring a standard setter to establish the evidentiary basis for their decision rather than justifying all of their reasoning. It is a very different approach to that taken by the AB in EC—Hormones or by the Panel in Australia—Salmon. The Panel’s decision in Japan—Agricultural Products II was appealed to the AB, and Japan argued, among other things, that the Panel had erred in law in its interpretation of Article 2.2. The AB did not accept this appeal and roughly agreed with the Panel’s interpretation of Article 2.2 although strengthening its RI aspects by referring to the need for measures to have a ‘scientific justification’ as set out in Article 3.3 as well.158 Moreover, it understood the obligations in Article 2.2 in even more RI terms. Thus it noted: In the light of the above considerations based on the text and context of Article 2.2 of the SPS Agreement, we agree with the Panel that the obligation in Article 2.2 that an SPS measure not be maintained without sufficient scientific evidence requires that there be a rational or objective relationship between the SPS measure and the scientific evidence. Whether there is a rational relationship between an SPS measure and the scientific evidence is to be determined on a case-by-case basis and will depend upon the particular circumstances of the case, including the characteristics of the measure at issue and the quality and quantity of the scientific evidence.159

This is a restatement of the Panel’s conclusion, but the AB also required that the rationality of the relationship depended primarily upon the ‘quality and quantity of scientific evidence’ rather than any form of rational reasoning. In light of this conclusion it would seem that there was very little role for Article 5.1, its process requirements being subsumed by the proof requirements of Article 2.2, but the AB did conclude that the Panel had been incorrect in not analysing the varietal testing requirement’s consistency with Article 5.1 for certain products because the Panel’s conclusion in relation to Article 2.2 had not been in relation to these.160 The AB thus decided to analyse this issue and used as its starting point the three-step test of risk assessment set out in Australia—Salmon.161 In doing so, it understood those tests more as evidentiary requirements rather than a framework

157

Ibid at para 8.42. Appellate Body Report, Japan—Measures Affecting Agricultural Products, above n 30 at paras 75–83. 159 Ibid at para 83. 160 Ibid at para 111. 161 Ibid at para 112. 158

(H) Fisher Ch5

31/5/07

196

09:58

Page 196

Risk Assessment, The WTO and Administrative Constitutionalism

for logical reasoning.162 As there was no risk assessment that evaluated the efficacy of the SPS measure then there was no compliance with Article 5.1.163 The next dispute was in 2003 and concerned Japanese measures in relation to US apple imports so as to prevent the spread of fire blight disease in Japan—the Japan—Apples dispute.164 As such, this case was once again about the first definition of risk assessment in the Annex A.4 definition of risk assessment. Among the many arguments that the US put was that Japan had not based its measures on either Article 2.2 or Article 5.1. Once again, we can see the approach being taken by the Panel to be an RI one in which the focus was on the scientific evidence requirements of Article 2.2 rather than upon the process requirements of Article 5.1. In interpreting the concept of ‘sufficient scientific evidence’ the Panel stated: a) First, the very notion of ‘scientific evidence’ seems to exclude elements of information that cannot be considered as ‘evidence’. The same notion also seems to exclude any evidence that is not ‘scientific’. b) Second, the term ‘sufficient’ seems to address not only the quantity and quality of the evidence as such, but also the ‘causal link’ between the phytosanitary measure at issue and the scientific evidence establishing a phytosanitary risk and justifying the measure.165

This is an even narrower interpretation of ‘sufficient scientific evidence’ than seen in Japan—Varietal Products II and requires a Member State to prove that there is a risk and that the measure will reduce that risk. Moreover, the Panel went on to note that the concept of ‘scientific’ meant ‘evidence gathered through scientific methods’166 and that the concept of scientific evidence ‘excludes in essence not only insufficiently substantiated information but also such things as a nondemonstrated hypothesis’167 As such the Panel did not see it as appropriate to rely on ‘purely circumstantial evidence’, although it did recognise that in this area this was often the only type of evidence which existed.168 In response to the argument that this would lead to Member States being able to pass measures only in relation to proven risks, it pointed to the role of Article 5.7 and then also stated: requiring ‘scientific evidence’ does not limit the field of scientific evidence available to Members to support their measures. ‘Direct’ or ‘indirect’ evidence may be equally considered. The only difference is not one of scientific quality, but one of probative value within the legal meaning of the term, since it is obvious that evidence which does not directly prove a fact might not have as much weight as evidence directly proving it, if it is available.169

162 Appellate Body Report, Japan—Measures Affecting Agricultural Products, above n 30 at paras 76–79. 163 Ibid at para 113. 164 Panel Report, Japan—Measures Affecting the Importation of Apples, above n 74. 165 Ibid at para 8.82. 166 Ibid at para 8.92. 167 Ibid at para 8.93 168 Ibid at paras 8.94–8.95. 169 Ibid at para 8.98.

(H) Fisher Ch5

31/5/07

09:58

Page 197

Defining Risk Assessment in Disputes since EC—Hormones

197

This clarification is not particularly helpful in that the Panel are combining legal and scientific concepts of proof. The Panel also noted and agreed with the AB’s interpretation of sufficiency in Japan—Agricultural Products II.170 Moreover, they stated that: scientific evidence relates to a risk and is supposed to confirm the existence of a given risk.171

A risk was thus understood as a scientifically identifiable phenomenon and standard setting as a fact-finding exercise.172 After a lengthy analysis of the issues, the Panel concluded that the measure was not consistent with Article 2.2.173 The Panel then went on to consider whether Japan had also breached Article 5.1. While it recognised the close relationship between Article 5.1 and Article 2.2, it also saw it as important to consider them separately.174 Moreover, the Panel also identified a number of different risk assessment techniques developed by international organisations which it saw as being relevant.175 The Panel adopted the definition of risk assessment in Australia—Salmon176 and found that because within the relevant risk assessment there had been no ‘particular assessment of the degree of likelihood of an event’ there had been no evaluation of the probability and that the conclusions about the measures were too cursory.177 Again, in doing so, there was little focus on risk assessment procedures, and the overall analysis was dominated by the ‘sufficient scientific evidence’ requirement of Article 2.2. Thus much of the analysis of the risk assessment was an analysis of the evidence before the Panel.178 As with the other cases, Japan—Apples was appealed to the AB. One of the arguments put by Japan was that the Panel had erred in law by not giving ‘a certain degree of discretion’ to Japan in evaluating evidence and that it had relied too heavily on the views of the appointed experts.179 The AB rejected this argument and restated its approach in Japan—Agricultural Products II that ‘sufficient’ means a ‘rational or objective’ relationship.180 Moreover, the AB also agreed with the Panel’s interpretation of Article 5.1, and that in this case the Japanese measure had not been based on a risk assessment.181 It also noted that Article 5.1 did not dictate the methodology of a risk assessment, although in saying that it was not suggesting differences in scientific or analytical approach but rather differences in

170 171 172 173 174 175 176 177 178 179 180 181

Ibid at paras 8.101–8.103. Ibid at para 8.104. Although see ibid at para 8.105. Ibid at para 8.198. Ibid at para 8.230. Ibid at paras 8.241–8.244. Ibid at para 8.250. Ibid at paras 8.275 and 8.287–8.289. Ibid at paras 8.270, 8.284, and 8.288. Appellate Body Report, above n 124 at para 161. Ibid at para 162. Ibid at paras 203, 206, and 209.

(H) Fisher Ch5

31/5/07

198

09:58

Page 198

Risk Assessment, The WTO and Administrative Constitutionalism

how risk assessments were arranged.182 A Member State still needed to ‘attribute a likelihood of entry, establishment or spread of the disease to each agent specifically’,183 as well as provide a comprehensive analytical analysis of the consequences of the SPS measure.184 In other words, the AB was interpreting risk assessment as a tool for proving a risk and that an SPS measure would reduce that risk. The combined effect of Japan—Varietal Products II and Japan—Apples on WTO jurisprudence in this area is highly significant. In thinking about the reasonableness of SPS measures, the focus in these disputes is on the ‘sufficient scientific evidence’ requirement of Article 2.2 rather than the process requirements of Article 5.1. As such, both sets of decisions clearly promote the RI paradigm but they do so not because the Panel and the AB understand that the SPS Agreement raises issues of administrative constitutionalism, but because they understand dispute settlement as an exercise in proving the scientific basis for SPS measures. By doing this they are characterising SPS standard setting as primarily a process of applying the law to the facts, and thus in RI terms. This characterisation would seem to be under challenge, however, in the most recent dispute, EC—Biotech Products, in which at the time of writing only the Panel had reported.185 This case involved the US, Canada, and Argentina bringing actions to challenge a number of different EU measures in relation to products which contained genetically modified organisms (GMOs) on the basis that they were not consistent with the SPS Agreement and the TBT Agreement. The EU measures included: a general EU-wide moratorium on approvals; various EU product-specific measures; and various EU Member State safeguards. The report was issued only after lengthy and complex proceedings in which there was the submission of a large quantity of material, extensive fact finding, and the putting forward of many different arguments by the parties, including amicus curiae briefs.186 The final report ran to over 1,000 pages, not including annexes. Indeed, the case is a good example of the concept of a Panel as a neutral ‘fact finder’ being pushed to its limit. The approach of the Panel in EC—Biotech Products is very different from that seen in Japan—Varietal Products and Japan—Apples in that the Panel was far more focused on the overall process requirements of the SPS Agreement than upon a Member State providing scientific proof to justify its decisions. This is perhaps not surprising considering that many arguments before the Panel were not about proof but about the scope of the Agreement. The focus on the process can be seen in a number of different ways. 182

Appellate Body Report, above n 124 at para 204. Ibid at para 204. 184 Ibid at para 208. 185 Panel Report, above n 15. 186 Ibid at paras 7.37–7.43. See also L. Busch, et al, Amicus Curiae Brief Submitted to the Dispute Settlement Panel of the World Trade Organisation in the case of EC: Measures Affecting the Approval and Marketing of Biotech Products (2004). 183

(H) Fisher Ch5

31/5/07

09:58

Page 199

Defining Risk Assessment in Disputes since EC—Hormones

199

First, the Panel understood Article 2.2 as including three distinct requirements rather than just the requirement that measures cannot be maintained without ‘sufficient scientific evidence’.187 Although two of those requirements related to the scientific aspects of standard setting, this approach by the Panel placed those requirements in a broader context. Second, while recognising that Article 5.1 and Article 2.2 were related, the Panel also recognised that they operated independently of each other.188 This conclusion was inevitable because they found that a number of the different EU actions challenged were not measures (to which only Article 5.1 applies). The practical result was that Article 5.1 was not seen as interchangeable with Article 2.2. Third, and following on from this, the Panel understood the requirement for a risk assessment as part of a reasoning process that a Member State had to undertake—the risk assessment requirement being central to that process.189 This is not to say that the Member State itself had to undertake the risk assessment 190 but that the reasoning involved in a risk assessment was essential in justifying a measure. Fourth, in interpreting the concept ‘based on’, the Panel reiterated the DC ‘rational relationship’ test seen in the AB’s decision in EC—Hormones.191 In relation to the dispute before the Panel, the issue of Article 5.1 arose in the context of the EC Member State safeguard measures, and the Panel evaluated each safeguard measure to see if there had been a risk assessment, and where there had been, whether the measure was ‘based on’ it. In so doing, it did not discuss the nature of risk assessment in any detail except to note that risk assessment procedures may vary and that the requirements of Article 5.1 should be interpreted flexibly.192 It did, however, conclude that the risk assessments conducted as part of the EU regulatory process were risk assessments for the purposes of Article 5.1 because: they evaluated the likelihood of potential adverse effects on human health and/or the environment, as well as the associated potential consequences, according to the proposed use of the specific biotech product under consideration.193

This was even though such assessments were qualitative in nature. The Panel then proceeded to analyse each national safeguard measure to see whether any of the measures had been based on a risk assessment. It concluded that they had not, but in evaluating the consistency of each safeguard measure with Article 5.1 the Panel focused more on whether the measures were based on coherent reasoning processes rather than issues of proof.194 They were also more 187

Panel Report, above n 15 at paras 7.1424–7.1441. Ibid at paras 7.1751–1760. 189 Ibid at paras 7.3019–7.3021 190 Ibid at para 7.3024. 191 Ibid at para 7.0328. 192 Ibid at paras 7.3023, 7.3031, and 7.3032. Note the later discussion of the Australia—Salmon risk assessment definition at para 7.3039. 193 Ibid at para 7.3027 194 Ibid at paras 7.3040–7.3041 and 7.3077. 188

(H) Fisher Ch5

31/5/07

200

09:58

Page 200

Risk Assessment, The WTO and Administrative Constitutionalism

cognisant of scientific uncertainty195 and the normative nature of SPS standard setting.196 In relation to this last point, there was recognition that the same risk assessment could support different SPS measures, particularly if a Member State chose to take a precautionary approach in evaluating the assessment.197 In assessing whether any of the safeguard measures were ‘based on’ a risk assessment, the Panel set out more clearly its reasons. Thus, for example, in relation to the Austrian safeguard measure for Bt-176 maize, it recognised that the measure could be based on the Community risk assessments even though the Community institutions had come to a different conclusion. However, the Panel ruled that the Austrian measure could not be based on the Community risk assessment in this case because: a) we are not aware of any divergent opinions expressed in the risk assessments which were conducted by the lead CA [competent Member State authority under the relevant directive] and the risk assessments which were conducted by the SCPE, the SCAN or the SCF [the EU scientific communities involved] with regard to Bt176 maize; b) the European Communities or Austria did not explain, by reference to these risk assessments, how and why Austria assessed the risks differently, and did not provide a revised or supplemental assessment of the risks; c) the European Communities has not identified possible uncertainties or constraints in the risk assessments in question, and has not explained why, in view of any such uncertainties or constraints, Austria’s prohibition is warranted by the relevant risk assessments; and d) there is no apparent rational relationship between Austria’s safeguard measure, which imposes a prohibition, and risk assessments which found no evidence that Bt-176 maize will give rise to any adverse effects on human or animal health and the environment.198

These requirements are far more about Austria’s reasoning process rather than its providing proof, and the reasoning of the Panel in EC—Biotech Products is a shift away from understanding Article 5.1 solely in terms of the ‘sufficient scientific requirement’ contained in Article 2.2. Moreover, the Panel’s reasoning is also more in the DC terms seen in the AB’s report in EC—Hormones.

V The Problems with the WTO SPS Jurisprudence on Article 5.1 Before proceeding further it is useful to sum up my analysis so far. The WTO SPS Agreement has been understood as an Agreement that imposes scientific restraints 195 196 197 198

Ibid at paras 7.3064 and 7.3085. Ibid at para 7.3069. Ibid at para 7.3085. Ibid at para 7.3085. See also paras 7.3106, 7.3127, 7.3137, 7.3157, 7.3177, 7.3197, and 7.3211.

(H) Fisher Ch5

31/5/07

09:58

Page 201

The Problems with the WTO SPS Jurisprudence on Article 5.1

201

on the democratic wills of Member States. This, I have argued, is an incorrect characterisation because the Agreement is, in actual fact, regulating understandings of administrative constitutionalism and can be interpreted in terms of both the RI and DC paradigms of administrative constitutionalism. This can be seen quite starkly in the Panel and AB decisions in EC—Hormones. In dispute settlement since then, particularly in Japan—Varietal Products II and Japan—Apples, the focus has been on issues of proof. This has resulted in the inadvertent promotion of the RI paradigm: inadvertent because in these decisions there seems to be little realisation that the Panel and the AB are directing how public administration is being constituted, limited, and held to account in the different WTO Member States. The decision of the Panel in EC—Biotech Products represents a step away from the approach seen in those decisions, but time will only tell how this line of reasoning will be developed. My analysis so far has thus been mainly concerned with showing the incorrect nature of the current approach to characterising and interpreting the SPS Agreement but, as stressed in the introduction to this chapter, the present case-law is also problematic and many of the problems with it stem from the mischaracterisation of the Agreement in terms of the science/democracy dichotomy. I am not alone in identifying those problems, and most commentators see the operation of the SPS Agreement as one of the more legally troubling areas of WTO law.199 What is unique about my analysis is that I see that those problems stem from viewing the Agreement through an incorrect lens. Basically, there are four main problems with the current jurisprudence. The first is that such jurisprudence is not particularly coherent. As the authors of one leading WTO law textbook note about this case-law: In the WTO context, beyond banal statement, we are still unclear as to what distinguishes science from non-science.200

Thus, while a risk assessment needs some form of evaluation of ‘likelihood’ or ‘possibility’, the only guidance on what this means has been patchy and often inconsistent. An evaluation of probability can theoretically be qualitative but it is not clear exactly how. Likewise, the concept of a measure being ‘based on’ a risk assessment is also not clear, and the relationship between Article 2.2 and Article 5.1 is ambiguous, particularly after Japan—Varietal Products II and Japan— Apples. Much of this incoherence, I would argue, is due to the fact that understanding the Agreement as imposing a set of scientific requirements bears no relationship to how standard setting actually occurs. This is due to both the fact that risk assessment is not solely a scientific activity and that science may have many different definitions.201 199 T Cottier and P Mavroidis (eds), The Role of the Judge in International Trade Regulation: Experience and Lessons for the WTO (Ann Arbor, University of Michigan Press, 2003) at Part III; and Joerges and Petersmann (eds), above n 3 at section II. 200 Matsushita et al, above n 18 at 515. 201 Although see Winickoff et al, above n 49.

(H) Fisher Ch5

31/5/07

202

09:58

Page 202

Risk Assessment, The WTO and Administrative Constitutionalism

The second problem is that dispute settlement has largely been understood as requiring Member States to ‘prove’ that their standard has a scientific basis.202 This conceptualisation conflates the main question a Panel needs to be asking— has a Member State acted consistently with the Agreement?—with a secondary procedural question—how much evidence is necessary for a Member State to prove it has acted consistently? The SPS Agreement requirements have thus been reduced to the set of evidentiary requirements before a Panel. The AB has pointed out the problem with this203 but, despite this, SPS standard setting continues to be understood as the process of justification conducted in dispute settlement.204 The practical result with this can be seen in EC—Biotech Products, in which a Panel is faced with an overwhelming amount of material submitted by the parties. Again, this situation is due to the fact that SPS standard setting is understood as a factfinding exercise rather than an administrative exercise. Not only is risk evaluation not just to do with scientific analysis but the role of a Panel or AB is to assess the compatibility of a national measure with the SPS Agreement not just to assess its factual basis. The third problem with this body of jurisprudence is related to this in that the task of the Panel is presently understood as requiring it to assess whether a measure is based on ‘sufficient scientific evidence’.205 This requires a Panel to engage in a form of scientific analysis it is not qualified to carry out.206 Indeed, as seen above, in many cases the Panels defer to the judgment of experts they appoint, particularly in determining whether there was a risk assessment or evaluation of a risk.207 Again, this situation is a product of the science/democracy dichotomy in that the Panel being burdened with factual analysis is an inevitable consequence of understanding the SPS Agreement as imposing science. Moreover, it has resulted in there being a lack of appreciation that the SPS Agreement is open to divergent interpretations and that the type of ‘reasonable’ action it dictates is not fixed.208 The final problem with the approach seen above is that it has contributed to the schism that already exists between transnational trade governance and risk regulation. Science is understood to be the rationality and vocabulary of trade regulation while democracy is understood as the rationality and vocabulary of risk regulation. This is deeply problematic when one considers that dispute settlement 202 S Harlow, ‘Science-Based Trade Disputes: A New Challenge in Harmonizing the Evidentiary Systems of Law and Science’ (2004) 24 Risk Analysis 443; Christoforou, above n 28; and the questions asked in the ‘dumb duck disease’ case study in Cottier and Mavroidis (eds), above n 199 at the Appendix. 203 Appellate Body Report, above n 2 at paras 101–8 and 132–43. 204 Appellate Body Report, Japan—Measures Affecting Agricultural Products, above n 30 at para 141. 205 Bohanes, above n 45; and J Peel, ‘Risk Regulation Under the WTO SPS Agreement: Science as an International Normative Yardstick’ (2004) Jean Monnet Working Papers 02/04. 206 von Bogdandy, above n 80 at 641–3; and Cottier and Mavroidis (eds), above n 199 at 5. 207 Panel Report, above n 125 at paras 8.75, 8.79, and 8.87; and Panel Report, Japan—Measures Affecting Agricultural Products, above n 74 at para 8.32. See also Christoforou, above n 28 at 624, and Cottier and Mavroidis (eds), above n 199 at 5. 208 Although see Perez, above n 42.

(H) Fisher Ch5

31/5/07

09:58

Page 203

Future Lines of Inquiry

203

is a new forum for debating administrative constitutionalism and there needs to be far greater scrutiny of the implications of this for national risk regulation regimes beyond dispute settlement infusing trade ‘concerns’ into risk regulation debates. Thus, for example, WTO dispute settlement is perceived as a new source of legal authority. This can be seen from the European Commission’s interpretation of the precautionary principle, which reflects the RI thinking of the Panel’s decision in EC—Hormones.209 Dispute settlement decisions are not only having an effect on national and supranational policy but also on judicial review doctrine210 and the institutional structures of national risk regulation standard setters.211

VI Future Lines of Inquiry It is useful to ponder why the WTO jurisprudence has been so influenced by the science/democracy dichotomy. Broadly speaking, there are at least three possible reasons, all of which have already been discussed. The first is that the focus on science is a product of understanding the Panels’ ‘standard of review’ as requiring them to ‘engage in an objective assessment of facts’. The second is that the approach is due to Panels and the AB wishing to make the WTO dispute settlement process as neutral as possible. The third possible reason is the domination of the science/democracy dichotomy in the academic and policy literature.212 Panels and the AB have understood their task in such terms because that has been the prevailing ethos, and in a legal culture with no tradition of administrative constitutionalism such a characterisation makes sense.213 What all this highlights is that the failure to recognise that the SPS Agreement raises issues of administrative constitutionalism is to do with issues embedded deep within WTO legal culture. The promotion of the RI paradigm has been to do more with the reasons given above than the Panels and the AB following the reasoning set down by the Panel in EC—Hormones, and to argue that the present jurisprudence promotes certain understandings of administrative constitutionalism would make little sense to most of those who practise and study this area of law. This means that if there is to be a reorientation of WTO jurisprudence and scholarship it is not simply a matter of looking in different directions. It also must 209 Commission of the European Communities, Communication from the Commission on the Precautionary Principle, COM (2000) 1 final. See ch 6 below. 210 Case T-13/99 Pfizer Animal Health SA v Council [2002] ECR II-3305. For an analysis of this see E Fisher, ‘Precaution, Law and Principles of Good Administration’ (2005) 52 Water Science and Technology 19, and ch 6 below. 211 The European Food Safety Authority: see Commission of the European Communities, White Paper on Food Safety, COM (1999) 719 final. 212 See section I above. 213 Harlow, above n 202, and E Anderson and C St Hilaire, ‘The Contrast Between Risk Assessment and Rules of Evidence in the Context of International Trade Disputes: Can the US Experience Inform the Process?’ (2004) 24 Risk Analysis 449.

(H) Fisher Ch5

31/5/07

204

09:58

Page 204

Risk Assessment, The WTO and Administrative Constitutionalism

mean a change in mindsets. While I make no attempt in this chapter to provide a comprehensive framework for how such a culture shift would occur, what I can do is suggest two future lines of inquiry. The first concerns the WTO regime and the role that dispute settlement plays within that regime. There needs to be recognition of the fact that the SPS Agreement is not a neutral document, and dispute settlement is not just about an objective assessment of the facts. The Agreement’s interpretation and the resolution of disputes both require normative choices to be made. Likewise, there is a mutually constitutive interrelationship between paradigms of administrative constitutionalism, risk regulation regimes, WTO trade law, and WTO dispute settlement proceedings in which each are co-produced alongside the others. Thus, as seen above, understandings about the trade law purposes of the SPS Agreement are co-produced with paradigms of administrative constitutionalism. Implicit in this recognition of co-production must be a recognition of the thickness of WTO legal culture, and dispute settlement can no longer be understood as a practical compromise between the ‘security’ of the rule of law and the ‘friendly accommodation’ of diplomatic dispute resolution.214 Dispute settlement is creating a substantive body of normative rules, and Panels and the AB are not neutral arbiters but are regulating understandings of good administration. This is consistent with what other commentators are saying more generally about WTO dispute settlement, and thus there is nothing particularly radical in such an argument.215 In relation to the SPS Agreement it means that there needs to be a rethink of the roles of the Panel and the AB in relation to dispute settlement. Those roles are essentially assessing whether there has been consistency with the Agreement and, in this sense, their task is akin to, although different from, a national court carrying out judicial review of administrative action.216 It is akin in the sense that such review raises issues of relative institutional and ‘constitutional’ competence 217 as well as acting as an arena for debating administrative constitutionalism. It is different because such review will be shaped by understandings of both horizontal and vertical competence and be bounded by principles of trade liberalisation.218 As this is the case, then there needs to be greater consideration of those different competences and what the limits of AB and Panel powers are in reviewing SPS measures. Such a rethink is likely to be an uncomfortable one, highlighting as it 214 D McRae, ‘What is the Future of WTO Dispute Settlement?’ (2004) 7 Journal of International Economic Law 3 at 7. 215 Ibid; J Greenwald, ‘WTO Dispute Settlement: An Exercise in Trade Law Legislation?’ (2003) 6 Journal of International Economic Law 113; and J Pauwelyn, ‘The Sutherland Report: A Missed Opportunity for Genuine Debate on Trade, Globalisation and Reforming the WTO’ (2005) 8 Journal of International Economic Law 329. 216 For a paper that makes that argument see Peel, above n 205. 217 C Sunstein, Designing Democracy: What Constitutions Do (New York, Oxford University Press, 2001); and Jowell, above n 9. 218 For an interesting discussion of this in relation to standards of review see Oesch, above n 69 at ch 2.

(H) Fisher Ch5

31/5/07

09:58

Page 205

Future Lines of Inquiry

205

does the fact that both bodies are playing a normative role in determining how national risk regulators should be constituted, limited, and held to account. The legitimacy of such a role may be questionable and it is not enough, as some have done, to argue for a procedure-based approach to interpreting the Agreement, because that is just replacing the neutral paradigm of science with the neutral paradigm of procedure.219 The second aspect of recognising that the SPS Agreement regulates administrative constitutionalism is to do with the interrelationship between WTO law and national risk regulation regimes. As an arena for disputing administrative constitutionalism, WTO dispute settlement is operating across jurisdictional boundaries, which brings with it challenges to how legal norms and legal culture are understood.220 As already noted, this may be illegitimate to some, but it is also an inevitable consequence of globalisation. Indeed, WTO dispute settlement is an example of the fact that globalisation brings with it a need to reconcile different understandings of administrative constitutionalism derived from different legal cultures. There may be a tendency to try and eradicate such a plurality of theories of administrative constitutionalism but, as seen throughout this book, heterogeneity is an omnipresent feature of risk regulation and not only is it not easily eradicated but it is also arguably undesirable to do so. At the same time, the challenge of even finding a common discourse for these different understandings of administrative constitutionalism should not be underestimated. Thus, while commentators such as Howse might quite rightly point to the need of understanding the SPS Agreement as ‘enhancing’ the quality of rational deliberation about risk regulation, it needs to be appreciated that such a process of enhancement will not start or end with a uniform understanding of the issues.221 Related to these concerns is the need to think more carefully about how debates over the legitimacy of standard-setting regimes at the national and transnational level interact. Transnational standard-setting regimes will also be shaped by debates over administrative constitutionalism, and the question is how such debates interact with national discourses. This is something that is beyond the scope of this chapter but the importance of it can be seen in the disputes considered. Thus, for example, in Australia—Salmon both the Panel and the AB deferred to the OIE’s definition of risk assessment,222 yet that definition could be understood as a product of the OIE’s own legitimacy concerns as a transnational administrative organisation. In this sense there is an important relationship between the constitutionalisation process at the transnational level and at the national level, 219 For those arguments see Bohanes, above n 45; Winickoff et al, above n 49 at 90; and Guzman, above n 45 at 4. 220 A point that has been made about environmental principles: see N de Sadeleer, Environmental Principles: From Political Slogans to Legal Rules (Oxford, Oxford University Press, 2002). 221 Howse, above n 49. 222 Panel Report, above n 125 at para 8.48; and Appellate Body Report, Australia—Measures Affecting Importation of Salmon, above n 30 at para 123.

(H) Fisher Ch5

31/5/07

206

09:58

Page 206

Risk Assessment, The WTO and Administrative Constitutionalism

although it is one that is not about democratic constitutionalism but administrative constitutionalism.223 Highlighting these two aspects of understanding the SPS Agreement in terms of administrative constitutionalism does not provide any simple formulae for thinking about the SPS Agreement, WTO jurisprudence, or the interrelationship between the WTO and national risk regulation. What it does do, however, is provide more productive lines of inquiry for thinking about these issues than those presently provided for by the science/democracy dichotomy. Moreover, and perhaps more significantly, it also highlights that the challenges of globalisation are not likely to be resolved with reference to science.

VII Conclusion As stressed in the introduction to this chapter, my aim in examining these decisions has not been to provide a comprehensive analysis of WTO jurisprudence but rather to reorientate understandings about the role and nature of the SPS Agreement. My argument in this chapter has been that the present understanding of the WTO SPS Agreement in terms of the science/democracy dichotomy is incorrect and unhelpful. Rather, that Agreement needs to be understood as regulating administrative constitutionalism. In putting this argument I am not saying that there is a need to recognise my theories but rather something far more modest—that there is a need to recognise that the Agreement regulates understandings of good public administration. If that point is appreciated then a whole new agenda is opened up in the study of the WTO SPS Agreement.

223

See ch 6 below.

(I) Fisher Ch6

31/5/07

09:59

Page 207

6 The Precautionary Principle and Administrative Constitutionalism in the European Union: Asking Some Difficult Questions* In this last case study I examine the development and operation of the precautionary principle as a jurisprudential principle in the European Union (EU) through the lens of administrative constitutionalism. My aim in doing so is to provide an example of how supranational integration raises some difficult questions about administrative integration. In particular, I wish to make two points. First, the operation of the precautionary principle in the EU context highlights a complex and interrelated web of diverse administrative institutions, and thus discourses over administrative constitutionalism. Second, at the same time, European integration, due to overlaps and interrelationships between different spheres of operation, promotes a single definition of the precautionary principle and thus a single understanding of administrative constitutionalism. The recognition of this tension raises some difficult questions for European integration and administrative constitutionalism. The structure of this chapter is as follows. In the first section, I contextualise the precautionary principle in the EU context by showing that it is operating in at least six fundamentally different legal contexts in the EU. The implication of this is that a study of the principle highlights the diversity of administrative institutions and of administrative constitutionalism in the EU. In the second section, however, I show that overlaps, interrelationships, and the operation of common ideas of good administration have resulted in a uniform understanding of the principle. In the third, fourth and fifth sections I provide a short case study of this tension. In the third section, I consider how the European Court of Justice (ECJ) and the Court of First Instance (CFI) reviewed precautionary reasoning of Community institutions and Member States before the publication of the European Commission’s * An earlier and shorter version of this chapter can be found in E Fisher, ‘Opening Pandora’s Box: Contextualising the Precautionary Principle in the European Union’ in E Vos et al (eds), Uncertain Risks Regulated: National, EU and International Regulatory Models Compared (London, RoutledgeCavendish, forthcoming).

(I) Fisher Ch6

31/5/07

09:59

Page 208

208 Precautionary Principle and Administrative Constitutionalism in the EU Communication on the Precautionary Principle (the Communication) in 2000.1 Much of that case-law was based on deliberative-constitutive (DC) understandings of administrative constitutionalism, but the ECJ had developed different approaches depending on the context. In the fourth section, I examine the Commission’s Communication to show how it is a document that promotes a rational-instrumental (RI) understanding of administrative constitutionalism for Community administrative institutions. In the fifth section, I will show that since its publication it has been used as a basis for promoting a uniform RI definition of the precautionary principle in all contexts. In the last section, I reflect on this state of affairs. In particular, I show that a study of the precautionary principle through the lens of administrative constitutionalism highlights some difficult questions about administrative constitutionalism and administrative integration in the EU. A number of caveats should be made before starting. First, in thinking about the DC and RI paradigms, it is important to remember that they are being used as shorthand to describe the different normative assumptions underlying divergent legal and policy positions. Second, this chapter does not contain a basic discussion of the precautionary principle, its relevance to administrative constitutionalism, and how it can be interpreted in DC and RI terms. That discussion can be found in chapter one.2 Third, it is important to keep in mind that this chapter is an exercise in reorientation rather than comprehensive analysis. My focus is mainly upon the development of the precautionary principle as a jurisprudential principle by the ECJ and CFI, but even then, this chapter is not an exhaustive examination of those cases.

I Contextualising the Precautionary Principle in the European Union In this chapter, my concern is to study the precautionary principle in the EU through the lens of administrative constitutionalism. Before doing so, it is useful to begin by restating a number of points. As discussed in chapter one,3 the precautionary principle is a principle concerned with administrative constitutionalism that regulates understandings of reasonable administrative discretion in circumstances of scientific uncertainty. As such, how the principle is defined and applied will vary from legal culture to legal culture.4 The principle can be inter-

1 Commission of the European Communities, Communication from the Commission on the Precautionary Principle, COM (2000) 1 final. 2 See section V of ch 1. 3 Ibid. 4 E Fisher, ‘Precaution, Precaution Everywhere: Developing a “Common Understanding” of the Precautionary Principle in the European Community’ (2002) 9 Maastricht Journal of European and Comparative Law 7.

(I) Fisher Ch6

31/5/07

09:59

Page 209

Contextualising the Precautionary Principle in the European Union

209

preted in both DC and RI terms.5 As a DC principle it is understood as an enabling principle which requires deliberation and the flexible exercise of discretion. As a RI principle it is a limited exception to action needing to have a firm factual basis which operates only when certain conditions precedent are met. The principle also applies to the exercise of both internal and external state sovereignty. From an internal perspective, the principle governs what is understood as legitimate administrative action in circumstances of scientific uncertainty. From an external perspective, the principle may be used as a reason to require sovereign states to take action, or as a reason for a sovereign state to derogate from its international law obligations. As seen in the previous chapter, in a different context, the operation of external state sovereignty also raises issues of administrative constitutionalism.6 The precautionary principle, in its present form, originated in West German environmental policy in the late 1970s,7 although similar types of philosophies can be seen in other Northern European Member States.8 On becoming a principle with international prominence, particularly in European regional agreements,9 the principle was included in the Treaty establishing the European Community (TEC) in 1992 as one of a number of principles that environmental policy ‘shall be based on’.10 Moreover, it has been officially recognised in the EU context to apply not just to environmental risks but also to health risks.11 What all of this has meant is that the precautionary principle has been developed in parallel in the legal cultures of Member States and the EU.12 From an EU perspective, the inclusion of the principle occurred at a time when EU technological risk regulation was being consolidated. The rise of the precautionary principle and the rise of EU risk regulation regimes have thus occurred in tandem.13 5 Ch 4 above; and E Fisher and R Harding, ‘The Precautionary Principle and Administrative Constitutionalism: The Development of Frameworks for Applying the Precautionary Principle’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). 6 See ch 5. 7 For the its operation in West German law see K von Moltke, ‘The Vorsorgeprinzip in West German Environmental Policy’ in Royal Commission on Environmental Pollution (ed), Best Practicable Environmental Option—12th Report (London, HMSO, 1988); and S Boehmer-Christiansen, ‘The Precautionary Principle in Germany: Enabling Government’ in T O’Riordan and J Cameron (eds); Interpreting the Precautionary Principle (London, Earthscan Publications, 1994). 8 B Wahlstrom, ‘The Precautionary Approach to Chemicals Management: A Swedish Perspective’ in C Raffensberger and J Tickner (eds). Protecting Public Health and the Environment: Implementing the Precautionary Principle (Washington DC, Island Press, 1999). 9 R Harding and E Fisher, ‘Introducing the Precautionary Principle’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999) at 5. 10 Art 174(2) TEC. 11 Case T-74/00 Artegodan GmbH v Commission [2002] ECR II-4965 at para 184. Art 6 TEC is often used as a justification for the principle applying in other sectors. 12 A good example of this can be seen in O Godard, ‘The Precautionary Principle and Catastrophism on Tenterhooks: Lessons From Constitutional Reform in France’ in E Fisher et al (eds), Applying the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). 13 C Joerges, ‘Law, Science and the Management of Risks to Health at the National, European and International Level—Stories of Baby Dummies, Mad Cows and Hormones in Beef’ (2001) 7 Columbia Journal of European Law 1.

(I) Fisher Ch6

31/5/07

09:59

Page 210

210 Precautionary Principle and Administrative Constitutionalism in the EU Thus, while the EU has been regulating technological risks since the late 1960s,14 it was only in the 1990s that EU technological risk regulation regimes were explicitly understood as more than just accidental spillovers from market integration.15 In part this was to do with explicit Treaty recognition of the legitimacy of European risk regulation16 and in part it was to do with a growing sophistication of these institutions, including the development of multi-level authorisation regimes,17 an increased role for a range of different types of committees,18 the creation of new European agencies with limited powers,19 and the adoption of ‘new’ regulatory approaches.20 Integral to these developments was the exponential growth of a body of CFI and ECJ case-law.21 Before the mid 1990s most of the case-law concerned enforcement or Article 28 actions, but after that time both courts have increasingly been required to rule on the validity of Member State and Community action. The precautionary principle has been central to these developments and has increasingly been relied on by decisionmakers and has been the subject of litigation.22 Most significantly, it has been widely recognised as a defining principle of EU risk regulation,23 alongside the regulatory aim of a ‘high level of protection’.24 Indeed, critiques and evaluations of the precau14 Directive 67/548 on the classification, labelling and packaging of dangerous substances [1967] OJ 196/1. 15 E Vos, Institutional Frameworks of Community Health and Safety Legislation: Committees, Agencies and Private Bodies (Oxford, Hart Publishing, 1999) at ch 2; S Weatherill, EU Consumer Law and Policy (Cheltenham, Edward Elgar, 2005) at ch 1; and A Weale et al, Environmental Governance in Europe (Oxford, Oxford University Press, 2000) at Part 1. 16 For example: Art 175 TEC (environmental protection competence); Art 153 TEC (consumer protection competence); and Art 95(3) TEC (internal market measures should take as their base a high level of protection in relation to health, safety, environmental protection, and consumer protection). 17 That is, authorisation regimes where those being regulated apply to a Member State authority which then considers that application as part of a network of Member State and Community administrations. 18 Vos, above n 15; and G Schaäfer, ‘Linking Member State and European Administrations: The Role of Committees and Comitology’ in M Andenas and A Turk (eds), Delegated Legislation and the Role of Committees in the EC (The Hague, Kluwer Law International, 2000). 19 The European Food Safety Authority being the most obvious example: Regulation 178/2002 laying down the general principles and requirements of food law, establishing the European Food Safety Authority and laying down procedures in matters of food safety [2002] OJ L31/1. 20 E Fisher, ‘Unpacking the Toolbox: Or Why the Public/Private Divide is Important in EC Environmental Law’ in M Freedland and J-B Auby (eds), The Public Law/Private Law Divide: Une entente assez cordiale? (Oxford, Hart Publishing, 2006); and J Scott and J Holder, ‘Law and New Environmental Governance in the European Union’ in G De Búrca and J Scott (eds), Law and New Governance in the EU and the US (Oxford, Hart Publishing, 2006). 21 L Azoulay, ‘The Court of Justice and the Administrative Governance’ (2001) 7 European Law Journal 425; and H Somsen, ‘Discretion in European Community Environmental Law: An Analysis of ECJ Caselaw’ (2003) 40 Common Market Law Review 1413. 22 N de Sadeleer, ‘The Precautionary Principle in EC Health and Environmental Law’ (2006) 12 European Law Journal 139; and J da Cruz Vilaca, ‘The Precautionary Principle in EC Law’ (2004) 10 European Public Law 369. 23 P Craig, EU Administrative Law (Oxford, Oxford University Press, 2006) at ch 19; K Lenaerts, ‘“In the Union We Trust”: Trust Enhancing Principles of Community Law’ (2004) 41 Common Market Law Review 317; and T Christoforou, ‘The Origins and Content of the Precautionary Principle in European Community Law’ in C Leben and J Verhoven (eds), Le Principe de Precaution—Aspects de Droit International et Communautaire (Paris, Pantheon Assas, LGDJ Diffuseur, 2002). 24 Arts 2, 95(3), 153(1) and 174(2) TEC.

(I) Fisher Ch6

31/5/07

09:59

Page 211

Contextualising the Precautionary Principle in the European Union

211

tionary principle are invariably understood as critiques and evaluations of European risk regulation writ large.25 Despite this co-evolution of EU risk regulation and the precautionary principle there has been very little examination of the interrelationship between the two. Rather, the principle has tended to be treated as a free-standing principle, the legitimacy of which depends on its own internal logic rather than upon how it operates in particular institutional contexts.26 Thus, for example, de Sadeleer has described the principle as one of a series of directing principles that lie at the interface between modern and post-modern law which emphasises the ‘pragmatic, gradual, unstable, and reversible nature of rules’.27 Moreover, a crucial feature for de Sadeleer of post-modern law is the ‘mingling’ of norms from different legal systems.28 The problem with this approach is that, while it is true that principles such as the precautionary principle are crossing legal borders, they are operating very differently in distinct legal cultures and divergent institutional contexts.29 In relation to the precautionary principle this has to do with the fact that the principle is a principle concerned with administrative constitutionalism, and administrative constitutionalism is part of the thickness of legal culture. Indeed, studying the operation of the precautionary principle in the EU through the lens of administrative constitutionalism highlights that the principle is being applied in many different but interrelated contexts. This is because EU technological risk evaluation regimes involve a range of different Community and Member State institutions exercising discretion. Moreover, while some of these institutions are democratic legislative bodies, most are administrative in nature due to the institutional competence required for technological risk evaluation.30 Thus, at the Member State level it is mainly public administration that is involved in the detailed implementation of Community law.31 Moreover, at the Community level it is the Commission, committees, and agencies that take a central role in technological risk evaluation.32 In each case, however, each institution is operating in 25 G Marchant and K Mossman, Arbitrary and Capricious: The Precautionary Principle in the European Union Courts (Washington DC, The AEI Press, 2004); O Renn et al, The Application of the Precautionary Principle in the European Union: Regulatory Strategies and Research Needs to Compose and Specify a European Policy on the Application of the Precautionary Principle (PrecauPri) (Stuttgart, Centre for Technology Assessment, 2003); P van Zwanenberg and A Stirling, ‘Risk and Precaution in the US and Europe’ (2004) 3 Yearbook of European Environmental Law 43; and L Kogan, Exporting Precaution: How Europe’s Risk Free Regulatory Agenda Threatens American Free Enterprise (Washington DC, Washington Legal Foundation, 2005). 26 C Sunstein, Laws of Fear: Beyond the Precautionary Principle (Cambridge, Cambridge University Press, 2005); and G Majone, ‘What Price Safety? The Precautionary Principle and its Policy Implications’ (2002) 40 Journal of Common Market Studies 89. 27 N de Sadeleer, Environmental Principles: From Political Slogans to Legal Rules (Oxford, Oxford University Press, 2002) at 371. 28 Ibid at paras 248–9. 29 Fisher, above n 4. 30 See section II.A of ch 1. 31 C Demmke and M Unfried, European Environmental Policy: The Administrative Challenge for Member States (Maastricht, European Institute of Public Administration, 2001). 32 Vos above n 15.

(I) Fisher Ch6

31/5/07

09:59

Page 212

212 Precautionary Principle and Administrative Constitutionalism in the EU different legal contexts in that they are being governed by different rules concerning how they should be constituted, limited, and held to account. This is best illustrated by briefly mapping some of those different contexts. At least six can be identified. The first context is where the precautionary principle is applied by Community institutions in exercising their power. Community institutions may rely on the principle in the creating of a new legislative/regulatory scheme33 or, more likely, in the exercise of power in relation to a specific product or activity.34 In regards to the latter, the principle may be utilised by a range of different institutions, including the Commission, committees, and agencies.35 As noted already, most of these institutions are administrative in nature and they are subject to a discourse of administrative constitutionalism where those debates relate more generally to issues in the wider integration and constitutionalisation process. An important aspect of that discourse is the process of limiting and holding decision-makers to account by the CFI and ECJ.36 The second context in which the precautionary principle is operating is when it is being applied by Member State administrations that are operating pursuant to a Community regulatory regime. This may either be in implementing EU standards into national law or by a Member State authority operating pursuant to a Community-wide authorisation scheme. In such circumstances, the principle may be relevant to how Member State authorities choose to interpret a directive37 and/or to the type of discretion that they have pursuant to that directive.38 In most cases, the process of implementation is being carried on by national public administrations that will be subject to both national principles of administrative constitutionalism and EU legal principles relating to implementation.39 The third context is closely related to the second category and involves cases where there is a Community regulatory regime but a Member State wishes to rely on the principle in derogating from the obligations of that regime. This may be pursuant to specific provisions of a directive or regulation40 or more general opt-out provisions of the TEC such as Articles 95(4) and 176.41 33 Regulation No 2371/2002 on the conservation and sustainable exploitation of fisheries resources under the Common Fisheries Policy [2002] OJ L358/59. 34 Art 7 of Regulation No 178/2002, above n 19. 35 It is important to note that because of the rules concerning delegation in the EU these decisions will nearly always be attributed to the Council: see Case 9/56 Meroni v ECSC High Authority [1957–8] ECR 133 36 Craig, above n 23 at Part 2; and C Harlow, Accountability in the European Union (Oxford, Oxford University Press, 2002) at ch 6. 37 Case C-318/98 Fornasar [2000] ECR I-4785 at para 37; and Case C-9/00 Palin Granit OY and Vehmassalon Kansanterveystyön Kuntayhtymän Hallitus v Lounais-Suomen Ympäristökeskus [2002] ECR I-3533 at para 23. 38 Case C-236/01 Monsanto Agricoltura Italia SpA v Presidenza del Consiglio dei Ministri [2003] ECR I-8105 at paras 110–12; and Case C-127/02 Landelijke Vereniging tot Behoud van de Waddenzee v Staatssecretaris van Landbouw, Natuurbeheer en Visserij [2004] ECR I-9405. 39 Case C-293/97 R v Secretary of State for the Environment, ex parte Standley [1999] ECR I-2603; and Case C-318/98 Fornasar, above n 37. 40 Case C-6/99 Association Greenpeace France v Ministère de l’Agriculture et de la Pêche [2000] ECR I-1651. 41 Case C-3/00 Denmark v Commission [2003] ECR I-2643; and Case T-366/03 Land Oberosterreich v Commission [2005] ECR II-4005.

(I) Fisher Ch6

31/5/07

09:59

Page 213

Contextualising the Precautionary Principle in the European Union

213

The fourth context in which the precautionary principle operates is where it is applied by Member States where there is no Community regulatory regime but application conflicts with general Community obligations. The most obvious example here is where such application prima facie infringes Article 28 TEC.42 The legal question which arises is whether a Member State is justified in taking such action under either Article 30 TEC or the Cassis mandatory requirements doctrine.43 In such cases a Member State must establish that its restrictions shall not ‘constitute a means of arbitrary discrimination or a disguised restriction on trade between Member States’.44 Broadly speaking, this requires a Member State to establish that the measure is necessary, appropriate, and proportional. As seen in chapter five, what is ‘necessary, appropriate and proportional’ will depend on understandings of reasonable administrative action. Where those principles are derived from in the EU will be discussed in more detail below. The fifth context in which the precautionary principle is being applied is where Community institutions are carrying out their international obligations. This may include where the principle is being used to reinforce those obligations45 or, more controversially, where the principle is being used to derogate from those obligations.6 In most cases, such decisions will actually be exercises of internal Community power and thus the decisions will be of committees, agencies, the Commission, Parliament, and the Council.47 It is international agreements that will regulate what is understood by reasonable action, however.48 Thus, for example, the World Trade Organisation (WTO) Sanitary and Phytosanitary Agreement (SPS Agreement) is understood to regulate the reasonableness of regulatory action in relation to measures that protect human, plant, or animal health. This means that it is the understandings of administrative constitutionalism that are taken to be embodied in that Agreement which governs the exercise of Community power.49 The sixth and final category in which the principle is operating is where the principle is being applied by Member States in matters that bear no relationship 42 Case C-95/01 Criminal Proceedings Against Greenham and Abel [2004] ECR I-1333; and Case C-41/02 Commission v Netherlands [2004] ECR I-11375. 43 Case 120/79 Rewe-Zentral AG v Bundesmonopolverwaltung für Branntwein [1979] ECR 649. 44 Art 30 TEC; and ibid. 45 Cartagena Protocol on Biosafety, Protocol to the Convention on biological diversity signed in 2000. 46 Panel Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), WT/DS26/R/USA, 18 August 1997; and Appellate Body Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), WT/DS26/AB, 16 January 1998. 47 Also note that in some cases the action will be that of a Member State: see Panel Report, European Communities—Measures Affecting the Approval and Marketing of Biotech Products, WT/DS291-3/R, 29 September 2006. 48 Appellate Body Report, Australia—Measures Affecting Importation of Salmon, WT/DS18/AB/R, 20 October 1998. 49 See ch 5 above. An important related issue is the direct effect of WTO law in EU law: see Case C-149/96 Portugal v Council [1999] ECR I-8395. For discussion see G de Búrca and J Scott, ‘The Impact of the WTO on EU Decision-Making’ in G de Búrca and J Scott (eds), The EU and the WTO: Legal and Constitutional Issues (Oxford, Hart Publishing, 2001).

(I) Fisher Ch6

31/5/07

09:59

Page 214

214 Precautionary Principle and Administrative Constitutionalism in the EU with EU law. This will include areas such as town planning and will often be in the context of long established statutory regimes.50 This category is a large one, particularly because, in a number of Member States, the principle is well entrenched. Within these legal cultures, the principle gives rise to a rich and complex debate over administrative constitutionalism.51

II Overlaps, Interrelationships, and Administrative Integration The six categories above make an unwieldy set of groupings and what is immediately clear is that it is difficult to make generalisations about the principle. In each different category it is applying to a different set of ‘administrative’ institutions and in doing so giving rise to very distinct debates over administrative constitutionalism. It is also important to remember that references to the operation of discretion by Member States are references to 27 separate legal cultures. Moreover, in each context there is a multitude of different regulatory regimes and an array of different administrative institutions. Likewise, in talking about the international context there are also numerous different legal cultures in operation. To talk of the precautionary principle in light of all the above seems foolishly naïve. Likewise, those that have criticised the variation of the operation of the precautionary principle for showing it is an ‘arbitrary’ principle have failed to appreciate these different contexts.52 With that said, Member States and Community institutions are not acting independently or in isolation in these different contexts. Rather, the application of the precautionary principle in each context is part of a web of overlapping and interrelated institutions. Indeed, the consolidation of EU risk regulation regimes is really about the growing complexity of this web rather than the greater gaining of power at the Community level.53 These overlaps and interrelationships can be seen in a number of ways. To begin with, the same institution can find itself subject to two different discourses over administrative constitutionalism operating in different legal cultures. Thus, for example, Community institutions will invariably be subject to both WTO law and Community law in the evaluation and regulation of SPS risks.54 Likewise, 50 R (on the Application of Thomas Bates & Son Ltd) v Secretary of State for Transport Local Government and The Regions [2005] 2 P & CR 11; and R (on the application of Davies) v Carmarthenshire County Council [2004] EWHC 2847. 51 Godard, above n 12; E Fisher, ‘Is the Precautionary Principle Justiciable?’ (2001) 13 Journal of Environmental Law 317; and R Macrory (ed), Principles of European Environmental Law (Groningen, Europa Law Publishing, 2004). 52 Marchant and Mossman, above n 25; and Sunstein, above n 26 at 20–3. 53 Vos, above n 15, at ch2; and Weale et al, above n 15 at Part 1. 54 A good example is Appellate Body Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), above n 46; and Case C-331/88 R v Minister of Agriculture, Fisheries and Food and Secretary of State for Health, ex parte FEDESA [1990] ECR I-4023. See the analysis in

(I) Fisher Ch6

31/5/07

09:59

Page 215

Overlaps, Interrelationships, and Administrative Integration

215

taking unilateral action when the Community has not regulated will mean that a decision-maker is governed by both understandings of administrative constitutionalism embodied in Article 28 jurisprudence and by understandings of administrative constitutionalism operating within its own legal culture.55 Thus, for example, in Kemikalienspektionen v Toolex Alpha AB,56 a decision by Sweden to ban a certain chemical was entirely consistent with the principles of administrative constitutionalism operating in that legal culture. The question for the ECJ, however, was whether Sweden’s regulatory action was consistent with what were understood to be as valid administrative constitutionalism norms for Member States to rely on in derogating from Article 28. Moreover, a Member State acting pursuant to, or attempting to derogate from, a Community regulatory scheme is particularly problematic because it is not obvious whether the analytical starting point should be ideas of national or Community administrative constitutionalism.57 Following on from this, there is also a close interrelationship between these different exercises of administrative power. Thus, for example, because the exercise of Community power in this area is so heavily dependent on the operation of comitology committees, it means that Community administrative action often has as its impetus Member State action.58 Likewise, because many regulatory regimes are in the form of shared administration, they require the concurrent exercise of administrative power by different Member States and Community institutions.59 Moreover, at the Community level there are a range of different institutions including committees, agencies, and the Commission all operating in tandem. A good example of these interrelationships can be seen in the decision-making processes which were subject to review in Pfizer v Council (the Pfizer decision),60 one of the first CFI decisions to examine the precautionary principle in detail.61 The case concerned a regulation that banned the use of certain antibiotics in animal feeding stuffs because of the potential but unproven risk that such use may lead to antibiotic resistance in humans. The regulation was the result of a decisionmaking process under Directive 70/524/EEC concerning additives in feeding stuffs.62 The process was begun by Denmark (followed later by other Member States) exercising its right to apply a safeguard clause under Article 11 of that Directive by putting forward a dossier to the Standing Committee on Animal Feedingstuffs, a comitology committee.63 Further information was collected, the N McNelis, ‘The Role of the Judge in the EU and WTO: Lessons from the BSE and Hormones Cases’ (2001) 4 Journal of International Economic Law 189. 55 Case C-95/01 Greenham and Abel, above n 42 56 Case C-473/98 Kemikalienspektionen v Toolex Alpha AB [2000] ECR I-5681. 57 F Bignami, ‘Foreword’ (2004) 68 Law and Contemporary Problems 1. 58 Case C-434/02 Arnold André GmbH & Co KG v Landrat SWA Kreises Herford [2004] ECR I-4023; and Case C-154/04 R (on the application of Alliance for Natural Health) v Secretary of State for Health [2005] ECR I-6451. 59 See the regulatory action reviewed in Case C-6/99 Association Greenpeace, above n 40. 60 Case T-13/99 Pfizer Animal Health SA v Council [2002] ECR II-3305. 61 See Section V below for a detailed discussion of the case. 62 [1998] OJ L 351/ 4 (as amended). 63 Art 24 of the Directive; and Case T-13/99 Pfizer, above n 60 at paras 29–59.

(I) Fisher Ch6

31/5/07

09:59

Page 216

216 Precautionary Principle and Administrative Constitutionalism in the EU Scientific Committee on Animal Nutrition was consulted, and ultimately the Standing Committee voted on the measure. What can thus be seen is that the ultimate regulation was the product of decision-making by a series of interrelated administrative institutions. This is even more the case when one considers the wider set of institutions which were also involved in identifying antibiotic resistance as a problem.64 It is not only overlaps and interrelationships that can be seen between these different contexts of application. What can also be seen is that a side-product of integration is that the same norms are being applied to different contexts. Thus, for example, the Commission in assessing the validity of Member State action under Articles 95(4) and 95(5) TEC relies on the norms develop by the ECJ in Article 28 jurisprudence.65 Moreover, in Article 28 case-law the ECJ assesses Member State action against norms which are widely held in the Community and expressed in Community policy.66 Maduro has argued that this is due to the Court engaging in economic constitution building, and the Court has encouraged Member States to harmonise their regulatory approaches by promoting ‘majoritarianism’.67 This ‘majoritarianism’ is not only in relation to regulatory outcomes and general policies, however.68 It also concerns norms of administrative constitutionalism in that, as we will see below, the Court is scrutinising carefully the process by which risk regulations are set and assessing those processes against commonly understood ideas of administrative constitutionalism in the EU context. The focus of the Court in these decisions is upon the process of decision-making—the reasons for the decision, who the state administrative authority consulted, the information it relied on, and so on—to establish whether the state was exercising its power reasonably.69 Moreover, in making that determination the Court is clearly influenced by how similar decisions are being made in other Member States and at the Community level.70 In other words, the Court’s ‘majoritarian’ approach is in harmonising what is understood as a ‘reasonable’ administrative process on the part of Member States, taking into account ideas of administrative constitutionalism that exist in Member States and the Community. What all this means is that there is a tension in thinking about the precautionary principle in the EU. On the one hand, there is a need to engage with diversity— 64

Art 24 of the Directive; and Case T-13/99 Pfizer, above n 60 at paras 29–59. J Scott and E Vos, ‘The Juridification of Uncertainty: Observations on the Ambivalence of the Precautionary Principle within in the EU and WTO’ in C Joerges and R Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002) at 267–72. 66 Case C-2/90 Commission v Belgium [1992] ECR I-4431; and Case C-67/97 Anklagemyndigheden v Ditlev Bluhme [1998] ECR I-8033. 67 M Maduro, We, The Court: The European Court of Justice and the European Economic Constitution (Oxford, Hart Publishing, 1998). 68 Note that Maduro argues that the case-law concerning Member State risk regulation measures does not fit his thesis: ibid at 73–4. However, this is due only to the fact that he is focusing on outcome rather than processes. 69 Case 54/85 Mirepoix [1986] ECR 1067; Case 53/80 Koninklijke Kaasfabriek Eyssen BV [1981] ECR 409; and Case C-473/98 Toolex Alpha, above n 56. 70 Case 304/84 Muller [1985] ECR 1511; Case 247/84 Motte [1985] ECR 3887; Case 178/84 Commission v Germany [1987] ECR 1227; and Case 176/84 Commission v Greece [1987] ECR 1193. 65

(I) Fisher Ch6

31/5/07

09:59

Page 217

Overlaps, Interrelationships, and Administrative Integration

217

diversity of contexts, legal cultures, administrative institutions, and discourses of administrative constitutionalism. That diversity is made even more obvious when it is also remembered that Member States are simultaneously operating in two different legal cultures. On the other hand, however, these different contexts do interact, and integration encourages the promotion of uniform norms of administrative constitutionalism. The desire to recognise a common precautionary principle is thus not just wishful thinking on the part of commentators and policy-makers but rather is a by-product of the overlap and interrelationships between the different exercises of power within the EU. This tension between diversity and uniformity is, of course, the raison d’être of European lawyers.71 As, too, are the struggles between legal cultures for supremacy which are implicit in such a tension.72 The problem is that most of the focus has been on political, economic, and legal integration. A study of the precautionary principle shows that another important facet to integration is administrative integration, including the integration of norms of administrative constitutionalism.73 This fact has not gone ignored by commentators. Ladeur in reviewing the Pfizer decision noted, after considering the intergovernmental decision-making processes it involved, that: the multi-polar character of the European administrative system might even turn out to be well adapted to the setting up of such a network-like structure of decision-making, thus enabling learning processes, mutual comparison, and experimentation with different types of relationships and co-ordination.74

Ladeur’s comments suggest that the operation of the precautionary principle is an opportunity to explore more closely the nature of administrative power in the EU. I would wholeheartedly agree but would be more circumspect about the opportunities for ‘learning processes, mutual comparison, and experimentation’. My reasons for circumspectness can be seen in the next three sections. The discussion so far has been relatively abstract and what I now wish to do is chart the development of the ECJ’s and CFI’s case-law where these courts are reviewing decisions that have been based on precautionary reasoning or the precautionary principle.75 I say precautionary reasoning because before 2000 there was a body of 71 C Joerges and R Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002); M La Torre, ‘Legal Pluralism as Evolutionary Achievement of Community Law’ (1999) 12 Ratio Juris 182; and N Barber, ‘Legal Pluralism and the European Union’ (2006) 12 European Law Journal 306. 72 Brunner v European Union Treaty [1994] 1 CMLR 57; and J Weiler, The Constitution of Europe: Do the New Clothes Have an Emperor? (Cambridge, Cambridge University Press, 1999) at ch 9. 73 For a commentator who has highlighted the importance of this dimension see M Everson, ‘The Constitutionalisation of European Administrative Law: Legal Oversight of a Stateless Internal Market’ in C Joerges and E Vos (eds), EU Committees: Social Regulation, Law and Politics (Oxford, Hart Publishing, 1999), and M Everson, ‘Administering Europe’ (1998) 36 Journal of Common Market Studies 195. 74 K-H Ladeur, ‘The Introduction of the Precautionary Principle into EU Law: A Pyrrhic Victory for Environmental and Public Health Law? Decision-making Under Conditions of Complexity in Multilevel Political Systems’ (2003) 40 Common Market Law Review 1455 at 1479. 75 For obvious reasons this analysis does not consider the last two categories of the principle’s application identified above.

(I) Fisher Ch6

31/5/07

09:59

Page 218

218 Precautionary Principle and Administrative Constitutionalism in the EU case-law concerned with the validity of such reasoning but no explicit reference to the precautionary principle. A study of that case-law through the lens of administrative constitutionalism shows that over time the ECJ and CFI’s jurisprudence has shifted dramatically from the DC to the RI paradigm in relation to this area. A key tipping point in that shift was the Commission’s Communication on the principle in 2000. In the next section, I consider the case-law before the Communication; in the fourth section, I consider the Communication itself; and in the fifth section, I consider the case-law after the Communication.

III The Case Law Before the Commission’s Communication As already noted, it is only in the past decade that issues pertaining to technological risk regulation have begun to be heavily litigated before the ECJ and CFI. With that said, the ECJ, and, to a more limited extent, the CFI, were required to rule on the legal validity of precautionary action before that time, albeit most of this caselaw being from the 1990s. This arose in the first four contexts identified in Section I above—the review of Community action, the review of Member State implementation of Community laws, the review of Member States’ attempting to derogate from Community regimes, and the review of Member State unilateral action that prima face infringed Article 28—with most of the case-law falling into the last category. What can be seen is that, in each of these categories, both courts were generally guided by a DC understanding of administrative constitutionalism. The first category concerns the case-law in relation to Article 230 and, although the body of pre-2000 case-law is relatively small, what can be seen is that the ECJ and CFI understood technological risk regulation in DC terms. This is not surprising. Most of these legislative frameworks gave wide-ranging discretion to Community institutions, and the process of upgrading legislative measures was through comitology committees in consultation with scientific committees.76 Such an institutional framework emphasised deliberation over analysis. Three different assumptions can be seen in this case-law. First, the CFI and ECJ understood the regulation of such risks as driven by a set of normative concerns, and the responsibility of Community institutions was, in their terms, a ‘political’ one.77 In relation to this there was also awareness that Community legislation was pursuing a series of different prescriptive aims.78 With that said, a great deal of emphasis in these cases was put on the public health protection objective of many

76

Vos, above n 15. Case C-331/88 FEDESA, above n 54 per Mischo AG; and Case T-125 & 152/96 Boehringer v Council and Commission [1999] ECR II-3427 at para 84. 78 Case 68/86 United Kingdom v Council [1988] ECR 855 at para 12. 77

(I) Fisher Ch6

31/5/07

09:59

Page 219

The Case Law Before the Commission’s Communication

219

of these directives.79 Second, in this case-law it is obvious that they were acutely aware of the different dimensions of scientific and behavioural uncertainty and that such uncertainty made factual assessment difficult.80 The existence of safeguard and technical progress provisions in most technological risk regulation directives made it obvious to the courts that information about technological risks was dynamic. Third, the courts were also aware of the need for Community institutions to deliberate with other Community institutions and the institutions of Member States due to the normative and factual complexity of problems.81 Such deliberation also included a process of reflection.82 The courts were also acutely aware that such deliberation and reflection was occurring in a context where Member States and different Community institutions had divergent opinions about the nature of the risks.83 There were two main implications of this approach to review. The first is that the CFI and ECJ limited the scope of their review.84 The second is that Community institutions did not need to prove the factual basis of their decisions to establish the legal validity of such decisions. Thus, for example, in R v Minister of Agriculture, Fisheries and Food and Secretary of State for Health, ex parte FEDESA the applicants argued that there was evidence that the five hormones being banned were in actual fact safe. The Court noted that: faced with divergent appraisals by the national authorities of the Member States, reflected in the differences between existing national legislation, the Council remained within the limits of its discretionary power in deciding to adopt the solution of prohibiting the hormones in question, and respond in that way to the concerns expressed by the European Parliament, the Economic and Social Committee and by several consumer organizations.85

In other words, the ECJ was making clear that the authority of the Council was not arising due to there being factual evidence of a risk but rather due to its responsibility to address a particular problem. Moreover, what was important in these decisions was that Community institutions considered and reflected upon the views of different institutions.86 A significant problem with these cases, however, is the ECJ, or the Advocate Generals (AGs), while recognising the problems of scientific uncertainty and the 79 Case C-331/88 FEDESA, above n 54 per AG Mischo; and Case T-199/96 Laboratories Pharmaceutiques Bergaderm SA v Commission [1998] ECR II-2805 at para 64. 80 Case T-125 & 152/96 Boehringer, above n 77 at para 90; Case T-199/96 Bergaderm, above n 79 at para 66; Case C-331/88 FEDESA above n 54 at paras 15–16; Case C-180/96 United Kingdom v Commission [1998] ECR I-2265 at para 63; and Case C-284/95 Safety Hi-Tech [1998] ECR I-4301 at paras 52–3. 81 Case T-105/96 Pharos v Commission [1998] ECR II-285 at paras 69–70 and 74–7; and Case 68/86 United Kingdom v Council, above n 78 per Lenz AG at para 51. 82 Case T-199/96 Bergaderm, above n 79 at para 55. 83 This can be seen in Case C-331/88 FEDESA, above n 54, and Case T-105/96 Pharos, above n 81. 84 Case C-331/88 FEDESA, above n 54 at para 8 and Case C-284/95 Safety Hi-Tech, above n 80 at para 37. 85 Case C-331/88 FEDESA, above n 54 at para 9. 86 Case T-199/96 Bergaderm, above n 79; and Case C-331/88 FEDESA above n 54.

(I) Fisher Ch6

31/5/07

09:59

Page 220

220 Precautionary Principle and Administrative Constitutionalism in the EU complexity of technological risk problems did not engage in any lengthy analysis of how these problems affected administrative discretion or how the courts carried out their review.87 Thus, for example, the ECJ stated in United Kingdom v Commission that: Where there is uncertainty as to the existence or extent of risks to human health, the institutions may take protective measures without having to wait until the reality and seriousness of those risks become fully apparent.88

It provided very little explanation of why this was the case, however, or of the implications for either administrative discretion or the role of the courts. As such, this body of case law, while clearly promoting the DC paradigm, did not represent a very sophisticated discourse concerning administrative constitutionalism. In particular, it has left some commentators with a concern that neither the ECJ nor the CFI are effectively policing Community power.89 This is not surprising in that many of the regimes the ECJ and CFI were reviewing had developed in an ad hoc way with little discourse about their role or development.90 The second category of cases in which the ECJ considered precautionary reasoning is that of challenges to Member State implementation of Community law, where a Member State attempted to take precautionary action in the face of scientific uncertainty.91 This body of case-law is a relatively small one in which there is little analysis but the ECJ can be understood to be taking a DC approach in these cases in that it recognised the legitimacy of flexible discretion in such circumstances.92 In particular, what can be seen in these cases is the legitimation of the operation of wideranging discretion that promoted the regulatory aims of a directive.93 The third, related, category is where Member States attempted to derogate from Community regimes. Again, this is a small subset of case-law in which the ECJ tended to emphasis two things. First, it stressed the need for Member States to cooperate in the operation of regulatory schemes, and thus, while they could take precautionary action, they needed to consult with other Member States and Community institutions.94 Second, it held that, in most cases, the Community regime embodied a precautionary approach in that it allowed discretionary action in the face of uncertainty. As this was the case there was no need for a Member State to take precautionary action beyond that allowed by the regime.95 Both these 87

Case T-199/96 Bergaderm, above n 79. Case C-180/96 United Kingdom v Commission, above n 80 at para 99. See also Case T-199/96 Bergaderm, above n 79 at para 66. 89 V Heyvaert, ‘Facing the Consequences of the Precautionary Principle in European Community Law’ (2006) 31 European Law Review 185; and Ladeur, above n 74. 90 Ladeur, above n 74 at 1459. 91 Note this case-law needs to be distinguished from that very large body of Art 226 TEC enforcement case-law concerning the failure of Member States to implement Community laws. 92 Case C-318/98 Fornasar, above n 37 at para 37; and Case C-293/97 Standley, above n 39. 93 Case C-418–419/97 Arco Chemie Nederland Ltd v Minister Van Volkshuisvesting, Ruimtelijke Ordening En Milieubeheer [2000] ECR I-4475 at para 40. 94 Case C-6/99 Association Greenpeace, above n 40 at paras 39–47. 95 Case C-52/92 Commission v Portugal [1993] I-2961 at para 13. 88

(I) Fisher Ch6

31/5/07

09:59

Page 221

The Case Law Before the Commission’s Communication

221

restraints can be understood as flowing from Article 10 TEC, and what can be seen in these cases is that the Court was understanding that obligation in deliberative terms rather than as a strict duty to apply a certain methodology. The final and most significant category of cases is that concerning the ECJ’s case-law in relation to the right of Member States to take action which prima facie infringed Article 28. There are many cases from this era in which the ECJ found that Member State technological risk regulation measures were justified breaches of Article 28, and, as Weatherill has noted, this body of case law shows that Article 28 was not a ‘charter of irresistible deregulation’.96 It is also a body of case-law in which the ECJ was very comfortable with reviewing Member State measures passed in circumstances of scientific uncertainty.97 Again, its approach can be seen to be underpinned by the DC paradigm. Speaking generally, the reasoning of both the AGs and the Court in these cases can be categorised into a number of steps. The first, as in all Article 28 cases, was for the Court to ask whether the Community had already harmonised this area of the law, and in most cases it had not. Where there had been partial harmonisation the Court would draw on Community policy (particularly precautionary policy) in determining the validity of a Member State’s reasoning.98 The second step was that, in cases where there was no harmonisation and scientific uncertainty over the nature of the risk, the ECJ stated that the ‘degree’ of protection was a matter for the Member States.99 In putting forward this requirement, the Court appreciated the problems of scientific uncertainty100 and did not understand the concept of ‘degree’ in quantitative terms. Rather, what the Court emphasised was the right of a Member State to act in relation to a particular risk. The concept of ‘degree’ was thus referring to the ability of Member States to take action in relation to a risk but did not dictate what the regulatory outcome would be. The emphasis on the ‘embryonic’ stage of harmonisation in these cases strengthened this approach because it recognised that there was no Communitywide deliberation of these problems.101 The third stage of the Court’s reasoning in these cases was to emphasise that the burden of establishing the validity of the Member State’s justification was on the Member State.102 Talking in terms of burdens would superficially seem to suggest that the Court was emphasising the evidentiary basis of the Member State’s reasoning but this was not the case for a number of reasons. Not only does the ECJ not rule on facts in preliminary references but there was very little emphasis on 96 97 98

Weatherill, above n 15 at 43. Scott and Vos, above n 65 at 260. Case 174/82 Sandoz BV [1983] ECR 2445 at para 11; and Case 54/85 Mirepoix, above n 69 at para

13. 99 Case 174/82 Sandoz, above n 98 at para 16; Case 227/82 Bennekom [1983] ECR 3883 at para 36; and Case 176/84 Commission v Greece, above n 70 at para 36. 100 See also AG Warner in Case 53/80 Eyssen, above n 69, discussing the problems of establishing what is ‘necessary’. 101 Case 247/84 Motte, above n 70 per AG Mancini. 102 Case 174/82 Sandoz above n 98 at para 22.

(I) Fisher Ch6

31/5/07

09:59

Page 222

222 Precautionary Principle and Administrative Constitutionalism in the EU evidence in Article 28 cases due to the Dassonville ruling not requiring a litigant to ‘prove’ that there was an actual restriction to free movement of goods.103 Indeed, the AG in Eyssen stressed the fact that in the risk and scientific uncertainty cases the issue was less one of evidence and more one of the ECJ ruling on the acceptability of the measure.104 The fourth step in reasoning in these cases was the Court assessing the necessity of the measures. This required the ECJ to assess the scientific uncertainties involved and the nature of the Member State’s evidence and reasoning. A Member State which put forward no evidence at all was likely to fail in establishing the validity of its measures.105 One of the interesting features of these cases is that the Court took a sophisticated approach to conceptualising scientific uncertainty and did not simply understand it as a data gap. Thus, for example, it distinguished scientific uncertainties to do with: the acceptable limits;106 animal studies;107 the amount of residues in a product;108 climatic conditions;109 divergences in national diets;110 and the limits of scientific studies.111 In assessing the information that a Member State was relying on, the Court was also less concerned with assessing outcomes than with assessing who had carried it out. Thus, for example, the ECJ in these cases gave greater weight to information produced by international scientific bodies112 and by Community committees.113 A Member State’s reliance on evidence solely produced by that State’s experts and contradicted by others was to be viewed with some scepticism.114 Moreover, in assessing such information the Court appreciated the nature of the bodies that carried out research. Thus, for example, in Muller, AG Darmon noted that research conducted by a Community scientific committee was not likely to remove uncertainties, particularly given its limited consultative role.115 The final step in reasoning in these cases concerned the application of the test of proportionality. Here, the Court emphasised that the nature of the risk was not the only relevant factor to consider and other factors may include whether the product served a nutritional116 or technological need.117 Likewise, the Court also 103 Case 8-74 Procureur du Roi v Dassonville [1974] ECR 837 at para 5, stating that measures that infringed Art 28 could do so ‘actually or potentially’. 104 Case 53/80 Eyssen, above n 69 per AG Warner. 105 Case 176/84 Commission v Greece, above n 70 at para 42; also see AG Slynn’s opinion. 106 Case 54/85 Mirepoix, above n 69 per AG Mancini; Case 227/82 Bennekom, above n 99 per AG Rozes; and Case 174/82 Sandoz, above n 98 at para 11. 107 Case 54/85 Mirepoix, above n 69 per AG Mancini. 108 Ibid at para 14. 109 Ibid per AG Mancini. 110 Case 53/80 Eyssen, above n 69; Case 247/84 Motte, above n 70 per AG Mancini at para 20; and Case 304/84 Muller, above n 70 at para 26. 111 Case 304/84 Muller, above n 70 per AG Darmon. 112 Eg Case C-473/98 Toolex Alpha, above n 56; Case 304/84 Muller, above n 70 at para 26; and Case 247/84 Motte, above n 70 at para 20. 113 Case 304/84 Muller, above n 70 at para 24; and Case 176/84 Commission v Greece, above n 70 at para 38. 114 Case 178/84 Commission v Germany, above n 70 per AG Slynn at para 44. 115 Case 304/84 Muller, above n 70. 116 Case 174/82 Sandoz, above n 98. 117 Case 176/84 Commission v Greece, above n 70 at para 45.

(I) Fisher Ch6

31/5/07

09:59

Page 223

The Case Law Before the Commission’s Communication

223

placed an emphasis on the importance of Member States being able to revise their standards.118 Nowhere in these cases did the ECJ take a cost/benefit analysis approach in answering the question of whether a measure was proportional. Overall, the reasoning in these cases was clearly informed by the DC paradigm of administrative constitutionalism. The ECJ understood the role and nature of public administration in assessing risks to be one that could not easily be constrained by methodology or by a preordained quantitative level of risk to be achieved. The end result was that the Court and the AGs in these cases were quite wide ranging in their analysis and, while they upheld a significant number of Member State measures, it cannot be said that their scrutiny of such measures amounted to a rubber-stamping exercise. As AG Slynn noted in Commission v Germany, cases such as these could not be ‘judged on generalities and broad statements of principle’ but rather required the Court to assess the factual material.119 The important point is that such scrutiny was not so much in terms of quantitative proof or the adherence of the Member State to a methodology but rather focused on the general quality of the Member State’s reasoning. Looking at these four categories side by side, what can be seen is that before 2000 the ECJ and CFI were promoting the DC paradigm in these different contexts. In so doing, however, their approach was very different in each case. Thus, in reviewing decision-making by Community institutions, their review was relatively minimal and confirmed the wide discretion given by legislative frameworks. In reviewing Member State implementation of Community regimes, the approach was a purposive one. In considering the ability of Member States to derogate from such regimes, the ECJ focused upon ensuring that Member States co-operated; and in Article 28 case-law it was carrying out quite a rigorous review on the basis of the DC paradigm. In all cases, the concern of both courts was on the quality of reasoning rather than on the accuracy of factual analysis. An end-product of this was that it gave both Member States and Community institutions considerable deliberative discretion. Indeed, much of this case-law would confirm the views of those who have tended to understand the EU as a form of deliberative supranationalism, albeit not a democratic one.120 A number of different reasons for why the ECJ’s and CFI’s reasoning before 2000 was underpinned by the DC paradigm in this area can be identified. First, the promotion of the DC paradigm at the Community level arguably reflected the neo-functionalist roots of the Community.121 A bureaucracy had been set up to address a series of broad problems and in light of the intergovernmental nature of 118

Case 54/85 Mirepoix, above n 69. Case 178/84 Commission v Germany, above n 70. 120 C Joerges and J Neyer, ‘From Intergovernmental Bargains to Deliberative Political Processes: The Constitutionalisation of Comitology’ (1997) 3 European Law Journal 273; and O Gerstenberg and C Sabel, ‘Directly-Deliberative Polyarchy: An Institutional Ideal for Europe?’ in C Joerges and R Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002). 121 B Rosamond, Theories of European Integration (Basingstoke, Macmillan Press, 2000) at ch 3. 119

(I) Fisher Ch6

31/5/07

09:59

Page 224

224 Precautionary Principle and Administrative Constitutionalism in the EU many of the regulatory arrangements the courts did not see the need to constrain power along RI lines. Second, the DC paradigm had arguably been the historical framework for risk regulation in a number of Member States.122 As debate about administrative constitutionalism was still emerging there was a greater reliance on those Member State understandings of administrative constitutionalism. This perhaps also explains the relative paucity of reasoning in the Article 230 case-law— the Court was carrying out review in a context in which the issues it was addressing had received little attention. Third, the DC paradigm gave both Community institutions and Member States considerable space for action, albeit space restricted by the need to co-operate. As a form of integration it was thus relatively flexible. Moreover, its operation meant that the ECJ, in relation to Member States, was not required to rule on the legitimacy of a Member State’s administrative regime but rather on the relationship between that regime and Community institutions and other Member States. This allowed the ECJ to operate a softer, more politically sensitive form of integration.

IV The Commission’s Communication on the Precautionary Principle In light of the case-law seen in the section above, it comes as no surprise that when the precautionary principle rose to international prominence in the 1990s it was embraced by Community institutions and Member States.123 The principle, in so being embraced, was understood in DC terms as a principle that allowed broadranging discretion that required particular attention to be paid to the prescriptive norms on which regulatory regimes were based. Throughout the 1990s, however, there was little in the way of detailed articulation about the precautionary principle, and it was only in the late 1990s that there were demands for a set of guidelines for the principle’s application. Those demands must be seen against a background of emerging debate about administrative constitutionalism within the EU. In particular, by the late 1990s, there was increasing concern about the legitimacy of Community law-making and the ability of the Courts to hold Community decision-makers to account.124 Corruption in the Commission,125 the BSE 122

von Moltke, above n 7; and Macrory (ed), above n 51. Christoforou, above n 23; and P Harremoës et al (eds), The Precautionary Principle in the Twentieth Century: Late Lessons from Early Warnings (London, Earthscan Publications, 2002). 124 Harlow, above n 36 at ch 6; A Arnull, ‘Introduction: The European Union’s Accountability and Legitimacy Deficit’ in A Arnull and D Wincott (eds), Accountability and Legitimacy in the European Union (Oxford, Oxford University Press, 2002) at 1; and Commission of the European Communities, European Governance: A White Paper, COM(2001) 428 final at 8. Case C-376/98 Germany v Parliament and Council [2000] ECR I-8149 arguably had symbolic importance in this regard. 125 P Craig, ‘The Fall and Renewal of the Commission: Accountability, Contract and Administrative Organisation’ (2000) 6 European Law Journal 98 123

(I) Fisher Ch6

31/5/07

09:59

Page 225

The Commission’s Communication on the Precautionary Principle

225

crisis,126 new empirical insights into Community decision-making;127 as well as the increasing obviousness of a democratic deficit128 all meant that Community decision-making was coming under increased scrutiny.129 A key issue in this process of reflection and evaluation was technological risk decision-making and the role of science within it.130 Furthermore, the WTO dispute settlement in EC— Hormones had also called into question the legality, in WTO terms, of EU risk evaluation.131 The implications of all of this for the European Commission’s Communication concerning the Precautionary Principle which was published in 2000132 was that it was one of a number of Commission documents at the time which was concerned with limiting and holding Community decision-makers to account.133 As such, it was part of a new era of administrative constitutionalism within the EU where the growing concern was with articulating the role and nature of Community administrative institutions. That process of articulation was, in relation to the Communication, in RI terms. This ironical state of affairs was for two reasons. First, some, although not all,134 of the EU debate over administrative constitutionalism was promoting the RI paradigm. This was particularly in relation to technological risk decision-making, where there was greater policy emphasis placed on ensuring decisions were underpinned by objective and independent science.135 Much of this was in response to the perceived lack of democratic control of Community institutions,136 but also was arguably due to the influence of the US discourses on administrative constitutionalism which, as seen in chapter three, were by this time in RI terms.137 Second, a major catalyst for the Commission’s 126 G Chambers, ‘The BSE Crisis and the European Parliament’ in C Joerges and E Vos (eds), EU Committees: Social Regulation, Law and Politics (Oxford, Hart Publishing, 1999). 127 Vos, above n 15. 128 Weiler, above n 72. 129 Craig, above n 23 at Part 1; G Majone, ‘Delegation of Regulatory Powers in a Mixed Polity’ (2002) 8 European Law Journal 319; and F Bignami, ‘The Democratic Deficit in European Rulemaking: A Call For Notice and Comment in Comitology’ (1999) 40 Harvard International Law Journal 451. 130 G Majone, Dilemmas of European Integration: The Ambiguities and Pitfalls of Integration by Stealth (Oxford, Oxford University Press, 2005) at ch 6; and C Joerges et al (eds), Integrating Scientific Expertise into Regulatory Decision-Making (Baden-Baden, Nomos Verlagsgesellschaft, 1997). 131 Panel Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), above n 46; and Appellate Body Report, European Communities—Measures Concerning Meat and Meat Products (Hormones), above n 46. For a discussion of these see section III of ch 5. 132 Commission of the European Communities, above n 1. 133 Commission of the European Communities, First Report on the Harmonisation of Risk Assessment Procedures (2000); Commission of the European Communities, above n 124; and Commission of the European Communities, Communication from the Commission on Impact Assessment, COM(2002) 276 final. 134 R Dehousse, ‘Misfits: EU Law and the Transformation of European Governance’ in C Joerges and R Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002); and O de Schutter et al (eds), Governance in the European Union (Luxembourg, European Commission, 2001). 135 Commission of the European Communities, Communication on Consumer Health and Food Safety, COM(97)183 final. 136 F Vibert, Governance in the European Union: From Ideology to Evidence (London, European Policy Forum, 2001); and Commission of the European Communities, above n 124. 137 Majone,above n 129; and Bignami, above n 129.

(I) Fisher Ch6

31/5/07

09:59

Page 226

226 Precautionary Principle and Administrative Constitutionalism in the EU Communication was the EU’s loss in the EC—Hormones decision.138 The Commission, however, mainly used the Panel’s decision, not the Appellate Body’s decision, as the template of reasonable regulatory action.139 As seen in the previous chapter, that decision promoted the RI paradigm.140 Throughout the Communication, the Commission stresses that one of the explicit aims of the Communication is to ‘avoid unwarranted recourse to the precautionary principle, as a disguised from of protectionism’141 and that the precautionary principle should never be used as a justification for arbitrary decision-making.142 The important point is that the concept of arbitrariness is being defined in RI terms.143 In the Communication, decision-makers are understood to be carrying out a series of discrete tasks, and environmental and public health decision-making is characterised as a three step process: a scientific process of risk assessment; a political process of risk management; and a process of risk communication.144 This characterisation of risk evaluation was not in operation in any of the Community legislative frameworks beforehand and is clearly taken from the EC—Hormones Panel decision. The Commission states that the principle applies only to the second of these steps,145 but also states that the principle applies only to risks where a risk assessment has identified a ‘potential risk’ and the uncertainties surrounding it.146 An assessment of the risk requires: reliable scientific data and logical reasoning, leading to a conclusion which expresses the possibility of occurrence and the severity of a hazard’s impact on the environment, or health of a given population including the extent of possible damage, persistency, reversibility, and delayed effect.147

While it is recognised that scientific uncertainty may make this problematic, the Commission stresses that as ‘complete a comprehensive assessment of risk’ as possible should be made.148 Risk assessment is also defined in strict procedural terms as consisting of hazard identification, hazard characterisation, appraisal of exposure, and risk characterisation.149 This means that while the document places significant emphasis on risk assessment and attempts to define what risk assessment is, the definition provided is ambiguous. The Communication is mainly silent on how scientific uncertainty should be identified and communicated as part of the risk assessment process. The Commission does, however, state that any ‘prudential’ approach

138 139 140 141 142 143 144 145 146 147 148 149

Majone, above n 130 at 128–31. Panel Report, above n 46. See section III of ch 5. Commission of the European Communities, above n 1 at 2. Ibid at 12, 15 and 21. Fisher and Harding, above n 5. Commission of the European Communities, above n 1 at 4. Ibid at 12. Ibid at 13. Ibid at 13. Ibid at 13 and 15. Ibid at 13 and Annex III.

(I) Fisher Ch6

31/5/07

09:59

Page 227

The Commission’s Communication on the Precautionary Principle

227

of taking into account of scientific uncertainty as part of the risk assessment process is distinct from the application of the precautionary principle.150 When a risk assessment does identify a potential ‘negative effect’ after a scientific evaluation has been done, then, as part of risk management, there must be a decision to act or not to act, and the precautionary principle will directly apply to this.151 In deciding this question, the Commission stresses the need for there to be an assessment of the uncertainties involved in the evaluation, and an assessment of the possible consequences of inaction or waiting for more scientific information.152 The process of applying the principle must be transparent and inclusive, particularly in assessing the consequences of different forms of action and inaction.153 The Commission, however, describes this as a ‘political decision’ and provides few guidelines for this process of evaluation beyond suggesting that it must have some scientific basis. Thus, while it states: The absence of scientific proof of the existence of a cause–effect relationship, a quantifiable dose/response relationship or a quantitative evaluation of the probability of the emergence of adverse effects following exposure should not be used to justify inaction ...

it goes on to note that: Even if scientific advice is supported only by a minority fraction of the scientific community, due account should be taken of their views, provided the credibility and reputation of this fraction are recognized.154

Not only is this a relatively limited understanding of scientific uncertainty but it would also seem clear that there must be some evidentiary basis for the decision. There must be some clear body of proof rather than evidence suggesting that there may be a problem. Moreover, the Commission does set out a series of standards which the measures to be taken must be judged against. Any measures taken pursuant to the principle must be proportional, non-discriminatory, consistent, based on an examination of potential costs and benefits, subject to revision in light of new data, and capable of assigning responsibility for the production of more scientific evidence.155 This list broadly reflects requirements in WTO and EU law, but there is no recognition of that fact in the Communication, or of the fact that these concepts are jurisprudentially complex and defined differently in each legal culture.156 Overall they amount to a RI duty to carry out a form of regulatory impact assessment to ensure that costs do not exceed benefits. 150

Ibid at 12. Ibid at 14. 152 Ibid at 15. 153 Ibid at 16. 154 Ibid at 16. 155 Ibid at 17–21. 156 For an example of the differences see G de Búrca, ‘Unpacking the Concept of Discrimination in EC and International Trade Law’ in C Barnard and J Scott (eds), The Law of the Single European Market: Unpacking the Premises (Oxford, Hart Publishing, 2002). 151

(I) Fisher Ch6

31/5/07

09:59

Page 228

228 Precautionary Principle and Administrative Constitutionalism in the EU The Communication is thus creating a rigid RI framework for decision-makers to follow if they wish to apply the precautionary principle. Decision-makers are being assigned two specific tasks of assessing and managing risks, and in regard to each are constrained by methodologies such as risk assessment and regulatory impact assessment. Not only must a decision-maker use those methodologies in making those decisions but a decision made pursuant to those methodologies is held to be valid because, so long as a decision-maker has followed these preordained processes, a decision is held to be within the power of the decision-maker. Moreover, the overall legitimacy of the decision is judged by the factual basis. Decision-making pursuant to the principle should start with ‘a scientific evaluation’,157 must be based on ‘a scientific evaluation, as complete as possible’, 158 and will probably involve commissioning scientists to perform ‘as objective and complete as possible scientific evaluation’.159 In other words, scientific evaluation, according to this document, is the central activity in the application of the precautionary principle. Before moving on, it is worth noting that the Communication is a highly cryptic set of guidelines. First, as already noted, there seems to be no fixed understandings of risk assessment and uncertainty informing the document. Moreover, although the Communication would appear to relate to all areas where the precautionary principle may apply, the discussion of risk assessment in Annex III applies to the health field, where the definition of risk assessment has been taken from the WTO and US context.160 Second, it is ambiguous how each of the steps identified by the Commission in the Communication relates to the others. The Communication divides up its guidelines into ‘the constituent parts of the precautionary principle’ and ‘guidelines for applying the precautionary principle’ but it is unclear when, and how, each step in the decision-making process must occur. Policy and scholarly responses to the Communication have been in terms of whether it promotes the model of public administration that the responder deems valid.161 Graham and Hsia praised the Communication for the orderly process it created, which ‘requires the kind of formal policy analysis’ taught in ‘schools of public policy and public administration’, while also criticising it for allowing the unchecked exercise of discretion.162 The European Council published a resolution in reply to the Communication in which it broadly agreed with it, although it placed greater stress on the importance of deliberation and the role of values, thus encouraging a more DC approach to standard setting.163 A similar type of 157

Commission of the European Communities, above n 1 at 15. Ibid at 15. 159 Ibid at 15. 160 E Fisher, ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds), Environmental Law for Sustainability (Oxford, Hart Publishing, 2006) at 121–2. 161 Majone, above n 26; and Sunstein, above n 26 at 121. 162 J Graham and S Hsia, ‘Europe’s Precautionary Principle: Promise and Pitfalls’ (2002) 5 Journal of Risk Research 371. 163 European Council, Council Resolution on the Precautionary Principle (Nice European Council Meeting, Presidency Conclusions, 2000). 158

(I) Fisher Ch6

31/5/07

09:59

Page 229

Case-law after the Commission’s Communication

229

response could also be seen from the Committee on the Environment, Public Health, and Consumer Policy of the European Parliament.164 In all cases the focus was on whether the Communication would aid ‘rational’ decision-making where what was understood as ‘rational’ was defined in accordance with understandings of administrative constitutionalism.

V Case-law after the Commission’s Communication Strictly speaking, the Commission’s Communication is meant only to set out how the Commission applies or intends to apply the precautionary principle,165 but the guidelines contained in it have become generic and have been applied by both Member States166 and other Community institutions. The most obvious example of the latter has been the regulatory framework of the European Food Safety Authority.167 This is not surprising. As already seen, integration involves the harmonisation of policy and the norms of administrative constitutionalism. Moreover, in the Communication, the Commission stress that it wishes to promote a ‘common understanding’ of how to assess and manage risk within the Community.168 The Communication has also, however, brought about a significant shift in thinking about administrative constitutionalism. While before its publication the precautionary principle and precautionary reasoning were understood in DC terms, after its publication the principle has been understood in RI terms. This can be seen in many different areas but my focus here is only upon the jurisprudence of the ECJ and CFI. What can be seen is that the Communication guidelines on the principle have been used by the CFI and ECJ as the framework for judging both Community and Member State action. As those guidelines understood the role of public administration in RI terms, this represented a dramatic shift in the case-law from the DC to the RI paradigm. The first significant judicial application of the Communication was not in relation to the control of Community institutions, however. Nor was it in the EU context. In 2001 the Communication clearly influenced a European Free Trade Association (EFTA) Court case concerning whether Norway was justified in banning vitamin- and iron-enriched cornflakes, EFTA v Norway.169 The jurisprudential framework for assessing whether Norway was justified in such a ban was 164 Committee on the Environment, Public Health, and Consumer Policy of the European Parliament, Report on the Commission Communication on the Precautionary Principle, A5-0352/2000 Final, 2000. 165 Commission of the European Communities, above n 1 at 3 and 9. 166 Interdepartmental Liaison Group on Risk Assessment, The Precautionary Principle: Policy and Application (2002); and R (on the application of Amvac Chemical UK Ltd) v Secretary of State for Environment, Food and Rural Affairs [2001] EWHC Admin 1011. 167 Regulation 178/2002, above n 19. 168 Commission of the European Communities, above n 1 at 3 and 9. 169 Case E-3/00 EFTA Surveillance Authority v Norway [2001] 2 CMLR 47.

(I) Fisher Ch6

31/5/07

09:59

Page 230

230 Precautionary Principle and Administrative Constitutionalism in the EU the same as that under Article 28 TEC.170 The Communication was not cited in the case, but was cited in the Judge-Rapporteur’s report.171 The Court stated: A proper application of the precautionary principle presupposes, firstly, an identification of potentially negative health consequences arising, in the present case, from a proposed fortification, and, secondly, a comprehensive evaluation of the risk to health based on the most recent scientific information. When the insufficiency, or the inconclusiveness, or the imprecise nature of the conclusions to be drawn from those considerations make it impossible to determine with certainty the risk or hazard, but the likelihood of considerable harm still persists were the negative eventuality to occur, the precautionary principle would justify the taking of restrictive measures. Such restrictive measures must be non-discriminatory and objective, and must be applied within the framework of a policy based on the best available scientific knowledge at any given time. The precautionary principle can never justify the adoption of arbitrary decisions, and the pursuit of the objective of ‘zero risk’ only in the most exceptional circumstances.172

This is a paraphrasing of the Communication’s RI guidelines, although these guidelines are not applying to the Commission but to a state that is not even a member of the EU. Despite this, and the fact that the Communication was at odds with previous case-law,173 the EFTA Court showed no qualms in applying the guidelines. Rather, the guidelines were taken as an authoritative account of the precautionary principle and thus an authoritative template for the appropriate paradigm of administrative constitutionalism for technological risk evaluation. EFTA v Norway is an important case because, while the Communication itself has not been cited heavily in CFI or ECJ case-law, this case has been.174 What can be seen in this case-law is the embracement by both courts of the precautionary principle defined in RI terms. This has been in all the different contexts seen above. Before looking at those contexts it is useful to make a few general points about that case-law. First, as shall be seen below, the precautionary principle is now understood in RI terms. Second, in doing this, neither court has acknowledged that the Communication departs from their previous approach in these cases.175 Rather, the promotion of the RI definition of the precautionary principle has been part of 170

Case E-3/00 EFTA Surveillance Authority v Norway [2001] 2 CMLR 47 at paras 23–4. EFTA Court, Report for the Hearing in Case E-3/00 prepared by Carl Baudenbacher, JudgeRapporteur (E-3/00/53, 2000) at paras 31, 72, 152, and 156. Also note the Commission was a party to the action. 172 Case E-3/00 Norway, above n 169 at paras 30–32. 173 In particular, Case 174/82 Sandoz, above n 98 which the EFTA Court cited: Case E-3/00 Norway, above n 169 at para 24. 174 Case C-192/01 Commission v Denmark [2003] ECR I-9693 at paras 47-01; Case C-95/01 Greenham and Abel per AG Mischo at para 55; Case C-286/02 Bellio F.lli Srl v Prefettura di Treviso [2004] ECR I-3465 at paras 57–60; and Case C41/02 Commission v Netherlands, above n 42 per AG Maduro at para 29. For examples where the Communication has been cited, see Case C-434/02 Arnold André, above n 58 per AG Geelhoed at para 97; Case C-127/02 Waddenzee, above n 38 per AG Kokott at para 100; and Case C-236/01 Monsanto, above n 38 at para 79. 175 See the interpretation of Case 174/82 Sandoz, above n 98 in the AG’s Opinion in Case C-192/01 Commission v Denmark, above n 174. 171

(I) Fisher Ch6

31/5/07

09:59

Page 231

Case-law after the Commission’s Communication

231

their ongoing promotion of precautionary reasoning more generally.176 Third, in examining precautionary reasoning, both courts rely on case-law deriving from different contexts. Thus, Article 230 cases are relied on in Article 28 cases, and vice versa.177 Fourth, as will also be seen, there is a certain pedantry in the application of the principle, and there have been a number of cases in which the principle has not been found to apply, because the court concluded that the principle should only apply: to provisional risk management measures;178 where there is a something more than a ‘hypothetical’ risk;179 and where there is scientific, as opposed to other types of, uncertainty,180 and should not apply to cases where the risk is well known.181 Indeed, it would seem that the principle is being treated more as a ‘bright line’ rule which dictates certain action in particular situations rather than a flexible principle that might result in a variety of outcomes. Fifth, and perhaps most significantly, what can be seen in all these cases is that it is not only the principle which is being interpreted in RI terms but also the whole process of risk evaluation. In particular, while there was historically no mention of risk assessment and no expectation of quantitative analysis, there now is.182 In some cases this is because risk assessment procedures have been included in Community legislation,183 but in most cases it would be because the risk assessment/risk management divide is now understood by the Court to be ‘ubiquitous’.184 All of this is best seen by example, and, while I cannot provide a comprehensive analysis of the case-law in this area, it is useful to highlight the approach taken in some of these cases. I do so by again looking at the four different contexts I analysed in Section III: review of Community action, review of Member State implementation of Community law, review of attempts by Member States to derogate from Community regimes, and Article 28 case-law. While before 2000 the body of case-law concerning judicial review of precautionary Community action was a small one, since that time the case-law has grown significantly. As well, there has been an increasing sophistication in the case-law concerning the control of Community institutions.185 The most high-profile cases they have considered in this category so far in relation to the precautionary 176 Case T-13/99 Pfizer, above n 60 at para 114; Case T-74/00 Artegodan, above n 11 at paras 185–6; Case C-236/01 Monsanto, above n 38 at paras 111–12. 177 Case C41/02 Commission v Netherlands, above n 42 at para 45; and Case C-244/03 France v Parliament [2005] ECR I-4021 per AG Geelhoed at para 106. 178 Case C-453/03 ABNA Ltd v Secretary of State for Health [2005] ECR I-10423 per AG Tizzano at para 129. 179 Case C-244/03 France v Parliament, above n 177 per AG Geelhoed at para 107. 180 Case C-434/02 Arnold André, above n 58 per AG Geelhoed at para 107. 181 Case E-4/04 Pedicel AS v Sosial-OG Helsedirektoratet (Directorate for Health and Social Affairs) [2005] 2 CMLR 7 182 Case E-3/00 Norway, above n 169 at paras 30–32; Case C-95/01 Greenham and Abel, above n 42; and Case C-24/00 Commission v France [2004] ECR I-1277 at paras 55–6. 183 Case C-132/03 Minstero della Salute v CODACONS [2005] ECR I-4167 at para 63. 184 Case C-192/01 Commission v Denmark, above n 174 per AG Mischo at para 89. 185 Azoulay, above n 21; Craig, above n 23 at Part 2; and Case C-378/00 Commission v Council [2003] ECR I-937.

(I) Fisher Ch6

31/5/07

09:59

Page 232

232 Precautionary Principle and Administrative Constitutionalism in the EU principle have been Pfizer and Alpharma v Council.186 Both these cases are highly complex but the analysis of the principle is the same in each case. For ease of analysis, I focus on the reasoning of the CFI in Pfizer.187 The factual matrix in relation to Pfizer was considered above.188 Besides the facts related there, it is important to note that under the directive there was no requirement of a risk assessment and not even a mandatory role for a scientific committee,189 although there was other authority that decisions should be based on ‘sound scientific advice’.190 In passing the Regulation, the Council had taken a precautionary approach, although had not explicitly applied the precautionary principle. Thus, in relation to one antibiotic, it noted that while there was no ‘real immediate risk’ to human health, there was a ‘potential’ risk.191 In other words, the actions of the Community institutions in this case were entirely consistent with the DC paradigm and with the previous judicial review case-law of the ECJ and CFI.192 There were a number of different grounds of review, including that the decision to withdraw authorisation was based on a manifest error of assessment because, among other things, the Community institutions had incorrectly applied the precautionary principle. The CFI concluded that the Commission’s Communication on the Precautionary Principle was not directly binding as it only came into force after the challenged decision.193 However, the Communication was taken to reflect the law at the time because of statements by the European Commission in the Communication and before the Court in this case.194 The CFI began its analysis by stating it must first ‘define the ‘risk’ which must be assessed when the precautionary principle is applied.195 This is an obvious and very logical starting point, but what is interesting is that no straightforward definition was forthcoming. The Court noted that ‘risk’ is not referred to in the Directive but that it was the relevant terminology to use, particularly as the Regulation was framed in terms of it. The risk in this case, it noted, was the ‘possibility that the use of virginiamycin [the relevant antibiotic] as an additive to 186 Case T-13/99 Pfizer, above n 60; and Case T-70/99 Alpharma v Council [2002] ECR II-3495. See W Douma, ‘Fleshing out the Precautionary Principle by the Court of First Instance’ (2003) 15 Journal of Environmental Law 372; C MacMaolain, ‘Using the Precautionary Principle to Protect Public Health: Pfizer v Council’ (2003) 28 European Law Review 723; E Vos, ‘Antibiotics, the Precautionary Principle and the Court of First Instance’ (2004) 11 Maastricht Journal of European and Comparative Law 187; E Fisher, ‘Precaution, Law and Principles of Good Administration’ (2005) 52 Water Science and Technology 19; and Craig, above n 23 at 722–7. 187 That reasoning is the same in Case T-70/99 Alpharma, above n 186. 188 See section II. 189 Art 8 of Directive 70/524/EEC allowed matters to be referred to the Scientific Committee for Animal Nutrition but did not require it. 190 Commission of the European Communities, above n 135; and Commission Decision 97/579/EC setting up Scientific Committees in the field of consumer health and food safety [1997] OJ L237/18. 191 Council Regulation No 2821/98 amending, as regards withdrawal of the authorisation of certain antibiotics, Directive 70/224/EEC concerning additives in feeding stuffs [1998] OJ L351/4. 192 Fisher, above n 186 at 21–2. 193 Case T-13/99 Pfizer, above n 160 at para 122. 194 Ibid at para 123. 195 Ibid at para 135.

(I) Fisher Ch6

31/5/07

09:59

Page 233

Case-law after the Commission’s Communication

233

feeding stuffs will give risk to adverse effects on human health’.196 This was quite a broad interpretation and could accommodate many different considerations. The CFI then stated that EU institutions did have a right to take action in the face of scientific uncertainty and stated that: in a situation in which the precautionary principle is applied, which by definition coincides with a situation in which there is scientific uncertainty, a risk assessment cannot be required to provide the Community institutions with conclusive scientific evidence of the reality of the risk and the seriousness of the potential adverse effects were that risk to become a reality.197

Such a statement is entirely consistent with the DC paradigm but a little further on in the judgment the Court took a different approach and one clearly inspired by the Communication. It stated: a preventive measure may be taken only if the risk, although the reality and extent thereof have not been ‘fully’ demonstrated by conclusive scientific evidence, appears nevertheless to be adequately backed up by the scientific data available at the time when the measure was taken.198

As such, it stated that the principle could not apply to a ‘purely hypothetical risk’199 and can therefore only apply in situations in which there is a risk . . . although it is not founded on mere hypotheses that have not been scientifically confirmed, has not yet been fully demonstrated.200

The CFI then went on to define risk as the ‘function of the probability that the use of a product or a procedure will adversely affect the interests safeguarded by the legal order’.201 Then it stated that: The purpose of a risk assessment is to assess the degree of probability of a certain product or procedure having adverse effects on human health and the seriousness of any such adverse effects.202

This approach was far more grounded in the RI paradigm. Risk was being defined in quantitative terms and to actually establish such a risk, even if it was ‘not fully demonstrable’, required a level of certainty that was unlikely to exist if an institution was wishing to apply the precautionary principle. Likewise, the CFI was defining risk assessment in very narrow terms and in a way in which it would be very difficult to take other factors into account. The analysis in Pfizer does highlight many of the problems in the Commission’s Communication and, in particular, how that Communication relates to existing 196 197 198 199 200 201 202

Ibid at para 138. Ibid at para 142. Ibid at para 144. Ibid at para 145. Ibid at para 146. Ibid at para 147. Ibid at para 148.

(I) Fisher Ch6

31/5/07

09:59

Page 234

234 Precautionary Principle and Administrative Constitutionalism in the EU decision-making frameworks.203 In other decisions, the analysis has been briefer and far more in straightforward RI terms. Thus in Artegodan GmbH v Commission204 the CFI understood that the precautionary principle applied to the Commission’s assessment of evidence and stated: The precautionary principle requires the suspension or withdrawal of a marketing authorisation where new data[s] give rise to serious doubts as to either the safety or the efficacy of the medicinal product in question and those doubts lead to an unfavourable assessment of the benefit/risk balance of that medicinal product . . . Against that background, the competent authority [in this case the Commission] need do no more than provide, in accordance with the general rules of evidence, solid and convincing evidence which, while not resolving the scientific uncertainty, may reasonably raise doubts as to the safety and/or efficacy of the medicinal product.205

Two things are clear from this statement. First, compared with the pre-2000 caselaw, precautionary reasoning is understood to have a far narrower scope. This approach has been followed in other cases.206 Second, the principle is understood to be a principle that applies to the exercise of power by both Member States and Community institutions pursuant to a Community regime. This is also consistent with the fact that the CFI in this decision recognised that the principle was a ‘general principle of Community law’.207 What can also be seen in this body of case law is that the precautionary principle is raised in a number of cases and then dismissed often because it has been found to be irrelevant.208 This approach reflects an interpretation of the principle being a ‘bright line’ RI rule. Thus in Arnold André GmbH & Co KG v Landrat SWA Kreises Herford 209AG Geelhoed, in considering the application of the precautionary principle, stated: It is in my view beyond doubt that the Community legislature can base its action on the precautionary principle if three cumulative conditions are fulfilled. There must be scientific uncertainty about the risk, the risk must be analysed and proved to be realistic and the risk must have substantial consequences for the public interest. When it comes to the substance of a measure, a measure relying on the precautionary principle may not go as far as banning all risk.210

This is a very narrow RI reading of the principle and, because he concluded that the uncertainties involved were not scientific, he did not find the principle relevant.211 At the same time, however, in other cases there seems to be an attempt to 203 M van Asselt and E Vos, ‘The Precautionary Principle and the Uncertainty Paradox’ (2006) 9 Journal of Risk Research 313. 204 Case T-74/00 Artegodan, above n 11. 205 Ibid at para 192. 206 Case T-392/02 Solvay Pharmaceuticals BV v Council [2003] ECR II-4555 at para 129. 207 Case T-74/00 Artegodan, above n 11 at para 184. 208 Case C-453/03 ABNA, above n 178 per AG Tizzano at paras 123–31. 209 Case C-434/02 Arnold André, above n 58. 210 Ibid per AG Geelhoed at para 103. 211 Ibid per AG Geelhoed at para 107.

(I) Fisher Ch6

31/5/07

09:59

Page 235

Case-law after the Commission’s Communication

235

recognise that the principle is distinct from a more general precautionary competence entrusted to Community institutions.212 Such an approach would seem to be an attempt to reconcile the RI understanding of the principle with more traditional DC understandings of Community power. The second category of cases where the precautionary principle is considered by the ECJ is in the implementation of Community law by Member States.213 As with the pre-2000 case-law there is little substantial analysis in these cases, and most of it is in the context of specific legislative frameworks. What is clear is that the precautionary principle is again being understood in RI terms. In particular, the focus in these cases is upon the treatment of evidence rather than upon the general exercise of discretion by a Member State. This can be seen clearly in the ECJ’s interpretation of the Habitats Directive.214 In Landelijke Vereniging tot Behoud van de Waddenzee v Nederlandse Vereniging tot Bescherming van Vogels (the Waddenzee decision)215 the ECJ considered how the first sentence of Article 6(3) of the directive should be interpreted in light of the precautionary principle. That sentence reads: Any plan or project not directly connected with or necessary to the management of the site but likely to have a significant effect thereon, either individually or in combination with other plans or projects, shall be subject to appropriate assessment of its implications for the site in view of the site’s conservation objectives.

The Court interpreted the provision in the following RI terms: It follows that the first sentence of Art 6(3) of the Habitats Directive subordinates the requirement for an appropriate assessment of the implications of a plan or project to the condition that there be a probability or a risk that the latter will have significant effects on the site concerned. In the light, in particular, of the precautionary principle, which is one of the foundations of the high level of protection pursued by Community policy on the environment, in accordance with the first subparagraph of Art 174(2) EC, and by reference to which the Habitats Directive must be interpreted, such a risk exists if it cannot be excluded on the basis of objective information that the plan or project will have significant effects on the site concerned. Such an interpretation of the condition to which the assessment of the implications of a plan or project for a specific site is subject, which implies that in case of doubt as to the absence of significant effects such an assessment must be carried out, makes it possible to ensure effectively that plans or projects which adversely affect the integrity of the site concerned are not authorised, and thereby contributes to achieving, in accordance with the third recital in the preamble to the Habitats Directive and Art 2(1) thereof, its main aim, namely, ensuring biodiversity through the conservation of natural habitats and of wild fauna and flora.216 212

Case C-453/03 ABNA, above n 178 per AG Tizzano at para 130. For a discussion of some of these cases see Craig, above n 23 at 73–3. 214 Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora [1992] OJ L 206/7. 215 Case C-127/02 Waddenzee, above n 38. 216 Ibid at paras 43–4. 213

(I) Fisher Ch6

31/5/07

09:59

Page 236

236 Precautionary Principle and Administrative Constitutionalism in the EU In other words, the Court is stating that in certain evidentiary circumstances (where ‘it cannot be excluded on the basis of objective information that the plan or project will have significant effects on the site’) the precautionary principle will require a Member State to make a certain assumption (that ‘a risk exists’). This case has been followed in a number of other cases217 and casts the discretion of Member State authority in highly RI terms. The principle is operating as a rule that requires a certain course of action to be taken when certain conditions precedent are met. The third category in which the ECJ and CFI have considered the application of the precautionary principle is where Member States have attempted to rely on the principle in derogating from a Community measure. In Denmark v Commission the principle was relied on by Denmark in maintaining certain national food safety provisions, but the ECJ found that in harmonised areas the principle was to be applied by Community institutions and should not be used to support unilateral provisions.218 The principle also clearly does not overrule the requirement of Article 95(5).219 Again, as in the other cases, the court understood the obligation as one of risk assessment rather than deliberation. Where there has been more substantive consideration of the precautionary principle has been in relation to Member States relying on specific provisions of a directive to put in place more precautionary measures than those of the Community. Again, however, the Court has tended to understand the principle as already embodied in Community legislation.220 In Monsanto Agricoltura Italia SpA v Presidenza del Consiglio dei Ministri 221 the ECJ found that the precautionary principle was relevant in the interpretation of a safeguard clause concerned. It stated: protective measures may be taken pursuant to Article 12 of Regulation No 258/97 interpreted in the light of the precautionary principle even if it proves impossible to carry out as full a risk assessment as possible in the particular circumstances of a given case because of the inadequate nature of the available scientific data (see to that effect Pfizer Animal Health v Council, cited above, paragraphs 160 and 162, and Alpharma v Council, cited above, paragraphs 173 and 175). Such measures presuppose, in particular, that the risk assessment available to the national authorities provides specific evidence which, without precluding scientific uncertainty, makes it possible reasonably to conclude on the basis of the most reliable scientific evidence available and the most recent results of international research that the implementation of those measures is necessary in order to avoid novel foods which pose potential risks to human health being offered on the market.222

217 Case C-6/04 Commission v United Kingdom [2005] ECR I-9017 at para 54; and Case C-98/03 Commission v Germany [2006] ECR I-53 at para 40. 218 Case C-3/00 Denmark v Commission, above n 41 at para 103. 219 Case T-366/03 Land Oberosterreich, above n 41 at para 71. 220 Case C-132/03 CODACONS, above n 183 at paras 61–3. 221 Case C-236/01 Monsanto, above n 38. 222 Ibid at paras 112–13.

(I) Fisher Ch6

31/5/07

09:59

Page 237

Case-law after the Commission’s Communication

237

Again, this understanding of the principle is in RI terms, and the operation of the precautionary principle is closely aligned with the operation of risk assessment. Moreover, the ECJ is using cases from another context—that of judicial review of Community institutions—to back up its interpretation. The final category where the ECJ has considered the application of the precautionary principle is in the context of Article 28 case-law. As seen in Section III above, the ECJ had developed a sophisticated body of jurisprudence in relation to Member States justifying their action on the basis of precautionary reasoning. That case-law has been replaced, however, with the type of reasoning seen in other contexts. Thus, for example, in Commission v Denmark,223 Denmark relied on the previous ECJ authority Sandoz 224 to justify its application of the precautionary principle in banning food enriched with vitamins. AG Mischo concluded that Sandoz was an application of the precautionary principle ‘before the fact’,225 but neither he nor the Court followed the reasoning in that case. Rather, the application of the precautionary principle required the same set of steps seen above. Thus, the Court stated: A proper application of the precautionary principle presupposes, in the first place, the identification of the potentially negative consequences for health of the proposed addition of nutrients, and, secondly, a comprehensive assessment of the risk to health based on the most reliable scientific data available and the most recent results of international research. Where it proves to be impossible to determine with certainty the existence or extent of the alleged risk because of the insufficiency, inconclusiveness or imprecision of the results of studies conducted, but the likelihood of real harm to public health persists should the risk materialise, the precautionary principle justifies the adoption of restrictive measures.226

This is a very different test from that in Sandoz. It is requiring a Member State to operate its risk evaluation processes in line with the RI paradigm of administrative constitutionalism. This recasting of risk evaluation can be seen in other cases as well, and the quantitative assessment of risk is taken to be a necessary and central feature of all decision-making processes.227 Thus, for example, in Commission v France the Court stated: A decision to prohibit the marketing of a fortified foodstuff, which is in fact the most restrictive obstacle to trade in products lawfully manufactured and marketed in other Member States, can be adopted only if the alleged real risk for public health appears to be sufficiently established on the basis of the latest scientific data available at the date of the adoption of such decision. In such a context, the object of the risk assessment to be 223

Case C-192/01 Commission v Denmark, above n 174. Case 174/82 Sandoz, above n 98. See section III for a discussion 225 Case C-192/01 Commission v Denmark, above n 174 per AG Mischo at para 83. 226 Ibid at para 52. 227 Case C-95/01 Greenham and Abel, above n 42 at para 43; and Case C41/02 Commission v Netherlands, above n 42 at paras 46–9. 224

(I) Fisher Ch6

31/5/07

09:59

Page 238

238 Precautionary Principle and Administrative Constitutionalism in the EU carried out by the Member State is to appraise the degree of probability of harmful effects on human health from the addition of certain nutrients to foodstuffs and the seriousness of those potential effects. It is clear that such an assessment of the risk could reveal that scientific uncertainty persists as regards the existence or extent of real risks to human health. In such circumstances, it must be accepted that a Member State may, in accordance with the precautionary principle, take protective measures without having to wait until the existence and gravity of those risks are fully demonstrated. However, the risk assessment cannot be based on purely hypothetical considerations.228

The decision-making process is being characterised in a very different way from that in which it was characterised in the earlier Article 28 case-law. The Court and the AGs identify the need for a Member State to identify a ‘real’ risk,229 and a risk assessment must be carried out so as to ‘appraise the degree of probability of harmful effects on human health’.230 The concept of degree is now being defined in quantitative rather than qualitative terms, and neither the Court nor the AGs engage in a particularly careful analysis of the scientific uncertainties involved. Rather uncertainty is understood to be in terms of the limits of risk assessment. Moreover, the doctrine of proportionality is understood in far more cost/benefit terms than in previous cases.231 It should be noted that all these Article 28 cases have been in relation to food, and, in particular, food supplements, where there now exists at the Community level a regime which includes a requirement for risk assessment.232 While that framework was not in force when these cases were litigated, the role given to risk assessment was influential. With that said, that directive was also influenced by the Communication as well as the more general shift to the RI paradigm at the Community level.233

VI Reflections What can be seen in the last three sections is that over the past decade there has been a dramatic shift in the ECJ’s jurisprudence. That shift has had three different dimensions. First, there has been a shift from assessing the validity of precautionary reasoning in particular contexts to applying the overarching precautionary 228

Case C-24/00 Commission v France, above n 182 at paras 55 and 56. Although note Case C-192/01 Commission v Denmark, above n 174 per AG Mischo at para 102, arguing that such a risk only need be plausible. 230 Ibid at paras 46 and 48. Also see Case C41/02 Commission v Netherlands, above n 42 at paras 47–8; and Case C-95/01 Greenham and Abel, above n 42 at paras 40–41. 231 Case C41/02 Commission v Netherlands, above n 42 at para 46; and Case C-24/00 Commission v France, above n 182 at para 52. 232 Directive 2002/46 on the approximation of the laws of the Member States relating to food supplements [2002] OJ L183/51. 233 Case C-3/00 Denmark v Commission, above n 41 per AG Mischo at paras 60 and 89. 229

(I) Fisher Ch6

31/5/07

09:59

Page 239

Reflections

239

principle. Thus there has been a blurring of the different contexts of application. This blurring has not been universal, however, and in Commission v Netherlands AG Maduro, arguably applying his own theories concerning Article 28,234 argued the need to recognise the different contexts of the principle’s application. He stated: Different consequences ensue from recourse to the precautionary principle depending on whether it is invoked by the Community institutions or by the Member States. If, for example, a State relies on the precautionary principle, its decision will lead to a partitioning of the single market. Furthermore, even if the measure adopted is not guided by protectionist considerations, the view of the other Member States cannot be taken into account, unlike in the case of a Community institution’s adoption of a decision under the precautionary principle. In my view, that explains the case law of the Court which set about imposing stringent restrictions on use of the precautionary principle where it is invoked by Member States.235

This view does not seem to be shared by others, however, and looking at the caselaw examined in the previous section, the main difference of approach has been dependent on different legislative frameworks rather than upon different legal contexts. The second shift that can be seen is that there has been a shift from understanding the principle and risk evaluation in DC terms to understanding them both in RI terms. The focus is less upon deliberative problem-solving than upon accurate analysis and the application of methodology. This shift is consistent with more general regulatory developments at the Community level and has meant that the courts have gone from understanding precautionary reasoning as broad ranging and deliberative to understanding it as an exception that operates in specific circumstances. Likewise, risk evaluation has gone from being understood as being an analytical-deliberative process to being divided neatly into risk assessment and risk management. Again, this shift is not universal. Thus, for example, returning to the comment of AG Maduro in Commission v Netherlands quoted above, we can see that he gave greater scope to Community institutions in applying the principle because they could deliberate with all Member States and their decisions could reflect those deliberations. Moreover, he held that risk assessments need only be taken into account.236 His analysis is very much grounded in the DC paradigm. Moreover, as already noted, it is also clear that the courts wish to recognise the precautionary principle, or some version of it, as enabling the more general exercise of Community power.237 The third dimension of this shift is that the ECJ and CFI in these more recent cases are relying overtly on how the Commission has interpreted the precautionary principle. Thus, what can be seen in these cases is the Commission’s own 234

Maduro, above n 67. Case C41/02 Commission v Netherlands, above n 42 per AG Maduro at para 30. 236 Ibid per AG Maduro at fn 52. 237 R (on the application of: Alliance for Natural Health), above n 58 at para 68; and Case C-453/03 ABNA, above n 178 at para 130. 235

(I) Fisher Ch6

31/5/07

09:59

Page 240

240 Precautionary Principle and Administrative Constitutionalism in the EU perspective on the precautionary principle being applied. This is even though in some cases the ECJ recognises problems with that perspective.238 This state of affairs is problematic when one considers that the precautionary principle is meant to be a ‘general principle of Community law’.239 If it is, then one would expect that its definition and application by courts would be as nuanced and complex as the definition and application of other legal principles such as fundamental rights and proportionality.240 Yet what has seemingly occurred is that the ECJ and CFI have deferred to the Commission’s Communication. In so doing, they have failed to appreciate that that Communication was the product of particular understandings of administrative constitutionalism. At the same time, however, such deference is not surprising. The harmonisation of policy norms and principles of administrative constitutionalism are features of integration. In such circumstances, it is perhaps natural that a document such as the Communication has been given such authority. As with previous chapters, this analysis does not provide the reader with any easy answers. Rather, it raises some very difficult questions where those questions relate to the interrelationship between different administrative institutions. To modify fashionable terminology, what are highlighted are problems with the ‘thru-put’ legitimacy of EU administrative institutions, rather than with output or input legitimacy, which have tended to be the focus of European integration scholarship.241 In thinking about those issues of ‘thru-put legitimacy’ it is necessary to return to the tension highlighted in Section II of this chapter. On the one hand, within multi-level networked systems of governance such as the EU there must be an appreciation that there is a range of different administrative institutions operating in different legal cultures and shaped by different understandings of administrative constitutionalism. In light of this diversity there is a need to recognise both legal pluralism and administrative pluralism. Integration cannot simply be the process of applying the same rules and the same decision-making processes across such a network. Such decision-making processes are expressions of administrative constitutionalism, and administrative constitutionalism is part of the thickness of legal culture. In this regard, it is not enough just to say that principles such as the precautionary principle represent a ‘mingling’ of norms from different legal systems.242 Rather, there needs to be greater recognition of the fact that such principles are embedded within theories of administrative constitutionalism. At the same time, however, there is pressure for understanding principles such as the precautionary principle in uniform terms. Indeed, as we have seen, an essential part of European integration is the integration of policy and norms of administrative 238

Case C-192/01 Commission v Denmark, above n 174 per AG Mischo at para 101. Case T-74/00 Artegodan, above n 11 at para 184. 240 T Tridimas, The General Principles of European Community Law, 2nd edn (Oxford, Oxford University Press, 2006). 241 F Scharpf, Governing in Europe: Effective and Democratic? (Oxford, Oxford University Press, 1999). 242 de Sadeleer, above n 27, at 248–9. 239

(I) Fisher Ch6

31/5/07

09:59

Page 241

Conclusion

241

constitutionalism. It is that pressure which had led to the Commission’s Communication having such significant authority. It is the intractability of this tension which makes me so circumspect about Ladeur’s comment that the European administrative system may be well suited to enabling ‘learning processes, mutual comparison and experimentation with different types of relationships and co-ordination’.243 Such a comment, while admirable in its intent, is ignoring the normative complexities of European integration from an administrative perspective. On the one hand, administrative constitutionalism is so embedded in legal culture that mutual comparison and experimentation is difficult. On the other, the pressure for uniformity also discourages such mutual learning. I am perhaps being unduly negative. Whatever the case, there is a real need for greater engagement with these issues. Such engagement can be seen in places,244 but what is needed is a far more comprehensive debate about these issues.

VII Conclusion My aim in this chapter was to provide a study of the development and operation of the precautionary principle as a jurisprudential principle in the EU. That study has shown that the principle highlights that there is a real need to think more carefully about the role that administrative constitutionalism plays in the process of European integration. In particular, I have highlighted that there is a fundamental tension in the operation of the precautionary principle. On the one hand, thinking about the precautionary principle in the EU context highlights the complexity and diversity of Community and Member State administrative institutions, and thus discourses over administrative constitutionalism. At the same time, European integration promotes a single definition of the precautionary principle and thus a single understanding of administrative constitutionalism. Over the last several years that promotion of a single definition has resulted in a shift from the DC to the RI paradigm. While identifying this state of affairs I have not identified any answers or solutions to this tension beyond needing to take a harder look at these issues. What I have done, however, is to show that much of this has to do with the operation of discourses of administrative constitutionalism, and those discourses are operating far and beyond the realms of the national administrative state.

243 244

Ladeur, above n 74 at 1479. Everson, ‘above n 23; and Case C41/02 Commission v Netherlands, above n 42 per AG Maduro.

(I) Fisher Ch6

31/5/07

09:59

Page 244

(J) Fisher Conclusion

31/5/07

09:59

Page 245

7 Beyond the Science/Democracy Dichotomy This book has been a project of reorientation, and my aim has been to recast the way in which debates over technological risk regulation have been characterised. In particular, I have made two points. First, technological risk disputes are not disputes over whether ‘scientific’ or ‘democratic’ approaches are better ways to regulate risk but rather over what should be the role and nature of public administration in regulating risk. These debates raise issues about how public administration should be constituted, limited, and held to account and are thus debates over administrative constitutionalism. Those debates have polarised around two ideal types—the rational-instrumental (RI) and deliberativeconstitutive (DC) paradigms of administrative constitutionalism. Second, the role of law in these disputes is a significant one because administrative constitutionalism is part of the thickness of legal culture. The combined effect of these two points has been to provide a new lens through which to view technological risk disputes. The utility of that lens I have shown in the last five chapters by providing a series of case studies. Those case studies have been ‘snapshots’ in which I have shown how thinking about technological risk disputes in terms of administrative constitutionalism highlights new and important dimensions of those disputes. What I have not done is provide any answers to the many problems that I have identified in the different case studies. I have not done so because that was never the aim of this book. With that said, in this last, brief chapter it is useful to summarise some of the issues seen in these case studies, to reflect on these studies, and provide some tentative suggestions for next steps. This I do in the next three sections.

I A Summary of Analysis My starting point in chapter one was the science/democracy dichotomy. In writing this book it was not a starting point I originally chose, but this dichotomy so dominates present technological risk discourse that for many people it is the only way that decision-making about technological risk issues can be conceptualised. My description of that dichotomy will be, for many scholars, overly simplistic, but it has been such a governing theoretical construct that it is only through providing

(J) Fisher Conclusion

246

31/5/07

09:59

Page 246

Beyond the Science/Democracy Dichotomy

a rough sketch of it that one can see both its appeal and its irrelevance. What the case studies in this book overwhelmingly show is the latter. In all these case studies, decision-making has been a mix of expertise and democracy, science and values, and politics and knowledge. It has not been the case that there has been a stark choice to be made between ‘science’ or ‘democracy’, as all decision-making regimes encompass the two. Rather, debates have been over how these terms are defined, and what I have shown is that they can be defined in many different ways. Thus, for example, in chapter two I showed how the expertise of the Southwood Working Party was defined in both DC terms and RI terms, and in chapter three I showed how the concept of ‘substantial evidence’ could mean two different things. Likewise, in chapter five I showed that ‘risk assessment’ is also open to interpretation.1 What the science/democracy dichotomy has really blinded us all to is that technological risk evaluation is an activity of public administration. In other words, what this book has shown is that the science/democracy dichotomy is not only irrelevant but also an obstructive dichotomy.2 Thus, for example, much of the analysis of the BSE crisis has been so focused on demarcating the line between science and democracy that most commentators have ignored the fact that decisionmaking was the product of public administration and that there exists one of the most comprehensive studies of administrative decision-making to prove that fact.3 Likewise, in chapter five I illustrated how the dichotomy is distorting understandings about the World Trade Organisation (WTO) Sanitary and Phytosanitary Agreement (SPS Agreement). If I have one hope for this book it is that it moves debate beyond much of the fruitless discourse that is presently being carried out. That leads me on to the real starting point of this book—public administration. What can be seen in each of these case studies is that some form of public administration plays a central role in technological risk evaluation. Those forms differ dramatically and range from central government departments in the UK to independent agencies in the US to the web of committees that can be found in the European Union (EU). We should not be quick to confine our understandings of public administration to Weberian concepts of bureaucracy.4 Whilst these institutions differ in their nature and structure, what they have in common is that they 1 This is a point I have also made in E Fisher, ‘The Rise of the Risk Commonwealth and the Challenge for Administrative Law’ (2003) Public Law 455; and E Fisher, ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds), Environmental Law for Sustainability (Oxford, Hart Publishing, 2006). 2 See H Collins and R Evans, ‘The Third Wave of Science Studies: Studies of Expertise and Experience’ (2002) 32 Social Studies of Science 235 and the responses to it for a similar recognition of this problem. Those responses include S Jasanoff, ‘Breaking the Waves in Science Studies: Comment on H.M. Collins and Robert Evans, “The Third Wave of Science Studies”’ (2003) 33 Social Studies of Science 389, and B Wynne, ‘Seasick on the Third Wave? Subverting the Hegemony of Propositionalism: Response to Collins & Evans (2002)’ (2003) 33 Social Studies of Science 401. 3 Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report (London, HMSO, 2000). 4 M Weber, From Max Weber: Essays In Sociology (London, Routledge, 1991) at ch 7.

(J) Fisher Conclusion

31/5/07

09:59

Page 247

A Summary of Analysis

247

are all public bodies which have been delegated their responsibilities from a primary law-maker. That process of delegation may also take many forms. Thus in the EU it takes a form so Byzantine that few scholars have been able to make sense of it and, when they do, they do so in different ways.5 The reality is that while public administration is highlighted in much empirical work on technological risk decision-making it has been virtually written out of more theoretical and prescriptive narratives.6 Indeed, recognising public administration as a fundamental feature of contemporary democracies is a very academically unfashionable thing to do. What literature there is on it, and the related concept of governance, tends to be focused on its democratisation rather than upon accepting its unelected nature as a given.7 This is not surprising—the handing of power to an unelected institution in a democracy committed to liberal constitutionalism is an anathema to democracy and the rule of law. But this book should not be read as my wanting to hand power to public administration.8 Rather, what I am arguing is that we need to recognise that technological risk decision-making is inescapably the province of public administration and that public administration faces an ongoing legitimacy problem. As I noted in chapter one, technological risk evaluation raises directly the challenge posed by Ackerman of needing to develop a ‘constitutional design’ that reconciles administrative power and liberal constitutionalism.9 This is the real issue—the animating of normative ideals about public administration and how it should be constituted, limited, and held to account so as to ensure its legitimacy. What I have shown in this book is that it is those ideals which have shaped the process of technological risk evaluation. The significance of those ideals is due to the fact that governing is an inherently normative enterprise, and the evolution of the administrative state has been an evolution of different theories about its role and nature.10 Technological risk disputes have been part of that evolution. Thus the creation of the large-scale science bureaucracies in the US has been a central feature of the development of US Federal government.11 Likewise, the growth of technological risk regulation regimes in the EU has given rise to a more general debate about EU governance.12 In all these debates it has been ideals 5 Some of these theories are well summarised in P Craig, EU Administrative Law (Oxford, Oxford University Press, 2006) at ch 4. 6 An exception is S Breyer, Breaking the Vicious Circle: Towards Effective Risk Regulation (Cambridge, Harvard University Press, 1993). 7 I make this point in relation to EU governance debates in E Fisher, ‘The European Union in the Age of Accountability’ (2004) 24 Oxford Journal of Legal Studies 495. 8 Of course, there are those that argue that such acceptance is doing exactly that: see G Frug, ‘The Ideology of Bureaucracy in American Law’ (1984) 97 Harvard Law Review 1276. 9 B Ackerman, ‘The New Separation of Powers’ (2000) 113 Harvard Law Review 633 at 696. See section II.B of ch 1 for discussion on this point. 10 P Craig, Public Law and Democracy in the United Kingdom and the United States of America (Oxford, Clarendon Press, 1990). 11 C Sunstein, After the Rights Revolution (Cambridge, Harvard University Press, 1990). 12 Commission of the European Communities, European Governance: A White Paper, COM(2001) 428 final.

(J) Fisher Conclusion

248

31/5/07

09:59

Page 248

Beyond the Science/Democracy Dichotomy

which have been influential rather than actualities, and to capture the normative, pluralistic, and significant nature of those theories, and debate about those theories, I have used the term ‘administrative constitutionalism’. That concept denotes the normative and ‘essentially contested’13 nature of public administration, and how discourses over it are separate from the meta discourses of constitutionalism. Moreover, I have provided an account of two ‘paradigms’ of administrative constitutionalism—the RI and DC paradigms. These paradigms are rough sketches and are more vehicles for elucidation and understanding than accurate depictions of reality.14 This is particularly because, as shown in this book, they mean very different things in different legal cultures. Moreover, while these two paradigms are a useful starting point I do not think that they exhaust the possible normative understandings of public administration. My analysis has not just been limited to the nation state, and one of the most significant features of administrative constitutionalism that I have shown is that discourses concerning it are operating beyond the confines of national jurisdictions. In chapter five I showed how paradigms of administrative constitutionalism shaped how the WTO SPS Agreement has been interpreted and in chapter six I showed how paradigms of administrative constitutionalism are operating in many different contexts in the multi-level polity of the EU. Administrative constitutionalism is not operating as an autonomous discourse, however. Rather, it is, in nearly all cases, part of legal culture, and the implications of this are that law is not only playing a significant role in constituting, limiting, and holding public administration to account but is also the discourse in which these debates are being carried out. Moreover, these discourses are shaping how technological risks are characterised and evaluated. Thus in chapter three we saw that the Supreme Court’s decision in Industrial Union Dept AFL-CIO v American Petroleum Institute (the Benzene decision)15 directly led to the National Research Council report Risk Assessment in the Federal Government: Managing the Process16—a report which has shaped decision-making not only in the US but also in other legal cultures, including the WTO.17 Likewise, as seen in chapter four, understandings of merits review have had a direct impact on how the precautionary principle has been defined in Australia. To recognise the legal nature of administrative constitutionalism is thus not to relegate it to the realm of lawyers but to recognise the importance of legal culture and open it up as a discourse to scrutiny. This book has been about that process of opening up, and it has highlighted three main things about administrative constitutionalism as a form of legal culture. First, how debates over administrative constitutionalism manifest themselves 13

WB Gaillie, ‘Essentially Contested Concepts’ (1956) 56 Proceedings of the Aristotelian Society 167. In this regard, I am thinking of them in terms of the use of models in some areas of scientific theory. See M Hesse, Models and Analogies in Science (Notre Dame, University of Notre Dame Press, 1966). 15 Industrial Union Dept AFL-CIO v American Petroleum Institute 448 US 607 (1980). 16 National Research Council, Risk Assessment in the Federal Government: Managing the Process (Washington DC, National Academy Press, 1983). 17 See ch 5 above, and Fisher, above n 1. 14

(J) Fisher Conclusion

31/5/07

09:59

Page 249

A Summary of Analysis

249

will vary from legal culture to legal culture. Thus I showed that in the US much of the debate was carried on in the context of judicial review, while in the UK there has been very little ‘conventional’ legal discourse. In contrast, debate in Australia has been carried on in the sui generis realm of environmental courts and generalist administrative tribunals. Furthermore, in the WTO context it is occurring in the forum of dispute settlement, which is still in the process of evolution. The notion that any simple comparison can be done between these different legal cultures is hopelessly naïve and I have not even begun to attempt such a study in this book. I chose these different case studies because I felt that as ‘snapshots’ of legal cultures they provided the best accounts of the diversity of disputes over administrative constitutionalism. The second thing I have highlighted is that administrative constitutionalism as a form of legal culture is, together with the rest of legal culture, a thick cultural phenomenon.18 Law is not just an instrumental set of rules but a culture unto itself, and what can be seen in several of these case studies is how the operation of aspects of legal culture not directly related to technological risk issues have shaped administrative constitutionalism discourse in relation to those issues. That can be seen most obviously in relation to adjudicative/adversarial procedure in chapters three and four but can also be seen in relation to how judges and those reviewing decisions perceive themselves in chapters three, four, and five. In each of these chapters, judges were reviewing and conceptualising technological risk evaluation with reference to general public law principles. Just as technological risk regulation has been closely intertwined with the development of the administrative state, so too have debates over technological risk evaluation been closely intertwined with public law. Indeed, as I noted above, technological risk evaluation, due its discretionary and contested nature, provides one of the most difficult conundrums for scholars of the administrative state. It is for this reason that it has become a focus for many mainstream administrative law scholars.19 The third, and most significant, thing I have shown, is that understandings of technological risk, public administration, and law have been co-produced. This is because technological risk decision-making is shaped by administrative constitutionalism, and administrative constitutionalism is a form of legal culture. This process of co-production was sketched in my descriptions of the RI and DC paradigms and can be seen in every case study. Thus I showed how understandings about the BSE ‘problem’ evolved alongside understandings about how public administration should deal with those problems. I illustrated in chapter three that different legal understandings of ‘hard look’ review resulted in different understandings of technological risk, and in chapter four that understandings of public administration and technological risk were co-produced alongside understandings 18

C Geertz, Local Knowledge (New York, Fontana Press, 1993). C Sunstein, Risk and Reason: Safety, Law and the Environment (Cambridge, Cambridge University Press, 2002); C Joerges, ‘Law, Science and the Management of Risks to Health at the National, European and International Level—Stories of Baby Dummies, Mad Cows and Hormones in Beef ’ (2001) 7 Columbia Journal of European Law 1; and Craig, above n 5 at ch 19. 19

(J) Fisher Conclusion

250

31/5/07

09:59

Page 250

Beyond the Science/Democracy Dichotomy

of merits review. In chapter five I showed that understandings of administrative constitutionalism and understandings of trade law were co-produced, and in chapter six I illustrated the multi-level co-production of understandings of law and public administration. These processes of co-production and the shifts between them are unmistakable and thus, while one may wish to quibble with the details of the RI and DC paradigms, what is readily apparent is the close interrelationship between how technological risks are understood, how they are evaluated by public administration, and how public administration is legally legitimated.

II Some Preliminary Findings My discussion so far has largely been a restatement of my argument in chapter one. The question arises whether the case studies in this book, as a whole, raise other issues. They do, and I consider some of those in this section. Again, it should be stressed that I do not see the purpose of this book as coming to any final conclusions of the interrelationship between risk regulation and administrative constitutionalism. With that said, it is useful to highlight five preliminary findings from these case studies. First, despite the prevalence of the science/democracy dichotomy in policy and scholarship, what actors really debate is the role of public administration. That can be seen in every chapter of this book. Thus, my argument has never been about changing practices of technological risk evaluation but rather about recognising them for what they really are. Technological risk evaluation is the province of public administration, and the promotion of different models of technological risk evaluation is really the promotion of different models of public administration. Moreover, conflicts over technological risk evaluation are conflicts over the role and nature of public administration. As this is the case, my ascribing the RI and DC paradigms to different points of view has not been difficult. I have simply examined the assumptions on which those points of views have been based and given expression to those assumptions. Recognising the significance of administrative constitutionalism in debates over technological risk evaluation does not thus require some dramatic rewriting of the whole field. Rather, it is about providing a more accurate lens through which to understand these debates. The second thing that can be seen in every chapter of this book is that the RI paradigm has been promoted over the DC paradigm. This finding will not be surprising to many and will lead them to conclude that this phenomenon is a product of the growing global dominance of RI public administration. Yet I would issue a word of caution, because the reasons for the dominance of the RI paradigm would seem to be different in every case. Thus in the context of the UK it has been a by-product of new public management reforms. In the US it has been brought about by similar reforms, but also by the curious legal features of technological risk regulation frameworks. In Australia the precautionary principle is being pro-

(J) Fisher Conclusion

31/5/07

09:59

Page 251

Some Preliminary Findings

251

moted in RI terms because of understandings of merits review, and in the WTO context such promotion is occurring for a range of reasons including a failure to appreciate that the SPS Agreement regulates administrative constitutionalism and not just science. Finally, in the EU setting such a promotion is occurring because of the authority being given to the European Commission’s Communication on the Precautionary Principle.20 I do not necessarily deny the existence of a grand RI narrative, but these case studies suggest a far more complex picture. Moreover, when one turns to theories that chart the dominance of RI thinking they ascribe this development to many different causes.21 If I did have to explain why the RI paradigm has dominated administrative constitutionalism discourses in these different legal cultures I would say it is because, as an ideal of public administration, it promises a simple model of control and accountability, and accountability is the obsession of the contemporary age.22 The simplicity and effectiveness of the model is, of course, open to question, but what can be seen in these different chapters is that the promotion of the RI paradigm was rarely due to some crude ideological deregulatory impulse but rather because courts, legislators, and administrators were trying to establish the legitimacy of public administration through restraining it. Thus, in the UK many of the reforms in the late 1980s were a response to what was seen as the undesirable Yes Minister culture.23 In the EU, the embracement of the RI paradigm has been part of the attempts to legitimise sprawling multi-level administrative frameworks. In a more complex situation, the RI paradigm has been promoted in Australia by environmental courts and generalist administrative tribunals as a means of restraining their own power. This is not to say that ideology has not had a role to play:24 many of the legal actors in these different chapters have exploited the RI paradigm,25 but the focus should not be on ideology at the expense of seeing how administrative constitutionalism as a form of legal culture is operating in a far more nuanced way. This relates to my third finding in this book, which is that there is no perfect answer to the legitimacy dilemmas of technological risk regulation. Both the RI and DC paradigms have their strengths and their weaknesses. The DC paradigm promises effective problem solving at the cost of forgoing a simple means of holding decision-makers to account. The RI paradigm promises accountability 20 Commission of the European Communities, Communication from the Commission on the Precautionary Principle, COM (2000) 1 final. 21 See my discussion of this point in Fisher, above n 1 at 114–15. 22 Fisher, above n 7; and E Fisher, ‘Drowning by Numbers: Standard Setting in Risk Regulation and the Pursuit of Accountable Public Administration’ (2000) 20 Oxford Journal of Legal Studies 109. 23 B Castle, ‘Mandarin Power’ in P Barberis (ed), The Whitehall Reader (Buckingham, Open University Press, 1996). 24 Clearly it has in relation to the precautionary principle: see L Kogan, Exporting Precaution: How Europe’s Risk Free Regulatory Agenda Threatens American Free Enterprise (Washington DC, Washington Legal Foundation, 2005); and G Marchant and K Mossman, Arbitrary and Capricious: The Precautionary Principle in the European Union Courts (Washington DC, The AEI Press, 2004). 25 A prime example being case-law in the US: see Competitive Enterprise Institute v NHTSA 956 F 2d 321 (DC Cir 1992); Competitive Enterprise Institute v NHTSA 901 F 2d 107 (DC Cir 1990); and Competitive Enterprise Institute v NHTSA 45 F 3d 481 (DC Cir 1995).

(J) Fisher Conclusion

252

31/5/07

09:59

Page 252

Beyond the Science/Democracy Dichotomy

and control but at the cost of effective problem solving. For those who see the solutions to the dilemmas of technological risk regulation lying in the adoption of an ideal model of risk evaluation then this book will have been a profound disappointment. Nothing in the previous chapters points to there being any perfect frameworks for technological risk evaluation. Rather, my findings point to the dangers of utopianism. Thus, for example, while I have highlighted many of the problems with the operation of the RI paradigm,26 I am not sure that the DC paradigm provides any simple solution to these problems either. Its operation in both the UK and the EU was not a resounding success in that the effectiveness of problem solving was counterbalanced by the fact that there was not much to stop public administration getting it horribly wrong. To paraphrase what one highly experienced public administrator once said to me years ago, ‘it is difficult to know when I am exercising wise discretion or just old man’s prejudice’. Likewise, we should be hesitant to embrace too quickly the problem-solving powers of deliberation.27 This is not to say that the DC paradigm is so flawed as to be unusable, and it is clear that public administration can be effectively held to account under it.28 My point is that just as we should not be naïve about the RI paradigm then we should not be naïve about the DC paradigm either. This is an important point because, in an era when the RI paradigm dominates technological risk evaluation, many commentators argue for a full embracement of a DC approach. All of this is because the legitimacy problems inherent in technological risk evaluation are insolvable. Indeed, and this will be cold comfort to those working at the coalface of technological risk evaluation, the most legitimate state of affairs is the simultaneous operation of both paradigms. This is what is occurring in many contexts,29 and, while resulting in risk evaluation being a rather messy practice, it guarantees the most acceptable balance between enabling and restraining public administration. Indeed, it is in those legal cultures such as the US where the RI paradigm dominates so absolutely that the most serious problems can be seen. In other words, the most legitimate technological risk regulation regimes are those in which decision-makers are subject to competing theories about their role and nature and, as such, we need to recognise the legitimacy and value of both the RI and the DC paradigms. Or—to put the point a different way—the legitimacy of public administration is derived from there being a vibrant and dynamic discourse of administrative constitutionalism. Indeed, paradigms of administrative constitutionalism are essentially forms of framing. As Jasanoff notes, the process of framing, provides an effective way of accommodating the solidity as well as the interpretative flexibility of the worlds in which policy gets made.30 26

See, in particular, chs 3 and 6. This point is highlighted in C Sunstein, Infotopia: How Many Minds Produce Knowledge (Oxford, Oxford University Press, 2006). 28 Natural Resources Defense Council v Nuclear Regulatory Commission 547 F 2d 633 (DC Cir 1976) 29 E Fisher, ‘Precaution, Law and Principles of Good Administration’ (2005) 52 Water Science and Technology 19. 30 S Jasanoff, Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005) at 26. 27

(J) Fisher Conclusion

31/5/07

09:59

Page 253

Some Preliminary Findings

253

Such a framing role for administrative constitutionalism is inevitable and the important point is that, as communities, we recognise the significance and plurality of those frames. Thus, for example, the legal operation of the precautionary principle in Australia has resulted in it having divergent definitions but what is also clear from that situation is that courts and decision-makers are thinking very hard about these issues. The fourth preliminary finding arising from the case studies in this book concerns the comparative study of technological risk regulation. This form of study has become a booming enterprise, particularly comparisons between the US and the EU, but much of it has been in crude terms. The focus has been on regulatory outcomes and/or the transplantation of different legal institutions or rules from one legal culture to another, and much of that scholarship ignores fundamental tenets of comparative law methodology.31 What these case studies show is that technological risk evaluation is so embedded in particular legal cultures that it takes on fundamentally different forms in each. Thus while the RI and DC paradigms are crude generalisations about the role and nature of public administration they are manifested in very different ways. Thus, for example, in Australia the RI paradigm has been a product of adjudicative procedure, in the UK a product of new public management, and in the WTO context a product of a focus on science. The differences in these disparities struck me several years ago when attending a Europe-wide public law conference on risk. While within each legal culture the focus was on public administration, the concerns were very different. Thus, to oversimplify, German public lawyers were concerned with issues of delegation, French public lawyers with issues of responsibility, and English public lawyers with neither of these things.32 It is not just the case that the rules are different, the preoccupations dissimilar, or that the institutions have divergent structures, but that in each legal culture the process of technological risk evaluation is occurring in different forums, and legal discourses are so divergently different as to be incomparable. Thus in the US, debate over administrative constitutionalism has occurred in forums traditionally understood as legal, while in the UK debates have occurred in far less legalised accountability forums. In contrast, again, in Australia debate has been carried out in arenas which are forms of state-sponsored legal pluralism. This is not to say comparison is impossible. If it was, I would not be writing this book, and the RI and DC paradigms would have been impossible to apply. What needs to be recognised is the framing power of administrative constitutionalism and legal culture. The framing of technological risk issues has been well charted by commentators such as Sheila Jasanoff, and this book is really just confirming how embedded those frames are.33 What this book has also shown is that legal culture 31

O Kahn Freund, ‘On Uses and Misuses of Comparative Law’ (1974) 37 Modern Law Review 1. E Fisher, ‘General Conclusions: Risk and the Challenges for Administrative Law’ (2003) 15 Review of European Public Law 707. 33 Jasanoff, above n 30 at 24–6; S Jasanoff, ‘Technological Risk and Cultures of Rationality’ in National Research Council (ed), Incorporating Science, Economics and Sociology in Developing Sanitary 32

(J) Fisher Conclusion

254

31/5/07

09:59

Page 254

Beyond the Science/Democracy Dichotomy

plays a powerful role in the framing process, and that framing is not just an ‘intensely social activity’ but also an intensely legal one.34 The final preliminary finding arising from the case studies in this book concerns globalisation. What can be seen in chapters five and six is that discourses of administrative constitutionalism do not disappear with the growing impact of trade law, the creation of transnational standard-setting bodies, and the evolution of a multilevel polity such as the EU. Rather, new forums are created and these forums do not displace national discourses but rather overlap and interact with them. That process of overlap and interaction is a profoundly complex one which I only began to chart in chapters five and six. What can be seen is not only that economic integration, no matter how minimal, must embody ideas of administrative constitutionalism, but that with the creation of each new transnational institution there is the creation of a new discourse of administrative constitutionalism which interrelates with national discourses. The end result is a web of incredible intricacy where norms of administrative constitutionalism operating in fundamentally different legal cultures are also being shared. I do not have any solutions to this state of affairs beyond the point I made in the introduction to Part Two of this book, which is that recognition of administrative constitutionalism in this context opens up a Pandora’s box of issues for scrutiny. This is particularly the case when the divergences of legal culture seen in Part One of this book are taken into account. I am not the only scholar opening that box up,35 and the case studies in this book highlight the fundamentally intractable nature of some of the issues that emerge. In particular, the relationship between national and transnational discourses of administrative constitutionalism is not a one-way street.36

III Next Steps So, where does that leave thinking about technological risk regulation and administrative constitutionalism? Again, I stress that this book is only a first step, and in this last section I propose five further next steps in study. While I describe these lines of inquiry as steps, I should stress that I see them as quite independent from and Phytosanitary Standards in International Trade: Proceedings of a Conference (Washington DC, National Academies Press, 2000) at 62–72; and D Winickoff et al, ‘Adjudicating the GM Food Wars: Science, Risk and Democracy in World Trade Law’ (2005) 30 Yale Journal of International Law 81 at 94. 34 Jasanoff, above n 30 at 24, although Jasanoff, I am sure, would not disagree with this: see S Jasanoff, Science at the Bar: Law, Science and Technology in America (Cambridge, Harvard University Press, 1995). 35 N Krisch, ‘The Pluralism of Global Administrative Law’ (2006) 17 European Journal of International Law 247; and C Harlow, ‘Global Administrative Law: The Quest for Principles and Values’ (2006) 17 European Journal of International Law 187. 36 See also S Jasanoff and M Long Martello (eds), Earthly Politics: Local and Global in Environmental Governance (Cambridge, MIT Press, 2004)

(J) Fisher Conclusion

31/5/07

09:59

Page 255

Next Steps

255

each other, except for the last one, which relates to all of the others. In tentatively suggesting these future lines of inquiry, I am also implicitly recognising the limits of this book but I am entirely comfortable with those limits. My aim has been one of reorientation through examining the same issue in different legal cultures. It is a first, not a last, word on the issue. The first possible next step of inquiry is to broaden the examination of technological risk disputes in the different legal cultures that have been studied in this book. What I have provided in each of these chapters are ‘snapshots’ of disputes in these legal cultures. These ‘snapshots’ have been chosen because they each show starkly the significant role of administrative constitutionalism. What is also clear is that discourse over administrative constitutionalism is not isolated to the issues I have examined. Not only for the sake of space and clarity of argument have I left much material out but also technological risk regulations are too sprawling and varied in any legal culture to be subject to a single analysis. There is thus a need to dig deeper into each legal culture seen here. Part of that process of digging is to consider also how discourses of administrative constitutionalism in relation to technological risk regulation interrelate with broader discourses of administrative constitutionalism and with discourses of administrative constitutionalism in other areas of state activity such as criminal justice. The former interrelationship I have briefly touched upon in some chapters,37 and the latter is particularly important in an era in which ideas of risk evaluation are being shifted from one realm of public life to another.38 Thus, for example, the concept of the precautionary principle is increasingly being advocated in the terrorism and criminal justice context.39 As I have expressed elsewhere,40 I have grave doubts about the ability to transfer ideas of risk evaluation and administrative constitutionalism from one regulatory context to another because it is obviously clear that the role and nature of public administration in these different contexts is so fundamentally different as to be incomparable.41 But that difference makes such a study all the more important. A second future step of inquiry is to look at how the interface between technological risk regulation and administrative constitutionalism operates in different legal cultures. My focus here has been a very narrow one—Anglo-American legal cultures and the sui generis legal cultures of the WTO and EU. Could such interrelationships also be found in legal systems remarkably different from these, such as those in Europe, Africa, South America, and Asia? I suspect that they could be, but in very different forms, and this once again emphasises the challenges of interrelating these different discourses in the transnational context. With that said, one 37

See chs 2 and 3. Fisher, above n 1. 39 C Sunstein, Laws of Fear: Beyond the Precautionary Principle (Cambridge, Cambridge University Press, 2005) at ch 9, discussing this shift. 40 Fisher, above n 1. 41 Compare the studies in this book with issues in the criminal justice and terrorism area: see L Zedner, ‘Securing Liberty in the Face of Terror: Reflections from Criminal Justice’ (2005) 32 Journal of Law and Society 507. 38

(J) Fisher Conclusion

256

31/5/07

09:59

Page 256

Beyond the Science/Democracy Dichotomy

also needs to be wary of the fact that accountability as understood in this book is perhaps curiously ‘Anglo-Saxon’ in nature, and care should be taken not to drag it into cultures where it does not naturally belong.42 This is not to say that legal cultures are inviolable and that there should not be any transferring of concepts from one culture to another—chapter six shows the incorrectness of such an assumption—but that in making sense of administrative constitutionalism in a legal culture a broad-minded approach needs to be taken. The third future step of inquiry is in relation to the role of private bodies. The analysis in this book has focused upon public administration but it is also the case that private institutions also play a role in technological risk evaluation, particularly in the form of industry standardisation bodies.43 An interesting question is whether their private nature means that they do not operate against a background of understandings of administrative constitutionalism. I suspect not, and elsewhere I argued that the privatisation of regulatory decision-making does not oust ideas of administrative constitutionalism but only makes them more complex.44 The legitimacy of delegation to these bodies must depend upon prescriptive theories about their role and nature, and these theories are theories of administrative constitutionalism, albeit adapted to the private law sphere.45 The fourth possible next step is in relation to the transnational context. There is clearly a need to study more closely the interactions and interrelationships between the growing global network of forums in which administrative constitutionalism is being debated. There is also a need to study more closely the debates over administrative constitutionalism in relation to institutions such as the Codex Alimentarius Commission. As noted in chapter five, some very thoughtful work has been done in this area,46 but there has not yet been a full recognition of the administrative nature of these organisations, and the focus has tended to be on constitutionalism writ large and processes of democratisation. The final and most important next step that needs to be taken is a further opening up of a conversation between law and the disciplines that relate to technological risk evaluation. Not only have lawyers and legal scholars treated science and technological risk as immutable black boxes 47 but other disciplines have treated 42 M McDonald, ‘Accountability, Anthropology, and the European Commission’ in M Strathern (ed), Audit Cultures: Anthropological Studies in Accountability, Ethics and the Academy (London, Routledge, 2000). 43 H Schepel, The Constitution of Private Governance (Oxford, Hart Publishing, 2005). 44 E Fisher, ‘Unpacking the Toolbox: Or Why the Public/Private Divide is Important in EC Environmental Law’ in M Freedland and J-B Auby (eds), The Public Law/Private Law Divide: Une entente assez cordiale? (Oxford, Hart Publishing, 2006). 45 For a thoughtful study of this in the context of the Private Finance Initiative in the UK see M. Freedland, ‘Public Law and Private Finance - Placing the Private Finance Initiative in a Public Law Frame’ (1998) Public Law 288. 46 M Livermore, ‘Authority and Legitimacy in Global Governance: Deliberation, Institutional Differentiation, and the Codex Alimentarius’ (2006) 81 New York University Law Review 766; and T Huller and M Maier, ‘Fixing the Codex? Global Food Safety Governance Under Review’ in C Joerges and E-U Petersmann (eds), Constitutionalism, Multi-Level Trade Governance and Social Regulation (Oxford, Hart Publishing, 2006). 47 Fisher, above n 1 at 466–70.

(J) Fisher Conclusion

31/5/07

09:59

Page 257

Conclusion

257

law in a ‘plug-and-play’ manner. Law is understood as a series of instrumental rules that are perceived to have little cultural thickness. The most obvious example of this is the keenness with which social scientists and policy-makers have described the precautionary principle as a ‘shifting of the burden of proof’.48 Chapter four is a perfect example of the naivety and dangers of such an approach. I do not pretend that the development of such a discourse is a simple thing, but this book goes a long way in proving the importance of such a conversation.

IV Conclusion In 1965 the great US administrative law scholar Louis Jaffe wrote ‘administration is not the remnant of an alien royalty but the most intimate and complete mirror of our people’.49 If this book has done anything, it is to prove the accuracy of this assertion. While public administration in the technological risk regulation context will never be democratic, what can be seen in each of these chapters is that how it is constituted, limited, and held to account does reflect normative expectations within a community about its role and nature. Admittedly, those expectations are operating through legal culture, but they are operating all the same. Moreover, those expectations are conflicting. Again, this is not surprising. To return to Jaffe: Administration, then, as the active principle of choosing, or preferring . . . has in it the inherent power to hurt, to awaken resentment, to stir the sense of injustice.50

In the technological risk evaluation context, that process of ‘choosing and preferring’ is not just simply favouring one interest over another but choosing one understanding of technological risk over another, one understanding of public administration over another, and one understanding of the role of law over another. As such, as a ‘complete mirror of a people’ it reflects all the complexity inherent in the messy business of governing risk.

48 49 50

See section V.C of ch 1 for a discussion of this. L Jaffe, Judicial Control of Administrative Action (Boston, Little, Brown, & Co, 1965) at 322. Ibid at 323.

(J) Fisher Conclusion

31/5/07

09:59

Page 260

(K) Fisher Biblio

31/5/07

09:59

Page 259

BIBLIOGRAPHY B ACKERMAN, ‘The New Separation of Powers’ (2000) 113 Harvard Law Review 633. B ACKERMAN and W HASSLER, Clean Coal/Dirty Air: Or How the Clean Air Act Became a Multimillion Dollar Bail Out for High Sulfur Coal Producers and What should be Done about it (New Haven, Yale University Press, 1981). Administrative Review Council, Better Decisions: Review of the Commonwealth Merits Review Tribunals, ARC 39 (Canberra, AGPS, 1993). —— Review of Commonwealth Environmental Impact Assessment Decisions: Discussion Paper (Canberra, AGPS, 1993). —— Environmental Decisions and the Administrative Appeals Tribunal, ARC 36 (Canberra, AGPS, 1994). Agriculture Committee, Salmonella in Eggs, HC 108-I (London, HMSO, 1989). M ALLARS, ‘Neutrality, the Judicial Paradigm and Tribunal Procedure’ (1991) 13 Sydney Law Review 377. K AMBROSE, ‘Science and the WTO’ (2000) 31 Law and Policy in International Business 861. E ANDERSON and C ST HILAIRE, ‘The Contrast between Risk Assessment and Rules of Evidence in the Context of International Trade Disputes: Can the US Experience Inform the Process?’ (2004) 24 Risk Analysis 449. D ANDREWS, Statement of Evidence to the BSE Inquiry (The BSE Inquiry/Statement 281, 1998). J APPLEGATE, ‘A Beginning Not an End In Itself: The Role of Risk Assessment in Environmental Decision Making’ (1995) 63 University of Cincinnati Law Review 1643. W ARMSTRONG, The Role and Character of the Civil Service (London, Oxford University Press, 1970). A ARNULL, ‘Introduction: The European Union’s Accountability and Legitimacy Deficit’ in A Arnull and D Wincott (eds), Accountability and Legitimacy in the European Union (Oxford, Oxford University Press, 2002). M ARONSON, ‘A Public Lawyer’s Response to Privatisation and Outsourcing’ in M Taggart (ed), The Province of Administrative Law (Oxford, Hart Publishing, 1997). M ARONSON et al, Judicial Review of Administrative Action (Sydney, Law Book Co, 2004). H ARTHURS, ‘Rethinking Administrative Law: A Slightly Dicey Business’ (1979) 17 Osgoode Hall Law Journal 1. —— ‘Without the Law’: Administrative Justice and Legal Pluralism in Nineteenth Century England (Toronto, University of Toronto Press, 1985). Attorney General’s Committee on Administrative Procedure, Final Report of the Attorney General’s Committee on Administrative Procedure (Washington DC, Government Printing Office, 1941). C AUBERACH, ‘Informal Rulemaking: A Proposed Relationship Between Administrative Procedures and Judicial Review’ (1977) 72 North Western University Law Review 15. Australian Law Reform Commission, Review of the Adversarial System of Litigation: Federal Tribunal Proceedings, Issues Paper No 24 (Canberra, 1998)

(K) Fisher Biblio

260

31/5/07

09:59

Page 260

Bibliography

I AYRES and J BRAITHWAITE, Responsive Regulation—Transcending the Deregulation Debate (New York, Oxford University Press, 1992). L AZOULAY, ‘The Court of Justice and the Administrative Governance’ (2001) 7 European Law Journal 425. N BARBER, ‘Legal Pluralism and the European Union’ (2006) 12 European Law Journal 306. C BARNARD, The Substantive Law of the EU: The Four Freedoms (Oxford, Oxford University Press, 2004). D BAZELON, ‘Coping with Technology through the Legal Process’ (1977) Cornell Law Review 817. —— ‘Science and Uncertainty: A Jurist’s View’ (1981) 5 Harvard Environmental Law Review 209. U BECK, Risk Society: Towards A New Modernity (London, Sage Publications, 1992). —— ‘The Reinvention of Politics: Towards a Theory of Reflexive Modernization’ in U Beck et al (eds), Reflexive Modernization: Politics, Tradition and Aesthetics in the Modern Social Order (Cambridge, Polity Press, 1994). —— ‘Risk Society and the Provident State’ in S Lash and B Szerszynski (eds), Risk, Environment and Modernity: Towards a New Ecology (London, Sage Publications, 1996). M BERNSTEIN, Regulating Business by Independent Commission (Princeton, Princeton University Press, 1955). H BEVIS, ‘Administrative Commissions and the Administration of Justice’ (1928) 2 University of Cincinnati Law Review 1. F BIGNAMI, ‘The Democratic Deficit in European Rulemaking: A Call For Notice and Comment in Comitology’ (1999) 40 Harvard International Law Journal 451. —— ‘Foreword’ (2004) 68 Law and Contemporary Problems 1. N BIGNOLD, ‘NSW Land and Environment Court—Its Contribution to Australia’s Development of Environmental Law’ (2001) 18 Environmental Planning and Law Journal 256. S BOEHMER-CHRISTIANSEN, ‘The Precautionary Principle in Germany: Enabling Government’ in T O’Riordan and J Cameron (eds), Interpreting the Precautionary Principle (London, Earthscan Publications, 1994). S BOEHMER-CHRISTIANSEN and J SKEA, Acid Politics: Environmental and Energy Policies in Britain and Germany (London, The Belhaven Press, 1991). J BOHANES, ‘Risk Regulation in WTO Law: A Procedure-Based Approach to the Precautionary Principle’ (2002) 40 Columbia Journal of Transnational Law 323. T BONYHADY, Places Worth Keeping: Conservationists, Politics and the Law (St Leonards, Allen & Unwin, 1993). B BOYER, ‘Alternatives to Administrative Trial Type Hearings for Resolving Complex Scientific, Economic and Social Issues’ (1972) 71 Michigan Law Review 111. J BRADBURY et al, ‘Trust and Public Participation in Risk Policy Issues’ in G Cvetkovich and R Lofstedt (eds), Social Trust and the Management of Risk (London, Earthscan, 1999). J BRAITHWAITE and P DRAHOS, Global Business Regulation (Cambridge, Cambridge University Press, 2000). J BRAND, Doctors and the State: The British Medical Profession and Government Action in Public Health, 1870–1912 (Baltimore, Johns Hopkins Press, 1965). G BRENNAN, ‘Twentieth Anniversary of the AAT: Opening Address’ in J McMillan (ed), The AAT Twenty Years Forward (Canberra, Australian Institute of Administrative Law, 1998).

(K) Fisher Biblio

31/5/07

09:59

Page 261

Bibliography

261

—— ‘The AAT: Early Issues’ (2001) 9 Australian Journal of Administrative Law 5. S BREYER, Breaking the Vicious Circle: Towards Effective Risk Regulation (Cambridge, Harvard University Press, 1993). V BROWN, ‘Top Down, Ground Up or Inside Out? Community Practice and the Precautionary Principle’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). L BUSCH et al, ‘Amicus Curiae Brief Submitted to the Dispute Settlement Panel of the World Trade Organisation in the case of EC: Measures Affecting the Approval and Marketing of Biotech Products’ (2004). C BUTTON, The Power to Protect: Trade, Health and Uncertainty in the WTO (Oxford, Hart Publishing, 2004). Cabinet Office, ‘The Armstrong Memorandum’ in P Barberis (ed), The Whitehall Reader (Buckingham, Open University Press, 1996). J CAMERON, ‘The Precautionary Principle: Core Meaning, Constitutional Framework and Procedures for Implementation’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). J CAMERON and J ABOUCHAR, ‘The Precautionary Principle: A Fundamental Principle for the Protection of the Global Environment’ (1991) 14 Boston College International and Comparative Law Review 1. P CANE, ‘Merits Review and Judicial Review—The AAT as a Trojan Horse’ (2000) 28 Federal Law Review 213. —— ‘Theory and Values in Public Law’ in P Craig and R Rawlings (eds), Law and Administration in Europe: Essays in Honour of Carol Harlow (Oxford, Oxford University Press, 2003). —— ‘The Making of Australian Administrative Law’ in P Cane (ed), Centenary Essays for the High Court of Australia (Sydney, LexisNexis Butterworths, 2004). Carnegie Commission on Science Technology and Government, Risk and The Environment: Improving Regulatory Decision Making (New York, 1993). R CARPENTER, ‘Limitations in Measuring Ecological Sustainability’ in T Trzyna (ed), A Sustainable World: Defining and Measuring Sustainable Development (Sacramento, IUCN/World Conservation Union, 1995). D CASS, The Constitutionalisation of the World Trade Organisation: Legitimacy, Democracy, and Community in the International Trading System (Oxford, Oxford University Press, 2005). B CASTLE, ‘Mandarin Power’ in P Barberis (ed), The Whitehall Reader (Buckingham, Open University Press, 1996). E CHADWICK, Report on the Sanitary Condition of the Labouring Population of Great Britain (Edinburgh, Edinburgh University Press, 1964). G CHAMBERS, ‘The BSE Crisis and the European Parliament’ in C Joerges and E Vos (eds), EU Committees: Social Regulation, Law and Politics (Oxford, Hart Publishing, 1999). A CHAYES, ‘The Role of the Judge in Public Law Litigation’ (1976) 89 Harvard Law Review 1281. Chief Inspector of the Alkali Inspectorate, Second Report of the Alkali Inspectorate (London, HMSO, 1866). —— Third Report of the Alkali Inspectorate (London, HMSO, 1867). —— Fourth Report of the Alkali Inspectorate (London, HMSO, 1868).

(K) Fisher Biblio

262

31/5/07

09:59

Page 262

Bibliography

E CHITI, ‘The Emergence of a Community Administration: The Case of European Agencies’ (2000) 37 Common Market Law Review 309. E CHRISTIE, ‘The Eternal Triangle: the Biodiversity Convention, Endangered Species Legislation and the Precautionary Principle’ (1993) 10 Environmental and Planning Law Journal 470. P CHRISTOFF, ‘Out of Chaos, a Shining Star?: Towards a Typology of Green States’ in J Barry and R Eckersley (eds), The State and the Global Ecological Crisis (Cambridge, MIT Press, 2005). T CHRISTOFOROU, ‘Settlement of Science-Based Trade Disputes in the WTO: A Critical Review of the Developing Case Law in the Face of Scientific Uncertainty’ (2000) 8 New York University Environmental Law Journal 622. —— ‘The Origins and Content of the Precautionary Principle in European Community Law’ in C Leben and J Verhoven (eds), Le Principle de Precaution—Aspects de Droit International et Communautaire (Paris, Pantheon Assas, LGDJ Diffuseur, 2002). L CLARKE and J SHORT, ‘Social Organization and Risk: Some Current Controversies’ (1993) 19 Annual Review of Sociology 375. J COHEN and A ARATO, Civil Society and Political Theory (Cambridge, MIT Press, 1992). H COLLINS and R EVANS, ‘The Third Wave of Science Studies: Studies of Expertise and Experience’ (2002) 32 Social Studies of Science 235. Commission of the European Communities, Communication on Consumer Health and Food Safety, COM (97)183 final. —— White Paper on Food Safety, COM (1999) 719 final. —— Communication from the Commission on the Precautionary Principle, COM (2000) 1 final. —— First Report on the Harmonisation of Risk Assessment Procedures (2000). —— European Governance: A White Paper, COM (2001) 428 final. —— Communication from the Commission on Impact Assessment, COM (2002) 276 final. Committee on Ministers’ Powers, Report of Committee of Ministers’ Powers, Cmd 4060 (London, HMSO, 1932). Committee on Safety and Health at Work, Health and Safety at Work (London, HMSO, 1972). Committee on the Environment, Public Health and Consumer Policy of the European Parliament, Report on the Commission Communication on the Precautionary Principle, A5-0352/2000 Final, 2000. B COOK, Bureaucracy and Self Government: Reconsidering the Role of Public Administration in American Government (Baltimore, Johns Hopkins University Press, 1996). C CORREA, ‘Implementing National Health Policies in the Framework of WTO Agreements’ (2000) 34 Journal of World Trade 89. T COTTIER and P MAVROIDIS (eds), The Role of the Judge in International Trade Regulation: Experience and Lessons for the WTO (Ann Arbor, University of Michigan Press, 2003). P CRAIG, Public Law and Democracy in the United Kingdom and the United States of America (Oxford, Clarendon Press, 1990). —— ‘The Nature of the Community: Integration, Democracy and Legitimacy’ in P Craig and G de Búrca (eds), The Evolution of EU Law (Oxford, Oxford University Press, 1998). —— ‘The Fall and Renewal of the Commission: Accountability, Contract and Administrative Organisation’ (2000) 6 European Law Journal 98. —— Administrative Law, 5th edn (London, Thompson/Sweet & Maxwell, 2003).

(K) Fisher Biblio

31/5/07

09:59

Page 263

Bibliography

263

—— EU Administrative Law (Oxford, Oxford University Press, 2006). R CRAMTON, ‘A Comment on Trial Type Hearings in Nuclear Power Plant Siting’ (1972) 58 Virginia Law Review 585. C CRANOR, Regulating Toxic Substances: A Philosophy of Science and the Law (Oxford, Oxford University Press, 1993). R CREYKE and J MCMILLAN, Control of Government Action: Text, Cases and Commentary (Sydney, LexisNexis Butterworths, 2005). H CROLY, Progressive Democracy (New York, Macmillan Co, 1915). G CVETKOVICH and R LOFSTEDT (eds), Social Trust and the Management of Risk (London, Earthscan, 1999). J DA CRUZ VILACA, ‘The Precautionary Principle in EC Law’ (2004) 10 European Public Law 369. R DAHL, ‘Democracy’ in N Smelser and P Baltes (eds), International Encyclopaedia of the Social and Behavioural Sciences (Oxford, Elsevier Science Ltd, 2001). F DAINTON, The Future of the Research Council System (The BSE Inquiry Report, Evidence M19 tab 2B, 1971). —— Doubts and Certainties: A Personal Memoir of the Twentieth Century (Sheffield, Sheffield Academic Press, 2001). W DAVEY, ‘The WTO Dispute Settlement System: The First Ten Years’ (2005) 8 Journal of International Economic Law 17. A DAVIES, Accountability: A Public Law Analysis of Government by Contract (Oxford, Oxford University Press, 2001). KC DAVIS, ‘An Approach to Problems of Evidence in the Administrative Process’ (1942) 55 Harvard Law Review 364. —— Administrative Law Treatise (St Paul, West Publishing, 1958). —— ‘English Administrative Law—An American View’ (1962) Public Law 139. —— Discretionary Justice: A Preliminary Enquiry (Urbana, University of Illinois Press, 1965). —— ‘Facts in Law-making’ (1980) 80 Columbia Law Review 931. KC DAVIS and R PIERCE, Administrative Law Treatise, 3rd edn (San Diego, KC Davis Publishing Co, 1994). J DAVISON, ‘Administration and Judicial Self-Limitation’ (1936) 4 George Washington Law Review 291. G DE BU´RCA, ‘Unpacking the Concept of Discrimination in EC and International Trade Law’ in C Barnard and J Scott (eds), The Law of the Single European Market: Unpacking the Premises (Oxford, Hart Publishing, 2002). G DE BU´RCA and J SCOTT, ‘The Impact of the WTO on EU Decision-Making’ in G de Búrca and J Scott (eds), The EU and the WTO: Legal and Constitutional Issues (Oxford, Hart Publishing, 2001). B DE MARCHI and J RAVETZ, ‘Risk Management and Governance: A Post-Normal Science Approach’ (1999) 31 Futures 743. N DE SADELEER, Environmental Principles: From Political Slogans to Legal Rules (Oxford, Oxford University Press, 2002). —— ‘The Precautionary Principle in EC Health and Environmental Law’ (2006) 12 European Law Journal 139. O DE SCHUTTER et al (eds), Governance in the European Union (Luxembourg, European Commission, 2001).

(K) Fisher Biblio

264

31/5/07

09:59

Page 264

Bibliography

R DEHOUSSE, ‘Misfits: EU Law and the Transformation of European Governance’ in C Joerges and R Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002). J DELONG, ‘Informal Rulemaking and the Integration of Law and Policy’ (1979) 65 Virginia Law Review 257. C DEMMKE and M UNFRIED, European Environmental Policy: The Administrative Challenge for Member States (Maastricht, European Institute of Public Administration, 2001). A DEVILLE and R HARDING, Applying the Precautionary Principle (Sydney, Federation Press, 1997). P DEVLIN, ‘The Common Law, Public Policy and the Executive’ [1956] Current Legal Problems 1. AV DICEY, ‘The Development of Administrative Law in England’ (1915) 31 Law Quarterly Review 148. S DILLION, ‘Observations on Trade Law and Globalisation’ (2005) 33 International Journal of Legal Information 103. D DINAN, An Ever Closer Union: An Introduction to European Integration, 2nd edn (Basingstoke, Macmillan, 1999). M DORF and C SABEL, ‘A Constitution of Democratic Experimentalism’ (1998) 98 Columbia Law Review 267. M DOUGLAS, Risk Acceptability According to the Social Sciences (London, Routledge & Kegan Paul, 1985). —— Risk and Blame: Essays in Cultural Theory (London, Routledge, 1992). M DOUGLAS and A WILDASKY, Risk and Culture (Berkeley, University of California Press, 1982). W DOUMA, ‘Fleshing Out the Precautionary Principle by the Court of First Instance’ (2003) 15 Journal of Environmental Law 372. S DOVERS and J HANDMER, ‘Ignorance, Sustainability and the Precautionary Principle’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). G DOWNES, ‘Expert Evidence: The Value of Single or Court-Appointed Experts’, Australian Institute of Judicial Administration: Expert Evidence Seminar (Melbourne, 2005). R DWORKIN, Taking Rights Seriously (London, Duckworth, 1977). D EASTWOOD, ‘Amplifying the Province of the Legislature: The Flow of Information and the English State in the Early Nineteenth Century’ (1989) 62 Historical Journal 275. M EDELSTEIN, ‘When the Honeymoon is Over: Environmental Stigma and Distrust in the Siting of a Hazardous Waste Disposal Facility in Niagara Falls, New York’ (1993) 5 Research Issues in Social Problems and Public Policy 75. C EDLEY, Administrative Law: Rethinking Judicial Control of Bureaucracy (New Haven, Yale University Press, 1990). Efficiency Unit, Improving Management in Government: The Next Steps (London, HMSO, 1988). M EGAN, Constructing a European Market: Standards, Regulation and Governance (Oxford, Oxford University Press, 2001). C-D EHLERMANN, ‘Reflections on the Appellate Body of the WTO’ (2003) 6 Journal of International Economic Law 695. C-D EHLERMANN and N LOCKHART, ‘Standard of Review in WTO Law’ (2004) 7 Journal of International Economic Law 491.

(K) Fisher Biblio

31/5/07

09:59

Page 265

Bibliography

265

D ELLIOTT, ‘Reinventing Rulemaking’ (1992) 41 Duke Law Journal 1490. D ELLIOTT et al, ‘Science, Agencies and the Courts: Is Three a Crowd?’ (2001) 31 Environmental Law Reporter 10125. K ERIKSON, A New Species of Trouble: The Human Experience of Modern Disasters (New York, WW Norton & Co, 1994). ESRC Global Environmental Change Programme, The Politics of GM Food: Risk, Science and Public Trust, Special Briefing No 5 (1999). European Council, Council Resolution on the Precautionary Principle (Nice European Council Meeting, Presidency Conclusions, 2000). European Institute of Public Administration, Governance by Committee: The Role of Committees in European Policy Making and Policy, Research Paper 00/GHA (Maastricht, European Institute of Public Administration, 2000). M EVERSON, ‘Administering Europe’ (1998) 36 Journal of Common Market Studies 195. —— ‘The Constitutionalisation of European Administrative Law: Legal Oversight of a Stateless Internal Market’ in C Joerges and E Vos (eds), EU Committees: Social Regulation, Law and Politics (Oxford, Hart Publishing, 1999). F EWALD, ‘Risk in Contemporary Society’ (2000) 6 Connecticut Insurance Law Journal 365. C FARINA, ‘The Consent of the Governed: Against Simple Rules for a Complex World’ (1997) 72 Chicago Kent Law Review 987. D FARRIER, ‘Factoring Biodiversity Conservation into Decision-Making Processes: The Role of the Precautionary Principle’ in R Harding and E Fisher (eds) Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). S FINER, The Life and Times of Sir Edwin Chadwick (London, Methuen & Co Ltd, 1952). D FIORINO, ‘Environmental Risk and Democratic Process: A Critical Review’ (1989) 14 Columbia Law Review 501. DE FISHER, Environmental Law: Text, Cases and Materials (Sydney, Law Book Company, 1993). E FISHER, ‘The Risks of Quantifying Justice: The Use of the Substantial Evidence Test in Judicial Review of OSHA Rule-Making’ in R Baldwin (ed), Law and Uncertainty, Risks and Legal Processes (The Hague, Kluwer Law International, 1997). —— Risk, Expertise and Judicial Review: Scope of Review and Decision-Making under Scientific Uncertainty, Thesis in Law Submitted for a D.Phil Degree (Oxford, University of Oxford, 1998). —— ‘Changing the Mindset? Implementing the Precautionary Principle in Australia’ (1999) 7 Research in Social Problems and Public Policy 183. —— ‘Drowning by Numbers: Standard Setting in Risk Regulation and the Pursuit of Accountable Public Administration’ (2000) 20 Oxford Journal of Legal Studies 109. —— ‘Is the Precautionary Principle Justiciable?’ (2001) 13 Journal of Environmental Law 317. —— ‘Precaution, Precaution Everywhere: Developing a “Common Understanding” of the Precautionary Principle in the European Community’ (2002) 9 Maastricht Journal of European and Comparative Law 7. —— ‘General Conclusions: Risk and the Challenges for Administrative Law’ (2003) 15 Review of European Public Law 707. —— ‘The Rise of the Risk Commonwealth and the Challenge for Administrative Law’ (2003) Public Law 455. —— ‘The European Union in the Age of Accountability’ (2004) 24 Oxford Journal of Legal Studies 495.

(K) Fisher Biblio

266

31/5/07

09:59

Page 266

Bibliography

E FISHER, ‘Precaution, Law and Principles of Good Administration’ (2005) 52 Water Science and Technology 19. —— ‘Beyond the Science/Democracy Dichotomy: The World Trade Organisation Sanitary and Phytosanitary Agreement and Administrative Constitutionalism’ in C Joerges and E-U Petersmann (eds), Transnational Trade Governance and Social Regulation: Tensions and Interdependencies (Oxford, Hart Publishing, 2006). —— ‘Risk and Environmental Law: A Beginner’s Guide’ in B Richardson and S Wood (eds), Environmental Law for Sustainability (Oxford, Hart Publishing, 2006). —— ‘Unpacking the Toolbox: or Why the Public/Private Divide is Important in EC Environmental Law’ in M Freedland and J-B Auby (eds), The Public Law/Private Law Divide: Une entente assez cordiale? (Oxford, Hart Publishing, 2006). —— ‘Opening Pandora’s Box: Contextualising the Precautionary Principle in the European Union’ in E Vos et al (eds), Uncertain Risks Regulated: National, EU and International Regulatory Models Compared (London, Routledge-Cavendish, forthcoming). E FISHER and R HARDING, ‘The Precautionary Principle: Towards a Deliberative, Transdisciplinary, Problem-Solving Process’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). —— —— ‘The Precautionary Principle and Administrative Constitutionalism: The Development of Frameworks for Applying the Precautionary Principle’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). E FISHER et al, ‘Implementing the Precautionary Principle: Perspectives and Prospects’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). E FISHER and P SCHMIDT, ‘Seeing the Blindspots in Administrative Law: Theory, Practice, and Rulemaking Settlements in the United States’ (2001) 30 Common Law World Review 272. J FITZGERALD, ‘Aspects of the System of Fact-Finding and Rules of Evidence’ in K Cole (ed), Administrative Law and Public Administration: Form vs Substance (Canberra, Australian Institute of Administrative Law, 1996). G FLEMING, ‘Administrative Review and the “Normative” Goal—Is Anybody Out There?’ (2000) 28 Federal Law Review 61. C FOSTER and C PLOWDEN, The State under Stress (Buckingham, Open University Press, 1996). J FRAIBERG and M TREBILCOCK, ‘Risk Regulation: Technocratic and Democratic Tools for Regulatory Reform’ (1998) 43 McGill Law Journal 835. J FRANK, If Men Were Angels: Some Aspects of Government in a Democracy (New York, Harper & Bros, 1942). F FRANKFURTER, ‘The Task of Administrative Law’ (1927) 75 University of Pennsylvania Law Review 614. —— The Public and its Government (New Haven, Yale University Press, 1930). M FREEDLAND, ‘Public Law and Private Finance—Placing the Private Finance Initiative in a Public Law Frame’ (1998) Public Law 288. —— ‘The Evolving Approach to the Public/Private Distinction in English Law’ in M Freedland and J-B Auby (eds), The Public Law/Private Law Divide: Une entente assez cordiale (Oxford, Hart Publishing, 2006). J FREEMAN, ‘Collaborative Governance in the Administrative State’ (1997) 45 UCLA Law Review 1.

(K) Fisher Biblio

31/5/07

09:59

Page 267

Bibliography

267

—— ‘The Private Role in Public Governance’ (2000) 75 New York University Law Review 543. J FREEMAN and L LANGBEIN, ‘Regulatory Negotiation and the Legitimacy Benefit’ (2000) 9 New York University Environmental Law Journal 60. E FREIDSON, Professionalism Reborn: Theory, Prophecy, and Policy (Cambridge, Polity Press, 1994). W FREUDENBURG, ‘Risk and Recreancy: Weber, the Division of Labour, and the Rationality of Risk Perceptions’ (1993) 71 Social Forces 909. W FREUDENBURG and S PASTOR, ‘Public Responses to Technological Risks: Towards A Sociological Perspective’ (1992) 33 Sociological Quarterly 389. H FRIENDLY, The Federal Administrative Agencies: The Need for Better Definition of Standards (Cambridge, Harvard University Press, 1962). G FRUG, ‘The Ideology of Bureaucracy in American Law’ (1984) 97 Harvard Law Review 1276. R FUCHS, ‘The Formulation and Review of Regulations under the Food Drug and Cosmetic Act’ (1939) 6 Law and Contemporary Problems 43. L FULLER, ‘The Forms and Limits of Adjudication’ (1978) 92 Harvard Law Review 353. Fulton Committee, The Civil Service—Volume One, Cmnd 3638 (London, HMSO, 1968). S FUNTOWICZ and J RAVETZ, ‘Three Types of Risk Assessment and the Emergence of Post Normal Science’ in S Krimsky and D Golding (eds), Social Theories of Risk (Westport, Praeger, 1992). WB GAILLIE, ‘Essentially Contested Concepts’ (1956) 56 Proceedings of the Aristotelian Society 167. B GALLIGAN and C FLETCHER, New Federalism: Intergovernmental Relations and Environmental Policy (Canberra, Resource Assessment Commission, 1993). D GALLIGAN, Due Process and Fair Procedures (Oxford, Clarendon Press, 1996). G GANZ, Administrative Procedures (London, Sweet and Maxwell, 1974). R GASKINS, Burdens of Proof in Modern Discourse (New Haven, Yale University Press, 1992). C GEERTZ, Local Knowledge (New York, Fontana Press, 1993). E GELHORN and G ROBINSON, ‘Perspectives on Administrative Law’ (1975) 75 Columbia Law Review 771. R GERODIMOS, ‘The UK BSE Crisis as a Failure of Government’ (2004) 82 Public Administration 911. O GERSTENBERG and C SABEL, ‘Directly-Deliberative Polyarchy: An Institutional Ideal for Europe?’ in C Joerges and R Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002). O GODARD, ‘The Precautionary Principle and Catastrophism on Tenterhooks: Lessons from Constitutional Reform in France’ in E Fisher et al (eds), Applying the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). J GRAHAM, ‘Making Sense of Risk’ (2000) 20 Risk Analysis 302. J GRAHAM and S HSIA, ‘Europe’s Precautionary Principle: Promise and Pitfalls’ (2002) 5 Journal of Risk Research 371. M GRANDO, ‘Allocating the Burden of Proof in WTO Disputes: A Critical Analysis’ (2006) 9 Journal of International Economic Law 615. M GRANT, Environmental Court Project. Final Report. (London, Department of Transport, Environment and the Regions, 2000). J GREENWALD, ‘WTO Dispute Settlement: An Exercise in Trade Law Legislation?’ (2003) 6 Journal of International Economic Law 113.

(K) Fisher Biblio

268

31/5/07

09:59

Page 268

Bibliography

T GREY, ‘Constitutionalism: An Analytical Framework’ in J Pennock and J Chapman (eds), Constitutionalism: Nomos XX (New York, New York University Press, 1979). R GROVE WHITE et al, Uncertain World: Genetically Modified Organisms, Food and Public Attitudes in Britain (Lancaster, Lancaster University, 1997). W GULLETT, ‘The Precautionary Principle in Australia: Policy, Law and Potential Precautionary EIAs’ (2000) 11 Risk: Health, Safety And Environment 93. W GULLETT et al, ‘Substantive Precautionary Decision-Making: The Australian Fisheries Management Authority’s Lawful Pursuit of the Precautionary Principle’ (2001) 2 Australasian Journal of Natural Resources Law and Policy 95. P GUMMETT, Scientists in Whitehall (Manchester, Manchester University Press, 1980). A GUZMAN, ‘Food Fears: Health and Safety at the WTO’ (2004) 45 Virginia Law Review 1. J HABERMAS, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy (Cambridge, Polity Press, 1996). R HAMILTON, ‘Procedure for the Adoption of Rules of General Applicability: the Need for Procedural Innovation in Administrative Rulemaking’ (1972) 60 California Law Review 1276. C HAMLIN, Public Health and Social Justice in the Age of Chadwick, Britain 1800–1854 (Cambridge, Cambridge University Press, 1998). J HAMMITT et al, ‘Precautionary Regulation in Europe and the United States: A Quantitative Comparison’ (2005) 25 Risk Analysis 1215. R HARDING, ‘Sustainability: Principles to Practice—Outcomes’, Fenner Conference on the Environment 1994 (Canberra, 13–16 November 1994). R HARDING and E FISHER, ‘The Precautionary Principle in Australia’ in T O’Riordan and J Cameron (eds), Interpreting the Precautionary Principle (London, Earthscan, 1994). —— ‘Introducing the Precautionary Principle’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). —— (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). C HARLOW, ‘Changing the Mindset: The Place of Theory in English Administrative Law’ (1994) 14 Oxford Journal of Legal Studies 419. —— Accountability in the European Union (Oxford, Oxford University Press, 2002). —— ‘Global Administrative Law: The Quest for Principles and Values’ (2006) 17 European Journal of International Law 187. C HARLOW and R RAWLINGS, Law and Administration, 2nd edn (London, Butterworths, 1997). S HARLOW, ‘Science-Based Trade Disputes: A New Challenge in Harmonizing the Evidentiary Systems of Law and Science’ (2004) 24 Risk Analysis 443. P HARREMOE¨S et al, ‘Twelve Late Lessons’ in P Harremoës et al (eds), The Precautionary Principle in the Twentieth Century: Late Lessons from Early Warnings (London, Earthscan, 2002). P HARREMOE¨S et al (eds), The Precautionary Principle in the Twentieth Century: Late Lessons from Early Warnings (London, Earthscan Publications, 2002). K HAWKINS, Environment and Enforcement (Oxford, Oxford University Press, 1984). B HAYES and C TRENORDEN, Combined Jurisdiction for Development Appeals in States and Territories (Canberra, AGPS, 1990). Health and Safety Executive, Reducing Risks, Protecting People (London, Health and Safety Executive, 1999). P HENNESSY, Whitehall (London, Pimlico, 2001).

(K) Fisher Biblio

31/5/07

09:59

Page 269

Bibliography

269

A HERWIG, ‘Transnational Governance Regimes for Foods Derived from Bio-Technology and their Legitimacy’ in C Joerges et al (eds), Transnational Governance and Constitutionalism (Oxford, Hart Publishing, 2004). M HESSE, Models and Analogies in Science (Notre Dame, University of Notre Dame Press, 1966). G HEWART, The New Despotism (London, Ernest Benn Ltd, 1929). V HEYVAERT, ‘Facing the Consequences of the Precautionary Principle in European Community Law’ (2006) 31 European Law Review 185. C HILSON, Regulating Pollution: A UK and EC Perspective (Oxford, Hart Publishing, 2000). S HINCHLIFFE, ‘ “Indeterminacy In-Decisions”—Science, Policy and Politics in the BSE Crisis’ (2001) 26 Transactions of the Institute of British Geographers 182. HM Government, Guidelines on Scientific Analysis in Policy Making (London, HMSO, 2005). J HOLDER, Environmental Assessment: The Regulation of Decision-Making (Oxford, Oxford University Press, 2005). Hoover Commission, Report on Organisation of the Executive Branch of Government (New York, McGraw-Hill Book Co Inc, 1949). P HOWARD, The Death of Commonsense: How Law is Suffocating America (New York, Random House, 1994). R HOWSE, ‘Democracy, Science, and Free Trade: Risk Regulation on Trial at the WTO’ (2000) 98 Michigan Law Review 2329. T HULLER and M MAIER, ‘Fixing the Codex? Global Food Safety Governance Under Review’ in C Joerges and E-U Petersmann (eds), Constitutionalism, Multi-level Trade Governance and Social Regulation (Oxford, Hart Publishing, 2006). Inquiry into BSE and v-CJD in the United Kingdom, The BSE Inquiry Report (London, HMSO, 2000). —— The BSE Inquiry Report, Volume One: Findings and Conclusions (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Two: Science (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Three: The Early Years (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Four: The Southwood Working Party, 1988–9 (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Six: Human Health (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Eight: Variant Creutzfeldt-Jakob Disease (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Ten: Economic Impact and International Trade (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Eleven: Scientists After Southwood (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Fourteen: Responsibilities for Human and Animal Health (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Fifteen: Government and Administration (London, HMSO, 2000). —— The BSE Inquiry Report, Volume Sixteen: Reference Material (London, HMSO, 2000). Interdepartmental Liaison Group on Risk Assessment, Use of Risk Assessment within Government Departments (London, Health and Safety Executive, 1996). —— The Precautionary Principle: Policy and Application (London, Health and Safety Executive, 2002).

(K) Fisher Biblio

270

31/5/07

09:59

Page 270

Bibliography

A IRWIN, Citizen Science (London, Routledge, 1995). —— ‘The Politics of Talk: Coming to Terms with the “New” Scientific Governance’ (2006) 36 Social Studies of Science 299. J JACKSON, The Jurisprudence of GATT and the WTO (Cambridge, Cambridge University Press, 2000). M JACOB and T HELLSTRÖM, ‘Policy Understanding of Science, Public Trust, and the BSE-CJD Crisis’ (2000) 78 Journal of Hazardous Materials 303. C JAEGER et al, Risk, Uncertainty, and Rational Action (London, Earthscan, 2001). L JAFFE, Judicial Control of Administrative Action (Boston, Little, Brown, & Co, 1965). P JAMES, Food Standards Agency: An Interim Proposal (1997). S JASANOFF, ‘The Problem of Rationality in American Health and Safety Regulation’ in Smith and Wynne (eds), Expert Evidence: Interpreting Science in the Law (London, Routledge, 1989). —— The Fifth Branch: Science Advisers as Policy Makers (Cambridge, Harvard University Press, 1990). —— Science at the Bar: Law, Science and Technology in America (Cambridge, Harvard University Press, 1995). —— ‘Civilization and Madness: The Great BSE Scare of 1996’ (1997) 6 Public Understanding of Science 221. —— ‘Technological Risk and Cultures of Rationality’ in National Research Council (ed), Incorporating Science, Economics and Sociology in Developing Sanitary and Phytosanitary Standards in International Trade: Proceedings of a Conference (Washington DC, National Academies Press, 2000). —— ‘Breaking the Waves in Science Studies: Comment on HM Collins and Robert Evans, “The Third Wave of Science Studies” ’ (2003) 33 Social Studies of Science 389. —— Designs on Nature: Science and Democracy in Europe and the United States (Princeton, Princeton University Press, 2005). —— ‘Law’s Knowledge: Science for Justice in Legal Settings’ (2005) 95 American Journal of Public Health S49. S JASANOFF and M LONG MARTELLO (eds), Earthly Politics: Local and Global in Environmental Governance (Cambridge, MIT Press, 2004). S JASANOFF and B WYNNE, ‘Science and Decision Making’ in S Rayner and E Malone (eds), Human Choice and Climate Change—Volume One (Pacific Northwest National Laboratory, Battelle Press, 1998). M JEFFREY, ‘Appropriateness of Judicial and Non Judicial Determination of Environmental Issues’ (1988) 5 Environmental Planning and Law Journal 265. C JOERGES, ‘Law, Science and the Management of Risks to Health at the National, European and International Level—Stories of Baby Dummies, Mad Cows and Hormones in Beef’ (2001) 7 Columbia Journal of European Law 1. —— ‘Deliberative Supranationalism: Two Defences’ (2002) 8 European Law Journal 133. C JOERGES and R DEHOUSSE, (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002). C JOERGES et al (eds), Integrating Scientific Expertise into Regulatory Decision-Making (BadenBaden, Nomos Verlagsgesellschaft, 1997). —— (eds), Mountain or MoleHill: A Critical Appraisal of the Commission White Paper on Governance (Florence, European University Institute, 2002).

(K) Fisher Biblio

31/5/07

09:59

Page 271

Bibliography

271

C JOERGES and J NEYER, ‘From Intergovernmental Bargains to Deliberative Political Processes: The Constitutionalisation of Comitology’ (1997) 3 European Law Journal 273. C JOERGES and E-U PETERSMANN (eds), Constitutionalism, Multilevel Trade Governance and Social Regulation (Oxford, Oxford University Press, 2006). J JONES and S BRONITT, ‘The Burden and Standard of Proof in Environmental Regulation: The Precautionary Principle in an Australian Administrative Context’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006). A JORDAN, ‘The Impact on UK Environmental Administration’ in P Lowe and S Ward (eds) British Environmental Policy and Europe (London, Routledge, 1998). J JOWELL, ‘Of Vires Or Vacuums: The Constitutional Context of Judicial Review’ in C Forsyth (ed), Judicial Review and the Constitution (Oxford, Hart Publishing, 2000). R KAGAN, Adversarial Legalism: The American Way of Law (Cambridge, Harvard University Press, 2003). O KAHN FREUND, ‘On Uses and Misuses of Comparative Law’ (1974) 37 Modern Law Review 1. D KELEMEN, ‘Suing for Europe: Adversarial Legalism and European Governance’ (2006) 39 Comparative Political Studies 101. R KENNEDY, ‘Risk, Democracy and the Environment’ (2000) 20 Risk Analysis 306. Kerr Committee, Commonwealth Administrative Review: Committee Report, Parliamentary Paper No 144 (Canberra, Commonwealth Government Publishing Office, 1971). B KINGSBURY et al, ‘The Emergence of Global Administrative Law’ (2005) 68 Law and Contemporary Problems 63. A KLINKE and O RENN, ‘A New Approach to Risk Evaluation and Management: Risk-Based, Precaution Based, and Discourse Based Strategies’ (2002) 22 Risk Analysis 1071. L KOGAN, Exporting Precaution: How Europe’s Risk Free Regulatory Agenda Threatens American Free Enterprise (Washington DC, Washington Legal Foundation, 2005). S KRIMSKY, ‘Epistemic Considerations on the Value of Folk-Wisdom in Science and Technology’ (1984) 3 Policy Studies Review 246. N KRISCH, ‘The Pluralism of Global Administrative Law’ (2006) 17 European Journal of International Law 247. R KUEHN, ‘The Environmental Justice Implications of Quantitative Risk Assessment’ (1996) University of Illinois Law Review 103. M LA TORRE, ‘Legal Pluralism as Evolutionary Achievement of Community Law’ (1999) 12 Ratio Juris 182. R LACEY, Unfit for Human Consumption: Food in Crisis (London, Grafton, 1992). M LACKO, ‘The Data Quality Act: Prologue to a Farce or a Tragedy?’ (2004) 53 Emory Law Journal 305. K-H LADEUR, ‘The Introduction of the Precautionary Principle into EU Law: A Pyrrhic Victory for Environmental and Public Health Law? Decision-Making under Conditions of Complexity in Multi-level Political Systems’ (2003) 40 Common Market Law Review 1455. Land and Environment Court Working Party, Report of the Land and Environment Court Working Party (Sydney, Land and Environment Court, 2001). J LANDIS, The Administrative Process (New Haven, Yale University Press, 1938). —— Report on Regulatory Agencies to the President Elect (Washington DC, 1960) B LATOUR, We Have Never Been Modern (Cambridge, Harvard University Press, 1993). —— Politics of Nature: How To Bring the Sciences Into Democracy (Cambridge, Harvard University Press, 2004).

(K) Fisher Biblio

272

31/5/07

09:59

Page 272

Bibliography

P LEGRAND, Fragments on Law as Culture (Deventer, WEJ Tjenkk Willink, 1999). P LEGRAND and R MUNDAY, Comparative Legal Studies: Traditions and Transitions (Cambridge, Cambridge University Press, 2003). K LEITNER and S LESTER, ‘WTO Dispute Settlement 1995–2004: A Statistical Analysis’ (2005) 8 Journal of International Economic Law 231. K LENAERTS, ‘Interlocking Legal Orders in the European Union and Comparative Law’ (2003) 52 International and Comparative Law Quarterly 873. —— ‘ ‘‘In the Union We Trust’: Trust Enhancing Principles of Community Law’ (2004) 41 Common Market Law Review 317. K LENAERTS and A VERHOVEN, ‘Institutional Balance as a Guarantee for Democracy in EU Governance’ in C Joerges and R Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002). H LEVENTHAL, ‘Environmental Decision Making and the Role of the Courts’ (1974) 122 University of Pennsylvania Law Review 509. R LEVIN, ‘Scope of Review Legislation: The Lessons of 1995’ (1996) 31 Wake Forest Law Review 647. ND LEWIS, Law and Governance (London, Cavendish Publishing, 2001). R LEWIS, Edwin Chadwick and the Public Health Movement 1832–1854 (London, Longmans, Green & Co, 1952). P LIGHT, The Tides of Reform: Making Government Work 1945–1995 (New Haven, Yale University Press, 1997). P LINDSETH, ‘Democratic Legitimacy and the Administrative Character of Supranationalism: The Example of the European Community’ (1999) 99 Columbia Law Review 628. W LIPPMANN, Drift and Mastery (New York, Mitchell Kennerley, 1914). G LITTLE, ‘BSE and the Regulation of Risk’ (2001) Modern Law Review 730. M LIVERMORE, ‘Authority and Legitimacy in Global Governance: Deliberation, Institutional Differentiation, and the Codex Alimentarius’ (2006) 81 New York University Law Review 766. N LONG, ‘Bureaucracy and Constitutionalism’ (1952) 46 American Political Science Review 808. T LOWI, The End of Liberalism: The Second Republic of the United States, 2nd edn (New York, WW Norton & Co, 1979). N LUHMANN, Risk: A Sociological Theory (New York, Aldine de Gruyter, 1993). N MACCORMICK, Questioning Sovereignty (Oxford, Oxford University Press, 1999). O MACDONAGH, ‘The Nineteenth Century Revolution in Government: A Reappraisal’ (1958) 1 Historical Journal 52. J MACGREGOR, Statement of Evidence to the BSE Inquiry (The BSE Inquiry/Statement 302, 1998). R MACLEOD (ed), Government and Expertise: Specialists, Administrators, and Professionals 1860–1919 (Cambridge, Cambridge University Press, 1988). —— ‘Introduction’ in R Macleod (ed), Government and Expertise: Specialists, Administrators and Professionals 1860–1919 (Cambridge, Cambridge University Press, 1988). C MACMAOLAIN, ‘Using the Precautionary Principle to Protect Public Health: Pfizer v Council ’ (2003) 28 European Law Review 723. R MACRORY (ed), Principles of European Environmental Law (Groningen, Europa Law Publishing, 2004).

(K) Fisher Biblio

31/5/07

09:59

Page 273

Bibliography

273

M MADURO, We, The Court: The European Court of Justice and the European Economic Constitution (Oxford, Hart Publishing, 1998). G MAJONE, Regulating Europe (London, Routledge, 1996). —— ‘Delegation of Regulatory Powers in a Mixed Polity’ (2002) 8 European Law Journal 319. —— ‘What Price Safety? The Precautionary Principle and its Policy Implications’ (2002) 40 Journal of Common Market Studies 89. —— Dilemmas of European Integration: The Ambiguities and Pitfalls of Integration by Stealth (Oxford, Oxford University Press, 2005). G MARCHANT and K MOSSMAN, Arbitrary and Capricious: The Precautionary Principle in the European Union Courts (Washington DC, The AEI Press, 2004). A MARCUS, Promise and Performance: Choosing and Implementing an Environmental Policy (Westport, Greenwood Press, 1980). D MARQUAND, Decline of the Public (Cambridge, Polity Press, 2004). W MARUYAMA, ‘A New Pillar of the WTO: Sound Science’ (1998) 32 International Lawyer 651. J MASHAW, Greed, Chaos and Governance: Using Public Choice to Improve Public Law (New Haven, Yale University Press, 1997). —— ‘Deconstructing Debate, Reconstructing Law’ (2002) 87 Cornell Law Review 682. —— ‘Law and Engineering: In Search of the Law-Science Problem’ (2003) 66 Law and Contemporary Problems 135. M MATSUSHITA et al, The World Trade Organization: Law, Practice and Policy, 2nd edn (Oxford, Oxford University Press, 2006). L MCDONALD, ‘The Rule of Law in the “New Regulatory State” ’ (2004) 33 Common Law World Review 197. M MCDONALD, ‘Accountability, Anthropology, and the European Commission’ in M Strathern (ed), Audit Cultures: Anthropological Studies in Accountability, Ethics and the Academy (London, Routledge, 2000). T MCGARITY, ‘Substantive and Procedural Discretion in Administrative Resolution of Science Policy Questions: Regulating Carcinogens in EPA and OSHA’ (1979) 67 Georgetown Law Journal 729. —— Reinventing Rationality: The Role of Regulatory Analysis in the Federal Bureaucracy (Cambridge, Cambridge University Press, 1991). —— ‘Some Thoughts on “De-Ossifying” the Rulemaking Process’ (1992) 41 Duke Law Journal 1385. —— ‘A Cost Benefit State’ (1998) 50 Administrative Law Review 7. —— ‘Our Science is Sound Science and Their Science is Junk Science: Science-Based Strategies for Avoiding Accountability and Responsibility for Risk-Producing Products and Activities’ (2004) 52 University of Kansas Law Review 897. C MCGOWAN, ‘Congress, Court and the Control of Delegated Power’ (1977) 77 Columbia Law Review 1119. T MCGRAW, Prophets of Regulation (Cambridge, The Belknap Press, 1984). C MCILWAIN, Constitutionalism: Ancient and Modern (Ithaca, Cornell University Press, 1947). J MCMILLAN, ‘Merit Review and the AAT: A Concept Develops’ in J McMillan (ed), The AAT—Twenty Years Forward (Canberra, Australian Institute of Administrative Law Inc, 1998).

(K) Fisher Biblio

274

31/5/07

09:59

Page 274

Bibliography

N MCNELIS, ‘The Role of the Judge in the EU and WTO: Lessons from the BSE and Hormones Cases’ (2001) 4 Journal of International Economic Law 189. D MCRAE, ‘What is the Future of WTO Dispute Settlement?’ (2004) 7 Journal of International Economic Law 3. D MERCER, A Question of Balance: Natural Resources Conflict Issues in Australia, 3rd edn (Sydney, Federation Press, 2000). E MILLSTONE et al, Science in Trade Dispute Related to Potential Risks: Comparative Case Studies, EUR 21301 EN (Seville, European Commission, Directorate General: Joint Research Centre, European Science and Technology Observatory, Institute for Prospective Technological Studies, 2004). Ministry of Reconstruction, Report of the Machinery of Government Committee, Cd 9230 (London, HMSO, 1918) G MOONEY, ‘Professionalisation in Public Health and the Measurement of Scientific Progress in Nineteenth Century England and Wales’ (1997) 10 Social History of Medicine 53. B MORGAN, ‘Turning Off the Tap: Urban Water Service Delivery and the Social Construction of Global Administrative Law’ (2006) 17 European Journal of International Law 215. MG MORGAN and M HENRION, Uncertainty: A Guide to Dealing with Uncertainty in Quantitative Risk and Policy Analysis (Cambridge, Cambridge University Press, 1991). J MORONE, The Democratic Wish: Popular Participation and the Limits of American Government, revised edn (New Haven, Yale University Press, 1998). J MORRIS (ed), Rethinking Risk and the Precautionary Principle (Oxford, ButterworthHeinemann, 2000). D MOSS, When All Else Fails: Government as the Ultimate Risk Manager (Cambridge, Harvard University Press, 2002). National Research Council, Risk Assessment in the Federal Government: Managing the Process (Washington DC, National Academy Press, 1983). —— Science and Judgement in Risk Assessment (Washington DC, National Academy Press, 1994). —— Understanding Risk: Informing Decisions in a Democratic Society (Washington DC, National Academy Press, 1996). D Nelken, ‘Disclosing/Invoking Legal Culture’ (1995) 4 Social and Legal Studies 437. —— ‘Towards a Sociology of Legal Adaptation’ in D Nelken and J Feest (eds), Adapting Legal Cultures (Oxford, Hart Publishing, 2001). —— ‘Using the Concept of Legal Culture’ (2004) 29 Australian Journal of Legal Philosophy 1. Note, ‘Judicial Review under the Occupational Safety and Health Act: The Substantial Evidence Test as Applied to Informal Rulemaking’ (1974) Duke Law Journal 459. H NOWOTNY et al, Re-thinking Science: Knowledge and the Public in an Age of Uncertainty (Cambridge, Polity Press, 2001). A O’NEILL, ‘The Rules of Evidence and Administrative Law’ in K Cole (ed), Administrative Law and Public Administration: Form vs Substance (Canberra, Australian Institute of Administrative Law, 1996). Obituary, ‘Sir Richard Southwood’ (2005) The Telegraph, 3 November. M OESCH, Standards of Review in WTO Dispute Resolution (Oxford, Oxford University Press, 2003).

(K) Fisher Biblio

31/5/07

09:59

Page 275

Bibliography

275

Office of Science and Technology, Guidelines 2000: Scientific Advice and Policy Making (London, HMSO, 2000). —— Code of Practice for Scientific Advisory Committees (London, HMSO, 2001). E ORTS, ‘Reflexive Environmental Law’ (1995) 89 Northwestern University Law Review 1227. D OSBORNE and T GAEBLER, Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector (New York, Plume, 1992). R PACKER, The Politics of BSE (Basingstoke, Palgrave MacMillan, 2006). N PAIN, ‘Environmental Decision-Making Processes’ (1996) 79 Canberra Bulletin of Public Administration 73. D PALMETER and P MAVROIDIS, Dispute Settlement in the World Trade Organisation, 2nd edn (Cambridge, Cambridge University Press, 2004). H PARRIS, Constitutional Bureaucracy: The Development of the British Central Administration since the Eighteenth Century (London, George Allen & Unwin Limited, 1969). P PASCUAL, ‘Wresting Environmental Decisions From an Uncertain World’ (2005) 35 Environmental Law Reporter 10539. J PAULWELYN, ‘Evidence, Proof and Persuasion in WTO Dispute Settlement’ (1998) 1 Journal of International Economic Law 227. —— ‘The WTO Agreement on Sanitary and Phytosanitary (SPS) Measures As Applied in the First Three Disputes: EC—Hormones, Australia—Salmon and Japan—Varietals’ (1999) 2 Journal of International Economic Law 641. —— ‘Does the WTO Stand for “Deference to” or “Interference with” National Health Authorities When Applying the Agreement on Sanitary and Phytosanitary Measures (SPS Agreement)?’ in T Cottier and P Mavroidis (eds), The Role of the Judge in International Trade Regulation: Experience and Lessons for the WTO (Ann Arbor, University of Michigan Press, 2003). —— ‘The Sutherland Report: A Missed Opportunity for Genuine Debate on Trade, Globalisation and Reforming the WTO’ (2005) 8 Journal of International Economic Law 329. —— ‘The Transformation of World Trade’ (2005) 104 Michigan Law Review 1. L PEARSON, ‘Merits Review Tribunals’ in R Creyke and J McMillan (eds) Administrative Law: The Essentials (Canberra, Australian Institute of Administrative Law Inc., 2002). W PEDERSON, ‘Formal Records and Informal Rulemaking’ (1975) 85 Yale Law Journal 38. J PEEL, ‘Risk Regulation under the WTO SPS Agreement: Science as an International Normative Yardstick’ (2004) Jean Monnet Working Papers 02/04. —— The Precautionary Principle in Practice: Environmental Decision-Making and Scientific Uncertainty (Sydney, Federation Press, 2005). O PEREZ, Ecological Sensitivity and Global Legal Pluralism (Oxford, Hart Publishing, 2004). D PETERSON, ‘Precaution: Principles and Practice in Australian Environmental and Natural Resources Management’ (2006) 50 Australian Journal of Agricultural and Resource Economics 469. R PIERCE, ‘Legislative Reform of Judicial Review of Agency Actions’ (1995) 44 Duke Law Journal 1110. —— ‘Seven Ways to Deossify Agency Rulemaking’ (1995) 47 Administrative Law Review 59. —— ‘Judicial Review of Agency Actions in a Period of Diminishing Resources’ (1997) 49 Administrative Law Review 61. T PORTER, Trust in Numbers: The Pursuit of Objectivity in Science and Public Life (Princeton, Princeton University Press, 1995).

(K) Fisher Biblio

276

31/5/07

09:59

Page 276

Bibliography

R POSNER, The Problems of Jurisprudence (Cambridge, Harvard University Press, 1990). —— ‘The Rise and Fall of Administrative Law’ (1997) 72 Chicago Kent Law Review 953. R POUND, ‘The Growth of Administrative Justice’ (1924) 2 Wisconsin Law Review 321. M POWER, The Audit Society: Rituals of Verification (Oxford, Oxford University Press, 1997). S PRECHAL, ‘Institutional Balance: A Fragile Principle With Uncertain Contents’ in T Heukels et al (eds), The European Union After Amsterdam (The Hague, Kluwer Law International, 1998). President’s Commission on the Accident at Three Mile Island, Final Report (Washington DC, Government Printing Office, 1980). Presidential/Congressional Commission on Risk Assessment and Risk Management, Framework for Environmental Health Risk Management—Volume One (Washington DC, 1997). B PRESTON, ‘Judicial Implementation of the Principles of Ecologically Sustainable Development in Australia and Asia’, Law Society of New South Wales Regional Presidents Meeting (Sydney, 21 July 2006). —— ‘Ecologically Sustainable Development in the Courts in Australia and Asia’, Paper for a seminar held at Buddle Findlay, Lawyers (Wellington New Zealand, 28 August 2006). —— ‘Adequacy of Environmental Impact Statements in New South Wales’ (1986) 3 Environmental and Planning Law Journal 194. —— Environmental Litigation (Sydney, Law Book Company Ltd, 1989). H PROKUDA, ‘The Precautionary Principle—A Case Study’, The Precautionary Principle (Institute of Environmental Studies, University of New South Wales, Sydney, 1993). S RAYNER and R CANTOR, ‘How Fair is Safe Enough?: The Cultural Approach to Societal Technology Choice’ (1987) 7 Risk Analysis 39. S RAYNER and E MALONE (eds), Human Choice and Climate Change—Volume Four (Pacific Northwest National Laboratory, Battelle Press, 1998). C REICH, The Greening of America (New York, The Penguin Press, 1970). O RENN et al, The Application of the Precautionary Principle in the European Union: Regulatory Strategies and Research Needs to Compose and Specify a European Policy on the Application of the Precautionary Principle (PrecauPri) (Stuttgart, Centre for Technology Assessment, 2003). R REVESZ, ‘Environmental Regulation, Ideology and the DC Circuit’ (1997) 83 Virginia Law Review 1717. L RHOMBERG, A Survey of Methods for Chemical Risk Assessment Among Federal Regulatory Agencies (Washington DC, National Commission for Risk Assessment and Management, 1997). H RICHARDSON, Democratic Autonomy: Public Reasoning About the Ends of Policy (New York, Oxford University Press, 2002). D ROBERTS, Victorian Origins of the British Welfare State (New Haven, Yale University Press, 1960). D ROBERTS ‘Sanitary and Phytosanitary Risk Management in the Post-Uruguay Round Era: An Economic Perspective’ in National Research Council (ed), Incorporating Science, Economics, and Sociology in Developing Sanitary and Phytosanitary Standards in International Trade: Proceedings of a Conference (Washington DC, National Academies Press, 2000). J ROBERTS, ‘The Professional Expert and Administrative Control’ (1929) 7 Public Administration 247.

(K) Fisher Biblio

31/5/07

09:59

Page 277

Bibliography

277

D ROBERTSON, The Penguin Dictionary of Politics, 2nd edn (London, Penguin, 1993). B ROBINSON, ‘Is “Cleaner Production” the Answer to the Precautionary Principle?’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). W RODGERS, ‘A Hard Look at Vermont Yankee: Environmental Law under Close Scrutiny’ (1979) 67 Georgetown Law Journal 699. B ROSAMOND, Theories of European Integration (Basingstoke, Macmillan Press, 2000). J ROSSI, ‘Participation Run Amok: The Costs of Mass Participation for Deliberative Agency Decision-Making’ (1997) 92 North Western University Law Review 173. —— ‘Bargaining in the Shadow of Administrative Procedure: The Public Interest in Rulemaking Settlement’ (2001) 51 Duke Law Journal 1015. ROTHSCHILD, Lord, The Organisation and Management of Government R&D (BSE Inquiry Report, Evidence M19 tab 2A, 1972). Royal Commission on Environmental Pollution, Air Pollution Control: An Integrated Approach, Fifth Report, Cmnd 6371 (London, HMSO, 1976). —— Best Practicable Environmental Option, Twelfth Report, Cmnd 310 (London, HMSO, 1988). —— Setting Environmental Standards, 21st Report, Cmnd 4053 (London, HMSO, 1998). Royal Society, Risk, Analysis, Perception and Management (London, Royal Society, 1992) P RYAN, ‘Court of Hope and False Expectations: Land and Environment Court 21 Years On’ (2002) 14 Journal of Environmental Law 301. C SAMPFORD, ‘Law, Institutions and the Public/ Private Divide’ (1991) 20 Federal Law Review 185. G SAMPSON, ‘Risk and the WTO’ in D Robertson and A Kellow (eds), Globalisation and the Environment: Risk Assessment and the WTO (Cheltenham, Edward Elgar, 2001). A SCALIA, ‘Vermont Yankee: The APA, the DC Circuit, and the Supreme Court’ (1978) Supreme Court Review 345. A SCALIA and F GOODMAN, ‘Procedural Aspects of the Consumer Product Safety Act’ (1973) 20 University of California Law Review 899. G SCHAÄFER, ‘Linking Member State and European Administrations: The Role of Committees and Comitology’ in M Andenas and A Turk (eds), Delegated Legislation and the Role of Committees in the EC (The Hague, Kluwer Law International, 2000). F SCHARPF, Governing in Europe: Effective and Democratic? (Oxford, Oxford University Press, 1999). H SCHEPEL, The Constitution of Private Governance (Oxford, Hart Publishing, 2005). P SCHMIDT, Lawyers and Regulation: The Politics of the Administrative Process (Cambridge, Cambridge University Press, 2005). B SCHWARTZ and W WADE, Legal Control of Government (Oxford, Clarendon Press, 1972). E SCHWARTZBAUER and S SHINDELL, ‘Cancer and the Adjudicative Process: The Interface of Environmental Protection and Toxic Tort Law’ (1988) 14 American Journal of Law and Medicine 1. Science Advisory Board, Reducing Risk: Setting Priorities and Strategies for Environmental Protection (Washington DC, US Environmental Protection Agency, 1990). A SCORDO, ‘A No-Frills Approach to Rulemaking Leaves Workers at Risk: OSHA’s Failed Attempt to Enact “Generic” Air Contaminants Standard’ (1994) 16 Journal of Products and Toxics Liability 167. C SCOTT, ‘Accountability in the Regulatory State’ (2000) 27 Journal of Law and Society 38.

(K) Fisher Biblio

278

31/5/07

09:59

Page 278

Bibliography

C SCOTT, ‘Analysing Regulatory Space: Fragmented Resources and Institutional Design’ (2001) Public Law 329. J SCOTT and J HOLDER, ‘Law and New Environmental Governance in the European Union’ in G de Búrca and J Scott (eds), Law and New Governance in the EU and the US (Oxford, Hart Publishing, 2006). J SCOTT and E VOS, ‘The Juridification of Uncertainty: Observations on the Ambivalence of the Precautionary Principle within in the EU and WTO’ in C Joerges and R. Dehousse (eds), Good Governance in Europe’s Integrated Market (Oxford, Oxford University Press, 2002). L SCOTT, ‘Evolution of Public Law’ (1932) 14 (3rd Ser) Journal of Comparative Legislation 163. M SEIDENFELD, ‘A Civic Republican Justification for the Bureaucratic State’ (1992) 105 Harvard Law Review 1512. M SHAPIRO, The Supreme Court and Administrative Agencies (New York, Free Press, 1968). —— Who Guards the Guardians: Judicial Control of Administration (Athens, University of Georgia Press, 1988). —— ‘A Golden Anniversary? The Administrative Procedure Act of 1946’ (1996) 3 Regulation 40. K SHRADER-FRECHETTE, Burying Uncertainty: Risk and the Case Against the Geological Disposal of Nuclear Waste (Berkeley, University of California Press, 1993). S SKOWRONEK, Building A New American State: The Expansion of National Administrative Capacities 1877–1920 (New York, Cambridge University Press, 1982). A-M SLAUGHTER, A New World Order (Princeton, Princeton University Press, 2004). P SLOVIC, The Perception of Risk (London, Earthscan, 2000). S SMITH, A Review of the Land and Environment Court, Briefing Paper 13/2001 (Sydney, Parliament of New South Wales, 2001). M SMITHSON, ‘Ignorance and Science’ (1993) Knowledge: Creation, Diffusion and Utilization 133. CP SNOW, Science and Government (London, Oxford University Press, 1961). H SOMSEN, ‘Discretion in European Community Environmental Law: An Analysis of ECJ Caselaw’ (2003) 40 Common Market Law Review 1413. R SOUTHWOOD, Statement to BSE Inquiry (The BSE Inquiry/Statement No 1, 1998). R SOUTHWOOD et al, Report of the Working Party on Bovine Spongiform Encephalopathy (London, Department of Health/Ministry of Agriculture, Fisheries and Food, 1989). J STEELE, ‘Participation and Deliberation in Environmental Law: A Problem Solving Approach’ (2001) 21 Oxford Journal of Legal Studies 415. A STEIN, Foundations of Evidence Law (Oxford, Oxford University Press, 2005). P STEIN, ‘Are Decision-Makers Too Cautious with the Precautionary Principle?’ (2000) 17 Environmental and Planning Law Journal 3. —— ‘A Cautious Application of the Precautionary Principle’ (2000) 2 Environmental Law Review 1. —— ‘Specialist Environmental Courts: the Land and Environment Court of New South Wales, Australia’ (2002) 4 Environmental Law Review 5. R STEINBERG, ‘Judicial Lawmaking at the WTO: Discursive, Constitutional, and Political Constraints’ (2004) 98 American Journal of International Law 247. R STEWART, ‘The Reformation of American Administrative Law’ (1975) 88 Harvard Law Review 1661.

(K) Fisher Biblio

31/5/07

09:59

Page 279

Bibliography

279

—— ‘The Discontents of Legalism: Interest Group Relations in Administrative Regulation’ (1985) Wisconsin Law Review 655. —— ‘Beyond Delegation Doctrine’ (1987) 36 American University Law Review 323. A STIRLING et al, ‘A Framework for the Precautionary Governance of Food Safety: Integrating Science and Participation in the Social Appraisal of Risk’ in E Fisher et al (eds), Implementing the Precautionary Principle: Perspectives and Prospects (Cheltenham, Edward Elgar, 2006) A STONE SWEET, The Judicial Construction of Europe (Oxford, Oxford University Press, 2004). K STRADER, ‘OSHA’s Air Contaminants Standard Revision Succumbs to Substantial Evidence Test’ (1993) 62 University of Cincinnati Law Review 351. C SUNSTEIN, ‘Interest Groups in American Public Law’ (1985) 38 Stanford Law Review 29. —— ‘Factions, Self Interest and the APA: Four Lessons Since 1946’ (1986) 72 Virginia Law Review 271. —— ‘Beyond the Republican Revival’ (1988) 97 Yale Law Journal 1539. —— After the Rights Revolution (Cambridge, Harvard University Press, 1990). —— The Partial Constitution (Cambridge, Harvard University Press, 1993). —— Free Markets and Social Justice (New York, Oxford University Press, 1997). —— Designing Democracy: What Constitutions Do (New York, Oxford University Press, 2001). —— ‘Regulating Risks After ATA’ (2001) Supreme Court Review 1. —— ‘Book Review: The Perception of Risk’ (2002) 115 Harvard Law Review 1119. —— The Cost Benefit State: The Future of Regulatory Protection (Washington DC, American Bar Association, 2002). —— Risk and Reason: Safety, Law and the Environment (Cambridge, Cambridge University Press, 2002). —— Laws of Fear: Beyond the Precautionary Principle (Cambridge, Cambridge University Press, 2005). —— Infotopia: How Many Minds Produce Knowledge (Oxford, Oxford University Press, 2006). C SUNSTEIN and E ULLMANN-MARGALIT, ‘Second-Order Decisions’ (1999) 110 Ethics 5. A SYKES, ‘The (Limited) Role of Regulatory Harmonisation in International Goods and Services Markets’ (1999) 2 Journal of International Economic Law 49. —— ‘Regulatory Protectionism and the Law of International Trade’ (1999) 66 University of Chicago Law Review 1. C TAPPER, Cross and Tapper on Evidence, 10th edn (London, Lexis-Nexis, 2004). Task Force on Legal Services and Procedure, Task Force Report on Legal Services and Procedure (Washington DC, Commission on Organization of the Executive Branch of the Government, 1955). G TAYLOR, ‘The New Administrative Law’ (1977) 51 Australian Law Journal 804. R THOMAS, ‘Deprofessionalisation and the Postmodern State of Administrative Law Pedagogy’ (1992) 42 Journal of Legal Education 75. A TOMKINS, The Constitution after Scott: Government Unwrapped (Oxford, Oxford University Press, 1998). —— Public Law (Oxford, Clarendon Press, 2003). J TOULMIN SMITH, Government By Commissions: Illegal and Pernicious (London, Sweet, 1849). —— Local Government and Centralisation (London, John Chapman, 1851).

(K) Fisher Biblio

280

31/5/07

09:59

Page 280

Bibliography

P TOYNE, The Reluctant Nation: Environment, Law and Politics in Australia (Sydney, ABC Books, 1994). J TRACHTMAN, ‘International Trade as a Vector in Domestic Regulatory Reform: Discrimination, Cost–Benefit Analysis, and Negotiations’ (2000) 24 Fordham International Law Journal 726. Treasury Solicitor, The Judge Over Your Shoulder: A Guide to Judicial Review for UK Government Administrators, 3rd edn (London, Treasury Solicitor’s Department, 2000). M TREBILCOCK and R HOWSE, ‘Trade Liberalization and Regulatory Diversity: Reconciling Competitive Markets with Competitive Politics’ (1998) 6 European Journal of Law & Economics 5. T TRIDIMAS, The General Principles of European Community Law, 2nd edn (Oxford, Oxford University Press, 2006). W TWINING, Rethinking Evidence: Exploratory Essays (Evanston, Northwestern University Press, 1994). UK Government, The Food Standards Agency: A Force for Change, Cm 3830 (London, HMSO, 1998). A UNDERWOOD, ‘Precautionary Principles Require Changes in Thinking About and Planning Environmental Sampling’ in R Harding and E Fisher (eds), Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). United States Environmental Protection Agency, Unfinished Business: A Comparative Assessment of Environmental Problems (Washington DC, 1987) M VAN ASSELT and E VOS, ‘The Precautionary Principle and the Uncertainty Paradox’ (2006) 9 Journal of Risk Research 313. P VAN ZWANENBERG and E MILLSTONE, BSE: Risk, Science and Governance (Oxford, Oxford University Press, 2005). P VAN ZWANENBERG and A STIRLING, ‘Risk and Precaution in the US and Europe’ (2004) 3 Yearbook of European Environmental Law 43. F VIBERT, Governance in the European Union: From Ideology to Evidence (London, European Policy Forum, 2001). D VICTOR, ‘The Sanitary and Phytosanitary Agreement of the World Trade Organisation: An Assessment After Five Years’ (2000) 32 New York University Journal of International Law and Politics 865. MJC VILE, Constitutionalism and the Separation of Powers, 2nd edn (Indianapolis, Liberty Fund, 1998). D VOGEL, ‘Risk Regulation in Europe and the US’ (2004) 3 Yearbook of European Environmental Law 1. A VON BOGDANDY, ‘Law and Politics in the WTO: Strategies to Cope with a Deficent Relationship’ (2001) 5 Max Planck Yearbook of United Nations Law 609. K VON MOLTKE, ‘The Vorsorgeprinzip in West German Environmental Policy’ in Royal Commission on Environmental Pollution (ed), Best Practicable Environmental Option— 12th Report (London, HMSO, 1988). E VOS, Institutional Frameworks of Community Health and Safety Legislation: Committees, Agencies and Private Bodies (Oxford, Hart Publishing, 1999). —— ‘Antibiotics, the Precautionary Principle and the Court of First Instance’ (2004) 11 Maastricht Journal of European and Comparative Law 187. W WAGNER, ‘The Science Charade in Toxic Risk Regulation’ (1995) 95 Columbia Law Review 1613.

(K) Fisher Biblio

31/5/07

09:59

Page 281

Bibliography

281

B WAHLSTROM, ‘The Precautionary Approach to Chemicals Management: A Swedish Perspective’ in C Raffensberger and J Tickner (eds), Protecting Public Health and the Environment: Implementing the Precautionary Principle (Washington DC, Island Press, 1999). P WALD, ‘Regulation at Risk: Are Courts Part of the Solution or Most of the Problem?’ (1994) 67 Southern California Law Review 621. —— ‘Judicial Review in Midpassage: The Uneasy Partnership Between Courts and Agencies Plays On’ (1997) 32 Tulsa Law Journal 221. J WALDRON, ‘Is the Rule of Law an Essentially Contested Concept (in Florida)?’ (2004) 21 Law and Philosophy 137. K WALKER, ‘Introduction’ in K Walker (ed), Australian Environmental Policy: Ten Case Studies (Sydney, UNSW Press, 1992). —— ‘Statist Developmentalism in Australia’ in K Walker and K Crowley (eds), Australian Environmental Policy 2: Studies in Decline and Devolution (Sydney, UNSW Press, 1999). V WALKER, ‘Keeping the WTO From Becoming a World Trans-scientific Organisation’ (1998) 31 Cornell International Law Journal 251. —— ‘The Myth of Science as a “Neutral Arbiter” for Triggering Precautions’ (2003) 26 Boston College International and Comparative Law Review 197. P WATERMAN, ‘Possible Implications of the Precautionary Principle for Australian Local Government’, The Precautionary Principle (Institute of Environmental Studies, University of New South Wales, Sydney, 1993). A WEALE et al, Environmental Governance in Europe (Oxford, Oxford University Press, 2000). S WEATHERILL, EU Consumer Law and Policy (Cheltenham, Edward Elgar, 2005). J WEBBER, ‘Culture, Legal Culture, and Legal Reasoning: A Comment on Nelken’ (2004) 29 Australian Journal of Legal Philosophy 27. M WEBER, From Max Weber: Essays in Sociology (London, Routledge, 1991). J WEILER, The Constitution of Europe: Do the New Clothes Have an Emperor? (Cambridge, Cambridge University Press, 1999). —— ‘The Rule of Lawyers and the Ethos of Diplomats: Reflections on the Internal and External Legitimacy of WTO Dispute Settlement’ (2002) 13 American Review of International Arbitration 177. J WEILER and M WIND (eds), European Constitutionalism Beyond the State (Cambridge, Cambridge University Press, 2003). T WEILER, ‘International Regulatory Reform Obligations’ (2000) 34 Journal of World Trade Law 71. K WERHAN, ‘Delegalising Administrative Law’ (1996) University of Illinois Law Review 423. AN WHITEHEAD, Science and the Modern World (Cambridge, Cambridge University Press, 1928). J WHITEHOUSE, ‘Will the Precautionary Principle Affect Environmental Decision-Making and Impact Assessment?’ in R Harding and E Fisher (eds) Perspectives on the Precautionary Principle (Sydney, Federation Press, 1999). J WIENER and M ROGERS, ‘Comparing Precaution in the United States and Europe’ (2002) 5 Journal of Risk Research 317. J WIGMORE, ‘Administrative Boards and Commissions: Are the Jury-Trial Rules of Evidence in Force for Their Inquiries?’ (1922) 17 Illinois Law Review 263. B WILLIAMS and A MATHENY, Democracy, Dialogue and Environmental Disputes: The Contested Languages of Social Regulation (New Haven, Yale University Press, 1995).

(K) Fisher Biblio

282

31/5/07

09:59

Page 282

Bibliography

D WILLIAMS, ‘Environmental Law and Democratic Legitimacy’ (1994) 4 Duke Environmental Law and Policy Forum 1. S WILLIAMS, ‘ “Hybrid Rulemaking” Under the APA: A Legal and Empirical Analysis’ (1976) 42 University of Chicago Law Review 401. D WINICKOFF et al, ‘Adjudicating the GM Food Wars: Science, Risk and Democracy in World Trade Law’ (2005) 30 Yale Journal of International Law 81. D WIRTH, ‘The Role of Science in the Uruguay Round and NAFTA Trade Disciplines’ (1994) 27 Cornell International Law Journal 817. D WIRTH and E SILBERGELD, ‘Risky Reform’ (1995) 95 Columbia Law Review 1857. D WOODHOUSE, In Pursuit of Good Administration: Ministers, Civil Servants and Judges (Oxford, Clarendon Press, 1997). J WRIGHT, ‘The Courts and the Rulemaking Process: The Limits of Judicial Review’ (1974) 59 Cornell Law Review 375. B WYNNE, ‘Uncertainty and Environmental Learning’ (1992) 2 Global Environmental Change 111. —— ‘Seasick on the Third Wave? Subverting the Hegemony of Propositionalism: Response to Collins & Evans (2002)’ (2003) 33 Social Studies of Science 401. L ZEDNER, ‘Securing Liberty in the Face of Terror: Reflections from Criminal Justice’ (2005) 32 Journal of Law and Society 507. A ZIMMERN, ‘Democracy and the Expert’ (1930) 1 Political Quarterly 7. S ZUCKERMAN, Advice and Responsibility: Romanes Lecture for 1975 (Oxford, Clarendon Press, 1975).

(O) Fisher Index

4/7/07

12:13

Page 283

INDEX accountability: age of, 251 alternative forms, 25 Anglo-Saxon concept, 256 deliberative-constitutive paradigm, 31, 33 destabilising nature, 25 European Union, 172 financial, 73 law and public administration, 23 mechanisms, 15 ministerial, 73 multiple, 25, 73 rational-instrumental-paradigm, 29–30, 33 see also administrative constitutionalism; adversarial legalism adaptive management, 149, 157 adjudication: generally, 14, 22, 38, 184, 249 hybrid rulemaking, 99 Interstate Commerce Commission, 93–4 merits review, 139 nature of, 37–9 precautionary principle, 146–54, 253 public administration, 37–9, 107 records, 102 role under rational-instrumental paradigm, 38 Administrative Appeals Tribunal, 54, 134, 136, 138, 147, 150, 151 administrative constitutionalism: dangers of utopianism, 252 democracy, 34 as a form of legal culture, 35–9, 44, 52–5, 56, 107, 121, 124, 165, 240, 245, 248, 249 as an essentially contested concept, 24 Australia, 54–5, 125–1 51 case studies as snapshots, 5, 52 criminal justice, 255 debated, 25, 100 defined, 23–6 European Union, 171–2, 207–241 globalisation, 205, 254, 165–241 governance, 171 importance of the concept of reasonableness, 26–7 link with democracy, 24 normative nature of, 24, 27, 248 other areas, 255 other models, 27 paradigms as specific to risk evaluation, 27 paradigms described and compared, 27–35, 250–1, 251–2

private actors, 256 reasons for promotion of one paradigm over another, 250–1, 253 relationship of paradigms to reality, 39 relationship with constitutionalism, 24, 92, 247 relationship with trade law, 166, 180–5, 213, 214–8, 246, 248 risk assessment, 177 role of accountability, 24–5 scope of review, 92–3 simultaneous operation of both paradigms, 252–3 technological risk disputes, 26, 249 utility of concept, 39–40, 51, 165, 245 World Trade Organisation, 173–206 see also administrative law; deliberativeconstitutive paradigm; public administration; rational-instrumental paradigm administrative law: Australia, 54–5, 56, 57–8,128–9, 133–40 European Union, 172, 231 normative nature, 23, 24, 27 relationship with constitutional law, 24, 53 scholarship, 24, 25, 46 United Kingdom, 52–3, 55, 57 United States, 53–4, 56, 57, 90–2 see also administrative constitutionalism; judicial review; law; public administration administrative pluralism, 240 Administrative Review Council, 147 adversarial legalism: European Union, 171 United States, 56, 90, 91, 100 adversarial process, see adjudication Alkali Inspectorate and legislation, 66, 67 analytical opportunism, 90, 121–2 arbitrary and capricious: standard of review, 91, 111, 121 Administrative Procedure Act, 96 hard look review, 101–7 see also scope of review Armstrong memorandum, 70 as low as reasonably practicable, 75, 80, 84 Ash Council, 97 Atomic Energy Commission, 105 Attorney General’s Committee on Administrative Procedure report, 95, 96–7 audit, 15

(O) Fisher Index

4/7/07

12:13

Page 284

284

Index

Australia Fisheries Management Authority, 145, 151–2 Australia, 125–61 administrative constitutionalism, 54–5 deliberative-constitutive paradigm, 128–30, 140–6, 157, 159–60 environmental impact assessment, 128–30 environmental law, 127–8 judicial review doctrine, 56 legal culture, 54–5, 56, 57–8 local government, 127 merits review tribunals, 133–5 planning law, 127 rational instrument paradigm, 146–54 role of law in risk evaluation 56, 57–8 see also merits review; precautionary principle Bazelon, David, 90, 101, 104–7, 121, 124 behavioural uncertainty, 8, 63 Benzene decision, 112–5, 116 see also Occupational Safety and Health Administration; substantial evidence best practicable means, 67, 75 bovine spongiform encephalopathy (BSE), 8, 62–4, 76 BSE crisis: accountability mechanisms, 73 administrative context of, 71–6 comparison with other issues, 77–8 European Union, 224–5 importance of expertise, 65–6, 85–6 informing of senior civil service, 76–7 risk to babies, 79, 83–4 role of delegated legislation, 64 role of Parliament, 73 role of public administration, 64–6 slaughter and compensation policy, 79, 82, 84 specified offal ban, 79, 82, 83, 85 statements of UK government, 82 see also Southwood Working Party BSE Inquiry, 61 burden of proof: adjudicative procedure, 45–6 explained, 44–6 precautionary principle, 44–6, 146–54, 158, 159, 160, 257 WTO dispute settlement,184, 187 see also adjudication; precautionary principle Central Veterinary Laboratory, 76 Chadwick, Edwin, 66 Chernobyl, 8 Chief Medical Officer, 74–5, 78 Chief Veterinary Officer, 74, 75–6, 76–7, 82–3 Codex Alimentarius Commission, 176, 182, 186, 256 command and control regulation, 15

communication: technological risk evaluation, 20–1 role under deliberative-constitutive paradigm, 31, 33 role under rational-instrumental paradigm, 28–9, 33 Communication on the Precautionary Principle, see precautionary principle comparative law methodology, 158 risk regulation, 165, 253–4 see also legal culture constitution, political, 53 constitutional law, 24 see also constitutionalism constitutionalism: relationship with administrative constitutionalism, 24, 92, 247 transnational context, 256 Consumer Product Safety Commission, 98 co-production: law and public administration, 96 law and society, 36 law, technological risk and public administration, 27, 32, 36, 90, 100, 107, 124, 165, 249–50 legal culture, 39 merits review and the precautionary principle, 140, 154–5, 161, 248, 250 science and culture, 13 trade law, administrative law and technological risk, 204, 250 see also administrative constitutionalism cost/benefit analysis, 20, 119, 123, 157, 223 hard look review, 117 precautionary principle, 227, 238 cost/benefit state, 121 courts, limits in risk evaluation, 17, 21 delegation: European Union, 171 private actors, 256 see also administrative constitutionalism deliberation, 31, 105–6 See also deliberative-constitutive paradigm deliberative supranationalism, 172, 223 deliberative-constitutive paradigm: appeal, 34 Australia, 128–30, 140–6, 157, 159–60 BSE crisis 65, 79–81 committees, 128 compared with rational-instrumental paradigm, 32–35 democracy 34 described, 30–32 environmental impact assessment, 129–30 European Union, 218–24, 239 merits review, 140–6 precautionary principle, 43, 140–6

(O) Fisher Index

4/7/07

12:13

Page 285

Index problems with, 34–5, 252 Southwood Working Party, 79–81 United Kingdom, 67–8, 71–3, 104–6 United States, 93–7, 104–6, 108–12 World Trade Organisation, 183, 187–8 see also administrative constitutionalism; precautionary reasoning democracy: links with administrative constitutionalism, 24 public administration and, 34 role under deliberative-constitutive paradigm, 33 role under rational-instrumental paradigm, 33 transnational context, 256 Department of Health, 59, 71–6, 83–4 Deregulation Initiative, 73, 84 discretion: necessity of, 21, 26 role under deliberative-constitutive paradigm, 30–1, 33 role under rational-instrumental paradigm, 29, 33 District of Columbia Circuit of the Federal Court of Appeals, 54, 89, 101 ecologically sustainable development, 147, 158 environmental courts, 54, 134–5 see also merits review environmental impact assessment: Australia, 128–30, 158–60 deliberative –constitutive paradigm, 129–30, 140, 141 environmental impact statements, 129–30, 143 see also fauna impact statements, species impact statements Environmental Protection Agency, 98, 102–3 essentially contested concepts, 24 see also administrative constitutionalism European Food Safety Authority, 229 European Free Trade Association, 229–31 European Union, 207–41 accountability, 172 administrative constitutionalism, 171–2 administrative integration, 214–8 Commission, 168, 224 committees, 168, 210, 215, 218, 222 Council, 168, 228 delegation, 171–2 deliberative-constitutive paradigm, 218–4, 239 economic constitution, 216 fundamental freedoms, 170, 210, 213, 215, 221–4, 231 general principles of Community law, 170, 234, 240 governance arrangements, 171–2, 214–8, 247–8 high level of protection, 210

285

implementation of Community law by Member States 212, 220, 235–6 integration, 210, 240–1, 254 interface with World Trade Organisation law, 213, 214 international obligations, 213 judicial review of Community institutions, 212, 218–20, 231–5 legal culture, 170, 214 legal supremacy, 170 legitimacy of institutions, 224, 240 majoritarianism, 216 Member States taking unilateral regulatory action, 212–3, 220–4, 236–8 new regulatory approaches, 210 Parliament, 168, 229 precautionary principle, 207–41 precautionary principle, contexts of operation, 212–4, 239 precautionary principle, post modern law 211 public administration, 171–2, 211–2, 214–8, 247 rational-instrumental paradigm, 224–38 responses to Communication on the precautionary principle, 228–9 role of law, 170–1 safeguard clause, 236–7 technological risk regulation, 165–6, 209–10 evidence law, 14 see also adjudication expertise: BSE crisis, 65–6 definition, 20 expert evidence law, 156 role in risk evaluation, 20 role under deliberative-constitutive paradigm, 30, 33, 65, 104 role under rational-instrumental paradigm, 28, 33, 65, 102 Southwood Working Party, 79–84 United Kingdom, 66–71 Factories Inspectorate, 66 fauna impact statements, 140, 142 food safety crises, 63, 70 Food Standards Agency, 86–7 framing, 10, 64, 252–4 see also administrative constitutionalism; co-production Fulton Report, 70 generalist civil service, 57, 67, 69, 70, 71, 74 globalisation, 4,11 challenges for scholarship, 166–7, 254 legal culture, 37, 205 public administration, 21 technological risk regulation regimes, 37

(O) Fisher Index

4/7/07

12:13

Page 286

286

Index

governance, 3, 11, 17, 166 administrative constitutionalism, 25, 171 European Union, 171–2 public administration, 21 role of law, 15 Great Barrier Reef Marine Park Authority, 150 hard look review, 89, 101–7, 249 1980s, 117 analytical methodology, 117 deliberative-constitutive paradigm, 104–6 rational-instrumental paradigm, 101–3 see also administrative law; scope of review Henry VIII clause, 72 Hoover Commission report, 97, 102 ideology, 251 information: role in risk evaluation, 19 role under deliberative-constitutive paradigm, 30, 33 role under rational-instrumental paradigm, 29, 33 interest representation, 28–9, 120 Intergovernmental Agreement on the Environment, 131, 143 Interstate Commerce Commission, 93–4 judicial review: Australian doctrine, 56 concept, 91 disillusionment in, 123 forum for administrative constitutionalism, 100 public administration and law, 23 settlement, 123 United States, 91 see also accountability; administrative law; hard look review; merits review; scope of review Landis Report, 97 law: nature of, 18, 249, 257 public administration and, 23–4 relationship with science, 6, 15, 17, 21, 256 role of under science democracy dichotomy, 14–6 role under deliberative-constitutive paradigm, 33 role under rational-instrumental paradigm, 33 see also administrative constitutionalism; administrative law; legal culture legal culture: Australia, 54–5, 56, 57–8 co-production of law and society, 36 definition, 35–6 different jurisdictions, 37

European Union, 170, 214 framing power, 253–4 importance, 253, 255–6 international and supranational, 167 legal pluralism, 25–6, 55 regulatory styles, 189 scholarship, 35 thick cultural phenomenon, 35–6, 37, 57 United Kingdom, 52–3, 55–6, 57 United States, 53–4, 56, 57 World Trade Organisation, 182, 203–4 see also administrative constitutionalism legal pluralism, 25, 36, 161 Australia, 133 European Union, 240 state sponsored, 54, 133, 253 see also merits review legislation: guidance, 19 role under deliberative-constitutive paradigm, 30, 33 role under rational-instrumental paradigm, 28, 33 legislatures: delegating risk evaluation decisions, 19, 21, 23, 26 making risk evaluation decisions, 17 see also legislation legitimacy: output and input, 240 thru-put, 240 see also administrative constitutionalism Leventhal, Harold, 89, 90, 101–3, 106–7, 124 merits review, 54, 57, 135–40 adjudicative procedure, 139 as external review, 137–8, 145 as internal review, 138–9, 146–54 distinction between internal and external review, 136–7, 139 evidentiary principles, 138–9, 148–9 expert evidence, 143, 144–5, 149–50, 156 judicial review, 137 nature, 135–40, 161 precedents, 137 see also legal pluralism methodology: information collection and assessment, 20 role under rational-instrumental paradigm, 28, 33, 103–4 World Trade Organisation, 187 see also cost/benefit analysis; regulatory impact statements; risk assessment; science ministerial accountability, see accountability Ministry of Agriculture Fisheries and Food, 59, 71–8, 83–4 mobile phone masts, 155–8 Monday morning quarterbacking, 107

(O) Fisher Index

4/7/07

12:13

Page 287

Index most favoured nation principle, 167 multi-criteria analysis, 157 National Institute of Occupational Safety and Health, 110 National Research Council, 10 analytic-deliberative process, 13, 31, 239 Red Book, 116, 248 National Strategy for Ecologically Sustainable Development, 130–1 national treatment principle, 167 neo-functionalism, 223 networks, 167, 248, 251, 256 see also governance New Deal, 94–5, 98 New South Wales Land and Environmental Court, 134, 136, 138, 140–4, 151, 155–60 new social movements, 11 Next Steps Agencies, 85 Northcote Trevelyan report, 68, 76 Nuclear Regulatory Commission, 98, 105–6 Occupational Safety and Health Administration, 57, 98, 107–16 feasibility, 109, 115 generic carcinogenic policy, 113, 115 generic rulemaking, 117–9 judicial review, 107–16 methylene chloride rule, 122 ‘significant risk’, 113–4 standards 19, 107–8 vinyl chloride standard, 109, 113 see also Benzene, substantial evidence Office of Management and Budget, 116 Office international des epizooties, 191, 205 ossification, 122 parliamentary select committees, 73 planning law 127, 214 polycentricity, 22 precautionary principle: ‘cautious approach’, 132–3, 141–2, 143, 161 adaptive management, 149,157 adjudicative procedure, 146–7 administrative constitutionalism, 41, 42–6, 208 Australia, 55, 125–61 burdens of proof, 44–6, 146–7, 157, 160 Communication on the Precautionary Principle, 203, 224–9, 230, 232, 251 contexts of European Union operation, 212–4 cost/benefit analysis, 227, 238 definition under deliberative-constitutive paradigm, 43, 140–6, 157, 159–60, 218–24, 239 definition under rational-instrumental paradigm, 43–4, 146–54, 156–9, 224–38, 251 environmental impact assessment, 158

287

European Court of Justice and Court of First Instance case law, 218–24, 229–38 European Union 157, 166, 207–41 exercise of state authority, 41 expert evidence, 156, 160 general principle of Community law, 234, 240 history, 40, 209 inclusion in Australian environmental law, 130–3 legal culture, 40–1, 208–9, 211 legal principle, 41 legitimacy of public administration, 42–3 mobile phone masts, 155–8 overview of, 40–2 post modern law, 211 problematic characterisation as burden of proof, 14, 145 problems with a deliberative-constitutive definition, 146 problems with rational-instrumental interpretation, 143, 154 process, 41 regulatory impact assessment, 227, 228 risk assessment, 157, 158, 226, 228, 231, 236, 237 risk management, 227 risk-weighted consequences, 131, 148, 150 scientific evaluation, 228 scientific uncertainty, 41,153, 156 standard of proof, 153 tension between diversity and uniformity in interpretation, 216–8, 240–1 threat, 153, 156 trade law, 41–2 variations in definition, 214 precautionary reasoning, 218–24 prion theory, 62 see also BSE Progressive Era, 94 proof, see burden of proof; standard of proof public administration: administrative pluralism, 240 Australia, 54–5, 130–3 European Union, 171–2, 214–8 facts, 45–6 flexible institutional space, 21 International context, 254 legal validity, 26 legitimacy, 3, 22, 25, 190, 251, 257 polycentricity, 22 precautionary principle, 42–4 rise of, 21 role in BSE crisis, 64–6 role in risk evaluation, 18–26, 246 role under deliberative-constitutive paradigm, 30, 33 role under rational-instrumental paradigm, 28, 29, 33

(O) Fisher Index

4/7/07

12:13

Page 288

288

Index

public administration: (cont.): theories of, 25 United Kingdom, 52–3, 66–71 United States, 53–4, 90–9 why controversial, 22, 26 World Trade Organisation, 171, 182 see also administrative constitutionalism; administrative law Public Health Board, 66 public international law, 168, 181 public participation, see communication public reason, 9, 31, 35 public research laboratories, 17 public/private divide, 15, 256 Queensland Planning and Environmental Court, 134, 136, 149–50, 155 Railways Inspectorate, 66 rational-instrumental paradigm: adjudicative procedure, 38 appeal of, 34 Australia, 146–54, 155–9 BSE crisis, 65, 81–7 compared with deliberative-constitutive paradigm, 32–5 democracy, 34 described, 28–30 European Union, 224–38 merits review, 146–54 precautionary principle, 43–4 problems with, 34–5, 252 public participation, 120 United Kingdom, 66–7 United States, 101–3, 106–7, 112–24, 252 World Trade Organisation, 183, 186–7, 190–8 see also administrative constitutionalism reasonableness: importance of, 26–7 sanitary and phytosanitary measures, 175, 180–2, 185, 194 scope of review, 91–3 see also administrative constitutionalism reflexive law, 15 regulatory heterogeneity, 179 regulatory impact statements, 29, 116, 227 regulatory science, see science regulatory space, 172 regulatory state, 172 rights revolution, 53, 98 risk acceptability, 7, 8–9, 10, 11, 13, 18, 19 BSE, 63–4 communication, 20 risk assessment, 12, 20 BSE crisis, 80, 83, 85 comparative, 117, 122 definition, 177–8, 228, 246 hard look review, 117

precautionary principle, 158, 160 risk assessment/risk management distinction, 116, 160, 186, 187, 189, 239 Sanitary and Phytosanitary Agreement, 166 United Kingdom, 86–7 United States, 116 World Trade Organisation, 176–206 see also risk evaluation risk evaluation, 7, 10, 11, 16 collapsing of standard setting and enforcement, 55–6 democratic approaches, 12 elements, 19–22 legitimacy, 11, 16 scientific approaches, 12 see also administrative constitutionalism; technological risk risk, 6 see also technological risk Robens Report on Health and Safety at Work, 67 Rothschild report, 71, 75 Royal Commission on Environmental Pollution, 67 rule of law, 24 rulemaking: hybrid rulemaking, 99, 108 informal rulemaking, 95–6, 97, 99,105 sanitary and phytosanitary measures, 175 ‘based on’ a risk assessment, 186, 188, 192, 194, 199–200, 201 concept of probability, 186, 187, 191–2, 193, 197, 201 precautionary approach to setting, 200 reasonableness, 180–2, 185, 194 regulating administrative constitutionalism, 181–2 ‘sufficient scientific evidence’, 194, 195, 196, 197, 198, 199, 200, 202 three step risk assessment definition 191, 195–6 see also World Trade Organisation scholarship blindspots, 46, 51 science/democracy dichotomy, 3, 4, 11–3, 18, 21, 245–6, 250 administrative constitutionalism, 34 hold on risk evaluation scholarship, 18 problems with, 16–8, 32, 46, 52 public administration and, 23 role of law, 14–6 World Trade Organisation, 173, 174, 179–81, 185, 201, 202, 203 science: regulatory, 106 role under deliberative-constitutive paradigm, 30–1, 33 role under rational-instrumental paradigm, 28–9, 33

(O) Fisher Index

4/7/07

12:13

Page 289

Index United Kingdom, 69–70, 71 see also deliberative-constitutive paradigm; expertise; information; rationalinstrumental paradigm; science/democracy dichotomy scientific uncertainty, 7–8, 12 Australia, 128 BSE, 62–63 European Union, 219–20, 221–2, 226–7 nature under deliberative-constitutive paradigm, 30, 33 104 nature under rational-instrumental paradigm, 29, 33 Occupational Safety and Health Administration, 109, 114, 117, 118 precautionary principle, 41, 141, 143, 150, 153, 156, 159, 226–7 Southwood Working Party, 80–1 United States, 96–7, 122 World Trade organisation, 178, 193 see also technological risk scope of review: after Benzene, 116–20 challenges in relation to technological risk evaluation, 96–7 concept of, 91–3, 124–5 history of, 90–8 legislated grounds of review, 99 nature under deliberative-constitutive paradigm, 93–7, 104–6, 108–12 nature under rational-instrumental paradigm ,101–3, 106–7, 112–20 rational-instrumental and deliberativeconstitutive paradigms compared, 92–3 zone of reasonableness test, 95, 119 see also arbitrary and capricious; Benzene; hard look review; judicial review; substantial evidence; United States separation of powers, 19, 24 shark finning, 151 slaughterhouse practices, 63, 83 South Australian Environment Resources and Development Court, 134, 136, 147–9, 155 Southwood Working Party: bifurcated role of, 60, 76–84, 246 characterisation in terms of science/democracy dichotomy, 87–8 deliberative-constitutive understanding of, 79–81 events after, 84–7 first meeting, 79 rational-instrumental understanding of, 81–4 report, 59 79, 81 scientific uncertainty, 81 see also BSE crisis; United Kingdom species impact statement, 144 species protection, 140–3

289

Spongiform Encephalopathy Advisory Committee, 59, 85 standard of proof, 150, 153 see also burden of proof State Veterinary Service, 75 state: nature of, 11–2, 16 problem of the concept, 17 reform of, 11 substantial evidence: adjudicative procedure, 110–2, 114 Administrative Procedure Act, 96 deliberative-constitutive interpretation, 93–4, 95, 108–12 Interstate Commerce Commission, 94 jury trial, 94 New Deal, 95 Occupational Safety and Health Act, 108–16, 117–9 rational-instrumental interpretation, 112–5, 117–9 standard of review, 91, 246 see also Benzene; judicial review; scope of review; United States Tasmanian Resource Management and Planning Appeal Tribunal, 135, 136 technological risk: characterisation under deliberativeconstitutive paradigm, 30, 33 characterisation under rational-instrumental paradigm, 29, 33 collective management, 10 consideration of substitutes, 120 disputes over, 16, 22, 23, 26 European Union institutional arrangements, 211 how characterised, 18 interface with trade law, 202–3, 205–6 nature of, 7–1 1 polycentricity, 8 relationship between acceptability and identification, 10 scholarship, 11, 18, 256–7 Three Mile Island, 8 trade law, administrative constitutionalism, 25, 166 interface with regulatory law, 202–3, 205–6 precautionary principle, 42 see also European Union; World Trade Organisation transmissible spongiform encephalopathies, 62, 84 Treasury, 78 trust, 9, 10, 12 role under deliberative-constitutive paradigm, 30, 34–5

(O) Fisher Index

4/7/07

12:13

Page 290

290

Index

trust (cont.): role under rational-instrumental paradigm, 34–5 Tyrell, Committee, 85 uncertainty, see behavioural uncertainty and scientific uncertainty United Kingdom, 59–88 administrative constitutionalism, 52–3 deliberative-constitutive paradigm, 67–70, 72–4, 75–6, 79–81 food safety crises, 63 judicial review, 68 legal culture, 53. 55–6, 57, 253 local government, 67 new public management, 70–71, 78, 85, 253 policy making process, 73–74 rational-instrumental paradigm, 66–7, 70–1, 73, 74–5, 81–7 risk assessment, 86–7 role of law in risk evaluation, 55–6, 57 scientific committees, 75, 78, 79–84, 86–7 technological risk regulation in the nineteenth century, 66 see also BSE crsis; Southwood Working Party United States, 89–124 administrative agencies that regulate technological risks, 98, 246, 247 administrative constitutionalism, 53–4 de-legalisation, 123 deliberative-constitutive paradigm, 93–97, 104–6, 108–10 disillusionment in judicial review, 123 disillusionment in rulemaking processes, 122–3 distrust in administrative agencies, 97 expert evidence, 156 impact on World Trade Organisation law, 186, 189, 248 influence on European Union, 225 judicial review, 91, 100 legal culture, 53–4, 56, 57 rational-instrumental paradigm, 101–3, 106–7, 112–124, 252

role of law in risk evaluation, 56, 57 see also adversarial legalism; scope of review values, 12, 33 role under deliberative-constitutive paradigm, 30–1, 33 role under rational-instrumental paradigm, 28, 33 variant Creutzfeldt-Jakob disease: deaths from, 59 technological risk, 61–4 see also BSE Weber, Max, 28, 246 Western Australian Town Planning Appeal Tribunal, 135, 154 World Trade Organisation, 173–206 administrative constitutionalism, 166, 167–8, 171, 251, 253 role of law, 169–1 70 deliberative-constitutive paradigm, 183, 187–8 influence on European Union Law, 213, 214, 226 legal culture, 182 legitimacy, 174 problems with Sanitary and Phytosanitary Agreement jurisprudence, 201–3 regulatory heterogeneity, 179, 182, 189 rational-instrumental paradigm, 183, 186–7, 190–8 state sovereignty, 190 trade protectionism, 178–9, 182 see also sanitary and phytosanitary measures; World Trade Organisation dispute settlement World Trade Organisation dispute settlement, 168, 169–70, 183–4, 202, 204–5 Panels, 183, 184 standard of review, 183–4, 185, 189, 203 burdens of proof, 184, 187, 197, 201, 202 expert evidence, 184, 193, 197, 198, 202 limits of process, 198 problems with neutrality, 203–4 see also World Trade Organisation